At 15:26 12/12/2004 -0700, Dar Scott wrote:
On Dec 12, 2004, at 3:09 PM, Randy Padawer wrote:
Let's say a text file occasionally includes the words dogs, cats, and fleas. I need to "read until" any of these words without skipping any.
Unless the file is huge, read it in in one piece with the URL method.
Then repeatedly parse it with matchText or matchChunk. Match with a pattern includes what you want and also allows you to obtain the rest of the data for processing more. I'm guessing that the pattern you want includes the front of the string, a bunch of characters, and then either one of the words. The rest is everything after that. If you need help with the regex pattern, just ask.
It would be nice to give a char # offset to matchText or matchChunk, but I don't know how.
Can't you just do matchText(char currOffset to -1 of theString, tRegEx, tVars)
Or, if that won't work,
put "." & currOffset & tRegEx into tempRegEx
matchText(theString, tempRegEx, tVars)
(so the first match is the .N where N is the offset ) ?
(Ugly, but I've seen something like this used before in Perl, where ugliness seemed natural :-)
-- Alex.
_______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
