home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.programming
- Path: sparky!uunet!spool.mu.edu!darwin.sura.net!uvaarpa!murdoch!usenet
- From: jpw@sansfoy.lib.Virginia.EDU (John Price-Wilkin)
- Subject: lex help
- Message-ID: <1992Dec23.012520.12393@murdoch.acc.Virginia.EDU>
- Keywords: lex
- Sender: usenet@murdoch.acc.Virginia.EDU
- Organization: University of Virginia
- Date: Wed, 23 Dec 1992 01:25:20 GMT
- Lines: 28
-
- My apologies if this is misdirected, but I could use some basic help with
- lex. I'm writing a filter for a tagged file and have done all of the
- basic routines (i.e., the routines my *basic* skills will cover). Some of
- the tagging I'm trying to cover looks like the following:
- <doc name="Opera" author="Anselmus">
- but sometimes they're like:
- <doc author="Anselmus">
- and sometimes like:
- <doc name="Opera">
- They are always one of the three, and the contents of the quotes are of
- variable length, often containing spaces. I'd like to render them as:
- Name: Opera
- Author: Anselmus
- (or the corresponding versions with only one of the elements). There's
- another variation on this theme, where the hierarchical tag scheme "<divN"
- is used, so that I encounter
- <div1 name="Opera" author="Anselmus">
- as well as
- <div2 name="Opera" author="Anselmus">
- and so on, through div4. My first and most immediate concern is to get a
- lex routine that will take care of each instance (e.g,. a <doc tag that
- has any one of the three configurations -- both author and name, or either
- author or name), but an ideal solution would be to have something that did
- a <doc or any <divN as well. I'd be much obliged if someone would show me
- the light on this. Replying to me personally would be swell.
-
- John Price-Wilkin
- jpw@virginia.edu
-