| Commit message (Collapse) | Author | Age |
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
A page could have possibly hundreds of backlinks, when the cache is
outdated they should not be rererendered at once
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Quoted-Printable specifies a maximum line length and some mail tools
(Apple mail and Thunderbird) take this quite serious and will fail to
decode subjects encoded with quoted-printable when the subject exceeds
the length limit.
The correct fix would be to wrap the header into multiple lines. But
this seems not to be possible with mails() $subject variable.
This patch switches to Base64 encoding for long subjects.
A general decision if switching completely to Base64 is the best way
to go is still open. (see bugreport)
|
| |
|
| |
|
|
|
|
| |
All linebreaks before p_close have been removed.
|
|
|
|
|
| |
This fixes a PCRE backtrack error that occurred on large pages like
:users on dokuwiki.org.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
This research suggests that, the change does not help, but in fact
breaks previoulsy working setups:
https://bugs.dokuwiki.org/index.php?do=details&task_id=1755#comment3446
I'm still at loss on how to fix this bug.
This reverts commit 2ae68f97446ff6bae5fbbe463eb00312598be840.
|
| |
|
| |
|
|\ |
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | | |
-Simplify the algorithm. May improve performance.
-Treat footnote as pure block and section as pure stack.
-Remove post-p-open and pre-p-close linefeeds. Affects the effect of xbr plugin.
|
| | |
| | |
| | |
| | | |
starting single linebreaks.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Anika's merge did not pul in the individual patches as one would expect.
Then I messed up when trying to fix this by merging with danny's repo
again but used the wrong branch. So we're still missing two patches. To
have them apply cleanly I have to revert Anika's merge here. Another
merge for the missing two patches will follow.
This reverts commit b17e20ac9cca30b612968d02f06fa9c5df5c01f0.
|
|\| | |
|
| |\| |
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | | |
prevents plugins that ought to handle headers from working and require a lot of ugly workarounds (such as header2).
If there's a concern, "protected" may be another choice.
|
| | | |
|
|/ / |
|
| |
| |
| |
| |
| | |
A search equivalent to a new ID so it should show up in the bread
crumbs.
|
| |
| |
| |
| |
| | |
This makes it possible to find words that include soft-hyphens. However,
search higlighting will not work and I have no idea how to make it work.
|
| |
| |
| |
| |
| |
| |
| | |
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.
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Previously the rmdir could be executed when the lock directory had been
deleted by another indexer already. This could lead to a race condition
when another indexer call creates the lock again between the if and the
rmdir. This issue still exists for stale lock directories but they
normally shouldn't exist.
This also prevents the loop from becoming an endless loop when the lock
directory can't be created.
This change also fixes a syntax error in the indexer and prevents an
endless loop when the lock directory exists but can't be deleted.
|
| | |
|
| | |
|
|\ \ |
|
| | | |
|
| | | |
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This changes the HTTP stream to blocking while writing to the stream
using select() to handle timeouts. Addtionally, wwriting is done in 4k
block now (as it is done with reading).
This is supposed to fix a problem with writing to a SSL stream that is
not quite ready.
Reading from the stream continues to be non-blocking as before.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Changes:
* XML-RPC now correctly allows leading and trailing _
* Error messages from XML-RPC are correct
* MEDIA_UPLOAD_FINISH has a sixth param specifying the move function
* Not having upload rights when using media_upload throws a msg
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
manager
This patch makes the plugin manager not descend into symbolically linked
directories as this behaviour is not expected from a recursive delete.
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Functionality changes:
* deleteAttachment now triggers MEDIA_DELETE_FILE (closes FS#1568)
* deletion success msg in mediamanager is correct, even when the ns dir
was deleted
* media_delete changed quite a bit
|