Thunderstone Support Forums

How do you metaparse(get the timport expression) for an xml feed?
I'm using 'recexpr' - I was given a URL that returns results. The results seem congested with xml DTD's and I can't get it to work. What Am I doing wrong?

The xml url:
http://meta.7search.com/scripts/meta.as ... &r=10&pn=1

The format there would be better handled with an expression per field.
field Url varchar(80) />>\R<URL><\!\[CDATA\[\P=!]]>*\F]]> ''

Also make sure you use the multiple keyword.
You may want to set a recdelim of
recdelim \R\x0a</SITE>\x0a

Hello, I understand the "expression per field" for the format part. Can you elaborate on "use the multiple keyword" and "set a RECDELIM of \R\x0a</SITE>\x0a"
Where would that go under recexpr?

Thanks, Paul

I'm trying to incorporate parsed results from an xml feed and when I try to execute I receive the error message:
015 /webinator/narrow3:107: Argument must be single variable or literal

I tried to look up info on message boards without any luck. Can you help

The order of non-field keywords within the schema does not matter. Also, don't forget the online manual:
http://thunderstone.master.com/texis/ma ... q=recdelim

http://thunderstone.master.com/texis/ma ... literal%22

I'm having a difficult time figuring this out. I was wondering - do you guys have a service whereas I could pay you to parse the xml url for me?

I'm using 'recexpr' - for the URL :
http://meta.7search.com/scripts/meta.as ... &r=10&pn=1

lmk, Paul

As mentioned in the message at http://thunderstone.master.com/texis/ma ... m37a875ed0 we can not help create meta-search engines which operate against general internet search engines, although the software is capable. We can help with general timport questions.

I'm not "operating against general internet search engines" the search engine I'm dealing with has provided me with a customised xml feed. I just have to be able to parse these results into my page. I tried numerous times to no avail. At this point, I would rather pay someone who knows what they're doing then waste countless hours trying to "maybe" figure it out myself.

Hello I am trying to parse an xml feed
http://searchalot.overture.com/d/search ... words=cars

I tried using different expressions to make it pull out the data but nothing seems to work.

Here is what I have to just pull out the URL
<$imports= '#OVERTURE
multiple
recexpr
field Link varchar(80) />>\RclickUrl=\"\P=!\"\'>

But it doesn't work, any suggestions would be greatly appreciated.

Regards,
Gerald

You need to escape the '=' after clickUrl, otherwise it parsed as clickUrl immediately followed by the double-quote. Instead of !\" it is more efficient to say [^\"], and you probably want a + after it to get more than one character.

If you have full Texis with the rex program running rex -x and an expression can be useful to see what it will pick out.