summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2007-10-21* Correct a pair of logic errors that triggered if svnpath was empty.joey
2007-10-21web commit by JonasSmedegaard: Improved routines to grab localized underlays.joey
2007-10-21document exactly what goes on during a cgi edit or remote commit with gitjoey
2007-10-21add news item for ikiwiki 2.10joey
2007-10-21releasing version 2.10joey
2007-10-21suggest python, since the rst formatter is written in itjoey
2007-10-21responsejoey
2007-10-21A few comments. I think git-stash could be used now that it's available.joey
Also removed some stuff about the git support not being robust or well-tested, since it certainly is now.
2007-10-21updatejoey
2007-10-21reorg all the git cruft a bit, highlight setup page and de-highlight morejoey
complex or less perfect setup methods
2007-10-21recommend using the git setup documented in the setup page and not thisjoey
2007-10-21ok, think I figured it outjoey
2007-10-21question about setting up gitjoey
2007-10-21 * Support git authors of the form "joey <joey>", which is common whenjoey
importing from a svn repo.
2007-10-21clarificationjoey
2007-10-19web commit from 212.179.254.14: poll vote (It's fast enough)joey
2007-10-19web commit by ttestjoey
2007-10-19web commit by http://ygingras.livejournal.com/: poll vote (It's too slow!)joey
2007-10-17let's keep companies under projects, for now (Braawi is also a companyjoey
afair, so is linuxworld)
2007-10-17er, this was implemented over 1 year ago..joey
2007-10-17web commit by http://natalian.org/: poll vote (Accept only OpenID for logins)joey
2007-10-17web commit by http://natalian.org/: UNDOjoey
2007-10-17web commit by http://natalian.org/: poor grammar fixed somewhatjoey
2007-10-17web commit by http://natalian.org/: Demonstrating demandjoey
2007-10-17web commit by AndrewGreenberg: Added the TOVA Company - hope it's ok that I ↵joey
added a new section heading!
2007-10-17web commit by JonasSmedegaard: Clarify the relevant localized repositories ↵joey
at source.jones.dk
2007-10-17web commit by JonasSmedegaard: Added note about danish l10n of templates and ↵joey
basewiki.
2007-10-16* Danish translation from Jonas Smedegaard. Closes: #446952joey
2007-10-16complete analysisjoey
2007-10-16web commit from 80.196.110.134: poll vote (Accept both)joey
2007-10-16web commit by http://roktas.myopenid.com/: add pcc projectjoey
2007-10-15web commit by JonasSmedegaard: Typo: Wrong punctuation.joey
2007-10-15* Rewritten rst plugin by madduck is a python program that communicates withjoey
ikiwiki via XML RPC. This should be much faster than the old plugin that had to fork python for every rst page render. Note that if you use the rst plugin, you now need to have the RPC::XML perl module installed.
2007-10-14web commit by http://svend.myopenid.com/joey
2007-10-13* Add a "createlink" class attribute to the span for wikilinks pointingjoey
to not-yet-existing pages. I don't have a useful style defined for that though.
2007-10-13improve commentjoey
2007-10-13untaint plugin filename, will be tainted if specified at the command linejoey
2007-10-13failed to reproducejoey
2007-10-13web commit by http://madduck.myopenid.com/joey
2007-10-13web commit by http://solair.livejournal.com/: Request CSS class attribute ↵joey
for span tags on broken wikilinks.
2007-10-10* Fix strange stderr-hiding code in the git module, allow error messagesjoey
to be passed on to stderr. Also fixes a potential bug, since git error meesages were treated as if they came from git stdout.
2007-10-10Vcs- fields now officialjoey
2007-10-10web commit by http://subvert.org.uk/~bma/: reasoning.joey
2007-10-10responsejoey
2007-10-10web commit by http://subvert.org.uk/~bma/: link to patch.joey
2007-10-10web commit by http://subvert.org.uk/~bma/: Add a patch to the Meta plugin to ↵joey
allow inclusion of CVS/SVN-style $Id$ or similar keywords.
2007-10-10clear hashes in loadindex, so that it can safely be called more than oncejoey
even if pages have been deleted in between
2007-10-10* In the cgi edit path, reload the index file before rendering. A bugjoey
showed up where a web edit that added a page caused a near-concurrent web edit to fail in will_render. While it would be hard to reproduce this, my analysis is that the failing cgi started first, loaded the index file (prior to locking) then the other cgi created the new page and rendered it, and then the failing cgi choked on the new file when _it_ tried to render it. Ensuring that the index file is loaded after taking the lock will avoid this bug.
2007-10-10responsejoey
2007-10-10move to users directoryjoey