BOOTSTRAP (kutils) functions


LIBRARY ROUTINE

kparse_file_search_delimit - break up a line data into an array of strings based on some set of delimiters

LIBRARY CALL

char **kparse_file_search_delimit(
   kfile    *file,
   char     *key,
   int       mode,
   char     *delimiters,
   char     *cont,
   char    **key_format,
   ssize_t  *num,
   int      *status)

INPUT

OUTPUT

RETURN VALUE

This routine returns a pointer to an array of items we just broke apart from the input data string. NULL is returned when the kparse_file_search() or kparse_string_delimit() fails.

DESCRIPTION

This routine looks for a user specified key with a call to kparse_file_search(); then, it reads in the rest of the line into a string, up to a '\\n'. Once the line is read in, it breaks up the line according to the user's list of delimiters by calling kparse_string_delimit(). The user can specify a set of continuation characters. Continuation characters must appear as the last character of the line. Line continuation can be chained together at the end of each consecutive line you want this routine to parse as a single line.

ADDITIONAL INFORMATION

The regular expression syntax supported by the parser is: '.', '*', '[..]', '[^..]', '$', '^', and '\\'. The explainations for each syntax is explained below.

. Match any single character except newline

* Match the preceding character or ranger of characters 0 or more times. This matching includes items specified within a [...] construct.

[...] or [^...] Matches any one character contained within the brackets. If the first character after the is a ']', then it is included as a character to match. If the first character after the '[' is a '^', then it will match all characters NOT included within the []. Finally, the '-' will indicate a range of characters. For example, [a-z] specifies all characters between and including the ascii values 'a' and 'z'. If the '-' follows the '[', or is right before the ']', then it is interpreted literally.

^ If it is the first character of the regular expression, it matches the beginning of the line.

$ If it is the last character of the regular expression, it matches the end of line. However, it does NOT put the \\n in your key when the matched key is returned.

\e This escapes the meaning of a special character.

EXAMPLES

none

SIDE EFFECTS

This routine mallocs data and sets the value of key_format parameter. Thus, the user should pass in an address of an unused character pointer. The calling routine is responsible for freeing the space malloc'ed for the key_format parameter.

This routine creates a new array of strings, and the calling routine is responsible for freeing the space allocated while creating the array via a call to karray_free().

RESTRICTIONS

It does not support the following regular expression constructs: or'ing '|', grouping of regular expressions '()', match one or more times '+', or match n to m times '\\{n,m\\}'. Finally, the '\\number' and '\\(\\)' constructs have no meaning for these routines, so they are not supported either.

Search keys and the data file should not contain the values '\\001', '\\002', '\\003', or '\\004', because these values are used as special search parameters by the parser.

MODIFICATION

none

FILES

$BOOTSTRAP/objects/library/kutils/src/parse.c

SEE ALSO

kparse_file_search(kutils), kparse_string_delimit(kutils), karray_free(kutils)

COPYRIGHT

Copyright (C) 1993 - 1997, Khoral Research, Inc. ("KRI") All rights reserved.