Date: Add fractional seconds support in formatters #763

mattyork · 2017-08-03T18:24:06Z

Follows this algorithm:
http://www.unicode.org/reports/tr35/tr35-dates.html#Matching_Skeletons

Fixes #753

Follows this algorithm: http://www.unicode.org/reports/tr35/tr35-dates.html#Matching_Skeletons Fixes globalizejs#753

Ref globalizejs#753

jsf-clabot · 2017-08-03T18:24:07Z

All committers have signed the CLA.

rxaviers · 2017-08-03T21:11:14Z

test/compiler/cases/date.js

@@ -29,7 +29,7 @@ module.exports = {

 			{ formatter: Globalize.dateFormatter({ skeleton: "GyMMMEd" }), args: [ date ] },
 			{ formatter: Globalize.dateFormatter({ skeleton: "dhms" }), args: [ date ] },
-			{ formatter: Globalize.dateFormatter({ skeleton: "GyMMMEdhms" }), args: [ date ] },
+			{ formatter: Globalize.dateFormatter({ skeleton: "GyMMMEdhmsSSS" }), args: [ date ] },


Let's keep the original test without S and have the additional test with S.

sounds good!

rxaviers · 2017-08-03T21:12:31Z

src/date/expand-pattern/augment-format.js

 	}

-	return bestMatchFormat.join( "" );
+	return bestMatchFormatParts.join( "" );


Good name addition, i.e., Parts which they are.

rxaviers · 2017-08-03T21:18:24Z

src/date/expand-pattern/augment-format.js

-		for ( j = 0; j < requestedSkeleton.length; j++ ) {
-			requestedType = requestedSkeleton[j].charAt( 0 );
-			requestedLength = requestedSkeleton[j].length;
+	requestedSkeletonParts = skeletonWithoutFractionalSeconds.match( datePatternRe );


I'm wondering how this approach scales... Maybe, not worth worrying about it now, but for example, I'm wondering if handling timeZone would take a similar approach.

Does it make sense to leave this function mostly as is, and handle S by a parent function? Basically, by replacing the existing call stack:

- getBestMatchPattern() -> augmentFormat() + getBestMatchPattern() -> stuffThatHandlesS() -> augmentFormat()

Yeah, that makes sense. Is it cool if I keep both of those functions in augment-format.js? That way the file perfectly reflects the "adjustment" section of the spec:

Once a skeleton match is found, the corresponding pattern is used, but with adjustments. Consider the following dateFormatItem:

<dateFormatItem id="yMMMd">d MMM y</dateFormatItem>

If this is the best match for yMMMMd, pattern is automatically expanded to produce the pattern "d MMMM y" in response to the request. Of course, if the desired behavior is that a request for yMMMMd should produce something other than "d MMMM y", a separate dateFormatItem must be present, for example:

<dateFormatItem id="yMMMMd">d 'de' MMMM 'de' y</dateFormatItem>

If the requested skeleton included both seconds and fractional seconds and the dateFormatItem skeleton included seconds but not fractional seconds, then the seconds field of the corresponding pattern should be adjusted by appending the locale’s decimal separator, followed by the sequence of ‘S’ characters from the requested skeleton.

rxaviers · 2017-08-03T21:25:11Z

test/unit/date/expand-pattern/get-best-match-pattern.js

@@ -30,7 +30,7 @@ QUnit.test( "should get best match pattern", function( assert ) {
 	assert.equal( getBestMatchPattern( en, "cccd" ), "d EEE" );
 	assert.equal( getBestMatchPattern( en, "ccccd" ), "d EEEE" );

-	assert.equal( getBestMatchPattern( en, "hhmms" ), "hh:mm:ss a" );
+	assert.equal( getBestMatchPattern( en, "hhmmsS" ), "hh:mm:ss.S a" );


Can we preserve the original test and have yours as an additional one?

mattyork · 2017-08-04T00:10:09Z

src/date/expand-pattern/augment-format.js

+}
+
+// See: http://www.unicode.org/reports/tr35/tr35-dates.html#Matching_Skeletons
+return function( requestedSkeleton, bestMatchFormat, decimalSeparator ) {


@rxaviers how about this? Same file, two functions. Encapsulates all the "adjustments" to the bestMatchFormat

It's good except for our build process, which assumes all files have one export (and no outer code). In order to have something different, we need an update the onBuildWrite, specifically tell it this file is a "type b".

Got it. There's a problem though: if I make it a "type b", that capitalizes the first character of the variable name, which breaks the output, as the actual variable name does not have a capital first letter. Example output:

var DateExpandPatternAugmentFormat = (function() { function expandBestMatchFormat( skeletonWithoutFractionalSeconds, bestMatchFormat ) { ... } // See: http://www.unicode.org/reports/tr35/tr35-dates.html#Matching_Skeletons return function( requestedSkeleton, bestMatchFormat, decimalSeparator ) { ... bestMatchFormat = expandBestMatchFormat( skeletonWithoutFractionalSeconds, bestMatchFormat ); ... }; }());

Why does it capitalize the first letter in DateExpandPatternAugmentFormat? The only other "type b" is "util/globalize-date", but that file doesn't seem to exist anymore, so I can't look at the example to know what the intention of the capitalization is.

However, if I leave it as a "type a", everything still works. This is the output:

function expandBestMatchFormat( skeletonWithoutFractionalSeconds, bestMatchFormat ) { ... } var dateExpandPatternAugmentFormat = function( requestedSkeleton, bestMatchFormat, decimalSeparator ) { ... bestMatchFormat = expandBestMatchFormat( skeletonWithoutFractionalSeconds, bestMatchFormat ); ... };

This works just fine.

So, what should I do here?

Leave as a type a

Change how the build handles type b so it stops capitalizing. If I do this should I also remove the reference to util/globalize-date?

Something else? Move the other function to another file?

@rxaviers ?

Number 2. (sorry for the delay)

Yeap, existing "type b" is bugged. Let's fix it by stop capitalizing the variable and by also removing the reference to the non-existing util/globalize-date. Please, take a look at this old-but-correct "type b" implementation: here and here.

Thanks!

Ref globalizejs#753

mattyork · 2017-08-14T22:29:57Z

How bout meow?

rxaviers · 2017-08-15T00:23:39Z

test/functional/date/date-formatter.js

@@ -88,7 +88,7 @@ QUnit.test( "should return a formatter", function( assert ) {

 	assert.equal( Globalize.dateFormatter({ skeleton: "GyMMMEd" })( date ), "Wed, Sep 15, 2010 AD" );
 	assert.equal( Globalize.dateFormatter({ skeleton: "dhms" })( date ), "15, 5:35:07 PM" );
-	assert.equal( Globalize.dateFormatter({ skeleton: "GyMMMEdhms" })( date ), "Wed, Sep 15, 2010 AD, 5:35:07 PM" );
+	assert.equal( Globalize.dateFormatter({ skeleton: "GyMMMEdhmsSSS" })( date ), "Wed, Sep 15, 2010 AD, 5:35:07.369 PM" );


Let's not replace the above test, but add this as a new one.

rxaviers · 2017-08-15T00:23:49Z

test/functional/date/parse-date.js

-	date = startOf( date, "second" );
-	assertParseDate( assert, "Wed 5:35:07 PM", { skeleton: "Ehms" }, date );
+	date.setMilliseconds(734);
+	assertParseDate( assert, "Wed 5:35:07.734 PM", { skeleton: "EhmsSSS" }, date );


Same here...

rxaviers · 2017-08-15T00:25:02Z

Great @mattyork, one more test update and I think this is it.

Ref globalizejs#753

mattyork · 2017-08-16T16:22:40Z

Aye aye, tests updated

mattyork · 2017-09-07T20:07:06Z

Poke

mattyork · 2017-09-16T23:24:01Z

@rxaviers ?

rxaviers · 2017-09-22T04:50:40Z

Hi @mattyork, I didn't forget about you. Sorry about the delay. I will merge your PR. I hope the delay doesn't impact any of your work.

mattyork added 2 commits August 3, 2017 10:53

Date: Add fractional seconds support in formatters

b954458

Follows this algorithm: http://www.unicode.org/reports/tr35/tr35-dates.html#Matching_Skeletons Fixes globalizejs#753

Date: Add parse and compiler tests for fractional seconds

356d495

Ref globalizejs#753

rxaviers reviewed Aug 3, 2017

View reviewed changes

Date: address PR feedback

fbbf744

mattyork commented Aug 4, 2017

View reviewed changes

mattyork added 2 commits August 3, 2017 17:15

Date: fix lint errors

79e6f02

Date: address PR feedback regarding packaging

f37520f

Ref globalizejs#753

rxaviers reviewed Aug 15, 2017

View reviewed changes

Date: Address PR feedback, putting back old tests

c4fc7ae

Ref globalizejs#753

rxaviers added this to the 1.4.0 milestone Mar 9, 2018

rxaviers merged commit dceb8cd into globalizejs:master Jul 17, 2018

snyk-bot mentioned this pull request Nov 14, 2019

[Snyk] Upgrade globalize from 1.4.0 to 1.4.2 Matthelonianxl/framework#5

Open

X-oss-byte mentioned this pull request Sep 11, 2024

[Snyk] Upgrade: , , , cross-fetch, css-vars-ponyfill, globalize, intersection-observer, pepjs, requirejs-text, resize-observer-polyfill, tslib, web-animations-js, wicg-inert X-oss-byte/framework#20

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Date: Add fractional seconds support in formatters #763

Date: Add fractional seconds support in formatters #763

mattyork commented Aug 3, 2017

jsf-clabot commented Aug 3, 2017 •

edited

Loading

rxaviers Aug 3, 2017

mattyork Aug 3, 2017

rxaviers Aug 3, 2017

rxaviers Aug 3, 2017

rxaviers Aug 3, 2017

mattyork Aug 3, 2017

rxaviers Aug 3, 2017

mattyork Aug 3, 2017

mattyork Aug 4, 2017

rxaviers Aug 4, 2017

mattyork Aug 4, 2017

mattyork Aug 8, 2017

rxaviers Aug 14, 2017

mattyork Aug 14, 2017

mattyork commented Aug 14, 2017

rxaviers Aug 15, 2017

rxaviers Aug 15, 2017

rxaviers commented Aug 15, 2017

mattyork commented Aug 16, 2017

mattyork commented Sep 7, 2017

mattyork commented Sep 16, 2017

rxaviers commented Sep 22, 2017

Date: Add fractional seconds support in formatters #763

Date: Add fractional seconds support in formatters #763

Conversation

mattyork commented Aug 3, 2017

jsf-clabot commented Aug 3, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mattyork commented Aug 14, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rxaviers commented Aug 15, 2017

mattyork commented Aug 16, 2017

mattyork commented Sep 7, 2017

mattyork commented Sep 16, 2017

rxaviers commented Sep 22, 2017

jsf-clabot commented Aug 3, 2017 •

edited

Loading