RE: string search

It's probably feasible to do this in XSLT 2.0 using regular expressions,
assuming you can specify the problem more precisely - or develop the regular
expressions incrementally by trial and error over a range of source files
that are representative of the whole collection you need to process.

(If you can't specify the problem more precisely, then a different
technology isn't going to help.)

Doing it with XSLT 1.0 is likely to be more difficult.

Regards,

Michael Kay
http://www.saxonica.com/
http://twitter.com/michaelhkay

> -----Original Message-----
> From: Skopik Pavel [mailto:Pavel.Skopik@xxxxxxxxxx]
> Sent: 27 October 2009 17:33
> To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
> Subject:  string search
>
> Hello,
>
> I have come across an issue related to string search which
> seems to be much advanced then just testing its occurence or positon.
>
> My XML source has a node, which contains list of
> organizations and their roles. First comes the role, second
> the name of the subject. The subject and its role may be
> surounded by other text. What I need to do is to extract the
> name of the subject which comes after the name of the role.
>
> Source text:
> Text text text. Approves: Ministry of whatever. Text Text.
>
> Desired result:
> Ministry of whatever
>
> Of course there can be punctuation characters around the name
> of the subject, spaces etc.
>
> Do you think it is feasable even to try do something like
> this or just leave it off and try to solve this situation
> with different technology? (I am inclined to the second
> option, but would like to hear you opinion).
>
> Best regards,
>
> Pavel )kopmk
>
>
>
>
>
> __________ Informace od ESET NOD32 Antivirus, verze databaze
> 4548 (20091027) __________
>
> Tuto zpravu proveril ESET NOD32 Antivirus.
>
> http://www.eset.cz