According to Geoff Hutchison:
>Hi,
>
>I just cleaned out the bug database and discovered something nice. Peter D.
>Gray <[EMAIL PROTECTED]> sent us a "hack" for changing the htdig URL
>exclusion and limit attributes into full regexp.
>
>The patches are fairly small--he introduces a minimal Regex class and then
>changed the calls to StringMatch with calls to his Regex class. I'll send
>the patches if anyone wants.
>
>So here's a question... Do we want this full regex behavior? Do we want to
>replace StringMatch with a regex class? The bonus would be full regex
>support in a number of areas. People have particularly complained about the
>restrict/exclude behavior in htdig and htsearch.
Having regex in most areas would be nice and add a lot to the usablility
of ht://Dig as a search tool. Doing regex only for a limited range would
not be worth the effort.
>The negative is that we have to field problems with incorrect regex
>(ew...), and we will probably have to work on the Regex class some. We also
>have no idea if it would be faster/slower or use more/less memory than
>StringMatch.
I think that in case of a byte compiled regex, memory usage should increase
only little and speed should be near equivalent to StringMatch for trivial
expressions.
>My feeling is we should play with it and see how it works performance-wise.
>I don't think the changes would be significant to what's contributed.
If the performance drops significantly for trivial expressions it might
even be worth thinking of optional configuration attributes that can be
used to substitute the normal StringMatch stuff, i.e. if a regex is re-
quested, the regex code will be used otherwise ht://Dig stays with old
StringMatch.
just my 2cc,
Torsten
--
InWise - Wirtschaftlich-Wissenschaftlicher Internet Service GmbH
Waldhofstra�e 14 Tel: +49-4101-403605
D-25474 Ellerbek Fax: +49-4101-403606
E-Mail: [EMAIL PROTECTED] Internet: http://www.inwise.de
------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
[EMAIL PROTECTED] containing the single word "unsubscribe" in
the SUBJECT of the message.
