Home | History | Annotate | Download | only in base
      1 // Copyright (c) 2009 The Chromium Authors. All rights reserved.
      2 // Use of this source code is governed by a BSD-style license that can be
      3 // found in the LICENSE file.
      4 
      5 #include "net/base/net_util.h"
      6 
      7 #include "base/file_path.h"
      8 #include "base/format_macros.h"
      9 #include "base/string_util.h"
     10 #include "base/sys_string_conversions.h"
     11 #include "base/time.h"
     12 #include "googleurl/src/gurl.h"
     13 #include "net/base/sys_addrinfo.h"
     14 #include "testing/gtest/include/gtest/gtest.h"
     15 
     16 namespace {
     17 
     18 class NetUtilTest : public testing::Test {
     19 };
     20 
     21 struct FileCase {
     22   const wchar_t* file;
     23   const char* url;
     24 };
     25 
     26 struct HeaderCase {
     27   const wchar_t* header_name;
     28   const wchar_t* expected;
     29 };
     30 
     31 struct HeaderParamCase {
     32   const wchar_t* header_name;
     33   const wchar_t* param_name;
     34   const wchar_t* expected;
     35 };
     36 
     37 struct FileNameCDCase {
     38   const char* header_field;
     39   const char* referrer_charset;
     40   const wchar_t* expected;
     41 };
     42 
     43 const wchar_t* kLanguages[] = {
     44   L"",      L"en",    L"zh-CN",       L"ja",    L"ko",
     45   L"he",    L"ar",    L"ru",          L"el",    L"fr",
     46   L"de",    L"pt",    L"sv",          L"th",    L"hi",
     47   L"de,en", L"el,en", L"zh-TW,en",    L"ko,ja", L"he,ru,en",
     48   L"zh,ru,en"
     49 };
     50 
     51 struct IDNTestCase {
     52   const char* input;
     53   const wchar_t* unicode_output;
     54   const bool unicode_allowed[arraysize(kLanguages)];
     55 };
     56 
     57 // TODO(jungshik) This is just a random sample of languages and is far
     58 // from exhaustive.  We may have to generate all the combinations
     59 // of languages (powerset of a set of all the languages).
     60 const IDNTestCase idn_cases[] = {
     61   // No IDN
     62   {"www.google.com", L"www.google.com",
     63    {true,  true,  true,  true,  true,
     64     true,  true,  true,  true,  true,
     65     true,  true,  true,  true,  true,
     66     true,  true,  true,  true,  true,
     67     true}},
     68   {"www.google.com.", L"www.google.com.",
     69    {true,  true,  true,  true,  true,
     70     true,  true,  true,  true,  true,
     71     true,  true,  true,  true,  true,
     72     true,  true,  true,  true,  true,
     73     true}},
     74   {".", L".",
     75    {true,  true,  true,  true,  true,
     76     true,  true,  true,  true,  true,
     77     true,  true,  true,  true,  true,
     78     true,  true,  true,  true,  true,
     79     true}},
     80   {"", L"",
     81    {true,  true,  true,  true,  true,
     82     true,  true,  true,  true,  true,
     83     true,  true,  true,  true,  true,
     84     true,  true,  true,  true,  true,
     85     true}},
     86   // IDN
     87   // Hanzi (Traditional Chinese)
     88   {"xn--1lq90ic7f1rc.cn", L"\x5317\x4eac\x5927\x5b78.cn",
     89    {true,  false, true,  true,  false,
     90     false, false, false, false, false,
     91     false, false, false, false, false,
     92     false, false, true,  true,  false,
     93     true}},
     94   // Hanzi ('video' in Simplified Chinese : will pass only in zh-CN,zh)
     95   {"xn--cy2a840a.com", L"\x89c6\x9891.com",
     96    {true,  false, true,  false,  false,
     97     false, false, false, false, false,
     98     false, false, false, false, false,
     99     false, false, false, false,  false,
    100     true}},
    101   // Hanzi + '123'
    102   {"www.xn--123-p18d.com", L"www.\x4e00" L"123.com",
    103    {true,  false, true,  true,  false,
    104     false, false, false, false, false,
    105     false, false, false, false, false,
    106     false, false, true,  true,  false,
    107     true}},
    108   // Hanzi + Latin : U+56FD is simplified and is regarded
    109   // as not supported in zh-TW.
    110   {"www.xn--hello-9n1hm04c.com", L"www.hello\x4e2d\x56fd.com",
    111    {false, false, true,  true,  false,
    112     false, false, false, false, false,
    113     false, false, false, false, false,
    114     false, false, false, true,  false,
    115     true}},
    116   // Kanji + Kana (Japanese)
    117   {"xn--l8jvb1ey91xtjb.jp", L"\x671d\x65e5\x3042\x3055\x3072.jp",
    118    {true,  false, false, true,  false,
    119     false, false, false, false, false,
    120     false, false, false, false, false,
    121     false, false, false, true,  false,
    122     false}},
    123   // Katakana including U+30FC
    124   {"xn--tckm4i2e.jp", L"\x30b3\x30de\x30fc\x30b9.jp",
    125    {true, false, false, true,  false,
    126     false, false, false, false, false,
    127     false, false, false, false, false,
    128     false, false, false, true, false,
    129     }},
    130   // Katakana + Latin (Japanese)
    131   // TODO(jungshik): Change 'false' in the first element to 'true'
    132   // after upgrading to ICU 4.2.1 to use new uspoof_* APIs instead
    133   // of our IsIDNComponentInSingleScript().
    134   {"xn--e-efusa1mzf.jp", L"e\x30b3\x30de\x30fc\x30b9.jp",
    135    {false, false, false, true,  false,
    136     false, false, false, false, false,
    137     false, false, false, false, false,
    138     false, false, false, true, false,
    139     }},
    140   // Hangul (Korean)
    141   {"www.xn--or3b17p6jjc.kr", L"www.\xc804\xc790\xc815\xbd80.kr",
    142    {true,  false, false, false, true,
    143     false, false, false, false, false,
    144     false, false, false, false, false,
    145     false, false, false, true,  false,
    146     false}},
    147   // b<u-umlaut>cher (German)
    148   {"xn--bcher-kva.de", L"b\x00fc" L"cher.de",
    149    {true,  false, false, false, false,
    150     false, false, false, false, true,
    151     true,  false,  false, false, false,
    152     true,  false, false, false, false,
    153     false}},
    154   // a with diaeresis
    155   {"www.xn--frgbolaget-q5a.se", L"www.f\x00e4rgbolaget.se",
    156    {true,  false, false, false, false,
    157     false, false, false, false, false,
    158     true,  false, true, false, false,
    159     true,  false, false, false, false,
    160     false}},
    161   // c-cedilla (French)
    162   {"www.xn--alliancefranaise-npb.fr", L"www.alliancefran\x00e7" L"aise.fr",
    163    {true,  false, false, false, false,
    164     false, false, false, false, true,
    165     false, true,  false, false, false,
    166     false, false, false, false, false,
    167     false}},
    168   // caf'e with acute accent' (French)
    169   {"xn--caf-dma.fr", L"caf\x00e9.fr",
    170    {true,  false, false, false, false,
    171     false, false, false, false, true,
    172     false, true,  true,  false, false,
    173     false, false, false, false, false,
    174     false}},
    175   // c-cedillla and a with tilde (Portuguese)
    176   {"xn--poema-9qae5a.com.br", L"p\x00e3oema\x00e7\x00e3.com.br",
    177    {true,  false, false, false, false,
    178     false, false, false, false, false,
    179     false, true,  false, false, false,
    180     false, false, false, false, false,
    181     false}},
    182   // s with caron
    183   {"xn--achy-f6a.com", L"\x0161" L"achy.com",
    184    {true,  false, false, false, false,
    185     false, false, false, false, false,
    186     false, false, false, false, false,
    187     false, false, false, false, false,
    188     false}},
    189   // TODO(jungshik) : Add examples with Cyrillic letters
    190   // only used in some languages written in Cyrillic.
    191   // Eutopia (Greek)
    192   {"xn--kxae4bafwg.gr", L"\x03bf\x03c5\x03c4\x03bf\x03c0\x03af\x03b1.gr",
    193    {true,  false, false, false, false,
    194     false, false, false, true,  false,
    195     false, false, false, false, false,
    196     false, true,  false, false, false,
    197     false}},
    198   // Eutopia + 123 (Greek)
    199   {"xn---123-pldm0haj2bk.gr",
    200    L"\x03bf\x03c5\x03c4\x03bf\x03c0\x03af\x03b1-123.gr",
    201    {true,  false, false, false, false,
    202     false, false, false, true,  false,
    203     false, false, false, false, false,
    204     false, true,  false, false, false,
    205     false}},
    206   // Cyrillic (Russian)
    207   {"xn--n1aeec9b.ru", L"\x0442\x043e\x0440\x0442\x044b.ru",
    208    {true,  false, false, false, false,
    209     false, false, true,  false, false,
    210     false, false, false, false, false,
    211     false, false, false, false, true,
    212     true}},
    213   // Cyrillic + 123 (Russian)
    214   {"xn---123-45dmmc5f.ru", L"\x0442\x043e\x0440\x0442\x044b-123.ru",
    215    {true,  false, false, false, false,
    216     false, false, true,  false, false,
    217     false, false, false, false, false,
    218     false, false, false, false, true,
    219     true}},
    220   // Arabic
    221   {"xn--mgba1fmg.ar", L"\x0627\x0641\x0644\x0627\x0645.ar",
    222    {true,  false, false, false, false,
    223     false, true,  false, false, false,
    224     false, false, false, false, false,
    225     false, false, false, false, false,
    226     false}},
    227   // Hebrew
    228   {"xn--4dbib.he", L"\x05d5\x05d0\x05d4.he",
    229    {true,  false, false, false, false,
    230     true,  false, false, false, false,
    231     false, false, false, false, false,
    232     false, false, false, false, true,
    233     false}},
    234   // Thai
    235   {"xn--12c2cc4ag3b4ccu.th",
    236    L"\x0e2a\x0e32\x0e22\x0e01\x0e32\x0e23\x0e1a\x0e34\x0e19.th",
    237    {true,  false, false, false, false,
    238     false, false, false, false, false,
    239     false, false, false, true,  false,
    240     false, false, false, false, false,
    241     false}},
    242   // Devangari (Hindi)
    243   {"www.xn--l1b6a9e1b7c.in", L"www.\x0905\x0915\x094b\x0932\x093e.in",
    244    {true,  false, false, false, false,
    245     false, false, false, false, false,
    246     false, false, false, false, true,
    247     false, false, false, false, false,
    248     false}},
    249   // Invalid IDN
    250   {"xn--hello?world.com", NULL,
    251    {false, false, false, false, false,
    252     false, false, false, false, false,
    253     false, false, false, false, false,
    254     false, false, false, false, false,
    255     false}},
    256   // Unsafe IDNs
    257   // "payp<alpha>l.com"
    258   {"www.xn--paypl-g9d.com", L"payp\x03b1l.com",
    259    {false, false, false, false, false,
    260     false, false, false, false, false,
    261     false, false, false, false, false,
    262     false, false, false, false, false,
    263     false}},
    264   // google.gr with Greek omicron and epsilon
    265   {"xn--ggl-6xc1ca.gr", L"g\x03bf\x03bfgl\x03b5.gr",
    266    {false, false, false, false, false,
    267     false, false, false, false, false,
    268     false, false, false, false, false,
    269     false, false, false, false, false,
    270     false}},
    271   // google.ru with Cyrillic o
    272   {"xn--ggl-tdd6ba.ru", L"g\x043e\x043egl\x0435.ru",
    273    {false, false, false, false, false,
    274     false, false, false, false, false,
    275     false, false, false, false, false,
    276     false, false, false, false, false,
    277     false}},
    278   // h<e with acute>llo<China in Han>.cn
    279   {"xn--hllo-bpa7979ih5m.cn", L"h\x00e9llo\x4e2d\x56fd.cn",
    280    {false, false, false, false, false,
    281     false, false, false, false, false,
    282     false, false, false, false, false,
    283     false, false, false, false, false,
    284     false}},
    285   // <Greek rho><Cyrillic a><Cyrillic u>.ru
    286   {"xn--2xa6t2b.ru", L"\x03c1\x0430\x0443.ru",
    287    {false, false, false, false, false,
    288     false, false, false, false, false,
    289     false, false, false, false, false,
    290     false, false, false, false, false,
    291     false}},
    292   // One that's really long that will force a buffer realloc
    293   {"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"
    294        "aaaaaaa",
    295    L"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"
    296        L"aaaaaaaa",
    297    {true,  true,  true,  true,  true,
    298     true,  true,  true,  true,  true,
    299     true,  true,  true,  true,  true,
    300     true,  true,  true,  true,  true,
    301     true}},
    302   // Test cases for characters we blacklisted although allowed in IDN.
    303   // Embedded spaces will be turned to %20 in the display.
    304   // TODO(jungshik): We need to have more cases. This is a typical
    305   // data-driven trap. The following test cases need to be separated
    306   // and tested only for a couple of languages.
    307   {"xn--osd3820f24c.kr", L"\xac00\xb098\x115f.kr",
    308     {false, false, false, false, false,
    309      false, false, false, false, false,
    310      false, false, false, false, false,
    311      false, false, false, false, false,
    312      false}},
    313   {"www.xn--google-ho0coa.com", L"www.\x2039google\x203a.com",
    314     {false, false, false, false, false,
    315      false, false, false, false, false,
    316      false, false, false, false, false,
    317      false, false, false, false, false,
    318   }},
    319   {"google.xn--comabc-k8d", L"google.com\x0338" L"abc",
    320     {false, false, false, false, false,
    321      false, false, false, false, false,
    322      false, false, false, false, false,
    323      false, false, false, false, false,
    324      }},
    325 #if 0
    326   // These two cases are special. We need a separate test.
    327   // U+3000 and U+3002 are normalized to ASCII space and dot.
    328   {"xn-- -kq6ay5z.cn", L"\x4e2d\x56fd\x3000.cn",
    329     {false, false, true,  false, false,
    330      false, false, false, false, false,
    331      false, false, false, false, false,
    332      false, false, true,  false, false,
    333      true}},
    334   {"xn--fiqs8s.cn", L"\x4e2d\x56fd\x3002" L"cn",
    335     {false, false, true,  false, false,
    336      false, false, false, false, false,
    337      false, false, false, false, false,
    338      false, false, true,  false, false,
    339      true}},
    340 #endif
    341 };
    342 
    343 struct AdjustOffsetCase {
    344   size_t input_offset;
    345   size_t output_offset;
    346 };
    347 
    348 struct CompliantHostCase {
    349   const char* host;
    350   bool expected_output;
    351 };
    352 
    353 struct SuggestedFilenameCase {
    354   const char* url;
    355   const char* content_disp_header;
    356   const char* referrer_charset;
    357   const wchar_t* default_filename;
    358   const wchar_t* expected_filename;
    359 };
    360 
    361 struct UrlTestData {
    362   const char* description;
    363   const char* input;
    364   const std::wstring languages;
    365   bool omit;
    366   UnescapeRule::Type escape_rules;
    367   const std::wstring output;
    368   size_t prefix_len;
    369 };
    370 
    371 // Returns an addrinfo for the given 32-bit address (IPv4.)
    372 // The result lives in static storage, so don't delete it.
    373 // |bytes| should be an array of length 4.
    374 const struct addrinfo* GetIPv4Address(const uint8* bytes) {
    375   static struct addrinfo static_ai;
    376   static struct sockaddr_in static_addr4;
    377 
    378   struct addrinfo* ai = &static_ai;
    379   ai->ai_socktype = SOCK_STREAM;
    380   memset(ai, 0, sizeof(static_ai));
    381 
    382   ai->ai_family = AF_INET;
    383   ai->ai_addrlen = sizeof(static_addr4);
    384 
    385   struct sockaddr_in* addr4 = &static_addr4;
    386   memset(addr4, 0, sizeof(static_addr4));
    387   addr4->sin_port = htons(80);
    388   addr4->sin_family = ai->ai_family;
    389   memcpy(&addr4->sin_addr, bytes, 4);
    390 
    391   ai->ai_addr = (sockaddr*)addr4;
    392   return ai;
    393 }
    394 
    395 // Returns a addrinfo for the given 128-bit address (IPv6.)
    396 // The result lives in static storage, so don't delete it.
    397 // |bytes| should be an array of length 16.
    398 const struct addrinfo* GetIPv6Address(const uint8* bytes) {
    399   static struct addrinfo static_ai;
    400   static struct sockaddr_in6 static_addr6;
    401 
    402   struct addrinfo* ai = &static_ai;
    403   ai->ai_socktype = SOCK_STREAM;
    404   memset(ai, 0, sizeof(static_ai));
    405 
    406   ai->ai_family = AF_INET6;
    407   ai->ai_addrlen = sizeof(static_addr6);
    408 
    409   struct sockaddr_in6* addr6 = &static_addr6;
    410   memset(addr6, 0, sizeof(static_addr6));
    411   addr6->sin6_port = htons(80);
    412   addr6->sin6_family = ai->ai_family;
    413   memcpy(&addr6->sin6_addr, bytes, 16);
    414 
    415   ai->ai_addr = (sockaddr*)addr6;
    416   return ai;
    417 }
    418 
    419 
    420 // A helper for IDN*{Fast,Slow}.
    421 // Append "::<language list>" to |expected| and |actual| to make it
    422 // easy to tell which sub-case fails without debugging.
    423 void AppendLanguagesToOutputs(const wchar_t* languages,
    424                               std::wstring* expected,
    425                               std::wstring* actual) {
    426   expected->append(L"::");
    427   expected->append(languages);
    428   actual->append(L"::");
    429   actual->append(languages);
    430 }
    431 
    432 }  // anonymous namespace
    433 
    434 TEST(NetUtilTest, FileURLConversion) {
    435   // a list of test file names and the corresponding URLs
    436   const FileCase round_trip_cases[] = {
    437 #if defined(OS_WIN)
    438     {L"C:\\foo\\bar.txt", "file:///C:/foo/bar.txt"},
    439     {L"\\\\some computer\\foo\\bar.txt",
    440      "file://some%20computer/foo/bar.txt"}, // UNC
    441     {L"D:\\Name;with%some symbols*#",
    442      "file:///D:/Name%3Bwith%25some%20symbols*%23"},
    443     // issue 14153: To be tested with the OS default codepage other than 1252.
    444     {L"D:\\latin1\\caf\x00E9\x00DD.txt",
    445      "file:///D:/latin1/caf%C3%A9%C3%9D.txt"},
    446     {L"D:\\otherlatin\\caf\x0119.txt",
    447      "file:///D:/otherlatin/caf%C4%99.txt"},
    448     {L"D:\\greek\\\x03B1\x03B2\x03B3.txt",
    449      "file:///D:/greek/%CE%B1%CE%B2%CE%B3.txt"},
    450     {L"D:\\Chinese\\\x6240\x6709\x4e2d\x6587\x7f51\x9875.doc",
    451      "file:///D:/Chinese/%E6%89%80%E6%9C%89%E4%B8%AD%E6%96%87%E7%BD%91"
    452          "%E9%A1%B5.doc"},
    453     {L"D:\\plane1\\\xD835\xDC00\xD835\xDC01.txt",  // Math alphabet "AB"
    454      "file:///D:/plane1/%F0%9D%90%80%F0%9D%90%81.txt"},
    455 #elif defined(OS_POSIX)
    456     {L"/foo/bar.txt", "file:///foo/bar.txt"},
    457     {L"/foo/BAR.txt", "file:///foo/BAR.txt"},
    458     {L"/C:/foo/bar.txt", "file:///C:/foo/bar.txt"},
    459     {L"/some computer/foo/bar.txt", "file:///some%20computer/foo/bar.txt"},
    460     {L"/Name;with%some symbols*#", "file:///Name%3Bwith%25some%20symbols*%23"},
    461     {L"/latin1/caf\x00E9\x00DD.txt", "file:///latin1/caf%C3%A9%C3%9D.txt"},
    462     {L"/otherlatin/caf\x0119.txt", "file:///otherlatin/caf%C4%99.txt"},
    463     {L"/greek/\x03B1\x03B2\x03B3.txt", "file:///greek/%CE%B1%CE%B2%CE%B3.txt"},
    464     {L"/Chinese/\x6240\x6709\x4e2d\x6587\x7f51\x9875.doc",
    465      "file:///Chinese/%E6%89%80%E6%9C%89%E4%B8%AD%E6%96%87%E7%BD"
    466          "%91%E9%A1%B5.doc"},
    467     {L"/plane1/\x1D400\x1D401.txt",  // Math alphabet "AB"
    468      "file:///plane1/%F0%9D%90%80%F0%9D%90%81.txt"},
    469 #endif
    470   };
    471 
    472   // First, we'll test that we can round-trip all of the above cases of URLs
    473   FilePath output;
    474   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(round_trip_cases); i++) {
    475     // convert to the file URL
    476     GURL file_url(net::FilePathToFileURL(
    477         FilePath::FromWStringHack(round_trip_cases[i].file)));
    478     EXPECT_EQ(round_trip_cases[i].url, file_url.spec());
    479 
    480     // Back to the filename.
    481     EXPECT_TRUE(net::FileURLToFilePath(file_url, &output));
    482     EXPECT_EQ(round_trip_cases[i].file, output.ToWStringHack());
    483   }
    484 
    485   // Test that various file: URLs get decoded into the correct file type
    486   FileCase url_cases[] = {
    487 #if defined(OS_WIN)
    488     {L"C:\\foo\\bar.txt", "file:c|/foo\\bar.txt"},
    489     {L"C:\\foo\\bar.txt", "file:/c:/foo/bar.txt"},
    490     {L"\\\\foo\\bar.txt", "file://foo\\bar.txt"},
    491     {L"C:\\foo\\bar.txt", "file:///c:/foo/bar.txt"},
    492     {L"\\\\foo\\bar.txt", "file:////foo\\bar.txt"},
    493     {L"\\\\foo\\bar.txt", "file:/foo/bar.txt"},
    494     {L"\\\\foo\\bar.txt", "file://foo\\bar.txt"},
    495     {L"C:\\foo\\bar.txt", "file:\\\\\\c:/foo/bar.txt"},
    496 #elif defined(OS_POSIX)
    497     {L"/c:/foo/bar.txt", "file:/c:/foo/bar.txt"},
    498     {L"/c:/foo/bar.txt", "file:///c:/foo/bar.txt"},
    499     {L"/foo/bar.txt", "file:/foo/bar.txt"},
    500     {L"/c:/foo/bar.txt", "file:\\\\\\c:/foo/bar.txt"},
    501     {L"/foo/bar.txt", "file:foo/bar.txt"},
    502     {L"/bar.txt", "file://foo/bar.txt"},
    503     {L"/foo/bar.txt", "file:///foo/bar.txt"},
    504     {L"/foo/bar.txt", "file:////foo/bar.txt"},
    505     {L"/foo/bar.txt", "file:////foo//bar.txt"},
    506     {L"/foo/bar.txt", "file:////foo///bar.txt"},
    507     {L"/foo/bar.txt", "file:////foo////bar.txt"},
    508     {L"/c:/foo/bar.txt", "file:\\\\\\c:/foo/bar.txt"},
    509     {L"/c:/foo/bar.txt", "file:c:/foo/bar.txt"},
    510     // We get these wrong because GURL turns back slashes into forward
    511     // slashes.
    512     //{L"/foo%5Cbar.txt", "file://foo\\bar.txt"},
    513     //{L"/c|/foo%5Cbar.txt", "file:c|/foo\\bar.txt"},
    514     //{L"/foo%5Cbar.txt", "file://foo\\bar.txt"},
    515     //{L"/foo%5Cbar.txt", "file:////foo\\bar.txt"},
    516     //{L"/foo%5Cbar.txt", "file://foo\\bar.txt"},
    517 #endif
    518   };
    519   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(url_cases); i++) {
    520     net::FileURLToFilePath(GURL(url_cases[i].url), &output);
    521     EXPECT_EQ(url_cases[i].file, output.ToWStringHack());
    522   }
    523 
    524   // Unfortunately, UTF8ToWide discards invalid UTF8 input.
    525 #ifdef BUG_878908_IS_FIXED
    526   // Test that no conversion happens if the UTF-8 input is invalid, and that
    527   // the input is preserved in UTF-8
    528   const char invalid_utf8[] = "file:///d:/Blah/\xff.doc";
    529   const wchar_t invalid_wide[] = L"D:\\Blah\\\xff.doc";
    530   EXPECT_TRUE(net::FileURLToFilePath(
    531       GURL(std::string(invalid_utf8)), &output));
    532   EXPECT_EQ(std::wstring(invalid_wide), output);
    533 #endif
    534 
    535   // Test that if a file URL is malformed, we get a failure
    536   EXPECT_FALSE(net::FileURLToFilePath(GURL("filefoobar"), &output));
    537 }
    538 
    539 TEST(NetUtilTest, GetIdentityFromURL) {
    540   struct {
    541     const char* input_url;
    542     const wchar_t* expected_username;
    543     const wchar_t* expected_password;
    544   } tests[] = {
    545     {
    546       "http://username:password@google.com",
    547       L"username",
    548       L"password",
    549     },
    550     { // Test for http://crbug.com/19200
    551       "http://username:p@ssword@google.com",
    552       L"username",
    553       L"p@ssword",
    554     },
    555     { // Username contains %20.
    556       "http://use rname:password (at) google.com",
    557       L"use rname",
    558       L"password",
    559     },
    560     { // Keep %00 as is.
    561       "http://use%00rname:password@google.com",
    562       L"use%00rname",
    563       L"password",
    564     },
    565     { // Use a '+' in the username.
    566       "http://use+rname:password@google.com",
    567       L"use+rname",
    568       L"password",
    569     },
    570     { // Use a '&' in the password.
    571       "http://username:p&ssword@google.com",
    572       L"username",
    573       L"p&ssword",
    574     },
    575   };
    576   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
    577     SCOPED_TRACE(StringPrintf("Test[%" PRIuS "]: %s", i, tests[i].input_url));
    578     GURL url(tests[i].input_url);
    579 
    580     std::wstring username, password;
    581     net::GetIdentityFromURL(url, &username, &password);
    582 
    583     EXPECT_EQ(tests[i].expected_username, username);
    584     EXPECT_EQ(tests[i].expected_password, password);
    585   }
    586 }
    587 
    588 // Try extracting a username which was encoded with UTF8.
    589 TEST(NetUtilTest, GetIdentityFromURL_UTF8) {
    590   GURL url(WideToUTF16(L"http://foo:\x4f60\x597d@blah.com"));
    591 
    592   EXPECT_EQ("foo", url.username());
    593   EXPECT_EQ("%E4%BD%A0%E5%A5%BD", url.password());
    594 
    595   // Extract the unescaped identity.
    596   std::wstring username, password;
    597   net::GetIdentityFromURL(url, &username, &password);
    598 
    599   // Verify that it was decoded as UTF8.
    600   EXPECT_EQ(L"foo", username);
    601   EXPECT_EQ(L"\x4f60\x597d", password);
    602 }
    603 
    604 // Just a bunch of fake headers.
    605 const wchar_t* google_headers =
    606     L"HTTP/1.1 200 OK\n"
    607     L"Content-TYPE: text/html; charset=utf-8\n"
    608     L"Content-disposition: attachment; filename=\"download.pdf\"\n"
    609     L"Content-Length: 378557\n"
    610     L"X-Google-Google1: 314159265\n"
    611     L"X-Google-Google2: aaaa2:7783,bbb21:9441\n"
    612     L"X-Google-Google4: home\n"
    613     L"Transfer-Encoding: chunked\n"
    614     L"Set-Cookie: HEHE_AT=6666x66beef666x6-66xx6666x66; Path=/mail\n"
    615     L"Set-Cookie: HEHE_HELP=owned:0;Path=/\n"
    616     L"Set-Cookie: S=gmail=Xxx-beefbeefbeef_beefb:gmail_yj=beefbeef000beefbee"
    617         L"fbee:gmproxy=bee-fbeefbe; Domain=.google.com; Path=/\n"
    618     L"X-Google-Google2: /one/two/three/four/five/six/seven-height/nine:9411\n"
    619     L"Server: GFE/1.3\n"
    620     L"Transfer-Encoding: chunked\n"
    621     L"Date: Mon, 13 Nov 2006 21:38:09 GMT\n"
    622     L"Expires: Tue, 14 Nov 2006 19:23:58 GMT\n"
    623     L"X-Malformed: bla; arg=test\"\n"
    624     L"X-Malformed2: bla; arg=\n"
    625     L"X-Test: bla; arg1=val1; arg2=val2";
    626 
    627 TEST(NetUtilTest, GetSpecificHeader) {
    628   const HeaderCase tests[] = {
    629     {L"content-type", L"text/html; charset=utf-8"},
    630     {L"CONTENT-LENGTH", L"378557"},
    631     {L"Date", L"Mon, 13 Nov 2006 21:38:09 GMT"},
    632     {L"Bad-Header", L""},
    633     {L"", L""},
    634   };
    635 
    636   // Test first with google_headers.
    637   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
    638     std::wstring result = net::GetSpecificHeader(google_headers,
    639                                                  tests[i].header_name);
    640     EXPECT_EQ(result, tests[i].expected);
    641   }
    642 
    643   // Test again with empty headers.
    644   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
    645     std::wstring result = net::GetSpecificHeader(L"", tests[i].header_name);
    646     EXPECT_EQ(result, std::wstring());
    647   }
    648 }
    649 
    650 TEST(NetUtilTest, GetHeaderParamValue) {
    651   const HeaderParamCase tests[] = {
    652     {L"Content-type", L"charset", L"utf-8"},
    653     {L"content-disposition", L"filename", L"download.pdf"},
    654     {L"Content-Type", L"badparam", L""},
    655     {L"X-Malformed", L"arg", L"test\""},
    656     {L"X-Malformed2", L"arg", L""},
    657     {L"X-Test", L"arg1", L"val1"},
    658     {L"X-Test", L"arg2", L"val2"},
    659     {L"Bad-Header", L"badparam", L""},
    660     {L"Bad-Header", L"", L""},
    661     {L"", L"badparam", L""},
    662     {L"", L"", L""},
    663   };
    664   // TODO(mpcomplete): add tests for other formats of headers.
    665 
    666   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
    667     std::wstring header_value =
    668         net::GetSpecificHeader(google_headers, tests[i].header_name);
    669     std::wstring result =
    670         net::GetHeaderParamValue(header_value, tests[i].param_name);
    671     EXPECT_EQ(result, tests[i].expected);
    672   }
    673 
    674   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
    675     std::wstring header_value =
    676         net::GetSpecificHeader(L"", tests[i].header_name);
    677     std::wstring result =
    678         net::GetHeaderParamValue(header_value, tests[i].param_name);
    679     EXPECT_EQ(result, std::wstring());
    680   }
    681 }
    682 
    683 TEST(NetUtilTest, GetFileNameFromCD) {
    684   const FileNameCDCase tests[] = {
    685     // Test various forms of C-D header fields emitted by web servers.
    686     {"content-disposition: inline; filename=\"abcde.pdf\"", "", L"abcde.pdf"},
    687     {"content-disposition: inline; name=\"abcde.pdf\"", "", L"abcde.pdf"},
    688     {"content-disposition: attachment; filename=abcde.pdf", "", L"abcde.pdf"},
    689     {"content-disposition: attachment; name=abcde.pdf", "", L"abcde.pdf"},
    690     {"content-disposition: attachment; filename=abc,de.pdf", "", L"abc,de.pdf"},
    691     {"content-disposition: filename=abcde.pdf", "", L"abcde.pdf"},
    692     {"content-disposition: filename= abcde.pdf", "", L"abcde.pdf"},
    693     {"content-disposition: filename =abcde.pdf", "", L"abcde.pdf"},
    694     {"content-disposition: filename = abcde.pdf", "", L"abcde.pdf"},
    695     {"content-disposition: filename\t=abcde.pdf", "", L"abcde.pdf"},
    696     {"content-disposition: filename \t\t  =abcde.pdf", "", L"abcde.pdf"},
    697     {"content-disposition: name=abcde.pdf", "", L"abcde.pdf"},
    698     {"content-disposition: inline; filename=\"abc%20de.pdf\"", "",
    699      L"abc de.pdf"},
    700     // Whitespaces are converted to a space.
    701     {"content-disposition: inline; filename=\"abc  \t\nde.pdf\"", "",
    702      L"abc    de.pdf"},
    703     // %-escaped UTF-8
    704     {"Content-Disposition: attachment; filename=\"%EC%98%88%EC%88%A0%20"
    705      "%EC%98%88%EC%88%A0.jpg\"", "", L"\xc608\xc220 \xc608\xc220.jpg"},
    706     {"Content-Disposition: attachment; filename=\"%F0%90%8C%B0%F0%90%8C%B1"
    707      "abc.jpg\"", "", L"\U00010330\U00010331abc.jpg"},
    708     {"Content-Disposition: attachment; filename=\"%EC%98%88%EC%88%A0 \n"
    709      "%EC%98%88%EC%88%A0.jpg\"", "", L"\xc608\xc220  \xc608\xc220.jpg"},
    710     // RFC 2047 with various charsets and Q/B encodings
    711     {"Content-Disposition: attachment; filename=\"=?EUC-JP?Q?=B7=DD=BD="
    712      "D13=2Epng?=\"", "", L"\x82b8\x8853" L"3.png"},
    713     {"Content-Disposition: attachment; filename==?eUc-Kr?b?v7m8+iAzLnBuZw==?=",
    714      "", L"\xc608\xc220 3.png"},
    715     {"Content-Disposition: attachment; filename==?utf-8?Q?=E8=8A=B8=E8"
    716      "=A1=93_3=2Epng?=", "", L"\x82b8\x8853 3.png"},
    717     {"Content-Disposition: attachment; filename==?utf-8?Q?=F0=90=8C=B0"
    718      "_3=2Epng?=", "", L"\U00010330 3.png"},
    719     {"Content-Disposition: inline; filename=\"=?iso88591?Q?caf=e9_=2epng?=\"",
    720      "", L"caf\x00e9 .png"},
    721     // Space after an encode word should be removed.
    722     {"Content-Disposition: inline; filename=\"=?iso88591?Q?caf=E9_?= .png\"",
    723      "", L"caf\x00e9 .png"},
    724     // Two encoded words with different charsets (not very likely to be emitted
    725     // by web servers in the wild). Spaces between them are removed.
    726     {"Content-Disposition: inline; filename=\"=?euc-kr?b?v7m8+iAz?="
    727      " =?ksc5601?q?=BF=B9=BC=FA=2Epng?=\"", "",
    728      L"\xc608\xc220 3\xc608\xc220.png"},
    729     {"Content-Disposition: attachment; filename=\"=?windows-1252?Q?caf=E9?="
    730      "  =?iso-8859-7?b?4eI=?= .png\"", "", L"caf\x00e9\x03b1\x03b2.png"},
    731     // Non-ASCII string is passed through and treated as UTF-8 as long as
    732     // it's valid as UTF-8 and regardless of |referrer_charset|.
    733     {"Content-Disposition: attachment; filename=caf\xc3\xa9.png",
    734      "iso-8859-1", L"caf\x00e9.png"},
    735     {"Content-Disposition: attachment; filename=caf\xc3\xa9.png",
    736      "", L"caf\x00e9.png"},
    737     // Non-ASCII/Non-UTF-8 string. Fall back to the referrer charset.
    738     {"Content-Disposition: attachment; filename=caf\xe5.png",
    739      "windows-1253", L"caf\x03b5.png"},
    740 #if 0
    741     // Non-ASCII/Non-UTF-8 string. Fall back to the native codepage.
    742     // TODO(jungshik): We need to set the OS default codepage
    743     // to a specific value before testing. On Windows, we can use
    744     // SetThreadLocale().
    745     {"Content-Disposition: attachment; filename=\xb0\xa1\xb0\xa2.png",
    746      "", L"\xac00\xac01.png"},
    747 #endif
    748     // Failure cases
    749     // Invalid hex-digit "G"
    750     {"Content-Disposition: attachment; filename==?iiso88591?Q?caf=EG?=", "",
    751      L""},
    752     // Incomplete RFC 2047 encoded-word (missing '='' at the end)
    753     {"Content-Disposition: attachment; filename==?iso88591?Q?caf=E3?", "", L""},
    754     // Extra character at the end of an encoded word
    755     {"Content-Disposition: attachment; filename==?iso88591?Q?caf=E3?==",
    756      "", L""},
    757     // Extra token at the end of an encoded word
    758     {"Content-Disposition: attachment; filename==?iso88591?Q?caf=E3?=?",
    759      "", L""},
    760     {"Content-Disposition: attachment; filename==?iso88591?Q?caf=E3?=?=",
    761      "",  L""},
    762     // Incomplete hex-escaped chars
    763     {"Content-Disposition: attachment; filename==?windows-1252?Q?=63=61=E?=",
    764      "", L""},
    765     {"Content-Disposition: attachment; filename=%EC%98%88%EC%88%A", "", L""},
    766     // %-escaped non-UTF-8 encoding is an "error"
    767     {"Content-Disposition: attachment; filename=%B7%DD%BD%D1.png", "", L""},
    768     // Two RFC 2047 encoded words in a row without a space is an error.
    769     {"Content-Disposition: attachment; filename==?windows-1252?Q?caf=E3?="
    770      "=?iso-8859-7?b?4eIucG5nCg==?=", "", L""},
    771   };
    772   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
    773     EXPECT_EQ(tests[i].expected,
    774               UTF8ToWide(net::GetFileNameFromCD(tests[i].header_field,
    775                                                 tests[i].referrer_charset)));
    776   }
    777 }
    778 
    779 TEST(NetUtilTest, IDNToUnicodeFast) {
    780   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(idn_cases); i++) {
    781     for (size_t j = 0; j < arraysize(kLanguages); j++) {
    782       // ja || zh-TW,en || ko,ja -> IDNToUnicodeSlow
    783       if (j == 3 || j == 17 || j == 18)
    784         continue;
    785       std::wstring output(net::IDNToUnicode(idn_cases[i].input,
    786           strlen(idn_cases[i].input), kLanguages[j], NULL));
    787       std::wstring expected(idn_cases[i].unicode_allowed[j] ?
    788           idn_cases[i].unicode_output : ASCIIToWide(idn_cases[i].input));
    789       AppendLanguagesToOutputs(kLanguages[j], &expected, &output);
    790       EXPECT_EQ(expected, output);
    791     }
    792   }
    793 }
    794 
    795 TEST(NetUtilTest, IDNToUnicodeSlow) {
    796   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(idn_cases); i++) {
    797     for (size_t j = 0; j < arraysize(kLanguages); j++) {
    798       // !(ja || zh-TW,en || ko,ja) -> IDNToUnicodeFast
    799       if (!(j == 3 || j == 17 || j == 18))
    800         continue;
    801       std::wstring output(net::IDNToUnicode(idn_cases[i].input,
    802           strlen(idn_cases[i].input), kLanguages[j], NULL));
    803       std::wstring expected(idn_cases[i].unicode_allowed[j] ?
    804           idn_cases[i].unicode_output : ASCIIToWide(idn_cases[i].input));
    805       AppendLanguagesToOutputs(kLanguages[j], &expected, &output);
    806       EXPECT_EQ(expected, output);
    807     }
    808   }
    809 }
    810 
    811 TEST(NetUtilTest, IDNToUnicodeAdjustOffset) {
    812   const AdjustOffsetCase adjust_cases[] = {
    813     {0, 0},
    814     {2, 2},
    815     {4, 4},
    816     {5, 5},
    817     {6, std::wstring::npos},
    818     {16, std::wstring::npos},
    819     {17, 7},
    820     {18, 8},
    821     {19, std::wstring::npos},
    822     {25, std::wstring::npos},
    823     {34, 12},
    824     {35, 13},
    825     {38, 16},
    826     {39, std::wstring::npos},
    827     {std::wstring::npos, std::wstring::npos},
    828   };
    829   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(adjust_cases); ++i) {
    830     size_t offset = adjust_cases[i].input_offset;
    831     // "test.\x89c6\x9891.\x5317\x4eac\x5927\x5b78.test"
    832     net::IDNToUnicode("test.xn--cy2a840a.xn--1lq90ic7f1rc.test", 39, L"zh-CN",
    833                       &offset);
    834     EXPECT_EQ(adjust_cases[i].output_offset, offset);
    835   }
    836 }
    837 
    838 TEST(NetUtilTest, CompliantHost) {
    839   const CompliantHostCase compliant_host_cases[] = {
    840     {"", false},
    841     {"a", true},
    842     {"-", false},
    843     {".", false},
    844     {"a.", true},
    845     {"a.a", true},
    846     {"9.a", true},
    847     {"a.9", false},
    848     {"_9a", false},
    849     {"a.a9", true},
    850     {"a.9a", false},
    851     {"a+9a", false},
    852     {"1-.a-b", false},
    853     {"1-2.a_b", true},
    854     {"a.b.c.d.e", true},
    855     {"1.2.3.4.e", true},
    856     {"a.b.c.d.5", false},
    857     {"1.2.3.4.e.", true},
    858     {"a.b.c.d.5.", false},
    859   };
    860 
    861   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(compliant_host_cases); ++i) {
    862     EXPECT_EQ(compliant_host_cases[i].expected_output,
    863               net::IsCanonicalizedHostCompliant(compliant_host_cases[i].host));
    864   }
    865 }
    866 
    867 TEST(NetUtilTest, StripWWW) {
    868   EXPECT_EQ(L"", net::StripWWW(L""));
    869   EXPECT_EQ(L"", net::StripWWW(L"www."));
    870   EXPECT_EQ(L"blah", net::StripWWW(L"www.blah"));
    871   EXPECT_EQ(L"blah", net::StripWWW(L"blah"));
    872 }
    873 
    874 TEST(NetUtilTest, GetSuggestedFilename) {
    875   const SuggestedFilenameCase test_cases[] = {
    876     {"http://www.google.com/",
    877      "Content-disposition: attachment; filename=test.html",
    878      "",
    879      L"",
    880      L"test.html"},
    881     {"http://www.google.com/",
    882      "Content-disposition: attachment; filename=\"test.html\"",
    883      "",
    884      L"",
    885      L"test.html"},
    886     {"http://www.google.com/path/test.html",
    887      "Content-disposition: attachment",
    888      "",
    889      L"",
    890      L"test.html"},
    891     {"http://www.google.com/path/test.html",
    892      "Content-disposition: attachment;",
    893      "",
    894      L"",
    895      L"test.html"},
    896     {"http://www.google.com/",
    897      "",
    898      "",
    899      L"",
    900      L"www.google.com"},
    901     {"http://www.google.com/test.html",
    902      "",
    903      "",
    904      L"",
    905      L"test.html"},
    906     // Now that we use googleurl's ExtractFileName, this case falls back
    907     // to the hostname. If this behavior is not desirable, we'd better
    908     // change ExtractFileName (in url_parse).
    909     {"http://www.google.com/path/",
    910      "",
    911      "",
    912      L"",
    913      L"www.google.com"},
    914     {"http://www.google.com/path",
    915      "",
    916      "",
    917      L"",
    918      L"path"},
    919     {"file:///",
    920      "",
    921      "",
    922      L"",
    923      L"download"},
    924     {"non-standard-scheme:",
    925      "",
    926      "",
    927      L"",
    928      L"download"},
    929     {"http://www.google.com/",
    930      "Content-disposition: attachment; filename =\"test.html\"",
    931      "",
    932      L"download",
    933      L"test.html"},
    934     {"http://www.google.com/",
    935      "",
    936      "",
    937      L"download",
    938      L"download"},
    939     {"http://www.google.com/",
    940      "Content-disposition: attachment; filename=\"../test.html\"",
    941      "",
    942      L"",
    943      L"test.html"},
    944     {"http://www.google.com/",
    945      "Content-disposition: attachment; filename=\"..\"",
    946      "",
    947      L"download",
    948      L"download"},
    949     {"http://www.google.com/test.html",
    950      "Content-disposition: attachment; filename=\"..\"",
    951      "",
    952      L"download",
    953      L"test.html"},
    954     // Below is a small subset of cases taken from GetFileNameFromCD test above.
    955     {"http://www.google.com/",
    956      "Content-Disposition: attachment; filename=\"%EC%98%88%EC%88%A0%20"
    957      "%EC%98%88%EC%88%A0.jpg\"",
    958      "",
    959      L"",
    960      L"\uc608\uc220 \uc608\uc220.jpg"},
    961     {"http://www.google.com/%EC%98%88%EC%88%A0%20%EC%98%88%EC%88%A0.jpg",
    962      "",
    963      "",
    964      L"download",
    965      L"\uc608\uc220 \uc608\uc220.jpg"},
    966     {"http://www.google.com/",
    967      "Content-disposition: attachment;",
    968      "",
    969      L"\uB2E4\uC6B4\uB85C\uB4DC",
    970      L"\uB2E4\uC6B4\uB85C\uB4DC"},
    971     {"http://www.google.com/",
    972      "Content-Disposition: attachment; filename=\"=?EUC-JP?Q?=B7=DD=BD="
    973      "D13=2Epng?=\"",
    974      "",
    975      L"download",
    976      L"\u82b8\u88533.png"},
    977     {"http://www.example.com/images?id=3",
    978      "Content-Disposition: attachment; filename=caf\xc3\xa9.png",
    979      "iso-8859-1",
    980      L"",
    981      L"caf\u00e9.png"},
    982     {"http://www.example.com/images?id=3",
    983      "Content-Disposition: attachment; filename=caf\xe5.png",
    984      "windows-1253",
    985      L"",
    986      L"caf\u03b5.png"},
    987     {"http://www.example.com/file?id=3",
    988      "Content-Disposition: attachment; name=\xcf\xc2\xd4\xd8.zip",
    989      "GBK",
    990      L"",
    991      L"\u4e0b\u8f7d.zip"},
    992     // Invalid C-D header. Extracts filename from url.
    993     {"http://www.google.com/test.html",
    994      "Content-Disposition: attachment; filename==?iiso88591?Q?caf=EG?=",
    995      "",
    996      L"",
    997      L"test.html"},
    998     // about: and data: URLs
    999     {"about:chrome",
   1000      "",
   1001      "",
   1002      L"",
   1003      L"download"},
   1004     {"data:,looks/like/a.path",
   1005      "",
   1006      "",
   1007      L"",
   1008      L"download"},
   1009     {"data:text/plain;base64,VG8gYmUgb3Igbm90IHRvIGJlLg=",
   1010      "",
   1011      "",
   1012      L"",
   1013      L"download"},
   1014     {"data:,looks/like/a.path",
   1015      "",
   1016      "",
   1017      L"default_filename_is_given",
   1018      L"default_filename_is_given"},
   1019     {"data:,looks/like/a.path",
   1020      "",
   1021      "",
   1022      L"\u65e5\u672c\u8a9e",  // Japanese Kanji.
   1023      L"\u65e5\u672c\u8a9e"},
   1024     // Dotfiles. Ensures preceeding period(s) stripped.
   1025     {"http://www.google.com/.test.html",
   1026     "",
   1027     "",
   1028     L"",
   1029     L"test.html"},
   1030     {"http://www.google.com/.test",
   1031     "",
   1032     "",
   1033     L"",
   1034     L"test"},
   1035     {"http://www.google.com/..test",
   1036     "",
   1037     "",
   1038     L"",
   1039     L"test"},
   1040   };
   1041   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(test_cases); ++i) {
   1042 #if defined(OS_WIN)
   1043     FilePath default_name(test_cases[i].default_filename);
   1044 #else
   1045     FilePath default_name(
   1046         base::SysWideToNativeMB(test_cases[i].default_filename));
   1047 #endif
   1048     FilePath filename = net::GetSuggestedFilename(
   1049         GURL(test_cases[i].url), test_cases[i].content_disp_header,
   1050         test_cases[i].referrer_charset, default_name);
   1051 #if defined(OS_WIN)
   1052     EXPECT_EQ(std::wstring(test_cases[i].expected_filename), filename.value())
   1053 #else
   1054     EXPECT_EQ(base::SysWideToNativeMB(test_cases[i].expected_filename),
   1055               filename.value())
   1056 #endif
   1057       << "Iteration " << i << ": " << test_cases[i].url;
   1058   }
   1059 }
   1060 
   1061 // This is currently a windows specific function.
   1062 #if defined(OS_WIN)
   1063 namespace {
   1064 
   1065 struct GetDirectoryListingEntryCase {
   1066   const wchar_t* name;
   1067   const char* raw_bytes;
   1068   bool is_dir;
   1069   int64 filesize;
   1070   base::Time time;
   1071   const char* expected;
   1072 };
   1073 
   1074 }  // namespace
   1075 TEST(NetUtilTest, GetDirectoryListingEntry) {
   1076   const GetDirectoryListingEntryCase test_cases[] = {
   1077     {L"Foo",
   1078      "",
   1079      false,
   1080      10000,
   1081      base::Time(),
   1082      "<script>addRow(\"Foo\",\"Foo\",0,\"9.8 kB\",\"\");</script>\n"},
   1083     {L"quo\"tes",
   1084      "",
   1085      false,
   1086      10000,
   1087      base::Time(),
   1088      "<script>addRow(\"quo\\\"tes\",\"quo%22tes\",0,\"9.8 kB\",\"\");</script>"
   1089          "\n"},
   1090     {L"quo\"tes",
   1091      "quo\"tes",
   1092      false,
   1093      10000,
   1094      base::Time(),
   1095      "<script>addRow(\"quo\\\"tes\",\"quo%22tes\",0,\"9.8 kB\",\"\");</script>"
   1096          "\n"},
   1097     // U+D55C0 U+AE00. raw_bytes is empty (either a local file with
   1098     // UTF-8/UTF-16 encoding or a remote file on an ftp server using UTF-8
   1099     {L"\xD55C\xAE00.txt",
   1100      "",
   1101      false,
   1102      10000,
   1103      base::Time(),
   1104      "<script>addRow(\"\\uD55C\\uAE00.txt\",\"%ED%95%9C%EA%B8%80.txt\""
   1105          ",0,\"9.8 kB\",\"\");</script>\n"},
   1106     // U+D55C0 U+AE00. raw_bytes is the corresponding EUC-KR sequence:
   1107     // a local or remote file in EUC-KR.
   1108     {L"\xD55C\xAE00.txt",
   1109      "\xC7\xD1\xB1\xDB.txt",
   1110      false,
   1111      10000,
   1112      base::Time(),
   1113      "<script>addRow(\"\\uD55C\\uAE00.txt\",\"%C7%D1%B1%DB.txt\""
   1114          ",0,\"9.8 kB\",\"\");</script>\n"},
   1115   };
   1116 
   1117   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(test_cases); ++i) {
   1118     const std::string results = net::GetDirectoryListingEntry(
   1119         WideToUTF16(test_cases[i].name),
   1120         test_cases[i].raw_bytes,
   1121         test_cases[i].is_dir,
   1122         test_cases[i].filesize,
   1123         test_cases[i].time);
   1124     EXPECT_EQ(test_cases[i].expected, results);
   1125   }
   1126 }
   1127 
   1128 #endif
   1129 
   1130 TEST(NetUtilTest, ParseHostAndPort) {
   1131   const struct {
   1132     const char* input;
   1133     bool success;
   1134     const char* expected_host;
   1135     int expected_port;
   1136   } tests[] = {
   1137     // Valid inputs:
   1138     {"foo:10", true, "foo", 10},
   1139     {"foo", true, "foo", -1},
   1140     {
   1141       "[1080:0:0:0:8:800:200C:4171]:11",
   1142       true,
   1143       "[1080:0:0:0:8:800:200C:4171]",
   1144       11,
   1145     },
   1146     // Invalid inputs:
   1147     {"foo:bar", false, "", -1},
   1148     {"foo:", false, "", -1},
   1149     {":", false, "", -1},
   1150     {":80", false, "", -1},
   1151     {"", false, "", -1},
   1152     {"porttoolong:300000", false, "", -1},
   1153     {"usrname@host", false, "", -1},
   1154     {"usrname:password@host", false, "", -1},
   1155     {":password@host", false, "", -1},
   1156     {":password@host:80", false, "", -1},
   1157     {":password@host", false, "", -1},
   1158     {"@host", false, "", -1},
   1159   };
   1160 
   1161   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
   1162     std::string host;
   1163     int port;
   1164     bool ok = net::ParseHostAndPort(tests[i].input, &host, &port);
   1165 
   1166     EXPECT_EQ(tests[i].success, ok);
   1167 
   1168     if (tests[i].success) {
   1169       EXPECT_EQ(tests[i].expected_host, host);
   1170       EXPECT_EQ(tests[i].expected_port, port);
   1171     }
   1172   }
   1173 }
   1174 
   1175 TEST(NetUtilTest, GetHostAndPort) {
   1176   const struct {
   1177     GURL url;
   1178     const char* expected_host_and_port;
   1179   } tests[] = {
   1180     { GURL("http://www.foo.com/x"), "www.foo.com:80"},
   1181     { GURL("http://www.foo.com:21/x"), "www.foo.com:21"},
   1182 
   1183     // For IPv6 literals should always include the brackets.
   1184     { GURL("http://[1::2]/x"), "[1::2]:80"},
   1185     { GURL("http://[::a]:33/x"), "[::a]:33"},
   1186   };
   1187   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
   1188     std::string host_and_port = net::GetHostAndPort(tests[i].url);
   1189     EXPECT_EQ(std::string(tests[i].expected_host_and_port), host_and_port);
   1190   }
   1191 }
   1192 
   1193 TEST(NetUtilTest, GetHostAndOptionalPort) {
   1194   const struct {
   1195     GURL url;
   1196     const char* expected_host_and_port;
   1197   } tests[] = {
   1198     { GURL("http://www.foo.com/x"), "www.foo.com"},
   1199     { GURL("http://www.foo.com:21/x"), "www.foo.com:21"},
   1200 
   1201     // For IPv6 literals should always include the brackets.
   1202     { GURL("http://[1::2]/x"), "[1::2]"},
   1203     { GURL("http://[::a]:33/x"), "[::a]:33"},
   1204   };
   1205   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
   1206     std::string host_and_port = net::GetHostAndOptionalPort(tests[i].url);
   1207     EXPECT_EQ(std::string(tests[i].expected_host_and_port), host_and_port);
   1208   }
   1209 }
   1210 
   1211 
   1212 TEST(NetUtilTest, NetAddressToString_IPv4) {
   1213   const struct {
   1214     uint8 addr[4];
   1215     const char* result;
   1216   } tests[] = {
   1217     {{0, 0, 0, 0}, "0.0.0.0"},
   1218     {{127, 0, 0, 1}, "127.0.0.1"},
   1219     {{192, 168, 0, 1}, "192.168.0.1"},
   1220   };
   1221 
   1222   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
   1223     const addrinfo* ai = GetIPv4Address(tests[i].addr);
   1224     std::string result = net::NetAddressToString(ai);
   1225     EXPECT_EQ(std::string(tests[i].result), result);
   1226   }
   1227 }
   1228 
   1229 TEST(NetUtilTest, NetAddressToString_IPv6) {
   1230   const struct {
   1231     uint8 addr[16];
   1232     const char* result;
   1233   } tests[] = {
   1234     {{0xFE, 0xDC, 0xBA, 0x98, 0x76, 0x54, 0x32, 0x10, 0xFE, 0xDC, 0xBA,
   1235       0x98, 0x76, 0x54, 0x32, 0x10},
   1236      "fedc:ba98:7654:3210:fedc:ba98:7654:3210"},
   1237   };
   1238 
   1239   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
   1240     const addrinfo* ai = GetIPv6Address(tests[i].addr);
   1241     std::string result = net::NetAddressToString(ai);
   1242     // Allow NetAddressToString() to fail, in case the system doesn't
   1243     // support IPv6.
   1244     if (!result.empty())
   1245       EXPECT_EQ(std::string(tests[i].result), result);
   1246   }
   1247 }
   1248 
   1249 TEST(NetUtilTest, GetHostName) {
   1250   // We can't check the result of GetHostName() directly, since the result
   1251   // will differ across machines. Our goal here is to simply exercise the
   1252   // code path, and check that things "look about right".
   1253   std::string hostname = net::GetHostName();
   1254   EXPECT_FALSE(hostname.empty());
   1255 }
   1256 
   1257 TEST(NetUtilTest, FormatUrl) {
   1258   const UrlTestData tests[] = {
   1259     {"Empty URL", "", L"", true, UnescapeRule::NORMAL, L"", 0},
   1260 
   1261     {"Simple URL",
   1262      "http://www.google.com/", L"", true, UnescapeRule::NORMAL,
   1263      L"http://www.google.com/", 7},
   1264 
   1265     {"With a port number and a reference",
   1266      "http://www.google.com:8080/#\xE3\x82\xB0", L"", true,
   1267      UnescapeRule::NORMAL,
   1268      L"http://www.google.com:8080/#\x30B0", 7},
   1269 
   1270     // -------- IDN tests --------
   1271     {"Japanese IDN with ja",
   1272      "http://xn--l8jvb1ey91xtjb.jp", L"ja", true, UnescapeRule::NORMAL,
   1273      L"http://\x671d\x65e5\x3042\x3055\x3072.jp/", 7},
   1274 
   1275     {"Japanese IDN with en",
   1276      "http://xn--l8jvb1ey91xtjb.jp", L"en", true, UnescapeRule::NORMAL,
   1277      L"http://xn--l8jvb1ey91xtjb.jp/", 7},
   1278 
   1279     {"Japanese IDN without any languages",
   1280      "http://xn--l8jvb1ey91xtjb.jp", L"", true, UnescapeRule::NORMAL,
   1281      // Single script is safe for empty languages.
   1282      L"http://\x671d\x65e5\x3042\x3055\x3072.jp/", 7},
   1283 
   1284     {"mailto: with Japanese IDN",
   1285      "mailto:foo (at) xn--l8jvb1ey91xtjb.jp", L"ja", true, UnescapeRule::NORMAL,
   1286      // GURL doesn't assume an email address's domain part as a host name.
   1287      L"mailto:foo (at) xn--l8jvb1ey91xtjb.jp", 7},
   1288 
   1289     {"file: with Japanese IDN",
   1290      "file://xn--l8jvb1ey91xtjb.jp/config.sys", L"ja", true,
   1291      UnescapeRule::NORMAL,
   1292      L"file://\x671d\x65e5\x3042\x3055\x3072.jp/config.sys", 7},
   1293 
   1294     {"ftp: with Japanese IDN",
   1295      "ftp://xn--l8jvb1ey91xtjb.jp/config.sys", L"ja", true,
   1296      UnescapeRule::NORMAL,
   1297      L"ftp://\x671d\x65e5\x3042\x3055\x3072.jp/config.sys", 6},
   1298 
   1299     // -------- omit_username_password flag tests --------
   1300     {"With username and password, omit_username_password=false",
   1301      "http://user:passwd@example.com/foo", L"", false, UnescapeRule::NORMAL,
   1302      L"http://user:passwd@example.com/foo", 19},
   1303 
   1304     {"With username and password, omit_username_password=true",
   1305      "http://user:passwd@example.com/foo", L"", true, UnescapeRule::NORMAL,
   1306      L"http://example.com/foo", 7},
   1307 
   1308     {"With username and no password",
   1309      "http://user@example.com/foo", L"", true, UnescapeRule::NORMAL,
   1310      L"http://example.com/foo", 7},
   1311 
   1312     {"Just '@' without username and password",
   1313      "http://@example.com/foo", L"", true, UnescapeRule::NORMAL,
   1314      L"http://example.com/foo", 7},
   1315 
   1316     // GURL doesn't think local-part of an email address is username for URL.
   1317     {"mailto:, omit_username_password=true",
   1318      "mailto:foo (at) example.com", L"", true, UnescapeRule::NORMAL,
   1319      L"mailto:foo (at) example.com", 7},
   1320 
   1321     // -------- unescape flag tests --------
   1322     {"Do not unescape",
   1323      "http://%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB.jp/"
   1324      "%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"
   1325      "?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB", L"en", true,
   1326      UnescapeRule::NONE,
   1327      // GURL parses %-encoded hostnames into Punycode.
   1328      L"http://xn--qcka1pmc.jp/%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"
   1329      L"?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB", 7},
   1330 
   1331     {"Unescape normally",
   1332      "http://%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB.jp/"
   1333      "%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"
   1334      "?q=%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB", L"en", true,
   1335      UnescapeRule::NORMAL,
   1336      L"http://xn--qcka1pmc.jp/\x30B0\x30FC\x30B0\x30EB"
   1337      L"?q=\x30B0\x30FC\x30B0\x30EB", 7},
   1338 
   1339     {"Unescape normally including unescape spaces",
   1340      "http://www.google.com/search?q=Hello%20World", L"en", true,
   1341      UnescapeRule::SPACES,
   1342      L"http://www.google.com/search?q=Hello World", 7},
   1343 
   1344     /*
   1345     {"unescape=true with some special characters",
   1346     "http://user%3A:%40passwd@example.com/foo%3Fbar?q=b%26z", L"", false, true,
   1347     L"http://user%3A:%40passwd@example.com/foo%3Fbar?q=b%26z", 25},
   1348     */
   1349     // Disabled: the resultant URL becomes "...user%253A:%2540passwd...".
   1350 
   1351     // -------- view-source: --------
   1352     {"view-source",
   1353      "view-source:http://xn--qcka1pmc.jp/", L"ja", true, UnescapeRule::NORMAL,
   1354      L"view-source:http://\x30B0\x30FC\x30B0\x30EB.jp/", 12 + 7},
   1355 
   1356     {"view-source of view-source",
   1357      "view-source:view-source:http://xn--qcka1pmc.jp/", L"ja", true,
   1358      UnescapeRule::NORMAL,
   1359      L"view-source:view-source:http://xn--qcka1pmc.jp/", 12},
   1360   };
   1361 
   1362   for (size_t i = 0; i < arraysize(tests); ++i) {
   1363     size_t prefix_len;
   1364     std::wstring formatted = net::FormatUrl(
   1365         GURL(tests[i].input), tests[i].languages, tests[i].omit,
   1366         tests[i].escape_rules, NULL, &prefix_len, NULL);
   1367     EXPECT_EQ(tests[i].output, formatted) << tests[i].description;
   1368     EXPECT_EQ(tests[i].prefix_len, prefix_len) << tests[i].description;
   1369   }
   1370 }
   1371 
   1372 TEST(NetUtilTest, FormatUrlParsed) {
   1373   // No unescape case.
   1374   url_parse::Parsed parsed;
   1375   std::wstring formatted = net::FormatUrl(
   1376       GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"
   1377            "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),
   1378       L"ja", false, UnescapeRule::NONE, &parsed, NULL, NULL);
   1379   EXPECT_EQ(L"http://%E3%82%B0:%E3%83%BC@\x30B0\x30FC\x30B0\x30EB.jp:8080"
   1380       L"/%E3%82%B0/?q=%E3%82%B0#\x30B0", formatted);
   1381   EXPECT_EQ(L"%E3%82%B0",
   1382       formatted.substr(parsed.username.begin, parsed.username.len));
   1383   EXPECT_EQ(L"%E3%83%BC",
   1384       formatted.substr(parsed.password.begin, parsed.password.len));
   1385   EXPECT_EQ(L"\x30B0\x30FC\x30B0\x30EB.jp",
   1386       formatted.substr(parsed.host.begin, parsed.host.len));
   1387   EXPECT_EQ(L"8080", formatted.substr(parsed.port.begin, parsed.port.len));
   1388   EXPECT_EQ(L"/%E3%82%B0/",
   1389       formatted.substr(parsed.path.begin, parsed.path.len));
   1390   EXPECT_EQ(L"q=%E3%82%B0",
   1391       formatted.substr(parsed.query.begin, parsed.query.len));
   1392   EXPECT_EQ(L"\x30B0", formatted.substr(parsed.ref.begin, parsed.ref.len));
   1393 
   1394   // Unescape case.
   1395   formatted = net::FormatUrl(
   1396       GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"
   1397            "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),
   1398       L"ja", false, UnescapeRule::NORMAL, &parsed, NULL, NULL);
   1399   EXPECT_EQ(L"http://\x30B0:\x30FC@\x30B0\x30FC\x30B0\x30EB.jp:8080"
   1400       L"/\x30B0/?q=\x30B0#\x30B0", formatted);
   1401   EXPECT_EQ(L"\x30B0",
   1402       formatted.substr(parsed.username.begin, parsed.username.len));
   1403   EXPECT_EQ(L"\x30FC",
   1404       formatted.substr(parsed.password.begin, parsed.password.len));
   1405   EXPECT_EQ(L"\x30B0\x30FC\x30B0\x30EB.jp",
   1406       formatted.substr(parsed.host.begin, parsed.host.len));
   1407   EXPECT_EQ(L"8080", formatted.substr(parsed.port.begin, parsed.port.len));
   1408   EXPECT_EQ(L"/\x30B0/", formatted.substr(parsed.path.begin, parsed.path.len));
   1409   EXPECT_EQ(L"q=\x30B0",
   1410       formatted.substr(parsed.query.begin, parsed.query.len));
   1411   EXPECT_EQ(L"\x30B0", formatted.substr(parsed.ref.begin, parsed.ref.len));
   1412 
   1413   // Omit_username_password + unescape case.
   1414   formatted = net::FormatUrl(
   1415       GURL("http://\xE3\x82\xB0:\xE3\x83\xBC@xn--qcka1pmc.jp:8080/"
   1416            "%E3%82%B0/?q=%E3%82%B0#\xE3\x82\xB0"),
   1417       L"ja", true, UnescapeRule::NORMAL, &parsed, NULL, NULL);
   1418   EXPECT_EQ(L"http://\x30B0\x30FC\x30B0\x30EB.jp:8080"
   1419       L"/\x30B0/?q=\x30B0#\x30B0", formatted);
   1420   EXPECT_FALSE(parsed.username.is_valid());
   1421   EXPECT_FALSE(parsed.password.is_valid());
   1422   EXPECT_EQ(L"\x30B0\x30FC\x30B0\x30EB.jp",
   1423       formatted.substr(parsed.host.begin, parsed.host.len));
   1424   EXPECT_EQ(L"8080", formatted.substr(parsed.port.begin, parsed.port.len));
   1425   EXPECT_EQ(L"/\x30B0/", formatted.substr(parsed.path.begin, parsed.path.len));
   1426   EXPECT_EQ(L"q=\x30B0",
   1427       formatted.substr(parsed.query.begin, parsed.query.len));
   1428   EXPECT_EQ(L"\x30B0", formatted.substr(parsed.ref.begin, parsed.ref.len));
   1429 
   1430   // View-source case.
   1431   formatted = net::FormatUrl(
   1432       GURL("view-source:http://user:passwd@host:81/path?query#ref"),
   1433       L"", true, UnescapeRule::NORMAL, &parsed, NULL, NULL);
   1434   EXPECT_EQ(L"view-source:http://host:81/path?query#ref", formatted);
   1435   EXPECT_EQ(L"view-source:http",
   1436       formatted.substr(parsed.scheme.begin, parsed.scheme.len));
   1437   EXPECT_FALSE(parsed.username.is_valid());
   1438   EXPECT_FALSE(parsed.password.is_valid());
   1439   EXPECT_EQ(L"host", formatted.substr(parsed.host.begin, parsed.host.len));
   1440   EXPECT_EQ(L"81", formatted.substr(parsed.port.begin, parsed.port.len));
   1441   EXPECT_EQ(L"/path", formatted.substr(parsed.path.begin, parsed.path.len));
   1442   EXPECT_EQ(L"query", formatted.substr(parsed.query.begin, parsed.query.len));
   1443   EXPECT_EQ(L"ref", formatted.substr(parsed.ref.begin, parsed.ref.len));
   1444 }
   1445 
   1446 TEST(NetUtilTest, FormatUrlAdjustOffset) {
   1447   const AdjustOffsetCase basic_cases[] = {
   1448     {0, 0},
   1449     {3, 3},
   1450     {5, 5},
   1451     {6, 6},
   1452     {13, 13},
   1453     {21, 21},
   1454     {22, 22},
   1455     {23, 23},
   1456     {25, 25},
   1457     {26, std::wstring::npos},
   1458     {500000, std::wstring::npos},
   1459     {std::wstring::npos, std::wstring::npos},
   1460   };
   1461   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(basic_cases); ++i) {
   1462     size_t offset = basic_cases[i].input_offset;
   1463     net::FormatUrl(GURL("http://www.google.com/foo/"), L"en", true,
   1464                    UnescapeRule::NORMAL, NULL, NULL, &offset);
   1465     EXPECT_EQ(basic_cases[i].output_offset, offset);
   1466   }
   1467 
   1468   const struct {
   1469     const char* input_url;
   1470     size_t input_offset;
   1471     size_t output_offset;
   1472   } omit_auth_cases[] = {
   1473     {"http://foo:bar@www.google.com/", 6, 6},
   1474     {"http://foo:bar@www.google.com/", 7, 7},
   1475     {"http://foo:bar@www.google.com/", 8, std::wstring::npos},
   1476     {"http://foo:bar@www.google.com/", 10, std::wstring::npos},
   1477     {"http://foo:bar@www.google.com/", 11, std::wstring::npos},
   1478     {"http://foo:bar@www.google.com/", 14, std::wstring::npos},
   1479     {"http://foo:bar@www.google.com/", 15, 7},
   1480     {"http://foo:bar@www.google.com/", 25, 17},
   1481     {"http://foo@www.google.com/", 9, std::wstring::npos},
   1482     {"http://foo@www.google.com/", 11, 7},
   1483   };
   1484   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(omit_auth_cases); ++i) {
   1485     size_t offset = omit_auth_cases[i].input_offset;
   1486     net::FormatUrl(GURL(omit_auth_cases[i].input_url), L"en", true,
   1487                    UnescapeRule::NORMAL, NULL, NULL, &offset);
   1488     EXPECT_EQ(omit_auth_cases[i].output_offset, offset);
   1489   }
   1490 
   1491   const AdjustOffsetCase view_source_cases[] = {
   1492     {0, 0},
   1493     {3, 3},
   1494     {11, 11},
   1495     {12, 12},
   1496     {13, 13},
   1497     {19, 19},
   1498     {20, std::wstring::npos},
   1499     {23, 19},
   1500     {26, 22},
   1501     {std::wstring::npos, std::wstring::npos},
   1502   };
   1503   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(view_source_cases); ++i) {
   1504     size_t offset = view_source_cases[i].input_offset;
   1505     net::FormatUrl(GURL("view-source:http://foo@www.google.com/"), L"en", true,
   1506                    UnescapeRule::NORMAL, NULL, NULL, &offset);
   1507     EXPECT_EQ(view_source_cases[i].output_offset, offset);
   1508   }
   1509 
   1510   const AdjustOffsetCase idn_hostname_cases[] = {
   1511     {8, std::wstring::npos},
   1512     {16, std::wstring::npos},
   1513     {24, std::wstring::npos},
   1514     {25, 12},
   1515     {30, 17},
   1516   };
   1517   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(idn_hostname_cases); ++i) {
   1518     size_t offset = idn_hostname_cases[i].input_offset;
   1519     // "http://\x671d\x65e5\x3042\x3055\x3072.jp/foo/"
   1520     net::FormatUrl(GURL("http://xn--l8jvb1ey91xtjb.jp/foo/"), L"ja", true,
   1521                    UnescapeRule::NORMAL, NULL, NULL, &offset);
   1522     EXPECT_EQ(idn_hostname_cases[i].output_offset, offset);
   1523   }
   1524 
   1525   const AdjustOffsetCase unescape_cases[] = {
   1526     {25, 25},
   1527     {26, std::wstring::npos},
   1528     {27, std::wstring::npos},
   1529     {28, 26},
   1530     {35, std::wstring::npos},
   1531     {41, 31},
   1532     {59, 33},
   1533     {60, std::wstring::npos},
   1534     {67, std::wstring::npos},
   1535     {68, std::wstring::npos},
   1536   };
   1537   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(unescape_cases); ++i) {
   1538     size_t offset = unescape_cases[i].input_offset;
   1539     // "http://www.google.com/foo bar/\x30B0\x30FC\x30B0\x30EB"
   1540     net::FormatUrl(GURL(
   1541         "http://www.google.com/foo%20bar/%E3%82%B0%E3%83%BC%E3%82%B0%E3%83%AB"),
   1542         L"en", true, UnescapeRule::SPACES, NULL, NULL, &offset);
   1543     EXPECT_EQ(unescape_cases[i].output_offset, offset);
   1544   }
   1545 
   1546   const AdjustOffsetCase ref_cases[] = {
   1547     {30, 30},
   1548     {31, 31},
   1549     {32, std::wstring::npos},
   1550     {34, 32},
   1551     {37, 33},
   1552     {38, std::wstring::npos},
   1553   };
   1554   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(ref_cases); ++i) {
   1555     size_t offset = ref_cases[i].input_offset;
   1556     // "http://www.google.com/foo.html#\x30B0\x30B0z"
   1557     net::FormatUrl(GURL(
   1558         "http://www.google.com/foo.html#\xE3\x82\xB0\xE3\x82\xB0z"), L"en",
   1559         true, UnescapeRule::NORMAL, NULL, NULL, &offset);
   1560     EXPECT_EQ(ref_cases[i].output_offset, offset);
   1561   }
   1562 }
   1563 
   1564 TEST(NetUtilTest, SimplifyUrlForRequest) {
   1565   struct {
   1566     const char* input_url;
   1567     const char* expected_simplified_url;
   1568   } tests[] = {
   1569     {
   1570       // Reference section should be stripped.
   1571       "http://www.google.com:78/foobar?query=1#hash",
   1572       "http://www.google.com:78/foobar?query=1",
   1573     },
   1574     {
   1575       // Reference section can itself contain #.
   1576       "http://192.168.0.1?query=1#hash#10#11#13#14",
   1577       "http://192.168.0.1?query=1",
   1578     },
   1579     { // Strip username/password.
   1580       "http://user:pass@google.com",
   1581       "http://google.com/",
   1582     },
   1583     { // Strip both the reference and the username/password.
   1584       "http://user:pass@google.com:80/sup?yo#X#X",
   1585       "http://google.com/sup?yo",
   1586     },
   1587     { // Try an HTTPS URL -- strip both the reference and the username/password.
   1588       "https://user:pass@google.com:80/sup?yo#X#X",
   1589       "https://google.com:80/sup?yo",
   1590     },
   1591     { // Try an FTP URL -- strip both the reference and the username/password.
   1592       "ftp://user:pass@google.com:80/sup?yo#X#X",
   1593       "ftp://google.com:80/sup?yo",
   1594     },
   1595     { // Try an standard URL with unknow scheme.
   1596       "foobar://user:pass (at) google.com:80/sup?yo#X#X",
   1597       "foobar://google.com:80/sup?yo",
   1598     },
   1599   };
   1600   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(tests); ++i) {
   1601     SCOPED_TRACE(StringPrintf("Test[%" PRIuS "]: %s", i, tests[i].input_url));
   1602     GURL input_url(GURL(tests[i].input_url));
   1603     GURL expected_url(GURL(tests[i].expected_simplified_url));
   1604     EXPECT_EQ(expected_url, net::SimplifyUrlForRequest(input_url));
   1605   }
   1606 }
   1607 
   1608 TEST(NetUtilTest, SetExplicitlyAllowedPortsTest) {
   1609   std::wstring invalid[] = { L"1,2,a", L"'1','2'", L"1, 2, 3", L"1 0,11,12" };
   1610   std::wstring valid[] = { L"", L"1", L"1,2", L"1,2,3", L"10,11,12,13" };
   1611 
   1612   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(invalid); ++i) {
   1613     net::SetExplicitlyAllowedPorts(invalid[i]);
   1614     EXPECT_EQ(0, static_cast<int>(net::explicitly_allowed_ports.size()));
   1615   }
   1616 
   1617   for (size_t i = 0; i < ARRAYSIZE_UNSAFE(valid); ++i) {
   1618     net::SetExplicitlyAllowedPorts(valid[i]);
   1619     EXPECT_EQ(i, net::explicitly_allowed_ports.size());
   1620   }
   1621 }
   1622