summaryrefslogtreecommitdiff
path: root/coverage
Commit message (Collapse)AuthorAgeFilesLines
...
* | fixed an error where not specifying the cover package would cause a tracebackDavid Stanek2010-05-211-4/+6
| |
* | modified to use report as the default action - the way the old nose plugin ↵David Stanek2010-05-211-1/+1
| | | | | | | | worked
* | merged in latest changes from Ned's repoDavid Stanek2010-05-213-10/+303
|\ \ | |/ |/|
| * moved the nose plugin to a better nameDavid Stanek2010-02-201-0/+0
| | | | | | | | | | --HG-- rename : coverage/nose_coverage.py => coverage/noseplugin.py
| * fixed busted nose integrationDavid Stanek2010-02-202-6/+6
| |
| * integrate a patch by prologic (James Mills) to add some options.holger krekel2010-02-101-4/+32
| |
| * merging/unifying test plugin codeholger krekel2009-11-253-116/+56
| | | | | | | | | | | | | | | | | | | | | | * coverage/testplugin.py contains common test plugin options and pytest hooks (which don't induce "import py") * coverage/nose_coverage.py contains a basic Nose Plugin * test/test_testplugin.py contains a pytest-functional test and a nose-skeleton one. skipped as appropriate. --HG-- rename : coverage/runner.py => coverage/testplugin.py
| * * move pytest_coverage/nose_plugin to become directly importableholger krekel2009-11-253-68/+68
| | | | | | | | | | | | | | | | | | * fixing order of coverage options * pytest: only do coverage-processing if cover-action is defined --HG-- rename : coverage/test_plugins/nose_coverage.py => coverage/nose_coverage.py rename : coverage/test_plugins/pytest_coverage.py => coverage/pytest_coverage.py
| * Merging in changes from http://bitbucket.org/ned/coveragepy/rozza@x103086.gcapmedia.com2009-11-205-92/+128
| |\
| * | Typorozza@x103086.gcapmedia.com2009-11-201-1/+1
| | |
| * | Adding a CoverageTestWrapper to be used by test runners / plugins.rozza@x103086.gcapmedia.com2009-11-203-0/+318
| | | | | | | | | | | | | | | | | | | | | Added example plugins for pytest and nose Warning: I haven't tested the nose plugin against a test suite Warning: I use a file listing files / paths to omit from code coverage and use regexes to filter them out To ignore that remove the monkey patch..
| * | Moving file filter logic out so it can be monkey patched easilyrozza@x103086.gcapmedia.com2009-11-201-10/+17
| | |
* | | Format the code to pylint's liking, and fix up the docstrings for omit and ↵Ned Batchelder2010-05-159-92/+87
| | | | | | | | | | | | include.
* | | Hook up omit and include to the run command. Test the new cmdline behavior ↵Ned Batchelder2010-05-153-11/+16
| | | | | | | | | | | | and run behavior.
* | | Refactor common code in config.py, and start testing the new options.Ned Batchelder2010-05-151-36/+22
| | |
* | | Change 'requires' to 'include'.Ned Batchelder2010-05-159-94/+94
| | |
* | | Undo Zooko's version number change.Ned Batchelder2010-05-151-1/+1
| | |
* | | fix a couple of bugs in control.py such that handling of omit_prefixes ↵wonwinmcbrootles@Wonwin-McBrootles-Computer.local2010-05-092-8/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | and/or require_prefixes was wrong update cmdline tests so they don't go red due to an unexpected constructor argument 'require_prefixes' note that there is not a real test of the "--require" cmdline argument -- I just added "require_prefixes=None" to the tests, so there is not test which tests what happens when someone passes --require= to the summary commands, much less to the run command. --HG-- extra : transplant_source : %F7%B3%26E%27Va%A2W4%9EX%9F%02oV%3C%A2%27%E1
* | | add a --require option to specify directories which are required to be at ↵Zooko Ofsimplegeo2010-04-199-28/+168
| | | | | | | | | | | | | | | | | | | | | | | | | | | the beginning of the path for any file that is going to be included in code coverage make the --omit and --require options apply to code coverage generation as well as to reporting; This speeds up tests from 6 seconds to 1 second on my system, as well as making the resulting .coverage db include *only* the code that I care about, which helps with my code coverage progression/regression tool. --HG-- extra : transplant_source : %1F.4%81%E8%DA%0B%D0%D5%9D%89%DE%E1vY%E6%CD%1A%EB%C9
* | | Don't trace any files with names starting with less-than. They are all ↵Ned Batchelder2010-05-081-3/+5
| | | | | | | | | | | | fake, and can't be reported on anyway. Fixes #61.
* | | Properly record multiple exits separately. Fixes #62.Ned Batchelder2010-04-274-20/+35
| | |
* | | Use constants for return values to make it easier to see the various uses of ↵Ned Batchelder2010-04-261-17/+21
| | | | | | | | | | | | 0 and -1.
* | | Undo Ben Finney's patch to use the os.path methods. os.path.commonprefix is ↵Ned Batchelder2010-03-211-6/+6
| | | | | | | | | | | | both useless (beacause it's based on characters, not path components) and wrong (because to compute a relative path name, we need relative to a particular directory, or absolute, not some middle case).
* | | Not sure why I changed these from inherit to normal last week (something to ↵Ned Batchelder2010-03-201-2/+2
| | | | | | | | | | | | do with IE and the styling of the linked line numbers). 'normal' made the footer not be italic, and putting it back to 'inherit' doesn't seem to break anything.
* | | Calculate the pid suffix for data files at the end of the process so that ↵Ned Batchelder2010-03-202-29/+37
| | | | | | | | | | | | programs calling os.fork will collect data from both child and parent. Fixes issue #56.
* | | Source files can have more than one dot in them. Fixes issue #46.Ned Batchelder2010-03-141-2/+2
| | |
* | | Make this function name better.Ned Batchelder2010-03-142-2/+2
| | |
* | | Added a link back to the index page at the bottom of HTML source pages. ↵Ned Batchelder2010-03-143-3/+15
| | | | | | | | | | | | Closes issue #38.
* | | HTML line numbers are clickable, and highlight the line on arrival. Also, ↵Ned Batchelder2010-03-143-4/+30
| | | | | | | | | | | | noticed that IE8 didn't line up the line number properly, so added an unfortunate meta tag to make it right. Closes issue #55.
* | | Reports now emphasize missed lines over executed lines, since those are more ↵Ned Batchelder2010-03-133-13/+13
| | | | | | | | | | | | helpful for directing developers to improved test coverage.
* | | Minor cleanupNed Batchelder2010-03-131-3/+3
| | |
* | | I like this punctuation betterNed Batchelder2010-03-131-1/+2
| | |
* | | Merged Ben Finney's use-os-path-module fixes (again?)Ned Batchelder2010-03-131-2/+4
|\ \ \
| * | | Use ‘os.path.commonprefix’ to find the “relative directory” at the ↵Ben Finney2010-02-031-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | start of a path. --HG-- branch : use-os-path-module
* | | | Bump version numberNed Batchelder2010-03-131-1/+1
| | | |
* | | | Deal properly with dos-ended source files while tokenizing. Fixes issue #53.Ned Batchelder2010-03-131-1/+2
| | | |
* | | | Don't let parallel=True prevent reporting. Fixes #49.Ned Batchelder2010-03-061-9/+10
| | | |
* | | | We'll call this 3.3.1Ned Batchelder2010-02-281-1/+1
| | | |
* | | | If the user's code calls sys.exit(), honor the request and exit with that ↵Ned Batchelder2010-02-282-6/+15
| | | | | | | | | | | | | | | | status. Fixes issue #50.
* | | | When emulating the Python interpreter, don't print SystemExits tracebacks.Ned Batchelder2010-02-241-1/+6
| | | |
* | | | Move the version number to 3.3Ned Batchelder2010-02-241-1/+1
| | | |
* | | | Docs functionally complete for 3.3Ned Batchelder2010-02-241-2/+3
| | | |
* | | | Python source files that don't end with a newline can now be executed, ↵Ned Batchelder2010-02-222-10/+17
| | | | | | | | | | | | | | | | fixing #47.
* | | | SpacesNed Batchelder2010-02-221-3/+3
| | | |
* | | | Fiddling with docs for 3.3Ned Batchelder2010-02-221-6/+13
| | | |
* | | | Lint cleanupsNed Batchelder2010-02-201-1/+3
| | | |
* | | | If the product code throws an exception, 'coverage run' now produces the ↵Ned Batchelder2010-02-193-4/+28
| | | | | | | | | | | | | | | | same traceback as 'python' would, without the coverage-internal frames distracting from your code.
* | | | Clean up the lst += string stuff, whereby string being iterable means each ↵Ned Batchelder2010-02-073-21/+21
| | | | | | | | | | | | | | | | char is added as an element of the list. Also, apply the 'join is better than append' rule to other places, where, alas, it doesn't seem to have the same magic effect.
* | | | You know that thing they say about how ''.join(slist) is faster than s1 += ↵Ned Batchelder2010-02-071-2/+2
| | | | | | | | | | | | | | | | s? It's really true. This one change removed 90% of the templite time, and 59% of the html report generation time.
* | | | Update a few copyrights to 2010.Ned Batchelder2010-01-101-1/+1
| | | |