summaryrefslogtreecommitdiff
path: root/tests/__init__.py
Commit message (Collapse)AuthorAgeFilesLines
* WIP https://github.com/eventlet/eventlet/pull/346is-timeoutSergey Shepelev2017-01-041-0/+7
|
* test_import_patched_defaults bended to play with pyopenssl>=16.1.0Sergey Shepelev2016-12-311-14/+32
| | | | | | | Basically this patch replaces urllib with custom module that is guaranteed not imported before patching. https://github.com/eventlet/eventlet/issues/362 More general issue here https://github.com/eventlet/eventlet/issues/368
* tests: str/bytes typo on run_python timeoutSergey Shepelev2016-11-051-3/+6
|
* tests: configurable timeout for run_python/isolatedSergey Shepelev2016-08-101-5/+7
|
* patcher: patch existing threading locks; Thanks to Alexis LeeAlexis Lee2016-04-181-2/+10
| | | | | | | | | | | | | | | | | | In projects which dynamically determine whether to activate eventlet, it can be hard not to import a low level module like logging before eventlet. When logging is imported it initialises a threading.RLock which it uses to protect the logging configuration. If two greenthreads attempt to claim this lock, the second one will block the /native/ thread not just itself. As green systems usually only have one native thread, this will freeze the whole system. Search the GC for unsafe RLocks and replace their internal Lock with a safe one while monkey-patching. The tests pass, but were they to fail, the test process would never return. To deal with this, I've added a test dependency on subprocess32 which is a backport of the stdlib subprocess module from Python3. This offers a timeout option on Popen#communicate, which I've arbitrarily set at 30 seconds.
* little clean upSergey Shepelev2016-02-121-9/+14
|
* travis: Allow more CPU usageJakub Stasiak2016-01-071-0/+7
|
* greenio: socket.recv() could return str; Thanks to jerzyksocket_recv_strSergey Shepelev2015-09-061-4/+9
| | | | | It must always return bytes. https://github.com/eventlet/eventlet/issues/245
* tests/isolated for tests sandboxed in separate python processSergey Shepelev2015-02-251-30/+28
| | | | | | | | | | | | API: - write main test body into tests/isolated/filename.py - it must write 'pass' to stdout or 'skip:[optional reason]' - write a test with single line: tests.run_isolated('filename.py') FIXME: autorun all files in tests/isolated This deprecates tests.run_python and ProcessBase. TODO: rewrite old multiline string test bodies to isolated files TODO: add timeout to p.communicate() in run_python()
* greendns: IPv6 support, improved handling of /etc/hostsbb-40-greendns-ipv6Floris Bruynooghe2015-02-201-0/+16
| | | | | | https://github.com/eventlet/eventlet/issues/8 https://bitbucket.org/eventlet/eventlet/issue/105/name-resolution-needs-to-support-ipv6 https://bitbucket.org/eventlet/eventlet/pull-request/40/improve-asynchronous-ipv6-address
* Tests: Provide listener details on nonempty hubJakub Stasiak2015-02-121-4/+14
|
* Fix typos detected by toolkit misspellingsChangBo Guo(gcb)2015-02-071-1/+1
| | | | | | | * pip install misspellings * git ls-files | grep -v locale | misspellings -f - Closes GH #194
* hubs: drop Twisted supportSergey Shepelev2014-10-291-11/+0
|
* Reformat with autopep8Jakub Stasiak2014-10-111-1/+2
|
* tests: deprecated TestCase.assert_() -> assert keywordSergey Shepelev2014-07-161-8/+4
| | | | https://github.com/eventlet/eventlet/issues/101
* python3 compatibilitySergey Shepelev2014-04-241-5/+0
| | | | | | | | | | | | - __next__ for iterator interface - six.next() to get next item - list(dict.keys()) - popen2.popen4 -> subprocess - s2b -> b"..." literals - deprecated assertEquals -> assertEqual - hub_test test_fork using run_python - 1L -> 1 long literal - many PEP-8 fixes
* PEP-8 fixes, tox runs pep8 checkSergey Shepelev2014-04-231-13/+20
| | | | | For now, pep8 check is only run for some files known to be clean, we should clean the rest and enable pep8 check for all files then.
* pypy compatibilityJakub Stasiak2014-04-151-0/+3
|
* python3 compat: print() function syntaxSergey Shepelev2013-12-031-2/+4
|
* tests: kill all tpool workers in LimitedTestCaseJakub Stasiak2013-08-071-10/+3
| | | | Fixes GH-41 https://github.com/eventlet/eventlet/issues/41
* tests: extract run_python() utility functionSergey Shepelev2013-07-091-0/+21
| | | | Proposal is to use it for new subprocess tests where module is commited to repo
* tests: skip check_idle_cpu_usage on Windows; Thanks to Paul OppenheimSergey Shepelev2013-07-091-1/+9
| | | | TODO: use psutil package which ought to work on Windows
* tests: LimitedTestCase.set_alarm() detects busy loops like while True: passSergey Shepelev2013-01-111-9/+31
|
* tests: check_idle_cpu_usage measures getrusage() around sleep(); throws ↵Sergey Shepelev2013-01-101-0/+15
| | | | AssertionError
* tests: skip_if_no_ssl decorator always returned NoneSergey Shepelev2013-01-101-9/+11
|
* Skip SSL tests if SSL is not availableFloris Bruynooghe2012-09-051-0/+11
| | | | | This is particularly useful to not require the external OpenSSL module in order to run the tests on python2.5.
* Refactoring zmq support tests in light of lack of zeromq hub.Ryan Williams2011-04-091-17/+0
|
* Initial implementation of hubless zeromq support using ZMQ_FD and ZMQ_EVENTS.Zed Shaw2011-04-081-2/+2
|
* Skipping repeated selects test on zmq hub because it doesn't appear to ↵Ryan Williams2011-01-231-0/+17
| | | | support detecting bad filenos at select() time.
* Finally, eventlet.green.MySQLdb so we don't have to hassle with db_pool if ↵Ryan Williams2010-10-071-16/+23
| | | | we're not going to be using the pooling functionality.
* Support for SSL websockets, which also happens to improve our SSL support in ↵Ryan Williams2010-09-161-0/+3
| | | | WSGI generally. Fixes #62.
* Use built-in json for tests if available.Ryan Williams2010-07-231-5/+8
|
* Add support for the itimer module with blocking detectionR. Tyler Ballance2010-06-241-0/+10
| | | | | | This will allow for subsecond blocking detection thanks to the itimer module recently released by Slide, Inc: http://github.com/slideinc/itimer
* Cleaned up error output a bit better so diagnosis of failing psycopg tests ↵Ryan Williams2010-05-241-0/+1
| | | | doesn't take quite so long to track down t configuration problems again.
* Converted fully over to what the internet assures me is the American ↵Ryan Williams2010-05-051-1/+15
| | | | spelling of 'canceled' -- this only affects internal variables. Also cleaned up the meticulous timer accounting of the hub tests so they stop failing after other tests are run. Also added new parameter to the abort method because it seemed to be the dominant mode of operation.
* Database tests should be skipped if simplejson isn't installed.Ryan Williams2010-04-181-7/+13
|
* Integrated the dotfile auth lookup stuff with the new test.Ryan Williams2010-04-171-0/+35
|
* Fixed overlong duration of iterator test.Ryan Williams2010-03-251-0/+7
|
* py3k - more elegant s2b implementationamajorek2010-03-041-9/+3
|
* py3k socket send and receives byte type data. They are different from ↵amajorek2010-03-021-0/+10
| | | | | | | | | strings in 3.x. tests.s2b function added to convert string to bytes type on 3.x. No-op on 2.x. / replaced with // used when integer division was intedned. Testing for closed file made portable by changing if data=='': to if not data: ('' != b'') renamed bytes variable, not to clash with 3.x builtin type.
* Check that hubs are empty in tearDown. You have to run nosetests with -s to ↵Chris AtLee2010-02-231-1/+12
| | | | see these messages
* Fix some deprecation warningsChris AtLee2010-02-231-0/+10
|
* MergeRyan Williams2010-02-161-3/+3
|\
| * Imported Timeout class from gevent, deprecated exc_after, replaced ↵Ryan Williams2010-02-151-3/+3
| | | | | | | | TimeoutError with Timeout (should be seamless).
* | Disabling pyevent-problematic tests.Ryan Williams2010-02-161-3/+4
|/
* Fixed bug in tpool.killall, also another slight bug in the pollhub, and ↵Ryan Williams2010-01-181-0/+9
| | | | added a new test verification method which will work on every hub but pyevent.
* Refactored imap to make it even simpler and fix the limitation, added test ↵Ryan Williams2009-12-311-14/+0
| | | | for multiple iterators, removed SilencedTestCase because there was no need for it anymore.
* Moved get_hub, use_hub, get_default_hub to eventlet.hubs. This is a step in ↵Ryan Williams2009-12-171-7/+8
| | | | the direction of better factoring and fewer circular-ish dependencies.
* Renamed libevent hub to pyevent.Ryan Williams2009-12-161-5/+5
|
* Refactored __init__ a little, skipped a processes-based tests on Windows.Ryan Williams2009-12-071-20/+43
|