Subj : Re: <no subject> (How to parse markup)
To   : netscape.public.mozilla.jseng
From : Shanti Rao
Date : Sat Jul 17 2004 10:10 am

Here's my recipe:

1. Scan input until you encounter a <. Write all skipped characters to output.
2. When you find a <, read the next character. If it's whitspace, send the < 
and the space to output. Goto 1.
3. If it's a punctuation symbol, ignore the tag and send it to output. You 
don't want to parse <!DOCTYPE>, etc.
4. If it's not whitespace, it's the first character of the tag name. Keep 
reading from input until you reach whitespace, /, or >.
5. If the tag is terminated by whitespace, read attributes. Create a new 
object to be the hash table. Call Read Attributes.
6. Note whether the tag ended with /> or just >.

Sub Read Attributes

1. Read attribute name until any of "= \t\r\n/>"
2. If whitespace, skip to non-ws character
3. If / or >, tag is finished. Return.
4. If =, attribute value comes next.
5. Read from input until non-whitespace
6. If " or ', read until matching quote symbol. Goto 1.
7. Read until whitespace or / or >
8. If /, read until >
9. Done if >, else goto 1.




Axel Bernhardt wrote:
> hello,
> 
> has anybody experience with preparsing html? 
> i'm looking through httpUnit, try to work with jTidy. but i have no idea, where to start.
> 
> my js to be parsed is like that:
> <script type="text/javascript" language="javascript" src="http://a.as-eu.falkag.net/dat/dlv/aslmain.js">
> </script>
> 
> has anybody an example or a few codelines, how to start with parsing? i know, i asked before, but i still got no idea.
> 
> 
> Thanks for helping,
> juppi
> 
> _______________________________________________________
> WEB.DE Video-Mail - Sagen Sie mehr mit bewegten Bildern
> Informationen unter: http://freemail.web.de/?mc=021199
>

.