gdaLoad

Purpose

Loads variables in a GDA into the workspace.

Format

retcode = gdaLoad(filename, create, modify, rename, ftypes, errh, report)
Parameters:

  • filename (string) – name of data file.

  • create (scalar) –

    create flag:

    0

    do not create any new variables in the workspace.

    1

    create new variables in the workspace.

  • modify (scalar) –

    modify flag:

    0

    do not modify any variables in the workspace.

    1

    if the name of a variable in the data file matches the name of a variable already in the workspace, modify that variable.

  • rename (scalar) –

    rename flag:

    0

    do not rename a variable retrieved from the data file when copying it into the workspace.

    1

    rename variables retrieved from the data file when copying them into the workspace if there are name conflicts with existing variables, which may not be modified.

  • ftypes (scalar) –

    type force flag:

    0

    do not force a type change on any variables in the workspace when modifying.

    1

    force a type change on a variable in the workspace when modifying it with the data in a variable of the same name in the data file. Note that if ftypes is set to 1, gdaLoad() will follow regular type change rules. The types of sparse matrix and structure variables will NOT be changed.

  • errh (scalar) –

    controls the error handling of gdaLoad():

    0

    skip operations that cannot be performed, without setting an error return.

    1

    return an error code if operations are skipped.

    2

    terminate program if operations are skipped.

  • report (scalar) –

    controls reporting:

    0

    no reporting.

    1

    report only name changes and operations that could not be performed.

    2

    report type changes, name changes, and operations that could not be performed.

    3

    report everything.

Returns:

retcode (scalar) –

return code, 0 if successful, otherwise one of the following error codes:

4

File read error.

5

Invalid file type.

10

File contains no variables.

14

File too large to be read on current platform.

24

Variables skipped.

26

Cannot add structure definition.

27

Structure definition does not match.

Examples

/*
** Create new variables if same name
** does not exist
*/
create_new = 1;

/*
** Modify variable if name matches
** existing variable
*/
modify = 1;

/*
** Rename variables retrieved from the
** data file when copying them into the
** workspace if there are name conflicts
*/
rename = 1;

/*
** Force a type change on a variable in
** the workspace when modifying it
*/
ftypes = 1;

/*
** Return an error code if operations
** are skipped
*/
errh = 1;

/*
** Report only name changes and operations
** that could not be performed
*/
retcode = gdaLoad("myfile.gda", create_new, modify, rename, ftypes, errh, report)

This example loads the variables in myfile.gda into the workspace, creating a new variable if a variable of the same name does not already exist, modifying an existing variable if a variable of the same name does already exist and the modification does not result in an impossible type change, and renaming the variable if none of the above is possible. The example returns an error code if any variables in myfile.gda are skipped and reports all activity.

Remarks

For each variable in filename, gdaLoad() will first compare the name of the variable against the names of the variables already resident in the GAUSS workspace to see if there is a match. If there is not a match, and create is set to 1, it will create a new variable. Otherwise if create is set to 0, it will skip that variable.

If the variable name does match that of a variable already resident in the GAUSS workspace, and modify is set to 1, it will attempt to modify that variable. If the types of the two variables are different, and ftype is set to 1, it will force the type change if possible and modify the existing variable.

If it cannot modify the variable or modify is set to 0, it will check to see if rename is set to 1, and if so, attempt to rename the variable, appending an _ num to the variable name, beginning with \(num = 1\) and counting upward until it finds a name with which there are no conflicts. If the variable cannot be modified and rename is set to 0, then the variable will be skipped.

The rename argument also controls the handling of structure definitions. If a structure variable is encountered in the GDA file, and no variable of the same name exists in the workspace (or the variable is renamed), gdaLoad() will attempt to find a structure definition in the workspace that matches the one in the GDA. Note that in order for structure definitions to match, the structure definition names must be the same as well as the number, order, names, and types of their members.

If no matching structure definition is found, the definition in the file will be loaded into the workspace. If there is already a non-matching structure definition with the same name in the workspace and rename is set to 1, then gdaLoad() will attempt to rename the structure definition, using the same method as it does for variable names.

If a structure variable is encountered in the GDA file, a structure variable of the same name already exists in the workspace, and modify is set to 1, then gdaLoad() will modify the existing variable, providing that the structure definitions of the two variables match.

See also

Functions gdaSave()

previous

gdaIsCplx

next

gdaPack