Created attachment 162654 [details] Update math/py-pandas to 0.17.0 Changelog ========= * Update to 0.17.0 QA == * portlint See attached. * nosetests Ran 9043 tests in 1429.402s FAILED (SKIP=269, errors=2, failures=1) See attached for details. ERROR: test_pytables_native_read (pandas.io.tests.test_pytables.TestHDFStore) ERROR: test_to_hdf_with_object_column_names (pandas.io.tests.test_pytables.TestHDFStore) I believe these are errors in devel/py-tables. FAIL: test_sum (pandas.tests.test_panel4d.TestPanel4d) I have reported this upstream (https://github.com/pydata/pandas/issues/11495).
Created attachment 162655 [details] math/py-pandas: portlint output
Created attachment 162656 [details] poudriere testport log for py27-pandas-0.17.0
Created attachment 162657 [details] poudriere testport log for py34-pandas-0.17.0
Created attachment 162658 [details] Nosetests output (w/python2.7 and all options enabled)
@John Given the consistency and quality of your contributions to date, please feel free to just confirm that your changes pass QA in your initial issue description, without needing to attach logs. Something like the following is fine: * portlint: OK|WARN: output * testport: OK (poudriere, versions, archs, combinations tested) * unittest: OK (Ran 9043 tests in 1429.402s FAILED (SKIP=269, errors=2, failures=1) Don't forget to also set maintainer-approval on the attachments you want committers to address. You can mouse over the name of any flag to see what each of the values mean.
A commit references this bug: Author: wen Date: Sun Nov 1 13:12:40 UTC 2015 New revision: 400611 URL: https://svnweb.freebsd.org/changeset/ports/400611 Log: - Update to 0.17.0 PR: 204183 Submitted by: john@saltant.com(maintainer) Changes: head/math/py-pandas/Makefile head/math/py-pandas/distinfo
@koobs: > Given the consistency and quality of your contributions to date, please feel > free to just confirm that your changes pass QA in your initial issue > description, without needing to attach logs. Got it. Thank you. > Don't forget to also set maintainer-approval on the attachments you want > committers to address. This confuses me. Submitting as the maintainer, isn't my approval of my own patch implied? @wen: Thank you for committing!
(In reply to John W. O'Brien from comment #7) It is implied, but searches cant be made on that because the concept of a maintainer is in the ports tree, but not bugzilla, hence the flag. Explicitly setting it results in the issue showing up in reports and saved searches that committers look at.