Cactus 4.15
Reference Manual
commit 9032a73c39ded0e279a5200aa42d439d5c68a093
Documentation compiled on: November 20, 2024
This document will eventually be a complete reference manual for the Cactus Code. However, it is currently under development, so please be patient if you can’t find what you need. Please report omissions, errors, or suggestions to and of our contact addresses below, and we will try and fix them as soon as possible.
Overview of documentation
This guide covers the following topics
Here all the CCTK_*() Cactus flesh functions which are available to thorn writers are described.
Here all the Util_*() Cactus flesh functions which are available to thorn writers are described.
Here all the Driver_*() functions which a Driver may make available to thorn writers are described.
Other topics to be discussed in separate documents include:
This gives a general overview of the Cactus Computational Tool Kit, including overall design/architecture, how to get/configure/compile/run it, and general discussions of the how to program in Cactus.
This will contain details about the arrangements and thorns making up the Cactus Relativity Tool Kit, one of the major motivators, and still the driving force, for the Cactus Code.
This will contain all the gruesome details about the inner workings of Cactus, for all those who want or need to expand or maintain the core of Cactus.
Typographical Conventions
Is currently used for everything you type, for program names, and code extracts.
Indicates a compulsory argument.
Indicates an optional argument.
Indicates an exclusive or.
How to Contact Us
Please let us know of any errors or omissions in this guide, as well as suggestions for future editions. These can be reported via cactusmaint@cactuscode.org.
Acknowledgements
Hearty thanks to all those who have helped with documentation for the Cactus Code. Special thanks to those who struggled with the earliest sparse versions of this guide and sent in mistakes and suggestions, in particular John Baker, Carsten Gundlach, Ginny Hudak-David, Sai Iyer, Paul Lamping, Nancy Tran and Ed Seidel.
In this chapter all CCTK_* Cactus functions are described. These functions are callable from Fortran or C thorns. Note that whereas all functions are available from C, not all are currently available from Fortran.
[A27] Causes abnormal Cactus termination
[A32] Finds the thorn which activated a particular implementation
[A34] Returns the number of active timelevels from a group name
[A34] Returns the number of active timelevels from a group index
[A34] Returns the number of active timelevels from a group name
[A34] Returns the number of active timelevels from a variable index
[A34] Returns the number of active timelevels from a variable name
[A37] Returns a pointer to the local size for a group, given by its group name
[A40] Returns a pointer to the local size for a group, given by its group index
[A42] Synchronizes all processors
[A44] Registers a new named clock with the Flesh.
[A45] Turns two real numbers into a complex number (only C) [deprecated]
[A48] Returns the absolute value of a complex number (only C) [deprecated]
[A50] Returns the sum of two complex numbers (only C) [deprecated]
[A52] Returns the complex conjugate of a complex number (only C) [deprecated]
[A54] Returns the Cosine of a complex number (only C) [deprecated]
[A56] Returns the division of two complex numbers (only C) [deprecated]
[A58] Returns the Exponentiation of a complex number (only C) [deprecated]
[A60] Returns the imaginary part of a complex number (only C) [deprecated]
[A62] Returns the Logarithm of a complex number (only C) [deprecated]
[A64] Returns the multiplication of two complex numbers (only C) [deprecated]
[A66] Returns the real part of a complex number (only C) [deprecated]
[A68] Returns the Sine of a complex number (only C) [deprecated]
[A70] Returns the square root of a complex number (only C) [deprecated]
[A72] Returns the subtraction of two complex numbers (only C) [deprecated]
[A74] Gets the command line arguments.
[A75] Returns a formatted string containing the date stamp when Cactus was compiled
[A76] Returns a formatted string containing the datetime stamp when Cactus was compiled
[A77] Returns a formatted string containing the time stamp when Cactus was compiled
[A79] Return the name of the compiled implementation with given index
[A81] Return the name of the compiled thorn with given index
[A83] Give the direction for a given coordinate name (deprecated)
[A87] Give the grid variable index for a given coordinate (deprecated)
[A91] Return the global upper and lower bounds for a given coordinate name on a cctkGH (deprecated)
[A95] Register a coordinate as belonging to a coordinate system, with a given name and direction, and optionally with a grid variable (deprecated)
[A99] Saves the global upper and lower bounds for a given coordinate name on a cctkGH (deprecated)
[A103] Registers a coordinate system with a given dimension (deprecated)
[A107] Provides the dimension of a given coordinate system (deprecated)
[A111] Get the handle associated with a registered coordinate system (deprecated)
[A115] Provides the name of the coordinate system identified by its handle (deprecated)
[A117] Creates a directory
[A121] Gives the maximum number of timelevels for a group
[A125] Gives the maximum number of timelevels for a group
[A129] Gives the maximum number of timelevels for a group
[A130] Gives the maximum number of timelevels for a variable
[A134] Gives the maximum number of timelevels for a variable
[A138] Given the full name of a variable/group, separates the name returning both the implementation and the variable/group
[A140] Disable the communication for a group
[A141] Disable the communication for a group
[A143] Disable the storage for a group
[A144] Disable the storage for a group
[A145] Enable the communication for a group
[A147] Enable the communication for a group
[A149] Enable the storage for a group
[A150] Enable the storage for a group
[A152] Check a STRING or KEYWORD parameter for equality equality with a given string
[A158] Macro to print a single string as error message to standard error and stop the code
[A163] Function to print a single string as error message to standard error and stop the code
[A166] Causes normal Cactus termination
[A169] Given a group name returns the first variable index in the group
[A172] Given a group index returns the first variable index in the group
[A175] Copy the contents of a C string into a Fortran string variable
[A179] Given a group index, returns the group name
[A184] Given a variable index, returns the full name of the variable
[A181] Given a variable index, returns the full name of the variable
[A187] Given a pointer to a clock cTimerVal structure, returns the name of the clock.
[A188] Given a pointer to a clock cTimerVal structure, returns the resolution of the clock.
[A189] Given a pointer to a clock cTimerVal structure, returns the elapsed time.
[A190] Given the name of a clock, returns a pointer to the corresponding cTimerVal structure within the cTimerData structure.
[A191] Given the index of a clock, returns a pointer to the corresponding cTimerVal structure within the cTimerData structure.
[A193] Given a set of multidimensional indices compute the 1-dimensional index into a grid function.
[A195] Given a set of multidimensional indices compute the 2-dimensional index into a grid function.
[A197] Given a set of multidimensional indices compute the 3-dimensional index into a grid function.
[A199] Given a set of multidimensional indices compute the 4-dimensional index into a grid function.
[A201] Get the pointer to a registered extension to the Cactus GH structure
[A203] Get the handle associated with a extension to the Cactus GH structure
[A207] The name of the implementation of a grid array reduction operator, or NULL if the handle is invalid
[A209] Given a group index, return an array of the bounding box of the group for each face
[A209] Given a group name, return an array of the bounding box of the group for each face
[A212] Given a variable index, return an array of the bounding box of the variable for each face
[A212] Given a variable name, return an array of the bounding box of the variable for each face
[A215] Given a group index, returns information about the variables held in the group
[A219] Given a variable index, returns the dimension of all variables in the group associated with this variable
[A222] Given a group index, returns the dimension of variables in that group
[A225] Given a group index, returns information about the variables held in the group
[A227] Given a group index, returns the ghost size array of that group
[A230] Given a group index, return an array of the global size of the group in each dimension
[A230] Given a group name, return an array of the global size of the group in each dimension
[A233] Given a variable index, return an array of the global size of the variable in each dimension
[A233] Given a variable name, return an array of the global size of the variable in each dimension
[A236] Get the index number for a group name
[A240] Given a variable name, returns the index of the associated group
[A244] Given a variable index, returns the index of the associated group
[A248] Given a group index, return an array of the lower bounds of the group in each dimension
[A248] Given a group name, return an array of the lower bounds of the group in each dimension
[A251] Given a variable index, return an array of the lower bounds of the variable in each dimension
[A251] Given a variable name, return an array of the lower bounds of the variable in each dimension
[A254] Given a group index, return an array of the local size of the group in each dimension
[A254] Given a group name, return an array of the local size of the group in each dimension
[A257] Given a variable index, return an array of the local size of the variable in each dimension
[A257] Given a variable name, return an array of the local size of the variable in each dimension
[A260] Given a group index, return an array of the local allocated size of the group in each dimension
[A260] Given a group name, return an array of the local allocated size of the group in each dimension
[A263] Given a variable index, return an array of the local allocated size of the variable in each dimension
[A263] Given a variable name, return an array of the local allocated size of the variable in each dimension
[A266] Given a group index, returns the group name
[A268] Given a variable index, return the name of the associated group
[A270] Given a group index, return an array with the number of ghostzones in each dimension of the group
[A270] Given a group name, return an array with the number of ghostzones in each dimension of the group
[A273] Given a variable index, return an array with the number of ghostzones in each dimension of the variable’s group
[A273] Given a group variable, return an array with the number of ghostzones in each dimension of the variable’s group
[A276] Given a group index, returns the size array of that group
[A278] Decrease the active number of timelevels for a list of groups
[A280] Increase the active number of timelevels for a list of groups
[A282] Given a group name, return the table handle of the group’s tags table.
[A285] Given a group index, return the table handle of the group’s tags table.
[A288] Provides a group’s group type index given a variable index
[A292] Provides a group’s group type index given a group index
[A294] Given a group index, return an array of the upper bounds of the group in each dimension
[A294] Given a group name, return an array of the upper bounds of the group in each dimension
[A297] Given a variable index, return an array of the upper bounds of the variable in each dimension
[A297] Given a variable name, return an array of the upper bounds of the variable in each dimension
[A300] Given a variable index, returns the implementation name for a public or protected variable, the thorn name otherwise.
[A302] Return the ancestors for an implementation
[A304] Returns the name of one thorn providing an implementation
[A306] Return the thorns for an implementation
[A308] Macro to print a single string as an information message to screen
[A313] Function to print a single string as an information message to screen
[A317] Register one or more routines for dealing with information messages in addition to printing them to screen
[A321] Performs an interpolation on a list of CCTK grid variables, using a chosen external local interpolation operator
[A329] Returns the handle for a given interpolation operator
[A333] Interpolate a list of processor-local arrays which define a uniformly-spaced data grid
[A340] Registers a routine as a CCTK_InterpLocalUniform interpolation operator
[A343] Reports whether an aliased function has been provided
[A346] Reports whether an implementation was activated in a parameter file
[A348] Reports whether an implementation was compiled into a configuration
[A350] Reports whether a thorn was activated in a parameter file
[A353] Reports whether a thorn was compiled into a configuration
[A355] Returns the name of a registered reduction operator
[A357] Provide the implementation which provides an local array reduction operator
[A359] Returns the handle of a given local array reduction operator
[A361] Returns the maximum number of timeleves that were ever active from a group name
[A361] Returns the maximum number of timeleves that were ever active from a group index
[A361] Returns the maximum number of timeleves that were ever active from a group name
[A361] Returns the maximum number of timeleves that were ever active from a variable index
[A361] Returns the maximum number of timeleves that were ever active from a variable name
[A364] Get the maximum dimension of any grid variable
[A369] Get the maximum dimension of all grid functions
[A373] Decprecated. Use CCTK_DeclaredTimeLevels instead.
[A373] Get the local processor number
[A374] Get the total number of processors used
[A376] Returns a C-style NULL pointer value
[A378] Return the number of implementations compiled in
[A380] Return the number of thorns compiled in
[A382] The number of grid array reduction operators registered
[A384] Get the number of groups of variables compiled in the code
[A388] Returns the total number of I/O methods registered with the flesh
[A390] The number of local reduction operators registered
[A392] The number of global array reduction operators registered
[A394] Returns the number of clocks in a cTimerData structure.
[A395] Get the number of grid variables compiled in the code
[A399] Provides the number of variables in a group from the group name
[A403] Provides the number of variables in a group from the group index
[A407] Conditional output of all variables on a GH by all I/O methods
[A409] Output of a single variable by all I/O methods
[A412] Output of a single variable as an alias by all I/O methods
[A416] Output of a single variable as an alias by a single I/O method
[A419] Output of a single variable by a single I/O method
[A422] Initializes the parallel subsystem
[A423] Get parameter properties for given parameter/thorn pair
[A428] Get the data pointer to and type of a parameter’s value
[A425] Returns the parameter filename
[A430] Return the parameter checking level
[A431] Return number of times a parameter has been set
[A433] Sets the value of a parameter
[A437] Registers a parameter set operation notify callback
[A441] Unregisters a parameter set operation notify callback
[A445] Get the string representation of a parameter’s value
[A448] Walk through the list of parameters
[A450] Prints a warning from parameter checking, and possibly stops the code
[A455] Returns a pointer to a Fortran variable.
[A457] Prints a group name from its index
[A461] Prints a Cactus string to screen (from Fortran)
[A465] Prints a variable name from its index
[A469] Queries storage for a group given by its group name
[A471] Queries storage for a group given by its group name or index
[A474] Queries storage for a group given by its group index
[A476] Reduces a list of local arrays globally
[A483] Reduces a list of local arrays (new grid array reduction API)
[A489] Reduces a list of local arrays (new local array reduction API) Returns the address of a variable passed in by reference from a Fortran routine
[A497] Get the handle for a registered reduction operator
[A501] Perform a regular expression match of string against pattern
[A503] Register a banner for a thorn
[A507] Register the name of an extension to the Cactus GH
[A508] Register a routine for providing initialisation for an extension to the Cactus GH
[A509] Register a GH extension schedule traversal routine
[A511] Registers a function as a grid array reduction operator of a certain name
[A510] Register a routine for setting up an extension to the Cactus GH
[A513] Registers a new I/O method
[A515] Registers an I/O method’s routine for conditional output
[A516] Registers an I/O method’s routine for unconditional output
[A517] Register a routine for deciding if it is time to output for an IO method
[A518] Register a routine for dealing with trigger output for an IO method
[A519] Registers a function as a reduction operator of a certain name
[A521] Register a function as providing a global array reduction operation
[A523] Register a function as providing a reduction operation
[A524] Return the number of seconds since the run started
[A525] Output the timing results for a certain schedule item to stdout
[A527] Output the timing results for a certain schedule item to a file
[A529] Return the cFunctionData of the function currently executing via CCTK_CallFunction
[A531] Traverses a schedule point, and its entry and exit points if necessary
[A534] Sets up a CCTK grid hierarchy
[A535] Synchronize the ghost zones for a group of variables (identified by the group name)
[A539] Synchronize the ghost zones for a group of variables (identified by the group index)
[A544] Synchronize the ghost zones for a list of groups of variables (identified by their group indices)
[A549] Causes a Cactus simulation to terminate after the next iteration
[A551] Returns true if CCTK_TerminateNext has been called.
[A554] Returns the implementation provided by the thorn
[A557] Fills a timer cTimerData structure with current values of all clocks of a timer with a given name.
[A558] Create a timer with a given name, returns a timer index.
[A559] Allocates a timer cTimerData structure.
[A560] Create an unnamed timer, returns a timer index.
[A561] Reclaims resources for a timer with a given name.
[A562] Reclaims resources of a timer cTimerData structure.
[A563] Reclaims resources for a timer with a given index.
[A564] Fills a timer cTimerData structure with current values of all clocks of a timer with a given index.
[A565] Initialises the timer with a given name.
[A566] Initialises the timer with a given index.
[A567] Initialises the timer with a given name.
[A568] Initialises the timer with a given index.
[A569] Gets current values for all clocks of the timer with a given name.
[A570] Gets current values for all clocks of the timer with a given index.
[A575] Traverse through all variables and/or groups whose names appear in the given string.
[A577] Returns the data pointer for a grid variable
[A581] Returns the data pointer for a grid variable from the variable index or name
[A583] Returns the data pointer for a grid variable from the variable index
[A587] Get the index for a variable
[A591] Given a variable index, returns the variable name
[A594] Provides variable type index from the variable index
[A598] Provides variable type size in bytes from the variable type index
[A600] Given a set of vector and multidimensional indices compute the 1-dimensional index into a vector grid function.
[A603] Given a set of vector and multidimensional indices compute the 2-dimensional index into a vector grid function.
[A605] Given a set of vector and multidimensional indices compute the 3-dimensional index into a vector grid function.
[A607] Given a set of vector and multidimensional indices compute the 4-dimensional index into a vector grid function.
[A610] Macro to print a formatted string with a variable argument list as error message to standard error and stops the code
[A613] Prints a formatted string with a variable argument list as error message to standard error and stops the code
[A616] Macro to print a formatted string with a variable argument list as an information message to screen
[A619] Prints a formatted string with a variable argument list as an information message to screen
[A624] Prints a formatted string with a variable argument list as a warning from parameter checking, and possibly stops the code
[A622] Prints a formatted string with a variable argument list as a warning from parameter checking, and possibly stops the code
[A626] Macro to print a formatted string with a variable argument list as a warning message to standard error and possibly stops the code
[A630] Prints a formatted string with a variable argument list as a warning message to standard error and possibly stops the code
[A634] Macro to print a single string as a warning message to standard error and possibly stop the code
[A639] Function to print a single string as a warning message to standard error and possibly stop the code
[A642] Register one or more routines for dealing with warning messages in addition to printing them to standard error
Abnormal Cactus termination.
Synopsis
#include "cctk.h" int dummy = CCTK_Abort(const cGH *cctkGH, int exitcode);
#include "cctk.h" subroutine CCTK_Abort (dummy, cctkGH, exitcode) integer dummy CCTK_POINTER cctkGH integer exitcode end subroutine CCTK_Abort
Result
The function never returns, and hence never produces a result.
Parameters
GH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
exitcode Exit code that is passed to the operating system
Discussion
This routine causes an immediate, abnormal Cactus termination. It never returns to the caller.
See Also
CCTK_Exit [A166] Exit the code cleanly
CCTK_ERROR [A158] Macro to print a single string as error message and stop the code
CCTK_VError [A613] Prints a formatted string with a variable argument list as error message to standard error
and stops the code
CCTK_VWarn [A630] Prints a formatted string with a variable argument list as a warning message to standard
error and possibly stops the code
CCTK_WARN [A634] Macro to print a single string as a warning message and possibly stop the code
Errors
The function never returns, and hence never reports an error.
Examples
#include "cctk.h" CCTK_Abort (cctkGH);
#include "cctk.h" integer dummy call CCTK_Abort (dummy, cctkGH)
Finds the thorn which activated a particular implementation.
Synopsis
#include "cctk.h" const char *thorn = CCTK_ActivatingThorn(const char *name);
Result
thorn Name of activating thorn, or NULL if inactive
Parameters
name Implementation name
See Also
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
Errors
NULL The implementation is inactive, or an error occurred.
Returns the number of active time levels for a group.
Synopsis
#include "cctk.h" int timelevels = CCTK_ActiveTimeLevels(const cGH *cctkGH, const char *groupname); int timelevels = CCTK_ActiveTimeLevelsGI(const cGH *cctkGH, int groupindex); int timelevels = CCTK_ActiveTimeLevelsGN(const cGH *cctkGH, const char *groupname); int timelevels = CCTK_ActiveTimeLevelsVI(const cGH *cctkGH, int varindex); int timelevels = CCTK_ActiveTimeLevelsVN(const cGH *cctkGH, const char *varname);
#include "cctk.h" subroutine CCTK_ActiveTimeLevels(timelevels, cctkGH, groupname) integer timelevels CCTK_POINTER cctkGH character*(*) groupname end subroutine CCTK_ActiveTimeLevels subroutine CCTK_ActiveTimeLevelsGI(timelevels, cctkGH, groupindex) integer timelevels CCTK_POINTER cctkGH integer groupindex end subroutine CCTK_ActiveTimeLevelsGI subroutine CCTK_ActiveTimeLevelsGN(timelevels, cctkGH, groupname) integer timelevels CCTK_POINTER cctkGH character*(*) groupname end subroutine CCTK_ActiveTimeLevelsGN subroutine CCTK_ActiveTimeLevelsVI(timelevels, cctkGH, varindex) integer timelevels CCTK_POINTER cctkGH integer varindex end subroutine CCTK_ActiveTimeLevelsVI subroutine CCTK_ActiveTimeLevelsVN(timelevels, cctkGH, varname) integer timelevels CCTK_POINTER cctkGH character*(*) varname end subroutine CCTK_ActiveTimeLevelsVN
Result
timelevels The currently active number of timelevels for the group.
Parameters
GH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
groupname Name of the group.
groupindex Index of the group.
varname Name of a variable in the group.
varindex Index of a variable in the group.
Discussion
This function returns the number of timelevels for which storage has been activated, which is always equal to or less than the maximum number of timelevels which may have storage provided by CCTK_MaxActiveTimeLevels.
See Also
CCTK_MaxActiveTimeLevels [A361] Returns the maximum number of timeleves that were ever active from a
group name
CCTK_DeclaredTimeLevels [A121] Return the maximum number of active timelevels.
CCTK_GroupStorageDecrease [A278] Base function, overloaded by the driver, which decreases the number of
active timelevels, and also returns the number of active timelevels.
CCTK_GroupStorageIncrease [A280] Base function, overloaded by the driver, which increases the number of
active timelevels, and also returns the number of active timelevels.
Errors
timelevels \(<\) 0 Illegal arguments given.
Returns a pointer to the processor-local size for variables in a group, specified by its name, in a given dimension.
Synopsis
#include "cctk.h" int *size = CCTK_ArrayGroupSize(const cGH *cctkGH, int dir, const char *groupname);
Result
NULL A NULL pointer is returned if the group index or the dimension given are invalid.
Parameters
GH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dir (\(\ge \) 0) Which dimension of array to query.
groupname Name of the group.
Discussion
For a CCTK_ARRAY or CCTK_GF group, this routine returns a pointer to the processor-local size for variables in that group in a given direction. The direction is counted in C order (zero being the lowest dimension).
This function returns a pointer to the result for technical reasons; so that it will efficiently interface with Fortran. This may change in the future. Consider using CCTK_GroupgshGN instead.
See Also
CCTK_GroupgshGN [A230] Returns an array with the array size in all dimensions.
... There are many related functions which grab information from the GH, but many are not yet documented.
Returns a pointer to the processor-local size for variables in a group, specified by its index, in a given dimension.
Synopsis
#include "cctk.h" int *size = CCTK_ArrayGroupSizeI(const cGH *cctkGH, int dir, int groupi);
Result
NULL A NULL pointer is returned if the group index or the dimension given are invalid.
Parameters
GH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dir (\(\ge \) 0) Which dimension of array to query.
groupi The group index.
Discussion
For a CCTK_ARRAY or CCTK_GF group, this routine returns a pointer to the processor-local size for variables in that group in a given direction. The direction is counted in C order (zero being the lowest dimension).
This function returns a pointer to the result for technical reasons; so that it will efficiently interface with Fortran. This may change in the future. Consider using CCTK_GroupgshGI instead.
See Also
CCTK_GroupgshGI [A230] Returns an array with the array size in all dimensions.
... There are many related functions which grab information from the GH, but many are not yet documented.
Synchronizes all processors at a given execution point This routine synchronizes all processors in a parallel job at a given point of execution. No processor will continue execution until all other processors have called CCTK_Barrier. Note that this is a collective operation – it must be called by all processors otherwise the code will hang.
Synopsis
int istat = CCTK_Barrier(const cGH *cctkGH)
subroutine CCTK_Barrier (istat, cctkGH) integer itat CCTK_POINTER_TO_CONST cctkGH
Registers a named timer clock with the Flesh.
Synopsis
int err = CCTK_ClockRegister(name, functions)
Parameters
const char * name The name the clock will be given
const cClockFuncs * functions The structure holding the function pointers that define the clock
Discussion
The cClockFuncs structure contains function pointers defined by the clock module to be registered.
Errors
A negative return value indicates an error.
Turns two real numbers into a complex number (deprecated)
Synopsis
CCTK_COMPLEX cmpno = CCTK_Cmplx( CCTK_REAL realpart, CCTK_REAL imagpart)
Parameters
cmpno The complex number
realpart The real part of the complex number
imagpart The imaginary part of the complex number
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
cmpno = CCTK_Cmplx(re,im);
Absolute value of a complex number (deprecated)
Synopsis
CCTK_COMPLEX absval = CCTK_CmplxAbs( CCTK_COMPLEX inval)
Parameters
absval The computed absolute value
realpart The complex number who absolute value is to be returned
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
absval = CCTK_CmplxAbs(inval);
Sum of two complex numbers (deprecated)
Synopsis
CCTK_COMPLEX addval = CCTK_CmplxAdd( CCTK_COMPLEX inval1, CCTK_COMPLEX inval2)
Parameters
addval The computed added value
inval1 The first complex number to be summed
inval2 The second complex number to be summed
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
addval = CCTK_CmplxAdd(inval1,inval2);
Complex conjugate of a complex number (deprecated)
Synopsis
CCTK_COMPLEX conjgval = CCTK_CmplxConjg( CCTK_COMPLEX inval)
Parameters
conjval The computed conjugate
inval The complex number to be conjugated
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
conjgval = CCTK_CmplxConjg(inval);
Cosine of a complex number (deprecated)
Synopsis
CCTK_COMPLEX cosval = CCTK_CmplxCos( CCTK_COMPLEX inval)
Parameters
cosval The computed cosine
inval The complex number to be cosined
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
cosval = CCTK_CmplxCos(inval);
Division of two complex numbers (deprecated)
Synopsis
CCTK_COMPLEX divval = CCTK_CmplxDiv( CCTK_COMPLEX inval1, CCTK_COMPLEX inval2)
Parameters
divval The divided value
inval1 The enumerator
inval1 The denominator
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
divval = CCTK_CmplxDiv(inval1,inval2);
Exponent of a complex number (deprecated)
Synopsis
CCTK_COMPLEX expval = CCTK_CmplxExp( CCTK_COMPLEX inval)
Parameters
expval The computed exponent
inval The complex number to be exponented
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
expval = CCTK_CmplxExp(inval);
Imaginary part of a complex number (deprecated)
Synopsis
CCTK_REAL imval = CCTK_CmplxImag( CCTK_COMPLEX inval)
Parameters
imval The imaginary part
inval The complex number
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
The imaginary part of a complex number \(z=a+bi\) is \(b\).
Examples
imval = CCTK_CmplxImag(inval);
Logarithm of a complex number (deprecated)
Synopsis
CCTK_COMPLEX logval = CCTK_CmplxLog( CCTK_COMPLEX inval)
Parameters
logval The computed logarithm
inval The complex number
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
logval = CCTK_CmplxLog(inval);
Multiplication of two complex numbers (deprecated)
Synopsis
CCTK_COMPLEX mulval = CCTK_CmplxMul( CCTK_COMPLEX inval1, CCTK_COMPLEX inval2)
Parameters
mulval The product
inval1 First complex number to be multiplied
inval2 Second complex number to be multiplied
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
The product of two complex numbers \(z_1=a_1+b_1 i\) and \(z_2=a_2+b_2 i\) is \(z=(a_1 a_2 - b_1 b_2) + (a_1 b_2 + a_2 b_1)i\).
Examples
mulval = CCTK_CmplxMul(inval1,inval2);
Real part of a complex number (deprecated)
Synopsis
CCTK_REAL reval = CCTK_CmplxReal( CCTK_COMPLEX inval)
Parameters
reval The real part
inval The complex number
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
The real part of a complex number \(z=a+bi\) is \(a\).
Examples
reval = CCTK_CmplxReal(inval);
Sine of a complex number (deprecated)
Synopsis
CCTK_COMPLEX sinval = CCTK_CmplxSin( CCTK_COMPLEX inval)
Parameters
sinval The computed sine
inval The complex number to be Sined
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
sinval = CCTK_CmplxSin(inval);
Square root of a complex number (deprecated)
Synopsis
CCTK_COMPLEX sqrtval = CCTK_CmplxSqrt( CCTK_COMPLEX inval)
Parameters
expval The computed square root
inval The complex number to be square rooted
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
Examples
sqrtval = CCTK_CmplxSqrt(inval);
Subtraction of two complex numbers (deprecated)
Synopsis
CCTK_COMPLEX subval = CCTK_CmplxSub( CCTK_COMPLEX inval1, CCTK_COMPLEX inval2)
Parameters
addval The computed subtracted value
inval1 The complex number to be subtracted from
inval2 The complex number to subtract
Discussion
This function is deprecated in favor of the native complex number support in C99 and C++.
If \(z_1=a_1 + b_1 i\) and \(z_2 = a_2+ b_2 i\) then
Examples
subval = CCTK_CmplxSub(inval1,inval2);
Gets the command line arguments.
Synopsis
#include "cctk.h" int outargc = CCTK_CommandLine(char ***outargv)
Result
outargc The number of command line arguments.
Parameters
outargc Place to dump the command line arguments
See Also
CCTK_ParameterFilename [A425] Returns the parameter filename
Returns a formatted string containing the date stamp when Cactus was compiled
Synopsis
#include "cctk.h" const char *compile_date = CCTK_CompileDate ();
Result
compile_date formatted string containing the date stamp
See Also
CCTK_CompileTime [A77] Returns a formatted string containing the time stamp when Cactus was compiled
CCTK_CompileDateTime [A76] Returns a formatted string containing the datetime stamp when Cactus was
compiled
Returns a formatted string containing the datetime stamp when Cactus was compiled
Synopsis
#include "cctk.h" const char *compile_datetime = CCTK_CompileDateTime ();
Result
compile_datetime formatted string containing the datetime stamp
Discussion
If possible, the formatted string returned contains the datetime in a machine-processable format as defined in ISO 8601 chapter 5.4.
See Also
CCTK_CompileDate [A75] Returns a formatted string containing the date stamp when Cactus was compiled
CCTK_CompileTime [A77] Returns a formatted string containing the time stamp when Cactus was compiled
Returns a formatted string containing the time stamp when Cactus was compiled
Synopsis
#include "cctk.h" const char *compile_time = CCTK_CompileTime ();
Result
compile_time formatted string containing the time stamp
See Also
CCTK_CompileDate [A75] Returns a formatted string containing the date stamp when Cactus was compiled
CCTK_CompileDateTime [A76] Returns a formatted string containing the datetime stamp when Cactus was
compiled
Return the name of the compiled implementation with given index.
Synopsis
#include "cctk.h" const char *imp = CCTK_CompiledImplementation(int index);
Result
imp Name of the implementation
Parameters
index Implementation index, with \(0 \le \code {index} < \code {numimpls}\), where numimpls is returned by CCTK_NumCompiledImplementations.
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
Errors
NULL Error.
Return the name of the compiled thorn with given index.
Synopsis
#include "cctk.h" const char *thorn = CCTK_CompiledThorn(int index);
Result
thorn Name of the thorn
Parameters
index Thorn index, with \(0 \le \code {index} < \code {numthorns}\), where numthorns is returned by CCTK_NumCompiledThorns.
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
Errors
NULL Error.
Give the direction for a given coordinate.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int dir = CCTK_CoordDir( const char * coordname, const char * systemname)
call CCTK_CoordDir(dir , coordname, systemname ) integer dir character*(*) coordname character*(*) systemname
Parameters
dir The direction of the coordinate
coordname The name assigned to this coordinate
systemname The name of the coordinate system
Discussion
The coordinate name is independent of the grid function name.
Examples
direction = CCTK_CoordDir("xdir","cart3d");
call CCTK_COORDDIR(direction,"radius","spher3d")
Give the grid variable index for a given coordinate.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int index = CCTK_CoordIndex( int direction, const char * coordname, const char * systemname)
call CCTK_CoordIndex(index , direction, coordname, systemname ) integer index integer direction character*(*) coordname character*(*) systemname
Parameters
index The coordinates associated grid variable index
direction The direction of the coordinate in this coordinate system
coordname The name assigned to this coordinate
systemname The coordinate system for this coordinate
Discussion
The coordinate name is independent of the grid variable name. To find the index, the coordinate system name must be given, and either the coordinate direction or the coordinate name. The coordinate name will be used if the coordinate direction is given as less than or equal to zero, otherwise the coordinate name will be used.
Examples
index = CCTK_CoordIndex(-1,"xdir","cart3d");
call CCTK_COORDINDEX(index,one,"radius","spher2d")
Return the global upper and lower bounds for a given coordinate.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int ierr = CCTK_CoordRange( const cGH * cctkGH, CCTK_REAL * lower, CCTK_REAL * upper, int direction, const char * coordname, const char * systemname)
call CCTK_CoordRange(ierr , cctkGH, lower, upper, direction, coordname, systemname ) integer ierr CCTK_POINTER cctkGH CCTK_REAL lower CCTK_REAL upper integer direction character*(*) coordname character*(*) systemname
Parameters
ierr Error code
cctkGH pointer to CCTK grid hierarchy
lower Global lower bound of the coordinate (POINTER in C)
upper Global upper bound of the coordinate (POINTER in C)
direction Direction of coordinate in coordinate system
coordname Coordinate name
systemname Coordinate system name
Discussion
The coordinate name is independent of the grid function name. The coordinate range is registered by CCTK_CoordRegisterRange. To find the range, the coordinate system name must be given, and either the coordinate direction or the coordinate name. The coordinate direction will be used if is given as a positive value, otherwise the coordinate name will be used.
Examples
ierr = CCTK_CoordRange(cctkGH, &xmin, &xmax, -1, "xdir", "mysystem");
call CCTK_COORDRANGE(ierr, cctkGH, Rmin, Rmax, -1, "radius", "sphersystem")
Define a coordinate in a given coordinate system.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int ierr = CCTK_CoordRegisterData( int direction, const char * gvname, const char * coordname, const char * systemname)
call CCTK_CoordRegisterData(ierr , direction, gvname, coordname, systemname ) integer ierr integer direction character*(*) gvname character*(*) coordname character*(*) systemname
Parameters
ierr Error code
direction Direction of coordinate in coordinate system
gvname Name of grid variable associated with coordinate
coordname Name of this coordinate
systemname Name of this coordinate system
Discussion
There must already be a coordinate system registered, using CCTK_CoordRegisterSystem.
Examples
ierr = CCTK_CoordRegisterData(1,"coordthorn::myx","x2d","cart2d");
two = 2 call CCTK_COORDREGISTERDATA(ierr,two,"coordthorn::mytheta","spher3d")
Assign the global maximum and minimum values of a coordinate on a given grid hierachy.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int ierr = CCTK_CoordRegisterRange( const cGH * cctkGH, CCTK_REAL min, CCTK_REAL max, int direction, const char * coordname, const char * systemname)
call CCTK_CoordRegisterRange(ierr , cctkGH, min, max, direction, coordname, systemname ) integer ierr CCTK_POINTER cctkGH CCTK_REAL min CCTK_REAL max integer direction character*(*) coordname character*(*) systemname
Parameters
ierr Error code
dimension Pointer to CCTK grid hierachy
min Global minimum of coordinate
max Global maximum of coordinate
direction Direction of coordinate in coordinate system
coordname Name of coordinate in coordinate system
systemname Name of this coordinate system
Discussion
There must already be a coordinate registered with the given name, with CCTK_CoordRegisterData. The coordinate range can be accessed by CCTK_CoordRange.
Examples
ierr = CCTK_CoordRegisterRange(cctkGH,-1.0,1.0,1,"x2d","cart2d");
min = 0 max = 3.1415d0/2.0d0 two = 2 call CCTK_COORDREGISTERRANGE(ierr,min,max,two,"coordthorn::mytheta","spher3d")
Assigns a coordinate system with a chosen name and dimension.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int ierr = CCTK_CoordRegisterSystem( int dimension, const char * systemname)
call CCTK_CoordRegisterSystem(ierr , dimension, systemname ) integer ierr integer dimension character*(*) systemname
Parameters
ierr Error code
dimension Dimension of coordinate system
systemname Unique name assigned to coordinate system
Examples
ierr = CCTK_CoordRegisterSystem(3,"cart3d");
three = 3 call CCTK_COORDREGISTERSYSTEM(ierr,three,"sphersystem")
Give the dimension for a given coordinate system.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int dim = CCTK_CoordSystemDim( const char * systemname)
call CCTK_CoordSystemDim(dim , systemname ) integer dim character*(*) systemname
Parameters
dim The dimension of the coordinate system
systemname The name of the coordinate system
Examples
dim = CCTK_CoordSystemDim("cart3d");
call CCTK_COORDSYSTEMDIM(dim,"spher3d")
Returns the handle associated with a registered coordinate system.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
int handle = CCTK_CoordSystemHandle( const char * systemname)
call CCTK_CoordSystemHandle(handle , systemname ) integer handle character*(*) systemname
Parameters
handle The coordinate system handle
systemname Name of the coordinate system
Examples
handle = CCTK_CoordSystemHandle("my coordinate system");
call CCTK_CoordSystemHandle(handle,"my coordinate system")
Errors
negative A negative return code indicates an invalid coordinate system name.
Returns the name of a registered coordinate system.
All the CCTK_Coord* APIs are deprecated, and will probably be phased out fairly soon. New code should use the APIs provided by the CoordBase thorn instead (this lives in the CactusBase arrangement).
Synopsis
const char * systemname = CCTK_CoordSystemName( int handle)
Parameters
handle The coordinate system handle
systemname The coordinate system name
Discussion
No Fortran routine exists at the moment.
Examples
systemname = CCTK_CoordSystemName(handle); handle = CCTK_CoordSystemHandle(systemname);
Errors
NULL A NULL pointer is returned if an invalid handle was given.
Create a directory with required permissions
Synopsis
int ierr = CCTK_CreateDirectory( int mode, const char * pathname)
call CCTK_CreateDirectory(ierr , mode, pathname ) integer ierr integer mode character*(*) pathname
Parameters
ierr Error code
mode Permission mode for new directory as an octal number
pathname Directory to create
Discussion
To create a directory readable by everyone, but writeable only by the user running the code, the permission mode would be 0755. Alternatively, a permission mode of 0777 gives everyone unlimited access; the user’s umask setting should cut this down to whatever the user’s normal default permissions are anyway.
Note that (partly for historical reasons and partly for Fortran 77 compatability) the order of the arguments is the opposite of that of the usual Unix mkdir(2) system call.
Examples
ierr = CCTK_CreateDirectory(0755, "Results/New");
call CCTK_CREATEDIRECTORY(ierr,0755, "Results/New")
Errors
1 Directory already exists
0 Directory successfully created
-1 Memory allocation failed
-2 Failed to create directory
-3 Some component of pathname already exists but is not a directory
Gives the number of timelevels for a group
Synopsis
int numlevels = CCTK_DeclaredTimeLevels( const char * name)
call CCTK_DeclaredTimeLevels(numlevels , name ) integer numlevels character*(*) name
Parameters
name The full group name
numlevels The number of timelevels
Discussion
The group name should be in the form <implementation>::<group>
Examples
numlevels = CCTK_DeclaredTimeLevels("evolve::phivars");
call CCTK_DECLAREDTIMELEVELS(numlevels,"evolve::phivars")
Gives the number of timelevels for a group
Synopsis
int numlevels = CCTK_DeclaredTimeLevelsGI( int index)
call CCTK_DeclaredTimeLevelsGI(numlevels , index ) integer numlevels integer index
Parameters
numlevels The number of timelevels
index The group index
Examples
index = CCTK_GroupIndex("evolve::phivars") numlevels = CCTK_DeclaredTimeLevelsGI(index);
call CCTK_DECLAREDTIMELEVELSGI(numlevels,3)}
Gives the number of timelevels for a group
Synopsis
int retval = CCTK_DeclaredTimeLevelsGN(const char *group);
Result
The maximum number of timelevels this group has, or -1 if the group name is incorrect.
Parameters
group The variable group’s name
Discussion
This function and its relatives return the maximum number of timelevels that the given variable group can have active. This function does not tell you anything about how many time levels are active at the time.
Gives the number of timelevels for a variable
Synopsis
int numlevels = CCTK_DeclaredTimeLevelsVI( int index)
call CCTK_DeclaredTimeLevelsVI(numlevels , index ) integer numlevels integer index
Parameters
numlevels The number of timelevels
index The variable index
Examples
index = CCTK_VarIndex("evolve::phi") numlevels = CCTK_DeclaredTimeLevelsVI(index);
call CCTK_DECLAREDTIMELEVELSVI(numlevels,3)
Gives the number of timelevels for a variable
Synopsis
int numlevels = CCTK_DeclaredTimeLevelsVN( const char * name)
call CCTK_DeclaredTimeLevelsVN(numlevels , name ) integer numlevels character*(*) name
Parameters
name The full variable name
numlevels The number of timelevels
Discussion
The variable name should be in the form <implementation>::<variable>
Examples
numlevels = CCTK_DeclaredTimeLevelsVN("evolve::phi")
call CCTK_DECLAREDTIMELEVELSVN(numlevels,"evolve::phi")
Given the full name of a variable/group, separates the name returning both the implementation and the variable/group
Synopsis
int istat = CCTK_DecomposeName( const char * fullname, char ** imp, char ** name)
Parameters
istat Status flag returned by routine
fullname The full name of the group/variable
imp The implementation name
name The group/variable name
Discussion
The implementation name and the group/variable name must be explicitly freed after they have been used.
No Fortran routine exists at the moment.
Examples
istat = CCTK_DecomposeName("evolve::scalars",imp,name)
Turn communications off for a group of grid variables
Synopsis
int istat = CCTK_DisableGroupComm( cGH * cctkGH, const char * group)
Parameters
cctkGH pointer to CCTK grid hierarchy
Discussion
Turning off communications means that ghost zones will not be communicated during a call to CCTK_SyncGroup. By default communications are all off.
Turn communications off for a group of grid variables.
Synopsis
int istat = CCTK_DisableGroupCommI(cGH * cctkGH, int group);
Result
0 The Group has been disabled.
Parameters
cctkGH pointer to CCTK grid hierarchy
group number of group of grid variables to turn off
Discussion
Turning off communications means that ghost zones will not be communicated during a call to CCTK_SyncGroup. By default communications are all off.
See Also
CCTK_DisableGroupComm [A140] Turn communications off for a group of grid variables.
CCTK_EnableGroupCommI [A147] Turn communications on for a group of grid variables.
CCTK_EnableGroupComm [A145] Turn communications on for a group of grid variables.
Free the storage associated with a group of grid variables
Synopsis
int istat = CCTK_DisableGroupStorage( cGH * cctkGH, const char * group)
Parameters
cctkGH pointer to CCTK grid hierarchy
Deallocates memory for a group based upon its index
Synopsis
int CCTK_DisableGroupStorageI(const cGH *GH, int group);
Result
0 The group previously had storage
1 The group did not have storage to disable storage
-1 The decrease storage routine was not overloaded
Parameters
GH pointer to grid hierarchy
group index of the group to deallocate storage for
Discussion
The disable group storage routine should deallocate memory for a group and return the previous status of that memory. This default function checks for the presence of the newer GroupStorageDecrease function, and if that is not available it flags an error. If it is available it makes a call to it, passing -1 as the timelevel argument, which is supposed to mean disable all timelevels, i.e. preserving this obsolete behaviour.
Turn communications on for a group of grid variables
Synopsis
int istat = CCTK_EnableGroupComm( cGH * cctkGH, const char * group)
Parameters
cctkGH pointer to CCTK grid hierarchy
Discussion
Grid variables with communication enabled will have their ghost zones communicated during a call to CCTK_SyncGroup. In general, this function does not need to be used, since communication is automatically enabled for grid variables who have assigned storage via the schedule.ccl file.
Turn communications on for a group of grid variables.
Synopsis
int istat = CCTK_EnableGroupCommI(cGH * cctkGH, int group);
Result
0 The Group has been enabled.
Parameters
cctkGH pointer to CCTK grid hierarchy
group number of the group of grid variables to turn on
Discussion
Grid variables with communication enabled will have their ghost zones communicated during a call to CCTK_SyncGroup. In general, this function does not need to be used, since communication is automatically enabled for grid variables who have assigned storage via the schedule.ccl file.
See Also
CCTK_DisableGroupComm [A140] Turn communications off for a group of grid variables.
CCTK_DisableGroupCommI [A141] Turn communications off for a group of grid variables.
CCTK_EnableGroupComm [A147] Turn communications on for a group of grid variables.
Assign the storage for a group of grid variables
Synopsis
int istat = CCTK_EnableGroupStorage(cGH * cctkGH, const char * group);
Result
0 The Storage has been enabled.
Parameters
cctkGH pointer to CCTK grid hierarchy
group name of the group to allocate storage for
Discussion
In general this function does not need to be used, since storage assignment is best handled by the Cactus scheduler via a thorn’s schedule.ccl file.
Assign the storage for a group of grid variables
Synopsis
int istat = CCTK_EnableGroupStorageI(cGH * cctkGH, int group);
Result
0 The Storage has been enabled.
Parameters
cctkGH pointer to CCTK grid hierarchy
group Index of the group to allocate storage for
Discussion
In general this function does not need to be used, since storage assignment is best handled by the Cactus scheduler via a thorn’s schedule.ccl file.
Checks a STRING or KEYWORD parameter for equality with a given string
Synopsis
#include "cctk.h" int status = CCTK_Equals(const char* parameter, const char* value)
integer status CCTK_POINTER parameter character*(*) value status = CCTK_Equals(parameter, value)
Result
1 if the parameter is (case-independently) equal to the specified value
0 if the parameter is (case-independently) not equal to the specified value
Parameters
parameter The string or keyword parameter to compare; Cactus represents this as a CCTK_POINTER pointing to
the string value.
value The value against which to compare the string or keyword parameter. This is typically a string literal
(see the examples below).
Discussion
This function compares a Cactus parameter of type STRING or KEYWORD against a given string value. The comparison is performed case-independently, returning a 1 if the strings are equal, and zero if they differ.
Note that in Fortran code, STRING or KEYWORD parameters are passed as C pointers, and can not be treated as normal Fortran strings. Thus CCTK_Equals should be used to check the value of such a parameter. See the examples below for typical usage.
See Also
Util_StrCmpi [B19] compare two C-style strings case-independently
Errors
null pointer If either argument is passed as a null pointer, CCTK_Equals() aborts the Cactus run with an
error message. Otherwise, there are no error returns from this function.
Examples
#include "cctk.h" #include "cctk_Arguments.h" #include "cctk_Parameters.h" /* * assume this thorn has a string or keyword parameter my_parameter */ void MyThorn_some_function(CCTK_ARGUMENTS) { DECLARE_CCTK_ARGUMENTS; DECLARE_CCTK_PARAMETERS; if (CCTK_Equals(my_parameter, "option A")) { CCTK_VInfo(CCTK_THORNSTRING, "using option A"); } }
#include "cctk.h" #include "cctk_Arguments.h" #include "cctk_Functions.h" #include "cctk_Parameters.h" ! ! assume this thorn has a string or keyword parameter my_parameter ! subroutine MyThorn_some_routine(CCTK_ARGUMENTS) implicit none DECLARE_CCTK_ARGUMENTS DECLARE_CCTK_FUNCTIONS DECLARE_CCTK_PARAMETERS if (CCTK_Equals(my_parameter, "option A") /= 0) then call CCTK_INFO("using option A") end if end subroutine MyThorn_some_routine
Macro to print a single string as error message and stop the code
Synopsis
#include <cctk.h> CCTK_ERROR(const char *message);
#include "cctk.h" call CCTK_ERROR(message) character*(*) message
Parameters
message The error message to print
Discussion
This macro can be used by thorns to print a single string as error message to stderr.
CCTK_ERROR(message) expands to a call to a CCTK_Error() which is equivalent to CCTK_VError(), but without the variable-number-of-arguments feature (so it can be used from Fortran).1 The macro automatically includes details about the origin of the warning (the thorn name, the source code file name and the line number where the macro occurs).
To include variables in the error message from C, you can use the routine CCTK_VError which accepts a variable argument list. To include variables from Fortran, a string must be constructed and passed in a CCTK_ERROR macro.
See Also
CCTK_Abort [A27] Abort the code
CCTK_Exit [A166] Exit the code cleanly
CCTK_VERROR [A610] macro to print an error message with a variable argument list
CCTK_VWARN [A626] macro to print a formatted string with a variable argument list as a warning message to
standard error and possibly stops the code
CCTK_WARN [A634] Macro to print a single string as a warning message and possibly stop the code
Examples
#include <cctk.h> CCTK_ERROR("Divide by 0");
#include "cctk.h" integer myint CCTK_REAL myreal character*200 message write(message, ’(A32, G12.7, A5, I8)’) & ’Your error message, including ’, myreal, ’ and ’, myint call CCTK_ERROR(message)
Function to print a single string as error message and stop the code
Synopsis
#include <cctk.h> void CCTK_Error(int line_number, const char* file_name, const char* thorn_name,const char* message)
#include "cctk.h" call CCTK_Error(line_number, file_name, thorn_name, message) integer line_number character*(*) file_name, thorn_name, message
Parameters
line_number The line number in the originating source file where the CCTK_VError call occured. You can use
the standardized __LINE__ preprocessor macro here.
file_name The file name of the originating source file where the CCTK_VError call occured. You can use the
standardized __FILE__ preprocessor macro here.
thorn_name The thorn name of the originating source file where the CCTK_VError call occured. You can use the
CCTK_THORNSTRING macro here (defined in cctk.h).
message The error message to print
Discussion
The macro CCTK_ERROR automatically includes the line number, file name and the name of the originating thorn in the info message. It is recommended that the macro CCTK_ERROR is used to print a message rather than calling CCTK_Error directly.
See Also
CCTK_Abort [A27] Abort the code
CCTK_Exit [A166] Exit the code cleanly
CCTK_VERROR [A610] macro to print an error message with a variable argument list
CCTK_VWARN [A626] macro to print a formatted string with a variable argument list as a warning message to
standard error and possibly stops the code
CCTK_WARN [A634] Macro to print a single string as a warning message and possibly stop the code
Exit the code cleanly
Synopsis
int istat = CCTK_Exit( cGH * cctkGH, int value)
call CCTK_Exit(istat , cctkGH, value ) integer istat CCTK_POINTER cctkGH integer value
Parameters
cctkGH pointer to CCTK grid hierarchy
value the return code to exit with
Discussion
This routine causes an immediate, regular termination of Cactus. It never returns to the caller.
See Also
CCTK_Abort [A27] Abort the code
CCTK_ERROR [A158] Macro to print a single string as error message and stop the code
CCTK_VError [A613] Prints a formatted string with a variable argument list as error message to standard error
and stops the code
CCTK_VWarn [A630] Prints a formatted string with a variable argument list as a warning message to standard
error and possibly stops the code
CCTK_WARN [A634] Macro to print a single string as a warning message and possibly stop the code
Given a group name, returns the first variable index in the group.
Synopsis
#include "cctk.h" int first_varindex = CCTK_FirstVarIndex(const char* group_name);
#include "cctk.h" integer first_varindex character*(*) group_name call CCTK_FirstVarIndex(first_varindex, group_name)
Result
first_varindex (\(\ge \) 0) The first variable index in the group.
Parameters
group_name (\(\ne \) NULL in C) For C, this is a non-NULL pointer to the character-string name of the group. For
Fortran, this is the character-string name of the group. In both cases this should be of the form
"implementation::group".
Discussion
If the group contains \(N > 0\) variables, and \(V\) is the value of first_varindex returned by this function, then the group’s variables have variable indices \(V\), \(V+1\), \(V+2\), …, \(V+N-1\).
See Also
CCTK_FirstVarIndexI() Given a group index, returns the first variable index in the group.
CCTK_GroupData() Get “static” information about a group (including the number of variables in the group).
CCTK_GroupDynamicData() Get “dynamic” information about a group.
Errors
-1 Group name is invalid.
-2 Group has no members.
Given a group index, returns the first variable index in the group.
Synopsis
#include "cctk.h" int first_varindex = CCTK_FirstVarIndexI(int group_index)
#include "cctk.h" integer first_varindex, group_index call CCTK_FirstVarIndexI(first_varindex, group_index)
Result
first_varindex (\(\ge \) 0) The first variable index in the group.
Parameters
group_index (\(\ge \) 0) The group index, e.g. as returned by CCTK_GroupIndex().
Discussion
If the group contains \(N > 0\) variables, and \(V\) is the value of first_varindex returned by this function, then the group’s variables have variable indices \(V\), \(V+1\), \(V+2\), …, \(V+N-1\).
See Also
CCTK_FirstVarIndex() Given a group name, returns the first variable index in the group.
CCTK_GroupData() Get “static” information about a group (including the number of variables in the group).
CCTK_GroupDynamicData() Get “dynamic” information about a group.
Errors
-1 Group index is invalid.
-2 Group has no members.
Copy the contents of a C string into a Fortran string variable
Synopsis
#include "cctk.h" int CCTK_FortranString (char const * c_string, char * fortran_string, int fortran_length);
#include "cctk.h" subroutine CCTK_FortranString (string_length, c_string, fortran_string) CCTK_INT string_length CCTK_POINTER_TO_CONST c_string character*(*) fortran_string end subroutine
Parameters
c_string This is (a pointer to) a standard C-style (NUL-terminated) string. Typically this argument is the
name of a Cactus keyword or string paramameter.
fortran_string [This is an output argument] A Fortran character variable into which this function copies the
C string (or as much of it as will fit).
fortran_length The length of the Fortran character variable.
Result
string_length This function sets this variable to the number of characters in the C string (not counting the
terminating NUL character). If this is larger than the declared length of fortran_string then the string was
truncated. If this is negative, then an error occurred.
Discussion
String or keyword parameters in Cactus are passed into Fortran routines as pointers to C strings, which can’t be directly used by Fortran code. This subroutine copies such a C string into a Fortran character*N string variable, from where it can be used by Fortran code.
Examples
# *** this is param.ccl for some thorn *** # This example shows how we can use a Cactus string parameter to # specify the contents of a Cactus key/value table, or the name of # a Fortran output file string our_parameters "parameter string" { ".*" :: "any string acceptable to Util_TableCreateFromString()" } "order=3" string output_file_name "name of our output file" { ".*" :: "any valid file name" } "foo.dat" c *** this is sample Fortran code in this same thorn *** #include "util_Table.h" #include "cctk.h" #include "cctk_Arguments.h" #include "cctk_Parameters.h" subroutine my_Fortran_subroutine(CCTK_ARGUMENTS) DECLARE_CCTK_ARGUMENTS DECLARE_CCTK_PARAMETERS CCTK_INT :: string_length integer :: status integer :: table_handle integer, parameter:: max_string_length = 500 character*max_string_length :: our_parameters_fstring character*max_string_length :: output_file_name_fstring c c create Cactus key/value table from our_parameters parameter c call CCTK_FortranString(string_length, $ our_parameters, $ our_parameters_fstring) if (string_length .gt. max_string_length) then call CCTK_WARN(CCTK_WARN_ALERT, "’our_parameters’ string too long!") end if call Util_TableCreateFromString(table_handle, our_parameters_fstring) c c open a Fortran output file named via output_file_name parameter c call CCTK_FortranString(string_length, $ output_file_name, $ output_file_name_fstring) if (string_length .gt. max_string_length) then call CCTK_WARN(CCTK_WARN_ALERT, "’output_file_name’ string too long!") end if open (unit=9, iostat=status, status=’replace’, $ file=output_file_name_fstring)
See Also
CCTK_FullVarName() Given a variable index, returns the full name of the variable.
Given a group index, returns the group name.
Synopsis
#include <cctk.h> const char *name = CCTK_FullGroupName(int index);
Result
thorn Name of group, or NULL if group index is invalid
Parameters
index The group index
Discussion
The group name must not be freed.
Examples
#include <cctk.h> #include <stdio.h> int index = CCTK_GroupIndex("evolve::scalars"); const char *name = CCTK_FullGroupName(index); printf ("Group name: %s", name);
See Also
CCTK_FullVarName [A181] Given a variable index, returns the variable name
CCTK_GroupName [A266] Given a group index, returns the group name
Errors
NULL The group index is invalid.
Given a variable index, returns the full name of the variable
Synopsis
const char * fullname = CCTK_FullVarName( int index)
#include "cctk.h" subroutine CCTK_FullVarName(nchars, index, fullname) integer nchars integer index character*(*) fullname end subroutine CCTK_FullVarName
Parameters
implementation The full variable name
index The variable index
Discussion
The full variable name must not be freed after it has been used since the storage is maintained by the flesh.
The full variable name is in the form <implementation>::<variable> for PUBLIC or PROTECTED variables and <thorn>::<variable> for PRIVATE variables.
Examples
index = CCTK_VarIndex("evolve::phi"); name = CCTK_FullVarName(index); printf ("Variable name: %s", name);
See Also
CCTK_FullName() Given a variable index, returns the full name of the variable.
Given a variable index, returns the full name of the variable
Synopsis
char * fullname = CCTK_FullName( int index)
#include "cctk.h" subroutine CCTK_FullName(nchars, index, fullname) integer nchars integer index character*(*) fullname end subroutine CCTK_FullName
Parameters
implementation The full variable name
index The variable index
Discussion
The full variable name must be explicitly freed after it has been used.
The full variable name is in the form <implementation>::<variable> for PUBLIC or PROTECTED variables and <thorn>::<variable> for PRIVATE variables.
Examples
index = CCTK_VarIndex("evolve::phi"); name = CCTK_FullName(index); printf ("Variable name: %s", name); free (name);
Given a pointer to the cTimerVal corresponding to a timer clock returns a pointer to a string that is the name of the clock
Synopsis
const char * CCTK_GetClockName(val)
Parameters
const cTimerVal * val timer clock value pointer
Discussion
Do not attempt to free the returned pointer directly. You must use the string before calling CCTK_TimerDestroyData on the containing timer info.
Given a pointer to the cTimerVal corresponding to a timer clock returns the resolution of the clock in seconds.
Synopsis
double CCTK_GetClockResolution(val)
Parameters
const cTimerVal * val timer clock value pointer
Discussion
Ideally, the resolution should represent a good lower bound on the smallest non-zero difference between two consecutive calls of CCTK_GetClockSeconds. In practice, it is sometimes far smaller than it should be. Often it just represents the smallest value representable due to how the information is stored internally.
Given a pointer to the cTimerVal corresponding to a timer clock returns a the elapsed time in seconds between the preceding CCTK_TimerStart and CCTK_TimerStop as recorded by the requested clock.
Synopsis
double CCTK_GetClockSeconds(val)
Parameters
const cTimerVal * val timer clock value pointer
Discussion
Be aware, different clocks measure different things (proper time, CPU time spent on this process, etc.), and have varying resolution and accuracy.
Given a name of a clock that is in the given cTimerData structure, returns a pointer to the cTimerVal structure holding the clock’s value.
Synopsis
const cTimerVal * CCTK_GetClockValue(name, info)
Parameters
const char * name Name of clock
const cTimerData * info Timer information structure containing clock.
Discussion
Do not attempt to free the returned pointer directly.
Errors
A null return value indicates an error.
Given a index of a clock that is in the given cTimerData structure, returns a pointer to the cTimerVal structure holding the clock’s value.
Synopsis
const cTimerVal * CCTK_GetClockValue(index, info)
Parameters
int index Index of clock
const cTimerData * info Timer information structure containing clock.
Discussion
Do not attempt to free the returned pointer directly.
Errors
A null return value indicates an error.
Given a set of multidimensional indices compute the 1-dimensional index into a grid function.
Synopsis
int CCTK_GFINDEX1D(const cGH *restrict cctkGH, int i)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (i=0; i<cctk_lsh[0]; i++) { int const ind1d = CCTK_GFINDEX1D(cctkGH,i); rho[ind1d] = exp(-pow(r[ind1d],2)); }
See Also
CCTK_VECTGFINDEX1D() Given a set of vector and multidimensional indices compute the 1-dimensional index
into a vector grid function.
Given a set of multidimensional indices compute the 2-dimensional index into a grid function.
Synopsis
int CCTK_GFINDEX2D(const cGH *restrict cctkGH, int i, int j)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
int j Index in the j direction
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (j=0; j<cctk_lsh[1]; j++) { for (i=0; i<cctk_lsh[0]; i++) { int const ind2d = CCTK_GFINDEX2D(cctkGH,i,j); rho[ind2d] = exp(-pow(r[ind2d],2)); } }
See Also
CCTK_VECTGFINDEX2D() Given a set of vector and multidimensional indices compute the 2-dimensional index
into a vector grid function.
Given a set of multidimensional indices compute the 3-dimensional index into a grid function.
Synopsis
int CCTK_GFINDEX3D(const cGH *restrict cctkGH, int i, int j, int k)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
int j Index in the j direction
int k Index in the k direction
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (k=0; k<cctk_lsh[2]; k++) { for (j=0; j<cctk_lsh[1]; j++) { for (i=0; i<cctk_lsh[0]; i++) { int const ind3d = CCTK_GFINDEX3D(cctkGH,i,j,k); rho[ind3d] = exp(-pow(r[ind3d],2)); } } }
See Also
CCTK_VECTGFINDEX3D() Given a set of vector and multidimensional indices compute the 3-dimensional index
into a vector grid function.
Given a set of multidimensional indices compute the 4-dimensional index into a grid function.
Synopsis
int CCTK_GFINDEX4D(const cGH *restrict cctkGH, int i, int j, int k, int l)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
int j Index in the j direction
int k Index in the k direction
int l Index in the l direction
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (l=0; l<cctk_lsh[3]; l++) { for (k=0; k<cctk_lsh[2]; k++) { for (j=0; j<cctk_lsh[1]; j++) { for (i=0; i<cctk_lsh[0]; i++) { int const ind4d = CCTK_GFINDEX4D(cctkGH,i,j,k,l); rho[ind4d] = exp(-pow(r[ind4d],2)); } } } }
See Also
CCTK_VECTGFINDEX4D() Given a set of vector and multidimensional indices compute the 4-dimensional index
into a vector grid function.
Get the pointer to a registered extension to the Cactus GH structure
Synopsis
void * extension = CCTK_GHExtension( const GH * cctkGH, const char * name)
Parameters
extension The pointer to the GH extension
cctkGH The pointer to the CCTK grid hierarchy
name The name of the GH extension
Discussion
No Fortran routine exists at the moment.
Examples
void *extension = CCTK_GHExtension(GH, "myExtension");
Errors
NULL A NULL pointer is returned if an invalid extension name was given.
Get the handle associated with a extension to the Cactus GH structure
Synopsis
int handle = CCTK_GHExtensionHandle( const char * name)
call CCTK_GHExtensionHandle(handle , name ) integer handle character*(*) name
Parameters
handle The GH extension handle
group The name of the GH extension
Examples
handle = CCTK_GHExtension("myExtension") ;
call CCTK_GHExtension(handle,"myExtension")
The name of the implementation of the registered grid array reduction operator, NULL if none is registered
Synopsis
#include "cctk.h" const char *ga_reduc_imp = CCTK_GridArrayReductionOperator();
Result
ga_reduc_imp Returns the name of the implementation of the registered grid array reduction operator or NULL
if none is registered
Discussion
We only allow one grid array reduction operator currently. This function can be used to check if any grid array reduction operator has been registered.
See Also
CCTK_ReduceGridArrays() Performs reduction on a list of distributed grid arrays
CCTK_RegisterGridArrayReductionOperator() Registers a function as a grid array reduction operator of a
certain name
CCTK_NumGridArrayReductionOperators() The number of grid array reduction operators registered
Given a group index or name, return an array of the bounding box of the group for each face
Synopsis
#include "cctk.h" int status = CCTK_GroupbboxGI(const cGH *cctkGH, int dim, int *bbox, int groupindex); int status = CCTK_GroupbboxGN(const cGH *cctkGH, int dim, int *bbox, const char *groupname);
call CCTK_GroupbboxGI(status, cctkGH, dim, bbox, groupindex) call CCTK_GroupbboxGN(status, cctkGH, dim, bbox, groupname) integer status CCTK_POINTER cctkGH integer dim integer bbox(dim) integer groupindex character*(*) groupname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid group index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of group.
bbox (\(\ne \) NULL) Pointer to array which will hold the return values.
groupindex Group index.
groupname Group’s full name.
Discussion
The bounding box for a given group is returned in a user-supplied array buffer.
See Also
CCTK_GroupbboxVI, CCTK_GroupbboxVN Returns the lower bounds for a given variable.
Given a variable index or name, return an array of the bounding box of the variable for each face
Synopsis
#include "cctk.h" int status = CCTK_GroupbboxVI(const cGH *cctkGH, int dim, int *bbox, int varindex); int status = CCTK_GroupbboxVN(const cGH *cctkGH, int dim, int *bbox, const char *varname);
call CCTK_GroupbboxVI(status, cctkGH, dim, bbox, varindex) call CCTK_GroupbboxVN(status, cctkGH, dim, bbox, varname) integer status CCTK_POINTER cctkGH integer dim integer bbox(dim) integer varindex character*(*) varname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid variable index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of variable.
bbox (\(\ne \) NULL) Pointer to array which will hold the return values.
varindex Group index.
varname Group’s full name.
Discussion
The bounding box for a given variable is returned in a user-supplied array buffer.
See Also
CCTK_GroupbboxGI, CCTK_GroupbboxGN Returns the upper bounds for a given group.
Given a group index, returns information about the group and its variables.
Synopsis
#include "cctk.h" int status = CCTK_GroupData(int group_index, cGroup* group_data_buffer);
Result
0 success
Parameters
group_index The group index for which the information is desired.
group_data_buffer (\(\ne \) NULL) Pointer to a cGroup structure in which the information should be stored. See the
”Discussion” section below for more information about this structure.
Discussion
The cGroup structure2 contains (at least) the following members:3
int grouptype; /* group type, as returned by CCTK_GroupTypeNumber() */ int vartype; /* variable type, as returned by CCTK_VarTypeNumber() */ int disttype; /* distribution type, */ /* as returned by CCTK_GroupDistribNumber() */ int dim; /* dimension (rank) of the group */ /* e.g. 3 for a group of 3-D variables */ int numvars; /* number of variables in the group */ int numtimelevels; /* declared number of time levels for this group’s variables */ int vectorgroup; /* 1 if this is a vector group, 0 if it’s not */ int vectorlength; /* vector length of group */ /* (i.e. number of vector elements) */ /* (it is numvars = vectorlength * num_basevars, */ /* where num_basevars is the number of */ /* variables that have been given names in the */ /* interface.ccl) */ /* 1 if this isn’t a vector group */ int tagstable; /* handle to the group’s tags table; */ /* this is a Cactus key-value table used to store */ /* metadata about the group and its variables, */ /* such as the variables’ tensor types */
See Also
"interface.ccl" Defines variables, groups, tags tables, and lots of other things.
CCTK_FullGroupName [A179] Gets the group name for a given group index.
CCTK_GroupDynamicData [A225] Gets grid-size information for a group’s variables.
CCTK_GroupIndex [A236] Gets the group index for a given group name.
CCTK_GroupIndexFromVar [A240] Gets the group index for a given variable name.
CCTK_GroupName [A266] Gets the group name for a given group index.
CCTK_GroupNameFromVarI [A268] Gets the group name for a given variable name.
CCTK_GroupTypeI [A292] Gets a group type index for a given group index.
CCTK_GroupTypeFromVarI [A288] Gets a group type index for a given variable index.
Errors
-1 group_index is invalid.
-2 group_data_buffer is NULL.
Examples
#include <stdio.h> #include "cctk.h" cGroup group_info; int group_index, status; group_index = CCTK_GroupIndex("BSSN_MoL::ADM_BSSN_metric"); if (group_index < 0) CCTK_VWarn(CCTK_WARN_ABORT, "error return %d trying to get BSSN metric’s group index!", group_index); /*NOTREACHED*/ status = CCTK_GroupData(group_index, &group_info); if (status < 0) CCTK_VWarn(CCTK_WARN_ABORT, "error return %d trying to get BSSN metric’s group information!", status); /*NOTREACHED*/ printf("this group’s arrays are %-dimensional and have %d time levels\n", group_info.dim, group_info.numtimelevels);
Given a variable index, returns the dimension of all variables in the corresponding group.
Synopsis
#include "cctk.h" int dim = CCTK_GroupDimFromVarI(int varindex);
call CCTK_GroupDimFromVarI(dim, varindex)
Result
positive the dimension of the group
-1 invalid variable index
Parameters
varindex Variable index
Discussion
The dimension of all variables in a group associcated with the given variable is returned.
See Also
CCTK_GroupDimI Returns the dimension for a given group
Given a group index, returns the dimension of that group.
Synopsis
#include "cctk.h" int dim = CCTK_GroupDimI(int groupindex);
call CCTK_GroupDimI(dim, groupindex)
Result
positive the dimension of the group
-1 invalid group index
Parameters
groupindex Group index
Discussion
The dimension of variables in the given group is returned.
See Also
CCTK_GroupDimFromVarI Returns the dimension for a group given by a member variable index
Returns the driver’s internal data for a given group
Synopsis
#include "cctk.h" int retval = CCTK_GroupDynamicData (const cGH *GH, int group, cGroupDynamicData *data);
Result
0 Sucess
-1 the given pointer to the data structure data is null
-3 the givenGH pointer is invalid
-77 the requested group has zero variables
Parameters
GH a valid initialized GH structure for your driver
group the index of the group you’re interested in
data a pointer to a caller-supplied data structure to store the group data
Discussion
This function returns information about the given grid hierarchy. The data structure used to store the information in is of type cGroupDynamicData. The members of this structure that are set are:
dim: The number of dimensions in this group.
lsh: The (process-)local size.
ash: The (process-)local allocated size.
gsh: The global grid size.
lbnd: The lowest index of the local grid as seen on the global grid. (These use zero based indexing.)
ubnd: The largest index of the local grid as seen on the global grid. (These use zero based indexing.)
tile_min: The lowest index of the local grid that should be written by this thread. (These use zero based indexing.)
tile_max: The largest index minus one of the local grid that should be written by this thread. (These use zero based indexing.)
nghostzones: The number of ghostzones for each dimension.
bbox: Values indicating whether these are inter-process boundaries (0) or physical boundaries (1).
activetimelevels: The number of active time levels.
-
Given a group index, return a pointer to an array containing the ghost sizes of the group in each dimension.
Synopsis
#include "cctk.h" CCTK_INT **ghostsizes = CCTK_GroupGhostsizesI(int groupindex);
Result
non-NULL a pointer to the ghost size array
NULL invalid group index
Parameters
groupindex Group index
Discussion
The ghost sizes in each dimension for a given group are returned as a pointer reference.
See Also
CCTK_GroupDimI Returns the dimension for a group.
CCTK_GroupSizesI Returns the size arrays for a group.
Given a group index or name, return an array of the global size of the group in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GroupgshGI(const cGH *cctkGH, int dim, int *gsh, int groupindex); int status = CCTK_GroupgshGN(const cGH *cctkGH, int dim, int *gsh, const char *groupname);
call CCTK_GroupgshGI(status, cctkGH, dim, gsh, groupindex) call CCTK_GroupgshGN(status, cctkGH, dim, gsh, groupname) integer status CCTK_POINTER cctkGH integer dim integer gsh(dim) integer groupindex character*(*) groupname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid group name
Parameters
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of group.
gsh (\(\ne \) NULL) Pointer to array which will hold the return values.
groupindex Index of the group.
groupname Name of the group.
Discussion
The global size in each dimension for a given group is returned in a user-supplied array buffer.
See Also
CCTK_GroupgshVI, CCTK_GroupgshVN Returns the global size for a given variable.
CCTK_GrouplshGI, CCTK_GrouplshGN Returns the local size for a given group.
CCTK_GrouplshVI, CCTK_GrouplshVN Returns the local size for a given variable.
CCTK_GroupashGI, CCTK_GroupashGN Returns the local allocated size for a given group.
CCTK_GroupashVI, CCTK_GroupashVN Returns the local allocated size for a given variable.
Given a variable index or its full name, return an array of the global size of the variable in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GroupgshVI(const cGH *cctkGH, int dim, int *gsh, int varindex); int status = CCTK_GroupgshVN(const cGH *cctkGH, int dim, int *gsh, const char *varname);
call CCTK_GroupgshVI(status, cctkGH, dim, gsh, varindex) call CCTK_GroupgshVN(status, cctkGH, dim, gsh, varname) integer status CCTK_POINTER cctkGH integer dim integer gsh(dim) integer varindex chararacter*(*) varname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid variable index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of variable.
gsh (\(\ne \) NULL) Pointer to array which will hold the return values.
varindex Variable index.
varname Variable’s full name.
Discussion
The global size in each dimension for a given variable is returned in a user-supplied array buffer.
See Also
CCTK_GroupgshGI, CCTK_GroupgshGN Returns the global size for a given group.
CCTK_GrouplshGI, CCTK_GrouplshGN Returns the local size for a given group.
CCTK_GrouplshVI, CCTK_GrouplshVN Returns the local size for a given variable.
CCTK_GroupashGI, CCTK_GroupashGN Returns the local size for a given group.
CCTK_GroupashVI, CCTK_GroupashVN Returns the local size for a given variable.
Get the index number for a group name
Synopsis
int index = CCTK_GroupIndex( const char * groupname)
call CCTK_GroupIndex(index , groupname ) integer index character*(*) groupname
Parameters
groupname The name of the group
Discussion
The group name should be the given in its fully qualified form, that is <implementation>::<group> for a public or protected group, and <thornname>::<group> for a private group.
Examples
index = CCTK_GroupIndex("evolve::scalars");
call CCTK_GroupIndex(index,"evolve::scalars")
Given a variable name, returns the index of the associated group
Synopsis
int groupindex = CCTK_GroupIndexFromVar( const char * name)
call CCTK_GroupIndexFromVar(groupindex , name ) integer groupindex character*(*) name
Parameters
groupindex The index of the group
name The full name of the variable
Discussion
The variable name should be in the form <implementation>::<variable>
Examples
groupindex = CCTK_GroupIndexFromVar("evolve::phi") ;
call CCTK_GROUPINDEXFROMVAR(groupindex,"evolve::phi")
Given a variable index, returns the index of the associated group
Synopsis
int groupindex = CCTK_GroupIndexFromVarI( int varindex)
call CCTK_GroupIndexFromVarI(groupindex , varindex ) integer groupindex integer varindex
Parameters
groupindex The index of the group
varindex The index of the variable
Examples
index = CCTK_VarIndex("evolve::phi"); groupindex = CCTK_GroupIndexFromVarI(index);
call CCTK_VARINDEX("evolve::phi") CCTK_GROUPINDEXFROMVARI(groupindex,index)
Given a group index or name, return an array of the lower bounds of the group in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GrouplbndGI(const cGH *cctkGH, int dim, int *lbnd, int groupindex); int status = CCTK_GrouplbndGN(const cGH *cctkGH, int dim, int *lbnd, const char *groupname);
call CCTK_GrouplbndGI(status, cctkGH, dim, lbnd, groupindex) call CCTK_GrouplbndGN(status, cctkGH, dim, lbnd, groupname) integer status CCTK_POINTER cctkGH integer dim integer lbnd(dim) integer groupindex character*(*) groupname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid group index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of group.
lbnd (\(\ne \) NULL) Pointer to array which will hold the return values.
groupindex Group index.
groupname Group’s full name.
Discussion
The lower bounds in each dimension for a given group is returned in a user-supplied array buffer.
See Also
CCTK_GrouplbndVI, CCTK_GrouplbndVN Returns the lower bounds for a given variable.
CCTK_GroupubndGI, CCTK_GroupubndGN Returns the upper bounds for a given group.
CCTK_GroupubndVI, CCTK_GroupubndVN Returns the upper bounds for a given variable.
Given a variable index or name, return an array of the lower bounds of the variable in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GrouplbndVI(const cGH *cctkGH, int dim, int *lbnd, int varindex); int status = CCTK_GrouplbndVN(const cGH *cctkGH, int dim, int *lbnd, const char *varname);
call CCTK_GrouplbndVI(status, cctkGH, dim, lbnd, varindex) call CCTK_GrouplbndVN(status, cctkGH, dim, lbnd, varname) integer status CCTK_POINTER cctkGH integer dim integer lbnd(dim) integer varindex character*(*) varname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid variable index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of variable.
lbnd (\(\ne \) NULL) Pointer to array which will hold the return values.
varindex Group index.
varname Group’s full name.
Discussion
The lower bounds in each dimension for a given variable is returned in a user-supplied array buffer.
See Also
CCTK_GrouplbndGI, CCTK_GrouplbndGN Returns the lower bounds for a given group.
CCTK_GroupubndGI, CCTK_GroupubndGN Returns the upper bounds for a given group.
CCTK_GroupubndVI, CCTK_GroupubndVN Returns the upper bounds for a given variable.
Given a group index or name, return an array of the local size of the group in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GrouplshGI(const cGH *cctkGH, int dim, int *lsh, int groupindex); int status = CCTK_GrouplshGN(const cGH *cctkGH, int dim, int *lsh, const char *groupname);
call CCTK_GrouplshGI(status, cctkGH, dim, lsh, groupindex) call CCTK_GrouplshGN(status, cctkGH, dim, lsh, groupname) integer status CCTK_POINTER cctkGH integer dim integer lsh(dim) integer groupindex character*(*) groupname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid group name
Parameters
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of group.
lsh (\(\ne \) NULL) Pointer to array which will hold the return values.
groupindex Index of the group.
groupname Name of the group.
Discussion
The local size in each dimension for a given group is returned in a user-supplied array buffer.
See Also
CCTK_GroupgshGI, CCTK_GroupgshGN Returns the global size for a given group.
CCTK_GroupgshVI, CCTK_GroupgshVN Returns the global size for a given variable.
CCTK_GrouplshVI, CCTK_GrouplshVN Returns the local size for a given variable.
CCTK_GroupashGI, CCTK_GroupashGN Returns the local allocated size for a given group.
CCTK_GroupashVI, CCTK_GroupashVN Returns the local allocated size for a given variable.
Given a variable index or its full name, return an array of the local size of the variable in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GrouplshVI(const cGH *cctkGH, int dim, int *lsh, int varindex); int status = CCTK_GrouplshVN(const cGH *cctkGH, int dim, int *lsh, const char *varname);
call CCTK_GrouplshVI(status, cctkGH, dim, lsh, varindex) call CCTK_GrouplshVN(status, cctkGH, dim, lsh, varname) integer status CCTK_POINTER cctkGH integer dim integer lsh(dim) integer varindex character*(*) varname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid variable index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of variable.
lsh (\(\ne \) NULL) Pointer to array which will hold the return values.
varindex Variable index.
varname Variable’s full name.
Discussion
The local size in each dimension for a given variable is returned in a user-supplied array buffer.
See Also
CCTK_GroupgshGI, CCTK_GroupgshGN Returns the global size for a given group.
CCTK_GroupgshVI, CCTK_GroupgshVN Returns the global size for a given variable.
CCTK_GrouplshGI, CCTK_GrouplshGN Returns the local size for a given group.
CCTK_GroupashGI, CCTK_GroupashGN Returns the local allocated size for a given group.
CCTK_GroupashVI, CCTK_GroupashVN Returns the local allocated size for a given variable.
Given a group index or name, return an array of the local allocated size of the group in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GroupashGI(const cGH *cctkGH, int size, int *ash, int groupindex); int status = CCTK_GroupashGN(const cGH *cctkGH, int size, int *ash, const char *groupname);
call CCTK_GroupashGI(status, cctkGH, size, ash, groupindex) call CCTK_GroupashGN(status, cctkGH, size, ash, groupname) integer status CCTK_POINTER cctkGH integer size integer ash(size) integer groupindex character*(*) groupname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid group name
Parameters
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
size (\(\ge 1\)) Size of output array, should be at least dimension of group.
ash (\(\ne \) NULL) Pointer to array which will hold the return values.
groupindex Index of the group.
groupname Name of the group.
Discussion
The local allocated size in each dimension for a given group is returned in a user-supplied array buffer.
See Also
CCTK_GroupgshGI, CCTK_GroupgshGN Returns the global size for a given group.
CCTK_GroupgshVI, CCTK_GroupgshVN Returns the global size for a given variable.
CCTK_GrouplshGI, CCTK_GrouplshGN Returns the local size for a given group.
CCTK_GrouplshVI, CCTK_GrouplshVN Returns the local size for a given variable.
CCTK_GroupashVI, CCTK_GroupashVN Returns the local allocated size for a given variable.
Given a variable index or its full name, return an array of the local allocated size of the variable in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GroupashVI(const cGH *cctkGH, int size, int *ash, int varindex); int status = CCTK_GroupashVN(const cGH *cctkGH, int size, int *ash, const char *varname);
call CCTK_GroupashVI(status, cctkGH, size, ash, varindex) call CCTK_GroupashVN(status, cctkGH, size, ash, varname) integer status CCTK_POINTER cctkGH integer size integer ash(size) integer varindex character*(*) varname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid variable index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
size (\(\ge 1\)) Size of output array, should be at least dimension of group.
ash (\(\ne \) NULL) Pointer to array which will hold the return values.
varindex Variable index.
varname Variable’s full name.
Discussion
The local allocated size in each dimension for a given variable is returned in a user-supplied array buffer.
See Also
CCTK_GroupgshGI, CCTK_GroupgshGN Returns the global size for a given group.
CCTK_GroupgshVI, CCTK_GroupgshVN Returns the global size for a given variable.
CCTK_GrouplshGI, CCTK_GrouplshGN Returns the local size for a given group.
CCTK_GrouplshVI, CCTK_GrouplshVN Returns the local size for a given variable.
CCTK_GroupashGI, CCTK_GroupashGN Returns the local allocated size for a given group.
Given a group index, returns the group name
Synopsis
char * name = CCTK_GroupName( int index)
Parameters
name The group name
index The group index
Discussion
The group name must be explicitly freed after it has been used.
Examples
index = CCTK_GroupIndex("evolve::scalars"); name = CCTK_GroupName(index); printf ("Group name: %s", name); free (name);
Given a variable index, return the name of the associated group
Synopsis
char * group = CCTK_GroupNameFromVarI(int varindex)
Parameters
group The name of the group
varindex The index of the variable
Examples
index = CCTK_VarIndex("evolve::phi"); group = CCTK_GroupNameFromVarI(index) ;
Given a group index or name, return an array with the number of ghostzones in each dimension of the group
Synopsis
#include "cctk.h" int status = CCTK_GroupnghostzonesGI(const cGH *cctkGH, int dim, int *nghostzones, int groupindex) int status = CCTK_GroupnghostzonesGN(const cGH *cctkGH, int dim, int *nghostzones, const char *groupname)
call CCTK_GroupnghostzonesGI(status, cctkGH, dim, nghostzones, groupindex) call CCTK_GroupnghostzonesGN(status, cctkGH, dim, nghostzones, groupname) integer status CCTK_POINTER cctkGH integer dim integer nghostzones(dim) integer groupindex character*(*) groupname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of group.
nghostzones (\(\ne \) NULL) Pointer to array which will hold the return values.
groupindex Group index.
groupname Group name.
Discussion
The number of ghostzones in each dimension for a given group is returned in a user-supplied array buffer.
See Also
CCTK_GroupnghostzonesVI, CCTK_GroupnghostzonesVN Returns the number of ghostzones for a given
variable.
Given a variable index or its full name, return an array with the number of ghostzones in each dimension of the variable
Synopsis
#include "cctk.h" int status = CCTK_GroupnghostzonesVI(const cGH *cctkGH, int dim, int *nghostzones, int varindex) int status = CCTK_GroupnghostzonesVN(const cGH *cctkGH, int dim, int *nghostzones, const char *varname)
call CCTK_GroupnghostzonesVI(status, cctkGH, dim, nghostzones, varindex) call CCTK_GroupnghostzonesVN(status, cctkGH, dim, nghostzones, varname) integer status CCTK_POINTER cctkGH integer dim integer nghostzones(dim) integer varindex character*(*) varname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of group.
nghostzones (\(\ne \) NULL) Pointer to array which will hold the return values.
varindex Variable index.
varname Variable’s full name.
Discussion
The number of ghostzones in each dimension for a given variable is returned in a user-supplied array buffer.
See Also
CCTK_GroupnghostzonesGI, CCTK_GroupnghostzonesGN Returns the number of ghostzones for a given group.
Given a group index, return a pointer to an array containing the sizes of the group in each dimension.
Synopsis
#include "cctk.h" CCTK_INT **ghostsizes = CCTK_GroupSizesI(int groupindex);
Result
non-NULL a pointer to the size array
NULL invalid group index
Parameters
groupindex Group index
Discussion
The sizes in each dimension for a given group are returned as a pointer reference.
See Also
CCTK_GroupDimI Returns the dimension for a group.
CCTK_GroupGhostsizesI Returns the size arrays for a group.
Decrease the number of timelevels allocated for the given variable groups.
Synopsis
int numTL = CactusDefaultGroupStorageDecrease (const cGH *GH, int n_groups, const int *groups, const int *timelevels, int *status);
Result
The new total number of timelevels with storage enabled for all groups queried or modified.
Parameters
GH pointer to grid hierarchy
n_groups Number of groups
groups list of group indices to reduce storage for
timelevels number of time levels to reduce storage for for each group
groups list of group indices to allocate storage for
status optional return array which, if not NULL, will, on return, contain the number of timelevels which were
previously allocated storage for each group
Discussion
The decrease group storage routine decreases the memory allocated to the specified number of timelevels for each listed group, returning the previous number of timelevels enabled for that group in the status array, if that is not NULL. It never increases the number of timelevels enabled, i.e., if it is asked to reduce to more timelevels than are enabled, it does not change the storage for that group.
There is a default implementation which checks for the presence of the older DisableGroupStorage function, and if that is not available it flags an error. If it is available it makes a call to it, and puts its return value in the status flag for the group. Usually, a driver has overloaded the default implementation.
A driver should replace the appropriate GV pointers on the cGH structure when it changes the storage state of a GV.
Increases the number of timelevels allocated for the given variable groups.
Synopsis
int numTL = CactusDefaultGroupStorageIncrease (const cGH *GH, int n_groups, const int *groups, const int *timelevels, int *status);
Result
The new total number of timelevels with storage enabled for all groups queried or modified.
Parameters
GH pointer to grid hierarchy
n_groups Number of groups
groups list of group indices to allocate storage for
timelevels number of time levels to allocate storage for for each group
groups list of group indices to allocate storage for
status optional return array which, if not NULL, will, on return, contain the number of timelevels which were
previously allocated storage for each group
Discussion
The increase group storage routine increases the allocated memory to the specified number of timelevels of each listed group, returning the previous number of timelevels enabled for that group in the status array, if that is not NULL. It never decreases the number of timelevels enabled, i.e., if it is asked to enable less timelevels than are already enabled it does not change the storage for that group.
There is a default implementation which checks for the presence of the older EnableGroupStorage function, and if that is not available it flags an error. If it is available it makes a call to it, and puts its return value in the status flag for the group. Usually, a driver has overloaded the default implementation.
A driver should replace the appropriate GV pointers on the cGH structure when it changes the storage state of a GV.
Given a group name, return the table handle of the group’s tags table.
Synopsis
#include "cctk.h" int table_handle = CCTK_GroupTagsTable(const char* group_name);
#include "cctk.h" integer table_handle character*(*) group_name call CCTK_VarIndex(table_handle, group_name)
Result
table_handle The table handle of the group’s tags table.
Parameters
group_name The character-string name of group. This should be given in its fully qualified form, that is
implementation::group_name or thorn_name::group_name.
See Also
CCTK_GroupData [A215] This function returns a variety of “static” information about a group (“static” in the
sense that it doesn’t change during a Cactus run).
CCTK_GroupDynamicData [A225] This function returns a variety of “dynamic” information about a group
(“dynamic” in the sense that a driver can (and often does) change this information during a Cactus run).
Errors
-1 no group exists with the specified name
Given a group name, return the table handle of the group’s tags table.
Synopsis
#include "cctk.h" int table_handle = CCTK_GroupTagsTableI(int group_index);
#include "cctk.h" integer table_handle integer group_index call CCTK_VarIndex(table_handle, group_index)
Result
table_handle The table handle of the group’s tags table.
Parameters
group_index The group index of the group.
See Also
CCTK_GroupData [A215] This function returns a variety of “static” information about a group (“static” in the
sense that it doesn’t change during a Cactus run).
CCTK_GroupDynamicData [A225] This function returns a variety of “dynamic” information about a group
(“dynamic” in the sense that a driver can (and often does) change this information during a Cactus run).
CCTK_GroupIndex [A236] Get the group index for a specified group name.
CCTK_GroupIndexFromVar [A240] Get the group index for the group containing the variable with a specified
name.
CCTK_GroupIndexFromVarI [A244] Get the group index for the group containing the variable with a specified
variable index.
Errors
-1 no group exists with the specified name
Provides a group’s group type index given a variable index
Synopsis
int type = CCTK_GroupTypeFromVarI( int index)
call CCTK_GroupTypeFromVarI(type , index ) integer type integer index
Parameters
type The group’s group type index
group The variable index
Discussion
The group’s group type index indicates the type of variables in the group. Either scalars, grid functions or arrays. The group type can be checked with the Cactus provided macros for CCTK_SCALAR, CCTK_GF, CCTK_ARRAY.
Examples
index = CCTK_GroupIndex("evolve::scalars") array = (CCTK_ARRAY == CCTK_GroupTypeFromVarI(index));
call CCTK_GROUPTYPEFROMVARI(type,3)
Provides a group’s group type index given a group index
Synopsis
#include "cctk.h" int group_type = CCTK_GroupTypeI(int group);
Result
-1 -1 is returned if the given group index is invalid.
Parameters
group Group index.
Discussion
A group’s group type index indicates the type of variables in the group. The three group types are scalars, grid functions, and grid arrays. The group type can be checked with the Cactus provided macros for CCTK_SCALAR, CCTK_GF, CCTK_ARRAY.
See Also
CCTK_GroupTypeFromVarI [A288] This function takes a variable index rather than a group index as its
argument.
Given a group index or name, return an array of the upper bounds of the group in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GroupubndGI(const cGH *cctkGH, int dim, int *ubnd, int groupindex); int status = CCTK_GroupubndGN(const cGH *cctkGH, int dim, int *ubnd, const char *groupname);
call CCTK_GroupubndGI(status, cctkGH, dim, ubnd, groupindex) call CCTK_GroupubndGN(status, cctkGH, dim, ubnd, groupname) integer status CCTK_POINTER cctkGH integer dim integer ubnd(dim) integer groupindex character*(*) groupname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid group index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of group.
ubnd (\(\ne \) NULL) Pointer to array which will hold the return values.
groupindex Group index.
groupname Group’s full name.
Discussion
The upper bounds in each dimension for a given group is returned in a user-supplied array buffer.
See Also
CCTK_GrouplbndGI, CCTK_GrouplbndGN Returns the lower bounds for a given group.
CCTK_GrouplbndVI, CCTK_GrouplbndVN Returns the lower bounds for a given variable.
CCTK_GroupubndVI, CCTK_GroupubndVN Returns the upper bounds for a given variable.
Given a variable index or name, return an array of the upper bounds of the variable in each dimension
Synopsis
#include "cctk.h" int status = CCTK_GroupubndVI(const cGH *cctkGH, int dim, int *ubnd, int varindex); int status = CCTK_GroupubndVN(const cGH *cctkGH, int dim, int *ubnd, const char *varname);
call CCTK_GroupubndVI(status, cctkGH, dim, ubnd, varindex) call CCTK_GroupubndVN(status, cctkGH, dim, ubnd, varname) integer status CCTK_POINTER cctkGH integer dim integer ubnd(dim) integer varindex character*(*) varname
Result
0 success
-1 incorrect dimension supplied
-2 data not available from driver
-3 called on a scalar group
-4 invalid variable index
Parameters
status Return value.
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dim (\(\ge 1\)) Number of dimensions of variable.
ubnd (\(\ne \) NULL) Pointer to array which will hold the return values.
varindex Group index.
varname Group’s full name.
Discussion
The upper bounds in each dimension for a given variable is returned in a user-supplied array buffer.
See Also
CCTK_GrouplbndGI, CCTK_GrouplbndGN Returns the lower bounds for a given group.
CCTK_GrouplbndVI, CCTK_GrouplbndVN Returns the lower bounds for a given variable.
CCTK_GroupubndGI, CCTK_GroupubndGN Returns the upper bounds for a given group.
Given a variable index, returns the implementation name
Synopsis
char * implementation = CCTK_ImpFromVarI( int index)
Parameters
implementation The implementation name if the argument is the index of a public or protected variable, the
thorn name otherwise.
index The variable index
Discussion
No Fortran routine exists at the moment
Examples
index = CCTK_VarIndex("evolve::phi"); implementation = CCTK_ImpFromVarI(index);
Return the ancestors for an implementation.
Synopsis
#include "cctk.h" uStringList *imps = CCTK_ImplementationRequires(const char *imp);
Result
imps (not documented)
Parameters
imp (not documented)
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
Errors
(not documented)
Returns the name of one thorn providing an implementation.
Synopsis
#include "cctk.h" const char *thorn = CCTK_ImplementationThorn(const char *name);
Result
thorn Name of the thorn or NULL
Parameters
name Name of the implementation
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
Errors
NULL Error.
Return the thorns for an implementation.
Synopsis
#include "cctk.h" t_sktree *thorns = CCTK_ImpThornList(const char *name);
Result
thorns (not documented)
Parameters
name Name of implementation
Discussion
(not documented)
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
Errors
(not documented)
Macro to print a single string as an information message to screen
Synopsis
#include <cctk.h> CCTK_INFO(const char *message);
#include "cctk.h" call CCTK_INFO(message) character*(*) message
Parameters
message The string to print as an info message
Discussion
This macro can be used by thorns to print a single string as an info message to screen.
The macro CCTK_INFO(message) expands to a call to the underlying function CCTK_Info:
CCTK_Info(CCTK_THORNSTRING, message)
So the macro automatically includes the name of the originating thorn in the info message. It is recommended that the macro CCTK_INFO is used to print a message rather than calling CCTK_Info directly.
To include variables in an info message from C, you can use the routine CCTK_VInfo which accepts a variable argument list. To include variables from Fortran, a string must be constructed and passed in a CCTK_INFO macro.
See Also
CCTK_ERROR [A158] macro to print an error message with a single string argument and stop the code
CCTK_VERROR [A610] macro to print a formatted string with a variable argument list as error message and stops
the code
CCTK_VINFO() [A616] macro to print a formatted string with a variable argument list as an info message to
screen
CCTK_VWARN [A626] macro to print a warning message with a variable argument list
CCTK_WARN [A634] macro to print a warning message with a single string argument and possibly stop the code
Examples
#include <cctk.h> CCTK_INFO("Output is disabled");
#include "cctk.h" integer myint real myreal character*200 message write(message, ’(A32, G12.7, A5, I8)’) & ’Your info message, including ’, myreal, ’ and ’, myint call CCTK_INFO(message)
Function to print a single string as an information message to screen
Synopsis
#include <cctk.h> CCTK_Info(const char *thorn, const char *message);
#include "cctk.h" call CCTK_INFO(thorn, message) character*(*) thorn, message
Parameters
message The string to print as an info message
Discussion
The macro CCTK_INFO automatically includes the name of the originating thorn in the info message. It is recommended that the macro CCTK_INFO is used to print a message rather than calling CCTK_Info directly.
See Also
CCTK_ERROR [A158] macro to print an error message with a single string argument and stop the code
CCTK_VERROR [A610] macro to print a formatted string with a variable argument list as error message and stops
the code
CCTK_VINFO [A616] macro to print a formatted string with a variable argument list as an info message to screen
CCTK_VWARN [A626] macro to print a warning message with a variable argument list
CCTK_WARN [A634] macro to print a warning message with a single string argument and possibly stop the code
Examples
#include <cctk.h> CCTK_INFO("Output is disabled");
#include "cctk.h" integer myint real myreal character*200 message write(message, ’(A32, G12.7, A5, I8)’) & ’Your info message, including ’, myreal, ’ and ’, myint call CCTK_INFO(message)
Register one or more routines for dealing with information messages in addition to printing them to screen
Synopsis
#include <cctk.h> CCTK_InfoCallbackRegister(void *data, cctk_infofunc callback);
Parameters
data The void pointer holding extra information about the registered call back routine
callback The function pointer pointing to the call back function dealing with information messages. The definition of the function pointer is:
typedef void (*cctk_infofunc)(const char *thorn, const char *message, void *data);
The argument list is the same as those in CCTK_Info() (see the discussion of CCTK_INFO() page A308) except
an extra void pointer to hold the information about the call back routine.
Discussion
This function can be used by thorns to register their own routines to deal with information messages. The registered function pointers will be stored in a pointer chain. When CCTK_VInfo() is called, the registered routines will be called in the same order as they get registered in addition to dumping warning messages to stderr.
The function can only be called in C.
See Also
CCTK_VInfo() prints a formatted string with a variable argument list as an info message to screen
CCTK_WarnCallbackRegister Register one or more routines for dealing with warning messages in addition to
printing them to standard error
Examples
/*DumpInfo will dump information messages to a file*/ void DumpInfo(const char *thorn, const char *message, void *data) { DECLARE_CCTK_PARAMETERS FILE *fp; char *str = (char *)malloc((strlen(thorn) +strlen(message) +100)*sizeof(char)); /*info_dump_file is a string set in the parameter file*/ if((fp = fopen (info_dump_file, "a"))==0) { fprintf(stderr, "fatal error: can not open the file %s\n",info_dump_file); return; } sprintf(str, "\n[INFO]\nThorn->%s\nMsg->%s\n",thorn,message); fprintf(fp, "%s", str); free(str); fclose(fp); } ... /*data = NULL; callback = DumpInfo*/ CCTK_InfoCallbackRegister(NULL,DumpInfo);
Interpolate a list of distributed grid variables
The computation is optimized for the case of interpolating a number of grid variables at a time; in this case all the interprocessor communication can be done together, and the same interpolation coefficients can be used for all the variables. A grid variable can be either a grid function or a grid array.
Synopsis
#include "cctk.h" int status = CCTK_InterpGridArrays(const cGH *cctkGH, int N_dims, int local_interp_handle, int param_table_handle, int coord_system_handle, int N_interp_points, const int interp_coords_type_code, const void *const interp_coords[], int N_input_arrays, const CCTK_INT input_array_variable_indices[], int N_output_arrays, const CCTK_INT output_array_type_codes[], void *const output_arrays[]);
call CCTK_InterpGridArrays(status, . cctkGH, . N_dims, . local_interp_handle, param_table_handle, . coord_system_handle, . N_interp_points, . interp_coords_type_code, interp_coords, . N_input_arrays, input_array_variable_indices, . N_output_arrays, output_array_type_codes, . output_arrays) integer status CCTK_POINTER cctkGH integer local_interp_handle, param_table_handle, coord_system_handle integer N_dims, N_interp_points, N_input_arrays, N_output_arrays CCTK_POINTER interp_coords(N_dims) integer interp_coords_type_code CCTK_INT input_array_variable_indices(N_input_arrays) CCTK_INT output_array_type_codes(N_output_arrays) CCTK_POINTER output_arrays(N_output_arrays)
Result
0 success
< 0 indicates an error condition (see Errors)
Parameters
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
N_dims (\(\ge 1\)) Number of dimensions in which to interpolate. This must be \(\leq \) the dimensionality of the coordinate
system defined by coord_system_handle. The default case is that it’s \(=\); see the discussion of the
interpolation_hyperslab_handle parameter-table entry for the \(<\) case.
local_interp_handle (\(\ge 0\)) Handle to the local interpolation operator as returned by CCTK_InterpHandle.
param_table_handle (\(\ge 0\)) Handle to a key-value table containing zero or more additional parameters for the
interpolation operation. The table is allowed to be modified by the local and/or global interpolation routine(s).
coord_system_handle (\(\ge 0\)) Cactus coordinate system handle defining the mapping between (usually
floating-point) coordinates and integer grid subscripts, as returned by CCTK_CoordSystemHandle.
N_interp_points (\(\ge 0\)) The number of interpolation points requested by this processor.
interp_coords_type_code One of the CCTK_VARIABLE_* type codes, giving the data type of the
interpolation-point coordinate arrays pointed to by interp_coords[]. All interpolation-point coordinate arrays
must be of the same data type. (In practice, this data type will almost always be CCTK_REAL or one of the
CCTK_REAL* types.)
interp_coords (\(\ne \) NULL) (Pointer to) an array of N_dims pointers to 1-D arrays giving the coordinates of the
interpolation points requested by this processor. These coordinates are with respect to the coordinate system
defined by coord_system_handle.
N_input_arrays (\(\ge 0\)) The number of input variables to be interpolated. If N_input_arrays is zero then no
interpolation is done; such a call may be useful for setup, interpolator querying, etc. Note that if the
parameter table entry operand_indices is used to specify a nontrivial (e.g. one-to-many) mapping
of input variables to output arrays, only the unique set of input variables should be given here.
input_array_variable_indices (\(\ne \) NULL) (Pointer to) an array of N_input_arrays CCTK grid variable
indices (as returned by CCTK_VarIndex) specifying the input grid variables for the interpolation. For any
element with an index value of -1 in the grid variable indices array, that interpolation is skipped.
This may be useful if the main purpose of the call is e.g. to do some query or setup computation.
N_output_arrays (\(\ge 0\)) The number of output arrays to be returned from the interpolation. If N_output_arrays
is zero then no interpolation is done; such a call may be useful for setup, interpolator querying, etc. Note that
N_output_arrays may differ from N_input_arrays, e.g. if the operand_indices parameter-table entry is used
to specify a nontrivial (e.g. many-to-one) mapping of input variables to output arrays. If such a mapping is
specified, only the unique set of output arrays should be given in the output_arrays argument.
output_array_type_codes (\(\ne \) NULL) (Pointer to) an array of N_output_arrays CCTK_VARIABLE_* type codes
giving the data types of the 1-D output arrays pointed to by output_arrays[].
output_arrays (\(\ne \) NULL) (Pointer to) an array of N_output_arrays pointers to the (user-supplied) 1-D output
arrays for the interpolation. If any of the pointers in the output_arrays array is NULL, then that interpolation
is skipped. This may be useful if the main purpose of the call is e.g. to do some query or setup computation.
Discussion
This function interpolates a list of CCTK grid variables (in a multiprocessor run these are generally distributed over processors) on a list of interpolation points. The grid topology and coordinates are implicitly specified via a Cactus coordinate system. The interpolation points may be anywhere in the global Cactus grid. In a multiprocessor run they may vary from processor to processor; each processor will get whatever interpolated data it asks for. The routine CCTK_InterpGridArrays does not do the actual interpolation itself but rather takes care of whatever interprocessor communication may be necessary, and – for each processor’s local component of the domain-decomposed grid variables – calls CCTK_InterpLocalUniform to invoke an external local interpolation operator (as identified by an interpolation handle).
Additional parameters for the interpolation operation of both CCTK_InterpGridArrays and CCTK_InterpLocalUniform can be passed in via a handle to a key/value options table. All interpolation operators should check for a parameter table entry with the key suppress_warnings which – if present – indicates that the caller wants the interpolator to be silent in case of an error condition and only return an appropriate error code. One common parameter-table option, which a number of interpolation operators are likely to support, is order, a CCTK_INT specifying the order of the (presumably polynomial) interpolation (1=linear, 2=quadratic, 3=cubic, etc). As another example, a table might be used to specify that the local interpolator should take derivatives, by specifying
const CCTK_INT operand_indices[N_output_arrays]; const CCTK_INT operation_codes[N_output_arrays];
Also, the global interpolator will typically need to specify some options of its own for the local interpolator.4 These will overwrite any entries with the same keys in the param_table_handle table. Finally, the parameter table can be used to pass back arbitrary information by the local and/or global interpolation routine(s) by adding/modifying appropriate key/value pairs.
Note that CCTK_InterpGridArrays is a collective operation, so in the multiprocessor case you must call this function in parallel on each processor, passing identical arguments except for the number of interpolation points, the interpolation coordinates, and the output array pointers. You may (and typically will) specify a different set of interpolation points on each processor’s call – you may even specify an empty set on some processors. The interpolation points may be “owned” by any processors (this function takes care of all interprocessor-communication issues), though it may be more efficient to have most or all of the interpolation points “owned” by the current processor.
In the multiprocessor case, the result returned by CCTK_InterpGridArrays is guaranteed to be the same on all processors. (All current implementations simply take the minimum of the per-processor results over all processors; this gives a result which is 0 if all processors succeeded, or which is the most negative error code encountered by any processor otherwise.)
The semantics of CCTK_InterpGridArrays are mostly independent of which Cactus driver is being used, but an implementation will most likely depend on, and make use of, driver-specific internals. For that reason, CCTK_InterpGridArrays is made an overloadable function. The Cactus flesh will supply only a dummy routine for it which – if called – does nothing but print a warning message saying that it wasn’t overloaded by another thorn, and stop the code. So one will always need to compile in and activate a driver-specific thorn which provides an interpolation routine for CCTK grid variables and properly overloads CCTK_InterpGridArrays with it at startup.
Details of the operation performed, and what (if any) inputs and/or outputs are specified in the parameter table, depend on which driver-specific interpolation thorn and interpolation operator (provided by a local interpolation thorn) you use. See the documentation on individual interpolator thorns (e.g. PUGHInterp in the CactusPUGH arrangement, CarpetInterp in the Carpet arrangement, LocalInterp in the CactusBase arrangement, and/or AEILocalInterp in the AEIThorns arrangement) for details.
Note that in a multiprocessor Cactus run, it’s the user’s responsibility to choose the interprocessor ghost-zone size (driver::ghost_size) large enough so that the local interpolator never has to off-center its molecules near interprocessor boundaries. (This ensures that the interpolation results are independent of the interprocessor decomposition, at least up to floating-point roundoff errors.) If the ghost-zone size is too small, the interpolator should return the CCTK_ERROR_INTERP_GHOST_SIZE_TOO_SMALL error code.
See Also
CCTK_InterpHandle() Get the interpolator handle for a given character-string name.
CCTK_InterpLocalUniform() Interpolate a list of processor-local arrays which define a uniformly-spaced data
grid
Errors
The following list of error codes indicates specific error conditions. For the complete list of possible error return
codes you should refer to the ThornGuide’s chapter of the corresponding interpolation thorn(s) you are using. To
find the numerical values of the error codes (or more commonly, to find which error code corresponds to a given
numerical value), look in the files cctk_Interp.h, util_ErrorCodes.h, and/or util_Table.h in the
src/include/ directory in the Cactus flesh.
CCTK_ERROR_INTERP_POINT_OUTSIDE one or more of the interpolation points is out of range (in this case
additional information about the out-of-range point may be reported through the parameter table; see
the Thorn Guide for whatever thorn provides the local interpolation operator for further details)
CCTK_ERROR_INTERP_GRID_TOO_SMALL one or more of the dimensions of the input arrays is/are smaller than the
molecule size chosen by the interpolator (based on the parameter-table options, e.g. the interpolation order)
CCTK_ERROR_INTERP_GHOST_SIZE_TOO_SMALL for a multi-processor run, the size of the interprocessor
boundaries (the ghostzone size) is smaller than the molecule size chosen by the interpolator (based on the
parameter-table options, e.g. the interpolation order).
This error code is also returned if a processor’s chunk of the global grid is smaller than the actual molecule size.
UTIL_ERROR_BAD_INPUT one or more of the input arguments is invalid (e.g. NULL pointer)
UTIL_ERROR_NO_MEMORY unable to allocate memory
UTIL_ERROR_BAD_HANDLE parameter table handle is invalid
other error codes this function may also return any error codes returned by the Util_Table* routines used
to get parameters from (and/or set results in) the parameter table
Examples
Here’s a simple example to do quartic 3-D interpolation of a real and a complex grid array, at 1000 interpolation points:
#include "cctk.h" #include "util_Table.h" #define N_DIMS 3 #define N_INTERP_POINTS 1000 #define N_INPUT_ARRAYS 2 #define N_OUTPUT_ARRAYS 2 const cGH *GH; int operator_handle, coord_system_handle; /* interpolation points */ CCTK_REAL interp_x[N_INTERP_POINTS], interp_y[N_INTERP_POINTS], interp_z[N_INTERP_POINTS]; const void *interp_coords[N_DIMS]; /* input and output arrays */ CCTK_INT input_array_variable_indices[N_INPUT_ARRAYS]; static const CCTK_INT output_array_type_codes[N_OUTPUT_ARRAYS] = { CCTK_VARIABLE_REAL, CCTK_VARIABLE_COMPLEX }; void *output_arrays[N_OUTPUT_ARRAYS]; CCTK_REAL output_for_real_array [N_INTERP_POINTS]; CCTK_COMPLEX output_for_complex_array[N_INTERP_POINTS]; operator_handle = CCTK_InterpHandle("generalized polynomial interpolation"); if (operator_handle < 0) { CCTK_WARN(CCTK_WARN_ABORT, "can’t get operator handle!"); } coord_system_handle = CCTK_CoordSystemHandle("cart3d"); if (coord_system_handle < 0) { CCTK_WARN(CCTK_WARN_ABORT, "can’t get coordinate-system handle!"); } interp_coords[0] = (const void *) interp_x; interp_coords[1] = (const void *) interp_y; interp_coords[2] = (const void *) interp_z; input_array_variable_indices[0] = CCTK_VarIndex("my_thorn::real_array"); input_array_variable_indices[1] = CCTK_VarIndex("my_thorn::complex_array"); output_arrays[0] = (void *) output_for_real_array; output_arrays[1] = (void *) output_for_complex_array; if (CCTK_InterpGridArrays(GH, N_DIMS, operator_handle, Util_TableCreateFromString("order=4"), coord_system_handle, N_INTERP_POINTS, CCTK_VARIABLE_REAL, interp_coords, N_INPUT_ARRAYS, input_array_variable_indices, N_OUTPUT_ARRAYS, output_array_type_codes, output_arrays) < 0) { CCTK_WARN(CCTK_WARN_ABORT, "error return from interpolator!"); }
Return the handle for a given interpolation operator
Synopsis
int handle = CCTK_InterpHandle( const char * operator)
call CCTK_InterpHandle(handle , operator ) integer handle character*(*) operator
Parameters
handle Handle for the interpolation operator
operator Name of interpolation operator
Examples
handle = CCTK_InterpHandle("my interpolation operator");
call CCTK_InterpHandle(handle,"my interpolation operator")
Errors
negative A negative value is returned for invalid/unregistered interpolation operator names.
Interpolate a list of processor-local arrays which define a uniformly-spaced data grid
The computation is optimized for the case of interpolating a number of arrays at a time; in this case the same interpolation coefficients can be used for all the arrays.
Synopsis
#include "util_ErrorCodes.h" #include "cctk.h" int status = CCTK_InterpLocalUniform(int N_dims, int operator_handle, int param_table_handle, const CCTK_REAL coord_origin[], const CCTK_REAL coord_delta[], int N_interp_points, int interp_coords_type_code, const void *const interp_coords[], int N_input_arrays, const CCTK_INT input_array_dims[], const CCTK_INT input_array_type_codes[], const void *const input_arrays[], int N_output_arrays, const CCTK_INT output_array_type_codes[], void *const output_arrays[]);
call CCTK_InterpLocalUniform(status, . N_dims, . operator_handle, . param_table_handle, . coord_origin, . coord_delta, . N_interp_points, . interp_coords_type_code, . interp_coords, . N_input_arrays, . input_array_dims, . input_array_type_codes, . input_arrays, . N_output_arrays, . output_array_type_codes, . output_arrays) integer status integer operator_handle, param_table_handle integer N_dims, N_interp_points, N_input_arrays, N_output_arrays CCTK_REAL coord_origin(N_dims), coord_delta(N_dims) integer interp_coords_type_code CCTK_POINTER interp_coords(N_dims) CCTK_INT input_array_dims(N_dims), input_array_type_codes(N_input_arrays) CCTK_POINTER input_arrays(N_input_arrays) CCTK_INT output_array_type_codes(N_output_arrays) CCTK_POINTER output_arrays(N_output_arrays)
Result
0 success
Parameters
N_dims (\(\ge 1\)) Number of dimensions in which to interpolate. Note that this may be less than the number of
dimensions of the input arrays if the storage is set up appropriately. For example, we might want to interpolate
along 1-D lines or in 2-D planes of a 3-D input array; here N_dims would be 1 or 2 respectively. For details,
see the section on “Non-Contiguous Input Arrays” in the Thorn Guide for thorn AEILocalInterp.
operator_handle (\(\ge 0\)) Handle to the interpolation operator as returned by CCTK_InterpHandle.
param_table_handle (\(\ge 0\)) Handle to a key-value table containing additional parameters for the
interpolator.
One common parameter-table option, which a number of interpolation operators are likely to support, is order, a CCTK_INT specifying the order of the (presumably polynomial) interpolation (1=linear, 2=quadratic, 3=cubic, etc).
See the Thorn Guide for the AEILocalInterp thorn for other parameters.
coord_origin (\(\ne \) NULL) (Pointer to) an array giving the coordinates of the data point with integer array
subscripts 0, 0, …, 0, or more generally (if the actual array bounds don’t include the all-zeros-subscript point) the
coordinates which this data point would have if it existed. See the “Discussion” section below for more on how
coord_origin[] is actually used.
coord_delta (\(\ne \) NULL) (Pointer to) an array giving the coordinate spacing of the data arrays. See the
“Discussion” section below for more on how coord_delta[] is actually used.
N_interp_points (\(\ge 0\)) The number of points at which interpolation is to be done.
interp_coords_type_code One of the CCTK_VARIABLE_* type codes, giving the data type of the 1-D
interpolation-point-coordinate arrays pointed to by interp_coords[]. (In practice, this data type will almost
always be CCTK_REAL or one of the CCTK_REAL* types.)
interp_coords (\(\ne \) NULL) (Pointer to) an array of N_dims pointers to 1-D arrays giving the coordinates of the
interpolation points. These coordinates are with respect to the coordinate system defined by coord_origin[]
and coord_delta[].
N_input_arrays (\(\ge 0\)) The number of input arrays to be interpolated. Note that if the parameter table entry
operand_indices is used to specify a 1-to-many mapping of input arrays to output arrays, only the unique set
of input arrays should be given here.
input_array_dims (\(\ne \) NULL) (Pointer to) an array of N_dims integers giving the dimensions of the
N_dims-D input arrays. By default all the input arrays are taken to have these dimensions, with [0] the
most contiguous axis and [N_dims-1] the least contiguous axis, and array subscripts in the range
0 <= subscript < input_array_dims[axis]. See the discussion of the input_array_strides optional
parameter (passed in the parameter table) for details of how this can be overridden.
input_array_type_codes (\(\ne \) NULL) (Pointer to) an array of N_input_arrays CCTK_VARIABLE_*
type codes giving the data types of the N_dims-D input arrays pointed to by input_arrays[].
input_arrays (\(\ne \) NULL) (Pointer to) an array of N_input_arrays pointers to the N_dims-D input arrays
for the interpolation. If any input_arrays[in] pointer is NULL, that interpolation is skipped.
N_output_arrays (\(\ge 0\)) The number of output arrays to be returned from the interpolation.
output_array_type_codes (\(\ne \) NULL) (Pointer to) an array of N_output_arrays CCTK_VARIABLE_* type codes
giving the data types of the 1-D output arrays pointed to by output_arrays[].
output_arrays (\(\ne \) NULL) (Pointer to) an array of N_output_arrays pointers to the (user-supplied) 1-D output
arrays for the interpolation. If any output_arrays[out] pointer is NULL, that interpolation is skipped.
Discussion
CCTK_InterpLocalUniform is a generic API for interpolating processor-local arrays when the data points’ \(xyz\) coordinates are linear functions of the integer array subscripts ijk (we’re describing this for 3-D, but the generalization to other numbers of dimensions should be obvious). The coord_origin[] and coord_delta[] arguments specify these linear functions:
\(x = \code {coord\_origin[0] + i*coord\_delta[0]}\)
\(y = \code {coord\_origin[1] + j*coord\_delta[1]}\)
\(z = \code {coord\_origin[2] + k*coord\_delta[2]}\)
The \((x,y,z)\) coordinates are used for the interpolation (i.e. the interpolator may internally use polynomials in these coordinates); interp_coords[] specifies coordinates in this same coordinate system.
Details of the operation performed, and what (if any) inputs and/or outputs are specified in the parameter table, depend on which interpolation operator you use. See the Thorn Guide for the AEILocalInterp thorn for further discussion.
See Also
CCTK_InterpHandle() Get the interpolator handle for a given character-string name.
CCTK_InterpGridArrays() Interpolate a list of Cactus grid arrays
CCTK_InterpRegisterOpLocalUniform() Register a CCTK_InterpLocalUniform interpolation operator
CCTK_InterpLocalNonUniform() Interpolate a list of processor-local arrays, with non-uniformly spaced data
points.
Errors
To find the numerical values of the error codes (or more commonly, to find which error code corresponds to a
given numerical value), look in the files cctk_Interp.h, util_ErrorCodes.h, and/or util_Table.h in the
src/include/ directory in the Cactus flesh.
CCTK_ERROR_INTERP_POINT_OUTSIDE one or more of the interpolation points is out of range (in this case
additional information about the out-of-range point may be reported through the parameter table; see the Thorn
Guide for the AEILocalInterp thorn for further details)
CCTK_ERROR_INTERP_GRID_TOO_SMALL one or more of the dimensions of the input arrays is/are smaller than the
molecule size chosen by the interpolator (based on the parameter-table options, e.g. the interpolation order)
UTIL_ERROR_BAD_INPUT one or more of the inputs is invalid (e.g. NULL pointer)
UTIL_ERROR_NO_MEMORY unable to allocate memory
UTIL_ERROR_BAD_HANDLE parameter table handle is invalid
other error codes this function may also return any error codes returned by the Util_Table* routines used
to get parameters from (and/or set results in) the parameter table
Examples
Here’s a simple example of interpolating a CCTK_REAL and a CCTK_COMPLEX \(10 \times 20\) 2-D array, at 5 interpolation points, using cubic interpolation.
Note that since C allows arrays to be initialized only if the initializer values are compile-time constants, we have to declare the interp_coords[], input_arrays[], and output_arrays[] arrays as non-const, and set their values with ordinary (run-time) assignment statements. In C++, there’s no restriction on initializer values, so we could declare the arrays const and initialize them as part of their declarations.
#define N_DIMS 2 #define N_INTERP_POINTS 5 #define N_INPUT_ARRAYS 2 #define N_OUTPUT_ARRAYS 2 /* (x,y) coordinates of data grid points */ #define X_ORIGIN ... #define X_DELTA ... #define Y_ORIGIN ... #define Y_DELTA ... const CCTK_REAL origin[N_DIMS] = { X_ORIGIN, Y_ORIGIN }; const CCTK_REAL delta [N_DIMS] = { X_DELTA, Y_DELTA }; /* (x,y) coordinates of interpolation points */ const CCTK_REAL interp_x[N_INTERP_POINTS]; const CCTK_REAL interp_y[N_INTERP_POINTS]; const void *interp_coords[N_DIMS]; /* see note above */ /* input arrays */ /* ... note Cactus uses Fortran storage ordering, i.e.\ X is contiguous */ #define NX 10 #define NY 20 const CCTK_REAL input_real [NY][NX]; const CCTK_COMPLEX input_complex[NY][NX]; const CCTK_INT input_array_dims[N_DIMS] = { NX, NY }; const CCTK_INT input_array_type_codes[N_INPUT_ARRAYS] = { CCTK_VARIABLE_REAL, CCTK_VARIABLE_COMPLEX }; const void *input_arrays[N_INPUT_ARRAYS]; /* see note above */ /* output arrays */ CCTK_REAL output_real [N_INTERP_POINTS]; CCTK_COMPLEX output_complex[N_INTERP_POINTS]; const CCTK_INT output_array_type_codes[N_OUTPUT_ARRAYS] = { CCTK_VARIABLE_REAL, CCTK_VARIABLE_COMPLEX }; void *const output_arrays[N_OUTPUT_ARRAYS]; /* see note above */ int operator_handle, param_table_handle; operator_handle = CCTK_InterpHandle("my interpolation operator"); if (operator_handle < 0) CCTK_WARN(CCTK_WARN_ABORT, "can’t get interpolation handle!"); param_table_handle = Util_TableCreateFromString("order=3"); if (param_table_handle < 0) CCTK_WARN(CCTK_WARN_ABORT, "can’t create parameter table!"); /* initialize the rest of the parameter arrays */ interp_coords[0] = (const void *) interp_x; interp_coords[1] = (const void *) interp_y; input_arrays[0] = (const void *) input_real; input_arrays[1] = (const void *) input_complex; output_arrays[0] = (void *) output_real; output_arrays[1] = (void *) output_complex; /* do the actual interpolation, and check for error returns */ if (CCTK_InterpLocalUniform(N_DIMS, operator_handle, param_table_handle, origin, delta, N_INTERP_POINTS, CCTK_VARIABLE_REAL, interp_coords, N_INPUT_ARRAYS, input_array_dims, input_array_type_codes, input_arrays, N_OUTPUT_ARRAYS, output_array_type_codes, output_arrays) < 0) CCTK_WARN(CCTK_WARN_ABORT, "error return from interpolator!");
Register a CCTK_InterpLocalUniform interpolation operator.
Synopsis
#include "cctk.h" int CCTK_InterpRegisterOpLocalUniform(cInterpOpLocalUniform operator_ptr, const char *operator_name, const char *thorn_name);
Result
handle (\(\ge 0\)) A cactus handle to refer to all interpolation operators registered under this operator name.
Parameters
operator_ptr (\(\ne \) NULL) Pointer to the CCTK_InterpLocalUniform interpolation operator. This argument must
be a C function pointer of the appropriate type; the typedef can be found in src/include/cctk_Interp.h in
the Cactus source code.
operator_name (\(\ne \) NULL) (Pointer to) a (C-style null-terminated) character string giving the name under which
to register the operator.
thorn_name (\(\ne \) NULL) (Pointer to) a (C-style null-terminated) character string giving the name of the thorn
which provides the interpolation operator.
Discussion
Only C functions (or other routines with C-compatible calling sequences) can be registered as interpolation operators.
See Also
CCTK_InterpHandle() Get the interpolator handle for a given character-string name.
CCTK_InterpLocalUniform() Interpolate a list of processor-local arrays, with uniformly spaced data points.
Errors
-1 NULL pointer was passed as interpolation operator routine
-2 interpolation handle could not be allocated
-3 Interpolation operator with this name already exists
Examples
/* prototype for function we want to register */ int AEILocalInterp_InterpLocalUniform(int N_dims, int param_table_handle, /***** coordinate system *****/ const CCTK_REAL coord_origin[], const CCTK_REAL coord_delta[], /***** interpolation points *****/ int N_interp_points, int interp_coords_type_code, const void *const interp_coords[], /***** input arrays *****/ int N_input_arrays, const CCTK_INT input_array_dims[], const CCTK_INT input_array_type_codes[], const void *const input_arrays[], /***** output arrays *****/ int N_output_arrays, const CCTK_INT output_array_type_codes[], void *const output_arrays[]); /* register it! */ CCTK_InterpRegisterOpLocalUniform(AEILocalInterp_InterpLocalUniform, "generalized polynomial interpolation", CCTK_THORNSTRING);
Reports whether an aliased function has been provided
Synopsis
int istat = CCTK_IsFunctionAliased( const char * functionname)
call CCTK_IsFunctionAliased(istat , functionname ) integer istat character*(*) functionname
Parameters
istat the return status
functionname the name of the function to check
Discussion
This function returns a non-zero value if the function given by functionname is provided by any active thorn, and zero otherwise.
Reports whether an implementation was activated in a parameter file
Synopsis
int istat = CCTK_IsImplementationActive( const char * implementationname)
CCTK_IsImplementationActive( istat, implementationname ) integer istat character*(*) implementationname
Parameters
istat the return status
implementationname the name of the implementation to check
Discussion
This function returns a non-zero value if the implementation given by implementationname was activated in a parameter file, and zero otherwise. See also CCTK_ActivatingThorn [A32], CCTK_CompiledImplementation [A79], CCTK_CompiledThorn [A81], CCTK_ImplementationRequires [A302], CCTK_ImplementationThorn [A304], CCTK_ImpThornList [A306], CCTK_IsImplementationCompiled [A348], CCTK_IsThornActive [A350], CCTK_NumCompiledImplementations [A378], CCTK_NumCompiledThorns [A380], CCTK_ThornImplementation [A554].
Reports whether an implementation was compiled into the configuration
Synopsis
int istat = CCTK_IsImplementationCompiled( const char * implementationname)
istat = CCTK_IsImplementationCompiled( implementationname ) integer istat character*(*) implementationname
Parameters
istat the return status
implementationname the name of the implementation to check
Discussion
This function returns a non-zero value if the implementation given by implementationname was compiled into the configuration, and zero otherwise. See also CCTK_ActivatingThorn [A32], CCTK_CompiledImplementation [A79], CCTK_CompiledThorn [A81], CCTK_ImplementationRequires [A302], CCTK_ImplementationThorn [A304], CCTK_ImpThornList [A306], CCTK_IsImplementationActive [A346], CCTK_IsThornActive [A350], CCTK_IsThornCompiled [A353], CCTK_NumCompiledImplementations [A378], CCTK_NumCompiledThorns [A380], CCTK_ThornImplementation [A554].
Reports whether a thorn was activated in a parameter file
Synopsis
#include "cctk.h" int status = CCTK_IsThornActive(const char* thorn_name);
#include "cctk.h" integer status character *(*) thorn_name status = CCTK_IsThornActive(thorn_name)
Result
status This function returns a non-zero value if thorn thorn_name was activated in a parameter file, and zero
otherwise.
Parameters
thorn_name The character-string name of the thorn, for example ”SymBase”.
Discussion
This function lets you find out at run-time whether or not a given thorn is active in the current Cactus run.
Reports whether a thorn was activated in a parameter file
Synopsis
int istat = CCTK_IsThornCompiled( const char * thornname)
istat = CCTK_IsThornCompiled( thornname ) integer istat character*(*) thornname
Parameters
istat the return status
thorname the name of the thorn to check
Discussion
This function returns a non-zero value if the implementation given by thornname was compiled into the configuration, and zero otherwise.
Returns the name of a registered reduction operator
Synopsis
#include "cctk.h" const char *name = CCTK_LocalArrayReduceOperator(int handle);
Result
name Returns the name of a registered local reduction operator of handle
handle or NULL if the handle is invalid
Parameters
handle The handle of a registered local reduction operator
Discussion
This function returns the name of a registered reduction operator given its handle. NULL is returned if the handle is invalid
See Also
CCTK_ReduceLocalArrays() Reduces a list of local arrays (new local array reduction API)
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_RegisterLocalArrayReductionOperator() Registers a function as a reduction operator of a certain
name
CCTK_LocalArrayReduceOperatorImplementation() Provide the implementation which provides an local
array reduction operator
CCTK_NumLocalArrayReduceOperators() The number of local reduction operators registered
Provide the implementation which provides an local array reduction operator
Synopsis
#include "cctk.h" const char *implementation = CCTK_LocalArrayReduceOperatorImplementation( int handle);
Result
implementation The name of the implementation implementing the local reduction operator of handle
handle
Parameters
handle The handle of a registered local reduction operator
Discussion
This function returns the implementation name of a registered reduction operator given its handle or NULL if the handle is invalid
See Also
CCTK_ReduceLocalArrays() Reduces a list of local arrays (new local array reduction API)
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_RegisterLocalArrayReductionOperator() Registers a function as a reduction operator of a certain
name
CCTK_LocalArrayReduceOperator() Returns the name of a registered reduction operator
CCTK_NumLocalArrayReduceOperators() The number of local reduction operators registered
Returns the handle of a given local array reduction operator
Synopsis
#include "cctk.h" int handle = CCTK_LocalArrayReductionHandle(const char *operator);
Result
handle The handle corresponding to the local reduction operator
Parameters
operator The reduction operation to be performed. If no matching registered operator is found, a warning is
issued and an error returned.
Discussion
This function returns the handle of the local array reduction operator. The local reduction handle is also used in the grid array reduction.
See Also
CCTK_ReduceLocalArrays() Reduces a list of local arrays (new local array reduction API)
CCTK_RegisterLocalArrayReductionOperator() Registers a function as a reduction operator of a certain
name
CCTK_LocalArrayReduceOperatorImplementation() Provide the implementation which provides an local
array reduction operator
CCTK_LocalArrayReduceOperator() Returns the name of a registered reduction operator
CCTK_NumLocalArrayReduceOperators() The number of local reduction operators registered
Returns the maximum number of time levels ever active for a group.
Synopsis
#include "cctk.h" int timelevels = CCTK_MaxActiveTimeLevels(const cGH *cctkGH, const char *groupname); int timelevels = CCTK_MaxActiveTimeLevelsGI(const cGH *cctkGH, int groupindex); int timelevels = CCTK_MaxActiveTimeLevelsGN(const cGH *cctkGH, const char *groupname); int timelevels = CCTK_MaxActiveTimeLevelsVI(const cGH *cctkGH, int varindex); int timelevels = CCTK_MaxActiveTimeLevelsVN(const cGH *cctkGH, const char *varname);
#include "cctk.h" subroutine CCTK_MaxActiveTimeLevels(timelevels, cctkGH, groupname) integer timelevels CCTK_POINTER cctkGH character*(*) groupname end subroutine CCTK_MaxActiveTimeLevels subroutine CCTK_MaxActiveTimeLevelsGI(timelevels, cctkGH, groupindex) integer timelevels CCTK_POINTER cctkGH integer groupindex end subroutine CCTK_MaxActiveTimeLevelsGI subroutine CCTK_MaxActiveTimeLevelsGN(timelevels, cctkGH, groupname) integer timelevels CCTK_POINTER cctkGH character*(*) groupname end subroutine CCTK_MaxActiveTimeLevelsGN subroutine CCTK_MaxActiveTimeLevelsVI(timelevels, cctkGH, varindex) integer timelevels CCTK_POINTER cctkGH integer varindex end subroutine CCTK_MaxActiveTimeLevelsVI subroutine CCTK_MaxActiveTimeLevelsVN(timelevels, cctkGH, varname) integer timelevels CCTK_POINTER cctkGH character*(*) varname end subroutine CCTK_MaxActiveTimeLevelsVN
Result
timelevels The largest number of timelevels that were ever active for the group or the result of
CCTK_DeclaredTimeLevels, whichever is larger.
Parameters
GH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
groupname Name of the group.
groupindex Index of the group.
varname Name of a variable in the group.
varindex Index of a variable in the group.
Discussion
This function returns the number of timelevels for which storage has ever been request either through an actual storage request or via the TIMELEVELS attribute in interface.ccl. This is always at least as large as CCTK_DeclaredTimeLevels.
See Also
CCTK_DeclaredTimeLevels [A121] Return the maximum number of timelevels in interface.ccl.
CCTK_GroupStorageDecrease [A278] Base function, overloaded by the driver, which decreases the number of
active timelevels, and also returns the number of active timelevels.
CCTK_GroupStorageIncrease [A280] Base function, overloaded by the driver, which increases the number of
active timelevels, and also returns the number of active timelevels.
Errors
timelevels \(<\) 0 Illegal arguments given.
Get the maximum dimension of any grid variable
Synopsis
int dim = CCTK_MaxDim()
call CCTK_MaxDim(dim ) integer dim
Parameters
dim The maximum dimension
Discussion
Note that the maximum dimension will depend only on the active thorn list, and not the compiled thorn list.
Examples
dim = CCTK_MaxDim()
call CCTK_MaxDim(dim)
Get the maximum dimension of all grid functions
Synopsis
int dim = CCTK_MaxGFDim()
call CCTK_MaxGFDim(dim ) integer dim
Parameters
dim The maximum dimension of all grid functions
Discussion
Note that the maximum dimension will depend only on the active thorn list, and not the compiled thorn list.
Examples
dim = CCTK_MaxGFDim();
call CCTK_MaxGFDim(dim)}
Decprecated. Use CCTK_DeclaredTimeLevels instead.
Synopsis
Returns the number of the local processor for a parallel run
Synopsis
int myproc = CCTK_MyProc( const cGH * cctkGH)
Parameters
cctkGH pointer to CCTK grid hierarchy
Discussion
For a single processor run this call will return zero. For multiprocessor runs, this call will return 0 \(\leq \) myproc \(<\) CCTK_nProcs(cctkGH).
Calling CCTK_MyProc(NULL) is safe (it will not crash). Current drivers (PUGH, Carpet, CarpetX) handle this case correctly (i.e. CCTK_MyProc(NULL) returns a correct result), but only a “best effort” is guaranteed for future drivers (or future revisions of current drivers).
Returns the number of processors being used for a parallel run
Synopsis
int nprocs = CCTK_nProcs( const cGH * cctkGH)
nprocs = CCTK_nProcs( cctkGH ) integer nprocs CCTK_POINTER cctkGH
Parameters
cctkGH pointer to CCTK grid hierarchy
Discussion
For a single processor run this call will return one.
Calling CCTK_nProcs(NULL) is safe (it will not crash). Current drivers (PUGH, Carpet, CarpetX) handle this case correctly (i.e. CCTK_nProcs(NULL) returns a correct result), but only a “best effort” is guaranteed for future drivers (or future revisions of current drivers).
Returns a C-style NULL pointer value.
Synopsis
#include "cctk.h" CCTK_POINTER pointer_var pointer_var = CCTK_NullPointer()
Result
pointer_var a CCTK_POINTER type variable which is initialized with a C-style NULL pointer
Discussion
Fortran doesn’t know the concept of pointers so problems arise when a C function is to be called which expects a pointer as one (or more) of it(s) argument(s).
In order to pass a NULL pointer from Fortran to C, a local CCTK_POINTER variable should be used which has been initialized before with CCTK_NullPointer.
Note that there is only a Fortran wrapper available for CCTK_NullPointer.
See Also
CCTK_PointerTo() Returns the address of a variable passed in by reference from a Fortran routine.
Examples
#include "cctk.h" integer ierror, table_handle CCTK_POINTER pointer_var pointer_var = CCTK_NullPointer() call Util_TableCreate(table_handle, 0) call Util_TableSetPointer(ierror, table_handle, pointer_var, "NULL pointer")
Return the number of implementations compiled in.
Synopsis
#include "cctk.h" int numimpls = CCTK_NumCompiledImplementations();
Result
numimpls Number of implementations compiled in.
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
Return the number of thorns compiled in.
Synopsis
#include "cctk.h" int numthorns = CCTK_NumCompiledThornss();
Result
numthorns Number of thorns compiled in.
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_ThornImplementation [A554] Returns the implementation provided by the thorn
The number of grid array reduction operators registered
Synopsis
#include "cctk.h" int num_ga_reduc = CCTK_NumGridArrayReductionOperators();
Result
num_ga_reduc The number of registered grid array reduction operators (currently either 1 or 0)
Discussion
This function returns the number of grid array reduction operators. Since we only allow one grid array reduction operator currently, this function can be used to check if a grid array reduction operator has been registered or not.
See Also
CCTK_ReduceGridArrays() Performs reduction on a list of distributed grid arrays
CCTK_RegisterGridArrayReductionOperator() Registers a function as a grid array reduction operator of a
certain name
CCTK_GridArrayReductionOperator() The name of the grid reduction operator, or NULL if none is registered
Get the number of groups of variables compiled in the code
Synopsis
int number = CCTK_NumGroups()
call CCTK_NumGroups(number ) integer number
Parameters
number The number of groups compiled from the thorns interface.ccl files
Examples
number = CCTK_NumGroups();
call CCTK_NumGroups(number);
Find the total number of I/O methods registered with the flesh
Synopsis
int num_methods = CCTK_NumIOMethods (void);
call CCTK_NumIOMethods (num_methods) integer num_methods
Parameters
num_methods number of registered IO methods
Discussion
Returns the total number of IO methods registered with the flesh.
The number of local reduction operators registered
Synopsis
#include "cctk.h" int num_ga_reduc = CCTK_NumLocalArrayReduceOperators();
Result
num_ga_reduc The number of registered local array operators
Discussion
This function returns the total number of registered local array reduction operators
See Also
CCTK_ReduceLocalArrays() Reduces a list of local arrays (new local array reduction API)
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_RegisterLocalArrayReductionOperator() Registers a function as a reduction operator of a certain
name
CCTK_LocalArrayReduceOperatorImplementation() Provide the implementation which provides an local
array reduction operator
CCTK_LocalArrayReduceOperator() Returns the name of a registered reduction operator
The number of global array reduction operators registered, either 1 or 0.
Synopsis
#include "cctk.h" int num_reduc = CCTK_NumReductionArraysGloballyOperators();
Result
num_reduc The number of registered global array operators
Discussion
This function returns the total number of registered global array reduction operators, it is either 1 or 0 as we do not allow multiple array reductions.
See Also
CCTK_ReduceArraysGlobally() Reduces a list of arrays globally
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_RegisterReduceArraysGloballyOperator() Registers a function as a reduction operator of a certain
name
Given a cTimerData structure, returns its number of clocks.
Synopsis
int err = CCTK_NumTimerClocks(info)
Parameters
const cTimerData * info The timer information structure whose clocks are to be counted.
Get the number of grid variables compiled in the code
Synopsis
int number = CCTK_NumVars()
call CCTK_NumVars(number ) integer number
Parameters
number The number of grid variables compiled from the thorn’s interface.ccl files
Examples
number = CCTK_NumVars();
call CCTK_NumVars(number)
Provides the number of variables in a group from the group name
Synopsis
int num = CCTK_NumVarsInGroup( const char * name)
call CCTK_NumVarsInGroup(num , name ) integer num character*(*) name
Parameters
num The number of variables in the group
group The full group name
Discussion
The group name should be given in the form <implementation>::<group>
Examples
numvars = CCTK_NumVarsInGroup("evolve::scalars")
call CCTK_NUMVARSINGROUP(numvars,"evolve::scalars")
Provides the number of variables in a group from the group index
Synopsis
int num = CCTK_NumVarsInGroupI( int index)
call CCTK_NumVarsInGroupI(num , index ) integer num integer index
Parameters
num The number of variables in the group
group The group index
Discussion
Examples
index = CCTK_GroupIndex("evolve::scalars")} firstvar = CCTK_NumVarsInGroupI(index)
call CCTK_NUMVARSINGROUPI(firstvar,3)
Output all variables living on the GH looping over all registered IO methods.
Synopsis
int istat = CCTK_OutputGH (const cGH *cctkGH);
call CCTK_OutputGH (istat, cctkGH) integer istat CCTK_POINTER cctkGH
Parameters
istat total number of variables for which output was done by all IO methods
cctkGH pointer to CCTK grid hierarchy
Discussion
The IO methods decide themselfes whether it is time to do output now or not.
Errors
0 it wasn’t time to output anything yet by any IO method
-1 if no IO methods were registered
Output a single variable by all I/O methods
Synopsis
int istat = CCTK_OutputVar (const cGH *cctkGH, const char *variable);
call CCTK_OutputVar (istat, cctkGH, variable) integer istat CCTK_POINTER cctkGH character*(*) variable
Parameters
istat return status
cctkGH pointer to CCTK grid hierarchy
variable full name of variable to output, with an optional options string in curly braces
Discussion
The output should take place if at all possible. If the appropriate file exists the data is appended, otherwise a new file is created.
Errors
0 for success
negative for some error condition (e.g. IO method is not registered)
Output a single variable as an alias by all I/O methods
Synopsis
int istat = CCTK_OutputVarAs (const cGH *cctkGH, const char *variable, const char *alias);
call CCTK_OutputVarAs (istat, cctkGH, variable, alias) integer istat CCTK_POINTER cctkGH character*(*) variable character*(*) alias
Parameters
istat return status
cctkGH pointer to CCTK grid hierarchy
variable full name of variable to output, with an optional options string in curly braces
alias alias name to base the output filename on
Discussion
The output should take place if at all possible. If the appropriate file exists the data is appended, otherwise a new file is created. Uses alias as the name of the variable for the purpose of constructing a filename.
Errors
positive the number of IO methods which did output of variable
0 for success
negative if no IO methods were registered
Synopsis
int istat = CCTK_OutputVarAsByMethod (const cGH *cctkGH, const char *variable, const char *method, const char *alias);
call CCTK_OutputVarAsByMethod (istat, cctkGH, variable, method, alias) integer istat CCTK_POINTER cctkGH character*(*) variable character*(*) method character*(*) alias
Parameters
istat return status
cctkGH pointer to CCTK grid hierarchy
variable full name of variable to output, with an optional options string in curly braces
method method to use for output
alias alias name to base the output filename on
Discussion
Output a variable variable using the method method if it is registered. Uses alias as the name of the variable for the purpose of constructing a filename. The output should take place if at all possible. If the appropriate file exists the data is appended, otherwise a new file is created.
Errors
0 for success
negative indicating some error (e.g. IO method is not registered)
Synopsis
int istat = CCTK_OutputVarByMethod (const cGH *cctkGH, const char *variable, const char *method);
call CCTK_OutputVarByMethod (istat, cctkGH, variable, method) integer istat CCTK_POINTER cctkGH character*(*) variable character*(*) method
Parameters
istat return status
cctkGH pointer to CCTK grid hierarchy
variable full name of variable to output, with an optional options string in curly braces
method method to use for output
Discussion
Output a variable variable using the IO method method if it is registered. The output should take place if at all possible. if the appropriate file exists the data is appended, otherwise a new file is created.
Errors
0 for success
negative indicating some error (e.g. IO method is not registered)
Initialize the parallel subsystem
Synopsis
int istat = CCTK_ParallelInit( cGH * cctkGH)
Parameters
cctkGH pointer to CCTK grid hierarchy
Discussion
Initializes the parallel subsystem.
Get parameter properties for given parameter/thorn pair.
Synopsis
#include "cctk.h" const cParamData *paramdata = CCTK_ParameterData (const char *name, const char *thorn);
Result
paramdata Pointer to parameter data structure
Parameters
name Parameter name
thorn Thorn name (for private parameters) or implementation name (for restricted parameters)
Discussion
The thorn or implementation name must be the name of the place where the parameter is originally defined. It is not possible to pass the thorn or implementation name of a thorn that merely declares the parameter as used.
See Also
CCTK_ParameterGet [A428] Get the data pointer to and type of a parameter’s value
CCTK_ParameterLevel [A430] Return the parameter checking level
CCTK_ParameterQueryTimesSet [A431] Return number of times a parameter has been set
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterValString [A445] Get the string representation of a parameter’s value
CCTK_ParameterWalk [A448] Walk through list of parameters
Errors
NULL No parameter with that name was found.
Returns the parameter filename.
Synopsis
#include "cctk.h" int retval = CCTK_ParameterFilename(int len, char *filename)
call CCTK_ParameterFilename(retval, len, filename) INTEGER :: retval, len CHARACTER(*) :: filename
Result
retval The length of the returned string.
Parameters
len The length of the incoming string
filename String to contain the filename
Discussion
Returns the name of the parameter file given to Cactus, up to len characters in length and guaranteed to be NUL terminated.
See Also
CCTK_CommandLine [A74] Gets the command line arguments.
Get the data pointer to and type of a parameter’s value.
Synopsis
#include "cctk.h" const void *paramval = CCTK_ParameterGet (const char *name, const char *thorn, int *type);
Result
paramval Pointer to the parameter value
Parameters
name Parameter name
thorn Thorn name (for private parameters) or implementation name (for restricted parameters)
type If not NULL, a pointer to an integer which will hold the type of the parameter
Discussion
The thorn or implementation name must be the name of the place where the parameter is originally defined. It is not possible to pass the thorn or implementation name of a thorn that merely declares the parameter as used.
See Also
CCTK_ParameterData [A423] Get parameter properties for given parameter/thorn pair
CCTK_ParameterLevel [A430] Return the parameter checking level
CCTK_ParameterQueryTimesSet [A431] Return number of times a parameter has been set
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterValString [A445] Get the string representation of a parameter’s value
CCTK_ParameterWalk [A448] Walk through list of parameters
Errors
NULL No parameter with that name was found.
Return the parameter checking level.
Synopsis
#include "cctk.h" int level = CCTK_ParameterLevel (void);
Result
level Parameter checking level now being used.
See Also
CCTK_ParameterData [A423] Get parameter properties for given parameter/thorn pair
CCTK_ParameterGet [A428] Get the data pointer to and type of a parameter’s value
CCTK_ParameterQueryTimesSet [A431] Return number of times a parameter has been set
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterValString [A445] Get the string representation of a parameter’s value
CCTK_ParameterWalk [A448] Walk through list of parameters
Return number of times a parameter has been set.
Synopsis
#include "cctk.h" int nset = CCTK_ParameterQueryTimesSet (const char *name, const char *thorn);
Result
nset Number of times the parameter has been set.
Parameters
name Parameter name
thorn Thorn name (for private parameters) or implementation name (for restricted parameters)
Discussion
The number of times that a parameter has been set is 0 if the parameter was not set in a parameter file. The number increases when CCTK_ParameterSet is called.
The thorn or implementation name must be the name of the place where the parameter is originally defined. It is not possible to pass the thorn or implementation name of a thorn that merely declares the parameter as used.
See Also
CCTK_ParameterData [A423] Get parameter properties for given parameter/thorn pair
CCTK_ParameterGet [A428] Get the data pointer to and type of a parameter’s value
CCTK_ParameterLevel [A430] Return the parameter checking level
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterValString [A445] Get the string representation of a parameter’s value
CCTK_ParameterWalk [A448] Walk through list of parameters
Errors
\(-1\) No parameter with that name exists.
Sets the value of a parameter.
Synopsis
#include "cctk.h" int ierr = CCTK_ParameterSet (const char *name, const char *thorn, const char *value);
call CCTK_ParameterSet (ierr, name, thorn, value) CCTK_INT ierr character*(*) name character*(*) thorn character*(*) value
Result
ierr Error code
Parameters
name Parameter name
thorn Thorn name (for private parameters) or implementation name (for restricted parameters)
value The new (stringified) value for the parameter parameter
Discussion
The thorn or implementation name must be the name of the place where the parameter is originally defined. It is not possible to pass the thorn or implementation name of a thorn that merely declares the parameter as used.
While setting a new parameter value is immediately reflected in Cactus’ database, the value of the parameter is not changed immediately in the routine that sets the new value: It is updated only the next time a routine is entered (or rather, when the DECLARE_CCTK_PARAMETERS is encountered the next time). It is therefore advisable to set the new parameter value in a routine scheduled at a time earlier to when the new value is required.
See Also
CCTK_ParameterData [A423] Get parameter properties for given parameter/thorn pair
CCTK_ParameterLevel [A430] Return the parameter checking level
CCTK_ParameterQueryTimesSet [A431] Return number of times a parameter has been set
CCTK_ParameterSetNotifyRegister [A437] Registers a parameter set operation notify callback
CCTK_ParameterSetNotifyUnregister [A441] Unregisters a parameter set operation notify callback
CCTK_ParameterValString [A445] Get the string representation of a parameter’s value
CCTK_ParameterWalk [A448] Walk through list of parameters
Errors
ierr
\(0\)
success
\(-1\)
parameter is out of range
\(-2\)
parameter was not found
\(-3\)
trying to steer a non-steerable parameter
\(-6\)
not a valid integer or float
\(-7\)
tried to set an accumulator parameter directly
\(-8\)
tried to set an accumulator parameter directly
\(-9\)
final value of accumulator out of range
Registers a parameter set operation notify callback
Synopsis
#include "cctk.h" int handle = CCTK_ParameterSetNotifyRegister (cParameterSetNotifyCallbackFn callback, void *data, const char *name, const char *thorn_regex, const char *param_regex
call CCTK_ParameterSetNotifyRegister (handle, callback, data, . name, thorn_regex, param_regex) integer handle external callback integer callback CCTK_POINTER data character*(*) name character*(*) thorn_regex character*(*) param_regex
Result
0 success
-1 another callback has already been registered under the given name
-2 memory allocation error
-3 invalid regular expression given for thorn_regex / param_regex
Parameters
callback Function pointer of the notify callback to be registered
data optional user-defined data pointer to associate with the notify callback
name Unique name under which the notify callback is to be registered
thorn_regex Optional regular expression string to match a thorn name in a full parameter name
param_regex Optional regular expression string to match a parameter name in a full parameter name
Discussion
Declaring a parameter steerable at runtime in its param.ccl definition requires a thorn writer to add extra logic to the code which checks if a parameter value has changed, either periodically in a scheduled function, or by direct notification from the flesh’s parameter set routine CCTK_ParameterSet().
With CCTK_ParameterSetNotifyRegister() thorns can register a callback function which in turn is automatically invoked by CCTK_ParameterSet() whenever a parameter is being steered. Each callback function gets passed the triple of thorn name, parameter name, and (stringified) new parameter value (as passed to CCTK_ParameterSet()), plus an optional callback data pointer defined by the user at registration time. When a callback function is registered with CCTK_ParameterSetNotify(), the calling routine may also pass an optional regular expression string for both a thorn name and a parameter name to match against in a parameter set notification; leave them empty or pass a NULL pointer to get notified about changes of any parameter.
Registered notification callbacks would be invoked by CCTK_ParameterSet() only after initial parameter setup from the parfile, and – in case of recovery – only after all parameters have been restored from the checkpoint file. The callbacks are then invoked just before the parameter is set to its new value so that they can still query its old value if necessary.
See Also
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterSetNotifyUnregister [A441] Unregisters a parameter set operation notify callback
Examples
#include <stdio.h> #include "cctk.h" static void ParameterSetNotify (void *unused, const char *thorn, const char *parameter, const char *new_value) { printf ("parameter set notification: %s::%s is set to ’%s’\n", thorn, parameter, new_value); } void RegisterNotifyCallback (void) { /* we are interested only in this thorn’s parameters so pass the thorn name in the ’thorn_regex’ argument */ if (CCTK_ParameterSetNotifyRegister (ParameterSetNotify, NULL, CCTK_THORNSTRING, CCTK_THORNSTRING, NULL)) { CCTK_VWarn (0, __LINE__, __FILE__, CCTK_THORNSTRING, "Couldn’t register parameter set notify callback"); } }
Unregisters a parameter set operation notify callback
Synopsis
#include "cctk.h" int ierr = CCTK_ParameterSetNotifyUnregister (const char *name);
call CCTK_ParameterSetNotifyUnregister (ierr, name) integer ierr character*(*) name
Result
0 success
-1 no callback was registered under the given name
Parameters
name Unique name under which the notify callback was registered
Discussion
Notify callbacks should be unregistered when not needed anymore.
See Also
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterSetNotifyRegister [A437] Registers a parameter set operation notify callback
Examples
#include "cctk.h" call CCTK_ParameterSetNotifyUnregister (CCTK_THORNSTRING)
Get the string representation of a parameter’s value.
Synopsis
#include "cctk.h" char *valstring = CCTK_ParameterValString (const char *name, const char *thorn);
subroutine CCTK_ParameterValString (nchars, name, thorn, value) integer nchars character*(*) name character*(*) thorn character*(*) value end subroutine
Result
valstring Pointer to parameter value as string. The memory for this string must be released with a call to
free() after it has been used.
Parameters
name Parameter name
thorn Thorn name (for private parameters) or implementation name (for restricted parameters)
nchars On exit, the number of characters in the stringified parameter value, or \(-1\) if the parameter doesn’t exist
value On exit, contains as many characters of the stringified parameter value as fit into the Fortran string
provided. You should check for truncation by comparing nchars against the length of your Fortran string.
Discussion
In C, the string valstring must be freed afterwards.
The thorn or implementation name must be the name of the place where the parameter is originally defined. It is not possible to pass the thorn or implementation name of a thorn that merely declares the parameter as used.
Real variables are formatted according to the C "%.20g" format.
See Also
CCTK_ParameterData [A423] Get parameter properties for given parameter/thorn pair
CCTK_ParameterGet [A428] Get the data pointer to and type of a parameter’s value
CCTK_ParameterLevel [A430] Return the parameter checking level
CCTK_ParameterQueryTimesSet [A431] Return number of times a parameter has been set
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterWalk [A448] Walk through list of parameters
Errors
NULL No parameter with that name was found.
Walk through the list of parameters.
Synopsis
#include "cctk.h" % int istat = CCTK_ParameterWalk (int first, const char *origin, char **fullname, const cParamData **paramdata);
Result
istat Zero for success, positive if parameter was not found, negative if initial startpoint was not set.
Parameters
origin Thorn name, or NULL for all thorns.
fullname Address of a pointer that will point to the full parameter name. This name must be freed after use.
paramdata Address of a pointer that will point to the parameter data structure.
Discussion
Gets parameters in order, restricted to ones from origin, or all if origin is NULL. Starts with the first parameter if first is true, otherwise gets the next one. Can be used for generating full help file, or for walking the list and checkpointing.
See Also
CCTK_ParameterData [A423] Get parameter properties for given parameter/thorn pair
CCTK_ParameterGet [A428] Get the data pointer to and type of a parameter’s value
CCTK_ParameterLevel [A430] Return the parameter checking level
CCTK_ParameterQueryTimesSet [A431] Return number of times a parameter has been set
CCTK_ParameterSet [A433] Sets the value of a parameter
CCTK_ParameterValString [A445] Get the string representation of a parameter’s value
Errors
negative The initial startpoint was not set.
Prints a warning from parameter checking, and possibly stops the code
Synopsis
= CCTK_PARAMWARN( const char * message)
call CCTK_PARAMWARN( , message ) character*(*) message
Parameters
message The warning message
Discussion
The call should be used in routines registered at the schedule point CCTK_PARAMCHECK to indicate that there is parameter error or conflict and the code should terminate. The code will terminate only after all the parameters have been checked.
The macro CCTK_PARAMWARN(message) expands to a call to the underlying function CCTK_ParamWarn:
CCTK_ParamWarn(CCTK_THORNSTRING, message)
So the macro automatically includes the name of the originating thorn in the info message. It is recommended that the macro CCTK_PARAMWARN is used to print a message rather than calling CCTK_ParamWarn directly.
To include variables in an info message from C, you can use the routine CCTK_VParamWarn which accepts a variable argument list. To include variables from Fortran, a string must be constructed and passed in a CCTK_PARAMWARN macro.
Examples
CCTK_PARAMWARN("Mass cannot be negative");
call CCTK_PARAMWARN("Inside interpolator")
See Also
CCTK_VParamWarn [A622] Prints a formatted string with a variable argument list as a warning from parameter
checking, and possibly stops the code
Returns a pointer to a Fortran variable.
Synopsis
#include "cctk.h" CCTK_POINTER addr, var addr = CCTK_PointerTo(var)
Result
addr the address of variable var
Parameters
var variable in the Fortran context from which to take the address
Discussion
Fortran doesn’t know the concept of pointers so problems arise when a C function is to be called which expects a pointer as one (or more) of it(s) argument(s).
To obtain the pointer to a variable in Fortran, one can use CCTK_PointerTo() which takes the variable itself as a single argument and returns the pointer to it.
Note that there is only a Fortran wrapper available for CCTK_PointerTo.
See Also
CCTK_NullPointer() Returns a C-style NULL pointer value.
Examples
#include "cctk.h" integer ierror, table_handle CCTK_POINTER addr, var addr = CCTK_PointerTo(var) call Util_TableCreate(table_handle, 0) call Util_TableSetPointer(ierror, table_handle, addr, "variable")
Prints a group name from its index
Synopsis
= CCTK_PrintGroup( int index)
call CCTK_PrintGroup( , index ) integer index
Parameters
index The group index
Discussion
This routine is for debugging purposes for Fortran programmers.
Examples
CCTK_PrintGroup(1)
call CCTK_PRINTGROUP(1)
Prints a Cactus string
Synopsis
= CCTK_PrintString( char * string)
call CCTK_PrintString( , string ) CCTK_STRING string
Parameters
string The string to print
Discussion
This routine can be used to print Cactus string variables and parameters from Fortran.
Examples
CCTK_PrintString(string_param)
call CCTK_PRINTSTRING(string_param)
Prints a variable name from its index
Synopsis
= CCTK_PrintVar( int index)
call CCTK_PrintVar( , index ) integer index
Parameters
index The variable index
Discussion
This routine is for debugging purposes for Fortran programmers.
Examples
CCTK_PrintVar(1)
call CCTK_PRINTVAR(1)
Query storage for a group given by its group name
Synopsis
int istat = CCTK_QueryGroupStorage( const cGH * cctkGH, const char * groupname)
call CCTK_QueryGroupStorage(istat , cctkGH, groupname ) integer istat CCTK_POINTER cctkGH character*(*) groupname
Parameters
cctkGH pointer to CCTK grid hierarchy
groupname the group to query, given by its full name
istat the return code
Discussion
This routine queries whether the variables in a group have storage assigned. If so it returns true (a positive value), otherwise false (zero).
Errors
negative A negative error code is returned for an invalid group name.
Synopsis
int storage = CCTK_QueryGroupStorageB( const cGH * cctkGH, int groupindex, const char * groupname)
Parameters
cctkGH pointer to CCTK grid hierarchy
groupindex the group to query, given by its index
groupname the group to query, given by its full name
istat the return code
Discussion
This routine queries whether the variables in a group have storage assigned. If so it returns true (a positive value), otherwise false (zero).
The group can be specified either through the group index groupindex, or through the group name groupname. The groupname takes precedence; only if it is passed as NULL, the group index is used.
Errors
negative A negative error code is returned for an invalid group name.
Query storage for a group given by its group index
Synopsis
int istat = CCTK_QueryGroupStorageI( const cGH * cctkGH, int groupindex)
call CCTK_QueryGroupStorageI(istat , cctkGH, groupindex ) integer istat cctkGH integer groupindex
Parameters
cctkGH pointer to CCTK grid hierarchy
groupindex the group to query, given by its index
istat the return code
Discussion
This routine queries whether the variables in a group have storage assigned. If so it returns true (a positive value), otherwise false (zero).
Errors
negative A negative error code is returned for an invalid group name.
Performs global reduction on a list of arrays
The computation is optimized for the case of reducing a number of grid arrays at a time; in this case all the interprocessor communication can be done together.
Synopsis
#include "cctk.h" int CCTK_ReduceArraysGlobally(const cGH *GH, int dest_proc, int local_reduce_handle, int param_table_handle, int N_input_arrays, const void * const input_arrays[], int input_dims, const CCTK_INT input_array_dims[], const CCTK_INT input_array_type_codes[], int M_output_values, const CCTK_INT output_value_type_codes[], void* const output_values[]);
call CCTK_ReduceArraysGlobally(status, . GH, . dest_proc, . local_reduce_handle, . param_table_handle, . N_input_arrays, . input_arrays, . input_dims, . input_array_dims, . input_array_type_codes, . M_output_values, . output_value_type_codes, . output_values) integer status CCTK_POINTER_TO_CONST GH integer dest_proc, integer local_reduce_handle integer param_table_handle integer N_input_arrays CCTK_INT input_arrays(N_input_arrays) integer input_dims CCTK_INT input_array_dims(input_dims) CCTK_INT input_array_type_codes(N_input_arrays) integer M_output_values CCTK_INT output_value_type_codes(M_output_values) CCTK_POINTER output_values(M_output_values)
Result
0 success
< 0 indicates an error condition
Parameters
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dest_processor The destination processor. \(-1\) will distribute the result to all processors.
local_reduce_handle (\(\ge 0\)) Handle to the local reduction operator as returned by
CCTK_LocalArrayReductionHandle(). It is the caller’s responsibility to ensure that the specified reducer
supports any optional parameter-table entries that
CCTK_ReduceGridArrays() passes to it. Each thorn providing a
CCTK_ReduceGridArrays() reducer should document what options it requires from the local reducer.
param_table_handle (\(\ge 0\)) Handle to a key-value table containing zero or more additional parameters for the
reduction operation. The table can be modified by the local and/or global reduction routine(s).
Also, the global reducer will typically need to specify some options of its own for the local reducer. These will override any entries with the same keys in the param_table_handle table. The discussion of individual table entries below says if these are modified in this manner.
Finally, the param_table_handle table can be used to pass back arbitrary information by the local and/or global
reduction routine(s) by adding/modifying appropriate key/value pairs.
N_input_arrays (\(\ge 0\)) The number of input arrays to be reduced. If N_input_arrays is zero, then no reduction is
done; such a call may be useful for setup, reducer querying, etc. If the operand_indices parameter table entry is
used to specify a nontrivial (eg 1-to-many) mapping of input arrays to output values, only the unique set of
input arrays should be given here.
input_arrays (Pointer to) an array of N_input_arrays local arrays specifying the input arrays for the
reduction.
input_dims (\(\ge 0\)) The number of dimensions of the input arrays
input_array_dims (\(\ge 0\)) (Pointer to) an array of size input_dims containing the dimensions of the arrays to be
reduced.
input_array_type_codes (\(\ge 0\)) (Pointer to) an array of input_dims CCTK_VARIABLE_* type codes giving the
data types of the arrays to be reduced.
M_output_values (\(\ge 0\)) The number of output values to be returned from the reduction. If N_input_arrays == 0
then no reduction is done; such a call may be useful for setup, reducer querying, etc. Note that
M_output_values may differ from N_input_arrays , eg if the operand_indices parameter table entry is used
to specify a nontrivial (eg many-to-1) mapping of input arrays to output values, If such a mapping is specified,
only the unique set of output values should be given here.
output_value_type_codes (Pointer to) an array of M_output_values CCTK_VARIABLE_* type codes giving the
data types of the output values pointed to by output_values[].
output_values (Pointer to) an array of M_output_values pointers to the (caller-supplied) output values for
the reduction. If output_values[out] is NULL for some index or indices out , then that reduction is skipped.
(This may be useful if the main purpose of the call is (eg) to do some query or setup computation.) These
pointers may (and typically will) vary from processor to processor in a multiprocessor Cactus run.
However, any given pointer must be either NULL on all processors, or non-NULL on all processors.
4
Discussion
This function reduces a list of CCTK local arrays globally. This function does not perform the actual reduction, it only handles interprocessor communication. The actual reduction is performed by the local reduction implementation, that is passed arguments and parameters from the grid array reduction implementation.
Note that CCTK_ReduceArraysGlobally is a collective operation, so in the multiprocessor case you must call this function in parallel on each processor, passing identical arguments.
See Also
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_RegisterGridArrayReductionOperator() Registers a function as a grid array reduction operator of a
certain name
CCTK_GridArrayReductionOperator() The name of the grid reduction operator, or NULL if the handle is
invalid
CCTK_GridArrayReductionOperator() The number of grid array reduction operators registered
Examples
Here’s a simple example to perform grid array reduction of two grids arrays of different types.
#include "cctk.h" #include "util_Table.h" #define N_INPUT_ARRAYS 2 #define M_OUTPUT_VALUES 2 const cGH *GH; /* input */ /* create empty parameter table */ const int param_table_handle = Util_CreateTable(UTIL_TABLE_FLAGS_CASE_INSENSITIVE); /* input arrays and output values */ const CCTK_INT input_array_variable_indices[N_INPUT_ARRAYS] = { CCTK_VarIndex("my_thorn::real_array"), /* no error checking */ CCTK_VarIndex("my_thorn::complex_array") }; /* here */ const CCTK_INT output_value_type_codes[M_OUTPUT_VALUES] = { CCTK_VARIABLE_REAL, CCTK_VARIABLE_COMPLEX }; void *const output_numbers[M_OUTPUT_values] = { (void *) output_for_real_values, (void *) output_for_complex_values }; const int status = CCTK_ReduceGridArrays(GH, 0, param_table_handle, N_INPUT_ARRAYS, input_array_variable_indices, M_OUTPUT_VALUES, output_value_type_codes, output_values); Util_TableDestroy(param_table_handle);
Performs reduction on a list of distributed grid arrays
The computation is optimized for the case of reducing a number of grid arrays at a time; in this case all the interprocessor communication can be done together.
Synopsis
#include "cctk.h" int status = CCTK_ReduceGridArrays(const cGH *GH, int dest_processor, int local_reduce_handle, int param_table_handle, int N_input_arrays, const CCTK_INT input_array_variable_indices[], int M_output_values, const CCTK_INT output_value_type_codes[], void* const output_values[]);
call CCTK_ReduceGridArrays(status, . GH, . dest_processor, . local_reduce_handle, . param_table_handle, . N_input_arrays, . input_array_variable_indices, . M_output_values, . output_value_type_codes, . output_values) integer status CCTK_POINTER_TO_CONST GH integer dest_processor integer local_reduce_handle integer param_table_handle integer N_input_arrays CCTK_INT input_array_variable_indices(N_input_arrays) integer M_output_values CCTK_INT output_value_type_codes(M_output_values) CCTK_POINTER output_values(M_output_values)
Result
0 success
< 0 indicates an error condition
Parameters
cctkGH (\(\ne \) NULL) Pointer to a valid Cactus grid hierarchy.
dest_processor The destination processor. \(-1\) will distribute the result to all processors.
local_reduce_handle (\(\ge 0\)) Handle to the local reduction operator as returned by
CCTK_LocalArrayReductionHandle(). It is the caller’s responsibility to ensure that the specified reducer
supports any optional parameter-table entries that
CCTK_ReduceGridArrays() passes to it. Each thorn providing a
CCTK_ReduceGridArrays() reducer should document what options it requires from the local reducer.
param_table_handle (\(\ge 0\)) Handle to a key-value table containing zero or more additional parameters for the
reduction operation. The table can be modified by the local and/or global reduction routine(s).
Also, the global reducer will typically need to specify some options of its own for the local reducer. These will override any entries with the same keys in the param_table_handle table. The discussion of individual table entries below says if these are modified in this manner.
Finally, the param_table_handle table can be used to pass back arbitrary information by the local and/or global
reduction routine(s) by adding/modifying appropriate key/value pairs.
N_input_arrays (\(\ge 0\)) The number of input arrays to be reduced. If N_input_arrays is zero, then no reduction is
done; such a call may be useful for setup, reducer querying, etc. If the operand_indices parameter table entry is
used to specify a nontrivial (eg 1-to-many) mapping of input arrays to output values, only the unique set of
input arrays should be given here.
input_array_variable_indices (Pointer to) an array of N_input_arrays Cactus variable indices (as returned
by CCTK_VarIndex() ) specifying the input grid arrays for the reduction. If
input_array_variable_indices[in] == -1 for some index or indices in , then that reduction is skipped.
(This may be useful if the main purpose of the call is (eg) to do some query or setup computation.)
M_output_values (\(\ge 0\)) The number of output values to be returned from the reduction. If N_input_arrays == 0
then no reduction is done; such a call may be useful for setup, reducer querying, etc. Note that
M_output_values may differ from N_input_arrays , eg if the operand_indices parameter table entry is used
to specify a nontrivial (eg many-to-1) mapping of input arrays to output values, If such a mapping is specified,
only the unique set of output values should be given here.
output_value_type_codes (Pointer to) an array of M_output_values CCTK_VARIABLE_* type codes giving the
data types of the output values pointed to by output_values[].
output_values (Pointer to) an array of M_output_values pointers to the (caller-supplied) output values for
the reduction. If output_values[out] is NULL for some index or indices out , then that reduction is skipped.
(This may be useful if the main purpose of the call is (eg) to do some query or setup computation.) These
pointers may (and typically will) vary from processor to processor in a multiprocessor Cactus run.
However, any given pointer must be either NULL on all processors, or non-NULL on all processors.
Discussion
This function reduces a list of CCTK grid arrays (in a multiprocessor run these are generally distributed over processors). This function does not perform the actual reduction, it only handles interprocessor communication. The actual reduction is performed by the local reduction implementation, that is passed arguments and parameters from the grid array reduction implementation.
Note that CCTK_ReduceGridArrays is a collective operation, so in the multiprocessor case you must call this function in parallel on each processor, passing identical arguments.
See Also
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_RegisterGridArrayReductionOperator() Registers a function as a grid array reduction operator of a
certain name
CCTK_GridArrayReductionOperator() The name of the grid reduction operator, or NULL if the handle is
invalid
CCTK_GridArrayReductionOperator() The number of grid array reduction operators registered
Examples
Here’s a simple example to perform grid array reduction of two grids arrays of different types.
#include "cctk.h" #include "util_Table.h" #define N_INPUT_ARRAYS 2 #define M_OUTPUT_VALUES 2 const cGH *GH; /* input */ /* create empty parameter table */ const int param_table_handle = Util_CreateTable(UTIL_TABLE_FLAGS_CASE_INSENSITIVE); /* input arrays and output values */ const CCTK_INT input_array_variable_indices[N_INPUT_ARRAYS] = { CCTK_VarIndex("my_thorn::real_array"), /* no error checking */ CCTK_VarIndex("my_thorn::complex_array") }; /* here */ const CCTK_INT output_value_type_codes[M_OUTPUT_VALUES] = { CCTK_VARIABLE_REAL, CCTK_VARIABLE_COMPLEX }; void *const output_numbers[M_OUTPUT_values] = { (void *) output_for_real_values, (void *) output_for_complex_values }; const int status = CCTK_ReduceGridArrays(GH, 0, param_table_handle, N_INPUT_ARRAYS, input_array_variable_indices, M_OUTPUT_VALUES, output_value_type_codes, output_values); Util_TableDestroy(param_table_handle);
Performs reduction on a list of local grid arrays
Synopsis
#include "cctk.h" int status = CCTK_ReduceLocalArrays(int N_dims, int operator_handle, int param_table_handle, int N_input_arrays, const CCTK_INT input_array_dims[], const CCTK_INT input_array_type_codes[], const void *const input_arrays[], int M_output_numbers, const CCTK_INT output_number_type_codes[], void *const output_values[]);
call CCTK_ReduceLocalArrays(status, . N_dims, operator_handle, . param_table_handle, N_input_arrays, . input_array_dims, . input_array_type_codes, . input_arrays, . M_output_numbers, . output_number_type_codes, . output_values) integer status integer N_dims integer operator_handle integer param_table_handle integer N_input_arrays CCTK_INT input_array_dims(N_dims) CCTK_INT input_array_type_codes(N_input_arrays) CCTK_POINTER input_arrays(N_input_arrays) integer M_output_values CCTK_INT output_value_type_codes(M_output_values) CCTK_POINTER output_values(M_output_values)
Result
0 success
< 0 indicates an error condition
Parameters
N_dims Number of dimensions of input arrays. This is required to find proper indices for arrays in memory
operator_handle Handle to the local reduction operator as returned by
CCTK_LocalArrayReductionHandle().
param_table_handle Handle to a key-value table containing zero or more additional parameters for the
reduction operation. The table can be modified by the local and/or global reduction routine(s).
The parameter table may be used to specify non-default storage indexing for input or output arrays, and/or
various options for the reduction itself. Some reducers may not implement all of these options. N_input_arrays
(\(\ge 0\)) The number of input arrays to be reduced. If N_input_arrays is zero, then no reduction is done; such a call
may be useful for setup, reducer querying, etc. If the operand_indices parameter table entry is used to specify a
nontrivial (eg 1-to-many) mapping of input arrays to output values, only the unique set of input arrays should
be given here.
input_array_dims array of input array dimensions (common to all input arrays) and of size N_dims
input_array_type_codes array of input array dimensions (common to all input arrays) and of size
N_input_arrays
M_output_values (\(\ge 0\)) The number of output values to be returned from the reduction. If N_input_arrays == 0
then no reduction is done; such a call may be useful for setup, reducer querying, etc. Note that
M_output_values may differ from N_input_arrays , eg if the operand_indices parameter table entry is used
to specify a nontrivial (eg many-to-1) mapping of input arrays to output values, If such a mapping is specified,
only the unique set of output values should be given here.
output_value_type_codes (Pointer to) an array of M_output_values CCTK_VARIABLE_* type codes giving the
data types of the output values pointed to by output_values[].
output_values (Pointer to) an array of M_output_values pointers to the (caller-supplied) output values for
the reduction. If output_values[out] is NULL for some index or indices out , then that reduction is skipped.
(This may be useful if the main purpose of the call is (eg) to do some query or setup computation.)
Discussion
Sometimes one of the arrays used by the reduction isn’t contiguous in memory. So, we use several optional table entries (these should be supported by all reduction operators):
For the input arrays, we use
const CCTK_INT input_array_offsets[N_input_arrays]; /* next 3 table entries are shared by all input arrays */ const CCTK_INT input_array_strides [N_dims]; const CCTK_INT input_array_min_subscripts[N_dims]; const CCTK_INT input_array_max_subscripts[N_dims];
Then for input array number a, the generic subscripting expression for the 3-D case is
data_pointer[offset + i*istride + j*jstride + k*kstride]
where
data_pointer = input_arrays[a] offset = input_array_offsets[a] (istride,jstride,kstride) = input_array_stride[]
and where (i,j,k) run from input_array_min_subscripts[] to input_array_max_subscripts[] inclusive.
The defaults are offset=0, stride=determined from input_array_dims[] in the usual Fortran manner, input_array_min_subscripts[] = 0, input_array_max_subscripts[] = input_array_dims[]-1. If the stride and max subscript are both specified explicitly, then the input_array_dims[] function argument is ignored.
See Also
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_RegisterLocalArrayReductionOperator() Registers a function as a reduction operator of a certain
name
CCTK_LocalArrayReduceOperatorImplementation() Provide the implementation which provides an local
array reduction operator
CCTK_LocalArrayReduceOperator() Returns the name of a registered reduction operator
CCTK_NumLocalArrayReduceOperators() The number of local reduction operators registered
Examples
Here’s a simple example, written in Fortran 77, to do reduction of a real and a complex local array in 3-D:
c input arrays: integer ni, nj, nk parameter (ni=..., nj=..., nk=...) CCTK_REAL real_array (ni,nj,nk) CCTK_COMPLEX complex_array(ni,nj,nk) c output numbers: CCTK_REAL My_real (M_reduce) CCTK_COMPLEX My_complex(M_reduce) integer status, dummy CCTK_INT input_array_type_codes(2) data input_array_type_codes /CCTK_VARIABLE_REAL, $ CCTK_VARIABLE_COMPLEX/ CCTK_INT input_array_dims(3) CCTK_POINTER input_arrays(2) CCTK_POINTER output_numbers(2) input_array_dims(1) = ni input_array_dims(2) = nj input_array_dims(3) = nk output_numbers(1) = Util_PointerTo(My_real) output_numbers(2) = Util_PointerTo(My_complex) call CCTK_ReduceLocalArrays $ (status, ! return code 3, ! number of dimensions operator_handle, N_reduce, 2, ! number of input arrays input_array_type_codes, input_array_dims, input_arrays, 2, ! number of output numbers output_numbers_type_codes, output_numbers) if (status .lt. 0) then call CCTK_WARN(CCTK_WARN_ABORT, "Error return from reducer!") end if
Handle for given reduction method
Synopsis
int handle = CCTK_ReductionHandle( const char * reduction)
handle = CCTK_ReductionHandle( reduction ) integer handle character*(*) reduction
Parameters
handle handle returned for this method
name name of the reduction method required
Discussion
Reduction methods should be registered at CCTK_STARTUP. Note that integer reduction handles are used to call CCTK_Reduce to avoid problems with passing Fortran strings. Note that the name of the reduction operator is case dependent.
Examples
handle = CCTK_ReductionHandle("maximum");
call CCTK_ReductionHandle(handle,"maximum")
Perform a regular expression match of string against pattern
Synopsis
success = CCTK_RegexMatch( const char *string, const char *pattern, const int nmatch, regmatch_t *pmatch)
Parameters
string String to match against
pattern Regex pattern
nmatch The size of the pmatch array
pmatch Array in which to place the matches
Result
0 pattern does not match
1 pattern matches
< 0 indicates an error condition (pattern did not compile as a regular expression)
Discussion
Perform a regular expression match of string against pattern. Also returns the specified number of matched substrings as give by regexec. This is a modified form of the example routine given in the SGI man page for regcomp.
Examples
#define R_BEGIN "(\\[|\\()?" #define R_VALUE "([^]):]*)" #define R_SEP ":" #define R_END "(\\]|\\))?" #define R_MAYBE(x) "(" x ")?" int matched; const char pattern[] = R_BEGIN R_VALUE R_MAYBE(R_SEP R_VALUE R_MAYBE(R_SEP R_VALUE)) R_END; if( (matched = CCTK_RegexMatch(range, pattern, 8, pmatch)) > 0) { CCTK_VInfo(CCTK_THORNSTRING, "’%s’ is a valid range specifier", range); } else if(!matched) { CCTK_VInfo(CCTK_THORNSTRING, "’%s’ is not a valid range specifier", range); } else { CCTK_VInfo(CCTK_THORNSTRING, "invalid pattern ’%s’", pattern); }
Register a banner for a thorn
Synopsis
void = CCTK_RegisterBanner( const char * message)
call CCTK_RegisterBanner( , message ) character*(*) message
Parameters
message String which will be displayed as a banner
Discussion
The banner must be registered during CCTK_STARTUP. The banners are displayed in the order in which they are registered.
Examples
CCTK_RegisterBanner("My Thorn: Does Something Useful");
call CCTK_REGISTERBANNER("*** MY THORN ***")
Register an extension to the CactusGH
Synopsis
int istat = CCTK_RegisterGHExtension( const char * name)
Register a function which will initialise a given extension to the Cactus GH
Synopsis
int istat = CCTK_RegisterGHExtensionInitGH( int handle, void * (*func)(cGH *))
Register a GH extension schedule traversal routine
Synopsis
int istat = CCTK_RegisterGHExtensionScheduleTraverseGH( int handle, int (*func)(cGH *,const char *))
Register a function which will set up a given extension to the Cactus GH
Synopsis
int istat = CCTK_RegisterGHExtensionSetupGH( int handle, void * (*func)(tFleshConfig *, int, cGH *))
Registers a function as a grid array reduction operator of a certain name
Synopsis
#include "cctk.h" int status = CCTK_RegisterGridArrayReductionOperator( cGridArrayReduceOperator operator)
Result
0 success
< 0 indicates an error condition
Parameters
operator The function to register as a global reduction function.
Discussion
This function simply registers a function as the grid array reduction. Currently we support a single function as a global reduction function (this can be modified to accomodate more functions if need be).
See Also
CCTK_ReduceGridArrays() Performs reduction on a list of distributed grid arrays
CCTK_GridArrayReductionOperator() The name of the grid reduction operator, or NULL if none is registered
CCTK_NumGridArrayReductionOperators() The number of grid array reduction operators registered
Register a new I/O method
Synopsis
int handle = CCTK_RegisterIOMethod( const char * name)
handle = CCTK_RegisterIOMethod( name ) integer handle name
Parameters
handle handle returned by registration
name name of the I/O method
Discussion
IO methods should be registered at CCTK_STARTUP.
Register a routine for an I/O method which will be called from CCTK_OutputGH.
Synopsis
int istat = CCTK_RegisterIOMethodOutputGH( int handle, int (* func)(const cGH *))
Register a routine for an I/O method which will provide aliased variable output
Synopsis
int istat = CCTK_RegisterIOMethodOutputVarAs( int handle, int (* func)(const cGH *,const char*, const char *))
Register a routine for an I/O method which will decide if it is time for the method to output.
Synopsis
int istat = CCTK_RegisterIOMethodTimeToOutput( int handle, int (* func)(const cGH *,int))
Register a routine for an I/O method which will handle trigger output
Synopsis
int istat = CCTK_RegisterIOMethodTriggerOutput( int handle, int (* func)(const cGH *,int))
Registers a function as a reduction operator of a certain name
Synopsis
#include "cctk.h" int handle = CCTK_RegisterLocalArrayReductionOperator( cLocalArrayReduceOperator operator, const char *name);
Result
handle The handle corresponding to the registered local reduction operator, -1 if an error occured.
Parameters
operator The function to be registered as a local reduction operator
name The name under which the operator is registered as a local reduction operator
Discussion
This function registers a local array reduction operator. It registers an operator under a name with the flesh and returns its assigned handle. If another reduction operator exists with the same name, an error is returned.
See Also
CCTK_ReduceLocalArrays() Reduces a list of local arrays (new local array reduction API)
CCTK_LocalArrayReductionHandle() Returns the handle of a given local array reduction operator
CCTK_LocalArrayReduceOperatorImplementation() Provide the implementation which provides an local
array reduction operator
CCTK_LocalArrayReduceOperator() Returns the name of a registered reduction operator
CCTK_NumLocalArrayReduceOperators() The number of local reduction operators registered
Registers a function as a reduction operator of a certain name
Synopsis
#include "cctk.h" int handle = CCTK_RegisterReduceArraysGloballyOperator( cReduceArraysGloballyOperator operator, const char *name);
Result
handle The handle corresponding to the registered global array reduction operator, -1 if an error occured.
Parameters
operator The function to be registered as a global array reduction operator
name The name under which the operator is registered as a global array reduction operator
Discussion
This function registers a global array reduction operator. It registers an operator under a name with the flesh and returns its assigned handle. If another reduction operator exists with the same name, an error is returned.
See Also
CCTK_ReduceArraysGlobally() Reduces a list of local arrays globally
Synopsis
CCTK_RegisterReductionOperator()
Synopsis
int CCTK_RunTime()
Result
seconds The number of seconds since the run started.
Output the timing results for a certain schedule item to stdout
Synopsis
#include "cctk.h" int status = CCTK_SchedulePrintTimes(const char *where)
Result
Return code of DoScheduleTraverse, or
0 Success.
Parameters
where Name of schedule item, or NULL to print the whole schedule
Discussion
Output the timing results for a certain schedule item to stdout. The schedule item is traversed recursively if it is a schedule group or a schedule bin.
This routine is used to produce the timing output when the parameter Cactus::cctk_timer_output is set to yes.
See Also
CCTK_SchedulePrintTimesToFile Output the timing results for a certain schedule item to a file
Examples
#include "cctk.h" int status = CCTK_SchedulePrintTimes("CCTK_ANALYSIS")
Output the timing results for a certain schedule item to a file
Synopsis
#include "cctk.h" int status = CCTK_SchedulePrintTimesToFile(const char *where, FILE *file)
Result
Return code of DoScheduleTraverse, or
0 Success.
Parameters
where Name of schedule item, or NULL to print the whole schedule
file File to which the results are output; the file must be open for writing
Discussion
Output the timing results for a certain schedule item to a file. The schedule item is traversed recursively if it is a schedule group or a schedule bin.
Note that each processor will output its results. You should either call this routine on only a single processor, or you should pass different files on different processors.
See Also
CCTK_SchedulePrintTimes Output the timing results for a certain schedule item to stdout
Examples
#include <stdio.h> #include "cctk.h" if (CCTK_MyProc(cctkGH)==3) { FILE *file = fopen("timing-results.txt", "a"); int status = CCTK_SchedulePrintTimesToFile("CCTK_ANALYSIS", file) fclose(file); }
Return the cFunctionData of the function currently executing via CCTK_CallFunction.
Synopsis
#include "cctk.h" const cFunctionData *CCTK_ScheduleQueryCurrentFunction(const cGH *GH)
Result
Data of last call to CCTK_CallFunction, or
NULL if not within a scheduled function.
Parameters
cctkGH Pointer to a Cactus grid hierarchy.
Discussion
Returns a data structure containing the thorn and routine name of the currently executing function as well as the Cactus schedule bin or schedule group name. If no function is currently executing, returns NULL. This is intended to be used by thorns providing callable functions to identify their caller when reporting errors.
See Also
CCTK_CallFunction Calls a function depending upon the data passed in the the fdata structure.
Examples
#include <stdio.h> #include "cctk.h" const cFunctionData *fdata = CCTK_ScheduleQueryCurrentFunction(cctkGH); printf("scheduled function: %s::%s AT %s\n", fdata->thorn, fdata->routine, fdata->where);
Traverses a schedule point, and its entry and exit points if necessary.
Synopsis
#include "cctk.h" typedef int (*CallFunction_t)(void *func, cFunctionData *attributes, void *data); int CCTK_ScheduleTraverse(const char *where, void *data, CallFunction_t CallFunction);
Result
0 success
1 memory failure
2 schedule item not found
3 unknown error
Parameters
where Schedule point
data user data passed alongto CallFunction as its last argument
CallFunction callback function with the same signature as CCTK_CallFunction. Pass NULL to use the default.
Discussion
This functions is intended to be used by driver thorns to iterate through the schedule and act on the scheduled functions. Using it in user code can lead to recursive function call warnings.
See Also
CCTK_CallFunction Calls a function depending upon the data passed in the the fdata structure.
Examples
#include <stdio.h> #include "cctk.h" int CallFunction(void *fun, cFunctionData *fdata, void *data) { int *count = data; *count += 1; printf("scheduled function: %s::%s AT %s\n", fdata->thorn, fdata->routine, fdata->where); return 1; // no need to SYNC } int count = 0; int ierr = CCTK_ScheduleTraverse("MoL_PostStep", CallFunction, &count); printf("There were %d functions scheduled\n", count);
Setup a new GH
Synopsis
cGH * cctkGH = CCTK_SetupGH( tFleshConfig config, int convlevel)
Synchronise the ghostzones for a group of grid variables (identified by the group name)
Synopsis
#include "cctk.h" int status = CCTK_SyncGroup(const cGH* GH, const char* group_name)
#include "cctk.h" integer status CCTK_POINTER GH character*(*) group_name call CCTK_SyncGroup(status, GH, group_name)
Result
0 Success.
Parameters
GH A pointer to a Cactus grid hierarchy.
group_name The full name (Implementation::group or Thorn::group) of the group to be synchronized.
Discussion
Only those grid variables which have communication enabled will be synchronised. This is usually equivalent to the variables which have storage assigned, unless communication has been explicitly turned off with a call to CCTK_DisableGroupComm.
Note that an alternative to calling CCTK_SyncGroup explicitly from within a thorn, is to use the SYNC keyword in a thorns schedule.ccl file to indicate which groups of variables need to be synchronised on exit from the routine. This latter method is the preferred method from synchronising variables.
Note that CCTK_SyncGroup is a collective operation, so in the multiprocessor case you must call this function in parallel on each processor, passing the same group_name argument.
See Also
CCTK_SyncGroupI [A539] Synchronise the ghostzones for a group of grid variables (identified by the group index)
CCTK_SyncGroupsI [A544] Synchronise the ghostzones for a list of groups of grid variables (identified by their
group indices)
Errors
-1 group_name was invalid.
-2 The driver returned an error on syncing the group.
Examples
#include "cctk.h" #include "cctk_Arguments.h" /* this function synchronizes the ADM metric */ void synchronize_ADM_metric(CCTK_ARGUMENTS) { DECLARE_CCTK_ARGUMENTS /* defines "magic variable" cctkGH */ const int status = CCTK_SyncGroup(cctkGH, "ADMBase::metric"); if (status < 0) CCTK_VWarn(CCTK_WARN_ABORT, __LINE__, __FILE__, CCTK_THORNSTRING, "***** synchronize_ADM_metric():\n" " failed to synchronize ADM metric!\n" " (CCTK_SyncGroup() returned error code %d)\n" , status); /*NOTREACHED*/ }
Synchronise the ghostzones for a group of grid variables (identified by the group index)
Synopsis
#include "cctk.h" int status = CCTK_SyncGroupI(const cGH* GH, int group_index)
#include "cctk.h" integer status CCTK_POINTER GH integer group_index call CCTK_SyncGroupI(status, GH, group_index)
Result
0 Success.
Parameters
GH A pointer to a Cactus grid hierarchy.
group_index The group index of the group to be synchronized.
Discussion
Only those grid variables which have communication enabled will be synchronised. This is usually equivalent to the variables which have storage assigned, unless communication has been explicitly turned off with a call to CCTK_DisableGroupComm.
Note that an alternative to calling CCTK_SyncGroupI explicitly from within a thorn, is to use the SYNC keyword in a thorns schedule.ccl file to indicate which groups of variables need to be synchronised on exit from the routine. This latter method is the preferred method from synchronising variables.
Note that CCTK_SyncGroupI is a collective operation, so in the multiprocessor case you must call this function in parallel on each processor, passing the same group_name argument.
See Also
CCTK_SyncGroup [A535] Synchronise the ghostzones for a group of grid variables (identified by the group name)
CCTK_SyncGroupsI [A544] Synchronise the ghostzones for a list of groups of grid variables (identified by their
group indices)
CCTK_GroupIndex [A236] Gets the group index for a given group name.
CCTK_GroupIndexFromVar [A240] Gets the group index for a given variable name.
Errors
-1 group_name was invalid.
-2 The driver returned an error on syncing the group.
Examples
#include "cctk.h" #include "cctk_Arguments.h" /* this function synchronizes the ADM metric */ void synchronize_ADM_metric(CCTK_ARGUMENTS) { DECLARE_CCTK_ARGUMENTS /* defines "magic variable" cctkGH */ int group_index, status; group_index = CCTK_GroupIndex("ADMBase::metric"); if (group_index < 0) CCTK_VWarn(CCTK_WARN_ABORT, __LINE__, __FILE__, CCTK_THORNSTRING, "***** synchronize_ADM_metric():\n" " couldn’t get group index for ADM metric!\n" " (CCTK_GroupIndex() returned error code %d)\n" , group_index); /*NOTREACHED*/ status = CCTK_SyncGroupI(cctkGH, group_index); if (status < 0) CCTK_VWarn(CCTK_WARN_ABORT, __LINE__, __FILE__, CCTK_THORNSTRING, "***** synchronize_ADM_metric():\n" " failed to synchronize ADM metric!\n" " (CCTK_SyncGroupI() returned error code %d)\n" , status); /*NOTREACHED*/ }
Synchronise the ghostzones for a list of groups of grid variables (identified by their group indices)
Synopsis
#include "cctk.h" int status = CCTK_SyncGroupsI(const cGH* GH, int num_groups, const int *groups)
#include "cctk.h" integer status CCTK_POINTER GH integer num_groups integer groups(num_groups) call CCTK_SyncGroupsI(status, GH, num_groups, groups)
Result
0 Returns the number of groups that have been synchronised.
Parameters
GH A pointer to a Cactus grid hierarchy.
num_groups The number of groups to be synchronised.
groups The group indices of the groups to be synchronized.
Discussion
Only those grid variables which have communication enabled will be synchronised. This is usually equivalent to the variables which have storage assigned, unless communication has been explicitly turned off with a call to CCTK_DisableGroupComm.
Note that an alternative to calling CCTK_SyncGroupsI explicitly from within a thorn, is to use the SYNC keyword in a thorns schedule.ccl file to indicate which groups of variables need to be synchronised on exit from the routine. This latter method is the preferred method from synchronising variables.
Note that CCTK_SyncGroupsI is a collective operation, so in the multiprocessor case you must call this function in parallel on each processor, passing the same number of groups in the same order.
See Also
CCTK_SyncGroup [A535] Synchronise the ghostzones for a single group of grid variables (identified by the group
name)
CCTK_SyncGroupI [A539] Synchronise the ghostzones for a single group of grid variables (identified by the group
index)
CCTK_GroupIndex [A236] Gets the group index for a given group name.
CCTK_GroupIndexFromVar [A240] Gets the group index for a given variable name.
Examples
#include "cctk.h" #include "cctk_Arguments.h" /* this function synchronizes the ADM metric and lapse */ void synchronize_ADM_metric_and_lapse(CCTK_ARGUMENTS) { DECLARE_CCTK_ARGUMENTS /* defines "magic variable" cctkGH */ int group_indices[2], status; group_indices[0] = CCTK_GroupIndex("ADMBase::metric"); group_indices[1] = CCTK_GroupIndex("ADMBase::lapse"); if (group_indices[0] < 0) CCTK_VWarn(CCTK_WARN_ABORT, __LINE__, __FILE__, CCTK_THORNSTRING, "***** synchronize_ADM_metric():\n" " couldn’t get group index for ADM metric!\n" " (CCTK_GroupIndex() returned error code %d)\n" , group_indices[0]); /*NOTREACHED*/ if (group_indices[1] < 0) CCTK_VWarn(CCTK_WARN_ABORT, __LINE__, __FILE__, CCTK_THORNSTRING, "***** synchronize_ADM_metric_and_lapse():\n" " couldn’t get group index for ADM lapse!\n" " (CCTK_GroupIndex() returned error code %d)\n" , group_indices[1]); /*NOTREACHED*/ status = CCTK_SyncGroupsI(cctkGH, 2, group_indices); if (status != 2) CCTK_VWarn(CCTK_WARN_ABORT, __LINE__, __FILE__, CCTK_THORNSTRING, "***** synchronize_ADM_metric_and_lapse():\n" " failed to synchronize ADM metric and lapse!\n" " (CCTK_SyncGroupsI() returned error code %d)\n" , status); /*NOTREACHED*/ }
Causes a Cactus simulation to terminate after present iteration finishes
Synopsis
#include "cctk.h" void CCTK_TerminateNext (const cGH *cctkGH)
#include "cctk.h" call CCTK_TerminateNext (cctkGH) CCTK_POINTER_TO_CONST cctkGH
Parameters
cctkGH Pointer to a Cactus grid hierarchy.
Discussion
This function triggers unconditional termination of Cactus after the present iteration. It bypasses all other termination conditions specified in the Cactus::terminate keyword parameter.
At this time, the cctkGH parameter does nothing.
See Also
CCTK_TerminationReached [A551] Returns true if CCTK_TerminateNext has been called.
Returns true if CCTK_TerminateNext has been called.
Synopsis
#include "cctk.h" void CCTK_TerminationReached (const cGH *cctkGH)
#include "cctk.h" call CCTK_TerminationReached (cctkGH) CCTK_POINTER_TO_CONST cctkGH
Parameters
cctkGH Pointer to a Cactus grid hierarchy.
Discussion
Returns true if Cactus has been requested to terminate after the present iteration by the CCTK_TerminateNext function.
At this time, the cctkGH parameter does nothing.
See Also
CCTK_TerminateNext [A549] Causes a Cactus simulation to terminate after the present iteration.
Returns the implementation provided by the thorn.
Synopsis
#include "cctk.h" const char *imp = CCTK_ThornImplementationThorn(const char *name);
Result
imp Name of the implementation or NULL
Parameters
name Name of the thorn
See Also
CCTK_ActivatingThorn [A32] Finds the thorn which activated a particular implementation
CCTK_CompiledImplementation [A79] Return the name of the compiled implementation with given index
CCTK_CompiledThorn [A81] Return the name of the compiled thorn with given index
CCTK_ImplementationRequires [A302] Return the ancestors for an implementation
CCTK_ImplementationThorn [A304] Returns the name of one thorn providing an implementation.
CCTK_ImpThornList [A306] Return the thorns for an implementation
CCTK_IsImplementationActive [A346] Reports whether an implementation was activated in a parameter file
CCTK_IsImplementationCompiled [A348] Reports whether an implementation was compiled into a configuration
CCTK_IsThornActive [A350] Reports whether a thorn was activated in a parameter file
CCTK_IsThornCompiled [A353] Reports whether a thorn was compiled into a configuration
CCTK_NumCompiledImplementations [A378] Return the number of implementations compiled in
CCTK_NumCompiledThorns [A380] Return the number of thorns compiled in
Errors
NULL Error.
Fills a cTimerData structure with timer clock info, for the timer specified by name.
Synopsis
int err = CCTK_Timer(name,info)
Parameters
const char * name Timer name
cTimerData * info Timer clock info pointer
Errors
A negative return value indicates an error.
Creates a timer with a given name, returns an index to the timer.
Synopsis
int index = CCTK_TimerCreate(name)
Parameters
const char * name timer name
Errors
< 0 A negative return value indicates an error.
Allocates the cTimerData structure, which is used to store timer clock info.
Synopsis
cTimerData * info = CCTK_TimerCreateData()
Errors
NULL A null return value indicates an error.
Creates an unnamed timer, returns an index to the timer.
Synopsis
int index = CCTK_TimerCreate()
Errors
< 0 A negative return value indicates an error.
Reclaims resources used by the given timer, specified by name.
Synopsis
int err = CCTK_TimerDestroy(name)
Parameters
const char * name timer name
Errors
< 0 A negative return value indicates an error.
Releases resources from the cTimerData structure, created by CCTK_TimerCreateData.
Synopsis
int err = CCTK_TimerDestroyData(info)
Parameters
cTimerData * info Timer clock info pointer
Errors
< 0 A negative return value indicates an error.
Reclaims resources used by the given timer, specified by index.
Synopsis
int err = CCTK_TimerDestroyI(index)
Parameters
int index timer index
Errors
< 0 A negative return value indicates an error.
Fills a cTimerData structure with timer clock info, for the timer specified by index.
Synopsis
int err = CCTK_TimerI(index,info)
Parameters
int index Timer index
cTimerData * info Timer clock info pointer
Errors
< 0 A negative return value indicates an error.
Gets values from all the clocks in the given timer, specified by name.
Synopsis
int err = CCTK_TimerReset(name)
Parameters
const char * name timer name
Errors
< 0 A negative return value indicates an error.
Gets values from all the clocks in the given timer, specified by index.
Synopsis
int err = CCTK_TimerResetI(index)
Parameters
int index timer index
Errors
< 0 A negative return value indicates an error.
Initialises all the clocks in the given timer, specified by name.
Synopsis
int err = CCTK_TimerStart(name)
Parameters
const char * name timer name
Errors
< 0 A negative return value indicates an error.
Initialises all the clocks in the given timer, specified by index.
Synopsis
int err = CCTK_TimerStartI(index)
Parameters
int index timer index
Errors
< 0 A negative return value indicates an error.
Gets values from all the clocks in the given timer, specified by name.
Synopsis
int err = CCTK_TimerStop(name)
Parameters
int name timer name
Discussion
Call this before getting the values from any of the timer’s clocks.
Errors
< 0 A negative return value indicates an error.
Gets values from all the clocks in the given timer, specified by index.
Synopsis
int err = CCTK_TimerStopI(index)
Parameters
int index timer index
Discussion
Call this before getting the values from any of the timer’s clocks.
Errors
< 0 A negative return value indicates an error.
Checks if a Cactus timer is running, given its name. Returns 0 of not (or in case of errors) and 1 if the timer is running.
Synopsis
int err = CCTK_TimerIsRunning(name)
call CCTK_TimerIsRunning(isrunning, name) integer isrunning character*(*) name
Parameters
char* name timer name
Discussion
Errors are treated as non-running timers: 0 is returned.
Checks if a Cactus timer is running, given its handle. Returns 0 of not (or in case of errors) and 1 if the timer is running.
Synopsis
int err = CCTK_TimerIsRunningI(index)
call CCTK_TimerIsRunningI(isrunning , index ) integer isrunning integer index
Parameters
int index timer index
Discussion
Errors are treated as non-running timers: 0 is returned.
Traverse through all variables and/or groups whose names appear in the given string, and call the callback routine with those indices and an optional option string appended to the variable/group name enclosed in square braces. The special keyword ”all” in the string can be used to indicate that the callback should be called for all variables/groups.
Synopsis
int err = CCTK_TraverseString(traverse_string, callback, callback_arg, selection)
Parameters
const char * traverse_string List of variable and/or group names
void (*callback) (int idx, const char *optstring, void *callback_arg) Routine to call for every
variable and/or group found. idx is the Cactus variable index, optstring is the optional ‘{}’ enclosed option
string after the variable name, and callback_arg is the arbitrary argument passed to CCTK_TraverseString.
void *callback_arg An arbitrary argument which gets passed to the callback routine
int selection Decides whether group and/or variable names are accepted in the string. Possible values are:
CCTK_VAR, CCTK_GROUP or CCTK_GROUP_OR_VAR.
Discussion
Use this to loop over a list of variables passed in by the user.
Result
number of variables positive for the number of traversed variables
Errors
-1 no callback routine was given
-2 option string is not associated with a group or variable
-3 unterminated option string
-4 garbage found at end of option string
-5 invalid token in traversed string found
Returns the data pointer for a grid variable
Synopsis
void * ptr = CCTK_VarDataPtr( const cGH * cctkGH, int timelevel, char * name)
call CCTK_VarDataPtr(ptr, cctkGH, timelevel, varname) CCTK_POINTER vardataptr CCTK_POINTER_TO_CONST cctkGH integer timelevel character*(*) varname
Parameters
ptr a void pointer to the grid variable data
cctkGH pointer to CCTK grid hierarchy
timelevel The timelevel of the grid variable
name The full name of the variable
Discussion
The variable name should be in the form <implementation>::<variable>.
Examples
myVar = (CCTK_REAL *)(CCTK_VarDataPtr(GH,0,"imp::realvar"))
CCTK_REAL, dimension(cctk_ash(1),cctk_ash(2),cctk_ash(3)) :: var CCTK_POINTER myVar pointer (myVar, var) call CCTK_VarDataPtr(myVar,GH,0,"imp::realvar")
Returns the data pointer for a grid variable from the variable index or the variable name
Synopsis
void * ptr = CCTK_VarDataPtrB( const cGH * cctkGH, int timelevel, int index, char * name)
Parameters
ptr a void pointer to the grid variable data
cctkGH pointer to CCTK grid hierarchy
timelevel The timelevel of the grid variable
index The index of the variable
name The full name of the variable
Discussion
If the name is NULL the index will be used, if the index is negative the name will be used.
Examples
myVar = (CCTK_REAL *)(CCTK_VarDataPtrB(GH,0,CCTK_VarIndex("imp::realvar"),NULL));
Returns the data pointer for a grid variable from the variable index
Synopsis
void * ptr = CCTK_VarDataPtrI( const cGH * cctkGH, int timelevel, int index)
call CCTK_VarDataPtrI(ptr, cctkGH, timelevel, index) CCTK_POINTER vardataptr CCTK_POINTER_TO_CONST cctkGH integer timelevel integer index
Parameters
cctkGH pointer to CCTK grid hierarchy
timelevel The timelevel of the grid variable
index The index of the variable
Examples
myVar = (CCTK_REAL *)(CCTK_VarDataPtr(GH,0,CCTK_VarIndex("imp::realvar")));
CCTK_REAL, dimension(cctk_ash(1),cctk_ash(2),cctk_ash(3)) :: var CCTK_POINTER myVar pointer (myVar, var) call CCTK_VarDataPtr(myVar,GH,0,CCTK_VarIndex("imp::realvar"))
Get the index for a variable.
Synopsis
#include "cctk.h" int index = CCTK_VarIndex(const char *varname);
call CCTK_VarIndex(index, varname) integer index character*(*) varname
Parameters
varname The name of the variable.
Discussion
The variable name should be the given in its fully qualified form, that is <implementation>::<variable> for a public or protected variable, and <thornname>::<variable> for a private variable. For vector variables, the zero-based component index should be included in square brackets after the variable name.
Errors
-1 no variable of this name exists
-2 failed to catch error code from Util_SplitString
-3 given full name is in wrong format
-4 memory allocation failed
Examples
index = CCTK_VarIndex("evolve::phi"); index = CCTK_VarIndex("evolve::vect[0]");
call CCTK_VarIndex(index,"evolve::phi") call CCTK_VarIndex(index,"evolve::vect[0]")
Given a variable index, returns the variable name
Synopsis
const char * name = CCTK_VarName( int index)
#include "cctk.h" subroutine CCTK_VarName(nchars, index, fullname) integer nchars integer index character*(*) name end subroutine CCTK_VarName
Parameters
name The variable name
index The variable index
Discussion
The pointer returned is part of a structure managed by Cactus and so must not be freed after use.
Examples
index = CCTK_VarIndex("evolve::phi"); name = CCTK_VarName(index);
Provides variable type index from the variable index
Synopsis
int type = CCTK_VarTypeI( int index)
call CCTK_VarTypeI(type , index ) integer type integer index
Parameters
type The variable type index
index The variable index
Discussion
The variable type index indicates the type of the variable. Either character, int, complex or real. The group type can be checked with the Cactus provided macros for CCTK_VARIABLE_INT, CCTK_VARIABLE_REAL, CCTK_VARIABLE_COMPLEX or CCTK_VARIABLE_CHAR.
Examples
index = CCTK_VarIndex("evolve::phi") real = (CCTK_VARIABLE_REAL == CCTK_VarTypeI(index)) ;
call CCTK_VARTYPEI(type,3)
Provides variable type size in bytes from the variable type index
Synopsis
#include "cctk.h" int CCTK_VarTypeSize(int vtype);
#include "cctk.h" CCTK_INT size, vtype call CCTK_VarTypeSize(size, vtype);
Parameters
vtype Variable type index.
Discussion
Given a CCTK_VARIABLE_* type code (e.g. CCTK_VARIABLE_INT, CCTK_VARIABLE_REAL, CCTK_VARIABLE_COMPLEX, etc.), this function returns the size in bytes of the corresponding data type (CCTK_INT, CCTK_REAL, CCTK_COMPLEX, etc.).
Errors
-1 vtype is not one of the CCTK_VARIABLE_* values.
Given a set of vector and multidimensional indices compute the 1-dimensional index into a vector grid function.
Synopsis
int CCTK_VECTGFINDEX1D(const cGH *restrict cctkGH, int i, int n)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
int n The vector index
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (i=0; i<cctk_lsh[0]; i++) { /* vector indices are 0, 1, 2 */ vel[CCTK_VECTGFINDEX1D(cctkGH,i,0)] = 1.0; vel[CCTK_VECTGFINDEX1D(cctkGH,i,1)] = 0.0; vel[CCTK_VECTGFINDEX1D(cctkGH,i,2)] = 0.0; }
See Also
CCTK_GFINDEX1D() Given a set of multidimensional indices compute the 1-dimensional index into a grid
function.
Given a set of vector and multidimensional indices compute the 2-dimensional index into a vector grid function.
Synopsis
int CCTK_VECTGFINDEX2D(const cGH *restrict cctkGH, int i, int j, int n)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
int j Index in the j direction
int n The vector index
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (j=0; j<cctk_lsh[1]; j++) { for (i=0; i<cctk_lsh[0]; i++) { /* vector indices are 0, 1, 2 */ vel[CCTK_VECTGFINDEX2D(cctkGH,i,j,0)] = 1.0; vel[CCTK_VECTGFINDEX2D(cctkGH,i,j,1)] = 0.0; vel[CCTK_VECTGFINDEX2D(cctkGH,i,j,2)] = 0.0; } }
See Also
CCTK_GFINDEX2D() Given a set of multidimensional indices compute the 2-dimensional index into a grid
function.
Given a set of vector and multidimensional indices compute the 3-dimensional index into a vector grid function.
Synopsis
int CCTK_VECTGFINDEX3D(const cGH *restrict cctkGH, int i, int j, int k, int n)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
int j Index in the j direction
int n The vector index
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (k=0; k<cctk_lsh[2]; k++) { for (j=0; j<cctk_lsh[1]; j++) { for (i=0; i<cctk_lsh[0]; i++) { /* vector indices are 0, 1, 2 */ vel[CCTK_VECTGFINDEX3D(cctkGH,i,j,k,0)] = 1.0; vel[CCTK_VECTGFINDEX3D(cctkGH,i,j,k,1)] = 0.0; vel[CCTK_VECTGFINDEX3D(cctkGH,i,j,k,2)] = 0.0; } } }
See Also
CCTK_GFINDEX3D() Given a set of multidimensional indices compute the 3-dimensional index into a grid
function.
Given a set of vector and multidimensional indices compute the 4-dimensional index into a vector grid function.
Synopsis
int CCTK_VECTGFINDEX4D(const cGH *restrict cctkGH, int i, int j, int k, int l, int n)
Parameters
const cGH *restrict cctkGH The pointer to the CCTK grid hierarchy
int i Index in the i direction
int j Index in the j direction
int k Index in the k direction
int l Index in the l direction
int n The vector index
Discussion
Grid functions are held in memory as 1-dimensional C arrays. These are laid out in memory as in Fortran. Cactus provides macros to find the 1-dimensional index which is needed from the multidimensional indices which are usually used. In Fortran, grid functions are accessed as Fortran arrays.
Examples
for (l=0; l<cctk_lsh[3]; l++) { for (k=0; k<cctk_lsh[2]; k++) { for (j=0; j<cctk_lsh[1]; j++) { for (i=0; i<cctk_lsh[0]; i++) { /* vector indices are 0, 1, 2 */ vel[CCTK_VECTGFINDEX4D(cctkGH,i,j,k,l,0)] = 1.0; vel[CCTK_VECTGFINDEX4D(cctkGH,i,j,k,l,1)] = 0.0; vel[CCTK_VECTGFINDEX4D(cctkGH,i,j,k,l,2)] = 0.0; } } } }
See Also
CCTK_GFINDEX4D() Given a set of multidimensional indices compute the 4-dimensional index into a grid
function.
Prints a formatted string with a variable argument list as error message and stops the code
Synopsis
#include <cctk.h> void CCTK_VERROR(const char *format, ...);
Parameters
format The printf-like format string to use for printing the warning message.
... The printf-style variable argument list.
Discussion
This routine can be used by thorns to print a formatted string followed by a variable argument list as error message to stderr. After printing the message, Cactus aborts the run (and CCTK_VERROR does not return to the caller).
This macro can be used by thorns to print an info message to screen.
The macro CCTK_VERROR(...) expands to a call to the underlying function CCTK_VError:
CCTK_VError(CCTK_THORNSTRING, ...)
So the macro automatically includes the name of the originating thorn in the info message. It is recommended that the macro CCTK_VERROR is used to print a message rather than calling CCTK_VError directly.
See Also
CCTK_Abort [A27] Abort the code
CCTK_ERROR [A158] macro to print an error message with a single string argument
CCTK_Exit [A166] Exit the code cleanly
CCTK_VWARN [A626] Possibly prints a formatted string with a variable argument list as warning message and/or
stops the code
CCTK_WARN [A634] macro to print a warning message with a single string argument
Examples
#include <cctk.h> const char *outdir; CCTK_VERROR("Output directory ’%s’ could not be created", outdir);
Prints a formatted string with a variable argument list as error message and stops the code
Synopsis
#include <cctk.h> void CCTK_VError(int line, const char *file, const char *thorn, const char *format, ...);
Parameters
line The line number in the originating source file where the CCTK_VError call occured. You can use the
standardized __LINE__ preprocessor macro here.
file The file name of the originating source file where the CCTK_VError call occured. You can use the
standardized __FILE__ preprocessor macro here.
thorn The thorn name of the originating source file where the CCTK_VError call occured. You can use the
CCTK_THORNSTRING macro here (defined in cctk.h).
format The printf-like format string to use for printing the warning message.
... The variable argument list.
Discussion
This routine can be used by thorns to print a formatted string followed by a variable argument list as error message to stderr. After printing the message, Cactus aborts the run (and CCTK_VError does not return to the caller).
See Also
CCTK_Abort [A27] Abort the code
CCTK_ERROR [A158] macro to print an error message with a single string argument
CCTK_Exit [A166] Exit the code cleanly
CCTK_VWarn [A630] Possibly prints a formatted string with a variable argument list as warning message and/or
stops the code
CCTK_WARN [A634] macro to print a warning message with a single string argument
Examples
#include <cctk.h> const char *outdir; CCTK_VError(__LINE__, __FILE__, CCTK_THORNSTRING, "Output directory ’%s’ could not be created", outdir);
Macro to print a printf-style variable argument list an information message to screen
Synopsis
#include <cctk.h> CCTK_VINFO(const char *format, ...);
Result
0 ok
Parameters
format The printf-like format string to use for printing the info message.
... The variable argument list.
Discussion
This macro can be used by thorns to print an info message to screen.
The macro CCTK_VINFO(...) expands to a call to the underlying function CCTK_VInfo:
CCTK_VInfo(CCTK_THORNSTRING, ...)
So the macro automatically includes the name of the originating thorn in the info message. It is recommended that the macro CCTK_VINFO is used to print a message rather than calling CCTK_VInfo directly.
See Also
CCTK_ERROR [A158] macro to print an error message with a single string argument and stop the
code CCTK_INFO() [A308] macro to print an info message with a single string argument to screen
CCTK_VERROR [A610] macro to print a formatted string with a variable argument list as error message and stops
the code
CCTK_VWARN [A626] macro to print a warning message with a variable argument list
CCTK_WARN [A634] macro to print a warning message with a single string argument and possibly stop the code
Examples
#include <cctk.h> const char *outdir; CCTK_VINFO("Output files will go to ’%s’", outdir);
Prints a formatted string with a variable argument list as an info message to sceen
Synopsis
#include <cctk.h> int status = CCTK_VInfo(const char *thorn, const char *format, ...);
Result
0 ok
Parameters
thorn The name of the thorn printing this info message. You can use the CCTK_THORNSTRING macro here
(defined in cctk.h).
format The printf-like format string to use for printing the info message.
... The variable argument list.
Discussion
This routine can be used by thorns to print a formatted string with a variable argument list as an info message to screen. The message will include the name of the originating thorn, otherwise its semantics is equivalent to printf.
See Also
CCTK_INFO [A308] macro to print an info message with a single string argument
CCTK_ERROR [A158] macro to print an error message with a single string argument and stop the code
CCTK_VError [A613] prints a formatted string with a variable argument list as error message and stops the code
CCTK_VWarn [A630] prints a warning message with a variable argument list
CCTK_WARN [A634] macro to print a warning message with a single string argument and possibly stop the code
Examples
#include "cctk.h" const char *outdir; CCTK_VInfo(CCTK_THORNSTRING, "Output files will go to ’%s’", outdir);
Prints a formatted string with a variable argument list as a warning from parameter checking, and possibly stops the code
Synopsis
#include <cctk.h> int status = CCTK_VParamWarn(const char *thorn, const char *format, ...);
Parameters
thorn Name of originating thorn
format Format for variable argument list
… Variable argument list
Discussion
The call should be used in routines registered at the schedule point CCTK_PARAMCHECK to indicate that there is parameter error or conflict and the code should terminate. The code will terminate only after all the parameters have been checked.
Examples
CCTK_VParamWarn(CCTK_THORNSTRING, "Mass cannot be negative: %g", (double)mass);
See Also
CCTK_PARAMWARN [A450] Prints a warning from parameter checking, and possibly stops the code
CCTK_VPARAMWARN [A624] Prints a formatted string with a variable argument list as a warning from parameter
checking, and possibly stops the code
Prints a formatted string with a variable argument list as a warning from parameter checking, and possibly stops the code
Synopsis
#include <cctk.h> int status = CCTK_VPARAMWARN(const char *format, ...);
Parameters
format Format for variable argument list
… Variable argument list
Discussion
The call should be used in routines registered at the schedule point CCTK_PARAMCHECK to indicate that there is parameter error or conflict and the code should terminate. The code will terminate only after all the parameters have been checked.
Examples
CCTK_VPARAMWARN("Mass cannot be negative: %g", (double)mass);
See Also
CCTK_PARAMWARN [A450] Prints a warning from parameter checking, and possibly stops the code
CCTK_VParamWarn [A622] Prints a formatted string with a variable argument list as a warning from parameter
checking, and possibly stops the code
Possibly prints a formatted string with a variable argument list as warning message and/or stops the code
Synopsis
#include <cctk.h> int status = CCTK_VWARN(int level, const char *format, ...);
Result
0 ok
Parameters
level (\(\ge 0\)) The warning level for the message to print, with level 0 being the severest level and greater levels
being less severe.
format The printf-like format string to use for printing the warning message.
... The printf-style variable argument list.
Discussion
This routine can be used by thorns to print a formatted string followed by a variable argument list as a warning message to stderr. If the message’s “warning level” is severe enough, then after printing the message Cactus aborts the run (and CCTK_VWARN does not return to the caller).
CCTK_VWARN(level, ...) expands to a call to CCTK_Warn() function which is equivalent to CCTK_VWarn(). The macro automatically includes details about the origin of the warning (the thorn name, the source code file name and the line number where the macro occurs).
Cactus’s behavior when CCTK_VWARN is called depends on the -W and -E command-line options:
Cactus prints any warning with a warning level \(\le \) the -W level to standard error (any warnings with warning levels \(>\) the -W level are silently discarded). The default -W level is 1, i.e. only level 0 and level 1 warnings will be printed.
Cactus stops (aborts) the current run for any warning with a warning level \(\le \) the -E level. The default -W level is 0, i.e. only level 0 warnings will abort the run.
Cactus guarantees that \(\mbox {the \code {-W}~level} \ge \mbox {the \code {-E}~level} \ge 0\). This implies that a message will always be printed for any warning that’s severe enough to halt the Cactus run. It also implies that a level 0 warning is guaranteed (to be printed and) to halt the Cactus run.
The severity level may actually be any integer, and a lot of existing code uses bare “magic number” integers for warning levels, but to help standardize warning levels across thorns, new code should probably use one of the following macros, defined in "cctk_WarnLevel.h" (which is #included by "cctk.h"):
#define CCTK_WARN_ABORT 0 /* abort the Cactus run */ #define CCTK_WARN_ALERT 1 /* the results of this run will probably */ /* be wrong, but this isn’t quite certain, */ /* so we’re not going to abort the run */ #define CCTK_WARN_COMPLAIN 2 /* the user should know about this, but */ /* the results of this run are probably ok */ #define CCTK_WARN_PICKY 3 /* this is for small problems that can */ /* probably be ignored, but that careful */ /* people may want to know about */ #define CCTK_WARN_DEBUG 4 /* these messages are probably useful */ /* only for debugging purposes */
For example, to provide a warning for a serious problem, which indicates that the results of the run are quite likely wrong, and which will be printed to the screen by default, a level CCTK_WARN_ALERT warning should be used.
In any case, the Boolean flesh parameter cctk_full_warnings determines whether all the details about the warning origin (processor ID, line number, source file, source thorn) are shown. The default is to print everything.
See Also
CCTK_Abort [A27] Abort the code
CCTK_ERROR [A158] macro to print an error message with a single string argument and stop the code
CCTK_Exit [A166] Exit the code cleanly
CCTK_INFO [A308] macro to print an info message with a single string argument
CCTK_VINFO() [A616] prints a formatted string with a variable argument list as an info message to screen
CCTK_VERROR [A610] prints a formatted string with a variable argument list as error message and stops the code
CCTK_WARN [A634] macro to print a warning message with a single string argument
Examples
#include <cctk.h> const char *outdir; CCTK_VWARN(CCTK_WARN_ALERT, "Output directory ’%s’ could not be created", outdir);
Possibly prints a formatted string with a variable argument list as warning message and/or stops the code
Synopsis
#include <cctk.h> int status = CCTK_VWarn(int level, int line, const char *file, const char *thorn, const char *format, ...);
Result
0 ok5
Parameters
level (\(\ge 0\)) The warning level for the message to print, with level 0 being the severest level and greater levels
being less severe.
line The line number in the originating source file where the CCTK_VWarn call occured. You can use the
standardized __LINE__ preprocessor macro here.
file The file name of the originating source file where the CCTK_VWarn call occured. You can use the
standardized __FILE__ preprocessor macro here.
thorn The thorn name of the originating source file where the CCTK_VWarn call occured. You can use the
CCTK_THORNSTRING macro here (defined in cctk.h).
format The printf-like format string to use for printing the warning message.
... The variable argument list.
Discussion
This routine can be used by thorns to print a formatted string followed by a variable argument list as a warning message to stderr. If the message’s “warning level” is severe enough, then after printing the message Cactus aborts the run (and CCTK_VWarn does not return to the caller).
Cactus’s behavior when CCTK_VWarn is called depends on the -W and -E command-line options:
Cactus prints any warning with a warning level \(\le \) the -W level to standard error (any warnings with warning levels \(>\) the -W level are silently discarded). The default -W level is 1, i.e. only level 0 and level 1 warnings will be printed.
Cactus stops (aborts) the current run for any warning with a warning level \(\le \) the -E level. The default -W level is 0, i.e. only level 0 warnings will abort the run.
Cactus guarantees that \(\mbox {the \code {-W}~level} \ge \mbox {the \code {-E}~level} \ge 0\). This implies that a message will always be printed for any warning that’s severe enough to halt the Cactus run. It also implies that a level 0 warning is guaranteed (to be printed and) to halt the Cactus run.
The severity level may actually be any integer, and a lot of existing code uses bare “magic number” integers for warning levels, but to help standardize warning levels across thorns, new code should probably use one of the following macros, defined in "cctk_WarnLevel.h" (which is #included by "cctk.h"):
#define CCTK_WARN_ABORT 0 /* abort the Cactus run */ #define CCTK_WARN_ALERT 1 /* the results of this run will probably */ /* be wrong, but this isn’t quite certain, */ /* so we’re not going to abort the run */ #define CCTK_WARN_COMPLAIN 2 /* the user should know about this, but */ /* the results of this run are probably ok */ #define CCTK_WARN_PICKY 3 /* this is for small problems that can */ /* probably be ignored, but that careful */ /* people may want to know about */ #define CCTK_WARN_DEBUG 4 /* these messages are probably useful */ /* only for debugging purposes */
For example, to provide a warning for a serious problem, which indicates that the results of the run are quite likely wrong, and which will be printed to the screen by default, a level CCTK_WARN_ALERT warning should be used.
In any case, the Boolean flesh parameter cctk_full_warnings determines whether all the details about the warning origin (processor ID, line number, source file, source thorn) are shown. The default is to print everything.
See Also
CCTK_Abort [A27] Abort the code
CCTK_ERROR [A158] macro to print an error message with a single string argument and stop the code
CCTK_Exit [A166] Exit the code cleanly
CCTK_INFO [A308] macro to print an info message with a single string argument
CCTK_VINFO() [A616] prints a formatted string with a variable argument list as an info message to screen
CCTK_VERROR [A610] prints a formatted string with a variable argument list as error message and stops the code
CCTK_WARN [A634] macro to print a warning message with a single string argument
Examples
#include <cctk.h> const char *outdir; CCTK_VWARN(CCTK_WARN_ALERT, "Output directory ’%s’ could not be created", outdir);
Macro to print a single string as a warning message and possibly stop the code
Synopsis
#include <cctk.h> CCTK_WARN(int level, const char *message);
#include "cctk.h" call CCTK_WARN(level, message) integer level character*(*) message
Parameters
level The warning level to use; see the description of CCTK_VWarn() on page A619 for a detailed discussion of
this parameter and the Cactus macros for standard warning levels
message The warning message to print
Discussion
This macro can be used by thorns to print a single string as a warning message to stderr.
CCTK_WARN(level, message) expands to a call to CCTK_Warn() function which is equivalent to CCTK_VWarn(), but without the variable-number-of-arguments feature (so it can be used from Fortran).6 The macro automatically includes details about the origin of the warning (the thorn name, the source code file name and the line number where the macro occurs).
To include variables in a warning message from C, you can use the routine CCTK_VWarn which accepts a variable argument list. To include variables from Fortran, a string must be constructed and passed in a CCTK_WARN macro.
See Also
CCTK_Abort [A27] Abort the code
CCTK_ERROR [A158] macro to print an error message with a single string argument and stop the code
CCTK_Exit [A166] Exit the code cleanly
CCTK_ERROR [A158] macro to print an error message an abort the code
CCTK_INFO [A308] macro to print an info message with a single string argument
CCTK_VERROR [A610] macro to print a formatted string with a variable argument list as error message and stops
the code
CCTK_VINFO [A616] macro to print a formatted string with a variable argument list as an info message to screen
CCTK_VWARN [A626] macro to print a warning message with a variable argument list
Examples
#include "cctk.h" CCTK_WARN(CCTK_WARN_ABORT, "Divide by 0");
#include "cctk.h" integer myint real myreal character*200 message write(message, ’(A32, G12.7, A5, I8)’) & ’Your warning message, including ’, myreal, ’ and ’, myint call CCTK_WARN(CCTK_WARN_ALERT, message)
Function to print a single string as error message and possibly stop the code
Synopsis
#include <cctk.h> void CCTK_Warn(int level, int line_number, const char* file_name, const char* thorn_name,const char* message)
#include "cctk.h" call CCTK_Warn(level, line_number, file_name, thorn_name, message) integer level, line_number character*(*) file_name, thorn_name, message
Parameters
level (\(\ge 0\)) The warning level for the message to print, with level 0 being the severest level and greater levels
being less severe.
line_number The line number in the originating source file where the CCTK_VWarn call occured. You can use the
standardized __LINE__ preprocessor macro here.
file_name The file name of the originating source file where the CCTK_VWarn call occured. You can use the
standardized __FILE__ preprocessor macro here.
thorn_name The thorn name of the originating source file where the CCTK_VWarn call occured. You can use the
CCTK_THORNSTRING macro here (defined in cctk.h).
message The error message to print
Discussion
The macro CCTK_WARN automatically includes the line number, file name and the name of the originating thorn in the info message. It is recommended that the macro CCTK_WARN is used to print a message rather than calling CCTK_Warn directly.
See Also
CCTK_Abort [A27] Abort the code
CCTK_Exit [A166] Exit the code cleanly
CCTK_INFO [A308] macro to print an info message
CCTK_VERROR [A610] macro to print an error message with a variable argument list
CCTK_VINFO [A616] macro to print an info message with a variable argument list
CCTK_VWARN [A626] macro to print a formatted string with a variable argument list as a warning message to
standard error and possibly stops the code
CCTK_WARN [A634] Macro to print a single string as a warning message and possibly stop the code
Register one or more routines for dealing with warning messages in addition to printing them to standard error
Synopsis
#include <cctk.h> CCTK_WarnCallbackRegister(int minlevel, int maxlevel, void *data, cctk_warnfunc callback);
Parameters
minlevel The minimum warning level to use.
You can find a detailed discussion of the Cactus macros for standard warning levels on page A619. Both
minlevel and maxlevel follow that definition.
maxlevel The maximum warning level to use
data The void pointer holding extra information about the registered call back routine
callback The function pointer pointing to the call back function dealing with warning messages. The definition of the function pointer is:
typedef void (*cctk_warnfunc)(int level, int line, const char *file, const char *thorn, const char *message, void *data);
The argument list is the same as those in CCTK_Warn() (see the footnote of CCTK_WARN() page A634) except an
extra void pointer to hold the information about the call back routine.
Discussion
This function can be used by thorns to register their own routines to deal with warning messages. The registered function pointers will be stored in a pointer chain. When CCTK_VWarn() is called, the registered routines will be called in the same order as they get registered in addition to dumping warning messages to stderr.
The function can only be called in C.
See Also
CCTK_InfoCallbackRegister() Register one or more routines for dealing with information messages in
addition to printing them to screen
CCTK_VWarn() Prints a formatted string with a variable argument list as a warning message to standard error
and possibly stops the code
Examples
/*DumpWarn will dump warning messages to a file*/ void DumpWarn(int level, int line, const char *file, const char *thorn, const char *message, void *data) { DECLARE_CCTK_PARAMETERS FILE *fp; char *str = (char *)malloc((strlen(file)+strlen(thorn)+strlen(message)+100); /*warn_dump_file is a string set in the parameter file*/ if((fp = fopen (warn_dump_file, "a"))==0) { fprintf(stderr, "fatal error: can not open the file %s\n",warn_dump_file); return; } sprintf(str, "\n[WARN]\nLevel->%d\nLine->%d\nFile->%s\nThorn->%s\nMsg->%s\n", level,line,file,thorn,message); fprintf(fp, "%s", str); free(str); fclose(fp); } ... /*minlevel = 0; maxlevel = 5; data = NULL; callback = DumpWarn*/ CCTK_WarnCallbackRegister(0,5,NULL,DumpWarn);
In this chapter all Util_*() Cactus utility functions are described. These are low-level functions mainly for more complicated programming, which are used by the rest of Cactus, but don’t depend heavily on it. Some of them are callable from Fortran or C, but many are C-only.
Here the functions are listed alphabetically within each section.
[B10] Fills string with current local date
[B12] Returns the current datetime in a machine-processable format as defined in ISO 8601 chapter 5.4.
[B14] Fills string with current local time
[B16] Deprecated; use snprintf instead.
[B16] Deprecated; use vsnprintf instead.
[B16] Sprintf with memory allocation. On input the buffer should point to a NULL area of memory.
[B19] Compare two strings, ignoring upper/lower case.
[B25] Deprecated; use strdup instead.
[B25] Concatenate two strings safely.
[B30] Copy a string safely.
[B35] Separate first token from a string.
[B40] Create a new table which is a “clone” (exact copy) of an existing table
[B45] Create a new (empty) table
[B49] Create a new table (with the case-insensitive flag set) and sets values in it based on a string argument (interpreted with “parameter-file” semantics)
[B53] Delete a specified key/value entry from a table
[B56] Destroy a table
[B60] This is a family of functions, one for each Cactus data type, to get the single (1-element array) value, or more generally the first array element of the value, associated with a specified key in a key/value table.
[B64] This is a family of functions, one for each Cactus data type, to get a copy of the value associated with a specified key, and store it (more accurately, as much of it as will fit) in a specified array
[B68] Get the single (1-element array) value, or more generally the first array element of the value, associated with a specified key in a key/value table; the value’s data type is generic
[B72] Get a copy of the value associated with a specified key, and store it (more accurately, as much of it as will fit) in a specified array; the array’s data type is generic
[B76] Gets a copy of the character-string value associated with a specified key in a table, and stores it (more accurately, as much of it as will fit) in a specified character string
[B80] Advance a table iterator to the next entry in the table
[B82] Creates a new table iterator which is a “clone” (exact copy) of an existing table iterator
[B86] Create a new table iterator
[B89] Destroy a table iterator
[B91] Query whether a table iterator is not in the “null-pointer” state
[B94] Query whether a table iterator is in the “null-pointer” state
[B97] Query the key and the type and number of elements of the value corresponding to that key, of the table entry to which an iterator points
[B101] Query what table a table iterator iterates over
[B103] Reset a table iterator to point to the starting table entry
[B105] Set a key/value iterator to point to a specified entry in the table.
[B107] Set a key/value iterator to the “null-pointer” state.
[B109] Query a table’s flags word
[B113] Query whether or not a specified key is in the table, and optionally the type and/or number of elements of the value corresponding to this key
[B118] Query the maximum key length in a table
[B121] Query the number of key/value entries in a table
[B123] This is a family of functions, one for each Cactus data type, to set the value associated with a specified key to be a specified single (1-element array) value
[B127] This is a family of functions, one for each Cactus data type, to set the value associated with a specified key to be a copy of a specified array
[B131] Sets values in a table based on a string argument (interpreted with “parameter-file” semantics)
[B136] Set the value associated with a specified key to be a specified single (1-element array) value, whose data type is generic
[B140] Set the value associated with a specified key to be a copy of a specified array, whose data type is generic
[B145] Sets the value associated with a specified key in a table, to be a copy of a specified C-style null-terminated character string
[B149] Print out a table and its data structures, using a verbose internal format meant for debugging
[B152] Print out all tables and their data structures, using a verbose internal format meant for debugging
[B154] Print out all table iterators and their data structures, using a verbose internal format meant for debugging
[B156] Print out a table, using a human-readable format similar to the one accepted by Util_TableCreateFromString
Fills string with current local date
Synopsis
#include "cctk.h" #include "cctk_Misc.h.h" int retval = Util_CurrentDate (int len, char *now);
Parameters
len length of the user-supplied string buffer
now user-supplied string buffer to write the date stamp to
Result
retval length of the string returned in now, or 0 if the string was truncated
See Also
Util_CurrentTime [B14] Fills string with current local time
Util_CurrentDateTime [B12] Returns the current datetime in a machine-processable format as defined in ISO
8601 chapter 5.4.
Returns the current datetime in a machine-processable format as defined in ISO 8601 chapter 5.4.
Synopsis
#include "cctk.h" #include "cctk_Misc.h.h" char *current_datetime = Util_CurrentDateTime ();
Result
current_datetime Pointer to an allocated formatted string containing the current datetime stamp. The pointer
should be freed by the caller.
Discussion
The formatted string returned contains the current datetime in a machine-processable format as defined in ISO 8601 chapter 5.4: "YYYY-MM-DDThh:mm:ss+hh:mm”
See Also
Util_CurrentDate [B10] Fills string with current local date
Util_CurrentTime [B14] Fills string with current local time
Fills string with current local time
Synopsis
#include "cctk.h" #include "cctk_Misc.h.h" int retval = Util_CurrentTime (int len, char *now);
Parameters
len length of the user-supplied string buffer
now user-supplied string buffer to write the time stamp to
Result
retval length of the string returned in now, or 0 if the string was truncated
See Also
Util_CurrentDate [B10] Fills string with current local date
Util_CurrentDateTime [B12] Returns the current datetime in a machine-processable format as defined in ISO
8601 chapter 5.4.
Deprecated; use snprintf instead.
Deprecated; use vsnprintf instead.
Sprintf with memory allocation. On input the buffer should point to a NULL area of memory.
Synopsis
#include "util_String.h" int count = Util_asprintf(char** buffer, const char* format, ...);
Result
count Number of characters (excluding the terminating NUL) that have been placed in the allocated buffer.
Parameters
buffer Buffer to which to print the string.
*buffer should be NULL on entry. The routine allocates the memory, so the previous contents of the
pointer are lost. On exit the buffer size will be count+1 (i.e the length of the string plus the \0).
format A (non-NULL pointer to a) C-style NUL-terminated string describing how to format any further
arguments
... Zero or more further arguments, with types as specified by the format argument.
Discussion
This function is identical to sprintf(), except that it allocates a buffer large enough to hold the output including the terminating null byte, and returns a pointer to it via the first argument. This pointer should be passed to free() to release the allocated storage when it is no longer needed.
See Also
asprintf() GNU/BSD C library function which this function tries to clone.
sprintf() Unsafe and dangerous C library function similar to snprintf(), which doesn’t check the buffer
length.
Compare two strings, ignoring upper/lower case.
Synopsis
#include "util_String.h" int cmp = Util_StrCmpi(const char *str1, const char *str2);
Result
cmp An integer which is:
\(<0\) | if \(\code {str1} < \code {str2}\) in lexicographic order ignoring upper/lower case distinctions |
\(0\) | if \(\code {str1} = \code {str2}\) ignoring upper/lower case distinctions |
\(>0\) | if \(\code {str1} > \code {str2}\) in lexicographic order ignoring upper/lower case distinctions |
Parameters
str1 A non-NULL pointer to a (C-style NUL-terminated) string to be compared.
str2 A non-NULL pointer to a (C-style NUL-terminated) string to be compared.
Discussion
The standard C library strcmp() function does a case-sensitive string comparison, i.e. strcmp("cactus", "Cactus") will find the two strings not equal. Sometimes it’s useful to do case-insensitive string comparison, where upper/lower case distinctions are ignored. Many systems provide a strcasecmp() or strcmpi() function to do this, but some systems don’t, and even on those that do, the name isn’t standardised. So, Cactus provides its own version, Util_StrCmpi().
Notice that the return value of Util_StrCmpi(), like that of strcmp(), is zero (logical “false” in C) for equal strings, and nonzero (logical “true” in C) for non-equal strings. Code of the form
if (Util_StrCmpi(str1, str2)) { /* strings differ */ }
or
if (!Util_StrCmpi(str1, str2)) { /* strings are identical apart from case distinctions */ }
may be confusing to readers, because the sense of the comparison isn’t immediately obvious. Writing an explicit comparison against zero make make things clearer:
if (Util_StrCmpi(str1, str2) != 0) { /* strings differ */ }
or
if (Util_StrCmpi(str1, str2) == 0) { /* strings are identical apart from case distinctions */ }
Unfortunately, the basic concept of “case-insensitive” string operations doesn’t generalize well to non-English character sets,1 where lower-case \(\leftrightarrow \) upper-case mappings may be context-dependent, many-to-one, and/or time-dependent.2 At present Cactus basically ignores these issues. :(
See Also
strcmp() Standard C library function (prototype in <string.h>) to compare two strings.
Examples
#include "util_String.h" /* does the Cactus parameter driver specify the PUGH driver? */ /* (Cactus parameters are supposed to be case-insensitive) */ if (Util_StrCmpi(driver, "pugh") == 0) { /* PUGH code */ } else { /* non-PUGH code */ }
Depredcated; use strdup instead.
Concatenate strings safely.
Synopsis
#include "util_String.h" size_t result_len = Util_Strlcat(char *dst, const char *src, size_t size);
Result
result_len The size of the string the function tried to create, i.e. the initial strlen(dst) plus strlen(src).
Parameters
dst A non-NULL pointer to the (C-style NUL-terminated) destination string.
src A non-NULL pointer to the (C-style NUL-terminated) source string.
size The size of the destination buffer.
Discussion
The standard strcat() and strcpy() functions provide no way to specify the size of the destination buffer, so code using these functions is often vulnerable to buffer overflows. The standard strncat() and strncpy() functions can be used to write safe code, but their API is cumbersome, error-prone, and sometimes surprisingly inefficient:
Their size arguments are the number of characters remaining in the destination buffer, which must often be calculated at run-time, and is prone to off-by-one errors.
strncpy() doesn’t always NUL-terminate the destination string.
strncpy() NUL-fills the remainder of the buffer not used for the source string; this NUL-filling can be very expensive.
To solve these problems, the OpenBSD project developed the strlcat() and strlcpy() functions. See http://www.openbsd.org/papers/strlcpy-paper.ps for a history and general discussion of these functions. Some other Unix systems (notably Solaris) now provide these, but many don’t, so Cactus provides its own versions, Util_Strlcat() and Util_Strlcpy().
Util_Strlcat() appends the NUL-terminated string src to the end of the NUL-terminated string dst. It will append at most size - strlen(dst) - 1 characters (hence it never overflows the destination buffer), and it always leaves dst string NUL-terminated.
See Also
strcat() Standard C library function (prototype in <string.h>) to concatenate two strings. This does not
check that the buffer is big enough to hold the result, and is thus very dangerous. Use Util_Strlcat() instead!
Util_Strlcpy() [B30] Safely copy a string.
Examples
#include "util_String.h" /* * safely concatenate strings s1,s2,s3 into buffer: * ... this code is safe (it will never overflow the buffer), but * quick-n-dirty in that it doesn’t give any error indication * if the result is truncated to fit in the buffer */ #define BUFFER_SIZE 1024 char buffer[BUFFER_SIZE]; Util_Strlcpy(buffer, s1, sizeof(buffer)); Util_Strlcat(buffer, s2, sizeof(buffer)); Util_Strlcat(buffer, s3, sizeof(buffer));
#include "util_String.h" #define OK 0 #define ERROR_TRUNC 1 /* * safely concatenate strings s1,s2,s3 into buffer[N_buffer]; * return OK if ok, ERROR_TRUNC if result was truncated to fit in buffer */ int cat3(int N_buffer, char buffer[], const char s1[], const char s2[], const char s3[]) { int length; length = Util_Strlcpy(buffer, s1, N_buffer); if (length >= N_buffer) return ERROR_TRUNC; /*** ERROR EXIT ***/ length = Util_Strlcat(buffer, s2, N_buffer); if (length >= N_buffer) return ERROR_TRUNC; /*** ERROR EXIT ***/ length = Util_Strlcat(buffer, s3, N_buffer); if (length >= N_buffer) return ERROR_TRUNC; /*** ERROR EXIT ***/ return OK; /*** NORMAL RETURN ***/ }
Copies a string safely.
Synopsis
#include "util_String.h" size_t result_len = Util_Strlcpy(char *dst, const char *src, size_t size);
Result
result_len The size of the string the function tried to create, i.e. strlen(src).
Parameters
dst A non-NULL pointer to the (C-style NUL-terminated) destination string.
src A non-NULL pointer to the (C-style NUL-terminated) source string.
size The size of the destination buffer.
Discussion
The standard strcat() and strcpy() functions provide no way to specify the size of the destination buffer, so code using these functions is often vulnerable to buffer overflows. The standard strncat() and strncpy() functions can be used to write safe code, but their API is cumbersome, error-prone, and sometimes surprisingly inefficient:
Their size arguments are the number of characters remaining in the destination buffer, which must often be calculated at run-time, and is prone to off-by-one errors.
strncpy() doesn’t always NUL-terminate the destination string.
strncpy() NUL-fills the remainder of the buffer not used for the source string; this NUL-filling can be very expensive.
To solve these problems, the OpenBSD project developed the strlcat() and strlcpy() functions. See http://www.openbsd.org/papers/strlcpy-paper.ps for a history and general discussion of these functions. Some other Unix systems (notably Solaris) now provide these, but many don’t, so Cactus provides its own versions, Util_Strlcat() and Util_Strlcpy().
Util_Strlcpy() copies up to size-1 characters from the source string to the destination string, followed by a NUL character (so dst is always NUL-terminated). Unlike strncpy(), Util_Strlcpy() does not fill any left-over space at the end of the destination buffer with NUL characters.
See Also
strcpy() Standard C library function (prototype in <string.h>) to copy a string to a buffer. This does not
check that the buffer is big enough to hold the string, and is thus very dangerous. Use Util_Strlcpy() instead!
Util_Strlcat() [B25] Safely concatenates two strings.
Examples
#include "util_String.h" /* * safely concatenate strings s1,s2,s3 into buffer: * ... this code is safe (it will never overflow the buffer), but * quick-n-dirty in that it doesn’t give any error indication * if the result is truncated to fit in the buffer */ #define BUFFER_SIZE 1024 char buffer[BUFFER_SIZE]; Util_Strlcpy(buffer, s1, sizeof(buffer)); Util_Strlcat(buffer, s2, sizeof(buffer)); Util_Strlcat(buffer, s3, sizeof(buffer));
#include "util_String.h" #define OK 0 #define ERROR_TRUNC 1 /* * safely concatenate strings s1,s2,s3 into buffer[N_buffer]; * return OK if ok, ERROR_TRUNC if result was truncated to fit in buffer */ int cat3(int N_buffer, char buffer[], const char s1[], const char s2[], const char s3[]) { int length; length = Util_Strlcpy(buffer, s1, N_buffer); if (length >= N_buffer) return ERROR_TRUNC; /*** ERROR EXIT ***/ length = Util_Strlcat(buffer, s2, N_buffer); if (length >= N_buffer) return ERROR_TRUNC; /*** ERROR EXIT ***/ length = Util_Strlcat(buffer, s3, N_buffer); if (length >= N_buffer) return ERROR_TRUNC; /*** ERROR EXIT ***/ return OK; /*** NORMAL RETURN ***/ }
Separate off the first token from a string.
Synopsis
#include "util_String.h" char* token = Util_StrSep(const char** string_ptr, const char* delim_set);
Result
token This function returns the original value of *string_ptr, or NULL if the end of the string is reached.
Parameters
string_ptr A non-NULL pointer to a (modifyable) non-NULL pointer to the (C-style NUL-terminated) string
to operate on.
delim_set A non-NULL pointer to a (C-style NUL-terminated) string representing a set of delimiter characters
(the order of these characters doesn’t matter).
Discussion
Many Unix systems define a function strsep() which provides a clean way of splitting a string into “words”. However, some systems only provide the older (and inferior-in-several-ways) strtok() function, so Cactus implements its own strsep() function, Util_StrSep().
Util_StrSep() finds the first occurence in the string pointed to by *string_ptr of any character in the string pointed to by delim_set (or the terminating NUL if there is no such character), and replaces this by NUL. The location of the next character after the NUL character just stored (or NULL, if the end of the string was reached) is stored in *string_ptr.
An “empty” field, i.e. one caused by two adjacent delimiter characters, can be detected (after Util_StrSep() returns) by the test **string_ptr == ’\0’, or equivalently strlen(*string_ptr) == 0.
See the example section below for the typical usage of Util_StrSep().
See Also
strsep() Some systems provide this in the standard C library (prototype in <string.h>); Util_StrSep() is a
clone of this.
strtok() Inferior API for splitting a string into tokens (defined by the ANSI/ISO C standard).
Examples
#include <stdio.h> #include <stdlib.h> #include <string.h> #include "util_String.h" /* prototypes */ int parse_string(char* string, int N_argv, char* argv[]); /* * Suppose we have a Cactus parameter gridfn_list containing a * whitespace-separated list of grid functions. This function * "processes" (here just prints the name of) each grid function. */ void process_gridfn_list(const char* gridfn_list) { #define MAX_N_GRIDFNS 100 int N_gridfns; int i; char* copy_of_gridfn_list; char* gridfn[MAX_N_GRIDFNS]; copy_of_gridfn_list = strdup(gridfn_list); N_gridfns = parse_string(copy_of_gridfn_list, MAX_N_GRIDFNS, gridfn); for (i = 0 ; i < N_gridfns ; ++i) { /* "process" (here just print the name of) each gridfn */ printf("grid function %d is \"%s\"\n", i, gridfn[i]); } free(copy_of_gridfn_list); } /* * This function parses a string containing whitespace-separated * tokens into a main()-style argument vector (of size N_argv ). * This function returns the number of pointers stored into argv[] . * * Adjacent sequences of whitespace are treated the same as single * whitespace characters. * * Note that this function this modifies its input string. */ int parse_string(char* string, int N_argv, char* argv[]) { int i; for (i = 0 ; i < N_argv ; ) { argv[i] = Util_StrSep(&string, " \t\n\r\v"); if (argv[i] == NULL) { break; } /* reached end-of-string */ if (*argv[i] == ’\0’) { /* * found a 0-length "token" (a sequence of * two or more adjacent whitespace characters) * ==> skip this "token" (don’t store it) * ==> no-op here */ } else { /* token has length > 0 ==> store it */ ++i; } } return i; }
Creates a new table which is a “clone” (exact copy) of an existing table
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int clone_handle = Util_TableClone(int handle);
call Util_TableClone(clone_handle, handle) integer clone_handle, handle
Result
clone_handle (\(\ge 0\)) A handle to the clone table
Parameters
handle Handle to the table to be cloned
Discussion
Viewing a table as a set of key/value pairs, this function creates a new table (with the same flags word as the original) containing copies of all the original table’s key/value pairs. The two tables are completely independent, i.e. future changes to one won’t affect the other.
Note that if there are any CCTK_POINTER and/or CCTK_FPOINTER values in the table, they are “shallow copied”, i.e. the (pointer) values in the table are copied. This results in the clone table’s pointer values pointing to the same places as the original table’s pointer values. Be careful with this! In particular, if you’re using pointer values in the table to keep track of malloc() memory, be careful not to free() the same block of memory twice!
Note that table iterators are not guaranteed to sequence through the original and clone tables in the same order. (This is a special case of the more general “non-guarantee” in the Section of table iterators in the Users’ Guide: the order of table iterators may differ even between different tables with identical key/value contents.)
See Also
Util_TableCreate() [B45] create a table
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableDestroy() [B56] destroy a table
Errors
UTIL_ERROR_NO_MEMORY unable to allocate memory
UTIL_ERROR_TABLE_BAD_FLAGS flags word is negative in the to-be-cloned table (this indicates an internal error
in the table routines, and should never happen)
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" /* * This function is passed (a handle to) a table containing some entries. * It needs to set some additional entries and pass the table to some * other function(s), but it also needs to leave the original table * intact for other use by the caller. The solution is to clone the * original table and work on the clone, leaving the original table * unchanged. */ int my_function(int handle, int x, int y) { int status; /* clone the table */ const int clone_handle = Util_TableClone(handle) if (clone_handle < 0) return clone_handle; /* error in cloning table */ /* now set our entries in the clone table */ status = Util_TableSetInt(clone_handle, x, "x"); if (status < 0) return status; /* error in setting x */ status = Util_TableSetInt(clone_handle, y, "y"); if (status < 0) return status; /* error in setting y */ /* ... code to use the clone table ... */ /* ... eg pass clone_handle to other functions ... */ /* we’re done with the clone now */ Util_TableDestroy(clone_handle); return 0; }
Creates a new (empty) table
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableCreate(int flags);
call Util_TableCreate(handle, flags) integer handle, flags
Result
handle (\(\ge 0\)) A handle to the newly-created table
Parameters
flags (\(\ge 0\)) A flags word for the table. This should be the inclusive-or of zero or more of the UTIL_TABLE_FLAGS_* bit
masks (defined in "util_Table.h"). For Fortran users, note that inclusive-or is the same as sum here, since the
bit masks are all disjoint.
Discussion
We require the flags word to be non-negative so that other functions can distinguish flags from (negative) error codes.
Any User-defined flag words should use only bit positions at or above
UTIL_TABLE_FLAGS_USER_DEFINED_BASE, i.e. all bit positions below this are reserved for present of future
Cactus use.
At present there is only a single flags-word bit mask defined in "util_Table.h":
UTIL_TABLE_FLAGS_CASE_INSENSITIVE
By default keys are treated as C-style character strings, and the table functions compare them with
the standard C strcmp function. However, by setting the
UTIL_TABLE_FLAGS_CASE_INSENSITIVE bit in the flags word, this table’s keys may be made
case-insensitive, i.e. the table routines then compare this table’s keys with Util_StrCmpi(). Note
that keys are still stored exactly as the caller specifies them (i.e. they are not forced into a canonical
case); it’s only their comparison that’s affected by this flag.
See Also
Util_StrCmpi() [B19] compare two strings, ignoring upper/lower case
Util_TableClone() [B40] create a new table which is a “clone” (exact copy) of an existing table
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableDestroy() [B56] destroy a table
Errors
UTIL_ERROR_NO_MEMORY unable to allocate memory
UTIL_ERROR_TABLE_BAD_FLAGS flags word is negative
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" /* create a table, simplest case */ int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); /* create a table whose keys will be treated as case-insensitive */ int handle2 = Util_TableCreate(UTIL_TABLE_FLAGS_CASE_INSENSITIVE);
Creates a new table (with the case-insensitive flag set) and sets values in it based on a string argument (interpreted with “parameter-file” semantics)
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableCreateFromString(const char *string);
call Util_TableCreateFromString(handle, string) integer handle character*(*) string
Result
handle (\(\ge 0\)) a handle to the newly-created table
Parameters
string a pointer to a C-style null-terminated string specifying the table contents; see the description
for Util_TableSetFromString() for a full description of the syntax and semantics of this string
See Also
Util_TableClone() [B40] Create a new table which is a “clone” (exact copy) of an existing table
Util_TableCreate() [B45] create a table
Util_TableSetFromString() [B131] sets values in a table based on a string argument
Errors
UTIL_ERROR_NO_MEMORY unable to allocate memory
UTIL_ERROR_BAD_KEY invalid input: key contains invalid character
UTIL_ERROR_BAD_INPUT invalid input: can’t parse input string
other error codes this function may also return any error codes returned by Util_TableCreate() or
Util_TableSetFromString()
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableCreateFromString("order = 3\t" "myreal = 42.314159\t" "mystring = ’hello’\t" "myarray = { 0 1 2 3 }"); /* equivalent code to the above */ int handle = Util_TableCreate(UTIL_TABLE_FLAGS_CASE_INSENSITIVE); Util_TableSetFromString(handle, "order = 3\t" "myreal = 42.314159\t" "mystring = ’hello’" "myarray = { 0 1 2 3 }"); /* also equivalent to the above */ int handle = Util_TableCreate(UTIL_TABLE_FLAGS_CASE_INSENSITIVE); CCTK_INT array[] = {0, 1, 2, 3}; Util_TableSetInt(handle, 3, "order"); Util_TableSetReal(handle, 42.314159, "myreal"); Util_TableSetString(handle, "hello", "mystring"); Util_TableSetIntArray(handle, 4, array, "myarray");
Deletes a specified key/value entry from a table
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int key_exists = Util_TableDeleteKey(int handle, const char *key);
call Util_TableDeleteKey(key_exists, handle, key) integer key_exists, handle character*(*) key
Result
0 ok (key existed before this call, and has now been deleted)
Parameters
handle (\(\ge 0\)) handle to the table
key a pointer to the key (a C-style null-terminated string)
Discussion
This function invalidates any iterators for the table which are not in the “null-pointer” state.
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_TABLE_NO_SUCH_KEY no such key in table
Destroys a table
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableDestroy(int handle);
call Util_TableDestroy(status, handle) integer status, handle
Result
0 ok
Parameters
handle (\(\ge 0\)) handle to the table
Discussion
Of course, this function invalidates any and all iterators for the table. :)
See Also
Util_TableClone() [B40] Create a new table which is a “clone” (exact copy) of an existing table
Util_TableCreate() [B45] create a table
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" /* create a table */ int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); /* do things with the table: put values in it, */ /* pass its handle to other functions, etc etc */ /* ... */ /* at this point we (and all other functions we */ /* may call in the future) are done with the table */ Util_TableDestroy(handle);
This is a family of functions, one for each Cactus data type, to get the single (1-element array) value, or more generally the first array element of the value, associated with a specified key in a key/value table.
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int N_elements = Util_TableGetXxx(int handle, CCTK_XXX *value, const char *key);
where XXX is one of POINTER, FPOINTER1 , CHAR, BYTE, INT, INT1, INT2, INT4, INT8, REAL, REAL4, REAL8, REAL16, COMPLEX, COMPLEX8, COMPLEX16, COMPLEX32 (not all of these may be supported on any given system)
call Util_TableGetXxx(N_elements, handle, value, key) integer N_elements, handle CCTK_XXX value character*(*) key
where CCTK_XXX may be any data type supported by C (above) except CCTK_CHAR (Fortran doesn’t have a separate “character” data type; use CCTK_BYTE instead)
Result
N_elements the number of array elements in the value
Parameters
handle (\(\ge 0\)) handle to the table
value a pointer to where this function should store a copy of the value (or more generally the first
array element of the value) associated with the specified key, or NULL pointer to skip storing this
key a pointer to the key (a C-style null-terminated string)
Discussion
Note that it is not an error for the value to actually have \(> 1\) array elements; in this case only the first element is stored. The rationale for this design is that the caller may know or suspect that the value is a large array, but may only want the first array element; in this case this design avoids the caller having to allocate a large buffer unnecessarily.
In contrast, it is an error for the value to actually be an empty (0-length) array, because then there is no “first array element” to get.
It is also an error for the value to actually have a different type than CCTK_XXX.
If any error code is returned, the user’s value buffer (pointed to by value if this is non-NULL) is unchanged.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*Array() get an array value
Util_TableGetString() [B76] get a character-string value
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_TABLE_NO_SUCH_KEY no such key in table
UTIL_ERROR_TABLE_WRONG_DATA_TYPE value has data type other than CCTK_TYPE
UTIL_ERROR_TABLE_VALUE_IS_EMPTY value is an empty (0-element) array
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" #define N_DIGITS 5 static const CCTK_INT pi_digits[N_DIGITS] = {3, 14, 159, 2653, 58979}; int N; CCTK_INT x; int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetIntArray(handle, N_DIGITS, pi_digits, "digits of pi"); Util_TableSetIntArray(handle, 0, pi_digits, "empty array"); /* gets N = 5, x = 3 */ N = Util_TableGetInt(handle, &x, "digits of pi"); /* gets N = UTIL_ERROR_TABLE_VALUE_IS_EMPTY */ N = Util_TableGetInt(handle, &x, "empty array");
This is a family of functions, one for each Cactus data type, to get a copy of the value associated with a specified key, and store it (more accurately, as much of it as will fit) in a specified array
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int N_elements = Util_TableGetXxxArray(int handle, int N_array, CCTK_XXX array[], const char *key);
where XXX is one of POINTER, FPOINTER2 , CHAR, BYTE, INT, INT1, INT2, INT4, INT8, REAL, REAL4, REAL8, REAL16, COMPLEX, COMPLEX8, COMPLEX16, COMPLEX32 (not all of these may be supported on any given system)
call Util_TableGetXxxArray(N_elements, handle, N_array, array, key) integer N_elements, handle, N_array CCTK_XXX(*) array character*(*) key
where CCTK_XXX may be any data type supported by C (above)
Result
N_elements the number of array elements in the value
Parameters
handle (\(\ge 0\)) handle to the table
N_array the number of array elements in array[] (must be \(\ge 0\) if array != NULL)
array a pointer to where this function should store (up to N_array elements of) a copy of the value associated
with the specified key, or NULL pointer to skip storing this
key a pointer to the key (a C-style null-terminated string)
Discussion
Note that it is not an error for the value to actually have \(> \code {N\_array}\) array elements; in this case only the first N_array elements are stored. The caller can detect this by comparing the return value with N_array. The rationale for this design is that the caller may know or suspect that the value is a large array, but may only want the first few array elements; in this case this design avoids the caller having to allocate a large buffer unnecessarily.
It is also not an error for the value to actually have \(< \code {N\_array}\) array elements; again the caller can detect this by comparing the return value with N_array.
It is an error for the value to actually have a different type than CCTK_XXX.
If any error code is returned, the user’s value buffer (pointed to by array if this is non-NULL) is unchanged.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_BAD_INPUT array != NULL and N_array \(< 0\)
UTIL_ERROR_TABLE_NO_SUCH_KEY no such key in table
UTIL_ERROR_TABLE_WRONG_DATA_TYPE value has data type other than CCTK_TYPE
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" #define N_STUFF 3 static const CCTK_REAL stuff[N_STUFF] = {42.0, 69.0, 105.5}; #define N_OUTPUT 2 CCTK_INT output[N_OUTPUT]; int N; int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetRealArray(handle, N_STUFF, stuff, "blah blah blah"); /* gets N = 3, output[0] = 42.0, output[1] = 69.0 */ N = Util_TableGetRealArray(handle, N_OUTPUT, output, "blah blah blah");
Get the single (1-element array) value, or more generally the first array element of the value, associated with a specified key in a key/value table; the value’s data type is generic. That is, the value is specified by a CCTK_VARIABLE_* type code and a void * pointer.
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int N_elements = Util_TableGetGeneric(int handle, int type_code, void *value, const char *key);
call Util_TableGetGeneric(N_elements, handle, type_code, value, key) integer N_elements, handle, type_code CCTK_POINTER value character*(*) key
Result
N_elements the number of array elements in the value
Parameters
handle (\(\ge 0\)) handle to the table
type_code the value’s type code (one of the CCTK_VARIABLE_* constants from "cctk_Constants.h")
value a pointer to where this function should store a copy of the value (or more generally the first
array element of the value) associated with the specified key, or NULL pointer to skip storing this
key a pointer to the key (a C-style null-terminated string)
Discussion
Note that it is not an error for the value to actually have \(> 1\) array elements; in this case only the first element is stored. The rationale for this design is that the caller may know or suspect that the value is a large array, but may only want the first array element; in this case this design avoids the caller having to allocate a large buffer unnecessarily.
In contrast, it is an error for the value to actually be an empty (0-length) array, because then there is no “first array element” to get.
It is also an error for the value to actually have a different type than that specified by type_code.
If any error code is returned, the user’s value buffer (pointed to by value if this is non-NULL) is unchanged.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value
Util_TableGet*Array() get an array value
Util_TableGetString() [B76] get a character-string value
Util_TableQueryValueInfo() [B113] query key present/absent in table, and optionally type and/or number of
elements
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_TABLE_NO_SUCH_KEY no such key in table
UTIL_ERROR_TABLE_WRONG_DATA_TYPE value has data type other than CCTK_TYPE
UTIL_ERROR_TABLE_VALUE_IS_EMPTY value is an empty (0-element) array
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" #include "cctk_Constants.h" #define N_DIGITS 5 static const CCTK_INT pi_digits[N_DIGITS] = {3, 14, 159, 2653, 58979}; int N; CCTK_INT x; void *xptr = (void *) &x; int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetIntArray(handle, N_DIGITS, pi_digits, "digits of pi"); Util_TableSetIntArray(handle, 0, pi_digits, "empty array"); /* gets N = 5, x = 3 */ N = Util_TableGetGeneric(handle, CCTK_VARIABLE_INT, &x, "the answer"); /* gets N = UTIL_ERROR_TABLE_VALUE_IS_EMPTY, leaves x unchanged */ N = Util_TableGetGeneric(handle, CCTK_VARIABLE_INT, &x, "empty array");
Get a copy of the value associated with a specified key, and store it (more accurately, as much of it as will fit) in a specified array; the array’s data type is generic. That is the array is specified by a CCTK_VARIABLE_* type code, a count of the number of array elements, and a void * pointer.
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int N_elements = Util_TableGetGenericArray(int handle, int type_code, int N_array, void *array, const char *key);
call Util_TableGetGenericArray(N_elements, . handle, . type_code, . N_array, array, . key) integer N_elements, handle, type_code, N_array CCTK_POINTER array character*(*) key
Result
N_elements the number of array elements in the value
Parameters
handle (\(\ge 0\)) handle to the table
type_code the value’s type code (one of the CCTK_VARIABLE_* constants from "cctk_Constants.h")
N_array the number of array elements in array[] (must be \(\ge 0\) if array != NULL)
array a pointer to where this function should store (up to N_array elements of) a copy of the value associated
with the specified key, or NULL pointer to skip storing this
key a pointer to the key (a C-style null-terminated string)
Discussion
Note that it is not an error for the value to actually have \(> \code {N\_array}\) array elements; in this case only the first N_array elements are stored. The caller can detect this by comparing the return value with N_array. The rationale for this design is that the caller may know or suspect that the value is a large array, but may only want the first few array elements; in this case this design avoids the caller having to allocate a large buffer unnecessarily.
It is also not an error for the value to actually have \(< \code {N\_array}\) array elements; again the caller can detect this by comparing the return value with N_array.
It is an error for the value to actually have a different type than that specified by type_code.
If any error code is returned, the user’s value buffer (pointed to by array if this is non-NULL) is unchanged.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableQueryValueInfo() [B113] query key present/absent in table, and optionally type and/or number of
elements
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_BAD_INPUT array != NULL and N_array \(< 0\)
UTIL_ERROR_TABLE_NO_SUCH_KEY no such key in table
UTIL_ERROR_TABLE_WRONG_DATA_TYPE value has data type other than CCTK_TYPE
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" #define N_STUFF 3 static const CCTK_REAL stuff[N_STUFF] = {42.0, 69.0, 105.5}; #define N_OUTPUT 2 CCTK_INT output[N_OUTPUT]; int N; int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetRealArray(handle, N_STUFF, stuff, "stuff"); /* gets N = UTIL_ERROR_TABLE_WRONG_DATA_TYPE, output[] unchanged */ N = Util_TableGetGenericArray(handle, CCTK_VARIABLE_INT, N_OUTPUT, output, "stuff"); /* gets N = 3, output[0] = 42.0, output[1] = 69.0 */ N = Util_TableGetGenericArray(handle, CCTK_VARIABLE_REAL, N_OUTPUT, output, "stuff");
Gets a copy of the character-string value associated with a specified key in a table, and stores it (more accurately, as much of it as will fit) in a specified character string
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int length = Util_TableGetString(int handle, int buffer_length, char buffer[], const char *key);
Result
Results are the same as all the other Util_TableGet*() functions:
length the length of the string (C strlen semantics, i.e. not including the terminating null character)
Parameters
handle (\(\ge 0\)) handle to the table
buffer_length the length (sizeof) of buffer[] (must be \(\ge 1\) if buffer != NULL)
buffer a pointer to a buffer into which this function should store (at most buffer_length-1 characters of) the
value, terminated by a null character as usual for C strings, or NULL pointer to skip storing this
key a pointer to the key (a C-style null-terminated string)
Discussion
This function assumes that the string is stored as an array of CCTK_CHARs, not including a terminating null character.
This function differs from Util_TableGetCharArray() in two ways: It explicitly provides a terminating null character for C-style strings, and it explicitly checks for the string being too long to fit in the buffer (in which case it returns UTIL_ERROR_TABLE_STRING_TRUNCATED).
If the error code UTIL_ERROR_TABLE_STRING_TRUNCATED is returned, then the first buffer_length-1 characters of the string are returned in the user’s buffer (assuming buffer is non-NULL), followed by a null character to properly terminate the string in the buffer. If any other error code is returned, the user’s value buffer (pointed to by buffer if this is non-NULL) is unchanged.
To find out how long the string is (and thus how big of a buffer you need to allocate to avoid having the string truncated), you can call this function with \(\code {buffer\_length} = 0\) and \(\code {buffer} = \code {NULL}\) (or actually anything you want); the return result will give the string length.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGet*Array() get an array value
Util_TableGetCharArray() [B64] get an array-of-CCTK_CHAR value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetString() [B145] set a character-string value
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetCharArray() [B127] set an array-of-CCTK_CHAR value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_BAD_INPUT buffer != NULL and buffer_length \(\le 0\)
UTIL_ERROR_TABLE_NO_SUCH_KEY no such key in table
UTIL_ERROR_TABLE_WRONG_DATA_TYPE value has data type other than CCTK_CHAR
UTIL_ERROR_TABLE_STRING_TRUNCATED buffer != NULL and value was truncated to fit in buffer[]
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" #define N_BUFFER 100 char buffer[N_BUFFER]; int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetString(handle, "relativity", "Einstein"); /* get length of string (= 10 here) */ int length = Util_TableGetString(handle, 0, NULL, "Einstein"); /* get null-terminated string into buffer, also returns 10 */ Util_TableGetString(handle, N_BUFFER, buffer, "Einstein");
Advance a table iterator to the next entry in the table
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int is_nonnull = Util_TableItAdvance(int ihandle);
Result
1 ok (iterator now points to some table entry)
0 ok (iterator has just advanced past the last table entry, and is now in the ”null-pointer” state)
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
If we view an iterator as an abstraction of a pointer into the table, then this function is the abstraction of the C “++” operation applied to the pointer, except that this function automagically sets the iterator to the ”null-pointer” state when it advances past the last table entry.
Note that bad things (garbage results, core dumps) may happen if you call this function on an iterator which has been invalidated by a change in the table’s contents.
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
Examples
/* walk through all entries of a table */ int ihandle; for ( ihandle = Util_TableItCreate(handle) ; Util_TableItQueryIsNonNull(ihandle) > 0 ; Util_TableItAdvance(ihandle) ) { /* do something with the table entry */ } Util_TableItDestroy(ihandle);
Creates a new table iterator which is a “clone” (exact copy) of an existing table iterator
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int clone_ihandle = Util_TableItClone(int ihandle);
Result
clone_ihandle (\(\ge 0\)) A handle to the clone table iterator
Parameters
ihandle handle to the table iterator to be cloned
Discussion
This function creates a new iterator which points to the same place in the same table as the original iterator. If the original iterator is in the “null-pointer” state, then the clone is also in this state.
Note that bad things (garbage results, core dumps) may happen if you call this function on an iterator which has been invalidated by a change in the table’s contents.
See Also
Util_TableClone() [B40] create a new table which is a “clone” (exact copy) of an existing table
Util_TableItCreate() [B86] create a table iterator
Util_TableItDestroy() [B89] destroy a table iterator
Errors
UTIL_ERROR_BAD_HANDLE iterator handle to be cloned, is invalid
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" /* * Apart from efficiency and slight differences in error return codes, * Util_TableItClone() could be simulated by the following code. */ int Util_TableItClone(int ihandle) { int status; /* to what table does the to-be-cloned iterator point? */ const int handle = Util_TableQueryTableHandle(ihandle); if (handle < 0) return handle; /* error in querying table handle */ /* create the to-be-cloned iterator */ /* (pointing into the same table as the original iterator) */ { const int clone_ihandle = Util_TableItCreate(handle); if (clone_ihandle < 0) return clone_ihandle; /* error in creating clone iterator */ /* how long is the key to which the to-be-cloned iterator points? */ { const int key_length = Util_TableItQueryKeyValueInfo(ihandle, 0, NULL, NULL, NULL); if (key_length == UTIL_TABLE_ITERATOR_IS_NULL) { /* to-be-cloned iterator is in "null-pointer" state */ Util_TableItSetToNull(clone_ihandle); return clone_ihandle; /* normal return */ } if (key_length < 0) return key_length; /* error in querying to-be-cloned iterator */ /* to what key does the to-be-cloned iterator point? */ { const int key_buffer_length = key_length + 1; char *const key_buffer = (char *) malloc(key_buffer_length); if (key_buffer == NULL) return UTIL_ERROR_NO_MEMORY; status = Util_TableItQueryKeyValueInfo(ihandle, key_buffer_length, key_buffer); if (status < 0) return status; /* error in querying to-be-cloned iterator */ /* set the clone iterator to point to the same key as the original */ status = Util_TableItSetToKey(clone_ihandle, key_buffer); free(key_buffer); return clone_ihandle; /* normal return */ } } } }
Create a new table iterator
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int ihandle = Util_TableItCreate(int handle);
Result
ihandle (\(\ge 0\)) handle to the table iterator
Parameters
handle (\(\ge 0\)) handle to the table over which the iterator should iterate
Discussion
This function creates a new table iterator. The iterator initially points at the starting table entry.
See Also
Util_TableItDestroy() [B89] destroy a table iterator
Errors
UTIL_ERROR_BAD_HANDLE table handle is invalid
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
/* walk through all entries of a table */ int ihandle; for ( ihandle = Util_TableItCreate(handle) ; Util_TableItQueryIsNonNull(ihandle) > 0 ; Util_TableItAdvance(ihandle) ) { /* do something with the table entry */ } Util_TableItDestroy(ihandle);
Destroy a table iterator
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableItDestroy(int ihandle);
Result
0 ok
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
See Also
Util_TableItCreate() [B86] create a table iterator
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
/* walk through all entries of a table */ int ihandle; for ( ihandle = Util_TableItCreate(handle) ; Util_TableItQueryIsNonNull(ihandle) > 0 ; Util_TableItAdvance(ihandle) ) { /* do something with the table entry */ } Util_TableItDestroy(ihandle);
Query whether a table iterator is not in the “null-pointer” state
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableItQueryIsNonNull(int ihandle);
Result
1 iterator is not in the “null-pointer” state, i.e. iterator points to some table entry
0 iterator is in the “null-pointer” state
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
If no errors occur, Util_TableItQueryIsNonNull(ihandle) is the same as
1 - Util_TableItQueryIsNull(ihandle).
Note that bad things (garbage results, core dumps) may happen if you call this function on an iterator which has been invalidated by a change in the table’s contents.
See Also
Util_TableItQueryIsNull() [B94] query whether a table iterator is in the “null-pointer” state
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
Examples
/* walk through all entries of a table */ int ihandle; for ( ihandle = Util_TableItCreate(handle) ; Util_TableItQueryIsNonNull(ihandle) > 0 ; Util_TableItAdvance(ihandle) ) { /* do something with the table entry */ } Util_TableItDestroy(ihandle);
Query whether a table iterator is in the “null-pointer” state
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableItQueryIsNull(int ihandle);
Result
1 iterator is in the “null-pointer” state
0 iterator is not in the “null-pointer” state, i.e. iterator points to some table entry
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
If no errors occur, Util_TableItQueryIsNull(ihandle) is the same as 1 - Util_TableItQueryIsNonNull(ihandle).
Note that bad things (garbage results, core dumps) may happen if you call this function on an iterator which has been invalidated by a change in the table’s contents.
See Also
Util_TableItQueryIsNonNull() [B91] query whether a table iterator is not in the “null-pointer” state,
i.e. whether the iterator points to some table entry
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
Examples
/* variant code to walk through all entries of a table */ int ihandle; for ( ihandle = Util_TableItCreate(handle) ; Util_TableItQueryIsNull(ihandle) == 0 ; Util_TableItAdvance(ihandle) ) { /* do something with the table entry */ } Util_TableItDestroy(ihandle);
Query the key and the type and number of elements of the value corresponding to that key, of the table entry to which an iterator points
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int key_length = Util_TableItQueryKeyValueInfo(int ihandle, int key_buffer_length, char key_buffer[], CCTK_INT *type_code, CCTK_INT *N_elements)
Result
key_length The string length of the key (this has C strlen semantics, i.e. it does not include a terminating
null character)
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
key_buffer_length the length (sizeof) of key_buffer[] (must be \(\ge 1\) if key_buffer != NULL)
key_buffer a pointer to a buffer into which this function should store (at most key_buffer_length-1
characters of) the key, terminated by a null character as usual for C strings, or NULL pointer to skip storing this
type_code a pointer to where this function should store the value’s type code (one of the CCTK_VARIABLE_*
constants from "cctk_Constants.h"), or a NULL pointer to skip storing this.
N_elements a pointer to where this function should store the number of array elements in the value, or a NULL
pointer to skip storing this.
Discussion
The usual use of an iterator is to iterate through all the entries of a table, calling this function on each entry, then taking further action based on the results.
Note that bad things (garbage results, core dumps) may happen if you call this function on an iterator which has been invalidated by a change in the table’s contents.
If the error code UTIL_ERROR_TABLE_STRING_TRUNCATED is returned, then the first key_buffer_length-1 characters of the key are returned in the user’s key buffer (assuming key_buffer is non-NULL), followed by a null character to properly terminate the string in the buffer. If any other error code is returned, the user’s key buffer (pointed to by key_buffer if this is non-NULL) is unchanged.
See Also
Util_TableQueryValueInfo() [B113] query key present/absent in table, and optionally type and/or number of
elements, but using the key instead of an iterator
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_ITERATOR_IS_NULL iterator is in ”null-pointer” state
UTIL_ERROR_TABLE_STRING_TRUNCATED key_buffer != NULL and key was truncated to fit in key_buffer
Examples
/* print out all entries in a table */ /* return 0 for ok, type code for any types we can’t handle, */ /* -ve for other errors */ #include <stdio.h> #include <stdlib.h> #include "util_ErrorCodes.h" #include "util_Table.h" #include "cctk.h" int print_table(int handle) { int max_key_length, N_key_buffer, ihandle; char *key_buffer; max_key_length = Util_TableQueryMaxKeyLength(handle); if (max_key_length < 0) return max_key_length; N_key_buffer = max_key_length + 1; key_buffer = (char *) malloc(N_key_buffer); if (key_buffer == NULL) return UTIL_ERROR_NO_MEMORY; for ( ihandle = Util_TableItCreate(handle) ; Util_TableItQueryIsNonNull(ihandle) > 0 ; Util_TableItAdvance(ihandle) ) { CCTK_INT type_code, N_elements; CCTK_INT value_int; CCTK_REAL value_real; Util_TableItQueryKeyValueInfo(ihandle, N_key_buffer, key_buffer, &type_code, &N_elements); printf("key = \"%s\"\n", key_buffer); switch (type_code) { case CCTK_VARIABLE_INT: Util_TableGetInt(handle, &value_int, key_buffer); printf("value[int] = %d\n", (int)value_int); break; case CCTK_VARIABLE_REAL: Util_TableGetReal(handle, &value_real, key_buffer); printf("value[real] = %g\n", (double)value_real); break; default: /* we don’t know how to handle this type */ Util_TableItDestroy(ihandle); free(key_buffer); return type_code; } } Util_TableItDestroy(ihandle); free(key_buffer); return 0; }
Query what table a table iterator iterates over
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableItQueryTableHandle(int ihandle);
Result
handle (\(\ge 0\)) handle to the table over which the iterator iterates
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
Note that it is always ok to call this function, regardless of whether or not the iterator is in the “null-pointer” state.
It’s also ok to call this function even when the iterator has been invalidated by a change in the table’s contents.
See Also
Util_TableItCreate() [B86] create an iterator (which iterates over a specified table)
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
Reset a table iterator to point to the starting table entry
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableItResetToStart(int ihandle);
Result
Results are the same as calling Util_TableItQueryIsNonNull() on the iterator after the reset:
1 iterator is not in the “null-pointer” state, i.e. iterator points to some table entry
0 iterator is in the “null-pointer” state (this happens if and only if the table is empty)
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
Note that it is always ok to call this function, regardless of whether or not the iterator is in the “null-pointer” state.
It’s also ok to call this function even when the iterator has been invalidated by a change in the table’s contents.
See Also
Util_TableItSetToNull() [B107] set an iterator to the “null-pointer” state
Util_TableItSetToKey() [B105] set an iterator to point to a specified table entry
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
Set a table iterator to point to a specified table entry
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableItSetToKey(int ihandle, const char *key);
Result
0 ok
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
This function has the same effect as calling Util_TableItResetToStart() followed by calling Util_TableItAdvance() zero or more times to make the iterator point to the desired table entry. However, this function will typically be (much) more efficient than that sequence.
Note that it is always ok to call this function, regardless of whether or not the iterator is in the “null-pointer” state.
It’s also ok to call this function even when the iterator has been invalidated by a change in the table’s contents.
See Also
Util_TableItResetToStart() [B103] reset an iterator to point to the starting table entry
Util_TableItSetToNull() [B107] set a table iterator to the ”null-pointer” state
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_TABLE_NO_SUCH_KEY no such key in table
Set a table iterator to the ”null-pointer” state
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableItSetToNull(int ihandle);
Result
0 ok
Parameters
ihandle (\(\ge 0\)) handle to the table iterator
Discussion
Note that it is always ok to call this function, regardless of whether or not the iterator is already in the “null-pointer” state.
It’s also ok to call this function even when the iterator has been invalidated by a change in the table’s contents.
See Also
Util_TableItResetToStart() [B103] reset an iterator to point to the starting table entry
Util_TableItSetToKey() [B105] set an iterator to point to a specified table entry
Errors
UTIL_ERROR_BAD_HANDLE iterator handle is invalid
Query a table’s flags word
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int flags = Util_TableQueryFlags(int handle);
call Util_TableQueryFlags(flags, handle) integer flags, handle
Result
flags (\(\ge 0\)) the flags word
Parameters
handle (\(\ge 0\)) handle to the table
Discussion
See Util_TableCreate() for further discussion of the semantics of flag words.
See Also
Util_TableClone() [B40] create a new table which is a “clone” (exact copy) of an existing table
Util_TableCreate() [B45] create a table (flags word specified explicitly)
Util_TableCreateFromString() [B49] convenience routine to create a table (with certain default flags) and set
key/value entries in it based on a parameter-file–like character string
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
Examples
#include <string.h> #include "util_ErrorCodes.h" #include "util_String.h" #include "util_Table.h" /* compare two strings, doing the comparison with the same */ /* case-sensitive/insensitive semantics as a certain table uses */ int compare_strings(int handle, const char *str1, const char *str2) { int flags = Util_TableQueryFlags(handle); return (flags & UTIL_TABLE_FLAGS_CASE_INSENSITIVE) ? Util_StrCmpi(str1, str2) : strcmp (str1, str2); }
Query whether or not a specified key is in the table, and optionally the type and/or number of elements of the value corresponding to this key
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int key_exists = Util_TableQueryValueInfo(int handle, CCTK_INT *type_code, CCTK_INT *N_elements, const char *key);
call Util_TableQueryValueInfo(key_exists, . handle, . type_code, N_elements, . key) integer key_exists, handle CCTK_INT type_code, N_elements character*(*) key
Result
1 ok (key is in table)
0 ok (no such key in table)
(in this case nothing is stored in *type_code and *N_elements)
Parameters
handle (\(\ge 0\)) handle to the table
type_code a pointer to where this function should store the value’s type code (one of the CCTK_VARIABLE_*
constants from "cctk_Constants.h"), or a NULL pointer to skip storing this.
N_elements a pointer to where this function should store the number of array elements in the value, or a NULL
pointer to skip storing this.
key a pointer to the key (a C-style null-terminated string)
Discussion
Unlike all the other table query functions, this function returns 0 for “no such key in table”. The rationale for this design is that by passing NULL pointers for type_code and N_elements, this function is then a Boolean “is key in table?” predicate.
If any error code is returned, the user’s buffers (pointed to by type_code and N_elements if these are non-NULL) are unchanged.
See Also
Util_TableItQueryKeyValueInfo() [B97] query key present/absent in table, and optionally type and/or
number of elements, but using an iterator instead of the key
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ‘/’ character
Examples
#include <stdio.h> #include <assert.h> #include "util_ErrorCodes.h" #include "util_Table.h" static const int data[] = {314, 159, 265}; #define N_DATA (sizeof(data) / sizeof(data[0])) CCTK_INT type_code, N_elements; /* see whether or not "key" is in table */ if (Util_TableQueryValueInfo(handle, NULL, NULL, "key")) { /* key is in the table */ } else { /* key is not in the table */ } /* put "data" in table as 3-element integer array */ Util_TableSetIntArray(handle, N_DATA, data, "data"); /* query info about "data" value */ assert( Util_TableQueryValueInfo(handle, &type_code, &N_elements, "data") == 1 ); assert( type_code == CCTK_VARIABLE_INT ); assert( N_elements == N_DATA );
Query the maximum key length in a table
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int max_key_length = Util_TableQueryMaxKeyLength(int handle);
call Util_TableQueryMaxKeyLength(max_key_length, handle) integer max_key_length, handle
Result
max_key_length (\(\ge 0\)) The string length of the longest key in the table (this has C strlen semantics, i.e. it does
not include a terminating null character)
Parameters
handle (\(\ge 0\)) handle to the table
Discussion
This function is useful if you’re going to iterate through a table, and you want to allocate a buffer which is guaranteed to be big enough to hold any key in the table.
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
Examples
#include <stdlib.h> #include "util_ErrorCodes.h" #include "util_Table.h" #include "cctk.h" int max_key_length = Util_TableQueryMaxKeyLength(handle); int N_buffer = max_key_length + 1; char *const buffer = (char *) malloc(N_buffer); if (buffer == NULL) { CCTK_WARN(CCTK_WARN_ABORT, "couldn’t allocate memory for table key buffer!"); abort(); /* CCTK_Abort() would be better */ /* if we have a cGH* available */ } /* now buffer is guaranteed to be */ /* big enough for any key in the table */
Query the number of key/value entries in a table
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int N_Keys = Util_TableQueryNKeys(int handle);
call Util_TableQueryNKeys(N_Keys, handle) integer N_Keys, handle
Result
N_Keys (\(\ge 0\)) the number of key/value entries in the table
Parameters
handle (\(\ge 0\)) handle to the table
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
This is a family of functions, one for each Cactus data type, to set the value associated with a specified key to be a specified single (1-element array) value
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableSetXxx(int handle, CCTK_XXX value, const char *key);
where XXX is one of POINTER, FPOINTER3 , CHAR, BYTE, INT, INT1, INT2, INT4, INT8, REAL, REAL4, REAL8, REAL16, COMPLEX, COMPLEX8, COMPLEX16, COMPLEX32 (not all of these may be supported on any given system)
call Util_TableSetXxx(status, handle, value, key) integer status, handle CCTK_XXX value character*(*) key
where CCTK_XXX may be any data type supported by C (above) except CCTK_CHAR (Fortran doesn’t have a separate “character” data type; use CCTK_BYTE instead)
Result
1 ok (key was already in table before this call, old value was replaced)
(it doesn’t matter what the old value’s type_code and N_elements were, i.e. these do not have to match the
new value)
0 ok (key was not in table before this call)
Parameters
handle (\(\ge 0\)) handle to the table
value the value to be associated with the key
key a pointer to the key (a C-style null-terminated string)
Discussion
The key may be any C character string which does not contain a slash character (’/’).
The value is stored as a 1-element array.
This function invalidates any iterators for the table which are not in the “null-pointer” state.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGet*Array() get an array value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
#include <math.h> #include "util_ErrorCodes.h" #include "util_Table.h" CCTK_COMPLEX16 z; int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetInt(handle, 42, "the answer"); Util_TableSetReal(handle, 299792458.0, "speed of light"); z.Re = cos(0.37); z.Im = sin(0.37); Util_TableSetComplex16(handle, z, "my complex number");
This is a family of functions, one for each Cactus data type, to set the value associated with a specified key to be a copy of a specified array
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableSetXxxArray(int handle, int N_elements, const CCTK_XXX array[], const char *key);
where XXX is one of POINTER, FPOINTER4 , CHAR, BYTE, INT, INT1, INT2, INT4, INT8, REAL, REAL4, REAL8, REAL16, COMPLEX, COMPLEX8, COMPLEX16, COMPLEX32 (not all of these may be supported on any given system)
call Util_TableSetXxxArray(status, handle, N_elements, array, key) integer status, handle, N_elements CCTK_XXX(*) array character*(*) key
where CCTK_XXX may be any data type supported by C (above)
Result
1 ok (key was already in table before this call, old value was replaced)
(it doesn’t matter what the old value’s type_code and N_elements were, i.e. these do not have to match the
new value)
0 ok (key was not in table before this call)
Parameters
handle (\(\ge 0\)) handle to the table
N_elements (\(\ge 0\)) the number of array elements in array[]
array a pointer to the array (a copy of which) is to be associated with the key
key a pointer to the key (a C-style null-terminated string)
Discussion
The key may be any C character string which does not contain a slash character (’/’).
Note that empty (0-element) arrays are ok.
This function invalidates any iterators for the table which are not in the “null-pointer” state.
Note that the table makes (stores) a copy of the array you pass in, so it’s somewhat inefficient to store a large array (e.g. a grid function) this way. If this is a problem, consider storing a CCTK_POINTER (pointing to the array) in the table instead. (Of course, this requires that you ensure that the pointed-to data is still valid whenever that CCTK_POINTER is used.)
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGet*Array() get an array value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableSet*() set a single (1-element array) value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_BAD_INPUT N_elements \(< 0\)
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" #define N_DIGITS 5 static const CCTK_INT pi_digits[N_DIGITS] = {3, 14, 159, 2653, 58979}; int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetIntArray(handle, N_DIGITS, pi_digits, "digits of pi");
Sets values in a table based on a string argument, which is interpreted with “parameter-file” semantics
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int count = Util_TableSetFromString(int handle, const char *string);
call Util_TableSetFromString(count, handle, string) integer count, handle character*(*) string
Result
count (\(\ge 0\)) the number of key/value entries set
Parameters
string a pointer to a C-style null-terminated string specifying the table entries to be set (see below for details
on the string contents)
Discussion
The string should contain a sequence of zero or more key=value “assignments”, separated by whitespace. This function processes these assignments in left-to-right order, setting corresponding key/value entries in the table.
The present implementation only recognises integer, real, and character-string values (not complex), and integer
and real arrays. To be precise, the string must match the following BNF:
string \(\rightarrow \) | assign* |
assign \(\rightarrow \) | whitespace* |
assign \(\rightarrow \) | whitespace* key whitespace* = whitespace* value delimiter |
key \(\rightarrow \) | any string not containing ’/’ or ’=’ or whitespace |
value \(\rightarrow \) | array \(\big |\) int_value \(\big |\) real_value \(\big |\) string_value |
array \(\rightarrow \) | { int_value* }\(\big |\){ real_value } |
int_value \(\rightarrow \) | anything recognized as a valid integer by strtol(3) in base 10 |
real_value \(\rightarrow \) | anything not recognized as a valid integer by strtol(3) but recognized as valid by strdod(3) |
string_value \(\rightarrow \) | a C-style string enclosed in ”double quotes” (C-style character escape codes are allowed, i.e. bell (’∖a’), backspace (’∖b’), form-feed (’∖f’), newline (’∖n’), carriage-return (’∖r’), tab (’∖t’), vertical-tab (’∖v’), backslash (’∖∖’), single-quote (’∖’’), double-quote (’∖"’), question-mark (’∖?’)) |
string_value \(\rightarrow \) | a C-style string enclosed in ’single quotes’ (C-style character escape codes are not allowed, i.e. every character within the string is interpreted literally) |
delimiter \(\rightarrow \) | end-of-string \(\big |\) whitespace |
whitespace \(\rightarrow \) | blank (’ ’) \(\big |\) tab (’∖t’) \(\big |\) newline (’∖n’) \(\big |\) carriage-return (’∖r’) \(\big |\) form-feed (’∖f’) \(\big |\) vertical-tab (’∖v’) |
Notice also that the keys allowed by this function are somewhat more restricted than those allowed by the other Util_TableSet*() functions, in that this function disallows keys containing ’=’ and/or whitespace.
If any error code is returned, assignments lexicographically earlier in the input string than where the error was detected will already have been made in the table. Unfortunately, there is no easy way to find out where the error was detected. :(
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGet*Array() get an array value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_NO_MEMORY unable to allocate memory
UTIL_ERROR_BAD_KEY invalid input: key contains invalid character
UTIL_ERROR_BAD_INPUT invalid input: can’t parse input string
UTIL_ERROR_NO_MIXED_TYPE_ARRAY invalid input: different array values have different datatypes
other error codes this function may also return any error codes returned by Util_TableSetString(),
Util_TableSetInt(), Util_TableSetReal(), Util_TableSetIntArray(), or Util_TableSetRealArray().
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" /* suppose we have a table referred to by handle */ /* then the call... */ int count = Util_TableSetFromString(handle, "n = 6\t" "dx = 4.0e-5\t" "pi = 3.1\t" "s = ’my string’\t" "array = { 1 2 3 }"); /* ... will return count=5 ... */ /* ... and is otherwise equivalent to the five calls ... */ CCTK_INT array[] = {1, 2, 3}; Util_TableSetInt(handle, 6, "n"); Util_TableSetReal(handle, 4.0e-5, "dx"); Util_TableSetReal(handle, 3.1, "pi"); Util_TableSetString(handle, "my string", "s"); Util_TableSetIntArray(handle, 3, array, "array");
Set the value associated with a specified key to be a specified single (1-element array) value, whose data type is generic. That is, the value is specified by a CCTK_VARIABLE_* type code and a void * pointer.
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableSetGeneric(int handle, int type_code, const void *value, const char *key);
call Util_TableSetGeneric(status, handle, type_code, value, key) integer status, handle, type_code CCTK_POINTER value character*(*) key
Result
1 ok (key was already in table before this call, old value was replaced)
(it doesn’t matter what the old value’s type_code and N_elements were, i.e. these do not have to match the
new value)
0 ok (key was not in table before this call)
Parameters
handle (\(\ge 0\)) handle to the table
type_code the array elements’ type code (one of the CCTK_VARIABLE_* constants from "cctk_Constants.h")
value_ptr a pointer to the value to be associated with the key
key a pointer to the key (a C-style null-terminated string)
Discussion
The key may be any C character string which does not contain a slash character (’/’).
The value is stored as a 1-element array.
This function invalidates any iterators for the table which are not in the “null-pointer” state.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGet*Array() get an array value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_BAD_INPUT type_code is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
#include "util_Table.h" #include "cctk_Constants.h" const CCTK_INT i = 42; const void *iptr = (void *) &i; CCTK_INT icopy; const CCTK_REAL x = 299792458.0; const void *xptr = (void *) &x; CCTK_REAL xcopy; const int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetGeneric(handle, CCTK_VARIABLE_INT, iptr, "the answer"); Util_TableSetGeneric(handle, CCTK_VARIABLE_REAL, xptr, "speed of light"); /* gets icopy to 42 */ Util_TableGetInt(handle, &icopy, "the answer"); /* gets xcopy to 299792458.0 */ Util_TableGetReal(handle, &xcopy, "speed of light");
Set the value associated with a specified key to be a copy of a specified array, whose data type is generic. That is, the array is specified by a CCTK_VARIABLE_* type code, a count of the number of array elements, and a void * pointer.
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableSetGenericArray(int handle, int type_code, int N_elements, const void *array, const char *key);
call Util_TableSetGenericArray(status, . handle, . type_code, . N_elements, array, . key) integer status, handle, type_code, N_elements CCTK_POINTER(*) array character*(*) key
Result
1 ok (key was already in table before this call, old value was replaced)
(it doesn’t matter what the old value’s type_code and N_elements were, i.e. these do not have to match the
new value)
0 ok (key was not in table before this call)
Parameters
handle (\(\ge 0\)) handle to the table
type_code the array elements’ type code (one of the CCTK_VARIABLE_* constants from "cctk_Constants.h")
N_elements (\(\ge 0\)) the number of array elements in array[]
value_ptr a pointer to the value to be associated with the key
key a pointer to the key (a C-style null-terminated string)
Discussion
The key may be any C character string which does not contain a slash character (’/’).
The value is stored as a 1-element array.
This function invalidates any iterators for the table which are not in the “null-pointer” state.
Note that the table makes (stores) a copy of the array you pass in, so it’s somewhat inefficient to store a large array (e.g. a grid function) this way. If this is a problem, consider storing a CCTK_POINTER (pointing to the array) in the table instead. (Of course, this requires that you ensure that the pointed-to data is still valid whenever that CCTK_POINTER is used.)
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGet*Array() get an array value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetFromString() [B131] convenience routine to set key/value entries in a table based on a
parameter-file–like character string
Util_TableSetString() [B145] set a character-string value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_BAD_INPUT type_code is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
#include "util_Table.h" #include "cctk_Constants.h" #define N_IARRAY 3 const CCTK_INT iarray[N_IARRAY] = {42, 69, 105}; const void *iarray_ptr = (void *) iarray; CCTK_INT iarray2[N_IARRAY]; #define N_XARRAY 2 const CCTK_REAL xarray[N_XARRAY] = {6.67e-11, 299792458.0}; const void *xarray_ptr = (void *) xarray; CCTK_REAL xarray2[N_XARRAY]; const int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetGenericArray(handle, CCTK_VARIABLE_INT, N_IARRAY, iarray_ptr, "my integer array"); Util_TableSetGenericArray(handle, CCTK_VARIABLE_REAL, N_XARRAY, xarray_ptr, "my real array"); /* gets iarray2[0] = 42, iarray2[1] = 69, iarray2[2] = 105 */ Util_TableGetIntArray(handle, N_IARRAY, iarray2, "my integer array"); /* gets xarray2[0] = 6.67e-11, xarray2[1] = 299792458.0 */ Util_TableGetRealArray(handle, N_XARRAY, xarray2, "my real array");
Sets the value associated with a specified key in a table, to be a copy of a specified C-style null-terminated character string
Synopsis
#include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TableSetString(int handle, const char *string, const char *key);
call Util_TableSetString(status, handle, string, key) integer status, handle character*(*) string, key
Result
Results are the same as all the other Util_TableSet*() functions:
1 ok (key was already in table before this call, old value was replaced)
(it doesn’t matter what the old value’s type_code and N_elements were, i.e. these do not have to match the
new value)
0 ok (key was not in table before this call)
Parameters
handle (\(\ge 0\)) handle to the table
string a pointer to the string (a C-style null-terminated string)
key a pointer to the key (a C-style null-terminated string)
Discussion
The key may be any C character string which does not contain a slash character (’/’).
The string is stored as an array of strlen(string) CCTK_CHARs. It does not include a terminating null character.
This function is very similar to Util_TableSetCharArray().
This function invalidates any iterators for the table which are not in the “null-pointer” state.
See Also
Util_TableCreateFromString() [B49] convenience routine to create a table and set key/value entries in it
based on a parameter-file–like character string
Util_TableGet*() get a single (1-element array) value, or more generally the first array element of an array
value
Util_TableGet*Array() get an array value
Util_TableGetGeneric() [B68] get a single (1-element array) value with generic data type
Util_TableGetGenericArray() [B72] get an array value with generic data type
Util_TableGetString() [B76] get a character-string value
Util_TableSetCharArray() [B127] get an array-of-CCTK_CHAR value
Util_TableSet*() set a single (1-element array) value
Util_TableSet*Array() set an array value
Util_TableSetGeneric() [B136] set a single (1-element array) value with generic data type
Util_TableSetGenericArray() [B140] set an array value with generic data type
Util_TableSetCharArray() [B127] set an array-of-CCTK_CHAR value
Errors
UTIL_ERROR_BAD_HANDLE handle is invalid
UTIL_ERROR_TABLE_BAD_KEY key contains ’/’ character
UTIL_ERROR_NO_MEMORY unable to allocate memory
Examples
#include "util_ErrorCodes.h" #include "util_Table.h" static const CCTK_CHAR array[] = {’r’, ’e’, ’l’, ’a’, ’t’, ’i’, ’v’, ’i’, ’t’, ’y’}; #define N_ARRAY (sizeof(array) / sizeof(array[0])) int handle = Util_TableCreate(UTIL_TABLE_FLAGS_DEFAULT); Util_TableSetString(handle, "relativity", "Einstein"); /* this produces the same table entry as the Util_TableSetString() */ Util_TableSetCharArray(handle, N_ARRAY, array, "Einstein");
Print out a table and its data structures, using a verbose internal format meant for debugging
Synopsis
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TablePrint(FILE *stream, int handle);
Result
0 ok
Parameters
stream (\(\ne 0\)) output stream, e.g. stdout
handle (\(\ge 0\)) handle to the table
Discussion
stream may be any output stream, e.g. stdout or stderr, or a file that has been opened for writing.
See Also
Util_TablePrintAll() [B152] Print out all tables and their data structures, using a verbose internal format
meant for debugging
Util_TablePrintAllIterators() [B154] Print out all table iterators and their data structures, using a verbose
internal format meant for debugging
Util_TablePrintPretty() [B156] Print out a table, using a human-readable format similar to the one accepted
by Util_TableCreateFromString
Examples
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableCreateFromString("ipar=1 dpar=2.0 spar=’three’"); Util_TablePrint(stdout, handle);
Print out all tables and their data structures, using a verbose internal format meant for debugging
Synopsis
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TablePrintAll(FILE *stream);
Result
0 ok
Parameters
stream (\(\ne 0\)) output stream, e.g. stdout
Discussion
stream may be any output stream, e.g. stdout or stderr, or a file that has been opened for writing.
See Also
Util_TablePrint() [B149] Print out a table and its data structures, using a verbose internal format meant for
debugging
Util_TablePrintAllIterators() [B154] Print out all table iterators and their data structures, using a verbose
internal format meant for debugging
Util_TablePrintPretty() [B156] Print out a table, using a human-readable format similar to the one accepted
by Util_TableCreateFromString
Examples
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableCreateFromString("ipar=1 dpar=2.0 spar=’three’"); Util_TablePrintAll(stdout);
Print out all table iterators and their data structures, using a verbose internal format meant for debugging
Synopsis
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TablePrintAllIterators(FILE *stream);
Result
0 ok
Parameters
stream (\(\ne 0\)) output stream, e.g. stdout
Discussion
stream may be any output stream, e.g. stdout or stderr, or a file that has been opened for writing.
See Also
Util_TablePrint() [B149] Print out a table and its data structures, using a verbose internal format meant for
debugging
Util_TablePrintAll() [B152] Print out all tables and their data structures, using a verbose internal format
meant for debugging
Util_TablePrintPretty() [B156] Print out a table, using a human-readable format similar to the one accepted
by Util_TableCreateFromString
Examples
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableCreateFromString("ipar=1 dpar=2.0 spar=’three’"); Util_TablePrintAllIterators(stdout);
Print out a table, using a human-readable format similar to the one accepted by Util_TableCreateFromString
Synopsis
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int status = Util_TablePrintPretty(FILE *stream, int handle);
Result
0 ok
Parameters
stream (\(\ne 0\)) output stream, e.g. stdout
handle (\(\ge 0\)) handle to the table
Discussion
stream may be any output stream, e.g. stdout or stderr, or a file that has been opened for writing.
See Also
Util_TableCreateFromString() [B49] Create a new table (with the case-insensitive flag set) and set values in
it based on a string argument (interpreted with “parameter-file” semantics)
Util_TablePrint() [B149] Print out a table and its data structures, using a verbose internal format meant for
debugging
Util_TablePrintAll() [B152] Print out all tables and their data structures, using a verbose internal format
meant for debugging
Util_TablePrintAllIterators() [B154] Print out all table iterators and their data structures, using a verbose
internal format meant for debugging
Examples
#include <stdio.h> #include "util_ErrorCodes.h" #include "util_Table.h" int handle = Util_TableCreateFromString("ipar=1 dpar=2.0 spar=’three’"); Util_TablePrintPretty(stdout, handle);
In this chapter all Driver_* Cactus functions are described. These functions are callable from C or Fortran thorns.
In the functions below, where refers to an integer whose values are defined in cctk_Constants.h. The basic values are WH_INTERIOR, WH_BOUNDARY, and WH_GHOSTS. Other values are created by bitwise or: WH_EXTERIOR = WH_INTERIOR | WH_GHOSTS, WH_EVERYWHERE = WH_INTERIOR | WH_EXTERIOR.
[C4] Gets the region where a grid function is valid.
[C6] Notify the driver as to which region of which grid function you have updated.
[C8] Ask the driver to provide valid data in the region you specify.
[C10] Tell the driver how to update the boundary conditions for the grid functions in a group.
[C13] Tell the driver how to update the boundary conditions for a grid functions.
[C16] Sets the region where a grid function is valid.
Gets the region where a grid function is valid.
Synopsis
#include "cctk.h" int where = CCTK_GetValidRegion(int variable_index ,int time_level);
Result
where Region where the given grid variable is valid at the given time level.
Parameters
variable_index The index of a grid function.
Parameters
time_level The time level of a grid function.
See Also
Driver_SetValidRegion [C16] Sets the region where a grid function is valid.
Errors
Abort Assertions may be triggered for invalid variables or time levels.
Notify the driver as to which region of which grid function you have updated.
Synopsis
#include "cctk.h" int zero = CCTK_NotifyDataModified(const cGH *cctkGH, int *variable_list, int *time_level_list, int num_variables, int *where_list);
Result
zero Always returns zero.
Parameters
cctkGH Pointer to CCTK grid hierarchy
Parameters
variable_list An array of num_variables grid function indexes.
Parameters
time_level_list An array of num_variables time levels.
Parameters
num_variables The number of variable index, time level, where tuples.
Parameters
where_list An array of num_variables where specifications describing where valid data was modified for a
given grid function and time level.
See Also
Driver_RequireValidData [C8] Ask the driver to provide valid data in the region you specify.
Errors
Abort Assertions may be triggered for invalid variables or time levels.
Synopsis
#include "cctk.h" int zero = CCTK_RequireValidData(const cGH *cctkGH, int *variable_list, int *time_level_list, int num_variables, int *where_list);
Result
zero Always returns zero.
Parameters
cctkGH Pointer to CCTK grid hierarchy
Parameters
variable_list An array of num_variables grid function indexes.
Parameters
time_level_list An array of num_variables time levels.
Parameters
num_variables The number of variable index, time level, where tuples.
Parameters
where_list An array of num_variables where specifications describing where valid data is needed for a given
grid function and time level.
See Also
Driver_NotifyDataModified [C6] Notify the driver as to which region of which grid function you have updated.
Errors
Abort Assertions may be triggered for invalid variables or time levels.
Tell the driver how to update the boundary conditions for the grid functions in a group.
Synopsis
#include "cctk.h" int err = Driver_SelectGroupForBC( const cGH *cctkGH, int faces, int width, int table handle, const char *group_name, const char *bc_name);
Result
err Returns -6 for an invalid group index, -2 for an invalid boundary condition name.
Parameters
cctkGH Pointer to CCTK grid hierarchy
Parameters
faces The choice of which faces the boundary condition applies to. Normally, this will be CCTK_ALL_FACES. If a
different specification is desired, consult the documentation on the Boundary thorn.
Parameters
width The number of zones from the edge filled in by the boundary condition.
Parameters
table_handle The table handle holds extra arguments for the boundary condition, if such are needed.
Use -1 (an invalid table handle) for boundary conditions which need no additional arguments.
Parameters
where_list An array of num_variables where specifications describing where valid data is needed for a given
grid function and time level.
Parameters
group_name The name of the group of grid functions to which this boundary condition applies.
Parameters
bc_name The name of the boundary condition registered physical boundary condition. See the documentation
for Boundary_RegisterPhysicalBC on the Boundary thorn.
See Also
Driver_SelectVarForBC [C13] Tell the driver how to update the boundary conditions for a grid functions.
Errors
Abort Assertions may be triggered for invalid variables or time levels.
Synopsis
#include "cctk.h" int err = Driver_SelectVarForBC( const cGH *cctkGH, int faces, int width, int table handle, const char *var_name, const char *bc_name);
Result
err Returns -7 for an invalid variable index, -2 for an invalid boundary condition name.
Parameters
cctkGH Pointer to CCTK grid hierarchy
Parameters
faces The choice of which faces the boundary condition applies to. Normally, this will be CCTK_ALL_FACES. If a
different specification is desired, consult the documentation on the Boundary thorn.
Parameters
width The number of zones from the edge filled in by the boundary condition.
Parameters
table_handle The table handle holds extra arguments for the boundary condition, if such are needed.
Use -1 (an invalid table handle) for boundary conditions which need no additional arguments.
Parameters
where_list An array of num_variables where specifications describing where valid data is needed for a given
grid function and time level.
Parameters
var_name The name of the grid function to which this boundary condition applies.
Parameters
bc_name The name of the boundary condition registered physical boundary condition. See the documentation
for Boundary_RegisterPhysicalBC on the Boundary thorn.
See Also
Driver_SelectVarForBC [C10] Tell the driver how to update the boundary conditions for the grid functions in a
group.
Errors
Abort Assertions may be triggered for invalid variables or time levels.
Gets the region where a grid function is valid.
Synopsis
#include "cctk.h" CCTK_SetValidRegion(int variable_index ,int time_level,int where);
Parameters
variable_index The index of a grid function.
Parameters
time_level The time level of a grid function.
Parameters
where The new where specification for the variable and time level.
See Also
Driver_GetValidRegion [C4] Gets the region where a grid function is valid.
Errors
Abort Assertions may be triggered for invalid variables or time levels.