Commit Graph

1313 Commits

Author SHA1 Message Date
Ronny Pfannschmidt 6c170201d6 Merge branch 'master' into features 2016-01-02 23:56:01 +01:00
Ronny Pfannschmidt bf4de4bd68 Merge pull request #1294 from nicoddemus/doctest-bytes-literals
Doctest bytes literals
2016-01-01 18:18:08 +01:00
Bruno Oliveira 1bdf71730a Complement #1255 by adding tests and docs
Fix #1242
2015-12-30 18:24:59 -02:00
jab 0ea8dc0d40 make --doctest-glob multi-allowed 2015-12-30 17:32:14 -02:00
Bruno Oliveira 638b3f5e39 Make monkeypatch calls O(1)
Fix #1292
2015-12-29 22:04:35 -02:00
Bruno Oliveira a0edbb75a4 Implement ALLOW_BYTES doctest option
Fix #1287
2015-12-29 20:55:19 -02:00
Ulrich Petri 8d49abb0d1 Make monkeypatch differentiate ImportError sources
Previously `monkeypatch` assumed that any `ImportError` was caused by
a mistake in the specified import path. However this assumption is false 
in case the import target itself causes an `ImportError`.

Fixes: #900
2015-12-27 22:44:23 +01:00
Bruno Oliveira 5afb61ad26 Fix trailing white-space 2015-12-25 17:51:55 -02:00
Bruno Oliveira 57bc14caa0 Apply readline workaround during initial conftest loading 2015-12-25 17:46:19 -02:00
Erik M. Bray 6e170a4a1c * Moved workaround to its own function, mostly for the sake of adding
a more descriptive docstring (the workaround itself is just to import
  readline earlier).  I removed the conditional on targetfd from the
  workaround since it doesn't really matter.

* Added # noqa marker.

* Added changelog entry, and self to authors.
2015-12-24 16:43:34 -05:00
Erik M. Bray 924a9667e1 Make sure readline has been imported before duping any stdio handles--otherwise pyreadline fails to connect to the correct handle for the console's stdout/in. 2015-12-24 14:56:57 -05:00
Bruno Oliveira 943099ddd1 Merge newinterpret into reinterpret.py 2015-12-16 18:31:43 -02:00
Bruno Oliveira 81588d7f63 Remove obsolete "oldinterpret" module 2015-12-16 16:16:22 -02:00
Bruno Oliveira af893aab26 Remove code related to support python <= 2.5
Fix #1226
2015-12-16 16:16:22 -02:00
Ronny Pfannschmidt 8bf7e7cc4b fixes #1259 - take finalized nodes out of the mapping
this allows double node id usage for file based items
2015-12-16 19:09:44 +01:00
Bruno Oliveira 2b2bec6b97 Remove experimental status from pytest_exception_interact 2015-12-16 14:29:39 -02:00
Bruno Oliveira 6ac31088c5 Merge branch 'master' into merge-master-into-features 2015-12-11 23:13:50 -02:00
Bruno Oliveira e87facfb22 Bump master to 2.8.6.dev1 2015-12-11 23:07:30 -02:00
Bruno Oliveira a124163425 Prepare for 2.8.5: bump version, CHANGELOG, announce 2015-12-11 19:20:07 -02:00
Bruno Oliveira 86b6ce5042 Merge remote-tracking branch 'upstream/master' into merge-master-into-features 2015-12-10 19:41:14 -02:00
Bruno Oliveira 8b61a332ba Merge remote-tracking branch 'bukzor/features-merge-master' into features 2015-12-10 19:03:55 -02:00
aselus-hub ec02f694ef Update python.py
updated dictionary itteration to create a list for generation, so that tests can be added in the generator functions under python3. This works fine as-is in python2 because python 2 already creates a list, whereas python3 returns an itterator. Forcing a list format for the return fixes python3 to work the same way as python2
2015-12-09 11:32:19 -08:00
Bruno Oliveira 1c46462991 Fix deprecated_call regression introduced in 2.8.4
Fix #1238
2015-12-08 22:40:05 -02:00
Buck Golemon eabf2f9091 Merge branch 'master' into features
Conflicts:
	AUTHORS
	_pytest/__init__.py
	_pytest/hookspec.py
	_pytest/recwarn.py
	testing/test_recwarn.py
2015-12-07 14:28:59 -08:00
David Vierra fbac936596 Add -rp and -rP options to report passing tests.
-rP is an alternative to `-s` for viewing the output of passing tests.
This causes the captured stdout/stderr of passing tests to be output in
the same way as that of failing tests.

-rp adds a simple one-line-per-test summary for passing tests.

Neither option is included by -ra.

Additional changes to `pytest_capturelog` and `pytest_catchlog` are
needed for this option to also output captured logs: They must be
changed to use `rep.sections.add` instead of `rep.longrepr.addsection`,
and to add these additional sections even if the test passes, since
passing tests don't seem to have a `longrepr` at report time.
2015-12-07 11:32:56 -10:00
Ronny Pfannschmidt 7b7737bf96 handle duplicate test ids via collection and xdist each reporting 2015-12-07 21:54:25 +01:00
Ronny Pfannschmidt 9ea7826427 Junitxml: correct node reporter attribute names 2015-12-07 21:54:25 +01:00
Ronny Pfannschmidt 09cc45b0c5 junitxml: correct docstring of make_properties_node 2015-12-07 21:54:25 +01:00
Ronny Pfannschmidt 0aa54101c9 junitxml: follow Bruno's attribute/method naming hint from the review 2015-12-07 21:54:25 +01:00
Ronny Pfannschmidt 518c88f149 finalize nodereporters by throwing away the intermediate xml nodes, fixes issue #1074 2015-12-07 21:54:25 +01:00
Ronny Pfannschmidt 5f5a7995b9 reintroduce junitxml report order and debug cleanups 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt 0528e5b45f junitxml: intermediate, move testcase generation to NodeReporter 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt 9b04958303 junitxml: keep track of custom property insert order
+ review: should we allow the same key multile times
2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt faed54d6c7 junitxml: use node.warn to ensure fslocations 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt 1f609f96e6 junitxml: introduce nodereporter and track durations/properties there 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt 0664ae137c junitxml: remove debug print 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt d0107c898e junitxml restrucutre stat generation - use node tags 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt 2a31df072b junitxml: reverse the if/else logic for failure appending 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt 02f5defd89 yapf junitxml 2015-12-07 21:54:24 +01:00
Ronny Pfannschmidt c727860241 bump to 2.8.5.dev 2015-12-06 20:31:14 +01:00
Ronny Pfannschmidt a3353c49fd prepare release 2.8.4 - changelog updates + version bump 2015-12-06 16:13:55 +01:00
Bruno Oliveira 14bc3c4009 Fix pastebin when captured output contains non-ascii characters
Fix #1219
2015-12-03 20:07:18 -02:00
Bruno Oliveira a54e4e64cd Merge remote-tracking branch 'upstream/master' into pastebin-py3 2015-12-01 23:51:14 -02:00
Bruno Oliveira 6a2ebddc7c Decode urlopen response in pastebin
Fix #1198
2015-12-01 23:33:37 -02:00
Jeff Widman 6be6798cdf Fix spelling: explicitely --> explicitly 2015-12-01 01:41:47 -08:00
Florian Bruhin 0d2668017d Fix spelling mistake in #1207. 2015-11-30 17:33:34 +01:00
Florian Bruhin aba55a0fb2 Fix terminal output if no tests were run.
Before:
====  in 0.00 seconds ====

After:
==== no tests run in 0.00 seconds ====
2015-11-30 17:24:40 +01:00
Ronny Pfannschmidt b5d65e5139 Merge pull request #1206 from The-Compiler/collect-getattr
Don't collect classes with truthy __getattr__.
2015-11-30 17:23:47 +01:00
Florian Bruhin ba9146c131 Don't collect classes with truthy __getattr__.
When we have a metaclass which returns something truthy (like a method) in its
__getattr__, we collected the class because pytest thought its __test__
attribute was set to True.

We can work around this to some degree by assuming __test__ will always be set
to an explicit True if that's what the user has intended, and if it's something
other than that, this is probably a mistake.

Fixes #1204.
2015-11-30 16:41:13 +01:00
Florian Bruhin c790f7475e Fix getting line number with nasty __getattr__.
When an object has a custom __getattr__ which always returns a non-int, we
tried to get compat_co_firstlineno from it and checked it was a integer, which
caused an exception if such a class is mistakenly collected.

If we still mistakenly collect such a class (which is likely to be something
other than a test), we now skip it with a warning (because it probably has an
__init__) instead of producing an error.

See #1204.
2015-11-30 16:13:15 +01:00