Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | fix: don't warn that dynamic plugins already imported their source files. #1150 | Ned Batchelder | 2021-05-01 | 4 | -2/+46 |
| | |||||
* | build: no more Windows 2.7? | Ned Batchelder | 2021-05-01 | 2 | -0/+6 |
| | | | | | Microsoft removed the vcpython27 code (because 2.7 isn't supported anymore). Discussion here: https://community.chocolatey.org/packages/vcpython27 | ||||
* | docs: fix code comment formatting (#1153) | Mayank Singhal | 2021-04-30 | 3 | -10/+10 |
| | | | | | | | | | | | | | | | | | | | | | | | | * docs(branch.rst): Line number comments not needed The topic `Branch Coverage Management` in this file already has a setting: :linenothreshold: 5 Using this setting, sphinx will automatically provide line numbers for code block longer than 5 lines. reference: https://www.sphinx-doc.org/en/master/usage/restructuredtext/directives.html#directive-option-highlight-linenothreshold * docs: Extra spaces in comment (maybe intentional) The lines edited in this commit might have been given extra indentation purposefully. As they are an instruction for coverage py and are immediately followed by another comment that is not for coveragepy. * docs: inconsistent spaces in comments Fix extra indentations or lack of indentations. | ||||
* | docs: clarify that loads/dumps are not related to data files | Ned Batchelder | 2021-04-19 | 1 | -1/+8 |
| | |||||
* | Use current_thread instead of currentThread that was deprecated in Python 3.10 | Karthikeyan Singaravelan | 2021-04-17 | 2 | -5/+5 |
| | |||||
* | build: make tags like 5.6.1 not coverage-5.6.1 | Ned Batchelder | 2021-04-15 | 5 | -7/+7 |
| | |||||
* | build: remove one leftover install_egg | Ned Batchelder | 2021-04-14 | 1 | -1/+1 |
| | |||||
* | build: suppress new 3.10 warnings | Ned Batchelder | 2021-04-14 | 1 | -0/+11 |
| | |||||
* | build: avoid pylint randomness | Ned Batchelder | 2021-04-14 | 1 | -1/+1 |
| | | | | | | | -j seems to introduce non-determinism: https://github.com/PyCQA/pylint/issues/4356 With -j4, we'd have occasional (1 in 20?) failures on CI. | ||||
* | test: improve zipfile test | Ned Batchelder | 2021-04-14 | 4 | -45/+50 |
| | | | | | | | Before this commit, the GetZipBytesTest.test_get_encoded_zip_files test was flaky on Python 3.10.0a7. Since I had just added new files to the common zip file, I tried splitting the newly added stuff into its own file, and that seemed to fix the problem. | ||||
* | build: run tests and quality on all branches | Ned Batchelder | 2021-04-14 | 2 | -4/+0 |
| | |||||
* | test: remove the .egg test | Ned Batchelder | 2021-04-14 | 8 | -48/+24 |
| | | | | | | People don't use .egg much anymore, distutils is showing deprecation warnings, and coverage.py only deals with them the same way it deals with .zip files, so let's just rely on a .zip test to cover that. | ||||
* | fix: correct slight mis-layout of the hotkey panels | Ned Batchelder | 2021-04-13 | 6 | -16/+16 |
| | |||||
* | build: bump version to 5.6b2 | Ned Batchelder | 2021-04-12 | 2 | -1/+7 |
| | |||||
* | build: version 5.6b1coverage-5.6b1v5.x | Ned Batchelder | 2021-04-12 | 3 | -6/+8 |
| | |||||
* | fix: pypy3 7.3.4 uses a non-empty sys.path[0] | Ned Batchelder | 2021-04-12 | 1 | -1/+4 |
| | |||||
* | build: report errors a little better in download_gha_artifacts.py | Ned Batchelder | 2021-04-11 | 1 | -0/+2 |
| | |||||
* | build: use 3.10.0a7 in CI | Ned Batchelder | 2021-04-11 | 2 | -2/+2 |
| | |||||
* | build: update build dependencies | Ned Batchelder | 2021-04-11 | 3 | -12/+12 |
| | |||||
* | build: update pylint | Ned Batchelder | 2021-04-11 | 2 | -3/+3 |
| | |||||
* | fix: adapt to 3.10.0a7's f_lasti field | Ned Batchelder | 2021-04-11 | 3 | -2/+10 |
| | |||||
* | fix: restore metacov functioning | Ned Batchelder | 2021-04-11 | 2 | -3/+1 |
| | | | | | | | The check for coverage files inside the --source check disables our metacoverage. Removing it means that coverage files will still not be measured, but the reason will be given as "is third-party" rather than "is part of coverage.py," which is a small price to pay. | ||||
* | fix: restore html report selection highlighting | Ned Batchelder | 2021-04-11 | 2 | -7/+6 |
| | |||||
* | build: keep the cog sample report working | Ned Batchelder | 2021-04-11 | 1 | -1/+1 |
| | | | | | We use PYTEST_ADDOPTS=-n8 locally, and for some reason that keeps the cog line from measuring any data. This keeps it working | ||||
* | fix: don't measure third-party scriptsnedbat/bin-exclude | Ned Batchelder | 2021-04-11 | 3 | -59/+134 |
| | | | | | | This finishes the last bit of #905 Also includes tighter logging of the reason for not tracing modules. | ||||
* | test: don't complain if an environment variable we don't want doesn't exist | Ned Batchelder | 2021-04-11 | 1 | -1/+1 |
| | |||||
* | debug: label each matcher with its role | Ned Batchelder | 2021-04-11 | 4 | -21/+24 |
| | |||||
* | fix: don't measure third-party packages | Ned Batchelder | 2021-04-10 | 5 | -13/+210 |
| | | | | | | | | | | | | Avoid measuring code located where third-party packages get installed. We have to take care to measure --source code even if it is installed in a third-party location. This also fixes #905, coverage generating warnings about coverage being imported when it will be measured. https://github.com/nedbat/coveragepy/issues/876 https://github.com/nedbat/coveragepy/issues/905 | ||||
* | fix: make TreeMatcher right for case-sensitive worlds | Ned Batchelder | 2021-04-10 | 2 | -2/+9 |
| | |||||
* | refactor: move stdlib and coverage location logic into functions | Ned Batchelder | 2021-04-10 | 1 | -29/+41 |
| | |||||
* | build: tox should be quiet | Ned Batchelder | 2021-04-10 | 1 | -3/+3 |
| | |||||
* | build: remove obsolete Tidelift release notes support | Ned Batchelder | 2021-04-10 | 3 | -55/+0 |
| | | | | | Tidelift removed their release notes API on 2021-03-09: https://forum.tidelift.com/t/release-notes-task/467 | ||||
* | refactor: nice_file can be used as a function | Ned Batchelder | 2021-03-22 | 3 | -18/+19 |
| | |||||
* | refactor: make_file can be used as a function | Ned Batchelder | 2021-03-22 | 2 | -39/+46 |
| | |||||
* | refactor: simplify temp dir cd code | Ned Batchelder | 2021-03-22 | 1 | -15/+10 |
| | |||||
* | docs: update the help in the docs | Ned Batchelder | 2021-03-21 | 1 | -2/+2 |
| | |||||
* | docs: update a pytest url to their new structure | Ned Batchelder | 2021-03-21 | 1 | -1/+1 |
| | |||||
* | style: remove commented-out sphinx extension we don't need | Ned Batchelder | 2021-03-19 | 1 | -1/+0 |
| | |||||
* | fix: remove debugging code I checked in by accident | Ned Batchelder | 2021-03-19 | 1 | -3/+0 |
| | |||||
* | style: correct two recent lint faux pas | Ned Batchelder | 2021-03-18 | 3 | -12/+15 |
| | |||||
* | docs: clarify the --source values | Ned Batchelder | 2021-03-18 | 2 | -3/+5 |
| | | | | | | On the Test & Code podcast (https://testandcode.com/148) Brian Okken explained why the old wording was confusing. I hope this makes it clearer. | ||||
* | refactor: remove unused encoding parameter | Ned Batchelder | 2021-03-14 | 1 | -2/+2 |
| | |||||
* | docs: note what pep626 is | Ned Batchelder | 2021-03-14 | 1 | -0/+2 |
| | |||||
* | tool: more information about the location of ast nodes when debugging | Ned Batchelder | 2021-03-13 | 1 | -1/+6 |
| | |||||
* | refactor: we no longer use unittest_mixins | Ned Batchelder | 2021-03-11 | 1 | -4/+0 |
| | |||||
* | refactor: correct a file name: test_mixins.py | Ned Batchelder | 2021-03-11 | 1 | -0/+0 |
| | |||||
* | refactor: pull module cleaning into here | Ned Batchelder | 2021-03-11 | 5 | -66/+86 |
| | | | | We don't need unittest_mixins' module cleaner anymore. | ||||
* | refactor: remove unused test class setting | Ned Batchelder | 2021-03-11 | 5 | -18/+0 |
| | | | | | | | | | unittest_mixins would check that files got created if a test made a temporary directory, so that we could trim down making temp dirs needlessly. But we don't use unittest_mixins any more, so this setting does nothing. Remove it. | ||||
* | refactor: our own change_dir context manager | Ned Batchelder | 2021-03-11 | 4 | -4/+20 |
| | | | | We don't need to use the one from unittest_mixins. | ||||
* | test: simplify how StopEverything is converted to skipnedbat/remove-unittest-testcase | Ned Batchelder | 2021-03-11 | 5 | -64/+12 |
| | | | | | | The auto-decorating metaclass was interfering with parameterized methods on Python 2.7. But we don't need it anymore anyway, since pytest will let us hook to deal with the exception in a simpler way. |