can evaluate them, check them in the wrapper right off the bat.
This doesn't prevent the deadlock in web commits that need to cvs
add directories, but I'm committing so Joey can take a look if he
wants.
This is necessary so that things that fork to the background,
like pinger, and inline ping, don't block other cgis from running.
Note that websetup also calls unlockwiki, before refreshing / rebuilding
the wiki. It makes perfect sense for that not to block other cgis.
Fixed by making the cgi wrapper wait on a cgilock.
If you had to set apache's MaxClients low to avoid ikiwiki thrashing
your server, you can now turn it up to a high value.
The downside to this is that a cgi call that doesn't need to call lockwiki
will be serialised by this so only one can run at a time. (For example,
do=search.) There are few such calls, and all of them call loadindex,
so each still eats gobs of memory, so serialising them still seems ok.
This speeds up web commits by 1/4th of a second or so, since perl does
not have to start up for the post commit hook.
perl's locking is completly FuBar, since it's impossible to tell what perl
flock() really does, and thus difficult to write code in other languages
that interoperates with perl's locking. (Let alone interoperating with
existing fcntl locking from perl...)
In this particular case, I think I was able to find a way to avoid the
insanity, mostly. The C code does a true flock(2), and if perl is using an
incompatable lock method that does not use the same locking primative at
the kernel level, then the C code's test will fail, and it will go ahead
and run the perl code. Then the perl code's test will test the right thing.
On Debian, at least lately, perl's flock() does a true flock(2), so the
optimisation does work.
to be created owned by some group other than the default. Useful
then there's a shared repository with access controlled by a group,
to let ikiwiki run setgid to that group.
* ikiwiki-mass-rebuild: Run build with the user in all their groups.
* Support building on systems that lack asprintf.
* mercurial getctime is currently broken, apparently by some change in
mercurial version 0.9.4. Turn the failing test case into a TODO test case.
using HTTP Authentication instead of ikiwiki's built in authentication.
Useful for eg, large sites with their own previously existing user auth
setup. Closes: #384534
<span>, so pages can use <h1> internally instead of needing to use <h2>.
* Updated all of ikiwiki's own wiki pages for that.
* Add pagetemplate hook, which can be used by plugins that want to mess
around with adding new stuff to the page template.
* Remove headercontent; the search plugin now adds the search box to the
header by registering a pagetemplate hook, and other plugins should do
similarly.
for other types of search engine plugins if wanted, and also opening up a
lot of new possibilities for other kinds of plugins later
some notable changes along the way:
- lots of new hook types: cgi, render, delete
- wrapper files fixed to support config strings with newlines in them
- HEADERCONTENT in page template useful for plugins. Probably needs to be
expanded to more such for other places plugins might want to add content.
- remove unnecessary wrappers field from config info stored in wrappers
to pave the way for adding other RCS support. This also changes the
setup files, where before they had svn => 1 or svn => 0, now they have
rcs => "svn" or rcs => "".
* Add a debian/NEWS file with upgrade notes.
* Load whatever rcs module is specified, so new ones can be just dropped in
as plugins and work.
* Add some basic docs about writing RCS modules.
This {gross,amazing} hack makes all wrapped uses of ikiwiki forgo any
option parsing at all. Options come in preparses via an env var from the
wrapper. As a bonus, Wrapper.pm no longer needs to be updated when command
line options are added to the program. Load time is sped up by around 10%.
ikiwikiwrap --params is no longer supported by this change. You will need
to rebuild your wrappers to take advantage of it.
kept in mind during this are a) to reduce load time for common cases like
cgi and post-commit and b) make the code easier to navigate.
This also modularises RCS support to the extent that it should be possible
to drop in a module for some RCS other than svn, add a switch for it, and
it pretty much just work.
High chance I missed an edge case that breaks something, this is only
barely tested at this point.
Amitai Schlair
Joey Hess
Simon McVittie
Joey Hess
joey