summaryrefslogtreecommitdiff
path: root/tests
Commit message (Collapse)AuthorAgeFilesLines
...
* test: remove some obsolete HTML scrubsNed Batchelder2021-10-171-5/+0
|
* chore: updated HTML gold filesNed Batchelder2021-10-1738-383/+626
|
* test: don't show these functions when HTML gold files failNed Batchelder2021-10-172-0/+3
|
* test: html scrubs need to account for pytest dir namingNed Batchelder2021-10-171-0/+1
|
* fix(html): adjust some colorsNed Batchelder2021-10-171-8/+8
| | | | | | | | - make the yellow border more visible - make the context-blue less vibrant - all colors are lower-case
* feat(html): sticky header on pyfileNed Batchelder2021-10-171-5/+15
|
* refactor(html): use header/main/footer html5 elementsNed Batchelder2021-10-171-8/+8
|
* test: fix the tests/gold/html/Makefile to copy actual outputNed Batchelder2021-10-171-4/+3
|
* test: improve this one html testNed Batchelder2021-10-171-3/+3
|
* feat: `coverage run` now sets the COVERAGE_RUN environment variableNed Batchelder2021-10-161-1/+16
|
* test: very minor clean upNed Batchelder2021-10-161-1/+4
|
* test: subprocess diagnostic for when we need itNed Batchelder2021-10-161-0/+6
|
* test: we don't need two sys.path restorersNed Batchelder2021-10-143-15/+6
|
* test: parametrize test_pathologically_long_code_objectNed Batchelder2021-10-141-21/+18
| | | | | While experimenting with packing branch ends into one int, this test showed a mistake I had made. Turns out it wasn't an obsolete test after all.
* test: don't use subprocesses where we don't have toNed Batchelder2021-10-133-56/+74
| | | | | | When running tests with `-n 0`, this shaves 4 seconds off the run time. When running the complete test suite with xdist, it takes 2 seconds longer!?
* refactor: use the get_report helperNed Batchelder2021-10-131-6/+1
|
* refactor: no longer need to explicitly omit site-packagesNed Batchelder2021-10-131-15/+0
| | | | 6.x omits site-packages automatically.
* refactor: convert existing imports to start_stop_importNed Batchelder2021-10-131-15/+6
|
* test: we really don't want to hear about imp being deprecatedNed Batchelder2021-10-121-0/+5
|
* test: add a test that self.stdout() works the way it says.Ned Batchelder2021-10-121-0/+9
|
* fix: use human sorting on human-readable thingsNed Batchelder2021-10-122-8/+39
|
* test: lightly test the ast_dump functionNed Batchelder2021-10-121-2/+24
|
* fix: find_writable_pth_directory must expect failure from open()Michał Górny2021-10-121-4/+4
| | | | | | | Fix find_writable_pth_directory() to expect the OSError from the open() call. In general, this is what's going to happen if the file is not writable. The current logic could only be triggered e.g. if the file was writable but the underlying filesystem run out of space.
* fix: source modules need to be re-imported. #1232Ned Batchelder2021-10-112-13/+11
|
* test: add more tests of run_python_fileNed Batchelder2021-10-112-5/+93
| | | | | | The tests in test_process run the exception handling in execfile.py, but only under coverage, so metacov can't see it. These smaller tests exercise the code without coverage on top.
* test: add a test of hash-based pyc filesNed Batchelder2021-10-111-2/+9
|
* fix: make third-party detection work with namespace packages. #1231Ned Batchelder2021-10-101-3/+74
|
* refactor: remove some left over test printsNed Batchelder2021-10-101-3/+0
|
* test: cover the last edge cases in sqldata.pyNed Batchelder2021-10-101-2/+12
|
* fix: raise CoverageException for SQLite connection errorsnedbat/sqldata-edge-testsNed Batchelder2021-10-101-0/+8
|
* refactor: simplify some strange string formattingNed Batchelder2021-10-102-5/+5
|
* test: test two add_file_tracer possibilitiesNed Batchelder2021-10-101-0/+22
|
* refactor: no need for maybe-u prefixes in test regexesNed Batchelder2021-10-104-11/+11
| | | | That was for Python 2, which we don't support anymore.
* test: run test_data.py with debugging onNed Batchelder2021-10-101-96/+110
|
* refactor: remove a mixin class from testsNed Batchelder2021-10-101-67/+64
|
* style: prefer explicit string concatenationNed Batchelder2021-10-1011-106/+100
|
* refactor: remove a Python 2 thingNed Batchelder2021-10-101-6/+1
|
* style: use the official designation for utf-8Ned Batchelder2021-10-104-4/+4
| | | | | Yes, this is completely unimportant. Don't ask me why I bothered, I'm not really sure.
* fix: contexts_by_lineno now returns a true dictNed Batchelder2021-10-091-3/+24
|
* test: add a test of touching files in an empty CoverageDataNed Batchelder2021-10-091-0/+6
|
* refactor: don't need this printNed Batchelder2021-10-091-1/+0
|
* test: set_query_contexts takes regex, make the examples look like regexNed Batchelder2021-10-091-6/+6
|
* test: this xdist defensiveness now seems unneededNed Batchelder2021-10-081-26/+0
|
* test: make metacov work with xdistNed Batchelder2021-10-081-1/+1
| | | | | | | | | I was using pytest.__file__ as a directory, but in 5.3.2, pytest changed to be a package, so that wasn't the site-packages directory anymore, and our .pth file was written someplace useless. Now we don't rely on the structure of pytest (why did we ever?), and it works again.
* test: canonicalize this bit of debug outputNed Batchelder2021-10-081-1/+1
|
* build: make "clean" targets be quietNed Batchelder2021-10-081-1/+1
|
* docs: note #1210 in the changelogNed Batchelder2021-10-061-1/+3
|
* Fix an incompatibility with pyarmorglacials2021-10-061-0/+6
|
* fix: pretend we didn't import third-party packages we use. #1228Ned Batchelder2021-10-061-1/+19
| | | | | | tomli couldn't use coverage themselves because we imported it early. Cleaning sys.modules means their own imports will actually execute after coverage has started, so their files will be properly measured.
* test: fix flaky multiprocessing testsNed Batchelder2021-09-291-1/+1
| | | | | Tests are failing because we expect to see three different pids, but only get two. Is that because the work is being completed too quickly?
View Lorry Controller Status