Re: analyze-string help?

On 10 June 2012 10:45, Michael Kay <mike@xxxxxxxxxxxx> wrote:
>
>
> On 10/06/2012 07:46, Syd Bauman wrote:
>
> I think this principle of always using the simplest regex that does the job
> is a very good one. Remember, even if the regex works on all your test
> cases, there's a good chance someone in a year's time is going to come back
> to it and need to change it. Complex regexes are intrinsically very hard to
> debug, so it's a good idea to avoid them if you can.

Then, a corollary of this argument would certainly be that the
simplest available
*technique* is a very good one. substring-before/after comes to my mind even
though this splits on the first occurrence. A fn:string-split would be
nice to have,
but it's easy to implement.

-W