summaryrefslogtreecommitdiff
path: root/coverage/collector.py
Commit message (Collapse)AuthorAgeFilesLines
...
* Touching files in an empty CoverageData shouldn't force lines. #399Ned Batchelder2015-08-201-3/+3
| | | | | | | | py.test --cov can write empty data to arcs, and then touch files due to --source. This would make the file be about lines instead of arcs, and then combining would fail. This changed a lot of CoverageData, including some of the methods names.
* Too many things named "arcs". Rename the booleans.Ned Batchelder2015-08-081-1/+1
|
* Why didn't I get rid of this TODO ages ago?Ned Batchelder2015-08-021-1/+2
|
* Implement FileDisposition in C for speed. Final part of fix for #387.Ned Batchelder2015-08-011-2/+12
|
* Use more specific names than 'plugins', since there will be more of them in ↵Ned Batchelder2015-07-241-5/+6
| | | | the future.
* Add license mention to the top of all files. #313.Ned Batchelder2015-07-241-1/+4
|
* Data files are now JSON instead of pickles. Fixes #236.Ned Batchelder2015-07-211-3/+3
|
* Refactor collector->data; data has only one of lines and arcs.Ned Batchelder2015-07-181-37/+15
| | | | | | Now the collector communicates directly with the data, and control is less involved. In the data, when measuring arcs, only arcs are stored. Lines are calculated as needed. This saves space in the data file, and is faster.
* Add some docstringsNed Batchelder2015-07-031-2/+10
|
* Clean up some pylintNed Batchelder2015-07-031-3/+3
|
* Correct the comment describing .should_trace_cacheNed Batchelder2015-06-161-3/+8
|
* Be more disciplined about the values in should_trace_cache. #374.Ned Batchelder2015-06-141-2/+1
|
* Don't report negative line numbers through get_line_dataNed Batchelder2015-02-191-2/+6
|
* Get the tracer right for metacovNed Batchelder2015-02-101-1/+5
|
* Plugin support is now only in the CTracer, not in the PyTracer.Ned Batchelder2015-02-091-0/+2
|
* Collect all the nudgy environment checks into coverage.envNed Batchelder2015-01-181-12/+9
|
* OK, I should really stop fiddling with spell-check...Ned Batchelder2014-11-271-2/+2
|
* Change how dynamic source filenames work in plugins.Ned Batchelder2014-11-241-0/+2
|
* Change all pylint message codes to message namesNed Batchelder2014-11-231-1/+1
|
* Upgraded pylint to 1.4, removed useless suppressionsNed Batchelder2014-11-231-2/+2
|
* Add comment and cleanups to Alex's PyPy change.Ned Batchelder2014-10-011-1/+15
|
* Merged in ↵Ned Batchelder2014-10-011-1/+9
|\ | | | | | | | | | | alex_gaynor/coveragepy/alex_gaynor/improve-performance-of-coverage-under-py-1411425050845 (pull request #40) Improve performance of coverage under PyPy.
| * Improve performance of coverage under PyPy.Alex Gaynor2014-09-221-1/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | An explanation is in order: should_trace_cache is a strictly growing key, which is to say once a key is in it, it never changes. Further, the keys used to access it are generally constant, given sufficient context. That is to say, at any given point _trace() is called, a Sufficient Smart Compiler (tm) is able to know the key. This is because the key is determined by the physical source code line, and that's obviously invariant with the call site. This property of a dict with immutable keys, combined call-site-constant keys is a match for PyPy's module dict, which is optimized for such workloads. This gives a 20% benefit on the workload described at https://bitbucket.org/pypy/pypy/issue/1871/10x-slower-than-cpython-under-coverage I have not benchmarked on a wider array of programs yet. --HG-- branch : alex_gaynor/improve-performance-of-coverage-under-py-1411425050845
* | Some error checking and more tests for concurrency control.Ned Batchelder2014-09-251-1/+3
| |
* | "concurrency" is a better name that "coroutine"Ned Batchelder2014-09-241-22/+22
| | | | | | | | | | --HG-- rename : tests/test_coroutine.py => tests/test_concurrency.py
* | Every comment makes the world slightly betterNed Batchelder2014-09-231-0/+1
|/
* More metacov, tox, and make cleanupNed Batchelder2014-09-201-1/+0
|
* More things we don't need with the latest versionsNed Batchelder2014-09-201-6/+5
|
* Coroutines are now only supported with the C tracer, and better error handlingNed Batchelder2014-09-201-18/+45
|
* Split this into two files, as it should have been years ago.Ned Batchelder2014-09-191-169/+3
| | | | | --HG-- rename : coverage/collector.py => coverage/pytracer.py
* Remove more commented-out code. #274.Ned Batchelder2014-09-171-6/+1
|
* Gevent, etc, support finally works. #149Ned Batchelder2014-09-171-36/+30
|
* Move dispositions closer to useful pluginsNed Batchelder2014-09-061-30/+45
|
* Rename extensions to plugins, since extensions reminds me of parts of filenamesNed Batchelder2014-08-201-16/+16
| | | | | --HG-- rename : coverage/extension.py => coverage/plugin.py
* Crazy debugging output for tracking down gevent weirdnessNed Batchelder2014-07-131-7/+33
|
* Crazy-ugly start to extensions for Django and MakoNed Batchelder2014-07-041-16/+24
| | | | | --HG-- branch : django
* should_trace now returns a FileDisposition objectNed Batchelder2014-06-151-6/+7
| | | | | --HG-- branch : django
* Hacked-in django template supportNed Batchelder2014-06-121-14/+24
| | | | | --HG-- branch : django
* Peter Portante's coroutine support, but it doesn't work yet.Ned Batchelder2014-01-091-2/+28
|
* The PyTracer can't always be stopped, so give it a flag to stop tracing when ↵Ned Batchelder2013-09-161-1/+8
| | | | that happens. This silences a noisome exception during the test suite.
* PyTracer's cache broke when I changed False to None from should_trace. Fixed.Ned Batchelder2013-09-041-2/+3
|
* New flag for run: --debug, can trace why files aren't being traced.Ned Batchelder2013-09-041-2/+2
|
* Don't issue spurious warnings about the trace function changing. Fixes #164Ned Batchelder2013-03-101-3/+7
|
* A more convenient way to check on the trace function.Ned Batchelder2013-01-111-3/+4
|
* Clean up debugging stuffNed Batchelder2012-12-211-4/+5
|
* Debugging to understand why 'trace function changed' appears.Ned Batchelder2012-12-201-2/+4
|
* Silence the last of the pylint warnings.Ned Batchelder2012-11-201-1/+1
|
* Fix #201: wrong default for traces0 being missing.Ned Batchelder2012-10-131-1/+1
|
* If fullcoverage is run with the wrong tracer, make it apparent.Ned Batchelder2012-08-301-1/+6
|
* Undo Brett's change to the fullcoverage feature: it should only be used with ↵Ned Batchelder2012-08-301-1/+2
| | | | the C tracer.
View Lorry Controller Status