| Age | Commit message (Collapse) | Author |
|---|
|
|
|
This may need to be reverted if CTRL-C handling breaks, but everything
works fine so far in testing.
Resolves issue #95.
|
|
|
|
computing logging data that may never be used. This is a HUGE performance improvement; in some of my test runs, unnecessary string creation was accounting for > 60% of all CPU time.
Note that using % in a string will _always_ perform the sting substitutions, because the strings are constructed before the function is called. So log.debug('%s' % expensiveoperation()) will take about the same CPU time whether or not the logging level is DEBUG or INFO. if you use , no substitutions are performed unless the string is actually logged
|
|
|
|
This should prevent some reference cycles that will cause garbage
collection issues.
|
|
Don't keep a reference to stanzas in Callback objects.
|
|
|
|
Waiting until the actual run step means that the handler is not
marked for deletion when checked in the __spawn_event() thread,
causing the callback to stay in the handler list.
|
|
|
|
Each module should now log into its own logger.
|
|
|
|
Cleaned up the __init__.py files.
|
|
|
|
|
|
Updated XMLStream to return True or False from removeHandler to indicate if the handler
existed and was removed.
Waiter handlers now unregister themselves after timing out.
|
|
xmlstream.matcher __init__.py files to make it simpler to import common classes.
|
|
references to nonexistant license.txt to LICENSE.
|
|
|
|
|
|
* added stanza unhandled (unhandled iqs now reply with feature-not-implemented)
* added stanza exceptions (stanzas may now reply with exceptions when their handler raises an exception)
|
|
|
|
* configuration-less nodes
|
|
|
|
* sleekxmpp no longer spawns threads for callback handlers -- there are now two threads: one for handlers and one for reading. callback handlers can get results from the read queue directly with the "wait" handler which is used in .send() for the reply catching argument.
|
|
|
