BOOTSTRAP (kutils) functions


LIBRARY ROUTINE

kparse_string_scan_delimit - Break a string into an array of strings

LIBRARY CALL

char **kparse_string_scan_delimit(
   char    *data,
   char    *key1,
   char    *key2,
   int      mode,
   char    *delimiters,
   char   **key1_format,
   char   **key2_format,
   ssize_t  *num,
   int     *status)

INPUT

OUTPUT

RETURN VALUE

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

DESCRIPTION

This routine looks in a data string for an area of text between two user specified match keys, then it delimits the section of text in to an array of smaller strings based on a set of character delimiters. Delimiters can be escaped by a '\\' if they need to appear in the text section. This routine is a combination of the calls kparse_string_scan() and kparse_string_delimit(). This routine cleans up the entries in the array via a kstring_cleanup() call.

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 the key1_format and key2_format parameters. Thus, the user should pass in addresses of an unused character pointers for them. The calling routine is responsible for freeing the space malloc'ed for the key1_format and key2_format parameters.

This routine mallocs the space for the return string; and hence, is responsible for freeing the that space via kfree_and_NULL() when they are done with it.

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_string_scan(kutils), kparse_string_delimit(kutils), karray_add(kutils), karray_free(kutils)

COPYRIGHT

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