From jurko.gospodnetic at pke.hr Sun Mar 2 23:36:56 2014 From: jurko.gospodnetic at pke.hr (=?windows-1250?Q?Jurko_Gospodneti=E6?=) Date: Sun, 02 Mar 2014 23:36:56 +0100 Subject: [pytest-dev] Displaying additional test failure information Message-ID: Hi all. I was wondering if anyone can tell me the cleanest way to make assertions in a specific part of my test code (e.g. a specific test class or module) report some extra information on test failure? I can add a regular assertion description string, and that displays the extra information but it also avoids regular pytest test failure result reporting which I would like to keep. I can also print out the extra information directly to stdout/stderr - but that can be hard to read, especially if the test prints out other data as well. Usage scenario: --------------- Imagine a class for comparing two XML structures. That class maintains some context information that would be useful to print out when an XML mismatch is detected. For example, if it detects two nodes with different namespaces, you want it to: - report exactly which nodes these are and not just 'there exist two nodes with mismatched namespaces' - make this information available without the developer having to drop into the debugger If possible, it would also be a great plus if this extra content could be integrated into the regular 'red' assertion failure information. Best regards, Jurko Gospodneti? From holger at merlinux.eu Mon Mar 3 12:07:45 2014 From: holger at merlinux.eu (holger krekel) Date: Mon, 3 Mar 2014 11:07:45 +0000 Subject: [pytest-dev] Parametrized tests funcargs missing on 'pytest_collection_modifyitems' hook (pytest 2.5.2) In-Reply-To: References: Message-ID: <20140303110745.GE1598@merlinux.eu> Hi Anton, On Wed, Feb 26, 2014 at 13:19 +0200, Anton P wrote: > Hi, > > For testing purposes I use hook 'pytest_collection_modifyitems' to get > values of all arguments passed in parametrized test function, e.g.: > > *def pytest_collection_modifyitems(items):* > * for item in items:* > * print item.funcargs* > > with code execution example: > > *{'doc': ['doc_1'], 'name': 'test_1'}* > *{'doc': ['doc_2'], 'name': 'test_2'}* > *{'doc': ['doc_3'], 'name': 'test_3'}* > > Example of code usage is attached. It works fine with pytest version 2.4.2. > However while moving to pytest version 2.5.2 i faced the following issue: > *item.funcargs *dictionary is empty > {} > {} > {} > > Could you please clarify how can I get *funcargs *in the > 'pytest_collection_modifyitems' hook in pytest 2.5.2? "item.funcargs" is now filled during test execution time, not test execution time. I think it was never documented either way so i don't regard it as an API promise. Also, funcargs/fixtures are setup during test execution so we can not generally make values available during collection. With 2.5.2 you might still fish for the values, i think it's on 'item.callspec.funcargs' or so but again, no promise. holger > Thanks in advance! > Anton > import pytest > > @pytest.mark.trylast > def pytest_collection_modifyitems(items): > for item in items: > print item.funcargs > import pytest > > > class TestClass(): > > argvals = [("test_1", ["doc_1", ]), ("test_2", ["doc_2", ]), ("test_3", ["doc_3", ])] > > @pytest.mark.parametrize("name, doc", argvals, ids=[x[0] for x in argvals]) > def test_params(self, name, doc): > pass > > _______________________________________________ > Pytest-dev mailing list > Pytest-dev at python.org > https://mail.python.org/mailman/listinfo/pytest-dev From ich at ronnypfannschmidt.de Mon Mar 3 08:04:32 2014 From: ich at ronnypfannschmidt.de (Ronny Pfannschmidt) Date: Mon, 03 Mar 2014 08:04:32 +0100 Subject: [pytest-dev] Displaying additional test failure information In-Reply-To: References: Message-ID: <37d6ae8c-021b-4876-907f-ae629f298cd1@ronnypfannschmidt.de> Hi, 2 ways come to mind a) using print b) using the == operator for the assertion and a assert_repr_compare hook see http://pytest.org/latest/assert.html#defining-your-own-assertion-comparison -- Ronny On Sunday, March 2, 2014 11:36:56 PM CEST, Jurko Gospodneti? wrote: > Hi all. > > I was wondering if anyone can tell me the cleanest way to > make assertions in a specific part of my test code (e.g. a > specific test class or module) report some extra information on > test failure? > > I can add a regular assertion description string, and that > displays the extra information but it also avoids regular pytest > test failure result reporting which I would like to keep. > > I can also print out the extra information directly to > stdout/stderr - but that can be hard to read, especially if the > test prints out other data as well. > > Usage scenario: > --------------- > Imagine a class for comparing two XML structures. That class > maintains some context information that would be useful to print > out when an XML mismatch is detected. For example, if it detects > two nodes with different namespaces, you want it to: > - report exactly which nodes these are and not just 'there > exist two nodes with mismatched namespaces' > - make this information available without the developer > having to drop into the debugger > > If possible, it would also be a great plus if this extra > content could be integrated into the regular 'red' assertion > failure information. > > Best regards, > Jurko Gospodneti? > From jurko.gospodnetic at pke.hr Mon Mar 3 12:32:22 2014 From: jurko.gospodnetic at pke.hr (=?UTF-8?B?SnVya28gR29zcG9kbmV0acSH?=) Date: Mon, 03 Mar 2014 12:32:22 +0100 Subject: [pytest-dev] Parametrized tests funcargs missing on 'pytest_collection_modifyitems' hook (pytest 2.5.2) In-Reply-To: <20140303110745.GE1598@merlinux.eu> References: <20140303110745.GE1598@merlinux.eu> Message-ID: Hi. > "item.funcargs" is now filled during test execution time, not test execution > time. Dang... this made me go into an infinite loop... took me 4 readings through the same sentence to finally see the problem. :-D Best regards, Jurko Gospodneti? From holger at merlinux.eu Mon Mar 3 12:41:17 2014 From: holger at merlinux.eu (holger krekel) Date: Mon, 3 Mar 2014 11:41:17 +0000 Subject: [pytest-dev] Parametrized tests funcargs missing on 'pytest_collection_modifyitems' hook (pytest 2.5.2) In-Reply-To: References: <20140303110745.GE1598@merlinux.eu> Message-ID: <20140303114117.GK1598@merlinux.eu> On Mon, Mar 03, 2014 at 12:32 +0100, Jurko Gospodneti? wrote: > Hi. > > >"item.funcargs" is now filled during test execution time, not test execution > >time. > > Dang... this made me go into an infinite loop... took me 4 > readings through the same sentence to finally see the problem. :-D sorry -- for others: "filled during test execution, not test collection time". holger > Best regards, > Jurko Gospodneti? > > > _______________________________________________ > Pytest-dev mailing list > Pytest-dev at python.org > https://mail.python.org/mailman/listinfo/pytest-dev From issues-reply at bitbucket.org Thu Mar 13 16:58:42 2014 From: issues-reply at bitbucket.org (David Szotten) Date: Thu, 13 Mar 2014 15:58:42 -0000 Subject: [pytest-dev] Issue #9: --cache copy-friendlier output (hpk42/pytest-cache) Message-ID: <20140313155842.16429.27112@app10.ash-private.bitbucket.org> New issue 9: --cache copy-friendlier output https://bitbucket.org/hpk42/pytest-cache/issue/9/cache-copy-friendlier-output David Szotten: it looks like `--cache` pprints the cache set, which isn't very friendly to copy/paste, since the first test is preceded by "`set([`". could we iterate over the set (maybe even sorted) and print each on its own line, (even skipping the quotes and the commas maybe) From holger at merlinux.eu Wed Mar 26 10:46:20 2014 From: holger at merlinux.eu (holger krekel) Date: Wed, 26 Mar 2014 09:46:20 +0000 Subject: [pytest-dev] please resolve issues / bug day Message-ID: <20140326094620.GK25870@merlinux.eu> Hi pytest users and contributors, Since 2.5.2 we've had around 40 new issues and we need help to resolve these: https://bitbucket.org/hpk42/pytest/issues?status=new&status=open Please see if you can find the time to look and possibly resolve a few through a Pull request. Further below i'll dump the titles and issue links for your convenience. Furthermore we should do a new pytest bug day. Here is a doodle: http://doodle.com/k26yek52yxvsmpc8#table The one choice, the 11th April, is during Pycon in Canada but i assume i'll still have a lot of Jet Lag and also can make time during Canadian day time. best, holger ---- new bug https://bitbucket.org/hpk42/pytest/issue/487/ request.param mentioned in docs but not listed in the request API docs ---- new bug https://bitbucket.org/hpk42/pytest/issue/486/ `--help` should not fail with an exception on conftest.py errors / handle conftest imports more gracefully/meaningful ---- new bug https://bitbucket.org/hpk42/pytest/issue/483/ 'function' object has no attribute 'im_func' with twisted TestCases on python 3 ---- new bug https://bitbucket.org/hpk42/pytest/issue/481/ Running doctests with -s fails? ---- new bug https://bitbucket.org/hpk42/pytest/issue/480/ SkipTest is not handled properly in setUpModule ---- new bug https://bitbucket.org/hpk42/pytest/issue/478/ --pyargs does not understand namespace packages ---- new bug https://bitbucket.org/hpk42/pytest/issue/477/ excluding name patterns from consideration ---- new bug https://bitbucket.org/hpk42/pytest/issue/476/ Fixtures defined in separate files are not available to tests ---- new bug https://bitbucket.org/hpk42/pytest/issue/475/ Misleading pytest exception description ---- new bug https://bitbucket.org/hpk42/pytest/issue/474/ Running tests with xdist (-n4) causes failures with Django TestCases (DB access denied) ---- new bug https://bitbucket.org/hpk42/pytest/issue/473/ Cannot use @mark and multiple @patch class decorators ---- new bug https://bitbucket.org/hpk42/pytest/issue/471/ pytest keeps running deleted tests ---- new bug https://bitbucket.org/hpk42/pytest/issue/470/ crash when a test changes the cwd to something that gets removed later on without propper cleanup ---- new bug https://bitbucket.org/hpk42/pytest/issue/469/ junitxml parses report.nodeid incorrectly ---- new bug https://bitbucket.org/hpk42/pytest/issue/466/ IOError when writing --junitxml report ---- new bug https://bitbucket.org/hpk42/pytest/issue/465/ running pytest.main(...) from python script breaks cov plugin ---- new bug https://bitbucket.org/hpk42/pytest/issue/462/ What is the correct spelling when using unicode in assertions? ---- new bug https://bitbucket.org/hpk42/pytest/issue/461/ Resources warning when combining capfd and skip() ---- new bug https://bitbucket.org/hpk42/pytest/issue/460/ AttributeError: 'SubRequest' object has no attribute 'param' ---- new bug https://bitbucket.org/hpk42/pytest/issue/459/ doctest-modules + pyreadline means default color becomes black ---- new bug https://bitbucket.org/hpk42/pytest/issue/458/ KeyError: 'COV_CORE_SOURCE' when clearing environment ---- new bug https://bitbucket.org/hpk42/pytest/issue/457/ parser.addoption(type="float") backwards incompatibility ---- new bug https://bitbucket.org/hpk42/pytest/issue/455/ Using xdist with -f option runs tests twice on a detected test source code change ---- new bug https://bitbucket.org/hpk42/pytest/issue/454/ Console output text color missing when pytest run using xdist's -f option ---- new bug https://bitbucket.org/hpk42/pytest/issue/453/ assertion rewriting fails with object whose __repr__ contains '{\n' ---- new bug https://bitbucket.org/hpk42/pytest/issue/452/ Incorrect capfd description/documentation ---- new bug https://bitbucket.org/hpk42/pytest/issue/451/ Incorrectly formatted PDF documentation tables in chapter 7 ---- new bug https://bitbucket.org/hpk42/pytest/issue/450/ pytest --fixtures console output should not include text formatted as rst source ---- new bug https://bitbucket.org/hpk42/pytest/issue/448/ pytest-xdist: AttributeError: 'module' object has no attribute '__all__' ---- new bug https://bitbucket.org/hpk42/pytest/issue/443/ Misleading documentation on the "skipped/xfail" page ---- new bug https://bitbucket.org/hpk42/pytest/issue/441/ parametrized fixture + module scope + failing test + '-x' option = invalid fixture finalizer output ---- new bug https://bitbucket.org/hpk42/pytest/issue/440/ parametrized fixture output captured inconsistently ---- new bug https://bitbucket.org/hpk42/pytest/issue/439/ capsys fixture does not collect the same test output as reported by pytest ---- new bug https://bitbucket.org/hpk42/pytest/issue/437/ Different tests collected on two nodes with xdist ---- new bug https://bitbucket.org/hpk42/pytest/issue/436/ Command-line arguments that are files which exist cause conftest.py not to be loaded from the usual location ---- new bug https://bitbucket.org/hpk42/pytest/issue/435/ python_files causes reload to fail? ---- new bug https://bitbucket.org/hpk42/pytest/issue/412/ crash on bad stdout ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/488/ Add ability to configure the "patience" of xprocess when waiting for process to start ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/485/ xdist: --boxed test crash throws away STDOUT and STDERR even if they have data ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/468/ guaranteed print to stdout and stderr ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/463/ Add alias for parametrize or provide better error reporting ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/456/ Support per-request scope for fixtures ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/449/ better indicate that there were xpassed test results in a test run ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/442/ Make reported captured output include function scope fixture finalizer output. ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/434/ Standalone test script should allow custom interpreter in shebang ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/430/ Config variable to set locations to try for conftest.py files ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/427/ Option to have pytest rewrite assertions in additional non-test modules ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/421/ Adding external tests to a collection ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/415/ Allow disabling the 'assertions disabled' warning. ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/411/ Documentation: misleading deep inspection ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/410/ No warning when class is skipped due to __init__ ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/401/ Reduce pytest's overhead for running tests ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/393/ non-parameterized fixtures not torn down after last use ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/391/ "native" ipdb as well ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/390/ support pdb.set_trace() with pytest-xdist ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/389/ enhance readability of assertion introspection on bound methods ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/387/ keyword-selection mechanism is too fuzzy ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/374/ plugins reading from shared py.test config file ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/363/ Using monkeypatch in non function scoped fixture ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/351/ Show repr() of exceptions passed to @parametrize ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/344/ Change doctest runner's flags ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/332/ Logging of fixture setup ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/331/ failure when callable object has test name prefix but no __name__ ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/327/ Add improved support for numpy testing ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/313/ Allow callable to be passed to @pytest.mark.parametrize in order to generate tests ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/309/ Doctest count ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/303/ add number of tests per file back into reporting ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/301/ serializing collection process (per host) on xdist to avoid conflicts/collection errors ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/297/ in docs, recwarn.txt lacks any :ref: targets ---- open enhancement https://bitbucket.org/hpk42/pytest/issue/295/ lazy import of doctest in pdb breaks debugging if the environment makes doctest unimportable ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/285/ Allow funcargs to be used in conjunction with default argument values ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/283/ traceback filtering hook ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/282/ allow disabling setuptools plugin discovery ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/277/ Markers as functions ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/272/ flush c buffers for fdcapture after each test ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/268/ isolation when importing test modules (nose-style) ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/263/ testscenarios support ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/259/ Test that unexpectedly passes doesn't produce exit code ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/254/ No visual separation between test failures in narrow terminals ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/253/ pytest should integrate with warnings module ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/252/ Allow fixtures to execute only once per session even if parallel running (xdist) is enabled ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/249/ document and/or change display of parametrization values in IDs ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/230/ Make getting test results from fixtures easier ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/228/ Recwarn empty ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/227/ yield and funcarg don't mix ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/207/ remember/return returnvalues of test functions ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/204/ allow to dynamically modify all ini-values from plugins ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/196/ error reprs not changable availiable via hooks ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/195/ traceback pruning not possible/changable via hooks ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/193/ xdist: support for --boxed on windows ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/175/ way to control how pytest-xdist runs tests in parallel? ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/174/ Configuration Files - Add an option to read them from command line parameter or an environment variable ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/170/ Add traceback matching for expected failures ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/167/ pytest.call_nocapture(func, *k, **kw) ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/166/ exposed api for generating test items ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/162/ Store terminal width on config object ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/157/ --verbose should report why a test case was skipped ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/155/ Bring japanese translation online / reorg website ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/152/ local plugin-list and early conftest opt-out ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/150/ improve pylint / pytest compatibility ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/147/ take a look into documenting this more complete setup.py testrunner ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/138/ support 3.3 chained exceptions ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/137/ Expose an "addfinalizer" to add callable items for cleanup ---- open enhancement https://bitbucket.org/hpk42/pytest/issue/133/ xdist does not color output ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/132/ Pygmentize native tracebacks ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/122/ allow to pass argvalues as an iterator to pytest.mark.parametrize() ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/116/ result log to include stdout\err ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/114/ Skip report refers to skipping plugin when using skipif marker ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/105/ testdir funcarg is slow during setup ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/103/ Move py.code to pytest.code ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/96/ a segfault in a test case causes the testrunner to crash with it ---- open enhancement https://bitbucket.org/hpk42/pytest/issue/82/ sys.path is extended with every dir containg conftest.py ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/62/ better html output for examples (sphinx highlighter) ---- new enhancement https://bitbucket.org/hpk42/pytest/issue/56/ Feature: report exceptions in threads as errors ---- new task https://bitbucket.org/hpk42/pytest/issue/183/ investigate armin rigos conftest ---- new task https://bitbucket.org/hpk42/pytest/issue/161/ investigate trouble around items with the same id ---- new task https://bitbucket.org/hpk42/pytest/issue/142/ move pastebin to plugin ---- new proposal https://bitbucket.org/hpk42/pytest/issue/467/ Cache fixtures which raise pytest.skip.Exception and pytest.fail.Exception ---- new proposal https://bitbucket.org/hpk42/pytest/issue/446/ Alternative syntax: allow no-argument fixture functions to be used as decorators ---- new proposal https://bitbucket.org/hpk42/pytest/issue/432/ Testing docs and doc tests ---- new proposal https://bitbucket.org/hpk42/pytest/issue/431/ opt-in to store the last n sets of reports ---- new proposal https://bitbucket.org/hpk42/pytest/issue/402/ let the capture plugin provide a way to run code with capture suspended ---- new proposal https://bitbucket.org/hpk42/pytest/issue/372/ raises regexp ---- new proposal https://bitbucket.org/hpk42/pytest/issue/364/ shorter traceback style by default ---- new proposal https://bitbucket.org/hpk42/pytest/issue/349/ Using fixtures in pytest.mark.parametrize ---- new proposal https://bitbucket.org/hpk42/pytest/issue/337/ there is no warning when tests have the same name ---- new proposal https://bitbucket.org/hpk42/pytest/issue/310/ Conditional skip and xfail in doctests ---- new proposal https://bitbucket.org/hpk42/pytest/issue/304/ add option to only show tracebacks on keyboard interrup ---- new proposal https://bitbucket.org/hpk42/pytest/issue/291/ tmpdir names get too long on Windows (exceeding PATH_MAX) ---- new proposal https://bitbucket.org/hpk42/pytest/issue/288/ Pass test result to finalizer for cleaning up debug data ---- new proposal https://bitbucket.org/hpk42/pytest/issue/270/ HINT: How to provide coloring in demo with failing tests in docs ---- new proposal https://bitbucket.org/hpk42/pytest/issue/255/ Add a flag to track assertion statistics ---- new proposal https://bitbucket.org/hpk42/pytest/issue/239/ not collect deselected tests ---- new proposal https://bitbucket.org/hpk42/pytest/issue/238/ Introduce "ids" argument to pytest.fixture ---- new proposal https://bitbucket.org/hpk42/pytest/issue/229/ implement something like the nose importer to avoid altering sys.path ---- new proposal https://bitbucket.org/hpk42/pytest/issue/218/ Add a new test hook that allows displaying custom information on an exception ---- new proposal https://bitbucket.org/hpk42/pytest/issue/153/ test intents ---- new proposal https://bitbucket.org/hpk42/pytest/issue/145/ filenames in junitxml output ---- open proposal https://bitbucket.org/hpk42/pytest/issue/136/ Make rsync roots optional not mandatory ---- new proposal https://bitbucket.org/hpk42/pytest/issue/121/ make pytest.mark available outside pytest ---- new proposal https://bitbucket.org/hpk42/pytest/issue/111/ add a hook for serializing/unserializing report sections ---- new proposal https://bitbucket.org/hpk42/pytest/issue/95/ assertion helpers for simplified value checks ---- new proposal https://bitbucket.org/hpk42/pytest/issue/78/ pytest_addoption - Command-line options not added ---- new proposal https://bitbucket.org/hpk42/pytest/issue/73/ sys.excepthook tracking for pytest.raises ---- open proposal https://bitbucket.org/hpk42/pytest/issue/16/ weird setupstate behaviour for yield tests From holger at merlinux.eu Thu Mar 27 08:23:51 2014 From: holger at merlinux.eu (holger krekel) Date: Thu, 27 Mar 2014 07:23:51 +0000 Subject: [pytest-dev] next bug day: Wednesday April 2nd, 2014 In-Reply-To: <20140326094620.GK25870@merlinux.eu> References: <20140326094620.GK25870@merlinux.eu> Message-ID: <20140327072351.GR25870@merlinux.eu> Hi all, it's settled, the next bug day is Wednesday coming week, 2nd April 2014. Bound to participate: holger krekel Jurko Gospodneti? Anatoly Bubenkov Floris Bruynooghe Ronny Pfannschmidt Everyone else: please consider participating as well! See you on #pylib, holger On Wed, Mar 26, 2014 at 09:46 +0000, holger krekel wrote: > Hi pytest users and contributors, > > Since 2.5.2 we've had around 40 new issues and we need help > to resolve these: > > https://bitbucket.org/hpk42/pytest/issues?status=new&status=open > > Please see if you can find the time to look and possibly resolve a few > through a Pull request. Further below i'll dump the titles and issue > links for your convenience. > > Furthermore we should do a new pytest bug day. Here is a doodle: > > http://doodle.com/k26yek52yxvsmpc8#table > > The one choice, the 11th April, is during Pycon in Canada but i assume i'll > still have a lot of Jet Lag and also can make time during Canadian > day time. > > best, > holger > > > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/487/ > request.param mentioned in docs but not listed in the request API docs > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/486/ > `--help` should not fail with an exception on conftest.py errors / handle conftest imports more gracefully/meaningful > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/483/ > 'function' object has no attribute 'im_func' with twisted TestCases on python 3 > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/481/ > Running doctests with -s fails? > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/480/ > SkipTest is not handled properly in setUpModule > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/478/ > --pyargs does not understand namespace packages > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/477/ > excluding name patterns from consideration > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/476/ > Fixtures defined in separate files are not available to tests > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/475/ > Misleading pytest exception description > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/474/ > Running tests with xdist (-n4) causes failures with Django TestCases (DB access denied) > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/473/ > Cannot use @mark and multiple @patch class decorators > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/471/ > pytest keeps running deleted tests > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/470/ > crash when a test changes the cwd to something that gets removed later on without propper cleanup > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/469/ > junitxml parses report.nodeid incorrectly > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/466/ > IOError when writing --junitxml report > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/465/ > running pytest.main(...) from python script breaks cov plugin > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/462/ > What is the correct spelling when using unicode in assertions? > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/461/ > Resources warning when combining capfd and skip() > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/460/ > AttributeError: 'SubRequest' object has no attribute 'param' > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/459/ > doctest-modules + pyreadline means default color becomes black > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/458/ > KeyError: 'COV_CORE_SOURCE' when clearing environment > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/457/ > parser.addoption(type="float") backwards incompatibility > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/455/ > Using xdist with -f option runs tests twice on a detected test source code change > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/454/ > Console output text color missing when pytest run using xdist's -f option > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/453/ > assertion rewriting fails with object whose __repr__ contains '{\n' > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/452/ > Incorrect capfd description/documentation > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/451/ > Incorrectly formatted PDF documentation tables in chapter 7 > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/450/ > pytest --fixtures console output should not include text formatted as rst source > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/448/ > pytest-xdist: AttributeError: 'module' object has no attribute '__all__' > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/443/ > Misleading documentation on the "skipped/xfail" page > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/441/ > parametrized fixture + module scope + failing test + '-x' option = invalid fixture finalizer output > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/440/ > parametrized fixture output captured inconsistently > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/439/ > capsys fixture does not collect the same test output as reported by pytest > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/437/ > Different tests collected on two nodes with xdist > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/436/ > Command-line arguments that are files which exist cause conftest.py not to be loaded from the usual location > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/435/ > python_files causes reload to fail? > ---- > new bug https://bitbucket.org/hpk42/pytest/issue/412/ > crash on bad stdout > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/488/ > Add ability to configure the "patience" of xprocess when waiting for process to start > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/485/ > xdist: --boxed test crash throws away STDOUT and STDERR even if they have data > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/468/ > guaranteed print to stdout and stderr > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/463/ > Add alias for parametrize or provide better error reporting > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/456/ > Support per-request scope for fixtures > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/449/ > better indicate that there were xpassed test results in a test run > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/442/ > Make reported captured output include function scope fixture finalizer output. > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/434/ > Standalone test script should allow custom interpreter in shebang > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/430/ > Config variable to set locations to try for conftest.py files > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/427/ > Option to have pytest rewrite assertions in additional non-test modules > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/421/ > Adding external tests to a collection > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/415/ > Allow disabling the 'assertions disabled' warning. > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/411/ > Documentation: misleading deep inspection > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/410/ > No warning when class is skipped due to __init__ > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/401/ > Reduce pytest's overhead for running tests > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/393/ > non-parameterized fixtures not torn down after last use > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/391/ > "native" ipdb as well > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/390/ > support pdb.set_trace() with pytest-xdist > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/389/ > enhance readability of assertion introspection on bound methods > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/387/ > keyword-selection mechanism is too fuzzy > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/374/ > plugins reading from shared py.test config file > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/363/ > Using monkeypatch in non function scoped fixture > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/351/ > Show repr() of exceptions passed to @parametrize > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/344/ > Change doctest runner's flags > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/332/ > Logging of fixture setup > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/331/ > failure when callable object has test name prefix but no __name__ > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/327/ > Add improved support for numpy testing > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/313/ > Allow callable to be passed to @pytest.mark.parametrize in order to generate tests > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/309/ > Doctest count > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/303/ > add number of tests per file back into reporting > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/301/ > serializing collection process (per host) on xdist to avoid conflicts/collection errors > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/297/ > in docs, recwarn.txt lacks any :ref: targets > ---- > open enhancement https://bitbucket.org/hpk42/pytest/issue/295/ > lazy import of doctest in pdb breaks debugging if the environment makes doctest unimportable > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/285/ > Allow funcargs to be used in conjunction with default argument values > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/283/ > traceback filtering hook > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/282/ > allow disabling setuptools plugin discovery > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/277/ > Markers as functions > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/272/ > flush c buffers for fdcapture after each test > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/268/ > isolation when importing test modules (nose-style) > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/263/ > testscenarios support > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/259/ > Test that unexpectedly passes doesn't produce exit code > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/254/ > No visual separation between test failures in narrow terminals > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/253/ > pytest should integrate with warnings module > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/252/ > Allow fixtures to execute only once per session even if parallel running (xdist) is enabled > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/249/ > document and/or change display of parametrization values in IDs > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/230/ > Make getting test results from fixtures easier > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/228/ > Recwarn empty > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/227/ > yield and funcarg don't mix > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/207/ > remember/return returnvalues of test functions > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/204/ > allow to dynamically modify all ini-values from plugins > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/196/ > error reprs not changable availiable via hooks > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/195/ > traceback pruning not possible/changable via hooks > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/193/ > xdist: support for --boxed on windows > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/175/ > way to control how pytest-xdist runs tests in parallel? > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/174/ > Configuration Files - Add an option to read them from command line parameter or an environment variable > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/170/ > Add traceback matching for expected failures > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/167/ > pytest.call_nocapture(func, *k, **kw) > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/166/ > exposed api for generating test items > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/162/ > Store terminal width on config object > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/157/ > --verbose should report why a test case was skipped > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/155/ > Bring japanese translation online / reorg website > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/152/ > local plugin-list and early conftest opt-out > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/150/ > improve pylint / pytest compatibility > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/147/ > take a look into documenting this more complete setup.py testrunner > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/138/ > support 3.3 chained exceptions > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/137/ > Expose an "addfinalizer" to add callable items for cleanup > ---- > open enhancement https://bitbucket.org/hpk42/pytest/issue/133/ > xdist does not color output > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/132/ > Pygmentize native tracebacks > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/122/ > allow to pass argvalues as an iterator to pytest.mark.parametrize() > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/116/ > result log to include stdout\err > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/114/ > Skip report refers to skipping plugin when using skipif marker > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/105/ > testdir funcarg is slow during setup > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/103/ > Move py.code to pytest.code > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/96/ > a segfault in a test case causes the testrunner to crash with it > ---- > open enhancement https://bitbucket.org/hpk42/pytest/issue/82/ > sys.path is extended with every dir containg conftest.py > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/62/ > better html output for examples (sphinx highlighter) > ---- > new enhancement https://bitbucket.org/hpk42/pytest/issue/56/ > Feature: report exceptions in threads as errors > ---- > new task https://bitbucket.org/hpk42/pytest/issue/183/ > investigate armin rigos conftest > ---- > new task https://bitbucket.org/hpk42/pytest/issue/161/ > investigate trouble around items with the same id > ---- > new task https://bitbucket.org/hpk42/pytest/issue/142/ > move pastebin to plugin > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/467/ > Cache fixtures which raise pytest.skip.Exception and pytest.fail.Exception > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/446/ > Alternative syntax: allow no-argument fixture functions to be used as decorators > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/432/ > Testing docs and doc tests > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/431/ > opt-in to store the last n sets of reports > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/402/ > let the capture plugin provide a way to run code with capture suspended > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/372/ > raises regexp > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/364/ > shorter traceback style by default > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/349/ > Using fixtures in pytest.mark.parametrize > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/337/ > there is no warning when tests have the same name > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/310/ > Conditional skip and xfail in doctests > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/304/ > add option to only show tracebacks on keyboard interrup > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/291/ > tmpdir names get too long on Windows (exceeding PATH_MAX) > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/288/ > Pass test result to finalizer for cleaning up debug data > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/270/ > HINT: How to provide coloring in demo with failing tests in docs > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/255/ > Add a flag to track assertion statistics > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/239/ > not collect deselected tests > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/238/ > Introduce "ids" argument to pytest.fixture > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/229/ > implement something like the nose importer to avoid altering sys.path > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/218/ > Add a new test hook that allows displaying custom information on an exception > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/153/ > test intents > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/145/ > filenames in junitxml output > ---- > open proposal https://bitbucket.org/hpk42/pytest/issue/136/ > Make rsync roots optional not mandatory > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/121/ > make pytest.mark available outside pytest > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/111/ > add a hook for serializing/unserializing report sections > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/95/ > assertion helpers for simplified value checks > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/78/ > pytest_addoption - Command-line options not added > ---- > new proposal https://bitbucket.org/hpk42/pytest/issue/73/ > sys.excepthook tracking for pytest.raises > ---- > open proposal https://bitbucket.org/hpk42/pytest/issue/16/ > weird setupstate behaviour for yield tests > _______________________________________________ > Pytest-dev mailing list > Pytest-dev at python.org > https://mail.python.org/mailman/listinfo/pytest-dev >