Test bugs oct2018

[anba]

faed4fb

test/built-ins/RegExp/prototype/Symbol.matchAll/isregexp-called-once.js

• Add missing return value for flags (see my PR from last month)
• Correct order of operations

test/intl402/ListFormat/constructor/constructor/locales-valid.js
test/intl402/RelativeTimeFormat/constructor/constructor/locales-valid.js

• The "best fit" locale matcher is not required to return the default locale for the language tag "x-private", so restrict "x-private" to only try the "lookup" locale matcher.

test/intl402/Locale/constructor-options-numeric-undefined.js
test/intl402/Locale/constructor-options-numeric-valid.js
test/intl402/Locale/getters.js

• "true" is now omitted in Unicode extension keywords, per UTS 35.

test/intl402/Locale/constructor-options-numeric-undefined.js

• SameValue(undefined, "true") is false, cf. https://tc39.github.io/proposal-intl-locale/#sec-Intl.Locale step 36.a

test/intl402/Locale/constructor-parse-twice.js
test/intl402/Locale/extensions-grandfathered.js
test/intl402/Locale/extensions-private.js
test/intl402/Locale/getters-grandfathered.js
test/intl402/Locale/getters-privateuse.js

• Update to current Intl.Locale proposal
• Note: Using Unicode locale ID vs BCP 47 in our spec proposal-intl-locale#63 may remove grandfathered and privateuse-only language tag support.

test/intl402/Locale/constructor-tag-tostring.js

• Remove assert message because it applies ToString on an object which throws when ToString is applied on it.

test/intl402/NumberFormat/constructor-default-value.js

• __proto__ is Annex-B, replace with Object.create.

test/intl402/NumberFormat/prototype/formatToParts/default-parameter.js

• Same issue as in my PR last month.

test/intl402/NumberFormat/prototype/formatToParts/value-tonumber.js

• assert.sameValue can't be used to compare two objects.

test/language/module-code/dynamic-import/eval-self-once-script.js

• Ensure $DONE is only called from the first script and not from the (module-)script when it's dynamically imported.

d3e5b81

• "invalid" is not a supported value in UTS 35 (cf. https://tc39.github.io/ecma402/#sec-internal-slots)
• That means https://tc39.github.io/ecma402/#sec-resolvelocale removes it from the language tag (step 8.h in ResolveLocale).
• Which means the tests are correct and V8 needs to be fixed to correctly implement this part of the Intl spec.

[leobalter]

nice!