Re: Processing two documents, which order?

I've been using expressions such as

<xsl:variable name="prefix"    select='"(^|\W)"'/>
<xsl:variable name="suffix"    select='"($|\W)"'/>
<xsl:variable name="pattern"   select="concat($prefix,$property,$suffix)"/>

Notice that
a) the hyphen is included in the set of stoppers, and
b) that you have to handle the cases  0word$, \Wword$, ^word\W and
\Wword\W separately so as to not lose a character matching the \W.

I think that the hyphen can be excluded by a character class
subtraction, but I'll have to study the syntax some more.

-W



On 8 April 2011 13:21, Dave Pawson <davep@xxxxxxxxxxxxx> wrote:
> On Fri, 08 Apr 2011 11:47:18 +0100
> David Carlisle <davidc@xxxxxxxxx> wrote:
>
>> On 08/04/2011 11:11, Dave Pawson wrote:
>> > I really couldn't contemplate writing such a regex to be honest
>> > David.
>>
>> if you have a whitespace list of words then the regex is
>> string-join(tokenize($list,'\s+'),'|')
>> is that so hard to contemplate?
>> or you could, as Wolfgang just commented, try to give the regex
>> engine a helping hand by pre-optimizing that regex by pulling out
>> common prefixes.
>>
>> David
>
> With Wolfgangs notes, I'm struggling with the regex for one.
>
>  <xsl:analyze-string select="."
>  regex="[^a-z]({$property})[\s.,&#x22;,&#x27;&lt;$]">
>
> Is currently missing >keyword ....
> and
> ... keyword<....
>
> Since checking for frisbee (where bee is the keyword)
> causes >bee .... to be missed.
>
> Ditto with the termination? I'm 'requiring'
> a termination, hence missing bee</element
>
> How to add ^ and $ to the start and termination classes?
>
> regards
>
> --
>
> regards
>
> --
> Dave Pawson
> XSLT XSL-FO FAQ.
> http://www.dpawson.co.uk