IdlHarness: Improve the formatting of generic types. #13142
Conversation
Ms2ger
force-pushed
the
idlharness-member-tostring
branch
from
8bff786
to
d273149
Compare
|
@Ms2ger can you rebase that and trigger som full runs so that it's easy to see what effect this will have? |
Ms2ger
force-pushed
the
idlharness-member-tostring
branch
from
d273149
to
c5f101b
Compare
|
@foolip rebased, but I don't know how to trigger full runs. |
|
@Ms2ger see https://web-platform-tests.org/running-tests/from-ci.html Chrome Dev, Firefox Nightly and Safari TP should be enough to catch anything interesting in the changes. |
|
Oh wait, the affected tests already includen the right tests, so no need to trigger full runs. |
|
https://wpt.fyi/results/?diff&filter=ADC&q=is%3Adifferent&run_id=349900001&run_id=340780001 is a good view to start from. Is it intentional that " |
|
No, that's a bug, probably caused by w3c/webidl2.js@f4292a7. I attempted a fix. CC @saschanaz |
|
The generated names look pretty good now, although I certainly haven't read through all of them. @Ms2ger this seems like should be very suited to unit testing, can you add some to show at least the basic cases? That'd also prevent regressions with future webidl2.js updates more effectively than skimming differences in results. |
This deduplicates some code, and improves the output. This ensures that unions and generics don't show up as [object Object], and it makes the output for other arguments more accurate.
Ms2ger
force-pushed
the
idlharness-member-tostring
branch
from
3d6dc61
to
2c9e585
Compare
|
Finally got around to this, and improved the output a bit more. re-r? @foolip |
There was a problem hiding this comment.
Yay! In https://wpt.fyi/results/?diff&filter=ADC&q=status%3Amissing&run_id=407600001&run_id=432290001 one can spot plenty of improved test names.
|
wpt-firefox-nightly-stability timed out because there were too many tests to run. If it hadn't timed out it probably would have failed because of #16175. Admin merging. |
No description provided.