Mercurial > p > roundup > code
view test/conftest.py @ 8458:15a92b0a9b79
fix: make user_src_input generate valid javascript
user_src_input used to generate False if edit_ok == False in this
statement:
tal:attributes="onblur python:edit_ok and 'split_name(this)';
but False isn't a boolean in javascript, so it throws an error in the
console. Changed to use:
tal:attributes="onblur python:'split_name(this)' if edit_ok else '';
which generates an empty onblur if the field is not editable.
| author | John Rouillard <rouilj@ieee.org> |
|---|---|
| date | Thu, 25 Sep 2025 23:30:07 -0400 |
| parents | 43dc610bbd24 |
| children |
line wrap: on
line source
# simple way to see if there are order dependencies in tests # can use if pytest-random-order --random-order mode isn't # usable (e.g. python2). # known issues: # python3 -m pytest -k Whoosh test/test_indexer.py fails starting with # first reversed (so last) test in module # # python3 -m pytest test/test_cgi.py # fails: FormTestCase::testCreatePermission # FormTestCase::testClassPermission # FormTestCase::testCheckAndPropertyPermission # # this failure results in a failure in test_action again with # bad permission application. Something run prior to these # tests is breaking the permission checks. #def pytest_collection_modifyitems(items): # items.reverse() # Add a marker for pg_schema tests. # They duplicate the postgresql tests exactly but uses a named # schema rather than the default 'public' schema. def pytest_configure(config): config.addinivalue_line( "markers", "pg_schema: tests using schema for postgres" ) # try to work around loss of testmon data when ^Cing out of tests. def pytest_unconfigure(config): if hasattr(config, "testmon_data"): config.testmon_data.db.con.close()
