chromium/chromium-use-fromUTF8-for-UnicodeString-construction.patch

   1 From e58fa0ba66272c5f28828b15d06c7e42a9882b3b Mon Sep 17 00:00:00 2001
   2 From: Jungshik Shin <jshin@chromium.org>
   3 Date: Sat, 16 Dec 2017 04:19:27 +0000
   4 Subject: [PATCH] Use fromUTF8() for UnicodeString construction from UTF-8
   5
   6 Chrome's copy of ICU is built with U_CHARSET_IS_UTF8=1 so that |char *|
   7 buffer is treated as UTF-8 when constructing UnicodeString() regardless
   8 of the default encoding of the current locale on Linux or non-Unicode code
   9 page on Windows.
  10
  11 However, some Linux distros do not set U_CHARSET_IS_UTF=1 when building
  12 ICU and Chromium build with system_icu crashes when Chromium is run in
  13 non-UTF-8 locale (e.g. 'C').
  14
  15 To make Chromium work in a non-UTF-8 locale (which is pretty rare these
  16 days), use 'icu::UnicodeString::fromUTF8(StringPiece)' instead of
  17 'icu::UnicodeString(const char*)'.
  18
  19 Bug: 772655
  20 Test: components_unittests --gtest_filter=*IDN*
  21 Test: Chromium built with system_icu does not crash in C locale.
  22 Change-Id: I0daa284ec06b8e83814fc70eb8e9e5c96444ebfa
  23 Reviewed-on: https://chromium-review.googlesource.com/831247
  24 Reviewed-by: Peter Kasting <pkasting@chromium.org>
  25 Commit-Queue: Jungshik Shin <jshin@chromium.org>
  26 Cr-Commit-Position: refs/heads/master@{#524586}
  27 ---
  28  components/url_formatter/idn_spoof_checker.cc | 10 +++++-----
  29  1 file changed, 5 insertions(+), 5 deletions(-)
  30
  31 diff --git a/components/url_formatter/idn_spoof_checker.cc b/components/url_formatter/idn_spoof_checker.cc
  32 index a88c5e8f8331..aee748d8a4d5 100644
  33 --- a/components/url_formatter/idn_spoof_checker.cc
  34 +++ b/components/url_formatter/idn_spoof_checker.cc
  35 @@ -110,8 +110,8 @@ IDNSpoofChecker::IDNSpoofChecker() {
  36
  37    // These Cyrillic letters look like Latin. A domain label entirely made of
  38    // these letters is blocked as a simplified whole-script-spoofable.
  39 -  cyrillic_letters_latin_alike_ =
  40 -      icu::UnicodeSet(icu::UnicodeString("[асԁеһіјӏорԛѕԝхуъЬҽпгѵѡ]"), status);
  41 +  cyrillic_letters_latin_alike_ = icu::UnicodeSet(
  42 +      icu::UnicodeString::fromUTF8("[асԁеһіјӏорԛѕԝхуъЬҽпгѵѡ]"), status);
  43    cyrillic_letters_latin_alike_.freeze();
  44
  45    cyrillic_letters_ =
  46 @@ -141,8 +141,8 @@ IDNSpoofChecker::IDNSpoofChecker() {
  47    UParseError parse_error;
  48    diacritic_remover_.reset(icu::Transliterator::createFromRules(
  49        UNICODE_STRING_SIMPLE("DropAcc"),
  50 -      icu::UnicodeString("::NFD; ::[:Nonspacing Mark:] Remove; ::NFC;"
  51 -                         " ł > l; ø > o; đ > d;"),
  52 +      icu::UnicodeString::fromUTF8("::NFD; ::[:Nonspacing Mark:] Remove; ::NFC;"
  53 +                                   " ł > l; ø > o; đ > d;"),
  54        UTRANS_FORWARD, parse_error, status));
  55
  56    // Supplement the Unicode confusable list by the following mapping.
  57 @@ -158,7 +158,7 @@ IDNSpoofChecker::IDNSpoofChecker() {
  58    //   - U+0D1F (ട) => s
  59    extra_confusable_mapper_.reset(icu::Transliterator::createFromRules(
  60        UNICODE_STRING_SIMPLE("ExtraConf"),
  61 -      icu::UnicodeString(
  62 +      icu::UnicodeString::fromUTF8(
  63            "ӏ > l; [кĸκ] > k; п > n; [ƅь] > b; в > b; м > m; н > h; "
  64            "т > t; [шщ] > w; ട > s;"),
  65        UTRANS_FORWARD, parse_error, status));
  66 --
  67 2.15.1
  68