These two slipped through the cracks in #2097:
test/intl402/Intl/getCanonicalLocales/non-iana-canon.js
- Variant subtag canonicalisation is currrently not allowed.
test/intl402/Locale/getters.js
- Only the first "loc.caseFirst" test in this file was updated in #2097.
intl402/Intl/getCanonicalLocales/canonicalized-tags.js
- Sign languages are no longer canonicalised.
- Variant subtags are sorted alphabetically.
intl402/Intl/getCanonicalLocales/preferred-grandfathered.js
- Canonical form of "cel-gaulish" is "xtg-x-cel-gaulish".
intl402/Intl/getCanonicalLocales/preferred-variant.js
- Variant subtags are no longer canonicalised.
constructor-non-iana-canon.js
- Variant subtag canonicalisation is currently no longer present.
constructor-options-region-valid.js
- Digit region codes are now canonicalised.
constructor-tag.js
- Variant subtags are now sorted alphabetically.
likely-subtags-grandfathered.js
- "cmn" is now canonicalised to "zh".
This test started failing when updating to ICU 64, because ICU supports "zh"
and "zh-Hans-CN", but not explicitly also "zh-Hans", which is required for this
test to pass. The same kind of error is reproducible with ICU <64 when "Guru"
is added to the list of script codes in 'testIntl.js', because ICU supports
"pa-Guru-IN", but "pa-IN" isn't explicitly supported, too.
So, change this test to also check 'byFallback' to see if a locale is supported.
Drive-by change:
- Modernise the test to make it more readable how subtags are combined.
- Also add "419" to the list of region codes to cover the digit region syntax.
harness/testIntl.js
- Add now invalid tags to getInvalidLanguageTags, these tags were previously used in test files changed in this commit.
- Update isCanonicalizedStructurallyValidLanguageTag regular expressions.
test/intl402/Intl/getCanonicalLocales/canonicalized-tags.js
- Moved five now invalid tags to getInvalidLanguageTags function in testIntl.js
test/intl402/Intl/getCanonicalLocales/preferred-grandfathered.js
- All irregular grandfathered tags are invalid now
- Regular grandfathered with extlang subtags are now also invalid
- Regular grandfathered with variant-like subtags are still valid
test/intl402/Intl/getCanonicalLocales/weird-cases.js
- Revert changes from last commit
- "x-u-foo" is now invalid and was moved to getInvalidLanguageTags function
test/intl402/ListFormat/constructor/constructor/locales-valid.js
test/intl402/RelativeTimeFormat/constructor/constructor/locales-valid.js
test/intl402/Segmenter/constructor/constructor/locales-valid.js
- Irregular grandfathered and privateuse only are no longer valid language tags
test/intl402/language-tags-canonicalized.js
- Same changes as in test/intl402/Intl/getCanonicalLocales/canonicalized-tags.js
test/intl402/language-tags-invalid.js
- Invalid tags list in this file was a subset of getInvalidLanguageTags, so replaced with getInvalidLanguageTags to get more coverage
test/intl402/language-tags-valid.js
- Same changes as in test/intl402/Intl/getCanonicalLocales/canonicalized-tags.js
Main changes:
- "cmn" is now canonicalised to "zh"
- "und-x-" is prepended before grandfathered tags without modern replacements
- "und-" is prepended before privateuse-only language tags
The following lines expect each unit has property "-1" and "1"
```
const expected = unit in exceptions
? [exceptions[unit]["1"], exceptions[unit]["0"], exceptions[unit]["0"], exceptions[unit]["-1"]]
: [`in 1 ${unit}`, `in 0 ${unit}s`, `0 ${unit}s ago`, `1 ${unit} ago`];
```
Per https://tc39.github.io/ecma402/#sec-canonicalizelanguagetag
6.2.3 CanonicalizeLanguageTag ( locale )
The CanonicalizeLanguageTag abstract operation returns the canonical and case-regularized form of the locale argument (which must be a String value that is a structurally valid Unicode BCP 47 Locale Identifier as verified by the IsStructurallyValidLanguageTag abstract operation). A conforming implementation shall take the steps specified in the “BCP 47 Language Tag to Unicode BCP 47 Locale Identifier” algorithm, from Unicode Technical Standard #35 LDML § 3.3.1 BCP 47 Language Tag Conversion.
And
http://unicode.org/reports/tr35/#BCP_47_Language_Tag_Conversion
"
2. If the BCP 47 primary language subtag matches the type attribute of a languageAlias element in Supplemental Data, replace the language subtag with the replacement value.
If there are additional subtags in the replacement value, add them to the result, but only if there is no corresponding subtag already in the tag.
"
and
https://www.unicode.org/repos/cldr/trunk/common/supplemental/supplementalMetadata.xml
...
<languageAlias type="aar" replacement="aa" reason="overlong"/>
<!-- [Afar] -->
...
<languageAlias type="ces" replacement="cs" reason="overlong"/>
<!-- [Czech] -->
...
<languageAlias type="heb" replacement="he" reason="overlong"/>
<!-- [Hebrew] -->
