Fossil SCM

Merge trunk. Restore fossil_utf8_to_filename signature by doing character conversions in utf-8 instead of unicode.

jan.nijtmans 2013-02-25 10:46 ticket-d17d6e5b17 merge
Commit ae1b1f4b5d97289ab1d6ca811350618ea688659d
Parent f1bb151e6f7a9e5…
2 files changed +34 -26 +34 -26
~ src/utf8.c ~ src/utf8.c
M src/utf8.c
+34 -26
--- src/utf8.c
+++ src/utf8.c
@@ -104,32 +104,37 @@
104104
 /*
105105
 ** Translate text from the filename character set into UTF-8.
106106
 ** Return a pointer to the translated text.
107107
 ** Call fossil_filename_free() to deallocate any memory used to store the
108108
 ** returned pointer when done.
109
+**
110
+** On Windows, translate some characters in the in the range
111
+** U+F001 - U+F07F (private use area) to ASCII. Cygwin sometimes
112
+** generates such filenames. See:
113
+** <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
109114
 */
110
-char *fossil_filename_to_utf8(void *zFilename){
115
+char *fossil_filename_to_utf8(const void *zFilename){
111116
 #if defined(_WIN32)
112
- int nByte;
113
- char *zUtf;
114
- wchar_t *wUnicode = zFilename;
115
- while( *wUnicode != 0 ){
116
- if ( (*wUnicode & 0xff80) == 0xf000 ){
117
- wchar_t converted = (*wUnicode & 0x7f);
118
- /* Only really convert it when the resulting char is in the given range*/
119
- if ( (converted < 32) || wcschr(L"\"*<>?|:", converted) ){
120
- *wUnicode = converted;
121
- }
122
- }
123
- ++wUnicode;
124
- }
125
- nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
126
- zUtf = sqlite3_malloc( nByte );
117
+ int nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
118
+ char *zUtf = sqlite3_malloc( nByte );
119
+ char *pUtf, *qUtf;
127120
 if( zUtf==0 ){
128121
 return 0;
129122
 }
130123
 WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, zUtf, nByte, 0, 0);
124
+ pUtf = qUtf = zUtf;
125
+ while( *pUtf ) {
126
+ if( *pUtf == (char)0xef ){
127
+ wchar_t c = ((pUtf[1]&0x3f)<<6)|(pUtf[2]&0x3f);
128
+ /* Only really convert it when the resulting char is in range. */
129
+ if ( c && ((c <= ' ') || wcschr(L"\"*.:<>?|", c)) ){
130
+ *qUtf++ = c; pUtf+=3; continue;
131
+ }
132
+ }
133
+ *qUtf++ = *pUtf++;
134
+ }
135
+ *qUtf = 0;
131136
 return zUtf;
132137
 #elif defined(__CYGWIN__)
133138
 char *zOut;
134139
 zOut = fossil_strdup(zFilename);
135140
 return zOut;
@@ -168,12 +173,12 @@
168173
 ** Return a pointer to the translated text.. Call fossil_filename_free()
169174
 ** to deallocate any memory used to store the returned pointer when done.
170175
 **
171176
 ** On Windows, characters in the range U+0001 to U+0031 and the
172177
 ** characters '"', '*', ':', '<', '>', '?' and '|' are invalid
173
-** to be used. Therefore, translated those to characters in the
174
-** (private use area), in the range U+F001 - U+F07F, so those
178
+** to be used. Therefore, translate those to characters in the
179
+** in the range U+F001 - U+F07F (private use area), so those
175180
 ** characters never arrive in any Windows API. The filenames might
176181
 ** look strange in Windows explorer, but in the cygwin shell
177182
 ** everything looks as expected.
178183
 **
179184
 ** See: <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
@@ -193,11 +198,11 @@
193198
 && (zUtf8[2]=='\\' || zUtf8[2]=='/')) {
194199
 zUnicode[2] = '\\';
195200
 wUnicode += 3;
196201
 }
197202
 while( *wUnicode != '\0' ){
198
- if ( (*wUnicode < 32) || wcschr(L"\"*<>?|:", *wUnicode) ){
203
+ if ( (*wUnicode < 32) || wcschr(L"\"*:<>?|", *wUnicode) ){
199204
 *wUnicode |= 0xF000;
200205
 }else if( *wUnicode == '/' ){
201206
 *wUnicode = '\\';
202207
 }
203208
 ++wUnicode;
@@ -239,11 +244,11 @@
239244
 ** to a file, -1 is returned and nothing is written
240245
 ** to the console.
241246
 */
242247
 int fossil_utf8_to_console(const char *zUtf8, int nByte, int toStdErr){
243248
 #ifdef _WIN32
244
- int nChar;
249
+ int nChar, written = 0;
245250
 wchar_t *zUnicode; /* Unicode version of zUtf8 */
246251
 DWORD dummy;
247252
248253
 static int istty[2] = { -1, -1 };
249254
 if( istty[toStdErr] == -1 ){
@@ -258,17 +263,20 @@
258263
 zUnicode = malloc( (nChar + 1) *sizeof(zUnicode[0]) );
259264
 if( zUnicode==0 ){
260265
 return 0;
261266
 }
262267
 nChar = MultiByteToWideChar(CP_UTF8, 0, zUtf8, nByte, zUnicode, nChar);
263
- if( nChar==0 ){
264
- free(zUnicode);
265
- return 0;
268
+ /* Split WriteConsoleW call into multiple chunks, if necessary. See:
269
+ * <https://connect.microsoft.com/VisualStudio/feedback/details/635230> */
270
+ while( written < nChar ){
271
+ int size = nChar-written;
272
+ if (size > 26000) size = 26000;
273
+ WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode+written,
274
+ size, &dummy, 0);
275
+ written += size;
266276
 }
267
- zUnicode[nChar] = '\0';
268
- WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode, nChar,
269
- &dummy, 0);
277
+ free(zUnicode);
270278
 return nChar;
271279
 #else
272280
 return -1; /* No-op on unix */
273281
 #endif
274282
 }
275283
--- src/utf8.c
+++ src/utf8.c
@@ -104,32 +104,37 @@
104 /*
105 ** Translate text from the filename character set into UTF-8.
106 ** Return a pointer to the translated text.
107 ** Call fossil_filename_free() to deallocate any memory used to store the
108 ** returned pointer when done.
 
 
 
 
 
109 */
110 char *fossil_filename_to_utf8(void *zFilename){
111 #if defined(_WIN32)
112 int nByte;
113 char *zUtf;
114 wchar_t *wUnicode = zFilename;
115 while( *wUnicode != 0 ){
116 if ( (*wUnicode & 0xff80) == 0xf000 ){
117 wchar_t converted = (*wUnicode & 0x7f);
118 /* Only really convert it when the resulting char is in the given range*/
119 if ( (converted < 32) || wcschr(L"\"*<>?|:", converted) ){
120 *wUnicode = converted;
121 }
122 }
123 ++wUnicode;
124 }
125 nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
126 zUtf = sqlite3_malloc( nByte );
127 if( zUtf==0 ){
128 return 0;
129 }
130 WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, zUtf, nByte, 0, 0);
 
 
 
 
 
 
 
 
 
 
 
 
131 return zUtf;
132 #elif defined(__CYGWIN__)
133 char *zOut;
134 zOut = fossil_strdup(zFilename);
135 return zOut;
@@ -168,12 +173,12 @@
168 ** Return a pointer to the translated text.. Call fossil_filename_free()
169 ** to deallocate any memory used to store the returned pointer when done.
170 **
171 ** On Windows, characters in the range U+0001 to U+0031 and the
172 ** characters '"', '*', ':', '<', '>', '?' and '|' are invalid
173 ** to be used. Therefore, translated those to characters in the
174 ** (private use area), in the range U+F001 - U+F07F, so those
175 ** characters never arrive in any Windows API. The filenames might
176 ** look strange in Windows explorer, but in the cygwin shell
177 ** everything looks as expected.
178 **
179 ** See: <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
@@ -193,11 +198,11 @@
193 && (zUtf8[2]=='\\' || zUtf8[2]=='/')) {
194 zUnicode[2] = '\\';
195 wUnicode += 3;
196 }
197 while( *wUnicode != '\0' ){
198 if ( (*wUnicode < 32) || wcschr(L"\"*<>?|:", *wUnicode) ){
199 *wUnicode |= 0xF000;
200 }else if( *wUnicode == '/' ){
201 *wUnicode = '\\';
202 }
203 ++wUnicode;
@@ -239,11 +244,11 @@
239 ** to a file, -1 is returned and nothing is written
240 ** to the console.
241 */
242 int fossil_utf8_to_console(const char *zUtf8, int nByte, int toStdErr){
243 #ifdef _WIN32
244 int nChar;
245 wchar_t *zUnicode; /* Unicode version of zUtf8 */
246 DWORD dummy;
247
248 static int istty[2] = { -1, -1 };
249 if( istty[toStdErr] == -1 ){
@@ -258,17 +263,20 @@
258 zUnicode = malloc( (nChar + 1) *sizeof(zUnicode[0]) );
259 if( zUnicode==0 ){
260 return 0;
261 }
262 nChar = MultiByteToWideChar(CP_UTF8, 0, zUtf8, nByte, zUnicode, nChar);
263 if( nChar==0 ){
264 free(zUnicode);
265 return 0;
 
 
 
 
 
266 }
267 zUnicode[nChar] = '\0';
268 WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode, nChar,
269 &dummy, 0);
270 return nChar;
271 #else
272 return -1; /* No-op on unix */
273 #endif
274 }
275
--- src/utf8.c
+++ src/utf8.c
@@ -104,32 +104,37 @@
104 /*
105 ** Translate text from the filename character set into UTF-8.
106 ** Return a pointer to the translated text.
107 ** Call fossil_filename_free() to deallocate any memory used to store the
108 ** returned pointer when done.
109 **
110 ** On Windows, translate some characters in the in the range
111 ** U+F001 - U+F07F (private use area) to ASCII. Cygwin sometimes
112 ** generates such filenames. See:
113 ** <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
114 */
115 char *fossil_filename_to_utf8(const void *zFilename){
116 #if defined(_WIN32)
117 int nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
118 char *zUtf = sqlite3_malloc( nByte );
119 char *pUtf, *qUtf;
 
 
 
 
 
 
 
 
 
 
 
 
120 if( zUtf==0 ){
121 return 0;
122 }
123 WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, zUtf, nByte, 0, 0);
124 pUtf = qUtf = zUtf;
125 while( *pUtf ) {
126 if( *pUtf == (char)0xef ){
127 wchar_t c = ((pUtf[1]&0x3f)<<6)|(pUtf[2]&0x3f);
128 /* Only really convert it when the resulting char is in range. */
129 if ( c && ((c <= ' ') || wcschr(L"\"*.:<>?|", c)) ){
130 *qUtf++ = c; pUtf+=3; continue;
131 }
132 }
133 *qUtf++ = *pUtf++;
134 }
135 *qUtf = 0;
136 return zUtf;
137 #elif defined(__CYGWIN__)
138 char *zOut;
139 zOut = fossil_strdup(zFilename);
140 return zOut;
@@ -168,12 +173,12 @@
173 ** Return a pointer to the translated text.. Call fossil_filename_free()
174 ** to deallocate any memory used to store the returned pointer when done.
175 **
176 ** On Windows, characters in the range U+0001 to U+0031 and the
177 ** characters '"', '*', ':', '<', '>', '?' and '|' are invalid
178 ** to be used. Therefore, translate those to characters in the
179 ** in the range U+F001 - U+F07F (private use area), so those
180 ** characters never arrive in any Windows API. The filenames might
181 ** look strange in Windows explorer, but in the cygwin shell
182 ** everything looks as expected.
183 **
184 ** See: <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
@@ -193,11 +198,11 @@
198 && (zUtf8[2]=='\\' || zUtf8[2]=='/')) {
199 zUnicode[2] = '\\';
200 wUnicode += 3;
201 }
202 while( *wUnicode != '\0' ){
203 if ( (*wUnicode < 32) || wcschr(L"\"*:<>?|", *wUnicode) ){
204 *wUnicode |= 0xF000;
205 }else if( *wUnicode == '/' ){
206 *wUnicode = '\\';
207 }
208 ++wUnicode;
@@ -239,11 +244,11 @@
244 ** to a file, -1 is returned and nothing is written
245 ** to the console.
246 */
247 int fossil_utf8_to_console(const char *zUtf8, int nByte, int toStdErr){
248 #ifdef _WIN32
249 int nChar, written = 0;
250 wchar_t *zUnicode; /* Unicode version of zUtf8 */
251 DWORD dummy;
252
253 static int istty[2] = { -1, -1 };
254 if( istty[toStdErr] == -1 ){
@@ -258,17 +263,20 @@
263 zUnicode = malloc( (nChar + 1) *sizeof(zUnicode[0]) );
264 if( zUnicode==0 ){
265 return 0;
266 }
267 nChar = MultiByteToWideChar(CP_UTF8, 0, zUtf8, nByte, zUnicode, nChar);
268 /* Split WriteConsoleW call into multiple chunks, if necessary. See:
269 * <https://connect.microsoft.com/VisualStudio/feedback/details/635230> */
270 while( written < nChar ){
271 int size = nChar-written;
272 if (size > 26000) size = 26000;
273 WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode+written,
274 size, &dummy, 0);
275 written += size;
276 }
277 free(zUnicode);
 
 
278 return nChar;
279 #else
280 return -1; /* No-op on unix */
281 #endif
282 }
283
M src/utf8.c
+34 -26
--- src/utf8.c
+++ src/utf8.c
@@ -104,32 +104,37 @@
104104
 /*
105105
 ** Translate text from the filename character set into UTF-8.
106106
 ** Return a pointer to the translated text.
107107
 ** Call fossil_filename_free() to deallocate any memory used to store the
108108
 ** returned pointer when done.
109
+**
110
+** On Windows, translate some characters in the in the range
111
+** U+F001 - U+F07F (private use area) to ASCII. Cygwin sometimes
112
+** generates such filenames. See:
113
+** <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
109114
 */
110
-char *fossil_filename_to_utf8(void *zFilename){
115
+char *fossil_filename_to_utf8(const void *zFilename){
111116
 #if defined(_WIN32)
112
- int nByte;
113
- char *zUtf;
114
- wchar_t *wUnicode = zFilename;
115
- while( *wUnicode != 0 ){
116
- if ( (*wUnicode & 0xff80) == 0xf000 ){
117
- wchar_t converted = (*wUnicode & 0x7f);
118
- /* Only really convert it when the resulting char is in the given range*/
119
- if ( (converted < 32) || wcschr(L"\"*<>?|:", converted) ){
120
- *wUnicode = converted;
121
- }
122
- }
123
- ++wUnicode;
124
- }
125
- nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
126
- zUtf = sqlite3_malloc( nByte );
117
+ int nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
118
+ char *zUtf = sqlite3_malloc( nByte );
119
+ char *pUtf, *qUtf;
127120
 if( zUtf==0 ){
128121
 return 0;
129122
 }
130123
 WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, zUtf, nByte, 0, 0);
124
+ pUtf = qUtf = zUtf;
125
+ while( *pUtf ) {
126
+ if( *pUtf == (char)0xef ){
127
+ wchar_t c = ((pUtf[1]&0x3f)<<6)|(pUtf[2]&0x3f);
128
+ /* Only really convert it when the resulting char is in range. */
129
+ if ( c && ((c <= ' ') || wcschr(L"\"*.:<>?|", c)) ){
130
+ *qUtf++ = c; pUtf+=3; continue;
131
+ }
132
+ }
133
+ *qUtf++ = *pUtf++;
134
+ }
135
+ *qUtf = 0;
131136
 return zUtf;
132137
 #elif defined(__CYGWIN__)
133138
 char *zOut;
134139
 zOut = fossil_strdup(zFilename);
135140
 return zOut;
@@ -168,12 +173,12 @@
168173
 ** Return a pointer to the translated text.. Call fossil_filename_free()
169174
 ** to deallocate any memory used to store the returned pointer when done.
170175
 **
171176
 ** On Windows, characters in the range U+0001 to U+0031 and the
172177
 ** characters '"', '*', ':', '<', '>', '?' and '|' are invalid
173
-** to be used. Therefore, translated those to characters in the
174
-** (private use area), in the range U+F001 - U+F07F, so those
178
+** to be used. Therefore, translate those to characters in the
179
+** in the range U+F001 - U+F07F (private use area), so those
175180
 ** characters never arrive in any Windows API. The filenames might
176181
 ** look strange in Windows explorer, but in the cygwin shell
177182
 ** everything looks as expected.
178183
 **
179184
 ** See: <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
@@ -193,11 +198,11 @@
193198
 && (zUtf8[2]=='\\' || zUtf8[2]=='/')) {
194199
 zUnicode[2] = '\\';
195200
 wUnicode += 3;
196201
 }
197202
 while( *wUnicode != '\0' ){
198
- if ( (*wUnicode < 32) || wcschr(L"\"*<>?|:", *wUnicode) ){
203
+ if ( (*wUnicode < 32) || wcschr(L"\"*:<>?|", *wUnicode) ){
199204
 *wUnicode |= 0xF000;
200205
 }else if( *wUnicode == '/' ){
201206
 *wUnicode = '\\';
202207
 }
203208
 ++wUnicode;
@@ -239,11 +244,11 @@
239244
 ** to a file, -1 is returned and nothing is written
240245
 ** to the console.
241246
 */
242247
 int fossil_utf8_to_console(const char *zUtf8, int nByte, int toStdErr){
243248
 #ifdef _WIN32
244
- int nChar;
249
+ int nChar, written = 0;
245250
 wchar_t *zUnicode; /* Unicode version of zUtf8 */
246251
 DWORD dummy;
247252
248253
 static int istty[2] = { -1, -1 };
249254
 if( istty[toStdErr] == -1 ){
@@ -258,17 +263,20 @@
258263
 zUnicode = malloc( (nChar + 1) *sizeof(zUnicode[0]) );
259264
 if( zUnicode==0 ){
260265
 return 0;
261266
 }
262267
 nChar = MultiByteToWideChar(CP_UTF8, 0, zUtf8, nByte, zUnicode, nChar);
263
- if( nChar==0 ){
264
- free(zUnicode);
265
- return 0;
268
+ /* Split WriteConsoleW call into multiple chunks, if necessary. See:
269
+ * <https://connect.microsoft.com/VisualStudio/feedback/details/635230> */
270
+ while( written < nChar ){
271
+ int size = nChar-written;
272
+ if (size > 26000) size = 26000;
273
+ WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode+written,
274
+ size, &dummy, 0);
275
+ written += size;
266276
 }
267
- zUnicode[nChar] = '\0';
268
- WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode, nChar,
269
- &dummy, 0);
277
+ free(zUnicode);
270278
 return nChar;
271279
 #else
272280
 return -1; /* No-op on unix */
273281
 #endif
274282
 }
275283
--- src/utf8.c
+++ src/utf8.c
@@ -104,32 +104,37 @@
104 /*
105 ** Translate text from the filename character set into UTF-8.
106 ** Return a pointer to the translated text.
107 ** Call fossil_filename_free() to deallocate any memory used to store the
108 ** returned pointer when done.
 
 
 
 
 
109 */
110 char *fossil_filename_to_utf8(void *zFilename){
111 #if defined(_WIN32)
112 int nByte;
113 char *zUtf;
114 wchar_t *wUnicode = zFilename;
115 while( *wUnicode != 0 ){
116 if ( (*wUnicode & 0xff80) == 0xf000 ){
117 wchar_t converted = (*wUnicode & 0x7f);
118 /* Only really convert it when the resulting char is in the given range*/
119 if ( (converted < 32) || wcschr(L"\"*<>?|:", converted) ){
120 *wUnicode = converted;
121 }
122 }
123 ++wUnicode;
124 }
125 nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
126 zUtf = sqlite3_malloc( nByte );
127 if( zUtf==0 ){
128 return 0;
129 }
130 WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, zUtf, nByte, 0, 0);
 
 
 
 
 
 
 
 
 
 
 
 
131 return zUtf;
132 #elif defined(__CYGWIN__)
133 char *zOut;
134 zOut = fossil_strdup(zFilename);
135 return zOut;
@@ -168,12 +173,12 @@
168 ** Return a pointer to the translated text.. Call fossil_filename_free()
169 ** to deallocate any memory used to store the returned pointer when done.
170 **
171 ** On Windows, characters in the range U+0001 to U+0031 and the
172 ** characters '"', '*', ':', '<', '>', '?' and '|' are invalid
173 ** to be used. Therefore, translated those to characters in the
174 ** (private use area), in the range U+F001 - U+F07F, so those
175 ** characters never arrive in any Windows API. The filenames might
176 ** look strange in Windows explorer, but in the cygwin shell
177 ** everything looks as expected.
178 **
179 ** See: <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
@@ -193,11 +198,11 @@
193 && (zUtf8[2]=='\\' || zUtf8[2]=='/')) {
194 zUnicode[2] = '\\';
195 wUnicode += 3;
196 }
197 while( *wUnicode != '\0' ){
198 if ( (*wUnicode < 32) || wcschr(L"\"*<>?|:", *wUnicode) ){
199 *wUnicode |= 0xF000;
200 }else if( *wUnicode == '/' ){
201 *wUnicode = '\\';
202 }
203 ++wUnicode;
@@ -239,11 +244,11 @@
239 ** to a file, -1 is returned and nothing is written
240 ** to the console.
241 */
242 int fossil_utf8_to_console(const char *zUtf8, int nByte, int toStdErr){
243 #ifdef _WIN32
244 int nChar;
245 wchar_t *zUnicode; /* Unicode version of zUtf8 */
246 DWORD dummy;
247
248 static int istty[2] = { -1, -1 };
249 if( istty[toStdErr] == -1 ){
@@ -258,17 +263,20 @@
258 zUnicode = malloc( (nChar + 1) *sizeof(zUnicode[0]) );
259 if( zUnicode==0 ){
260 return 0;
261 }
262 nChar = MultiByteToWideChar(CP_UTF8, 0, zUtf8, nByte, zUnicode, nChar);
263 if( nChar==0 ){
264 free(zUnicode);
265 return 0;
 
 
 
 
 
266 }
267 zUnicode[nChar] = '\0';
268 WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode, nChar,
269 &dummy, 0);
270 return nChar;
271 #else
272 return -1; /* No-op on unix */
273 #endif
274 }
275
--- src/utf8.c
+++ src/utf8.c
@@ -104,32 +104,37 @@
104 /*
105 ** Translate text from the filename character set into UTF-8.
106 ** Return a pointer to the translated text.
107 ** Call fossil_filename_free() to deallocate any memory used to store the
108 ** returned pointer when done.
109 **
110 ** On Windows, translate some characters in the in the range
111 ** U+F001 - U+F07F (private use area) to ASCII. Cygwin sometimes
112 ** generates such filenames. See:
113 ** <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
114 */
115 char *fossil_filename_to_utf8(const void *zFilename){
116 #if defined(_WIN32)
117 int nByte = WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, 0, 0, 0, 0);
118 char *zUtf = sqlite3_malloc( nByte );
119 char *pUtf, *qUtf;
 
 
 
 
 
 
 
 
 
 
 
 
120 if( zUtf==0 ){
121 return 0;
122 }
123 WideCharToMultiByte(CP_UTF8, 0, zFilename, -1, zUtf, nByte, 0, 0);
124 pUtf = qUtf = zUtf;
125 while( *pUtf ) {
126 if( *pUtf == (char)0xef ){
127 wchar_t c = ((pUtf[1]&0x3f)<<6)|(pUtf[2]&0x3f);
128 /* Only really convert it when the resulting char is in range. */
129 if ( c && ((c <= ' ') || wcschr(L"\"*.:<>?|", c)) ){
130 *qUtf++ = c; pUtf+=3; continue;
131 }
132 }
133 *qUtf++ = *pUtf++;
134 }
135 *qUtf = 0;
136 return zUtf;
137 #elif defined(__CYGWIN__)
138 char *zOut;
139 zOut = fossil_strdup(zFilename);
140 return zOut;
@@ -168,12 +173,12 @@
173 ** Return a pointer to the translated text.. Call fossil_filename_free()
174 ** to deallocate any memory used to store the returned pointer when done.
175 **
176 ** On Windows, characters in the range U+0001 to U+0031 and the
177 ** characters '"', '*', ':', '<', '>', '?' and '|' are invalid
178 ** to be used. Therefore, translate those to characters in the
179 ** in the range U+F001 - U+F07F (private use area), so those
180 ** characters never arrive in any Windows API. The filenames might
181 ** look strange in Windows explorer, but in the cygwin shell
182 ** everything looks as expected.
183 **
184 ** See: <http://cygwin.com/cygwin-ug-net/using-specialnames.html>
@@ -193,11 +198,11 @@
198 && (zUtf8[2]=='\\' || zUtf8[2]=='/')) {
199 zUnicode[2] = '\\';
200 wUnicode += 3;
201 }
202 while( *wUnicode != '\0' ){
203 if ( (*wUnicode < 32) || wcschr(L"\"*:<>?|", *wUnicode) ){
204 *wUnicode |= 0xF000;
205 }else if( *wUnicode == '/' ){
206 *wUnicode = '\\';
207 }
208 ++wUnicode;
@@ -239,11 +244,11 @@
244 ** to a file, -1 is returned and nothing is written
245 ** to the console.
246 */
247 int fossil_utf8_to_console(const char *zUtf8, int nByte, int toStdErr){
248 #ifdef _WIN32
249 int nChar, written = 0;
250 wchar_t *zUnicode; /* Unicode version of zUtf8 */
251 DWORD dummy;
252
253 static int istty[2] = { -1, -1 };
254 if( istty[toStdErr] == -1 ){
@@ -258,17 +263,20 @@
263 zUnicode = malloc( (nChar + 1) *sizeof(zUnicode[0]) );
264 if( zUnicode==0 ){
265 return 0;
266 }
267 nChar = MultiByteToWideChar(CP_UTF8, 0, zUtf8, nByte, zUnicode, nChar);
268 /* Split WriteConsoleW call into multiple chunks, if necessary. See:
269 * <https://connect.microsoft.com/VisualStudio/feedback/details/635230> */
270 while( written < nChar ){
271 int size = nChar-written;
272 if (size > 26000) size = 26000;
273 WriteConsoleW(GetStdHandle(STD_OUTPUT_HANDLE - toStdErr), zUnicode+written,
274 size, &dummy, 0);
275 written += size;
276 }
277 free(zUnicode);
 
 
278 return nChar;
279 #else
280 return -1; /* No-op on unix */
281 #endif
282 }
283

Keyboard Shortcuts

Open search /
Next entry (timeline) j
Previous entry (timeline) k
Open focused entry Enter
Show this help ?
Toggle theme Top nav button