diff options
Diffstat (limited to 'url')
| -rw-r--r-- | url/url.gyp | 1 | ||||
| -rw-r--r-- | url/url_canon_icu.cc | 66 | ||||
| -rw-r--r-- | url/url_canon_unittest.cc | 151 |
3 files changed, 195 insertions, 23 deletions
diff --git a/url/url.gyp b/url/url.gyp index fe4d5fd..2c81443 100644 --- a/url/url.gyp +++ b/url/url.gyp @@ -15,6 +15,7 @@ 'type': '<(component)', 'dependencies': [ '../base/base.gyp:base', + '../base/third_party/dynamic_annotations/dynamic_annotations.gyp:dynamic_annotations', '../third_party/icu/icu.gyp:icudata', '../third_party/icu/icu.gyp:icui18n', '../third_party/icu/icu.gyp:icuuc', diff --git a/url/url_canon_icu.cc b/url/url_canon_icu.cc index 8d6bdfc..cabbbf2 100644 --- a/url/url_canon_icu.cc +++ b/url/url_canon_icu.cc @@ -7,6 +7,7 @@ #include <stdlib.h> #include <string.h> +#include "base/lazy_instance.h" #include "base/logging.h" #include "third_party/icu/source/common/unicode/ucnv.h" #include "third_party/icu/source/common/unicode/ucnv_cb.h" @@ -71,6 +72,40 @@ class AppendHandlerInstaller { const void* old_context_; }; +// A wrapper to use LazyInstance<>::Leaky with ICU's UIDNA, a C pointer to +// a UTS46/IDNA 2008 handling object opened with uidna_openUTS46(). +// +// We use UTS46 with BiDiCheck to migrate from IDNA 2003 (with unassigned +// code points allowed) to IDNA 2008 with +// the backward compatibility in mind. What it does: +// +// 1. Use the up-to-date Unicode data. +// 2. Define a case folding/mapping with the up-to-date Unicode data as +// in IDNA 2003. +// 3. Use transitional mechanism for 4 deviation characters (sharp-s, +// final sigma, ZWJ and ZWNJ) for now. +// 4. Continue to allow symbols and punctuations. +// 5. Apply new BiDi check rules more permissive than the IDNA 2003 BiDI rules. +// 6. Do not apply STD3 rules +// 7. Do not allow unassigned code points. +// +// It also closely matches what IE 10 does except for the BiDi check ( +// http://goo.gl/3XBhqw ). +// See http://http://unicode.org/reports/tr46/ and references therein +// for more details. +struct UIDNAWrapper { + UIDNAWrapper() { + UErrorCode err = U_ZERO_ERROR; + // TODO(jungshik): Change options as different parties (browsers, + // registrars, search engines) converge toward a consensus. + value = uidna_openUTS46(UIDNA_CHECK_BIDI, &err); + if (U_FAILURE(err)) + value = NULL; + } + + UIDNA* value; +}; + } // namespace ICUCharsetConverter::ICUCharsetConverter(UConverter* converter) @@ -107,6 +142,9 @@ void ICUCharsetConverter::ConvertFromUTF16(const base::char16* input, } while (true); } +static base::LazyInstance<UIDNAWrapper>::Leaky + g_uidna = LAZY_INSTANCE_INITIALIZER; + // Converts the Unicode input representing a hostname to ASCII using IDN rules. // The output must be ASCII, but is represented as wide characters. // @@ -116,25 +154,33 @@ void ICUCharsetConverter::ConvertFromUTF16(const base::char16* input, // the length of the output will be set to the length of the new host name. // // On error, this will return false. The output in this case is undefined. +// TODO(jungshik): use UTF-8/ASCII version of nameToASCII. +// Change the function signature and callers accordingly to avoid unnecessary +// conversions in our code. In addition, consider using icu::IDNA's UTF-8/ASCII +// version with StringByteSink. That way, we can avoid C wrappers and additional +// string conversion. bool IDNToASCII(const base::char16* src, int src_len, CanonOutputW* output) { DCHECK(output->length() == 0); // Output buffer is assumed empty. + + UIDNA* uidna = g_uidna.Get().value; + DCHECK(uidna != NULL); while (true) { - // Use ALLOW_UNASSIGNED to be more tolerant of hostnames that violate - // the spec (which do exist). This does not present any risk and is a - // little more future proof. UErrorCode err = U_ZERO_ERROR; - int num_converted = uidna_IDNToASCII(src, src_len, output->data(), - output->capacity(), - UIDNA_ALLOW_UNASSIGNED, NULL, &err); - if (err == U_ZERO_ERROR) { - output->set_length(num_converted); + UIDNAInfo info = UIDNA_INFO_INITIALIZER; + int output_length = uidna_nameToASCII(uidna, src, src_len, output->data(), + output->capacity(), &info, &err); + if (U_SUCCESS(err) && info.errors == 0) { + output->set_length(output_length); return true; } - if (err != U_BUFFER_OVERFLOW_ERROR) + + // TODO(jungshik): Look at info.errors to handle them case-by-case basis + // if necessary. + if (err != U_BUFFER_OVERFLOW_ERROR || info.errors != 0) return false; // Unknown error, give up. // Not enough room in our buffer, expand. - output->Resize(output->capacity() * 2); + output->Resize(output_length); } } diff --git a/url/url_canon_unittest.cc b/url/url_canon_unittest.cc index 2e93f92..1b59a00 100644 --- a/url/url_canon_unittest.cc +++ b/url/url_canon_unittest.cc @@ -399,19 +399,138 @@ TEST(URLCanonTest, Host) { {"%ef%bc%85%ef%bc%90%ef%bc%90.com", L"%ef%bc%85%ef%bc%90%ef%bc%90.com", "%00.com", url_parse::Component(0, 7), CanonHostInfo::BROKEN, -1, ""}, // Basic IDN support, UTF-8 and UTF-16 input should be converted to IDN {"\xe4\xbd\xa0\xe5\xa5\xbd\xe4\xbd\xa0\xe5\xa5\xbd", L"\x4f60\x597d\x4f60\x597d", "xn--6qqa088eba", url_parse::Component(0, 14), CanonHostInfo::NEUTRAL, -1, ""}, + // See http://unicode.org/cldr/utility/idna.jsp for other + // examples/experiments and http://goo.gl/7yG11o + // for the full list of characters handled differently by + // IDNA 2003, UTS 46 (http://unicode.org/reports/tr46/ ) and IDNA 2008. + + // 4 Deviation characters are mapped/ignored in UTS 46 transitional + // mechansm. UTS 46, table 4 row (g). + // Sharp-s is mapped to 'ss' in UTS 46 and IDNA 2003. + // Otherwise, it'd be "xn--fuball-cta.de". + {"fu\xc3\x9f" "ball.de", L"fu\x00df" L"ball.de", "fussball.de", + url_parse::Component(0, 11), CanonHostInfo::NEUTRAL, -1, ""}, + // Final-sigma (U+03C3) is mapped to regular sigma (U+03C2). + // Otherwise, it'd be "xn--wxaijb9b". + {"\xcf\x83\xcf\x8c\xce\xbb\xce\xbf\xcf\x82", L"\x3c3\x3cc\x3bb\x3bf\x3c2", + "xn--wxaikc6b", url_parse::Component(0, 12), + CanonHostInfo::NEUTRAL, -1, ""}, + // ZWNJ (U+200C) and ZWJ (U+200D) are mapped away in UTS 46 transitional + // handling as well as in IDNA 2003. + {"a\xe2\x80\x8c" "b\xe2\x80\x8d" "c", L"a\x200c" L"b\x200d" L"c", "abc", + url_parse::Component(0, 3), CanonHostInfo::NEUTRAL, -1, ""}, + // ZWJ between Devanagari characters is still mapped away in UTS 46 + // transitional handling. IDNA 2008 would give xn--11bo0mv54g. + {"\xe0\xa4\x95\xe0\xa5\x8d\xe2\x80\x8d\xe0\xa4\x9c", + L"\x915\x94d\x200d\x91c", "xn--11bo0m", + url_parse::Component(0, 10), CanonHostInfo::NEUTRAL, -1, ""}, + // Fullwidth exclamation mark is disallowed. UTS 46, table 4, row (b) + // However, we do allow this at the moment because we don't use + // STD3 rules and canonicalize full-width ASCII to ASCII. + {"wow\xef\xbc\x81", L"wow\xff01", "wow%21", + url_parse::Component(0, 6), CanonHostInfo::NEUTRAL, -1, ""}, + // U+2132 (turned capital F) is disallowed. UTS 46, table 4, row (c) + // Allowed in IDNA 2003, but the mapping changed after Unicode 3.2 + {"\xe2\x84\xb2oo", L"\x2132oo", "%E2%84%B2oo", + url_parse::Component(0, 11), CanonHostInfo::BROKEN, -1, ""}, + // U+2F868 (CJK Comp) is disallowed. UTS 46, table 4, row (d) + // Allowed in IDNA 2003, but the mapping changed after Unicode 3.2 + {"\xf0\xaf\xa1\xa8\xe5\xa7\xbb.cn", L"\xd87e\xdc68\x59fb.cn", + "%F0%AF%A1%A8%E5%A7%BB.cn", + url_parse::Component(0, 24), CanonHostInfo::BROKEN, -1, ""}, + // Maps uppercase letters to lower case letters. UTS 46 table 4 row (e) + {"M\xc3\x9cNCHEN", L"M\xdcNCHEN", "xn--mnchen-3ya", + url_parse::Component(0, 14), CanonHostInfo::NEUTRAL, -1, ""}, + // Symbol/punctuations are allowed in IDNA 2003/UTS46. + // Not allowed in IDNA 2008. UTS 46 table 4 row (f). + {"\xe2\x99\xa5ny.us", L"\x2665ny.us", "xn--ny-s0x.us", + url_parse::Component(0, 13), CanonHostInfo::NEUTRAL, -1, ""}, + // U+11013 is new in Unicode 6.0 and is allowed. UTS 46 table 4, row (h) + // We used to allow it because we passed through unassigned code points. + {"\xf0\x91\x80\x93.com", L"\xd804\xdc13.com", "xn--n00d.com", + url_parse::Component(0, 12), CanonHostInfo::NEUTRAL, -1, ""}, + // U+0602 is disallowed in UTS46/IDNA 2008. UTS 46 table 4, row(i) + // Used to be allowed in INDA 2003. + {"\xd8\x82.eg", L"\x602.eg", "%D8%82.eg", + url_parse::Component(0, 9), CanonHostInfo::BROKEN, -1, ""}, + // U+20B7 is new in Unicode 5.2 (not a part of IDNA 2003 based + // on Unicode 3.2). We did allow it in the past because we let unassigned + // code point pass. We continue to allow it even though it's a + // "punctuation and symbol" blocked in IDNA 2008. + // UTS 46 table 4, row (j) + {"\xe2\x82\xb7.com", L"\x20b7.com", "xn--wzg.com", + url_parse::Component(0, 11), CanonHostInfo::NEUTRAL, -1, ""}, + // Maps uppercase letters to lower case letters. + // In IDNA 2003, it's allowed without case-folding + // ( xn--bc-7cb.com ) because it's not defined in Unicode 3.2 + // (added in Unicode 4.1). UTS 46 table 4 row (k) + {"bc\xc8\xba.com", L"bc\x23a.com", "xn--bc-is1a.com", + url_parse::Component(0, 15), CanonHostInfo::NEUTRAL, -1, ""}, + // BiDi check test + // "Divehi" in Divehi (Thaana script) ends with BidiClass=NSM. + // Disallowed in IDNA 2003 but now allowed in UTS 46/IDNA 2008. + {"\xde\x8b\xde\xa8\xde\x88\xde\xac\xde\x80\xde\xa8", + L"\x78b\x7a8\x788\x7ac\x780\x7a8", "xn--hqbpi0jcw", + url_parse::Component(0, 13), CanonHostInfo::NEUTRAL, -1, ""}, + // Disallowed in both IDNA 2003 and 2008 with BiDi check. + // Labels starting with a RTL character cannot end with a LTR character. + {"\xd8\xac\xd8\xa7\xd8\xb1xyz", L"\x62c\x627\x631xyz", + "%D8%AC%D8%A7%D8%B1xyz", url_parse::Component(0, 21), + CanonHostInfo::BROKEN, -1, ""}, + // Labels starting with a RTL character can end with BC=EN (European + // number). Disallowed in IDNA 2003 but now allowed. + {"\xd8\xac\xd8\xa7\xd8\xb1" "2", L"\x62c\x627\x631" L"2", + "xn--2-ymcov", url_parse::Component(0, 11), + CanonHostInfo::NEUTRAL, -1, ""}, + // Labels starting with a RTL character cannot have "L" characters + // even if it ends with an BC=EN. Disallowed in both IDNA 2003/2008. + {"\xd8\xac\xd8\xa7\xd8\xb1xy2", L"\x62c\x627\x631xy2", + "%D8%AC%D8%A7%D8%B1xy2", url_parse::Component(0, 21), + CanonHostInfo::BROKEN, -1, ""}, + // Labels starting with a RTL character can end with BC=AN (Arabic number) + // Disallowed in IDNA 2003, but now allowed. + {"\xd8\xac\xd8\xa7\xd8\xb1\xd9\xa2", L"\x62c\x627\x631\x662", + "xn--mgbjq0r", url_parse::Component(0, 11), + CanonHostInfo::NEUTRAL, -1, ""}, + // Labels starting with a RTL character cannot have "L" characters + // even if it ends with an BC=AN (Arabic number). + // Disallowed in both IDNA 2003/2008. + {"\xd8\xac\xd8\xa7\xd8\xb1xy\xd9\xa2", L"\x62c\x627\x631xy\x662", + "%D8%AC%D8%A7%D8%B1xy%D9%A2", url_parse::Component(0, 26), + CanonHostInfo::BROKEN, -1, ""}, + // Labels starting with a RTL character cannot mix BC=EN and BC=AN + {"\xd8\xac\xd8\xa7\xd8\xb1xy2\xd9\xa2", L"\x62c\x627\x631xy2\x662", + "%D8%AC%D8%A7%D8%B1xy2%D9%A2", url_parse::Component(0, 27), + CanonHostInfo::BROKEN, -1, ""}, + // As of Unicode 6.2, U+20CF is not assigned. We do not allow it. + {"\xe2\x83\x8f.com", L"\x20cf.com", "%E2%83%8F.com", + url_parse::Component(0, 13), CanonHostInfo::BROKEN, -1, ""}, + // U+0080 is not allowed. + {"\xc2\x80.com", L"\x80.com", "%C2%80.com", + url_parse::Component(0, 10), CanonHostInfo::BROKEN, -1, ""}, + // Mixed UTF-8 and escaped UTF-8 (narrow case) and UTF-16 and escaped // Mixed UTF-8 and escaped UTF-8 (narrow case) and UTF-16 and escaped // UTF-8 (wide case). The output should be equivalent to the true wide // character input above). - {"%E4%BD%A0%E5%A5%BD\xe4\xbd\xa0\xe5\xa5\xbd", L"%E4%BD%A0%E5%A5%BD\x4f60\x597d", "xn--6qqa088eba", url_parse::Component(0, 14), CanonHostInfo::NEUTRAL, -1, ""}, + {"%E4%BD%A0%E5%A5%BD\xe4\xbd\xa0\xe5\xa5\xbd", + L"%E4%BD%A0%E5%A5%BD\x4f60\x597d", "xn--6qqa088eba", + url_parse::Component(0, 14), CanonHostInfo::NEUTRAL, -1, ""}, // Invalid escaped characters should fail and the percents should be // escaped. - {"%zz%66%a", L"%zz%66%a", "%25zzf%25a", url_parse::Component(0, 10), CanonHostInfo::BROKEN, -1, ""}, + {"%zz%66%a", L"%zz%66%a", "%25zzf%25a", url_parse::Component(0, 10), + CanonHostInfo::BROKEN, -1, ""}, // If we get an invalid character that has been escaped. - {"%25", L"%25", "%25", url_parse::Component(0, 3), CanonHostInfo::BROKEN, -1, ""}, - {"hello%00", L"hello%00", "hello%00", url_parse::Component(0, 8), CanonHostInfo::BROKEN, -1, ""}, + {"%25", L"%25", "%25", url_parse::Component(0, 3), + CanonHostInfo::BROKEN, -1, ""}, + {"hello%00", L"hello%00", "hello%00", url_parse::Component(0, 8), + CanonHostInfo::BROKEN, -1, ""}, // Escaped numbers should be treated like IP addresses if they are. - {"%30%78%63%30%2e%30%32%35%30.01", L"%30%78%63%30%2e%30%32%35%30.01", "192.168.0.1", url_parse::Component(0, 11), CanonHostInfo::IPV4, 3, "C0A80001"}, - {"%30%78%63%30%2e%30%32%35%30.01%2e", L"%30%78%63%30%2e%30%32%35%30.01%2e", "192.168.0.1", url_parse::Component(0, 11), CanonHostInfo::IPV4, 3, "C0A80001"}, + {"%30%78%63%30%2e%30%32%35%30.01", L"%30%78%63%30%2e%30%32%35%30.01", + "192.168.0.1", url_parse::Component(0, 11), CanonHostInfo::IPV4, 3, + "C0A80001"}, + {"%30%78%63%30%2e%30%32%35%30.01%2e", L"%30%78%63%30%2e%30%32%35%30.01%2e", + "192.168.0.1", url_parse::Component(0, 11), CanonHostInfo::IPV4, 3, + "C0A80001"}, // Invalid escaping should trigger the regular host error handling. {"%3g%78%63%30%2e%30%32%35%30%2E.01", L"%3g%78%63%30%2e%30%32%35%30%2E.01", "%253gxc0.0250..01", url_parse::Component(0, 17), CanonHostInfo::BROKEN, -1, ""}, // Something that isn't exactly an IP should get treated as a host and @@ -423,9 +542,12 @@ TEST(URLCanonTest, Host) { // Broken IP addresses get marked as such. {"192.168.0.257", L"192.168.0.257", "192.168.0.257", url_parse::Component(0, 13), CanonHostInfo::BROKEN, -1, ""}, {"[google.com]", L"[google.com]", "[google.com]", url_parse::Component(0, 12), CanonHostInfo::BROKEN, -1, ""}, - // Cyrillic letter followed buy ( should return punicode for ( escaped before punicode string was created. I.e. - // if ( is escaped after punicode is created we would get xn--%28-8tb (incorrect). - {"\xd1\x82(", L"\x0442(", "xn--%28-7ed", url_parse::Component(0, 11), CanonHostInfo::NEUTRAL, -1, ""}, + // Cyrillic letter followed by '(' should return punycode for '(' escaped + // before punycode string was created. I.e. + // if '(' is escaped after punycode is created we would get xn--%28-8tb + // (incorrect). + {"\xd1\x82(", L"\x0442(", "xn--%28-7ed", url_parse::Component(0, 11), + CanonHostInfo::NEUTRAL, -1, ""}, // Address with all hexidecimal characters with leading number of 1<<32 // or greater and should return NEUTRAL rather than BROKEN if not all // components are numbers. @@ -453,10 +575,13 @@ TEST(URLCanonTest, Host) { output.Complete(); EXPECT_EQ(host_cases[i].expected_family != CanonHostInfo::BROKEN, - success); - EXPECT_EQ(std::string(host_cases[i].expected), out_str); - EXPECT_EQ(host_cases[i].expected_component.begin, out_comp.begin); - EXPECT_EQ(host_cases[i].expected_component.len, out_comp.len); + success) << "for input: " << host_cases[i].input8; + EXPECT_EQ(std::string(host_cases[i].expected), out_str) << + "for input: " << host_cases[i].input8; + EXPECT_EQ(host_cases[i].expected_component.begin, out_comp.begin) << + "for input: " << host_cases[i].input8; + EXPECT_EQ(host_cases[i].expected_component.len, out_comp.len) << + "for input: " << host_cases[i].input8; } // Wide version. |