OLD | NEW |
1 // Copyright 2013 The Chromium Authors. All rights reserved. | 1 // Copyright 2013 The Chromium Authors. All rights reserved. |
2 // Use of this source code is governed by a BSD-style license that can be | 2 // Use of this source code is governed by a BSD-style license that can be |
3 // found in the LICENSE file. | 3 // found in the LICENSE file. |
4 | 4 |
5 // Functions for canonicalizing "path" URLs. Not to be confused with the path | 5 // Functions for canonicalizing "path" URLs. Not to be confused with the path |
6 // of a URL, these are URLs that have no authority section, only a path. For | 6 // of a URL, these are URLs that have no authority section, only a path. For |
7 // example, "javascript:" and "data:". | 7 // example, "javascript:" and "data:". |
8 | 8 |
9 #include "url/url_canon.h" | 9 #include "url/url_canon.h" |
10 #include "url/url_canon_internal.h" | 10 #include "url/url_canon_internal.h" |
11 | 11 |
12 namespace url { | 12 namespace url { |
13 | 13 |
14 namespace { | 14 namespace { |
15 | 15 |
16 // Canonicalize the given |component| from |source| into |output| and | 16 // Canonicalize the given |component| from |source| into |output| and |
17 // |new_component|. If |separator| is non-zero, it is pre-pended to |ouput| | 17 // |new_component|. If |separator| is non-zero, it is pre-pended to |ouput| |
18 // prior to the canonicalized component; i.e. for the '?' or '#' characters. | 18 // prior to the canonicalized component; i.e. for the '?' or '#' characters. |
19 template<typename CHAR, typename UCHAR> | 19 template<typename CHAR, typename UCHAR> |
20 bool DoCanonicalizePathComponent(const CHAR* source, | 20 bool DoCanonicalizePathComponent(const CHAR* source, |
21 const Component& component, | 21 const Component& component, |
22 CHAR seperator, | 22 char separator, |
23 CanonOutput* output, | 23 CanonOutput* output, |
24 Component* new_component) { | 24 Component* new_component) { |
25 bool success = true; | 25 bool success = true; |
26 if (component.is_valid()) { | 26 if (component.is_valid()) { |
27 if (seperator) | 27 if (separator) |
28 output->push_back(seperator); | 28 output->push_back(separator); |
29 // Copy the path using path URL's more lax escaping rules (think for | 29 // Copy the path using path URL's more lax escaping rules (think for |
30 // javascript:). We convert to UTF-8 and escape non-ASCII, but leave all | 30 // javascript:). We convert to UTF-8 and escape non-ASCII, but leave all |
31 // ASCII characters alone. This helps readability of JavaStript. | 31 // ASCII characters alone. This helps readability of JavaStript. |
32 new_component->begin = output->length(); | 32 new_component->begin = output->length(); |
33 int end = component.end(); | 33 int end = component.end(); |
34 for (int i = component.begin; i < end; i++) { | 34 for (int i = component.begin; i < end; i++) { |
35 UCHAR uch = static_cast<UCHAR>(source[i]); | 35 UCHAR uch = static_cast<UCHAR>(source[i]); |
36 if (uch < 0x20 || uch >= 0x80) | 36 if (uch < 0x20 || uch >= 0x80) |
37 success &= AppendUTF8EscapedChar(source, &i, end, output); | 37 success &= AppendUTF8EscapedChar(source, &i, end, output); |
38 else | 38 else |
(...skipping 18 matching lines...) Expand all Loading... |
57 | 57 |
58 // We assume there's no authority for path URLs. Note that hosts should never | 58 // We assume there's no authority for path URLs. Note that hosts should never |
59 // have -1 length. | 59 // have -1 length. |
60 new_parsed->username.reset(); | 60 new_parsed->username.reset(); |
61 new_parsed->password.reset(); | 61 new_parsed->password.reset(); |
62 new_parsed->host.reset(); | 62 new_parsed->host.reset(); |
63 new_parsed->port.reset(); | 63 new_parsed->port.reset(); |
64 // We allow path URLs to have the path, query and fragment components, but we | 64 // We allow path URLs to have the path, query and fragment components, but we |
65 // will canonicalize each of the via the weaker path URL rules. | 65 // will canonicalize each of the via the weaker path URL rules. |
66 success &= DoCanonicalizePathComponent<CHAR, UCHAR>( | 66 success &= DoCanonicalizePathComponent<CHAR, UCHAR>( |
67 source.path, parsed.path, 0, output, &new_parsed->path); | 67 source.path, parsed.path, '\0', output, &new_parsed->path); |
68 success &= DoCanonicalizePathComponent<CHAR, UCHAR>( | 68 success &= DoCanonicalizePathComponent<CHAR, UCHAR>( |
69 source.query, parsed.query, '?', output, &new_parsed->query); | 69 source.query, parsed.query, '?', output, &new_parsed->query); |
70 success &= DoCanonicalizePathComponent<CHAR, UCHAR>( | 70 success &= DoCanonicalizePathComponent<CHAR, UCHAR>( |
71 source.ref, parsed.ref, '#', output, &new_parsed->ref); | 71 source.ref, parsed.ref, '#', output, &new_parsed->ref); |
72 | 72 |
73 return success; | 73 return success; |
74 } | 74 } |
75 | 75 |
76 } // namespace | 76 } // namespace |
77 | 77 |
(...skipping 34 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
112 Parsed* new_parsed) { | 112 Parsed* new_parsed) { |
113 RawCanonOutput<1024> utf8; | 113 RawCanonOutput<1024> utf8; |
114 URLComponentSource<char> source(base); | 114 URLComponentSource<char> source(base); |
115 Parsed parsed(base_parsed); | 115 Parsed parsed(base_parsed); |
116 SetupUTF16OverrideComponents(base, replacements, &utf8, &source, &parsed); | 116 SetupUTF16OverrideComponents(base, replacements, &utf8, &source, &parsed); |
117 return DoCanonicalizePathURL<char, unsigned char>( | 117 return DoCanonicalizePathURL<char, unsigned char>( |
118 source, parsed, output, new_parsed); | 118 source, parsed, output, new_parsed); |
119 } | 119 } |
120 | 120 |
121 } // namespace url | 121 } // namespace url |
OLD | NEW |