Editorial: Various fine-tooth comb changes #2944
Conversation
|
P.S. https://www.npmjs.com/package/@js-temporal/polyfill is apparently a year old; should it be republished? |
Yeah, @12wrigja and I have some work to do to catch it up with the latest commits. Sorry, I haven't had time for this recently. |
|
I guess node-codecov finally broke, faster than I hoped. I'll have to do #2931 before merging this. |
|
Also, are the Test262 failures expected? |
|
No test failures, it's just the codecov uploader failing. |
|
#2945 should fix it. Reviews appreciated! |
This was a mistake in #2943. _targetTime_.[[Days]] cannot exceed the original value of _duration_.[[Days]] here because its absolute value can only be equal or smaller coming out of AddTime.
"Input era ce doesn't match calculated value gregory-inverse" is incredibly confusing. Printing either "bc" or "bce" instead of "gregory-inverse" would be much preferable.
This has been annoying me for a long time, it decreases readability for little benefit.
These accidentally used the names of Duration slots, where they should be using Time Record slots.
… [[EpochNanoseconds]] Another thing that has been annoying me for a long time. Rename these to [[EpochNanoseconds]] so they don't get confused with the Duration [[Nanoseconds]] internal slot.
Move UnnormalizeDuration out of RoundRelativeDuration and
Difference{Plain,Zoned}DateTimeWithRounding, and into the calling
operations. This fits the normalize-calc-unnormalize pattern better, and
we can actually simplify the choice of which largestUnit is used for
UnnormalizeDuration.
…elds The description still used property names, but that is not correct anymore because we are dealing with a Calendar Fields Record, not an Object.
This was a straggler from the removal of custom calendars - these should be enumeration values from the Calendar Fields Record table, not property keys.
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #2944 +/- ##
==========================================
- Coverage 97.47% 97.46% -0.02%
==========================================
Files 22 22
Lines 10677 10672 -5
Branches 1872 1871 -1
==========================================
- Hits 10407 10401 -6
- Misses 270 271 +1 ☔ View full report in Codecov by Sentry. |
Here's a batch of editorial changes that resulted from a look over the spec text with a fine-tooth comb.