FastSort Manual

Using FastSort Commands
FastSort Manual—429834-003
3-6
COLLATE Command
SUBSORT
deletes current parameters of all SUBSORT commands.
TO
deletes all parameters for the current output file.
Examples
CLEAR DESC
CLEAR FROM FILETEN
CLEAR TO
COLLATE Command
Use the COLLATE command to specify an alternate collating sequence during a sort or
merge run. This enables you to define the comparison of alphanumeric key fields or
string-type data.
If you do not specify an alternate collating sequence, FastSort uses the sequence of
the ASCII character set to order your results. For more information, see Appendix D,
ASCII Character Set.
filename
is the name of an EDIT file containing a list of characters assigned to the 256 byte
positions.
The sequence in which FastSort reads the file determines the character or
characters assigned to each byte position. For a description of the alternate
collating sequence file, see the text that follows.
The COLLATE command directs FastSort to read the EDIT file indicated by file and
to translate the character list into an alternate collating sequence table. You can use
the COLLATEOUT command to store this alternate collating sequence table in an
unstructured file that the SORTMERGESTART procedure can use.
FastSort orders characters in the assignment list in the order it reads the characters
from the alternate collating sequence file. The first character or character range in the
list ranks before the second character or character range, and so on.
Specifying an Alternate Collating Sequence
You can specify an alternate collating sequence in an EDIT file by assigning one or
more characters to each of the 256 byte positions. The EDIT file can contain only
ranges of character assignments for the collating sequence, commas to separate
COLLATE filename