gador/nixpkgs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

python3Packages.parsimonious: disable benchmark tests that may fail

Browse Source 
This fixes

```
=================================== FAILURES ===================================
______________________ TestBenchmarks.test_lists_vs_dicts ______________________

self = <parsimonious.tests.test_benchmarks.TestBenchmarks testMethod=test_lists_vs_dicts>

    def test_lists_vs_dicts(self):
        """See what's faster at int key lookup: dicts or lists."""
        list_time = timeit('item = l[9000]', 'l = [0] * 10000')
        dict_time = timeit('item = d[9000]', 'd = {x: 0 for x in range(10000)}')

        # Dicts take about 1.6x as long as lists in Python 2.6 and 2.7.
>       self.assertTrue(list_time < dict_time, '%s < %s' % (list_time, dict_time))
E       AssertionError: False is not true : 0.051286615896970034 < 0.0440241270698607

parsimonious/tests/test_benchmarks.py:16: AssertionError
=========================== short test summary info ============================
FAILED parsimonious/tests/test_benchmarks.py::TestBenchmarks::test_lists_vs_dicts
=================== 1 failed, 82 passed, 2 skipped in 1.19s ====================
```
This commit is contained in:
Ivan Kozik
2023-01-17 05:07:30 +00:00
parent 768a982bfc
commit e31dea3d14
1 changed files with 8 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pkgs/development/python-modules/parsimonious/default.nix
+8
View File
@@ -26,6 +26,14 @@ buildPythonPackage rec {
pytestCheckHook
];
disabledTests = [
# test_benchmarks.py tests are actually benchmarks and may fail due to
# something being unexpectedly slow on a heavily loaded build machine
"test_lists_vs_dicts"
"test_call_vs_inline"
"test_startswith_vs_regex"
];
postPatch = ''
substituteInPlace setup.py \
--replace "regex>=2022.3.15" "regex"