| Commit message (Collapse) | Author | Age |
|---|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
this was lost because the subscription branch, branched before this
change in master
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
This is untested and probably broken currently
|
| | |
|
| |
|
|
|
|
|
| |
We don't want to create a bunch of mails whenever a namespace is
subscribed. Only changes *after* the subscription should be considered.
This patch adds the timestamp to "every" style subscriptions as well,
though this data is ignored.
|
| | |
|
| |
|
|
|
| |
This readds the last part of the subscription system: the normal "every"
subscriptions.
|
| |
|
|
|
|
|
| |
now the actual sending of bulk messages (digest, list) is reimplemented
and partially tested.
Still not complete
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
This introduces a class for nicer wrapping and easier testing. Some
functions were changed to provide nicer APIs (no throwing around of
unescaped regexps) and to simplify things (hopefully).
The refactoring isn't completed yet, so this will break the subscription
system.
The goal is to move as much subscription related stuff to this class as
possible. Currently there is some code in lib/exe/indexer.php and maybe
elsewhere (common.php?). Additionally everything should be covered by
tests. A few tests are included here already.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
| |
Users with names like foo.bar still didn't receive the mail when user
foo edited the page. Now the two possibilities (space and end of line)
are tested instead of just a word boundary.
|
| |
|
|
|
|
| |
This fixes the subscription user regex to really only not to match the
author of the change but to still match users who have that username as
substring in their username.
|
| |
|
|
|
|
|
| |
There is no reason why the subscription should wait for other calls
because the lock is only for one page so once the other call has
finished the work has already been done. This simplifies the lock
mechanism so there is no more loop.
|
| |
|
|
|
|
|
|
|
|
|
| |
Filenames can't contain ":" on windows and the lock directory contained
the unescaped page id. The lock function tries in an endless loop to
create the lock directory when it fails and the directory doesn't exist.
Just escaping the directory name won't work as then the filename length
limit will be quickly hit when using deep namespace structures with utf8
names. Thus using the md5sum seems to be the best solution. Perhaps the
lock function could also be changed to create a file with that name that
contains the id so the id can be retrieved for debugging purposes.
|
| | |
|
| | |
|
| |
|
|
|
| |
- use directories since working with them is atomic
- delete stale locks after 5 minutes
|
| | |
|
| | |
|
| |
|
|
|
|
| |
If you subscribed to namespaces other than the root namespace using the
new develonly subscription feature you'll need to renew your
subscriptions.
|
| | |
|
| | |
|
| |\
| |
| |
| |
| | |
Conflicts:
inc/fulltext.php
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Classes are loaded throug PHP5's class autoloader, all other
includes are just loaded by default. This skips a lot of
require_once calls.
Parser and Plugin stuff isn't handled by the class loader yet.
|
| | | |
|
| |/ |
|
| | |
|
| | |
|
| | |
|
| |
|
