CVS log for search/cgi/search.pl

Request diff between arbitrary revisions

date search experiment - some backend

date search experiment - some backend

fixing a regression bug for index/list selection

more robust sorting algorithm, see http://www.namazu.org/pipermail/namazu-users-en/2007-January/000298.html

syntax

serving XHTML (no conneg) if no query has been made.

last-mod not working well yet, commenting oot

debug OK

debuggung

debuggung

basic caching info - proper location of script

typo

basic caching info

basic caching info

commenting out decoding of quoted-printable accents and stuff, as it tends to screw up when it sees uris

fixing cross-site-scripting problem with HTML::Entities::encode()
+ a small markup fix

cleaning up links for proxying of custom search

cleaning up links for proxying of custom search

testing omit_params with custom list name parameter (not needed)

testing omit_params with custom list name parameter (not needed)

testing omit_params with custom list name parameter (not needed)

using script name rather than param to trigger custom search

starting to experiment with the idea of customn search

additional hackery for proper conversion of content to utf8

sending utf-8

fixing typos

adding form switcher

fixed escaping bug
reported in
	http://lists.w3.org/Archives/Team/w3t-archive/2004Aug/0218.html

fixing the bug:
	http://www.w3.org/Team/bugzilla/show_bug.cgi?id=198
(messing with index-group parameters for FULL indexes)

By default, mail search does NOT use slash in parameters any more (but still understands them)

Making sort by date (most recent first) the default sort

changed to prefer (crudely) rdf over soap over n3 over html

updated namazu output parsing to reflect simplified template and added threading

fixed for multi-value properties

some presentation fixup inspired by mime-type request from Graham Klyne

merging from branch 1.33.2.10

improved parser

duh, forgot to get rid of the debugging open again

fixed ordinal parsing

fixed wierdo refs and added session-ids

switch to line-counting parser

fixed some perl typos

fixed self URI refs

ready for testing

abstracted presentation layer

normalized match phrases

reverse updates to main branch -- sorry

normalized match phrases

fix for multiple list selection and CGI.pm API change, bug #157

fixed problem with !%$! message-ids with spaces

july 29th version I had forgotten

july 29th version I had forgotten

adding features asked by Martin Duerst (and others) to avoid "multiple choices" when possible

did I commit this revision already?

fixing cache info

*** empty log message ***

bug fixes:
	- linewrap in result URLs
	- page links (again)
	- added summary to tables

page url was invalid.

*** empty log message ***

fixing a typo

UI modifs (minor)

a few modifs before public instance

UI improvements.

small rewording

added "pre-fill" for index (last step)

switched to XHTML output

fixed a small (hidden) bug with index selection

added a warning if the list selection is bogus

bug fixes:
- was searching "team" only when asking for "all" in advanced mode

- names with "@w3.org" are now allowed

- the URI for next/previous pages was broken because of proxy

change list sort method

added "disclaimer"

a few "small changes" after the installation of the team instance

adapted to its final location

changed "cgi-team" to actual "team-bin" in "auto-detect-realm" features.
Same for member.

new stylesheet for search results
using <div> instead of stupid <dd> now that I took care of this annoying namazu "feature"

little tweak to adapt to new environment on wiggum

finished cleaning the HTML output (4.01) before trying XHTML output

a lot of debugging

added page management

There was a stupid bug which broke search in anything else than group (realm) search. Now fixed, you can search any index again.

*** empty log message ***