Write a subset relation

saf_write_subset_relation is a function defined in rel.c.

Synopsis:

int saf_write_subset_relation(SAF_ParMode pmode, SAF_Rel *rel, hid_t A_type, void *A_buf, hid_t B_type, void *B_buf, SAF_Db *file)

Formal Arguments:

  • pmode: The parallel mode.
  • rel: The relation whose data is to be written.
  • A_type: The type of A_buf (if not already supplied through the saf_declare_subset_relation call).
  • A_buf: The data (if not already supplied through the saf_declare_subset_relation call).
  • B_type: The type of B_buf (if not already supplied through the saf_declare_subset_relation call.
  • B_buf: The data (if not already supplied through the saf_declare_subset_relation call).
  • file: The optional destination file to write the data to. A null pointer for this argument indicates that the data is to be written to the same file as rel.

Description: This call writes relation data to the specified file.

Preconditions:

  • pmode must be valid. (low-cost)
  • rel must be a valid relation handle. (low-cost)
  • A_buf should be specified either here or in the saf_declare_subset_relation. (low-cost)
  • B_buf, if present, should be specified either here or in the. (low-cost)

Return Value: The constant SAF__SUCCESS is returned when this function is successful. Otherwise this function either returns an error number or throws an exception, depending on the value of the library’s error handling property.

Parallel Notes: SAF_EACH mode is a collective call where each of the N tasks provides a unique relation. SAF will create a single HDF5 dataset to hold all the data and will create N blobs to point into nonoverlapping regions in that dataset.

Issues: Overwrite is not currently allowed.

See Also: