author | Lars Hjemli <hjemli@gmail.com> | 2007-11-05 23:38:18 (UTC) |
---|---|---|
committer | Lars Hjemli <hjemli@gmail.com> | 2007-11-05 23:38:18 (UTC) |
commit | 55ac326ecb01161bf62865ae3350acf85db97d63 (patch) (unidiff) | |
tree | e37bcd8fef5523aa627ec9fdcd3401105f8e579d /parsing.c | |
parent | d04c4734bcf40b1d17c55b18fba2aa8344678e8f (diff) | |
parent | a2ebbd6948da96172108db5e9c02c141923ad05c (diff) | |
download | cgit-55ac326ecb01161bf62865ae3350acf85db97d63.zip cgit-55ac326ecb01161bf62865ae3350acf85db97d63.tar.gz cgit-55ac326ecb01161bf62865ae3350acf85db97d63.tar.bz2 |
Merge branch 'iconv-rebased' of http://x2a.org/pub/git/cgit
* 'iconv-rebased' of http://x2a.org/pub/git/cgit:
Use utf8::reencode_string from git
Convert subject and message with iconv_msg.
Add iconv_msg function.
Set msg_encoding according to the header.
Add commit->msg_encoding, allocate msg dynamicly.
-rw-r--r-- | parsing.c | 25 |
1 files changed, 25 insertions, 0 deletions
@@ -1,307 +1,332 @@ | |||
1 | /* config.c: parsing of config files | 1 | /* config.c: parsing of config files |
2 | * | 2 | * |
3 | * Copyright (C) 2006 Lars Hjemli | 3 | * Copyright (C) 2006 Lars Hjemli |
4 | * | 4 | * |
5 | * Licensed under GNU General Public License v2 | 5 | * Licensed under GNU General Public License v2 |
6 | * (see COPYING for full license text) | 6 | * (see COPYING for full license text) |
7 | */ | 7 | */ |
8 | 8 | ||
9 | #include "cgit.h" | 9 | #include "cgit.h" |
10 | 10 | ||
11 | int next_char(FILE *f) | 11 | int next_char(FILE *f) |
12 | { | 12 | { |
13 | int c = fgetc(f); | 13 | int c = fgetc(f); |
14 | if (c=='\r') { | 14 | if (c=='\r') { |
15 | c = fgetc(f); | 15 | c = fgetc(f); |
16 | if (c!='\n') { | 16 | if (c!='\n') { |
17 | ungetc(c, f); | 17 | ungetc(c, f); |
18 | c = '\r'; | 18 | c = '\r'; |
19 | } | 19 | } |
20 | } | 20 | } |
21 | return c; | 21 | return c; |
22 | } | 22 | } |
23 | 23 | ||
24 | void skip_line(FILE *f) | 24 | void skip_line(FILE *f) |
25 | { | 25 | { |
26 | int c; | 26 | int c; |
27 | 27 | ||
28 | while((c=next_char(f)) && c!='\n' && c!=EOF) | 28 | while((c=next_char(f)) && c!='\n' && c!=EOF) |
29 | ; | 29 | ; |
30 | } | 30 | } |
31 | 31 | ||
32 | int read_config_line(FILE *f, char *line, const char **value, int bufsize) | 32 | int read_config_line(FILE *f, char *line, const char **value, int bufsize) |
33 | { | 33 | { |
34 | int i = 0, isname = 0; | 34 | int i = 0, isname = 0; |
35 | 35 | ||
36 | *value = NULL; | 36 | *value = NULL; |
37 | while(i<bufsize-1) { | 37 | while(i<bufsize-1) { |
38 | int c = next_char(f); | 38 | int c = next_char(f); |
39 | if (!isname && (c=='#' || c==';')) { | 39 | if (!isname && (c=='#' || c==';')) { |
40 | skip_line(f); | 40 | skip_line(f); |
41 | continue; | 41 | continue; |
42 | } | 42 | } |
43 | if (!isname && isspace(c)) | 43 | if (!isname && isspace(c)) |
44 | continue; | 44 | continue; |
45 | 45 | ||
46 | if (c=='=' && !*value) { | 46 | if (c=='=' && !*value) { |
47 | line[i] = 0; | 47 | line[i] = 0; |
48 | *value = &line[i+1]; | 48 | *value = &line[i+1]; |
49 | } else if (c=='\n' && !isname) { | 49 | } else if (c=='\n' && !isname) { |
50 | i = 0; | 50 | i = 0; |
51 | continue; | 51 | continue; |
52 | } else if (c=='\n' || c==EOF) { | 52 | } else if (c=='\n' || c==EOF) { |
53 | line[i] = 0; | 53 | line[i] = 0; |
54 | break; | 54 | break; |
55 | } else { | 55 | } else { |
56 | line[i]=c; | 56 | line[i]=c; |
57 | } | 57 | } |
58 | isname = 1; | 58 | isname = 1; |
59 | i++; | 59 | i++; |
60 | } | 60 | } |
61 | line[i+1] = 0; | 61 | line[i+1] = 0; |
62 | return i; | 62 | return i; |
63 | } | 63 | } |
64 | 64 | ||
65 | int cgit_read_config(const char *filename, configfn fn) | 65 | int cgit_read_config(const char *filename, configfn fn) |
66 | { | 66 | { |
67 | static int nesting; | 67 | static int nesting; |
68 | int len; | 68 | int len; |
69 | char line[256]; | 69 | char line[256]; |
70 | const char *value; | 70 | const char *value; |
71 | FILE *f; | 71 | FILE *f; |
72 | 72 | ||
73 | /* cancel deeply nested include-commands */ | 73 | /* cancel deeply nested include-commands */ |
74 | if (nesting > 8) | 74 | if (nesting > 8) |
75 | return -1; | 75 | return -1; |
76 | if (!(f = fopen(filename, "r"))) | 76 | if (!(f = fopen(filename, "r"))) |
77 | return -1; | 77 | return -1; |
78 | nesting++; | 78 | nesting++; |
79 | while((len = read_config_line(f, line, &value, sizeof(line))) > 0) | 79 | while((len = read_config_line(f, line, &value, sizeof(line))) > 0) |
80 | (*fn)(line, value); | 80 | (*fn)(line, value); |
81 | nesting--; | 81 | nesting--; |
82 | fclose(f); | 82 | fclose(f); |
83 | return 0; | 83 | return 0; |
84 | } | 84 | } |
85 | 85 | ||
86 | char *convert_query_hexchar(char *txt) | 86 | char *convert_query_hexchar(char *txt) |
87 | { | 87 | { |
88 | int d1, d2; | 88 | int d1, d2; |
89 | if (strlen(txt) < 3) { | 89 | if (strlen(txt) < 3) { |
90 | *txt = '\0'; | 90 | *txt = '\0'; |
91 | return txt-1; | 91 | return txt-1; |
92 | } | 92 | } |
93 | d1 = hextoint(*(txt+1)); | 93 | d1 = hextoint(*(txt+1)); |
94 | d2 = hextoint(*(txt+2)); | 94 | d2 = hextoint(*(txt+2)); |
95 | if (d1<0 || d2<0) { | 95 | if (d1<0 || d2<0) { |
96 | strcpy(txt, txt+3); | 96 | strcpy(txt, txt+3); |
97 | return txt-1; | 97 | return txt-1; |
98 | } else { | 98 | } else { |
99 | *txt = d1 * 16 + d2; | 99 | *txt = d1 * 16 + d2; |
100 | strcpy(txt+1, txt+3); | 100 | strcpy(txt+1, txt+3); |
101 | return txt; | 101 | return txt; |
102 | } | 102 | } |
103 | } | 103 | } |
104 | 104 | ||
105 | int cgit_parse_query(char *txt, configfn fn) | 105 | int cgit_parse_query(char *txt, configfn fn) |
106 | { | 106 | { |
107 | char *t, *value = NULL, c; | 107 | char *t, *value = NULL, c; |
108 | 108 | ||
109 | if (!txt) | 109 | if (!txt) |
110 | return 0; | 110 | return 0; |
111 | 111 | ||
112 | t = txt = xstrdup(txt); | 112 | t = txt = xstrdup(txt); |
113 | 113 | ||
114 | while((c=*t) != '\0') { | 114 | while((c=*t) != '\0') { |
115 | if (c=='=') { | 115 | if (c=='=') { |
116 | *t = '\0'; | 116 | *t = '\0'; |
117 | value = t+1; | 117 | value = t+1; |
118 | } else if (c=='+') { | 118 | } else if (c=='+') { |
119 | *t = ' '; | 119 | *t = ' '; |
120 | } else if (c=='%') { | 120 | } else if (c=='%') { |
121 | t = convert_query_hexchar(t); | 121 | t = convert_query_hexchar(t); |
122 | } else if (c=='&') { | 122 | } else if (c=='&') { |
123 | *t = '\0'; | 123 | *t = '\0'; |
124 | (*fn)(txt, value); | 124 | (*fn)(txt, value); |
125 | txt = t+1; | 125 | txt = t+1; |
126 | value = NULL; | 126 | value = NULL; |
127 | } | 127 | } |
128 | t++; | 128 | t++; |
129 | } | 129 | } |
130 | if (t!=txt) | 130 | if (t!=txt) |
131 | (*fn)(txt, value); | 131 | (*fn)(txt, value); |
132 | return 0; | 132 | return 0; |
133 | } | 133 | } |
134 | 134 | ||
135 | /* | 135 | /* |
136 | * url syntax: [repo ['/' cmd [ '/' path]]] | 136 | * url syntax: [repo ['/' cmd [ '/' path]]] |
137 | * repo: any valid repo url, may contain '/' | 137 | * repo: any valid repo url, may contain '/' |
138 | * cmd: log | commit | diff | tree | view | blob | snapshot | 138 | * cmd: log | commit | diff | tree | view | blob | snapshot |
139 | * path: any valid path, may contain '/' | 139 | * path: any valid path, may contain '/' |
140 | * | 140 | * |
141 | */ | 141 | */ |
142 | void cgit_parse_url(const char *url) | 142 | void cgit_parse_url(const char *url) |
143 | { | 143 | { |
144 | char *cmd, *p; | 144 | char *cmd, *p; |
145 | 145 | ||
146 | cgit_repo = NULL; | 146 | cgit_repo = NULL; |
147 | if (!url || url[0] == '\0') | 147 | if (!url || url[0] == '\0') |
148 | return; | 148 | return; |
149 | 149 | ||
150 | cgit_repo = cgit_get_repoinfo(url); | 150 | cgit_repo = cgit_get_repoinfo(url); |
151 | if (cgit_repo) { | 151 | if (cgit_repo) { |
152 | cgit_query_repo = cgit_repo->url; | 152 | cgit_query_repo = cgit_repo->url; |
153 | return; | 153 | return; |
154 | } | 154 | } |
155 | 155 | ||
156 | cmd = strchr(url, '/'); | 156 | cmd = strchr(url, '/'); |
157 | while (!cgit_repo && cmd) { | 157 | while (!cgit_repo && cmd) { |
158 | cmd[0] = '\0'; | 158 | cmd[0] = '\0'; |
159 | cgit_repo = cgit_get_repoinfo(url); | 159 | cgit_repo = cgit_get_repoinfo(url); |
160 | if (cgit_repo == NULL) { | 160 | if (cgit_repo == NULL) { |
161 | cmd[0] = '/'; | 161 | cmd[0] = '/'; |
162 | cmd = strchr(cmd + 1, '/'); | 162 | cmd = strchr(cmd + 1, '/'); |
163 | continue; | 163 | continue; |
164 | } | 164 | } |
165 | 165 | ||
166 | cgit_query_repo = cgit_repo->url; | 166 | cgit_query_repo = cgit_repo->url; |
167 | p = strchr(cmd + 1, '/'); | 167 | p = strchr(cmd + 1, '/'); |
168 | if (p) { | 168 | if (p) { |
169 | p[0] = '\0'; | 169 | p[0] = '\0'; |
170 | if (p[1]) | 170 | if (p[1]) |
171 | cgit_query_path = trim_end(p + 1, '/'); | 171 | cgit_query_path = trim_end(p + 1, '/'); |
172 | } | 172 | } |
173 | cgit_cmd = cgit_get_cmd_index(cmd + 1); | 173 | cgit_cmd = cgit_get_cmd_index(cmd + 1); |
174 | cgit_query_page = xstrdup(cmd + 1); | 174 | cgit_query_page = xstrdup(cmd + 1); |
175 | return; | 175 | return; |
176 | } | 176 | } |
177 | } | 177 | } |
178 | 178 | ||
179 | char *substr(const char *head, const char *tail) | 179 | char *substr(const char *head, const char *tail) |
180 | { | 180 | { |
181 | char *buf; | 181 | char *buf; |
182 | 182 | ||
183 | buf = xmalloc(tail - head + 1); | 183 | buf = xmalloc(tail - head + 1); |
184 | strncpy(buf, head, tail - head); | 184 | strncpy(buf, head, tail - head); |
185 | buf[tail - head] = '\0'; | 185 | buf[tail - head] = '\0'; |
186 | return buf; | 186 | return buf; |
187 | } | 187 | } |
188 | 188 | ||
189 | struct commitinfo *cgit_parse_commit(struct commit *commit) | 189 | struct commitinfo *cgit_parse_commit(struct commit *commit) |
190 | { | 190 | { |
191 | struct commitinfo *ret; | 191 | struct commitinfo *ret; |
192 | char *p = commit->buffer, *t = commit->buffer; | 192 | char *p = commit->buffer, *t = commit->buffer; |
193 | 193 | ||
194 | ret = xmalloc(sizeof(*ret)); | 194 | ret = xmalloc(sizeof(*ret)); |
195 | ret->commit = commit; | 195 | ret->commit = commit; |
196 | ret->author = NULL; | 196 | ret->author = NULL; |
197 | ret->author_email = NULL; | 197 | ret->author_email = NULL; |
198 | ret->committer = NULL; | 198 | ret->committer = NULL; |
199 | ret->committer_email = NULL; | 199 | ret->committer_email = NULL; |
200 | ret->subject = NULL; | 200 | ret->subject = NULL; |
201 | ret->msg = NULL; | 201 | ret->msg = NULL; |
202 | ret->msg_encoding = NULL; | ||
202 | 203 | ||
203 | if (p == NULL) | 204 | if (p == NULL) |
204 | return ret; | 205 | return ret; |
205 | 206 | ||
206 | if (strncmp(p, "tree ", 5)) | 207 | if (strncmp(p, "tree ", 5)) |
207 | die("Bad commit: %s", sha1_to_hex(commit->object.sha1)); | 208 | die("Bad commit: %s", sha1_to_hex(commit->object.sha1)); |
208 | else | 209 | else |
209 | p += 46; // "tree " + hex[40] + "\n" | 210 | p += 46; // "tree " + hex[40] + "\n" |
210 | 211 | ||
211 | while (!strncmp(p, "parent ", 7)) | 212 | while (!strncmp(p, "parent ", 7)) |
212 | p += 48; // "parent " + hex[40] + "\n" | 213 | p += 48; // "parent " + hex[40] + "\n" |
213 | 214 | ||
214 | if (!strncmp(p, "author ", 7)) { | 215 | if (!strncmp(p, "author ", 7)) { |
215 | p += 7; | 216 | p += 7; |
216 | t = strchr(p, '<') - 1; | 217 | t = strchr(p, '<') - 1; |
217 | ret->author = substr(p, t); | 218 | ret->author = substr(p, t); |
218 | p = t; | 219 | p = t; |
219 | t = strchr(t, '>') + 1; | 220 | t = strchr(t, '>') + 1; |
220 | ret->author_email = substr(p, t); | 221 | ret->author_email = substr(p, t); |
221 | ret->author_date = atol(++t); | 222 | ret->author_date = atol(++t); |
222 | p = strchr(t, '\n') + 1; | 223 | p = strchr(t, '\n') + 1; |
223 | } | 224 | } |
224 | 225 | ||
225 | if (!strncmp(p, "committer ", 9)) { | 226 | if (!strncmp(p, "committer ", 9)) { |
226 | p += 9; | 227 | p += 9; |
227 | t = strchr(p, '<') - 1; | 228 | t = strchr(p, '<') - 1; |
228 | ret->committer = substr(p, t); | 229 | ret->committer = substr(p, t); |
229 | p = t; | 230 | p = t; |
230 | t = strchr(t, '>') + 1; | 231 | t = strchr(t, '>') + 1; |
231 | ret->committer_email = substr(p, t); | 232 | ret->committer_email = substr(p, t); |
232 | ret->committer_date = atol(++t); | 233 | ret->committer_date = atol(++t); |
233 | p = strchr(t, '\n') + 1; | 234 | p = strchr(t, '\n') + 1; |
234 | } | 235 | } |
235 | 236 | ||
237 | if (!strncmp(p, "encoding ", 9)) { | ||
238 | p += 9; | ||
239 | t = strchr(p, '\n') + 1; | ||
240 | ret->msg_encoding = substr(p, t); | ||
241 | p = t; | ||
242 | } else | ||
243 | ret->msg_encoding = xstrdup(PAGE_ENCODING); | ||
244 | |||
236 | while (*p && (*p != '\n')) | 245 | while (*p && (*p != '\n')) |
237 | p = strchr(p, '\n') + 1; // skip unknown header fields | 246 | p = strchr(p, '\n') + 1; // skip unknown header fields |
238 | 247 | ||
239 | while (*p == '\n') | 248 | while (*p == '\n') |
240 | p = strchr(p, '\n') + 1; | 249 | p = strchr(p, '\n') + 1; |
241 | 250 | ||
242 | t = strchr(p, '\n'); | 251 | t = strchr(p, '\n'); |
243 | if (t) { | 252 | if (t) { |
244 | if (*t == '\0') | 253 | if (*t == '\0') |
245 | ret->subject = "** empty **"; | 254 | ret->subject = "** empty **"; |
246 | else | 255 | else |
247 | ret->subject = substr(p, t); | 256 | ret->subject = substr(p, t); |
248 | p = t + 1; | 257 | p = t + 1; |
249 | 258 | ||
250 | while (*p == '\n') | 259 | while (*p == '\n') |
251 | p = strchr(p, '\n') + 1; | 260 | p = strchr(p, '\n') + 1; |
252 | ret->msg = xstrdup(p); | 261 | ret->msg = xstrdup(p); |
253 | } else | 262 | } else |
254 | ret->subject = substr(p, p+strlen(p)); | 263 | ret->subject = substr(p, p+strlen(p)); |
255 | 264 | ||
265 | if(strcmp(ret->msg_encoding, PAGE_ENCODING)) { | ||
266 | t = reencode_string(ret->subject, PAGE_ENCODING, | ||
267 | ret->msg_encoding); | ||
268 | if(t) { | ||
269 | free(ret->subject); | ||
270 | ret->subject = t; | ||
271 | } | ||
272 | |||
273 | t = reencode_string(ret->msg, PAGE_ENCODING, | ||
274 | ret->msg_encoding); | ||
275 | if(t) { | ||
276 | free(ret->msg); | ||
277 | ret->msg = t; | ||
278 | } | ||
279 | } | ||
280 | |||
256 | return ret; | 281 | return ret; |
257 | } | 282 | } |
258 | 283 | ||
259 | 284 | ||
260 | struct taginfo *cgit_parse_tag(struct tag *tag) | 285 | struct taginfo *cgit_parse_tag(struct tag *tag) |
261 | { | 286 | { |
262 | void *data; | 287 | void *data; |
263 | enum object_type type; | 288 | enum object_type type; |
264 | unsigned long size; | 289 | unsigned long size; |
265 | char *p, *t; | 290 | char *p, *t; |
266 | struct taginfo *ret; | 291 | struct taginfo *ret; |
267 | 292 | ||
268 | data = read_sha1_file(tag->object.sha1, &type, &size); | 293 | data = read_sha1_file(tag->object.sha1, &type, &size); |
269 | if (!data || type != OBJ_TAG) { | 294 | if (!data || type != OBJ_TAG) { |
270 | free(data); | 295 | free(data); |
271 | return 0; | 296 | return 0; |
272 | } | 297 | } |
273 | 298 | ||
274 | ret = xmalloc(sizeof(*ret)); | 299 | ret = xmalloc(sizeof(*ret)); |
275 | ret->tagger = NULL; | 300 | ret->tagger = NULL; |
276 | ret->tagger_email = NULL; | 301 | ret->tagger_email = NULL; |
277 | ret->tagger_date = 0; | 302 | ret->tagger_date = 0; |
278 | ret->msg = NULL; | 303 | ret->msg = NULL; |
279 | 304 | ||
280 | p = data; | 305 | p = data; |
281 | 306 | ||
282 | while (p && *p) { | 307 | while (p && *p) { |
283 | if (*p == '\n') | 308 | if (*p == '\n') |
284 | break; | 309 | break; |
285 | 310 | ||
286 | if (!strncmp(p, "tagger ", 7)) { | 311 | if (!strncmp(p, "tagger ", 7)) { |
287 | p += 7; | 312 | p += 7; |
288 | t = strchr(p, '<') - 1; | 313 | t = strchr(p, '<') - 1; |
289 | ret->tagger = substr(p, t); | 314 | ret->tagger = substr(p, t); |
290 | p = t; | 315 | p = t; |
291 | t = strchr(t, '>') + 1; | 316 | t = strchr(t, '>') + 1; |
292 | ret->tagger_email = substr(p, t); | 317 | ret->tagger_email = substr(p, t); |
293 | ret->tagger_date = atol(++t); | 318 | ret->tagger_date = atol(++t); |
294 | } | 319 | } |
295 | p = strchr(p, '\n') + 1; | 320 | p = strchr(p, '\n') + 1; |
296 | } | 321 | } |
297 | 322 | ||
298 | while (p && *p && (*p != '\n')) | 323 | while (p && *p && (*p != '\n')) |
299 | p = strchr(p, '\n') + 1; // skip unknown tag fields | 324 | p = strchr(p, '\n') + 1; // skip unknown tag fields |
300 | 325 | ||
301 | while (p && (*p == '\n')) | 326 | while (p && (*p == '\n')) |
302 | p = strchr(p, '\n') + 1; | 327 | p = strchr(p, '\n') + 1; |
303 | if (p && *p) | 328 | if (p && *p) |
304 | ret->msg = xstrdup(p); | 329 | ret->msg = xstrdup(p); |
305 | free(data); | 330 | free(data); |
306 | return ret; | 331 | return ret; |
307 | } | 332 | } |