OLD | NEW |
1 This patch removes the usage of tolower() in fts code, which is not locale | 1 This patch removes the usage of tolower() in fts code, which is not locale |
2 neutral and causes problem in some locales such as Turkish. | 2 neutral and causes problem in some locales such as Turkish. |
3 See http://crbug.com/15261 for details. | 3 See http://crbug.com/15261 for details. |
4 An upstream ticket was also created for this issue: | 4 An upstream ticket was also created for this issue: |
5 http://www.sqlite.org/src/tktview/991789d9f3136a0460dc83a33e815c1aa9757c26 | 5 http://www.sqlite.org/src/tktview/991789d9f3136a0460dc83a33e815c1aa9757c26 |
6 | 6 |
7 Contains backport for upstream http://www.sqlite.org/src/ci/b8b465ed2c. | 7 Contains backport for upstream http://www.sqlite.org/src/ci/b8b465ed2c. |
8 | 8 |
9 Index: ext/fts3/fts3.c | 9 Index: ext/fts3/fts3.c |
10 =================================================================== | 10 =================================================================== |
(...skipping 74 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
85 | 85 |
86 @@ -191,7 +193,7 @@ static int simpleNext( | 86 @@ -191,7 +193,7 @@ static int simpleNext( |
87 ** case-insensitivity. | 87 ** case-insensitivity. |
88 */ | 88 */ |
89 unsigned char ch = p[iStartOffset+i]; | 89 unsigned char ch = p[iStartOffset+i]; |
90 - c->pToken[i] = ch<0x80 ? tolower(ch) : ch; | 90 - c->pToken[i] = ch<0x80 ? tolower(ch) : ch; |
91 + c->pToken[i] = (ch>='A' && ch<='Z') ? ch-'A'+'a' : ch; | 91 + c->pToken[i] = (ch>='A' && ch<='Z') ? ch-'A'+'a' : ch; |
92 } | 92 } |
93 *ppToken = c->pToken; | 93 *ppToken = c->pToken; |
94 *pnBytes = n; | 94 *pnBytes = n; |
95 Index: ext/fts1/simple_tokenizer.c | |
96 =================================================================== | |
97 --- ext/fts1/simple_tokenizer.c 2009-09-03 13:32:06.000000000 -0700 | |
98 +++ ext/fts1/simple_tokenizer.c 2009-09-02 11:40:21.000000000 -0700 | |
99 @@ -138,7 +138,7 @@ | |
100 ** case-insensitivity. | |
101 */ | |
102 char ch = c->pCurrent[ii]; | |
103 - c->zToken[ii] = (unsigned char)ch<0x80 ? tolower(ch) : ch; | |
104 + c->zToken[ii] = ((ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch); | |
105 } | |
106 c->zToken[n] = '\0'; | |
107 *ppToken = c->zToken; | |
108 Index: ext/fts1/fts1_tokenizer1.c | |
109 =================================================================== | |
110 --- ext/fts1/fts1_tokenizer1.c 2009-09-03 13:32:06.000000000 -0700 | |
111 +++ ext/fts1/fts1_tokenizer1.c 2009-09-02 11:40:21.000000000 -0700 | |
112 @@ -182,7 +182,7 @@ | |
113 ** case-insensitivity. | |
114 */ | |
115 unsigned char ch = p[iStartOffset+i]; | |
116 - c->pToken[i] = ch<0x80 ? tolower(ch) : ch; | |
117 + c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch; | |
118 } | |
119 *ppToken = c->pToken; | |
120 *pnBytes = n; | |
121 Index: ext/fts1/fts1.c | |
122 =================================================================== | |
123 --- ext/fts1/fts1.c 2009-09-04 13:37:41.000000000 -0700 | |
124 +++ ext/fts1/fts1.c 2009-09-14 18:16:55.000000000 -0700 | |
125 @@ -208,7 +208,7 @@ | |
126 return (c&0x80)==0 ? isspace(c) : 0; | |
127 } | |
128 static int safe_tolower(char c){ | |
129 - return (c&0x80)==0 ? tolower(c) : c; | |
130 + return (c>='A' && c<='Z') ? (c-'A'+'a') : c; | |
131 } | |
132 static int safe_isalnum(char c){ | |
133 return (c&0x80)==0 ? isalnum(c) : 0; | |
134 Index: ext/fts2/fts2.c | 95 Index: ext/fts2/fts2.c |
135 =================================================================== | 96 =================================================================== |
136 --- ext/fts2/fts2.c 2009-09-04 13:37:41.000000000 -0700 | 97 --- ext/fts2/fts2.c 2009-09-04 13:37:41.000000000 -0700 |
137 +++ ext/fts2/fts2.c 2009-09-14 18:17:02.000000000 -0700 | 98 +++ ext/fts2/fts2.c 2009-09-14 18:17:02.000000000 -0700 |
138 @@ -372,7 +372,7 @@ | 99 @@ -372,7 +372,7 @@ |
139 return (c&0x80)==0 ? isspace(c) : 0; | 100 return (c&0x80)==0 ? isspace(c) : 0; |
140 } | 101 } |
141 static int safe_tolower(char c){ | 102 static int safe_tolower(char c){ |
142 - return (c&0x80)==0 ? tolower(c) : c; | 103 - return (c&0x80)==0 ? tolower(c) : c; |
143 + return (c>='A' && c<='Z') ? (c-'A'+'a') : c; | 104 + return (c>='A' && c<='Z') ? (c-'A'+'a') : c; |
144 } | 105 } |
145 static int safe_isalnum(char c){ | 106 static int safe_isalnum(char c){ |
146 return (c&0x80)==0 ? isalnum(c) : 0; | 107 return (c&0x80)==0 ? isalnum(c) : 0; |
147 Index: ext/fts2/fts2_tokenizer1.c | 108 Index: ext/fts2/fts2_tokenizer1.c |
148 =================================================================== | 109 =================================================================== |
149 --- ext/fts2/fts2_tokenizer1.c 2009-09-03 13:32:06.000000000 -0700 | 110 --- ext/fts2/fts2_tokenizer1.c 2009-09-03 13:32:06.000000000 -0700 |
150 +++ ext/fts2/fts2_tokenizer1.c 2009-09-02 11:40:21.000000000 -0700 | 111 +++ ext/fts2/fts2_tokenizer1.c 2009-09-02 11:40:21.000000000 -0700 |
151 @@ -191,7 +191,7 @@ | 112 @@ -191,7 +191,7 @@ |
152 ** case-insensitivity. | 113 ** case-insensitivity. |
153 */ | 114 */ |
154 unsigned char ch = p[iStartOffset+i]; | 115 unsigned char ch = p[iStartOffset+i]; |
155 - c->pToken[i] = ch<0x80 ? tolower(ch) : ch; | 116 - c->pToken[i] = ch<0x80 ? tolower(ch) : ch; |
156 + c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch; | 117 + c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch; |
157 } | 118 } |
158 *ppToken = c->pToken; | 119 *ppToken = c->pToken; |
159 *pnBytes = n; | 120 *pnBytes = n; |
OLD | NEW |