Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | refactor: convert all skipping to pytest skips | Ned Batchelder | 2021-02-07 | 1 | -9/+7 |
| | |||||
* | refactor: remove unittest.assertCountEqual | Ned Batchelder | 2021-02-06 | 1 | -1/+2 |
| | | | | Another step toward removing unittest.TestCase. | ||||
* | refactor: remove reliance on unittest_mixins.StdStreamCapturingMixin | Ned Batchelder | 2021-02-02 | 1 | -2/+3 |
| | | | | This is another step toward removing unittest.TestCase as a base class. | ||||
* | test: keep multi-assert arc diffs working | Ned Batchelder | 2021-01-31 | 1 | -4/+0 |
| | | | | | We don't have a way to do multi-assert in the pytest we're running, so cobble it together ourselves. | ||||
* | style: fix long lines and avoid backslashesnedbat/unittest2pytest | Ned Batchelder | 2021-01-31 | 1 | -4/+4 |
| | |||||
* | style: singleton comparisons should use is | Ned Batchelder | 2021-01-31 | 1 | -1/+1 |
| | | | | | | I guess the original line was wrong, but it would have been nice for unittest2pytest to fix it for me: https://github.com/pytest-dev/unittest2pytest/issues/52 | ||||
* | refactor: unittest2pytest -w tests | Ned Batchelder | 2021-01-31 | 1 | -12/+8 |
| | | | | One step of moving to pure pytest tests. | ||||
* | A better test for 'if not __debug__' | Ned Batchelder | 2021-01-10 | 1 | -11/+13 |
| | |||||
* | PEP 626: constant tests are kept as no-ops | Ned Batchelder | 2021-01-10 | 1 | -28/+113 |
| | | | | | The conditionals are now getting unwieldy, perhaps we can simplify them in the future? | ||||
* | In 3.10, modules always have firstlineno==1 | Ned Batchelder | 2021-01-10 | 1 | -1/+3 |
| | |||||
* | Python 3.10 doesn't compile statments after unconditional jumps. | Ned Batchelder | 2021-01-10 | 1 | -7/+30 |
| | | | | This includes break/continue/return/raise. | ||||
* | More bitbucket->github urls | Ned Batchelder | 2020-08-18 | 1 | -6/+6 |
| | |||||
* | handle decorators for AsyncFunctionDefs. Closes #964 | Kjell Braden | 2020-03-24 | 1 | -0/+17 |
| | |||||
* | Remove a temporary xfail for 3.9a2 | Ned Batchelder | 2020-01-25 | 1 | -8/+1 |
| | |||||
* | Better xfail mechanism | Ned Batchelder | 2020-01-18 | 1 | -15/+8 |
| | |||||
* | bpo39166 is fixed | Ned Batchelder | 2020-01-10 | 1 | -1/+1 |
| | |||||
* | A better name for a thing I will soon delete | Ned Batchelder | 2020-01-04 | 1 | -6/+6 |
| | |||||
* | bpo-39114 will be fixed soon. | Ned Batchelder | 2019-12-31 | 1 | -2/+2 |
| | |||||
* | bpo 39114 broke 3.9, but maybe it's their bug | Ned Batchelder | 2019-12-21 | 1 | -0/+14 |
| | |||||
* | Exit arcs have to be corrected to first lines | Ned Batchelder | 2019-09-21 | 1 | -45/+25 |
| | | | | | | | | | | We've long remapped line numbers to the first line of a multi-line statement. But exit line numbers (negative numbers) were not remapped. This meant we were needlessly chasing weirdnesses in implementations. But the actual results of running coverage always remapped results to the first line, so there's no point in tracking the unmapped line numbers in our tests. | ||||
* | Adapt to optimization tweak in CPython 3.8.0b1 | Ned Batchelder | 2019-06-05 | 1 | -0/+3 |
| | |||||
* | Move more PYVERSION to be PYBEHAVIOR | Ned Batchelder | 2019-03-24 | 1 | -9/+13 |
| | |||||
* | One last decorator test change for 3.8 | Ned Batchelder | 2018-11-11 | 1 | -2/+10 |
| | |||||
* | Python 3.8 will optimize away "while True:" | Ned Batchelder | 2018-11-11 | 1 | -9/+28 |
| | |||||
* | Use a new event loop | Ned Batchelder | 2018-10-24 | 1 | -2/+2 |
| | |||||
* | Finally jumps back to exiting lines | Ned Batchelder | 2018-10-06 | 1 | -17/+53 |
| | | | | | | In Python 3.8, when a finally clause is run because a line in the try block is exiting the block, the exiting line is visited again after the finally block. | ||||
* | Update NOTICE link to GitHub. | Ned Batchelder | 2018-06-24 | 1 | -1/+1 |
| | |||||
* | No more support for 2.6 or 3.3 | Ned Batchelder | 2018-05-05 | 1 | -4/+0 |
| | |||||
* | Fine-tune the 3.7 decisions | Ned Batchelder | 2018-01-13 | 1 | -1/+1 |
| | |||||
* | Python 3.7 optimizes away 'not __debug__' | Ned Batchelder | 2018-01-10 | 1 | -3/+9 |
| | |||||
* | More-precise no-coverage pragmas | Ned Batchelder | 2017-01-19 | 1 | -2/+2 |
| | |||||
* | No test failures on Jython | Ned Batchelder | 2017-01-18 | 1 | -0/+7 |
| | | | | One or two of these are questionable accommodations, but there are no failures. | ||||
* | A few more lines covered | Ned Batchelder | 2017-01-17 | 1 | -3/+4 |
| | |||||
* | Properly handle if-statements optimized away. #522 | Ned Batchelder | 2017-01-16 | 1 | -15/+96 |
| | |||||
* | A cleaner way to exclude Jython concerns from meta-coverage. | Ned Batchelder | 2017-01-16 | 1 | -1/+1 |
| | |||||
* | More pragmas for test skips | Ned Batchelder | 2017-01-14 | 1 | -2/+2 |
| | |||||
* | Forgot a bunch of jython coverage pragmas | Ned Batchelder | 2017-01-14 | 1 | -1/+1 |
| | |||||
* | More Jython test fixes/skips | Ned Batchelder | 2017-01-14 | 1 | -0/+3 |
| | |||||
* | 2.6 and 2.7 are not different wrt constants in while loops | Ned Batchelder | 2016-12-24 | 1 | -11/+5 |
| | |||||
* | Move the test program into the test | Ned Batchelder | 2016-12-24 | 1 | -7/+10 |
| | |||||
* | also use AST for while constants in python-2.7 #502 | loic@dachary.org | 2016-12-15 | 1 | -2/+19 |
| | | | | | | | | | | | The node.id is set to False, True or None is python-2.7: there is no reason to only check for it with python-3. It is more reliable than using the DEFAULT_PARTIAL_ALWAYS regexps on source lines. close #502 --HG-- branch : issue-502-7 | ||||
* | Bug #496 wasn't about the single-line continue, just the constant while. | Ned Batchelder | 2016-12-18 | 1 | -3/+4 |
| | |||||
* | Fix #496, while-true loop with a continue. | Ned Batchelder | 2016-12-18 | 1 | -0/+16 |
| | |||||
* | Clean up #493 fix | Ned Batchelder | 2016-12-18 | 1 | -24/+9 |
| | |||||
* | finally happens before return in a try #493 | loic@dachary.org | 2016-12-18 | 1 | -0/+18 |
| | | | | | | | | | | | | | | | | | | | | | In a try block such as: if expr: try: return finally print pass the print happens before the return and cannot be followed by pass. The general case is that when the body/else/handlers in a try block all have return, break etc., the code behind finally: has no arc to the statement following the try block. close #493 --HG-- branch : issue-493-2 | ||||
* | Async protocol changed in 3.5.2. Remove a deprecation warning | Ned Batchelder | 2016-11-27 | 1 | -1/+1 |
| | |||||
* | Use standard skipTest, which unittest-mixins will override. | Ned Batchelder | 2016-06-12 | 1 | -5/+5 |
| | |||||
* | Better descriptions of missing one-line executables. Part of #475 | Ned Batchelder | 2016-03-27 | 1 | -0/+30 |
| | |||||
* | Entry arcs now use the negative first line of the code object instead of -1. | Ned Batchelder | 2016-03-02 | 1 | -45/+45 |
| | |||||
* | Better branch-missed messages for lambdas. | Ned Batchelder | 2016-02-25 | 1 | -0/+20 |
| |