jshint reports failure on javascript files and should be fixed. Whenever this is done, one can make the jslint job to be voting by editing integration/zuul-config.git layout.yaml file and remove the line that make the jslint job to be non voting.
Version: master
Severity: normal
| Subject | Repo | Branch | Lines +/- | |
|---|---|---|---|---|
| TMH now passes jslint | integration/config | master | +0 -2 |
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Paladox | T62619 All repositories should pass jshint test | |||
| Resolved | • Gilles | T63638 TimedMediaHandler should pass jshint |
Rumour has it a big chunk of the js is going to be replaced with something from upstream sometime soonish. Should probably wait for that before wotying about this.
If the Javascript code comes from upstream, one instruct jshint to ignore that code entirely by creating/editing .jshintignore
mdale wrote:
Upstream location: https://github.com/kaltura/mwEmbed/
Antoine, do you have link to the format of the jshitignore ?
I don't think we would be against passing jshint upstream, but we certainly want to get things in sync first, then worry about jshint consistency.
I guess the format is the usual one shell pattern per line :-D Ie something like:
/lib/3rdparty*
/src/bugged.js
@TheDJ made that happen in https://gerrit.wikimedia.org/r/#/c/173513/ and previous commits
Change 175668 had a related patch set uploaded (by Gilles):
TMH now passes jslint
That is only for our 'private modules' btw, the upstream parts are ignored right now. Not ideal, since we are actually heavily out of sync with upstream.
It's fair game to ignore 3rd-party libraries for linting. They're supposed to do that for themselves, with their own conventions.
TimedMediaHandler now pass jshint and @Gilles kindly updated the Zuul configuration to prevent regression (job is voting). Nothing left to be done. Thank you!