Current parser (pyparsing) is not parsing correctly those signed-off-by patches that contain unicode chars (like Aníbal Limón), thus giving false positives: > git pw mbox 337 | patchtest - --test-dir ~/repos/pt-oe/tests --no-apply --raw ---------------------------------------------------------------------- Ran 4 tests in 6.249s OK [["ID", "mbox.signed_off_by.test_signed_off_by.SignedOffBy.test_signed_off_by_format"], ["line", "Signed-off-by: An\u00edbal Lim\u00f3n <anibal.limon@linux.intel.com>"], ["column", 18]] [["ID", "mbox.subject.test_subject.Subject.test_subject_length"], ["Subject", "utils/qemurunner.py: QemuRunner.start() add support for specify extra kernel cmdline"], ["Length", "Current length 84 Max length 80"]] where .git/config contains [patchwork "default"] url = http://patchwork.openembedded.org project = oe-core
Instead of using pyparsing, standard (regular expression) python module can be used.
Using pyparsing regular expressions to fix this bug.