philip.liard@gmail.com | b905691 | 2011-08-18 11:41:24 +0000 | [diff] [blame] | 1 | // Copyright (C) 2011 The Libphonenumber Authors |
philip.liard@gmail.com | 1ad5e5b | 2011-07-01 08:22:06 +0000 | [diff] [blame] | 2 | // |
| 3 | // Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 | // you may not use this file except in compliance with the License. |
| 5 | // You may obtain a copy of the License at |
| 6 | // |
| 7 | // http://www.apache.org/licenses/LICENSE-2.0 |
| 8 | // |
| 9 | // Unless required by applicable law or agreed to in writing, software |
| 10 | // distributed under the License is distributed on an "AS IS" BASIS, |
| 11 | // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 12 | // See the License for the specific language governing permissions and |
| 13 | // limitations under the License. |
| 14 | |
| 15 | // Author: George Yakovlev |
| 16 | // Philippe Liard |
| 17 | |
philip.liard@gmail.com | 384682a | 2011-07-12 15:41:29 +0000 | [diff] [blame] | 18 | #include "phonenumbers/regexp_adapter_re2.h" |
philip.liard@gmail.com | 1ad5e5b | 2011-07-01 08:22:06 +0000 | [diff] [blame] | 19 | |
| 20 | #include <cstddef> |
| 21 | #include <string> |
| 22 | |
| 23 | #include <re2/re2.h> |
| 24 | #include <re2/stringpiece.h> |
| 25 | |
philip.liard@gmail.com | af4a2ce | 2013-04-30 11:35:55 +0000 | [diff] [blame] | 26 | #include "phonenumbers/base/basictypes.h" |
| 27 | #include "phonenumbers/base/logging.h" |
philip.liard@gmail.com | 1ad5e5b | 2011-07-01 08:22:06 +0000 | [diff] [blame] | 28 | #include "phonenumbers/stringutil.h" |
| 29 | |
| 30 | namespace i18n { |
| 31 | namespace phonenumbers { |
| 32 | |
| 33 | using re2::StringPiece; |
| 34 | |
| 35 | // Implementation of RegExpInput abstract class. |
| 36 | class RE2RegExpInput : public RegExpInput { |
| 37 | public: |
| 38 | explicit RE2RegExpInput(const string& utf8_input) |
| 39 | : string_(utf8_input), |
| 40 | utf8_input_(string_) {} |
| 41 | |
| 42 | virtual string ToString() const { |
| 43 | return utf8_input_.ToString(); |
| 44 | } |
| 45 | |
| 46 | StringPiece* Data() { |
| 47 | return &utf8_input_; |
| 48 | } |
| 49 | |
| 50 | private: |
| 51 | // string_ holds the string referenced by utf8_input_ as StringPiece doesn't |
| 52 | // copy the string passed in. |
| 53 | const string string_; |
| 54 | StringPiece utf8_input_; |
| 55 | }; |
| 56 | |
| 57 | namespace { |
| 58 | |
| 59 | template <typename Function, typename Input> |
| 60 | bool DispatchRE2Call(Function regex_function, |
| 61 | Input input, |
| 62 | const RE2& regexp, |
| 63 | string* out1, |
| 64 | string* out2, |
| 65 | string* out3) { |
| 66 | if (out3) { |
| 67 | return regex_function(input, regexp, out1, out2, out3); |
| 68 | } |
| 69 | if (out2) { |
| 70 | return regex_function(input, regexp, out1, out2); |
| 71 | } |
| 72 | if (out1) { |
| 73 | return regex_function(input, regexp, out1); |
| 74 | } |
| 75 | return regex_function(input, regexp); |
| 76 | } |
| 77 | |
| 78 | // Replaces unescaped dollar-signs with backslashes. Backslashes are deleted |
| 79 | // when they escape dollar-signs. |
| 80 | string TransformRegularExpressionToRE2Syntax(const string& regex) { |
| 81 | string re2_regex(regex); |
| 82 | if (GlobalReplaceSubstring("$", "\\", &re2_regex) == 0) { |
| 83 | return regex; |
| 84 | } |
| 85 | // If we replaced a dollar sign with a backslash and there are now two |
| 86 | // backslashes in the string, we assume that the dollar-sign was previously |
| 87 | // escaped and that we need to retain it. To do this, we replace pairs of |
| 88 | // backslashes with a dollar sign. |
| 89 | GlobalReplaceSubstring("\\\\", "$", &re2_regex); |
| 90 | return re2_regex; |
| 91 | } |
| 92 | |
| 93 | } // namespace |
| 94 | |
| 95 | // Implementation of RegExp abstract class. |
| 96 | class RE2RegExp : public RegExp { |
| 97 | public: |
| 98 | explicit RE2RegExp(const string& utf8_regexp) |
| 99 | : utf8_regexp_(utf8_regexp) {} |
| 100 | |
| 101 | virtual bool Consume(RegExpInput* input_string, |
| 102 | bool anchor_at_start, |
| 103 | string* matched_string1, |
| 104 | string* matched_string2, |
| 105 | string* matched_string3) const { |
| 106 | DCHECK(input_string); |
| 107 | StringPiece* utf8_input = |
| 108 | static_cast<RE2RegExpInput*>(input_string)->Data(); |
| 109 | |
| 110 | if (anchor_at_start) { |
| 111 | return DispatchRE2Call(RE2::Consume, utf8_input, utf8_regexp_, |
| 112 | matched_string1, matched_string2, |
| 113 | matched_string3); |
| 114 | } else { |
| 115 | return DispatchRE2Call(RE2::FindAndConsume, utf8_input, utf8_regexp_, |
| 116 | matched_string1, matched_string2, |
| 117 | matched_string3); |
| 118 | } |
| 119 | } |
| 120 | |
| 121 | virtual bool Match(const string& input_string, |
| 122 | bool full_match, |
| 123 | string* matched_string) const { |
| 124 | if (full_match) { |
| 125 | return DispatchRE2Call(RE2::FullMatch, input_string, utf8_regexp_, |
| 126 | matched_string, NULL, NULL); |
| 127 | } else { |
| 128 | return DispatchRE2Call(RE2::PartialMatch, input_string, utf8_regexp_, |
| 129 | matched_string, NULL, NULL); |
| 130 | } |
| 131 | } |
| 132 | |
| 133 | virtual bool Replace(string* string_to_process, |
| 134 | bool global, |
| 135 | const string& replacement_string) const { |
| 136 | DCHECK(string_to_process); |
| 137 | const string re2_replacement_string = |
| 138 | TransformRegularExpressionToRE2Syntax(replacement_string); |
| 139 | if (global) { |
| 140 | return RE2::GlobalReplace(string_to_process, utf8_regexp_, |
| 141 | re2_replacement_string); |
| 142 | } else { |
| 143 | return RE2::Replace(string_to_process, utf8_regexp_, |
| 144 | re2_replacement_string); |
| 145 | } |
| 146 | } |
| 147 | |
| 148 | private: |
| 149 | RE2 utf8_regexp_; |
| 150 | }; |
| 151 | |
philip.liard@gmail.com | 384682a | 2011-07-12 15:41:29 +0000 | [diff] [blame] | 152 | RegExpInput* RE2RegExpFactory::CreateInput(const string& utf8_input) const { |
philip.liard@gmail.com | 1ad5e5b | 2011-07-01 08:22:06 +0000 | [diff] [blame] | 153 | return new RE2RegExpInput(utf8_input); |
| 154 | } |
| 155 | |
philip.liard@gmail.com | 384682a | 2011-07-12 15:41:29 +0000 | [diff] [blame] | 156 | RegExp* RE2RegExpFactory::CreateRegExp(const string& utf8_regexp) const { |
philip.liard@gmail.com | 1ad5e5b | 2011-07-01 08:22:06 +0000 | [diff] [blame] | 157 | return new RE2RegExp(utf8_regexp); |
| 158 | } |
| 159 | |
| 160 | } // namespace phonenumbers |
| 161 | } // namespace i18n |