net/base/net_util_icu_unittest.cc - Issue 1258813002: Implement a new IDN display policy

Unified Diff: net/base/net_util_icu_unittest.cc

Issue 1258813002: Implement a new IDN display policy (Closed) Base URL: https://chromium.googlesource.com/chromium/src.git@master

Patch Set: add back languages to one more, update comments Created 5 years, 5 months ago

Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.

Jump to:

View side-by-side diff with in-line comments

Download patch

Index: net/base/net_util_icu_unittest.cc

diff --git a/net/base/net_util_icu_unittest.cc b/net/base/net_util_icu_unittest.cc

index f643426aa678ed48f36e96877adde254a26624b5..e58b0649d16454fab018f28dbb2cc6abe5149467 100644

--- a/net/base/net_util_icu_unittest.cc

+++ b/net/base/net_util_icu_unittest.cc

@@ -24,328 +24,120 @@ namespace net {

namespace {

const size_t kNpos = base::string16::npos;

-const char* const kLanguages[] = {

- "", "en", "zh-CN", "ja", "ko",

- "he", "ar", "ru", "el", "fr",

- "de", "pt", "sv", "th", "hi",

- "de,en", "el,en", "zh-TW,en", "ko,ja", "he,ru,en",

- "zh,ru,en"

-};

struct IDNTestCase {

const char* const input;

const wchar_t* unicode_output;

- const bool unicode_allowed[arraysize(kLanguages)];

+ const bool unicode_allowed;

};

// TODO(jungshik) This is just a random sample of languages and is far

// from exhaustive. We may have to generate all the combinations

// of languages (powerset of a set of all the languages).

const IDNTestCase idn_cases[] = {

- // No IDN

- {"www.google.com", L"www.google.com",

- {true, true, true, true, true,

- true, true, true, true, true,

- true}},

- {"www.google.com.", L"www.google.com.",

- {true, true, true, true, true,

- true, true, true, true, true,

- true}},

- {".", L".",

- {true, true, true, true, true,

- true, true, true, true, true,

- true}},

- {"", L"",

- {true, true, true, true, true,

- true, true, true, true, true,

- true}},

- // IDN

- // Hanzi (Traditional Chinese)

- {"xn--1lq90ic7f1rc.cn", L"\x5317\x4eac\x5927\x5b78.cn",

- {true, false, true, true, false,

- false, false, false, false, false,

- false, false, true, true, false,

- true}},

- // Hanzi ('video' in Simplified Chinese : will pass only in zh-CN,zh)

- {"xn--cy2a840a.com", L"\x89c6\x9891.com",

- {true, false, true, false, false,

- false, false, false, false, false,

- true}},

- // Hanzi + '123'

- {"www.xn--123-p18d.com", L"www.\x4e00" L"123.com",

- {true, false, true, true, false,

- false, false, false, false, false,

- false, false, true, true, false,

- true}},

- // Hanzi + Latin : U+56FD is simplified and is regarded

- // as not supported in zh-TW.

- {"www.xn--hello-9n1hm04c.com", L"www.hello\x4e2d\x56fd.com",

- {false, false, true, true, false,

- false, false, false, false, false,

- false, false, false, true, false,

- true}},

- // Kanji + Kana (Japanese)

- {"xn--l8jvb1ey91xtjb.jp", L"\x671d\x65e5\x3042\x3055\x3072.jp",

- {true, false, false, true, false,

- false, false, false, false, false,

- false, false, false, true, false,

- false}},

- // Katakana including U+30FC

- {"xn--tckm4i2e.jp", L"\x30b3\x30de\x30fc\x30b9.jp",

- {true, false, false, true, false,

- false, false, false, false, false,

- false, false, false, true, false,

- }},

- {"xn--3ck7a7g.jp", L"\u30ce\u30f3\u30bd.jp",

- {true, false, false, true, false,

- false, false, false, false, false,

- false, false, false, true, false,

- }},

- // Katakana + Latin (Japanese)

- // TODO(jungshik): Change 'false' in the first element to 'true'

- // after upgrading to ICU 4.2.1 to use new uspoof_* APIs instead

- // of our IsIDNComponentInSingleScript().

- {"xn--e-efusa1mzf.jp", L"e\x30b3\x30de\x30fc\x30b9.jp",

- {false, false, false, true, false,

- false, false, false, false, false,

- false, false, false, true, false,

- }},

- {"xn--3bkxe.jp", L"\x30c8\x309a.jp",

- {false, false, false, true, false,

- false, false, false, false, false,

- false, false, false, true, false,

- }},

- // Hangul (Korean)

- {"www.xn--or3b17p6jjc.kr", L"www.\xc804\xc790\xc815\xbd80.kr",

- {true, false, false, false, true,

- false, false, false, false, false,

- false, false, false, true, false,

- false}},

- // b<u-umlaut>cher (German)

- {"xn--bcher-kva.de", L"b\x00fc" L"cher.de",

- {true, false, false, false, false,

- false, false, false, false, true,

- true, false, false, false, false,

- false}},

- // a with diaeresis

- {"www.xn--frgbolaget-q5a.se", L"www.f\x00e4rgbolaget.se",

- {true, false, false, false, false,

- false, false, false, false, false,

- true, false, true, false, false,

- true, false, false, false, false,

- false}},

- // c-cedilla (French)

- {"www.xn--alliancefranaise-npb.fr", L"www.alliancefran\x00e7" L"aise.fr",

- {true, false, false, false, false,

- false, false, false, false, true,

- false, true, false, false, false,

- false, false, false, false, false,

- false}},

- // caf'e with acute accent' (French)

- {"xn--caf-dma.fr", L"caf\x00e9.fr",

- {true, false, false, false, false,

- false, false, false, false, true,

- false, true, true, false, false,

- false, false, false, false, false,

- false}},

- // c-cedillla and a with tilde (Portuguese)

- {"xn--poema-9qae5a.com.br", L"p\x00e3oema\x00e7\x00e3.com.br",

- {true, false, false, false, false,

- false, false, false, false, false,

- false, true, false, false, false,

- false, false, false, false, false,

- false}},

- // s with caron

- {"xn--achy-f6a.com", L"\x0161" L"achy.com",

- {true, false, false, false, false,

- false, false, false, false, false,

- false}},

- // TODO(jungshik) : Add examples with Cyrillic letters

- // only used in some languages written in Cyrillic.

- // Eutopia (Greek)

- {"xn--kxae4bafwg.gr", L"\x03bf\x03c5\x03c4\x03bf\x03c0\x03af\x03b1.gr",

- {true, false, false, false, false,

- false, false, false, true, false,

- false, false, false, false, false,

- false, true, false, false, false,

- false}},

- // Eutopia + 123 (Greek)

- {"xn---123-pldm0haj2bk.gr",

- L"\x03bf\x03c5\x03c4\x03bf\x03c0\x03af\x03b1-123.gr",

- {true, false, false, false, false,

- false, false, false, true, false,

- false, false, false, false, false,

- false, true, false, false, false,

- false}},

- // Cyrillic (Russian)

- {"xn--n1aeec9b.ru", L"\x0442\x043e\x0440\x0442\x044b.ru",

- {true, false, false, false, false,

- false, false, true, false, false,

- false, false, false, false, false,

- false, false, false, false, true,

- true}},

- // Cyrillic + 123 (Russian)

- {"xn---123-45dmmc5f.ru", L"\x0442\x043e\x0440\x0442\x044b-123.ru",

- {true, false, false, false, false,

- false, false, true, false, false,

- false, false, false, false, false,

- false, false, false, false, true,

- true}},

- // Arabic

- {"xn--mgba1fmg.ar", L"\x0627\x0641\x0644\x0627\x0645.ar",

- {true, false, false, false, false,

- false, true, false, false, false,

- false, false, false, false, false,

- false}},

- // Hebrew

- {"xn--4dbib.he", L"\x05d5\x05d0\x05d4.he",

- {true, false, false, false, false,

- true, false, false, false, false,

- false, false, false, false, false,

- false, false, false, false, true,

- false}},

- // Thai

- {"xn--12c2cc4ag3b4ccu.th",

- L"\x0e2a\x0e32\x0e22\x0e01\x0e32\x0e23\x0e1a\x0e34\x0e19.th",

- {true, false, false, false, false,

- false, false, false, false, false,

- false, false, false, true, false,

- false, false, false, false, false,

- false}},

- // Devangari (Hindi)

- {"www.xn--l1b6a9e1b7c.in", L"www.\x0905\x0915\x094b\x0932\x093e.in",

- {true, false, false, false, false,

- false, false, false, false, false,

- false, false, false, false, true,

- false, false, false, false, false,

- false}},

- // Invalid IDN

- {"xn--hello?world.com", NULL,

- {false, false, false, false, false,

- false, false, false, false, false,

- false}},

- // Unsafe IDNs

- // "payp<alpha>l.com"

- {"www.xn--paypl-g9d.com", L"payp\x03b1l.com",

- {false, false, false, false, false,

- false, false, false, false, false,

- false}},

- // google.gr with Greek omicron and epsilon

- {"xn--ggl-6xc1ca.gr", L"g\x03bf\x03bfgl\x03b5.gr",

- {false, false, false, false, false,

- false, false, false, false, false,

- false}},

- // google.ru with Cyrillic o

- {"xn--ggl-tdd6ba.ru", L"g\x043e\x043egl\x0435.ru",

- {false, false, false, false, false,

- false, false, false, false, false,

- false}},

- // h<e with acute>llo<China in Han>.cn

- {"xn--hllo-bpa7979ih5m.cn", L"h\x00e9llo\x4e2d\x56fd.cn",

- {false, false, false, false, false,

- false, false, false, false, false,

- false}},

- // <Greek rho><Cyrillic a><Cyrillic u>.ru

- {"xn--2xa6t2b.ru", L"\x03c1\x0430\x0443.ru",

- {false, false, false, false, false,

- false, false, false, false, false,

- false}},

- // One that's really long that will force a buffer realloc

- {"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"

- "aaaaaaa",

- L"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"

- L"aaaaaaaa",

- {true, true, true, true, true,

- true, true, true, true, true,

- true}},

- // Test cases for characters we blacklisted although allowed in IDN.

- // Embedded spaces will be turned to %20 in the display.

- // TODO(jungshik): We need to have more cases. This is a typical

- // data-driven trap. The following test cases need to be separated

- // and tested only for a couple of languages.

- {"xn--osd3820f24c.kr", L"\xac00\xb098\x115f.kr",

- {false, false, false, false, false,

- false, false, false, false, false,

- false}},

- {"www.xn--google-ho0coa.com", L"www.\x2039google\x203a.com",

- {false, false, false, false, false,

- false, false, false, false, false,

- }},

- {"google.xn--comabc-k8d", L"google.com\x0338" L"abc",

- {false, false, false, false, false,

- false, false, false, false, false,

- }},

- {"google.xn--com-oh4ba.evil.jp", L"google.com\x309a\x309a.evil.jp",

- {false, false, false, false, false,

- false, false, false, false, false,

- }},

- {"google.xn--comevil-v04f.jp", L"google.com\x30ce" L"evil.jp",

- {false, false, false, false, false,

- false, false, false, false, false,

- }},

- // Padlock icon spoof.

- {"xn--google-hj64e", L"\U0001f512google.com",

- {false, false, false, false, false,

- false, false, false, false, false,

- }},

- // Ensure that blacklisting "\xd83d\xdd12" did not inadvertently blacklist

- // all strings with the surrogate '\xdd12'.

- {"xn--fk9c.com", L"\U00010912.com",

- {true, false, false, false, false,

- false, false, false, false, false,

- }},

+ // No IDN

+ {"www.google.com", L"www.google.com", true},

+ {"www.google.com.", L"www.google.com.", true},

+ {".", L".", true},

+ {"", L"", true},

+ // IDN

+ // Hanzi (Traditional Chinese)

+ {"xn--1lq90ic7f1rc.cn", L"\x5317\x4eac\x5927\x5b78.cn", true},

+ // Hanzi ('video' in Simplified Chinese

+ {"xn--cy2a840a.com", L"\x89c6\x9891.com", true},

+ // Hanzi + '123'

+ {"www.xn--123-p18d.com",

+ L"www.\x4e00"

+ L"123.com",

+ true},

+ // Hanzi + Latin : U+56FD is simplified

+ {"www.xn--hello-9n1hm04c.com", L"www.hello\x4e2d\x56fd.com", true},

+ // Kanji + Kana (Japanese)

+ {"xn--l8jvb1ey91xtjb.jp", L"\x671d\x65e5\x3042\x3055\x3072.jp", true},

+ // Katakana including U+30FC

+ {"xn--tckm4i2e.jp", L"\x30b3\x30de\x30fc\x30b9.jp", true},

+ {"xn--3ck7a7g.jp", L"\u30ce\u30f3\u30bd.jp", true},

+ // Katakana + Latin (Japanese)

+ {"xn--e-efusa1mzf.jp", L"e\x30b3\x30de\x30fc\x30b9.jp", true},

+ {"xn--3bkxe.jp", L"\x30c8\x309a.jp", true},

+ // Hangul (Korean)

+ {"www.xn--or3b17p6jjc.kr", L"www.\xc804\xc790\xc815\xbd80.kr", true},

+ // b<u-umlaut>cher (German)

+ {"xn--bcher-kva.de",

+ L"b\x00fc"

+ L"cher.de",

+ true},

+ // a with diaeresis

+ {"www.xn--frgbolaget-q5a.se", L"www.f\x00e4rgbolaget.se", true},

+ // c-cedilla (French)

+ {"www.xn--alliancefranaise-npb.fr",

+ L"www.alliancefran\x00e7"

+ L"aise.fr",

+ true},

+ // caf'e with acute accent' (French)

+ {"xn--caf-dma.fr", L"caf\x00e9.fr", true},

+ // c-cedillla and a with tilde (Portuguese)

+ {"xn--poema-9qae5a.com.br", L"p\x00e3oema\x00e7\x00e3.com.br", true},

+ // s with caron

+ {"xn--achy-f6a.com",

+ L"\x0161"

+ L"achy.com",

+ true},

+ {"xn--kxae4bafwg.gr", L"\x03bf\x03c5\x03c4\x03bf\x03c0\x03af\x03b1.gr",

+ true},

+ // Eutopia + 123 (Greek)

+ {"xn---123-pldm0haj2bk.gr",

+ L"\x03bf\x03c5\x03c4\x03bf\x03c0\x03af\x03b1-123.gr", true},

+ // Cyrillic (Russian)

+ {"xn--n1aeec9b.ru", L"\x0442\x043e\x0440\x0442\x044b.ru", true},

+ // Cyrillic + 123 (Russian)

+ {"xn---123-45dmmc5f.ru", L"\x0442\x043e\x0440\x0442\x044b-123.ru", true},

+ // Arabic

+ {"xn--mgba1fmg.eg", L"\x0627\x0641\x0644\x0627\x0645.eg", true},

+ // Hebrew

+ {"xn--4dbib.he", L"\x05d5\x05d0\x05d4.he", true},

+ // Thai

+ {"xn--12c2cc4ag3b4ccu.th",

+ L"\x0e2a\x0e32\x0e22\x0e01\x0e32\x0e23\x0e1a\x0e34\x0e19.th", true},

+ // Devangari (Hindi)

+ {"www.xn--l1b6a9e1b7c.in", L"www.\x0905\x0915\x094b\x0932\x093e.in", true},

+ // Invalid IDN

+ {"xn--hello?world.com", NULL, false},

+ // Unsafe IDNs

+ // "payp<alpha>l.com"

+ {"www.xn--paypl-g9d.com", L"payp\x03b1l.com", false},

+ // google.gr with Greek omicron and epsilon

+ {"xn--ggl-6xc1ca.gr", L"g\x03bf\x03bfgl\x03b5.gr", false},

+ // google.ru with Cyrillic o

+ {"xn--ggl-tdd6ba.ru", L"g\x043e\x043egl\x0435.ru", false},

+ // h<e with acute>llo<China in Han>.cn

+ {"xn--hllo-bpa7979ih5m.cn", L"h\x00e9llo\x4e2d\x56fd.cn", false},

+ // <Greek rho><Cyrillic a><Cyrillic u>.ru

+ {"xn--2xa6t2b.ru", L"\x03c1\x0430\x0443.ru", false},

+ // One that's really long that will force a buffer realloc

+ {"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"

+ "aaaaaaa",

+ L"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"

+ L"aaaaaaaa",

+ true},

+ {"xn--osd3820f24c.kr", L"\xac00\xb098\x115f.kr", false},

+ {"www.xn--google-ho0coa.com", L"www.\x2039google\x203a.com", false},

+ // Test cases for characters we blacklisted although allowed in IDN.

+ {"google.xn--comabc-k8d",

+ L"google.com\x0338"

+ L"abc",

+ false},

+ {"google.xn--com-oh4ba.evil.jp", L"google.com\x309a\x309a.evil.jp", false},

+ {"google.xn--comevil-v04f.jp",

+ L"google.com\x30ce"

+ L"evil.jp",

+ false},

+ // Padlock icon spoof.

+ {"xn--google-hj64e", L"\U0001f512google.com", false},

+ // Ensure that blacklisting "\xd83d\xdd12" did not inadvertently blacklist

+ // all strings with the surrogate '\xdd12'.

+ {"xn--fk9c.com", L"\U00010912.com", false},

+ {"xn--g6h.com", L"\x2665.com", true},

+ {"xn--2ci.com", L"\x272a.com", true},

#if 0

// These two cases are special. We need a separate test.

// U+3000 and U+3002 are normalized to ASCII space and dot.

@@ -372,24 +164,12 @@ struct AdjustOffsetCase {

struct UrlTestData {

const char* const description;

const char* const input;

- const char* const languages;

FormatUrlTypes format_types;

UnescapeRule::Type escape_rules;

const wchar_t* output; // Use |wchar_t| to handle Unicode constants easily.

size_t prefix_len;

};

-// A helper for IDN*{Fast,Slow}.

-// Append "::<language list>" to |expected| and |actual| to make it

-// easy to tell which sub-case fails without debugging.

-void AppendLanguagesToOutputs(const char* languages,

- base::string16* expected,

- base::string16* actual) {

- base::string16 to_append = ASCIIToUTF16("::") + ASCIIToUTF16(languages);

- expected->append(to_append);

- actual->append(to_append);

// A pair of helpers for the FormatUrlWithOffsets() test.

void VerboseExpect(size_t expected,

size_t actual,

@@ -412,8 +192,8 @@ void CheckAdjustedOffsets(const std::string& url_string,

offsets.push_back(i);

offsets.push_back(500000); // Something larger than any input length.

offsets.push_back(std::string::npos);

- base::string16 formatted_url = FormatUrlWithOffsets(url, languages,

- format_types, unescape_rules, NULL, NULL, &offsets);

+ base::string16 formatted_url = FormatUrlWithOffsets(

+ url, std::string(), format_types, unescape_rules, NULL, NULL, &offsets);

for (size_t i = 0; i < url_length; ++i)

VerboseExpect(output_offsets[i], offsets[i], url_string, i, formatted_url);

VerboseExpect(formatted_url.length(), offsets[url_length], url_string,

@@ -426,51 +206,14 @@ void CheckAdjustedOffsets(const std::string& url_string,

} // anonymous namespace

-TEST(NetUtilTest, IDNToUnicodeFast) {

+TEST(NetUtilTest, IDNToUnicode) {

for (size_t i = 0; i < arraysize(idn_cases); i++) {

- for (size_t j = 0; j < arraysize(kLanguages); j++) {

- // ja || zh-TW,en || ko,ja -> IDNToUnicodeSlow

- if (j == 3 || j == 17 || j == 18)

- continue;

- base::string16 output(IDNToUnicode(idn_cases[i].input, kLanguages[j]));

- base::string16 expected(idn_cases[i].unicode_allowed[j] ?

- WideToUTF16(idn_cases[i].unicode_output) :

- ASCIIToUTF16(idn_cases[i].input));

- AppendLanguagesToOutputs(kLanguages[j], &expected, &output);

- EXPECT_EQ(expected, output) << "input: \"" << idn_cases[i].input

- << "\", languages: \"" << kLanguages[j]

- << "\"";

- }

-TEST(NetUtilTest, IDNToUnicodeSlow) {

- for (size_t i = 0; i < arraysize(idn_cases); i++) {

- for (size_t j = 0; j < arraysize(kLanguages); j++) {

- // !(ja || zh-TW,en || ko,ja) -> IDNToUnicodeFast

- if (!(j == 3 || j == 17 || j == 18))

- continue;

- base::string16 output(IDNToUnicode(idn_cases[i].input, kLanguages[j]));

- base::string16 expected(idn_cases[i].unicode_allowed[j] ?

- WideToUTF16(idn_cases[i].unicode_output) :

- ASCIIToUTF16(idn_cases[i].input));

- AppendLanguagesToOutputs(kLanguages[j], &expected, &output);

- EXPECT_EQ(expected, output) << "input: \"" << idn_cases[i].input

- << "\", languages: \"" << kLanguages[j]

- << "\"";

- }

-// ulocdata_getExemplarSet may fail with some locales (currently bn, gu, and

-// te), which was causing a crash (See http://crbug.com/510551). This may be an

-// icu bug, but regardless, that should not cause a crash.

-TEST(NetUtilTest, IDNToUnicodeNeverCrashes) {

- for (char c1 = 'a'; c1 <= 'z'; c1++) {

- for (char c2 = 'a'; c2 <= 'z'; c2++) {

- std::string lang = base::StringPrintf("%c%c", c1, c2);

- base::string16 output(IDNToUnicode("xn--74h", lang));

- }

+ base::string16 output(IDNToUnicode(idn_cases[i].input, std::string()));

+ base::string16 expected(idn_cases[i].unicode_allowed

+ ? WideToUTF16(idn_cases[i].unicode_output)

+ : ASCIIToUTF16(idn_cases[i].input));

+ EXPECT_EQ(expected, output) << "input # " << i << ": \""

+ << idn_cases[i].input << "\"";

}

@@ -554,181 +297,149 @@ TEST(NetUtilTest, GetDirectoryListingEntry) {

TEST(NetUtilTest, FormatUrl) {

FormatUrlTypes default_format_type = kFormatUrlOmitUsernamePassword;

const UrlTestData tests[] = {

- {"Empty URL", "", "", default_format_type, UnescapeRule::NORMAL, L"", 0},

- {"Simple URL",

- "http://www.google.com/", "", default_format_type, UnescapeRule::NORMAL,

- L"http://www.google.com/", 7},

- {"With a port number and a reference",

- "http://www.google.com:8080/#\xE3\x82\xB0", "", default_format_type,

- UnescapeRule::NORMAL,

- L"http://www.google.com:8080/#\x30B0", 7},

- // -------- IDN tests --------

- {"Japanese IDN with ja",

- "http://xn--l8jvb1ey91xtjb.jp", "ja", default_format_type,

- UnescapeRule::NORMAL, L"http://\x671d\x65e5\x3042\x3055\x3072.jp/", 7},

- {"Japanese IDN with en",

- "http://xn--l8jvb1ey91xtjb.jp", "en", default_format_type,

- UnescapeRule::NORMAL, L"http://xn--l8jvb1ey91xtjb.jp/", 7},

- {"Japanese IDN without any languages",

- "http://xn--l8jvb1ey91xtjb.jp", "", default_format_type,

- UnescapeRule::NORMAL,

- // Single script is safe for empty languages.

- L"http://\x671d\x65e5\x3042\x3055\x3072.jp/", 7},

- {"mailto: with Japanese IDN",

- "mailto:foo@xn--l8jvb1ey91xtjb.jp", "ja", default_format_type,

- UnescapeRule::NORMAL,

- // GURL doesn't assume an email address's domain part as a host name.

- L"mailto:foo@xn--l8jvb1ey91xtjb.jp", 7},

- {"file: with Japanese IDN",

- "file://xn--l8jvb1ey91xtjb.jp/config.sys", "ja", default_format_type,

- UnescapeRule::NORMAL,

- L"file://\x671d\x65e5\x3042\x3055\x3072.jp/config.sys", 7},

- {"ftp: with Japanese IDN",

- "ftp://xn--l8jvb1ey91xtjb.jp/config.sys", "ja", default_format_type,

- UnescapeRule::NORMAL,

- L"ftp://\x671d\x65e5\x3042\x3055\x3072.jp/config.sys", 6},

- // -------- omit_username_password flag tests --------

- {"With username and password, omit_username_password=false",

- "http://user:passwd@example.com/foo", "",

- kFormatUrlOmitNothing, UnescapeRule::NORMAL,

- L"http://user:passwd@example.com/foo", 19},

- {"With username and password, omit_username_password=true",

- "http://user:passwd@example.com/foo", "", default_format_type,

- UnescapeRule::NORMAL, L"http://example.com/foo", 7},

- {"With username and no password",

- "http://user@example.com/foo", "", default_format_type,

- UnescapeRule::NORMAL, L"http://example.com/foo", 7},

- {"Just '@' without username and password",

- "http://@example.com/foo", "", default_format_type, UnescapeRule::NORMAL,

- L"http://example.com/foo", 7},

- // GURL doesn't think local-part of an email address is username for URL.

- {"mailto:, omit_username_password=true",

- "mailto:foo@example.com", "", default_format_type, UnescapeRule::NORMAL,

- L"mailto:foo@example.com", 7},

- // -------- unescape flag tests --------

- {"Do not unescape",

- "http://%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB.jp/"

- "%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"

- "?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB", "en", default_format_type,

- UnescapeRule::NONE,

- // GURL parses %-encoded hostnames into Punycode.

- L"http://xn--qcka1pmc.jp/%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"

- L"?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB", 7},

- {"Unescape normally",

- "http://%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB.jp/"

- "%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"

- "?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB", "en", default_format_type,

- UnescapeRule::NORMAL,

- L"http://xn--qcka1pmc.jp/\x30B0\x30FC\x30B0\x30EB"

- L"?q=\x30B0\x30FC\x30B0\x30EB", 7},

- {"Unescape normally with BiDi control character",

- "http://example.com/%E2%80%AEabc?q=%E2%80%8Fxy", "en", default_format_type,

- UnescapeRule::NORMAL, L"http://example.com/%E2%80%AEabc?q=%E2%80%8Fxy", 7},

- {"Unescape normally including unescape spaces",

- "http://www.google.com/search?q=Hello%20World", "en", default_format_type,

- UnescapeRule::SPACES, L"http://www.google.com/search?q=Hello World", 7},

- /*

- {"unescape=true with some special characters",

- "http://user%3A:%40passwd@example.com/foo%3Fbar?q=b%26z", "",

- kFormatUrlOmitNothing, UnescapeRule::NORMAL,

- L"http://user%3A:%40passwd@example.com/foo%3Fbar?q=b%26z", 25},

- */

- // Disabled: the resultant URL becomes "...user%253A:%2540passwd...".

- // -------- omit http: --------

- {"omit http with user name",

- "http://user@example.com/foo", "", kFormatUrlOmitAll,

- UnescapeRule::NORMAL, L"example.com/foo", 0},

- {"omit http",

- "http://www.google.com/", "en", kFormatUrlOmitHTTP,

- UnescapeRule::NORMAL, L"www.google.com/",

- 0},

- {"omit http with https",

- "https://www.google.com/", "en", kFormatUrlOmitHTTP,

- UnescapeRule::NORMAL, L"https://www.google.com/",

- 8},

- {"omit http starts with ftp.",

- "http://ftp.google.com/", "en", kFormatUrlOmitHTTP,

- UnescapeRule::NORMAL, L"http://ftp.google.com/",

- 7},

- // -------- omit trailing slash on bare hostname --------

- {"omit slash when it's the entire path",

- "http://www.google.com/", "en",

- kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

- L"http://www.google.com", 7},

- {"omit slash when there's a ref",

- "http://www.google.com/#ref", "en",

- kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

- L"http://www.google.com/#ref", 7},

- {"omit slash when there's a query",

- "http://www.google.com/?", "en",

- kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

- L"http://www.google.com/?", 7},

- {"omit slash when it's not the entire path",

- "http://www.google.com/foo", "en",

- kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

- L"http://www.google.com/foo", 7},

- {"omit slash for nonstandard URLs",

- "data:/", "en", kFormatUrlOmitTrailingSlashOnBareHostname,

- UnescapeRule::NORMAL, L"data:/", 5},

- {"omit slash for file URLs",

- "file:///", "en", kFormatUrlOmitTrailingSlashOnBareHostname,

- UnescapeRule::NORMAL, L"file:///", 7},

- // -------- view-source: --------

- {"view-source",

- "view-source:http://xn--qcka1pmc.jp/", "ja", default_format_type,

- UnescapeRule::NORMAL, L"view-source:http://\x30B0\x30FC\x30B0\x30EB.jp/",

- 19},

- {"view-source of view-source",

- "view-source:view-source:http://xn--qcka1pmc.jp/", "ja",

- default_format_type, UnescapeRule::NORMAL,

- L"view-source:view-source:http://xn--qcka1pmc.jp/", 12},

- // view-source should omit http and trailing slash where non-view-source

- // would.

- {"view-source omit http",

- "view-source:http://a.b/c", "en", kFormatUrlOmitAll,

- UnescapeRule::NORMAL, L"view-source:a.b/c",

- 12},

- {"view-source omit http starts with ftp.",

- "view-source:http://ftp.b/c", "en", kFormatUrlOmitAll,

- UnescapeRule::NORMAL, L"view-source:http://ftp.b/c",

- 19},

- {"view-source omit slash when it's the entire path",

- "view-source:http://a.b/", "en", kFormatUrlOmitAll,

- UnescapeRule::NORMAL, L"view-source:a.b",

- 12},

+ {"Empty URL", "", default_format_type, UnescapeRule::NORMAL, L"", 0},

+ {"Simple URL", "http://www.google.com/", default_format_type,

+ UnescapeRule::NORMAL, L"http://www.google.com/", 7},

+ {"With a port number and a reference",

+ "http://www.google.com:8080/#\xE3\x82\xB0", default_format_type,

+ UnescapeRule::NORMAL, L"http://www.google.com:8080/#\x30B0", 7},

+ // -------- IDN tests --------

+ {"Japanese IDN", "http://xn--l8jvb1ey91xtjb.jp", default_format_type,

+ UnescapeRule::NORMAL, L"http://\x671d\x65e5\x3042\x3055\x3072.jp/", 7},

+ {"mailto: with Japanese IDN", "mailto:foo@xn--l8jvb1ey91xtjb.jp",

+ default_format_type, UnescapeRule::NORMAL,

+ // GURL doesn't assume an email address's domain part as a host name.

+ L"mailto:foo@xn--l8jvb1ey91xtjb.jp", 7},

+ {"file: with Japanese IDN", "file://xn--l8jvb1ey91xtjb.jp/config.sys",

+ default_format_type, UnescapeRule::NORMAL,

+ L"file://\x671d\x65e5\x3042\x3055\x3072.jp/config.sys", 7},

+ {"ftp: with Japanese IDN", "ftp://xn--l8jvb1ey91xtjb.jp/config.sys",

+ default_format_type, UnescapeRule::NORMAL,

+ L"ftp://\x671d\x65e5\x3042\x3055\x3072.jp/config.sys", 6},

+ // -------- omit_username_password flag tests --------

+ {"With username and password, omit_username_password=false",

+ "http://user:passwd@example.com/foo", kFormatUrlOmitNothing,

+ UnescapeRule::NORMAL, L"http://user:passwd@example.com/foo", 19},

+ {"With username and password, omit_username_password=true",

+ "http://user:passwd@example.com/foo", default_format_type,

+ UnescapeRule::NORMAL, L"http://example.com/foo", 7},

+ {"With username and no password", "http://user@example.com/foo",

+ default_format_type, UnescapeRule::NORMAL, L"http://example.com/foo", 7},

+ {"Just '@' without username and password", "http://@example.com/foo",

+ default_format_type, UnescapeRule::NORMAL, L"http://example.com/foo", 7},

+ // GURL doesn't think local-part of an email address is username for URL.

+ {"mailto:, omit_username_password=true", "mailto:foo@example.com",

+ default_format_type, UnescapeRule::NORMAL, L"mailto:foo@example.com", 7},

+ // -------- unescape flag tests --------

+ {"Do not unescape",

+ "http://%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB.jp/"

+ "%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"

+ "?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB",

+ default_format_type, UnescapeRule::NONE,

+ // GURL parses %-encoded hostnames into Punycode.

+ L"http://\x30B0\x30FC\x30B0\x30EB.jp/"

+ L"%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"

+ L"?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB",

+ 7},

+ {"Unescape normally",

+ "http://%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB.jp/"

+ "%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"

+ "?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB",

+ default_format_type, UnescapeRule::NORMAL,

+ L"http://\x30B0\x30FC\x30B0\x30EB.jp/\x30B0\x30FC\x30B0\x30EB"

+ L"?q=\x30B0\x30FC\x30B0\x30EB",

+ 7},

+ {"Unescape normally with BiDi control character",

+ "http://example.com/%E2%80%AEabc?q=%E2%80%8Fxy", default_format_type,

+ UnescapeRule::NORMAL, L"http://example.com/%E2%80%AEabc?q=%E2%80%8Fxy",

+ 7},

+ {"Unescape normally including unescape spaces",

+ "http://www.google.com/search?q=Hello%20World", default_format_type,

+ UnescapeRule::SPACES, L"http://www.google.com/search?q=Hello World", 7},

+ /*

+ {"unescape=true with some special characters",

+ "http://user%3A:%40passwd@example.com/foo%3Fbar?q=b%26z", "",

+ kFormatUrlOmitNothing, UnescapeRule::NORMAL,

+ L"http://user%3A:%40passwd@example.com/foo%3Fbar?q=b%26z", 25},

+ */

+ // Disabled: the resultant URL becomes "...user%253A:%2540passwd...".

+ // -------- omit http: --------

+ {"omit http with user name", "http://user@example.com/foo",

+ kFormatUrlOmitAll, UnescapeRule::NORMAL, L"example.com/foo", 0},

+ {"omit http", "http://www.google.com/", kFormatUrlOmitHTTP,

+ UnescapeRule::NORMAL, L"www.google.com/", 0},

+ {"omit http with https", "https://www.google.com/", kFormatUrlOmitHTTP,

+ UnescapeRule::NORMAL, L"https://www.google.com/", 8},

+ {"omit http starts with ftp.", "http://ftp.google.com/",

+ kFormatUrlOmitHTTP, UnescapeRule::NORMAL, L"http://ftp.google.com/", 7},

+ // -------- omit trailing slash on bare hostname --------

+ {"omit slash when it's the entire path", "http://www.google.com/",

+ kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

+ L"http://www.google.com", 7},

+ {"omit slash when there's a ref", "http://www.google.com/#ref",

+ kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

+ L"http://www.google.com/#ref", 7},

+ {"omit slash when there's a query", "http://www.google.com/?",

+ kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

+ L"http://www.google.com/?", 7},

+ {"omit slash when it's not the entire path", "http://www.google.com/foo",

+ kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

+ L"http://www.google.com/foo", 7},

+ {"omit slash for nonstandard URLs", "data:/",

+ kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

+ L"data:/", 5},

+ {"omit slash for file URLs", "file:///",

+ kFormatUrlOmitTrailingSlashOnBareHostname, UnescapeRule::NORMAL,

+ L"file:///", 7},

+ // -------- view-source: --------

+ {"view-source", "view-source:http://xn--qcka1pmc.jp/",

+ default_format_type, UnescapeRule::NORMAL,

+ L"view-source:http://\x30B0\x30FC\x30B0\x30EB.jp/", 19},

+ {"view-source of view-source",

+ "view-source:view-source:http://xn--qcka1pmc.jp/", default_format_type,

+ UnescapeRule::NORMAL, L"view-source:view-source:http://xn--qcka1pmc.jp/",

+ 12},

+ // view-source should omit http and trailing slash where non-view-source

+ // would.

+ {"view-source omit http", "view-source:http://a.b/c", kFormatUrlOmitAll,

+ UnescapeRule::NORMAL, L"view-source:a.b/c", 12},

+ {"view-source omit http starts with ftp.", "view-source:http://ftp.b/c",

+ kFormatUrlOmitAll, UnescapeRule::NORMAL, L"view-source:http://ftp.b/c",

+ 19},

+ {"view-source omit slash when it's the entire path",

+ "view-source:http://a.b/", kFormatUrlOmitAll, UnescapeRule::NORMAL,

+ L"view-source:a.b", 12},

};

for (size_t i = 0; i < arraysize(tests); ++i) {

size_t prefix_len;

- base::string16 formatted = FormatUrl(

- GURL(tests[i].input), tests[i].languages, tests[i].format_types,

- tests[i].escape_rules, NULL, &prefix_len, NULL);

+ base::string16 formatted =

+ FormatUrl(GURL(tests[i].input), std::string(), tests[i].format_types,

+ tests[i].escape_rules, NULL, &prefix_len, NULL);

EXPECT_EQ(WideToUTF16(tests[i].output), formatted) << tests[i].description;

EXPECT_EQ(tests[i].prefix_len, prefix_len) << tests[i].description;

}

@@ -737,11 +448,11 @@ TEST(NetUtilTest, FormatUrl) {

TEST(NetUtilTest, FormatUrlParsed) {

// No unescape case.

url::Parsed parsed;

- base::string16 formatted = FormatUrl(

- GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"

- "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),

- "ja", kFormatUrlOmitNothing, UnescapeRule::NONE, &parsed, NULL,

- NULL);

+ base::string16 formatted =

+ FormatUrl(GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"

+ "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),

+ std::string(), kFormatUrlOmitNothing, UnescapeRule::NONE,

+ &parsed, NULL, NULL);

EXPECT_EQ(WideToUTF16(

L"http://%E3%82%B0:%E3%83%BC@\x30B0\x30FC\x30B0\x30EB.jp:8080"

L"/%E3%82%B0/?q=%E3%82%B0#\x30B0"), formatted);

@@ -761,11 +472,11 @@ TEST(NetUtilTest, FormatUrlParsed) {

formatted.substr(parsed.ref.begin, parsed.ref.len));

// Unescape case.

- formatted = FormatUrl(

- GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"

- "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),

- "ja", kFormatUrlOmitNothing, UnescapeRule::NORMAL, &parsed, NULL,

- NULL);

+ formatted =

+ FormatUrl(GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"

+ "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),

+ std::string(), kFormatUrlOmitNothing, UnescapeRule::NORMAL,

+ &parsed, NULL, NULL);

EXPECT_EQ(WideToUTF16(L"http://\x30B0:\x30FC@\x30B0\x30FC\x30B0\x30EB.jp:8080"

L"/\x30B0/?q=\x30B0#\x30B0"), formatted);

EXPECT_EQ(WideToUTF16(L"\x30B0"),

@@ -784,11 +495,11 @@ TEST(NetUtilTest, FormatUrlParsed) {

formatted.substr(parsed.ref.begin, parsed.ref.len));

// Omit_username_password + unescape case.

- formatted = FormatUrl(

- GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"

- "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),

- "ja", kFormatUrlOmitUsernamePassword, UnescapeRule::NORMAL, &parsed,

- NULL, NULL);

+ formatted =

+ FormatUrl(GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"

+ "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),

+ std::string(), kFormatUrlOmitUsernamePassword,

+ UnescapeRule::NORMAL, &parsed, NULL, NULL);

EXPECT_EQ(WideToUTF16(L"http://\x30B0\x30FC\x30B0\x30EB.jp:8080"

L"/\x30B0/?q=\x30B0#\x30B0"), formatted);

EXPECT_FALSE(parsed.username.is_valid());

« net/base/net_util_icu.cc ('K') | « net/base/net_util_icu.cc ('k') | no next file » | no next file with comments »