a46bc03591
In Chromium tree, ICU is rolled to 63.1. And, auto-roller will soon try to roll ICU in v8 to 63.1. Due to a nodejs trybot issue, autoroll needs a manual intervention. In the meantime, this CL will get rid of other blocking issues for ICU update. Prepare for the ICU roll by revising test/intl as following: * Line breaking loose mode is now supported in the Chromium's copy of ICU. Adjust the test expectation. * ICU's uloc_* can handle overlong locale ids. Drop tests that are not valid any more. Once ICU is rolled, a couple of TSAN-suppressed tests can be unsuppressed, but that has to be done in a separate CL. Bug: chromium:893196,v8:8272, v8:8110 Test: intl/*, test262/test402/* Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng Change-Id: I17f11457b61376b1e8d41bbbc951fa6cd3355a54 Reviewed-on: https://chromium-review.googlesource.com/c/1289369 Commit-Queue: Jungshik Shin <jshin@chromium.org> Reviewed-by: Michael Achenbach <machenbach@chromium.org> Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Cr-Commit-Position: refs/heads/master@{#57105}
58 lines
2.6 KiB
JavaScript
58 lines
2.6 KiB
JavaScript
// Copyright 2018 the V8 project authors. All rights reserved.
|
||
// Use of this source code is governed by a BSD-style license that can be
|
||
// found in the LICENSE file.
|
||
|
||
// Flags: --harmony-intl-segmenter
|
||
|
||
let breakCounts = {};
|
||
for (const locale of ["en", "fr", "ja", "zh", "ko"]) {
|
||
for (const lineBreakStyle of ["strict", "normal", "loose"]) {
|
||
const seg = new Intl.Segmenter(
|
||
[locale], {granularity: "line", lineBreakStyle: lineBreakStyle});
|
||
let opportunity = 0;
|
||
for (const text of [
|
||
// We know the following data caused different line break results between
|
||
// different modes.
|
||
// https://www.w3.org/TR/css-text-3/#propdef-line-break
|
||
// Japanese small kana or the Katakana-Hiragana prolonged sound mark
|
||
"あぁーぃーあーいーぁーぃー",
|
||
// hyphens:
|
||
// ‐ U+2010, – U+2013, 〜 U+301C, ゠ U+30A0
|
||
"ABC‐DEF–GHI〜JKL゠MNO",
|
||
// iteration marks:
|
||
// 々 U+3005, 〻 U+303B, ゝ U+309D, ゞ U+309E, ヽ U+30FD, ヾ U+30FE
|
||
"あ々あ〻あゝあゞあヽあヾあ",
|
||
// centered punctuation marks:
|
||
// ・ U+30FB, ： U+FF1A, ； U+FF1B, ･ U+FF65, ‼ U+203C
|
||
"ABC・DEF：GHI；JKL･MNO‼PQR",
|
||
// centered punctuation marks:
|
||
// ⁇ U+2047, ⁈ U+2048, ⁉ U+2049, ！ U+FF01, ？ U+FF1F
|
||
"ABC⁇DEF⁈GHI⁉JKL！MNO？PQR",
|
||
]) {
|
||
const iter = seg.segment(text);
|
||
while (!iter.following()) {
|
||
opportunity++;
|
||
}
|
||
}
|
||
breakCounts[locale + "-" + lineBreakStyle] = opportunity;
|
||
}
|
||
}
|
||
// In Japanese
|
||
// Just test the break count in loose mode is greater than normal mode.
|
||
assertTrue(breakCounts["ja-loose"] > breakCounts["ja-normal"]);
|
||
// and test the break count in normal mode is greater than strict mode.
|
||
assertTrue(breakCounts["ja-normal"] > breakCounts["ja-strict"]);
|
||
// In Chinese
|
||
// Just test the break count in loose mode is greater than normal mode.
|
||
assertTrue(breakCounts["zh-loose"] > breakCounts["zh-normal"]);
|
||
// and test the break count in normal mode is greater than strict mode.
|
||
assertTrue(breakCounts["zh-normal"] > breakCounts["zh-strict"]);
|
||
// In English, French and Korean
|
||
assertTrue(breakCounts["en-loose"] >= breakCounts["en-normal"]);
|
||
assertTrue(breakCounts["fr-loose"] >= breakCounts["fr-normal"]);
|
||
assertTrue(breakCounts["ko-loose"] >= breakCounts["ko-normal"]);
|
||
// and test the break count in normal mode is greater than strict mode.
|
||
assertTrue(breakCounts["en-normal"] > breakCounts["en-strict"]);
|
||
assertTrue(breakCounts["fr-normal"] > breakCounts["fr-strict"]);
|
||
assertTrue(breakCounts["ko-normal"] > breakCounts["ko-strict"]);
|