Commit Graph

2455 Commits

Author SHA1 Message Date
Ran Benita 9640c9c9eb skipping: use plain compile() instead of _pytest._code.compile()
eval() is used for evaluating string conditions in skipif/xfail e.g.

    @pytest.mark.skipif("1 == 0")

This is the only code that uses `_pytest._code.compile()`, so removing
its last use enables us to remove it entirely.

In this case it doesn't add much. Plain compile() gives a good enough
error message.

For regular exceptions, the message is the same.

For SyntaxError exceptions, e.g. "1 ==", the previous code adds a little
bit of useful context:

```
invalid syntax (skipping.py:108>, line 1)

The above exception was the direct cause of the following exception:
1 ==
     ^
(code was compiled probably from here: <0-codegen /pytest/src/_pytest/skipping.py:108>) (line 1)

During handling of the above exception, another exception occurred:
Error evaluating 'skipif' condition
    1 ==
         ^
SyntaxError: invalid syntax
```

The new code loses it:

```
unexpected EOF while parsing (<skipif condition>, line 1)

During handling of the above exception, another exception occurred:
Error evaluating 'skipif' condition
    1 ==
        ^
SyntaxError: invalid syntax
```

Since the old message is a minor improvement to an unlikely error
condition in a deprecated feature, I think it is not worth all the code
that it requires.
2020-07-01 20:20:12 +03:00
Ran Benita 4a27d7d973 code/source: remove unused method Source.putaround() 2020-07-01 20:20:11 +03:00
Ran Benita a7303b52db code/source: remove unused method Source.isparseable() 2020-07-01 20:20:11 +03:00
Ran Benita a127a22d13 code/source: remove support for comparing Source with str
Cross-type comparisons like this are a bad idea. This isn't used.
2020-07-01 20:20:10 +03:00
Ran Benita 2b99bfbc60 code/source: remove support for passing multiple parts to Source
It isn't used, so keep it simple.
2020-07-01 20:20:09 +03:00
Ran Benita c83e16ab2e code/source: remove unneeded assert
inspect.getsource() definitely returns str.
2020-07-01 20:20:09 +03:00
Ran Benita c6083ab970 code/source: remove old IndentationError workaround in getsource()
This has been there since as far as the git history goes (2007), is not
covered by any test, and says "Buggy python version consider upgrading".
Hopefully everyone have upgraded...
2020-07-01 20:20:08 +03:00
Ran Benita 4108174777 code/source: remove Source(deindent: bool) parameter
Not used, except in tests.
2020-07-01 20:20:07 +03:00
Ran Benita 2fe178488a code/source: expose deindent kwarg in signature
Probably was done to avoid the shadowing issue, but work around it
instead.
2020-07-01 20:20:06 +03:00
Ran Benita 7b1ba7c0db pytester: slightly clean up LsofFdLeakChecker 2020-07-01 19:39:48 +03:00
Ruaridh Williamson 304f2916fc logging: use unique handlers for caplog and reports
Setting log_level via the CLI or .ini will control the log level of the
report that is dumped upon failure of a test.

If caplog modified the log level during the execution of that test, it
should not impact the level that is displayed upon failure in the
"captured log report" section.

[
ran:
- rebased
- reused handler
- changed store keys also to "caplog_handler_*"
- added changelog
all bugs are mine :)
]
2020-07-01 19:30:03 +03:00
Ran Benita ae83dbd4cf python: remove ancient Function.repr_failure(outerr) parameter
This has been asserted like this since 04e9197fd6
(i.e. 11 years, pytest 1.0), seems safe to simply remove at this point.
2020-06-30 13:14:03 +03:00
Ran Benita 40c355f8c3 python: pytest_pycollect_makeitem doesn't need to be a hookwrapper
A trylast is more appropriate for this usecase.

hookwrappers are more complicated and more expensive than regular
hookimpls, so better avoided.
2020-06-30 13:14:03 +03:00
Ran Benita e492b1d567 python: don't pass entire Item for generating ID
Just the nodeid is enough for the error messages.
This removes an import cycle.
2020-06-30 13:13:45 +03:00
Ran Benita e6e300e729
Merge pull request #7396 from gnikonorov/issue_7295
Refactor src/_pytest/config/__init__.py to use the warnings module instead of stderr for warnings
2020-06-28 18:02:07 +03:00
Gleb Nikonorov 49ec2aed0f change stacklevel in warnings from 2 to 3 2020-06-28 10:48:33 -04:00
Ran Benita 992a7a8d38
Merge pull request #7397 from lgeiger/reduce-ihook-calls
Reduce calls to item.ihook
2020-06-28 15:01:43 +03:00
Gleb Nikonorov c39655725a change if else structure of _warn_bout_missing_assertion 2020-06-27 19:49:19 -04:00
Gleb Nikonorov 72a39117a8 Merge remote-tracking branch 'origin' into issue_7295 2020-06-27 19:43:03 -04:00
Bruno Oliveira 289197ff22 Remove package scope experimental status
Close #7389
2020-06-27 11:16:50 -03:00
Ran Benita 7450b6dd95
Merge pull request #7418 from bluetech/typing-3
More typing work
2020-06-27 10:54:29 +03:00
gdhameeja 03230b4002 Fix-6906: Added code-highlight option to disable highlighting optionally
Co-authored-by: Ran Benita <ran@unusedvar.com>
2020-06-26 13:24:56 -07:00
Lukas Geiger 97d2c711e6 Reduce calls of Node.ihook 2020-06-26 14:46:37 +02:00
Ran Benita f382a6bb20 hookspec: remove unused hookspec pytest_doctest_prepare_content()
It's been unused for 10 years at lest from bb50ec89a9.
2020-06-25 17:44:56 +03:00
Ran Benita 256a5d8b14 hookspec: improve typing of some remaining hooks 2020-06-25 17:33:21 +03:00
Ran Benita 97a11726e2 freeze_support: type annotate 2020-06-25 15:28:34 +03:00
Ran Benita 8f8f472379 python_api: type annotate some parts of pytest.approx() 2020-06-25 15:15:24 +03:00
Ran Benita 142d8963e6 recwarn: type annotate pytest.deprecated_call
Also improve its documentation.
2020-06-25 14:40:26 +03:00
Ran Benita 653c83e127 recwarn: type annotate recwarn fixture 2020-06-25 14:30:42 +03:00
Ran Benita 4d813fdf5e recwarn: improve return type annotation of non-contextmanager pytest.warns
It returns the return value of the function.
2020-06-25 14:30:42 +03:00
Ran Benita f00bec2a12 Replace yield_fixture -> fixture in internal code
`yield_fixture` is a deprecated alias to `fixture`.
2020-06-25 14:05:46 +03:00
Gleb Nikonorov 0dd77b3e39 Merge remote-tracking branch 'origin/master' into issue_7295 2020-06-23 23:21:36 -04:00
David Diaz Barquero 617bf8be5b
Add details to error message for junit (#7390)
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
2020-06-23 13:03:46 -03:00
Ran Benita 3624acb665
Merge pull request #7401 from bluetech/typing-config
config: improve typing
2020-06-23 18:58:21 +03:00
Ran Benita 8994e1e3a1 config: make _get_plugin_specs_as_list a little clearer and more general 2020-06-23 11:47:52 +03:00
Ran Benita 04a6d37823 nodes: fix string possibly stored in Node.keywords instead of MarkDecorator
This mistake was introduced in 7259c453d6.
2020-06-22 16:51:11 +03:00
Ran Benita 4655b79985 config: improve typing 2020-06-22 16:39:14 +03:00
Ran Benita 83891d9022
Merge pull request #7387 from cool-RR/2020-06-11-raise-from
Fix exception causes all over the codebase
2020-06-21 22:42:18 +03:00
Ran Benita 99d34ba029
Merge pull request #7388 from bluetech/mark-evaluate
skipping: refactor mark evaluation
2020-06-21 21:47:56 +03:00
Ran Benita 27492cf7a0
Merge pull request #7379 from bluetech/typing-builtin-marks
Type annotate pytest.mark.{skip,skipif,xfail,parametrize,usefixtures,filterwarnings}
2020-06-21 20:44:27 +03:00
Ran Benita b3fb5a2d47 Type annotate pytest.mark.* builtin marks 2020-06-21 20:18:52 +03:00
Ran Benita 7d8d1b4440 skipping: better links in --markers output
Suggested by Bruno.
2020-06-21 20:15:10 +03:00
Ran Benita c9737ae914 skipping: simplify xfail handling during call phase
There is no need to do the XPASS check here, pytest_runtest_makereport
already handled that (the current handling there is dead code).

All the hook needs to do is refresh the xfail evaluation if needed, and
check the NOTRUN condition again.
2020-06-21 19:54:29 +03:00
Gleb Nikonorov ac89d6532a replace stderr warnings with the warnings module 2020-06-20 12:15:58 -04:00
Ran Benita 3e6fe92b7e skipping: refactor skipif/xfail mark evaluation
Previously, skipif/xfail marks were evaluated using a `MarkEvaluator`
class. I found this class very difficult to understand.

Instead of `MarkEvaluator`, rewrite using straight functions which are
hopefully easier to follow.

I tried to keep the semantics exactly as before, except improving a few
error messages.
2020-06-20 16:15:28 +03:00
Ram Rachum dd446bee5e Fix exception causes all over the codebase 2020-06-19 22:02:24 +03:00
Ran Benita 6072c9950d skipping: move MarkEvaluator from _pytest.mark.evaluate to _pytest.skipping
This type was actually in `_pytest.skipping` previously, but was moved to
`_pytest.mark.evaluate` in cf40c0743c.

I think the previous location was more appropriate, because the
`MarkEvaluator` is not a generic mark facility, it is explicitly and
exclusively used by the `skipif` and `xfail` marks to evaluate their
particular set of arguments. So it is better to put it in the plugin
code.

Putting `skipping` related functionality into the core `_pytest.mark`
module also causes some import cycles which we can avoid.
2020-06-19 13:33:54 +03:00
Ran Benita a1f841d5d2 skipping: use pytest_runtest_call instead of pytest_pyfunc_call
`@pytest.mark.xfail` is meant to work with arbitrary items, and there is
a test `test_mark_xfail_item` which verifies this.

However, the code for some reason uses `pytest_pyfunc_call` for the
call phase check, which only works for Function items. The test
mentioned above only passed "accidentally" because the
`pytest_runtest_makereport` hook also runs a `evalxfail.istrue()` which
triggers and evaluation, but conceptually it shouldn't do that.

Change to `pytest_runtest_call` to make the xfail checking properly
generic.
2020-06-19 13:33:53 +03:00
Gleb Nikonorov 4cc4ebf3c9
Don't treat ini keys defined in conftest.py as invalid (#7384) 2020-06-18 12:58:41 -03:00
Andrew a67c553beb
Disable caching when evaluating expressions in marks (#7373) 2020-06-16 12:39:36 +03:00
Ran Benita b6fd89ef31
Merge pull request #7370 from bluetech/typing3
Some type annotations, doc improvements
2020-06-16 11:23:57 +03:00
Bruno Oliveira db00612b84
Merge pull request #7364 from nicoddemus/parseoutcomes-plural-6505
assertoutcomes() only accepts plural forms
2020-06-15 20:27:40 -03:00
Ronny Pfannschmidt 4e5a7b7d2f
Merge pull request #7366 from gnikonorov/issue_7346
Allow users to provide version information to the required_plugins INI key
2020-06-15 20:19:05 +02:00
Ran Benita 7b77fc086a
Merge pull request #7368 from bluetech/teardown-in-setup
runner: don't try to teardown previous items from pytest_runtest_setup
2020-06-15 19:41:21 +03:00
Ran Benita 4f4c2638d0
Merge pull request #7351 from cool-RR/2020-06-11-raise-from
Fix exception causes in config/__init__.py
2020-06-14 18:03:56 +03:00
Ran Benita da1124eb98 hookspec: improve runtest hooks documentation 2020-06-14 17:31:38 +03:00
Ran Benita 2a38ca8a0c doc/reference: add CollectReport
CollectReport appears in several hooks, so we should document it. It's
runtest equivalent TestReport is already documented.
2020-06-14 16:55:41 +03:00
Ran Benita bb878a2b13 runner: don't try to teardown previous items from pytest_runtest_setup
While working on improving the documentation of the
`pytest_runtest_setup` hook, I came up with this text:

> Called to perform the setup phase of the test item.
>
> The default implementation runs ``setup()`` on item and all of its
> parents (which haven't been setup yet). This includes obtaining the
> values of fixtures required by the item (which haven't been obtained
> yet).

But upon closer inspection I noticed this line at the start of
`SetupState.prepare` (which is what does the actual work for
`pytest_runtest_setup`):

    self._teardown_towards(needed_collectors)

which implies that the setup phase of one item might trigger teardowns
of *previous* items. This complicates the simple explanation. It also
seems like a completely undesirable thing to do, because it breaks
isolation between tests -- e.g. a failed teardown of one item shouldn't
cause the failure of some other items just because it happens to run
after it.

So the first thing I tried was to remove that line and see if anything
breaks -- nothing did. At least pytest's own test suite runs fine. So
maybe it's just dead code?
2020-06-14 13:48:05 +03:00
Ran Benita 314d00968a hookspec: type annotate pytest_runtest_log{start,finish} 2020-06-14 12:51:06 +03:00
Ran Benita bb7b3af9b9 hookspec: fix return type annotation of pytest_runtest_makereport 2020-06-13 22:30:00 +03:00
Ran Benita 25064eba7a pytest.collect: type annotate (backward compat module)
This is just to satisfy typing coverage.
2020-06-13 22:29:24 +03:00
Gleb Nikonorov 320625527a Add more tests and docs 2020-06-13 11:22:18 -04:00
Bruno Oliveira ab6dacf1d1
Introduce --import-mode=importlib (#7246)
Fix #5821

Co-authored-by: Ran Benita <ran@unusedvar.com>
2020-06-13 11:29:01 -03:00
Bruno Oliveira 2c37585f58
Merge pull request #7258 from piotrhm/issue_6471 2020-06-13 11:14:18 -03:00
Gleb Nikonorov 8a022c0ce3 test to make sure precommit is fixed 2020-06-13 09:57:13 -04:00
Gleb Nikonorov 7ea116d74c Merge remote-tracking branch 'origin' into issue_7346 2020-06-13 09:54:23 -04:00
Bruno Oliveira 3151219785 assertoutcomes() only accepts plural forms
Fix #6505
2020-06-13 10:29:22 -03:00
Ran Benita f551cab877
Merge pull request #7358 from bluetech/typing2
More type annotations, fix some typing bugs
2020-06-13 10:44:11 +03:00
Gleb Nikonorov 6f8633cc17 add in solution barring documentation 2020-06-13 02:47:15 -04:00
Bruno Oliveira aaa6f1c3fa
Merge pull request #7330 from gnikonorov/issue_7305 2020-06-12 19:35:38 -03:00
Ram Rachum caa984c029 Fix exception causes in config/__init__.py 2020-06-12 18:49:57 +03:00
Ran Benita a5ab7c19fb config: reject minversion if it's a list instead of a single string
Fixes:

src/_pytest/config/__init__.py:1071: error: Argument 1 to "Version" has incompatible type "Union[str, List[str]]"; expected "str"  [arg-type]
2020-06-12 17:34:31 +03:00
Ran Benita 1cf9405075 Fix some type errors around py.path.local
These errors are found using a typed version of py.path.local.
2020-06-12 17:34:31 +03:00
Ran Benita 0256cb3aae hookspec: type annotate pytest_internalerror
Also switch to using ExceptionRepr instead of
`Union[ReprExceptionInfo, ExceptionChainRepr]`
which is somewhat annoying and less future proof.
2020-06-12 17:34:31 +03:00
Ran Benita 7081ed19b8 hookspec: type annotate pytest_keyboard_interrupt 2020-06-12 17:34:31 +03:00
Ran Benita b4f046b777 monkeypatch: add type annotations 2020-06-12 17:34:31 +03:00
Ran Benita f84ffd9747 Remove unused type: ignores
Not needed since update from mypy 0.770 -> 0.780.
2020-06-12 17:34:31 +03:00
Zac Hatfield-Dodds 0f30103d9c
Merge pull request #7356 from Zac-HD/emancipate
Finish deprecation of "slave"
2020-06-12 23:13:11 +10:00
Zac-HD 564b2f707d Finish deprecation of "slave" 2020-06-12 22:49:33 +10:00
Gleb Nikonorov 2c8e356174 rename _emit_warning_or_fail to _warn_or_fail_if_strict and fix a doc typo 2020-06-12 08:27:55 -04:00
Bruno Oliveira ab331c906e Suppress errors while removing tmpdir's lock files
Fix #5456
2020-06-11 18:59:51 -03:00
Gleb Nikonorov c18afb59f5 final touches 2020-06-10 19:09:24 -04:00
Gleb Nikonorov 95cb7fb676 review feedback 2020-06-10 00:44:22 -04:00
ibriquem e2e7f15b71 Make dataclasses/attrs comparison recursive, fixes #4675 2020-06-09 14:39:40 -03:00
piotrhm f0e47c1ed6 Fix typo 2020-06-09 11:17:31 -03:00
piotrhm bde0ebcda9 Replace cleanup_numbered_dir with atexit.register 2020-06-09 11:17:31 -03:00
Bruno Oliveira 357f9b6e83 Add type annotations 2020-06-08 22:26:14 -03:00
piotrhm 51fb11c1d1 Added tests 2020-06-08 22:26:14 -03:00
piotrhm 0b70300ba4 Added requested modifications 2020-06-08 22:26:14 -03:00
Bruno Oliveira fcbaab8b0b
Allow tests to override "global" `log_level` (rebased) (#7340)
Co-authored-by: Ruaridh Williamson <ruaridh.williamson@flexciton.com>
2020-06-08 22:05:46 -03:00
Prashant Anand e78207c936
7119: data loss with mistyped --basetemp (#7170)
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
Co-authored-by: Ran Benita <ran@unusedvar.com>
2020-06-08 21:54:22 -03:00
Fabio Zadrozny 322190fd84
Fix issue where working dir becomes wrong on subst drive on Windows. Fixes #5965 (#6523)
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
2020-06-08 10:56:40 -03:00
Bruno Oliveira c17d50829f
Add pyproject.toml support (#7247) 2020-06-08 10:03:10 -03:00
Gleb Nikonorov d2bb67bfda validate plugins before keys in config files 2020-06-07 14:10:20 -04:00
Gleb Nikonorov f1746c50ea Merge remote-tracking branch 'origin/master' into issue_7305 2020-06-06 12:37:56 -04:00
Gleb Nikonorov 3f6b3e7faa update help for --strict-config 2020-06-06 11:33:28 -04:00
Gleb Nikonorov f760b105ef Touchup pre-PR 2020-06-06 11:06:08 -04:00
Gleb Nikonorov 2a3c21645e Commit solution thus far, needs to be polished up pre PR 2020-06-06 02:38:18 -04:00
Ran Benita 1deaa74345 mark/expression: prevent creation of illegal Python identifiers
This is rejected by Python DEBUG builds, as well as regular builds in
future versions.
2020-06-05 15:57:48 +03:00
Ran Benita 2b05faff0a Improve types around repr_failure() 2020-06-05 11:34:20 +03:00
Ran Benita 54ad048be7 Enable check_untyped_defs mypy option for testing/ too 2020-06-05 11:34:20 +03:00
Ran Benita 71dfdca4df Enable check_untyped_defs mypy option for src/
This option checks even functions which are not annotated. It's a good
step to ensure that existing type annotation are correct.

In a Pareto fashion, the last few holdouts are always the ugliest,
beware.
2020-06-05 11:34:20 +03:00
Ran Benita 848ab00663 Type annotate `@pytest.mark.foo` 2020-06-05 11:34:20 +03:00
Ran Benita c0af19d8ad Type annotate more of _pytest.terminal 2020-06-05 11:34:20 +03:00
Ran Benita 2833884688 Type annotate pytest.fixture and more improvements to _pytest.fixtures 2020-06-05 11:34:20 +03:00
Ran Benita 8bcf1d6de1 Remove duplicated conversion of pytest.fixture() params argument
The FixtureFunctionMarker attrs class already converts the params
itself.

When adding types, the previous converter composition causes some type
error, but extracting it to a standalone function fixes the issue (a
lambda is not supported by the mypy plugin, currently).
2020-06-05 11:34:20 +03:00
Ran Benita 1bd7d025d9 Type annotate more of _pytest.fixtures 2020-06-05 11:34:20 +03:00
Ran Benita f8bb61ae5b Type annotate _pytest.warnings 2020-06-05 11:34:20 +03:00
Ran Benita 01797e6370 Type annotate _pytest.debugging (a bit) 2020-06-05 11:34:20 +03:00
Ran Benita 216a010ab7 Type annotate _pytest.junitxml 2020-06-05 11:34:20 +03:00
Ran Benita 3e351afeb3 Type annotate _pytest.capture 2020-06-05 11:34:20 +03:00
Ran Benita b51ea4f1a5 Type annotate _pytest.unittest 2020-06-05 11:34:20 +03:00
Ran Benita db52928684 Type annotate _pytest.logging 2020-06-05 11:34:20 +03:00
Ran Benita 90e58f8961 Type annotate some parts related to runner & reports 2020-06-05 11:34:20 +03:00
Ran Benita 709bcbf3c4 Type annotate _pytest.mark.evaluate 2020-06-05 11:34:20 +03:00
Ran Benita fc325bc0c3 Type annotate more of _pytest.nodes 2020-06-05 11:34:19 +03:00
Ran Benita 32dd0e87cb Type annotate _pytest.doctest 2020-06-05 11:34:19 +03:00
Ran Benita 387d9d04f7 Type annotate tricky reorder_items() function in fixtures.py 2020-06-05 11:34:19 +03:00
Ran Benita e68a26199c Type annotate misc functions 2020-06-05 11:34:19 +03:00
Ran Benita d95132178c Type annotate _pytest.assertion 2020-06-05 11:34:19 +03:00
Ran Benita 30e3d473c4 Type annotate _pytest._io.saferepr 2020-06-05 11:34:19 +03:00
Ran Benita 247c4c0482 Type annotate some more hooks & impls 2020-06-05 11:34:19 +03:00
Ran Benita ef34729541 Type annotate fixtures.py & related 2020-06-05 11:34:19 +03:00
Ran Benita be00e12d47 Type annotate main.py and some parts related to collection 2020-06-05 11:34:19 +03:00
Ran Benita f8de424241 Type annotate CallSpec2 2020-06-05 11:34:19 +03:00
Ran Benita 0fb081aec6 Type annotate some hookspecs & impls
Annotate some "easy" arguments of hooks that repeat in a lot of internal
plugins.

Not all of the arguments are annotated fully for now.
2020-06-05 11:34:19 +03:00
Ran Benita ff8b7884e8 Type annotate ParameterSet 2020-06-05 11:34:19 +03:00
Ran Benita 43fa1ee8f9 Type annotate some misc places with no particular connection 2020-06-05 11:34:19 +03:00
Ran Benita e7c26a92d7
Merge pull request #7253 from bluetech/fixture-special-case
fixtures: remove special cases when deciding when pytest.fixture() is a direct decoration
2020-06-05 11:33:43 +03:00
Bruno Oliveira cf640a9b67
Merge pull request #7292 from kerizane/7259_iniconfig 2020-06-03 19:00:48 -03:00
Bruno Oliveira 61a6ac9705
Merge pull request #7242 from nicoddemus/fix-flaky-test
Introduce _pytest.timing as a way to control timing during internal tests
2020-06-02 20:25:02 -03:00
Bruno Oliveira 8ac18bbecb Show invalid ini keys sorted
Otherwise this relies on the dictionary order of `config.inicfg`, which
is insertion order in py36+ but "random" order in py35.
2020-06-02 16:01:47 -03:00
Ran Benita 85b5a289f0
warnings: fix missing None in existing hook & add some docs (#7288) 2020-06-02 19:59:25 +03:00
Bruno Oliveira 7a8b8a5c9f
Merge branch 'master' into fix-flaky-test 2020-06-02 11:33:15 -03:00
Ronny Pfannschmidt 8faf1e8eca
Merge pull request #7286 from gnikonorov/issue_6856
Output a warning to stderr when an invalid key is read from an INI config file
2020-06-02 15:47:02 +02:00
Tor Colvin fe64093411
Fix removal of very long paths on Windows (#6755)
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
2020-06-02 08:56:33 -03:00
Keri Volans 2748feed38 7291: Replace py.iniconfig with iniconfig 2020-06-01 18:09:57 +01:00
Bruno Oliveira 70b5bdf4ba
Merge pull request #7264 from bluetech/wcwidth
Improve our own wcwidth implementation and remove dependency on wcwidth package
2020-05-31 12:37:58 -03:00
Gleb Nikonorov 92d15c6af1 review feedback 2020-05-31 11:33:31 -04:00
Ronny Pfannschmidt c16ede5ce6
Merge pull request #7255 from gnikonorov/issue_4049
Add new hook pytest_warning_recorded
2020-05-31 12:47:46 +02:00
Gleb Nikonorov db203afba3 Add in --strict-config flag to force warnings to errors 2020-05-31 02:45:40 -04:00
Gleb Nikonorov eef4f87e7b Output a warning to stderr when an invalid key is read from an INI config file 2020-05-30 20:36:02 -04:00
Bruno Oliveira 757bded135 Use Path() instead of str for path comparison
On Windows specifically is common to have drives diverging just by
casing ("C:" vs "c:"), depending on the cwd provided by the user.
2020-05-30 20:05:21 -03:00
Daniel Hahler fb9f277a99 Node._repr_failure_py: use abspath with changed cwd
Fixes https://github.com/pytest-dev/pytest/issues/6428.
2020-05-30 20:02:47 -03:00
Bruno Oliveira 56bf819c2f
Do not call TestCase.tearDown for skipped tests (#7236)
Fix #7215
2020-05-30 14:33:22 -03:00
Katarzyna Król 94c7b8b47c
Issue 1316 - longrepr is a string when pytrace=False (#7100) 2020-05-30 14:10:58 +03:00
Ran Benita 2ee90887b7 code: remove last usage of py.error
`str(self.path)` can't raise at all, so it can just be removed.
2020-05-28 12:12:10 +03:00
Gleb Nikonorov 14de08011b fix the unit tests, add the proper deprecation warning, and add in a changelog entry 2020-05-27 23:03:07 -04:00
Anthony Sottile b3db440d4c
Merge pull request #7257 from DahlitzFlorian/fix-issue-6956
Prevent pytest from printing ConftestImportFailure traceback
2020-05-27 13:49:43 -07:00
Ran Benita 7a704288df capture: remove unneeded getattr
This attribute is set in __init__ and not deleted. Other methods do it
already but this one wasn't updated.
2020-05-27 15:27:16 +03:00
Ran Benita a35800c2e1 capture: formalize and check allowed state transition in capture classes
There are state transitions start/done/suspend/resume and two additional
operations snap/writeorg.

Previously it was not well defined in what order they can be called, and
which operations are idempotent.

Formalize this and enforce using assert checks with informative error
messages if they fail (rather than random AttributeErrors).
2020-05-27 15:27:15 +03:00
Ran Benita fd3ba053cf capture: don't assume that the tmpfile is backed by a BytesIO
Since tmpfile is a parameter to SysCapture, it shouldn't assume things
unnecessarily, when there is an alternative.
2020-05-27 15:19:30 +03:00
Ran Benita 97bcf5a3a2 capture: reorder file into sections and avoid forward references
Make it easier to read the file in progression, and avoid forward
references for upcoming type annotations.

There is one cycle, CaptureManager <-> CaptureFixture, which is hard to
untangle.

(This commit should be added to `.gitblameignore`).
2020-05-27 15:19:28 +03:00
Gleb Nikonorov d742b386c3 provide missing location parameter, and add type annotations to the hookspec 2020-05-27 00:53:31 -04:00
Ran Benita aca534c67d Improve our own wcwidth implementation and remove dependency on wcwidth package
`TerminalWriter`, imported recently from `py`, contains its own
incomplete wcwidth (`char_with`/`get_line_width`) implementation. The
`TerminalReporter` also needs this, but uses the external `wcwidth`
package.

This commit brings the `TerminalWriter` implementation up-to-par with
`wcwidth`, moves to implementation to a new file `_pytest._io.wcwidth`
which is used everywhere, and removes the dependency.

The differences compared to the `wcwidth` package are:

- Normalizes the string before counting.

- Uses Python's `unicodedata` instead of vendored Unicode tables. This
  means the data corresponds to the Python's version Unicode version
  instead of the `wcwidth`'s package version.

- Apply some optimizations.
2020-05-26 17:14:01 +03:00
Ran Benita 5507752c53 fixtures: remove special cases when deciding when pytest.fixture() is a direct decoration
pytest.fixture() can be used either as

    @pytest.fixture
    def func(): ...

or as

    @pytest.fixture()
    def func(): ...

or (while maybe not intended)

    func = pytest.fixture(func)

so it needs to inspect internally whether it got a function in the first
positional argument or not.

Previously, there were was oddity. In the following,

    func = pytest.fixture(func, autouse=True)
    # OR
    func = pytest.fixture(func, parms=['a', 'b'])

The result is as if `func` wasn't passed.

There isn't any reason for this special that I can understand, so remove
it.
2020-05-26 11:48:40 +03:00
Florian Dahlitz 95bd232e57
Apply suggestions from @bluetech 2020-05-26 10:31:53 +02:00
Ran Benita ea3f44894f capture: replace TeeSysCapture with SysCapture(tee=True)
This is more straightforward and does not require duplicating the
initialization logic.
2020-05-26 00:25:49 +03:00
Ran Benita 02c95ea624 capture: remove unused FDCapture tmpfile argument 2020-05-26 00:25:49 +03:00
Ran Benita 2695b41df3 capture: inline _capturing_for_request to simplify the control flow
With straight code, it is a little easier to understand, and simplify
further.
2020-05-26 00:25:49 +03:00
Ran Benita 491239d9b2 capture: remove some indirection in MultiCapture
Removing this indirection enables some further clean ups.
2020-05-26 00:25:49 +03:00
Florian Dahlitz 5ebcb34fb5
Move ConftestImportFailure check to correct position and add typing 2020-05-25 20:19:28 +02:00
Gleb Nikonorov 125b663f20 Address all feedback, minus the empty sring v None nodeid which is being discussed 2020-05-25 11:18:24 -04:00
Florian Dahlitz 6546d1f725
Prevent pytest from printing ConftestImportFailure traceback 2020-05-25 13:57:03 +02:00
Gleb Nikonorov 088d400b2d rename pytest_warning_record -> pytest_warning_recorded 2020-05-24 20:43:23 -04:00
Gleb Nikonorov b02d087dbd cleanup code pre pr 2020-05-24 20:26:14 -04:00
Gleb Nikonorov 9ee6550181 Add in a new hook pytest_warning_recorded for warning capture
communication
2020-05-24 19:05:24 -04:00
Anthony Sottile 45f53266e6
Merge pull request #7244 from DahlitzFlorian/fix-issue-7150
Prevent hiding underlying exception when ConfTestImportFailure is raised
2020-05-23 12:09:42 -07:00
Bruno Oliveira 1780924b27 Introduce _pytest.timing as a way to control timing during tests
_pytest.timing is an indirection to 'time' functions, which pytest production
code should use instead of 'time' directly.

'mock_timing' is a new fixture which then mocks those functions, allowing us
to write time-reliable tests which run instantly and are not flaky.

This was triggered by recent flaky junitxml tests on Windows related to timing
issues.
2020-05-23 11:50:33 -03:00
Claire Cecil 79701c65ed
Added support for less verbose version information (#7169) 2020-05-23 11:27:58 -03:00
Simon K 05c22ff823
7154-Improve-testdir-documentation-on-makefiles (#7239) 2020-05-23 11:27:06 -03:00
Ran Benita b38edec60f
Merge pull request #7238 from bluetech/micro-optimizations-1
A few tiny micro-optimizations/simplifications
2020-05-23 12:14:24 +03:00
Florian Dahlitz d0eb86cfa6
Prevent hiding underlying exception when ConfTestImportFailure is raised 2020-05-22 22:58:35 +02:00
Ran Benita 4a1557fa0e
Merge pull request #7091 from bluetech/capture-invalid-fd
Perform FD capturing even if the FD is invalid
2020-05-22 14:36:49 +03:00
Ran Benita f1f9c7792b Import `packaging` package lazily 2020-05-22 14:33:29 +03:00
Ran Benita 796fba6788 terminal: remove redundant write_fspath_result call
This is already done in pytest_runtest_logstart, so the fspath is
already guaranteed to have been printed (for xdist, it is disabled
anyway).

write_fspath_result is mildly expensive so it is worth avoiding calling
it twice.
2020-05-22 14:33:29 +03:00
Ran Benita 139a029b5e terminal: remove a redundant line
`write_fspath_result` already does this split.
2020-05-22 14:33:29 +03:00
Ran Benita 8d841ab0b8 nodes: remove unused argument from FSHookProxy 2020-05-22 14:33:28 +03:00
Ran Benita 919ac2239d
Merge pull request #7231 from bluetech/logging-error
logging: propagate errors during log message emits
2020-05-22 14:27:26 +03:00
Ran Benita eaeafd7c30 Perform FD capturing even if the FD is invalid
The `FDCapture`/`FDCaptureBinary` classes, used by `capfd`/`capfdbinary`
fixtures and the `--capture=fd` option (set by default), redirect FDs
1/2 (stdout/stderr) to a temporary file. To do this, they need to save
the old file by duplicating the FD before redirecting it, to be restored
once finished.

Previously, if this duplicating (`os.dup()`) failed, most likely due to
that FD being invalid, the FD redirection would silently not be done. The
FD capturing also performs python-level redirection (monkeypatching
`sys.stdout`/`sys.stderr`) which would still be done, but direct writes
to the FDs would fail.

This is not great. If pytest is run with `--capture=fd`, or a test is
using `capfd`, it expects writes to the FD to work and be captured,
regardless of external circumstances.

So, instead of disabling FD capturing, keep the redirection to a
temporary file, just don't restore it after closing, because there is
nothing to restore to.
2020-05-20 19:32:37 +03:00
Bruno Oliveira 0a03217903
Merge pull request #7046 from blueyed/k-skip-session-upstream 2020-05-19 20:09:36 -03:00
Daniel Hahler 691a7fceea Revisit some help texts with regard to newlines 2020-05-19 19:34:08 -03:00
Daniel Hahler 87423d3cc8 Keep explicit newlines with help texts
This makes a difference for e.g. pytest-xdist:

Before:
```
  --dist=distmode       set mode for distributing tests to exec environments. each: …
                        available environment. loadscope: …
                        grouped by file to any available environment. (default) no: …
```

After:
```
  --dist=distmode       set mode for distributing tests to exec environments.
                        each: send each test to all available environments.
                        load: load balance by sending any pending test to any available environment.
                        …
                        (default) no: run tests inprocess, don't distribute.
```

This might also result in unexpected changes (hard wrapping), when line
endings where used unintentionally, e.g. with:

```
help="""
    some long
    help text
    """
```

But the benefits from that are worth it, and it is easy to fix, as will
be done for the internal `assertmode` option.
2020-05-19 19:34:08 -03:00
Bruno Oliveira 5a6296a2d7
Merge pull request #7226 from nicoddemus/remove-function-args 2020-05-19 18:53:23 -03:00
Ran Benita b13fcb23d7 logging: propagate errors during log message emits
Currently, a bad logging call, e.g.

    logger.info('oops', 'first', 2)

triggers the default logging handling, which is printing an error to
stderr but otherwise continuing.

For regular programs this behavior makes sense, a bad log message
shouldn't take down the program. But during tests, it is better not to
skip over such mistakes, but propagate them to the user.
2020-05-19 11:16:39 +03:00
Ran Benita 85a06cfafb
Merge pull request #7227 from bluetech/logging-reuse-handler
logging: reuse LoggingCaptureHandler instance since it's expensive to create
2020-05-18 23:16:51 +03:00
Ran Benita fe2febfff2
Merge pull request #7019 from bluetech/rm-pyobj_property
Remove pyobj_property helper, inline it instead
2020-05-18 23:13:52 +03:00
Ran Benita 694fdc6554 Remove pyobj_property helper, inline it instead
It doesn't save much code but adds indirection which makes it a bit
harder to follow and to type.
2020-05-18 20:54:11 +03:00
Bruno Oliveira ad3169428b Remove unused Function.__init__ 'args' parameter 2020-05-18 14:30:48 -03:00
Daniel Hahler 8b9b81c3c0
Function: use `originalname` in `_getobj` and make it default to `name` (#7035) 2020-05-18 14:08:47 -03:00
Ran Benita d2d11a8bdc logging: reuse LoggingCaptureHandler instance since it's expensive to create
Previously, a LoggingCaptureHandler was instantiated for each test's
setup/call/teardown which turns out to be expensive.

Instead, only keep one instance and reset it between runs.
2020-05-18 11:27:50 +03:00
Ran Benita e27228a4e4
Merge pull request #7224 from bluetech/logging-simplifications
logging: some simplifications/cleanups
2020-05-18 00:15:59 +03:00
Ran Benita f71ec8cc90 logging: order hookimpl's in chronological order
Makes it easier to understand what's going on.
2020-05-17 21:41:38 +03:00
Ran Benita 3f8200676f logging: remove deprecated --no-print-logs option/ini
This option was deprecated in 5.4.0 and was marked for removal in 6.0.0.
2020-05-17 21:41:38 +03:00
Ran Benita bd5e3f042d logging: move log_file_handler cleanup from sessionend to unconfigure
It is set-up in configure, so match it.
2020-05-17 21:41:38 +03:00
Ran Benita 43c465c9bf logging: use dummy handler when CLI logging is disabled instead of None
This makes the code cleaner by removing conditionals and making the CLI
and file logging completely analogous.

Doesn't affect performance.
2020-05-17 21:41:38 +03:00
Ran Benita bd657bab3f logging: don't use _runtest_for for the pytest_log* hooks
The logstart/logreport/logfinish hooks don't need the stuff in
_runtest_for. The test capturing catching_logs call is irrelevant for
them, and the item-conditional sections are gone.
2020-05-17 21:41:38 +03:00
Ran Benita b13af52bbe logging: call set_when() in a consistent manner 2020-05-17 21:41:38 +03:00
Ran Benita 075903dafa logging: simplify log-file handling
- Instead of making it optional, always set up a handler, but possibly
  going to /dev/null. This simplifies the code by removing a lot of
  conditionals. It also can replace the NullHandler() we already add.

- Change `set_log_path` to just change the stream, instead of recreating
  one. Besides plugging a resource leak, it enables the next item.

- Remove the capturing_logs from _runtest_for, since it sufficiently
  covered by the one in pytest_runtestloop now, which wraps all other
  _runtest_for calls.

The first item alone would have had an adverse performance impact, but
the last item removes it.
2020-05-17 21:41:37 +03:00
Ran Benita e48ac692de logging: optimize catching_logs slightly
Remove usage of `@contextmanager` as it is a bit slower than
hand-rolling, and also disallows re-entry which we want to use.

Removing protections around addHandler()/removeHandler(), because
logging already checks that internally.
2020-05-17 20:42:06 +03:00
Ran Benita eceb28e4be logging: set formatter on handler creation, not in catching_logs
Conceptually it doesn't check per catching_logs (and catching_logs
doesn't restore the older one either). It is just something that is
defined for each handler once.
2020-05-17 20:42:06 +03:00
Ran Benita ce0f218793 logging: yield from _runtest_for instead of contextmanager
Avoid the slight overhead of contextmanager.
2020-05-17 20:42:06 +03:00
Ran Benita 9effbe7425 logging: inline _runtest_for_main into _runtest_for
This avoids a little bit of overhead, and makes the code a bit clearer
too.
2020-05-17 20:42:06 +03:00
Ran Benita ac6c02f1e2 logging: use item's store for private attributes
This makes things type-safe and properly private.
2020-05-17 20:42:05 +03:00
Bruno Oliveira c26f389c09 Refactor handling of non-top-level pytest_plugins handling
Decided to move the 'if' logic together with the error message, as this leaves
the _importconftest method cleaner.
2020-05-17 11:26:30 -03:00
Bruno Oliveira 9e1e7fcabe Use a nice string repr for ConftestImportFailure
The default message is often hard to read:

    E   _pytest.config.ConftestImportFailure: (local('D:\\projects\\pytest\\.tmp\\root\\foo\\conftest.py'), (<class 'RuntimeError'>, RuntimeError('some error',), <traceback object at 0x000001CCC3E39348>))

Using a shorter message is better:

    E   _pytest.config.ConftestImportFailure: RuntimeError: some error (from D:\projects\pytest\.tmp\root\foo\conftest.py)

And we don't really lose any information due to exception chaining.
2020-05-17 11:26:30 -03:00
Bruno Oliveira 5eaebc1900 Remove one of the tracebacks from conftest import failures
This removes the KeyError from the traceback chain when an
conftest fails to import:

    return self._conftestpath2mod[key]
    E   KeyError: WindowsPath('D:/projects/pytest/.tmp/root/foo/conftest.py')

    During handling of the above exception, another exception occurred:
    ...
        raise RuntimeError("some error")
    E   RuntimeError: some error

    During handling of the above exception, another exception occurred:
    ...
    E   _pytest.config.ConftestImportFailure: (...)

By slightly changing the code, we can remove the first chain, which is often
very confusing to users and doesn't help with anything.

Fix #7223
2020-05-17 11:26:02 -03:00
Bruno Oliveira c5b367b4f4 Package.name now contains only basname of the package
Previously it contained the entire path, which made '-k' match
against any name in the full path of the package.

Fix #7040
2020-05-16 15:04:07 -03:00
Daniel Hahler 3d3b9511fd -k should not match session name
Fixes https://github.com/pytest-dev/pytest/issues/7040.
2020-05-16 14:15:57 -03:00
Bruno Oliveira f506682abe
Merge pull request #6919 from nicoddemus/backport-6914 2020-05-16 12:42:15 -03:00
Daniel Hahler d530d70128 Fix regressions with `--lf` plugin
Only filter with known failures, and explicitly keep paths of passed
arguments.

This also displays the "run-last-failure" status before collected files,
and does not update the cache with "--collect-only".

Fixes https://github.com/pytest-dev/pytest/issues/6968.
2020-05-15 18:35:38 -03:00
Ran Benita f0f552d60c
Merge pull request #6283 from felixn/master
Create LogCaptureHandler if necessary (closes #6240)
2020-05-15 09:13:57 +03:00
Ran Benita 2ac28f6c65
Merge pull request #7205 from lancelote/7126
Issue 7126 - "saferepr" to avoid BytesWarning when using --setup-show
2020-05-14 16:04:22 +03:00
Felix Nieuwenhuizen 4dfc461036 Create LogCaptureHandler if necessary (closes #6240) 2020-05-13 20:38:14 +02:00
Ran Benita c714f05ad7 mark: reuse compiled expression for all items in -k/-m
The previous commit made this possible, so utilize it.

Since legacy.py becomes pretty bare, I inlined it into __init__.py. I'm
not sure it's really "legacy" anyway!

Using a simple 50000 items benchmark with `--collect-only -k nomatch`:

Before (two commits ago):

   ======================== 50000 deselected in 10.31s =====================
         19129345 function calls (18275596 primitive calls) in 10.634 seconds

   Ordered by: cumulative time

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
        1    0.001    0.001    2.270    2.270 __init__.py:149(pytest_collection_modifyitems)
        1    0.036    0.036    2.270    2.270 __init__.py:104(deselect_by_keyword)
    50000    0.055    0.000    2.226    0.000 legacy.py:87(matchkeyword)

After:

   ======================== 50000 deselected in 9.37s =========================
         18029363 function calls (17175972 primitive calls) in 9.701 seconds

   Ordered by: cumulative time

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
        1    0.000    0.000    1.394    1.394 __init__.py:239(pytest_collection_modifyitems)
        1    0.057    0.057    1.393    1.393 __init__.py:162(deselect_by_keyword)

The matching itself can be optimized more but that's a different story.
2020-05-12 12:55:37 +03:00
Ran Benita 622c4ce02e mark/expression: support compiling once and reusing for multiple evaluations
In current pytest, the same expression is matched against all items. But
it is re-parsed for every match.

Add support for "compiling" an expression and reusing the result. Errors
may only occur during compilation.

This is done by parsing the expression into a Python `ast.Expression`,
then `compile()`ing it into a code object. Evaluation is then done using
`eval()`.

Note: historically we used to use `eval` directly on the user input --
this is not the case here, the expression is entirely under our control
according to our grammar, we just JIT-compile it to Python as a
(completely safe) optimization.
2020-05-12 12:53:12 +03:00
Ran Benita 645aaa728d python_api: reduce scope of a `except BaseException` in ApproxNumpy
I'm not sure if it can even raise at all, but catching BaseException
would swallow ctrl-C and such and is definitely inappropriate here.
2020-05-12 12:07:56 +03:00
Ran Benita 23c9856857 Remove no longer needed noqa's 2020-05-12 09:29:47 +03:00
Ran Benita 59a12e9ab3 Replace bare `except`s with `except BaseException`
Mostly I wanted to remove uses of `noqa`.

In Python 3 the two are the same.
2020-05-12 09:29:47 +03:00
Ran Benita c4f9eaa5de mark: deprecate a couple undocumented -k syntaxes
The `-k '-expr'` syntax is an old alias to `-k 'not expr'`. It's also
not a very convenient to have syntax that start with `-` on the CLI.
Deprecate it and suggest replacing with `not`.

---

The `-k 'expr:'` syntax discards all items until the first match and
keeps all subsequent, e.g. `-k foo` with

    test_bar
    test_foo
    test_baz

results in `test_foo`, `test_baz`. That's a bit weird, so deprecate it
without a replacement. If someone complains we can reconsider or devise
a better alternative.
2020-05-11 16:07:52 +03:00
Pavel Karateev 7b196747dd Use saferepr for all types 2020-05-10 12:47:26 +03:00
Pavel Karateev 903e2ab6ee Fix #7126 - saferepr for bytes params
bytes parametrize parameters cause error when --setup-show is used
and Python is called with -bb flag
2020-05-09 13:57:17 +03:00
Bruno Oliveira 5dd987e10f
Merge pull request #6999 from bluetech/simplify-fixture-compat 2020-05-08 08:06:35 -03:00
Bruno Oliveira 9657798c12
Merge pull request #7158 from CarycaKatarzyna/issue_7076 2020-05-08 07:56:03 -03:00
Ran Benita 857b5c9ccd
Merge pull request #7185 from bluetech/sigpipe
Handle SIGPIPE/BrokenPipeError in pytest's CLI
2020-05-08 13:32:34 +03:00
Ran Benita 73448f265d Handle EPIPE/BrokenPipeError in pytest's CLI
Running `pytest | head -1` and similar causes an annoying error to be
printed to stderr:

    Exception ignored in: <_io.TextIOWrapper name='<stdout>' mode='w' encoding='utf-8'>
    BrokenPipeError: [Errno 32] Broken pipe

(or possibly even a propagating exception in older/other Python versions).

The standard UNIX behavior is to handle the EPIPE silently. To
recommended method to do this in Python is described here:
https://docs.python.org/3/library/signal.html#note-on-sigpipe

It is not appropriate to apply this recommendation to `pytest.main()`,
which is used programmatically for in-process runs. Hence, change
pytest's entrypoint to a new `pytest.console_main()` function, to be
used exclusively by pytest's CLI, and add the SIGPIPE code there.

Fixes #4375.
2020-05-08 12:51:02 +03:00
Anthony Sottile b238845d0f Fix _is_setup_py for files encoded differently than locale 2020-05-07 13:14:58 -07:00
Ran Benita 65963d2066 warnings: speed up work done in catch_warnings_for_item()
When setting up the warnings capture, filter strings (with the general
form `action:message:category:module:line`) are collected from the
cmdline, ini and item and applied. This happens for every test and other
cases.

To apply a string it needs to be parsed into a tuple, and it turns out
this is slow. Since we already vendor the parsing code from Python's
warnings.py, we can speed it up by caching the result. After splitting
the parsing part from the applying part, the parsing is pure and is
straightforward to cache.

An alternative is to parse ahead of time and reuse the result, however
the caching solution turns out cleaner and more general in this case.

On this benchmark:

    import pytest
    @pytest.mark.parametrize("x", range(5000))
    def test_foo(x): pass

Before:

============================ 5000 passed in 14.11s =============================
         14365646 function calls (13450775 primitive calls) in 14.536 seconds

After:

============================ 5000 passed in 13.61s =============================
         13290372 function calls (12375498 primitive calls) in 14.034 seconds
2020-05-07 12:19:13 +03:00
Ran Benita 81da5dac48
Merge pull request #7122 from bluetech/matcher-parser
Stop using Python's eval() for -m and -k
2020-05-07 09:23:29 +03:00
Bruno Oliveira 6c2d358918
Merge pull request #7135 from pytest-dev/terminalwriter 2020-05-06 18:26:44 -03:00
Bruno Oliveira 4d439760ad
Merge pull request #7168 from nicoddemus/saferepr-getattr-fail-7145 2020-05-06 17:57:38 -03:00
Ran Benita d16ae0bbdc
Merge pull request #7171 from bluetech/code-import-cycles
code: fix import cycles between code.py and source.py
2020-05-06 18:15:32 +03:00
Ran Benita 89eee90b5f python: optimize PythonCollector.collect 2020-05-06 12:05:04 +03:00
Ran Benita b90f34569f nodes: micro-optimize Node attribute access 2020-05-06 11:58:30 +03:00
Ran Benita 5702c86f4c nodes: micro-optimize hash(node)
Turns out it's called alot, and saving the function call makes it
faster.
2020-05-06 11:58:30 +03:00
Ran Benita fcc473ab1c Use dict instead of OrderedDict on Python 3.7
OrderedDict is quite a bit heavier than just a dict.
2020-05-06 11:58:28 +03:00
Ran Benita 69143fe5b0 code: fix import cycles between code.py and source.py
These two files were really intertwined. Make it so code.py depends on
source.py without a reverse dependency.

No functional changes.
2020-05-06 11:01:32 +03:00
Bruno Oliveira d0022b5a13 'saferepr' handles classes with broken __getattribute__
Fix #7145
2020-05-05 19:22:39 -03:00
Ran Benita 94400a68b4 terminal: fix non-deterministic warning summary order in Python 3.5
In Python 3.5, collections.Counter() does not preserve insertion order.
2020-05-05 23:08:44 +03:00
Ran Benita 4b91617002
Merge pull request #7020 from blueyed/fix-warnings-summary
Fix warnings summary
2020-05-05 21:58:57 +03:00
Ran Benita b4be6cd4b7
Merge pull request #6765 from blueyed/capture-refactor-1
Refactor Capture classes: move/rename CaptureIO classes
2020-05-05 21:46:18 +03:00
Daniel Hahler 8cae78a18b Fix warnings summary
- replace "tests with warnings" with just warnings: they a) might not
  come from a test in the first place, and b) a single test might have
  multiple warnings.
- fix usage of (unused) argument to `collapsed_location_report`

Co-authored-by: Ran Benita <ran@unusedvar.com>
2020-05-05 21:40:34 +03:00
Ran Benita d7d627b1e8
Merge pull request #7064 from blueyed/fix-_printcollecteditems-doc-upstream
Fix/improve printing of docs for collected items
2020-05-05 21:31:21 +03:00
Daniel Hahler 7647d1c836 Move Capture classes from compat to capture and improve naming
Move {Passthrough,CaptureIO} to capture module, and rename Passthrough
-> Tee to match the existing terminology.

Co-authored-by: Ran Benita <ran@unusedvar.com>
2020-05-05 21:24:59 +03:00
Ran Benita 46d768503e
Merge pull request #6839 from blueyed/typing-get_dirs_from_args-parseconfig
typing: get_dirs_from_args, parseconfig
2020-05-05 21:11:26 +03:00
Ran Benita 85732e2b08
Merge pull request #6993 from blueyed/doc-pytest_collection-upstream
[WIP] doc: pytest_collection: has to set `session.items`
2020-05-05 21:05:56 +03:00
Bruno Oliveira 16a44823eb Use reportinfo() instead of location in skipping message 2020-05-04 19:50:40 -03:00
Katarzyna 402ee6fb9d Relative path to invocationdir instead rootdir. 2020-05-03 22:56:38 +02:00
Bruno Oliveira 80e5098408
Merge pull request #7155 from kerizane/make_numbered_dir 2020-05-03 09:48:36 -03:00
Keri Volans 678440e46d 7018: Use internal version of make_numbered_dir 2020-05-03 10:45:06 +01:00
Bruno Oliveira 5c2e96c0e6 Fix cleanup functions not being invoked on test failures
Also delay calling tearDown() when --pdb is given, so users still have
access to the instance variables (which are usually cleaned up during tearDown())
when debugging.

Fix #6947
2020-05-02 15:26:55 -03:00
Bruno Oliveira 2b51ed46d5
Merge pull request #7153 from nicoddemus/xunit-warning-update 2020-05-02 13:31:34 -03:00
Bruno Oliveira 095a195d71 Improve docs about junit_family warning message
From discussion in #6178
2020-05-02 13:01:15 -03:00
Bruno Oliveira 0b78983197
Merge pull request #7141 from bluetech/rm-nonzero 2020-05-02 10:00:35 -03:00
Bruno Oliveira e3643751bc
Merge pull request #7144 from nicoddemus/async-testcase-7110 2020-05-01 16:44:10 -03:00
Bruno Oliveira fd2f172258 Let unittest frameworks deal with async functions
Instead of trying to handle unittest-async functions in pytest_pyfunc_call,
let the unittest framework handle them instead.

This lets us remove the hack in pytest_pyfunc_call, with the upside that
we should support any unittest-async based framework.

Also included 'asynctest' as test dependency for py37-twisted, and renamed
'twisted' to 'unittestextras' to better reflect that we install 'twisted' and
'asynctest' now.

This also fixes the problem of cleanUp functions not being properly called
for async functions.

Fix #7110
Fix #6924
2020-05-01 15:10:28 -03:00
Bruno Oliveira 7f5978c34c Allow File.from_parent to forward custom parameters to the constructor 2020-05-01 11:00:52 -03:00
Ran Benita a718ad6363 Stop using Python's eval() for -m and -k
Previously, the expressions given to the `-m` and `-k` options were
evaluated with `eval`. This causes a few issues:

- Python keywords cannot be used.

- Constants like numbers, None, True, False are not handled correctly.

- Various syntax like numeric operators and `X if Y else Z` is supported
  unintentionally.

- `eval()` is somewhat dangerous for arbitrary input.

- Can fail in many ways so requires `except Exception`.

The format we want to support is quite simple, so change to a custom
parser. This fixes the issues above, and gives us full control of the
format, so can be documented comprehensively and even be extended in the
future if we wish.
2020-05-01 12:59:06 +03:00
Ran Benita 409ffcef17 Remove a couple Python 2 __nonzero__ definitions
It's called __bool__ in Python 3.
2020-05-01 11:41:21 +03:00
Ran Benita d8558e87c5 terminalwriter: clean up markup function a bit 2020-04-30 16:44:03 +03:00
Ran Benita 414a87a53f config/argparsing: use our own get_terminal_width() 2020-04-30 16:44:03 +03:00
Ran Benita d5584c7207 terminalwriter: compute width_of_current_line lazily
Currently this property is computed eagerly, which means
get_line_width() is computed on everything written, but that is a slow
function.

Compute it lazily, so that get_line_width() only runs when needed.
2020-04-30 16:44:03 +03:00
Ran Benita dd32c72ff0 terminalwriter: remove unused property chars_on_current_line 2020-04-30 16:44:03 +03:00
Ran Benita 1bc4170e63 terminalwriter: don't flush implicitly; add explicit flushes
Flushing on every write is somewhat expensive.

Rely on line buffering instead (if line buffering for stdout is
disabled, there must be some reason...), and add explicit flushes when
not outputting lines.

This is how regular `print()` e.g. work so should be familiar.
2020-04-30 16:44:03 +03:00
Ran Benita d9b43647b7 terminalwriter: inline function _update_chars_on_current_line 2020-04-30 16:44:03 +03:00
Ran Benita 8e04d35a33 terminalwriter: remove unneeded hasattr use 2020-04-30 16:44:03 +03:00
Ran Benita e8fc5f99fa terminalwriter: add type annotations 2020-04-30 16:44:03 +03:00
Ran Benita f6564a548a terminalwriter: remove win32 specific code in favor of relying on colorama
On Windows we already depend on colorama, which takes care of all of
this custom code on its own.
2020-04-30 16:44:03 +03:00
Ran Benita 8d2d1c40f8 terminalwriter: inline function _escaped
Doesn't add much.
2020-04-30 16:44:03 +03:00
Ran Benita 66ee755649 terminalwriter: remove TerminalWriter's stringio argument
Had a mark indicating it should be removed, and I agree, it's better to
just use the `file` argument.
2020-04-30 16:44:02 +03:00
Ran Benita 94a57d2353 io: combine _io.TerminalWriter and _io.terminalwriter.TerminalWriter
Previously it extended an external type but now it come move to the type
itself.
2020-04-30 16:44:02 +03:00
Ran Benita dac05ccd9a terminalwriter: remove support for passing callable as file in TerminalWriter
Not used.
2020-04-30 16:44:02 +03:00
Ran Benita 0528307ebf terminalwriter: remove unused function TerminalWriter.reline 2020-04-30 16:44:02 +03:00
Ran Benita a6819726cd terminalwriter: remove unused function ansi_print 2020-04-30 16:44:02 +03:00
Ran Benita b6cc90e0af terminalwriter: remove support for writing bytes directly
It is not used and slows things down.
2020-04-30 16:44:02 +03:00
Ran Benita 9a59970cad terminalwriter: optimize get_line_width() a bit
This function is called a lot when printing a lot of text, and is very
slow -- this speeds it up a bit.
2020-04-30 16:44:02 +03:00
Ran Benita 6c1b6a09b8 terminalwriter: simplify get_terminal_width()
The shutil.get_terminal_size() handles everything this did already.
2020-04-30 16:44:02 +03:00
Ran Benita c749e44efc terminalwriter: remove custom win32 screen width code
Python 3 does this on its own so we can use the shared code:
bcf2b59fb5
2020-04-30 16:44:02 +03:00
Ran Benita 1d596b27a7 terminalwriter: move Win32ConsoleWriter definition under win32 conditional
This way non-Windows platforms skip it. It also uses things defined
inside the `if`.
2020-04-30 16:44:02 +03:00
Ran Benita 5e2d820308 terminalwriter: fix lints 2020-04-30 16:44:02 +03:00
Ran Benita 3014d9a3f7 terminalwriter: auto-format 2020-04-30 16:43:55 +03:00
Ran Benita 276405a039 terminalwriter: vendor TerminalWriter from py
Straight copy from py 1.8.1. Doesn't pass linting yet.
2020-04-30 16:43:54 +03:00
Ran Benita c72a1b2933 config: replace usage of py.io.dupfile
As part of the effort to remove uses of `py`.
2020-04-29 18:04:37 +03:00
Ran Benita e5d6e58625
Merge pull request #7130 from bluetech/cacheprovider-nf-set
cacheprovider: speed up NFPlugin when --nf is not enabled
2020-04-29 17:28:20 +03:00
Ran Benita 4a42afdc2f cacheprovider: speed up NFPlugin when --nf is not enabled
The code used an O(n^2) loop. Replace list with set to make it O(n).

For backward compatibility the filesystem cache still remains a list.

On this test:

    import pytest
    @pytest.mark.parametrize("x", range(5000))
    def test_foo(x): pass

run with `pytest --collect-only`:

Before: 0m1.251s
After:  0m0.921s
2020-04-29 13:02:38 +03:00
Ran Benita 9bf3efb8e9
Merge pull request #7120 from bluetech/update-dev
pre-commit: update some tool revisions
2020-04-25 23:07:15 +03:00
Ran Benita 3cd97d50f9 pre-commit: update pyupgrade 1.18.0 -> 2.2.1 2020-04-24 21:57:38 +03:00
Ran Benita 38c9d59ddc pre-commit: update blacken-docs 1.0.0 -> 1.6.0 2020-04-24 21:53:37 +03:00
Ran Benita 289e6c1d36
Update src/_pytest/deprecated.py
Co-Authored-By: Ronny Pfannschmidt <opensource@ronnypfannschmidt.de>
2020-04-24 21:37:08 +03:00
Ran Benita 907e29a47b fixtures: deprecate pytest._fillfuncargs function
This function is exposed and kept alive for the oejskit plugin which is
abandoned and no longer works with recent plugins, so let's prepare to
completely remove it.
2020-04-24 18:54:46 +03:00
Ran Benita 7d5f5a8785
Merge pull request #7050 from blueyed/fix-doc
doc: internal: remove references to old "newinterpret" module
2020-04-17 17:11:31 +03:00
Ran Benita 731c340be6
Merge pull request #7058 from blueyed/doc-store
doc: minor fixes for Store
2020-04-17 17:10:35 +03:00
Katarzyna Król 7789b51acb
Issue 4677 - always relative path in skip report (#6953) 2020-04-17 08:28:36 +03:00
Simon K de6c28ed1f
Improve error handling around yieldctx fixtures which do not yield a value (#7083) 2020-04-15 12:17:13 +03:00
Ran Benita 9ddf38369d
Merge pull request #7069 from bluetech/faulthandler-doc-windows
Remove note saying faulthandler_timeout is not available on Windows
2020-04-14 10:24:28 +03:00
Simon K f479cbce10
Document pytester fixtures so --fixtures makes more sense (#7080) 2020-04-13 20:58:50 +03:00
Ran Benita c08cff3770
Merge pull request #7082 from symonk/4583-better-ux-with-eval-fails
Gracefully handle eval() failure(s) for marker expressions
2020-04-13 17:03:37 +03:00
symonk 251e8f212e refactor mark tests, widen catching and make error msg more concise 2020-04-13 14:25:01 +01:00
symonk 87edc09dea Gracefully handle eval() failure(s) for marker expressions 2020-04-13 13:25:06 +01:00
Andreas Maier c9386ada29 Squash: Resolved 2nd round of review comments 2020-04-12 13:19:45 +02:00
Andreas Maier b2582b0314 Squash: Applied review comments 2020-04-12 12:12:00 +02:00
Andreas Maier 869c089887 Fixes #7077: Added & improved docs for repr_failure() in Node & Collector 2020-04-12 09:28:40 +02:00
Ran Benita b553ce54c8 Remove note saying faulthandler_timeout is not available on Windows
I think it is available in all Python versions we support, but at least
since Python 3.7 the docs[0] say:

    Changed in version 3.7: This function is now always available.

so let's just remove the note.

[0] https://docs.python.org/3/library/faulthandler.html#faulthandler.dump_traceback_later
2020-04-10 14:46:15 +03:00
Ronny Pfannschmidt 244c8e4a13
Merge pull request #6680 from RonnyPfannschmidt/fix-6294-more-docs-for-fromparent
doc: more docs for from_parent
2020-04-10 13:10:40 +02:00
Ronny Pfannschmidt 19c243f0fa
Merge pull request #6285 from earonesty/patch-1
Add _pytest.fixtures.FixtureLookupError to top level import
2020-04-10 07:58:58 +02:00
Ronny Pfannschmidt c5d636a1b2
Merge pull request #6992 from pytest-dev/revert-6767-tmpdir-cleanup-upstream
Revert "tmpdir: clean up indirection via config for factories"
2020-04-10 07:49:52 +02:00
Ronny Pfannschmidt 1eb2b45db5 Revert "tmpdir: clean up indirection via config for factories (#6767)"
This reverts commit 8a1633c3b4.

+ add changelog
2020-04-10 07:18:54 +02:00
Daniel Hahler 0b5d2ff526 Fix/improve printing of docs for collected items 2020-04-09 21:49:30 +02:00
Daniel Hahler accea46fa1 doc: minor fixes for Store 2020-04-09 17:08:47 +02:00
Daniel Hahler 413ca8a4d0
faulthandler: trylast=True (#7025)
It should happen as late as possible before the test runs.

Ref: https://github.com/pytest-dev/pytest/issues/7022
2020-04-09 08:53:35 +02:00
Daniel Hahler a98a62723e revisit 2020-04-09 01:57:46 +02:00
Daniel Hahler ad4c1071d9 doc: internal: remove references to old "newinterpret" module
This has been merged into the (only) assertrewrite mode.
2020-04-09 01:40:23 +02:00
Ran Benita 4344c61731 Slightly improve Mark and MarkDecorator documentation
Mostly I wanted to get rid of mentions of "MarkItem" which is something
that no longer exists, but I improved a little beyond that and annotated
some simple types.
2020-04-07 22:14:47 +03:00
Daniel Hahler ce806001b0
Fix doc for `numbered` arg with `TempPathFactory.mktemp` (#7014) 2020-04-07 11:44:54 +02:00
Ran Benita 259e5d0610
Merge pull request #7034 from bluetech/fix-testdurations-flaky
Fix flaky TestDurations test
2020-04-07 09:39:02 +03:00
Daniel Hahler 1fd14685c5
doc: document inversed lines with terminal report hooks (#7016)
It was surprising that `tryfirst=True` would not result in lines being
added to the beginning with `pytest_report_header`.
This is due to lines being reversed, and therefore the same applies to
`pytest_report_collectionfinish`.
2020-04-07 08:08:28 +02:00
Ran Benita c3e6e2e8c8 Fix flaky TestDurations test
TestDurations tests the `--durations=N` functionality which reports N
slowest tests, with durations <= 0.005s not shown by default.

The test relies on real time.sleep() (in addition to the code which uses
time.perf_counter()) which makes it flaky and inconsistent between
platforms.

Instead of trying to tweak it more, make it use fake time instead. The
way it is done is a little hacky but seems to work.
2020-04-06 23:49:46 +03:00
Ran Benita 1ce30fd38f Document the pytest_report_teststatus hook better and test uncovered functionality
This hook has some functionality to provide explicit markup for the test
status. It seemed unused and wasn't tested, so I was tempted to remove
it, but I found that the pytest-rerunfailures plugin uses it, so
document it and add a test instead.
2020-04-04 17:27:59 +03:00
Ran Benita 4a324ce920 Remove unused defaultfuncargprefixmarker
Unused since 1e80a9cb34.
2020-04-04 14:34:40 +03:00
Daniel Hahler 20f6331afd
Fix TerminalRepr instances to be hashable (#6988)
pytest-xdist assumes `ExceptionChainRepr` is hashable.

Fixes https://github.com/pytest-dev/pytest/issues/6925.
Fixes https://github.com/pytest-dev/pytest-xdist/issues/515.
2020-04-03 00:56:53 +02:00
Ran Benita 03451c397f Simplify positional arguments compatibility code in pytest.fixture()
The dynamic scope feature added in 10bf6aac76
necessitated some wrangling of arguments in pytest.fixture(). In
particular, it deprecated positional arguments in favor of keyword-only
arguments, while keeping backward compatibility.

The way it did this avoided some code duplication but ended up being
quite hard to follow and to annotate with types.

Replace it with some straightforward code, which is not very DRY but is
simple and easy to remove when the time comes.
2020-04-02 14:52:58 +03:00
Daniel Hahler 354602abe6
Update src/_pytest/hookspec.py
Co-Authored-By: Bruno Oliveira <nicoddemus@gmail.com>
2020-04-02 12:01:43 +02:00
Daniel Hahler 607f7603af doc: pytest_collection: has to set `session.items`
Would make sense to use its return value etc, but this helps for now.
2020-03-31 09:39:50 +02:00
Ronny Pfannschmidt 2d9dac95ec
Merge pull request #6927 from RonnyPfannschmidt/fix-6924-run-async-stdlib-unittests
running stdlib asyncio unittests again
2020-03-30 22:34:17 +02:00
Ronny Pfannschmidt 451aef65ac prepare tests and disable warnings for asyncio unittest cases
shoehorn unittest async results into python test result interpretation

changelog
2020-03-30 21:55:12 +02:00
Ronny Pfannschmidt f1d51ba1f5 deprecate the pytest.collect module
changelog

minimal unittest for collect module deprecations

\!fixup - changelog typo
2020-03-30 21:53:07 +02:00
Ronny Pfannschmidt eab2831671 fix #6951: allow to write TerminalReporter.writer 2020-03-30 21:31:53 +02:00
smarie 95fadd5740
Improved time counter used to compute test durations. (#6939)
Co-authored-by: Sylvain MARIE <sylvain.marie@se.com>
Co-authored-by: Ran Benita <ran@unusedvar.com>
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
2020-03-29 15:20:09 +03:00
Ran Benita a785754523 Change EnvironmentError, IOError to OSError - they are aliases
Since Python 3.3, these are aliases for OSError:
https://docs.python.org/3/whatsnew/3.3.html#pep-3151-reworking-the-os-and-io-exception-hierarchy
2020-03-27 18:40:23 +03:00
Ran Benita 83e18776f6
Merge pull request #6818 from blueyed/fix-assert-exc-upstream
assertion: rewrite: only catch EnvironmentError
2020-03-27 18:33:17 +03:00
Bruno Oliveira 429a28eca0
Merge pull request #6782 from nicoddemus/code-highlight-followup
Assorted improvements following up #6658
2020-03-27 08:58:50 -03:00
Daniel Hahler aae0579bcd
doc: use `envvar` directive for environment variables (#6874)
This changes the link anchors in "reference.html", from e.g.
`reference.html#pytest-current-test` to
`reference.html#envvar-PYTEST_CURRENT_TEST`, but I think that is OK, and
not worth adding labels for the old anchors.
2020-03-27 02:40:25 +01:00
Daniel Hahler 327ec54248
Doc fixes (#6861) 2020-03-27 02:22:03 +01:00
Ran Benita 817537523c Upgrade mypy 0.761 -> 0.770
https://mypy-lang.blogspot.com/2020/03/mypy-0770-released.html
2020-03-21 17:16:30 +02:00
Ran Benita ded30238f7
Merge pull request #6926 from bluetech/fix-capsysbin-print
Fix crash when printing while capsysbinary is active
2020-03-17 11:21:19 +02:00
Daniel Hahler 1fda861190 Fix crash when printing while capsysbinary is active
Previously, writing to sys.stdout/stderr in text-mode (e.g.
`print('foo')`) while a `capsysbinary` fixture is active, would crash
with:

    /usr/lib/python3.7/contextlib.py:119: in __exit__
        next(self.gen)
    E   TypeError: write() argument must be str, not bytes

This is due to some confusion in the types. The relevant functions are
`snap()` and `writeorg()`. The function `snap()` returns what was
captured, and the return type should be `bytes` for the binary captures
and `str` for the regular ones. The `snap()` return value is eventually
passed to `writeorg()` to be written to the original file, so it's input
type should correspond to `snap()`. But this was incorrect for
`SysCaptureBinary`, which handled it like `str`.

To fix this, be explicit in the `snap()` and `writeorg()`
implementations, also of the other Capture types.

We can't add type annotations yet, because the current inheritance
scheme breaks Liskov Substitution and mypy would complain. To be
refactored later.

Fixes: https://github.com/pytest-dev/pytest/issues/6871
Co-authored-by: Ran Benita (some modifications & commit message)
2020-03-16 18:21:33 +02:00
Ran Benita c26bbdfaf0
Merge pull request #6854 from gdhameeja/Fix-6822
Check invalid operations for -k
2020-03-15 09:53:36 +02:00
Ran Benita 29e4cb5d45 Remove safe_text_dupfile() and simplify EncodedFile
I tried to understand what the `safe_text_dupfile()` function and
`EncodedFile` class do. Outside tests, `EncodedFile` is only used by
`safe_text_dupfile`, and `safe_text_dupfile` is only used by
`FDCaptureBinary.__init__()`. I then started to eliminate always-true
conditions based on the single call site, and in the end nothing was
left except of a couple workarounds that are still needed.
2020-03-14 12:57:08 +02:00
Bruno Oliveira 010e711971 Merge pull request #6914 from nicoddemus/revert-6330
Revert "[parametrize] enforce explicit argnames declaration (#6330)"
2020-03-13 10:49:48 -03:00
Bruno Oliveira 2eedf9c2aa
Merge pull request #6910 from nicoddemus/resultlog-logreport
Handle unknown stats in pytest_report_teststatus hook
2020-03-13 09:30:04 -03:00
Bruno Oliveira 0c58ed2cc0 Handle unknown stats in pytest_report_teststatus hook
Noticed that the pytest_report_teststatus of reportlog was not properly
handling unknown statuses while taking a look at:

https://github.com/pytest-dev/pytest-rerunfailures/issues/103
2020-03-12 20:52:22 -03:00
Ran Benita 6954b3b0dc Assume os.dup is always available
The commit which added the checks for os.dup a15afb5e48
suggests it was done for Jython. But pytest doesn't support Jython
anymore (Jython is Python 2 only).

Furthermore, it looks like the faulthandler plugin (bundled in pytest
and enabled by default) uses os.dup() unprotected and there have not
been any complaints.

So seems better to just remove these checks, and only add if someone
with a legitimate use case complains.
2020-03-12 16:47:15 +02:00
Ran Benita 4f8fff9cab
Merge pull request #6868 from bluetech/simplify-exc
Simplify some exception handling code
2020-03-10 20:32:08 +02:00
Daniel Hahler ac7ebfa22e
doc: internal: fix `MultiCapture.readouterr` (#6878)
Remove wrong docstring: it might actually return bytes.
Replace it with a type annotation which is clear enough.
2020-03-08 12:38:21 +01:00
Daniel Hahler db92cea14c
black: --target-version py35 (#6872) 2020-03-08 01:24:31 +01:00
Daniel Hahler bce1d40fb0
tests: harden test_reprcompare_notin, factor out callop (#6764)
* tests: assertion: factor out `callop`, typing

* tests: harden test_reprcompare_notin
2020-03-08 01:23:19 +01:00
Daniel Hahler dc86fb6758 pre-commit run --all-files 2020-03-07 23:47:55 +01:00
gdhameeja 599bf075db Check invalid operations for -k
`KeywordMapping` returns a bool on lookup which when passed to eval
fail on certain operations such as index access and attribute access.
We catch all exceptions and raise a `UsageError`.
2020-03-07 20:17:01 +05:30
Ran Benita 978b315861 Simplify some exception handling code
Mostly avoid unnecessary usage of sys.exc_info(). Since Python3,
exception objects themselves have all that's needed. They are also
easier to type.
2020-03-07 13:18:54 +02:00
Ran Benita 580edc13e7
Merge pull request #6867 from bluetech/inline-pyobjmixin
Inline PyobjContext
2020-03-07 11:58:38 +02:00
Ran Benita b7f2e3d4f5 Inline PyobjContext
When it was introduced in 8adac2878f it
seems to have had some use, but now it doesn't.
2020-03-06 14:55:20 +02:00
Ran Benita f011bc642c Store mark's evalcache in config's store instead of attribute
Part of moving away from ad-hoc attributes to using the config's store.
2020-03-06 11:42:20 +02:00
Ran Benita b1d7a187f2 Add setdefault() method to Store
Can be useful in some cases.
2020-03-06 11:42:20 +02:00
Ran Benita 678d65f051 Store AssertionState in config's store instead of attribute
Part of moving away from ad-hoc attributes to using the config's store.
2020-03-06 11:27:01 +02:00
Daniel Hahler fcd3fad03d
Minor internal improvements to logging's log_level (#6849) 2020-03-06 03:12:57 +01:00
Bruno Oliveira 19bb2c6235
Merge pull request #6859 from nicoddemus/kw-mapping-attr-typing
Use attrs in KeywordMapping
2020-03-05 08:47:22 -03:00
Bruno Oliveira bc8e52c3c2 Use attrs in KeywordMapping
Also added type hinting.
2020-03-05 08:33:20 -03:00
Daniel Hahler d9a4626944 fixup! Fix documentation for Config/InvocationParams 2020-03-05 06:38:44 +01:00
Daniel Hahler ffa2658971 doc: reports: count_towards_summary: is a property
And therefore does not really `return`.  It confused me that there was
no `source` link in the docs, which is only there for functions.
2020-03-05 05:57:43 +01:00
Daniel Hahler c39a85e5f4 doc: revisit UnformattedWarning 2020-03-05 03:15:14 +01:00
Daniel Hahler b90f57d25c Remove wrong/outdated doc with UnformattedWarning
It was introduced in da6830f19 (added to `_pytest.deprecated`, but then
moved to `_pytest.warning_types`).
2020-03-05 03:13:28 +01:00
Daniel Hahler 77adb33ec6 doc: use show-inheritance with warnings, revisit docstrings
Revisits the docstring for `PytestExperimentalApiWarning` and
`PytestUnhandledCoroutineWarning`.
2020-03-05 02:56:18 +01:00
Daniel Hahler 1a8d427e98 doc: src/_pytest/deprecated.py: links 2020-03-05 02:47:21 +01:00
Kyle Altendorf dce2621710
Add pytest-twisted to list of async def handling plugins
https://github.com/pytest-dev/pytest-twisted/tree/v1.12#ensuredeferred

[pytest-twisted](https://github.com/pytest-dev/pytest-twisted) supports `async def` test functions and fixtures as well as `async def`/`yield` fixtures.
2020-03-04 17:42:08 -05:00
Daniel Hahler acec0b688f
Fix usage of pytester with doctests (#6802)
Use `request.node.name` instead of `request.function.__name__`:
`request.function` is `None` with `DoctestItem`s.
2020-03-04 05:33:50 +01:00
Daniel Hahler 37d074efc8
Fix (internal) documentation for pytester's LineComp (#6850) 2020-03-04 05:25:32 +01:00
Ronny Pfannschmidt 9fd71d6fe0
fix #571: deprecate pytest_collect_directory as ... (#6847)
Deprecate pytest_collect_directory

Fix #571

Co-authored-by: Daniel Hahler <github@thequod.de>
2020-03-03 18:58:14 -03:00
Daniel Hahler b11bfa106c
Use attrs with all Repr classes (#6739)
Co-authored-by: Ran Benita <ran234@gmail.com>
2020-03-03 21:53:28 +01:00
Daniel Hahler dc5219a9c0 Fix documentation for Config/InvocationParams 2020-03-03 21:15:06 +01:00
Daniel Hahler a42e85ed54 Fix documentation for _pytest.pytester.RunResult
When using `(i)var` in the class docstring it would link `duration` to
`_pytest.runner.TestReport.duration`.

This moves the docstrings to the attributes properly.
2020-03-03 18:12:12 +01:00
Bruno Oliveira aac11e5e75 Mention FixtureLookupError in getfixturevalue docs 2020-03-03 10:05:19 -03:00
earonesty b2d54fe6b1 Fix tox alpha order 2020-03-03 09:59:38 -03:00
earonesty 9b8ed8d9ad Update pytest.py 2020-03-03 09:59:38 -03:00
earonesty b40a9f9add Export FixtureLookupError to top level 2020-03-03 09:59:38 -03:00
Daniel Hahler ddc8edffbc logging: minor typing, rename function
`get_log_level_for_setting` reflects better what the function does, and
it does not require a (trivial) docstring.
2020-03-01 22:02:46 +01:00
Daniel Hahler bdd22fdd52 logging: improve help for log_level 2020-03-01 22:02:46 +01:00
Ronny Pfannschmidt 3d390940d1 refer the node-from-parent deprecation documentation in the warning
fixup: fix test for warning
2020-03-01 20:33:46 +01:00
Ronny Pfannschmidt 194b52145b
Merge pull request #6834 from RonnyPfannschmidt/fix-6833-summarize-warning-item-locations
summarize warning summaries if the number of locations is high
2020-03-01 20:21:36 +01:00
Ronny Pfannschmidt 23c43a37e0 summarize warning summaries if the number of locations is high 2020-03-01 14:08:15 +01:00
Bruno Oliveira 92767fec51
Merge pull request #6836 from bluetech/store
Add a typing-compatible mechanism for ad-hoc attributes on various objects
2020-03-01 09:30:10 -03:00
Ran Benita 2f60548e08 Mark AssertionRewritingHook as implementing importlib.abc.Loader
It implements the required methods. This is useful for typing but also a
nice indicator.
2020-02-29 16:55:56 +02:00
Daniel Hahler f10ab021e2
Move _collectfile to FSCollector (#6830)
Previously this was implemented both on `Session` and `Package`, where
the extra code in `Package._collectfile` was not covered/used.

Ref: https://github.com/pytest-dev/pytest/pull/6830#issuecomment-592663236
2020-02-29 11:40:11 +01:00
Daniel Hahler d99bfc18b8
Update src/_pytest/config/findpaths.py
Co-Authored-By: Ran Benita <ran234@gmail.com>
2020-02-29 11:39:33 +01:00
Bruno Oliveira ff7b5dbbde
Merge pull request #6784 from nicoddemus/deprecate-terminal-writer
Deprecate TerminalReporter.writer
2020-02-28 21:07:02 -03:00
Daniel Hahler 6092d3c6e1 typing: parseconfig 2020-02-28 20:40:39 +01:00
Daniel Hahler d2f9a73a29 typing: get_dirs_from_args 2020-02-28 19:58:15 +01:00
Daniel Hahler 5819536f00
Improve UsageError with invalid `-o` style (#6795)
This started from fixing the test, where `"xdist_strict True"` was used
as a single argument, although you typically would see `["xdist_strict",
"True"]`.

Improves the error message to mention the option that caused the error.
2020-02-28 19:41:56 +01:00
Ran Benita d636fcd557 Add a typing-compatible mechanism for ad-hoc attributes on various objects
pytest has several instances where plugins set their own attributes on
objects they receive in hooks, like nodes and config. Since plugins are
detached from these object's definition by design, this causes a problem
for type checking because these attributes are not defined and mypy
complains.

Fix this by giving these objects a "store" which can be used by plugins
in a type-safe manner.

Currently this mechanism is private. We can consider exposing it at a
later point.
2020-02-28 14:34:44 +02:00
Daniel Hahler 16c683dff9
Remove unused `setns` (#6804)
Last usage of it was removed in ba17363d7.
2020-02-26 15:35:14 +01:00
Daniel Hahler eac933acde assertion: rewrite: only catch EnvironmentError
This was changed unintentionally in 45c4a8fb3 (pytest 5.3.0), but only
EnvironmentErrors might have `errno`.
Since that is not really guaranteed and it is good to have more
information this uses the string representation of the exc in the trace
message.
2020-02-25 17:29:30 +01:00
Daniel Hahler ac3a42bafd
doctest: strip newlines with unexpected exceptions (#6801) 2020-02-24 15:18:08 +01:00
Daniel Hahler be23aeb989
Remove unused `pytester.getdecoded` (#6805)
Last usage was removed in 22dc47d9f.
2020-02-24 14:33:26 +01:00
Daniel Hahler 9928c7794b
minor: fix trace msg: s/loaded/loading (#6798) 2020-02-23 17:12:30 +01:00
Daniel Hahler 706ea86bba
capture: factor out _get_multicapture (#6788)
Ref: https://github.com/pytest-dev/pytest/pull/6671#issuecomment-588408992
2020-02-22 23:39:20 +01:00
Daniel Hahler 1d5a0ef284
Fix pytest_ignore_collect hooks: do not return False (#6778)
It should only return `True` when something is to be ignored, not
`False` otherwise typically.

This caused e.g. bad interaction with the cacheprovider (before
https://github.com/pytest-dev/pytest/pull/6448).
2020-02-22 23:35:47 +01:00
Daniel Hahler c8b4a1a471
Handle `match` with `pytest.raises()` (#6753)
Fixes https://github.com/pytest-dev/pytest/issues/6752.
2020-02-22 23:32:55 +01:00
Daniel Hahler 68fe0eb8f3
Minor cosmetic fixes (#6792) 2020-02-22 19:24:41 +01:00
Daniel Hahler de854c6ee1
doc: `pytest.raises`: remove mention of removed `message` (#6793)
Follow-up to 13f7f27fd.
2020-02-22 19:24:00 +01:00
Daniel Hahler 04126feea7
doc: `pytest.raises`: link to `python:try` (#6794) 2020-02-22 19:23:35 +01:00
Daniel Hahler 478a244f5e
main: `args` must be a list, not tuple (#6791)
Passing in a tuple crashes in `_prepareconfig`:

        def test_invoke_with_tuple(self):
    >       pytest.main(("-h",))

    src/_pytest/config/__init__.py:82: in main
        config = _prepareconfig(args, plugins)
    src/_pytest/config/__init__.py:229: in _prepareconfig
        return pluginmanager.hook.pytest_cmdline_parse(
    …
    src/_pytest/helpconfig.py:98: in pytest_cmdline_parse
        config = outcome.get_result()  # type: Config
    src/_pytest/config/__init__.py:808: in pytest_cmdline_parse
        self.parse(args)
    src/_pytest/config/__init__.py:1017: in parse
        self._preparse(args, addopts=addopts)
    _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

        def _preparse(self, args: List[str], addopts: bool = True) -> None:
            …
            if addopts:
                ini_addopts = self.getini("addopts")
                if ini_addopts:
    >               args[:] = self._validate_args(ini_addopts, "via addopts config") + args
    E               TypeError: can only concatenate list (not "tuple") to list

    addopts    = True
    args       = ('-h',)
    env_addopts = ''
    ini_addopts = ['-rfEX', …]

    src/_pytest/config/__init__.py:956: TypeError: can only concatenate list (not "tuple") to list

Might be worth handling (converting it to a list for example), but it
was documented to be a list to begin with when removing support for
strings (a7e401656).
2020-02-22 13:16:46 +01:00