2 * Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
3 * Copyright (C) 1999-2005 Hiroyuki Yamamoto
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License
16 * along with this program; if not, write to the Free Software
17 * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
29 #define HTMLBUFSIZE 8192
30 #define HR_STR "------------------------------------------------"
32 typedef struct _HTMLSymbol HTMLSymbol;
40 static HTMLSymbol symbol_list[] = {
51 static HTMLSymbol ascii_symbol_list[] = {
114 static GHashTable *default_symbol_table;
116 static HTMLState html_read_line (HTMLParser *parser);
117 static void html_append_char (HTMLParser *parser,
119 static void html_append_str (HTMLParser *parser,
122 static HTMLState html_parse_tag (HTMLParser *parser);
123 static void html_parse_special (HTMLParser *parser);
124 static void html_get_parenthesis (HTMLParser *parser,
129 HTMLParser *html_parser_new(FILE *fp, CodeConverter *conv)
133 g_return_val_if_fail(fp != NULL, NULL);
134 g_return_val_if_fail(conv != NULL, NULL);
136 parser = g_new0(HTMLParser, 1);
139 parser->str = g_string_new(NULL);
140 parser->buf = g_string_new(NULL);
141 parser->bufp = parser->buf->str;
142 parser->state = HTML_NORMAL;
144 parser->newline = TRUE;
145 parser->empty_line = TRUE;
146 parser->space = FALSE;
149 #define SYMBOL_TABLE_ADD(table, list) \
153 for (i = 0; i < sizeof(list) / sizeof(list[0]); i++) \
154 g_hash_table_insert(table, list[i].key, list[i].val); \
157 if (!default_symbol_table) {
158 default_symbol_table =
159 g_hash_table_new(g_str_hash, g_str_equal);
160 SYMBOL_TABLE_ADD(default_symbol_table, symbol_list);
161 SYMBOL_TABLE_ADD(default_symbol_table, ascii_symbol_list);
164 #undef SYMBOL_TABLE_ADD
166 parser->symbol_table = default_symbol_table;
171 void html_parser_destroy(HTMLParser *parser)
173 g_string_free(parser->str, TRUE);
174 g_string_free(parser->buf, TRUE);
175 g_free(parser->href);
179 gchar *html_parse(HTMLParser *parser)
181 parser->state = HTML_NORMAL;
182 g_string_truncate(parser->str, 0);
184 if (*parser->bufp == '\0') {
185 g_string_truncate(parser->buf, 0);
186 parser->bufp = parser->buf->str;
187 if (html_read_line(parser) == HTML_EOF)
191 while (*parser->bufp != '\0') {
192 switch (*parser->bufp) {
195 st = html_parse_tag(parser);
196 /* when we see an href, we need to flush the str
197 * buffer. Then collect all the chars until we
198 * see the end anchor tag
200 if (HTML_HREF_BEG == st || HTML_HREF == st)
201 return parser->str->str;
205 html_parse_special(parser);
211 if (parser->bufp[0] == '\r' && parser->bufp[1] == '\n')
215 if (!parser->newline)
216 parser->space = TRUE;
223 html_append_char(parser, *parser->bufp++);
227 return parser->str->str;
230 static HTMLState html_read_line(HTMLParser *parser)
232 gchar buf[HTMLBUFSIZE];
233 gchar buf2[HTMLBUFSIZE];
236 if (fgets(buf, sizeof(buf), parser->fp) == NULL) {
237 parser->state = HTML_EOF;
241 if (conv_convert(parser->conv, buf2, sizeof(buf2), buf) < 0) {
242 index = parser->bufp - parser->buf->str;
244 conv_utf8todisp(buf2, sizeof(buf2), buf);
245 g_string_append(parser->buf, buf2);
247 parser->bufp = parser->buf->str + index;
249 return HTML_CONV_FAILED;
252 index = parser->bufp - parser->buf->str;
254 g_string_append(parser->buf, buf2);
256 parser->bufp = parser->buf->str + index;
261 static void html_append_char(HTMLParser *parser, gchar ch)
263 GString *str = parser->str;
265 if (!parser->pre && parser->space) {
266 g_string_append_c(str, ' ');
267 parser->space = FALSE;
270 g_string_append_c(str, ch);
272 parser->empty_line = FALSE;
274 parser->newline = TRUE;
275 if (str->len > 1 && str->str[str->len - 2] == '\n')
276 parser->empty_line = TRUE;
278 parser->newline = FALSE;
281 static void html_append_str(HTMLParser *parser, const gchar *str, gint len)
283 GString *string = parser->str;
285 if (!parser->pre && parser->space) {
286 g_string_append_c(string, ' ');
287 parser->space = FALSE;
290 if (len == 0) return;
292 g_string_append(string, str);
295 Xstrndup_a(s, str, len, return);
296 g_string_append(string, s);
299 parser->empty_line = FALSE;
300 if (string->len > 0 && string->str[string->len - 1] == '\n') {
301 parser->newline = TRUE;
302 if (string->len > 1 && string->str[string->len - 2] == '\n')
303 parser->empty_line = TRUE;
305 parser->newline = FALSE;
308 static HTMLTag *html_get_tag(const gchar *str)
314 g_return_val_if_fail(str != NULL, NULL);
316 if (*str == '\0' || *str == '!') return NULL;
318 Xstrdup_a(tmp, str, return NULL);
320 tag = g_new0(HTMLTag, 1);
322 for (tmpp = tmp; *tmpp != '\0' && !g_ascii_isspace(*tmpp); tmpp++)
327 tag->name = g_strdup(tmp);
332 tag->name = g_strdup(tmp);
335 while (*tmpp != '\0') {
342 while (g_ascii_isspace(*tmpp)) tmpp++;
345 while (*tmpp != '\0' && !g_ascii_isspace(*tmpp) &&
348 if (*tmpp != '\0' && *tmpp != '=') {
350 while (g_ascii_isspace(*tmpp)) tmpp++;
355 while (g_ascii_isspace(*tmpp)) tmpp++;
357 if (*tmpp == '"' || *tmpp == '\'') {
362 if ((p = strchr(attr_value, quote)) == NULL) {
363 g_warning("html_get_tag(): syntax error in tag: '%s'\n", str);
368 while (g_ascii_isspace(*tmpp)) tmpp++;
372 while (*tmpp != '\0' && !g_ascii_isspace(*tmpp)) tmpp++;
379 g_strchomp(attr_name);
380 g_strdown(attr_name);
381 attr = g_new(HTMLAttr, 1);
382 attr->name = g_strdup(attr_name);
383 attr->value = g_strdup(attr_value);
384 tag->attr = g_list_append(tag->attr, attr);
390 static void html_free_tag(HTMLTag *tag)
395 while (tag->attr != NULL) {
396 HTMLAttr *attr = (HTMLAttr *)tag->attr->data;
400 tag->attr = g_list_remove(tag->attr, tag->attr->data);
405 static HTMLState html_parse_tag(HTMLParser *parser)
407 gchar buf[HTMLBUFSIZE];
410 html_get_parenthesis(parser, buf, sizeof(buf));
412 tag = html_get_tag(buf);
414 parser->state = HTML_UNKNOWN;
415 if (!tag) return HTML_UNKNOWN;
417 if (!strcmp(tag->name, "br")) {
418 parser->space = FALSE;
419 html_append_char(parser, '\n');
420 parser->state = HTML_BR;
421 } else if (!strcmp(tag->name, "a")) {
422 if (tag->attr && tag->attr->data &&
423 !strcmp(((HTMLAttr *)tag->attr->data)->name, "href")) {
424 g_free(parser->href);
426 g_strdup(((HTMLAttr *)tag->attr->data)->value);
427 parser->state = HTML_HREF_BEG;
429 } else if (!strcmp(tag->name, "/a")) {
430 parser->state = HTML_HREF;
431 } else if (!strcmp(tag->name, "p")) {
432 parser->space = FALSE;
433 if (!parser->empty_line) {
434 parser->space = FALSE;
435 if (!parser->newline) html_append_char(parser, '\n');
436 html_append_char(parser, '\n');
438 parser->state = HTML_PAR;
439 } else if (!strcmp(tag->name, "pre")) {
441 parser->state = HTML_PRE;
442 } else if (!strcmp(tag->name, "/pre")) {
444 parser->state = HTML_NORMAL;
445 } else if (!strcmp(tag->name, "hr")) {
446 if (!parser->newline) {
447 parser->space = FALSE;
448 html_append_char(parser, '\n');
450 html_append_str(parser, HR_STR "\n", -1);
451 parser->state = HTML_HR;
452 } else if (!strcmp(tag->name, "div") ||
453 !strcmp(tag->name, "ul") ||
454 !strcmp(tag->name, "li") ||
455 !strcmp(tag->name, "table") ||
456 !strcmp(tag->name, "tr") ||
457 (tag->name[0] == 'h' && g_ascii_isdigit(tag->name[1]))) {
458 if (!parser->newline) {
459 parser->space = FALSE;
460 html_append_char(parser, '\n');
462 parser->state = HTML_NORMAL;
463 } else if (!strcmp(tag->name, "/table") ||
464 (tag->name[0] == '/' &&
465 tag->name[1] == 'h' &&
466 g_ascii_isdigit(tag->name[1]))) {
467 if (!parser->empty_line) {
468 parser->space = FALSE;
469 if (!parser->newline) html_append_char(parser, '\n');
470 html_append_char(parser, '\n');
472 parser->state = HTML_NORMAL;
473 } else if (!strcmp(tag->name, "/div") ||
474 !strcmp(tag->name, "/ul") ||
475 !strcmp(tag->name, "/li")) {
476 if (!parser->newline) {
477 parser->space = FALSE;
478 html_append_char(parser, '\n');
480 parser->state = HTML_NORMAL;
485 return parser->state;
488 static void html_parse_special(HTMLParser *parser)
490 gchar symbol_name[9];
494 parser->state = HTML_UNKNOWN;
495 g_return_if_fail(*parser->bufp == '&');
498 for (n = 0; parser->bufp[n] != '\0' && parser->bufp[n] != ';'; n++)
500 if (n > 7 || parser->bufp[n] != ';') {
501 /* output literal `&' */
502 html_append_char(parser, *parser->bufp++);
503 parser->state = HTML_NORMAL;
506 strncpy2(symbol_name, parser->bufp, n + 2);
507 parser->bufp += n + 1;
509 if ((val = g_hash_table_lookup(parser->symbol_table, symbol_name))
511 html_append_str(parser, val, -1);
512 parser->state = HTML_NORMAL;
514 } else if (symbol_name[1] == '#' && g_ascii_isdigit(symbol_name[2])) {
517 ch = atoi(symbol_name + 2);
518 if ((ch > 0 && ch <= 127) ||
519 (ch >= 128 && ch <= 255 &&
520 parser->conv->charset == C_ISO_8859_1)) {
521 html_append_char(parser, ch);
522 parser->state = HTML_NORMAL;
527 /* http://www.w3schools.com/html/html_entitiesref.asp */
528 case 338: /* capital ligature OE Œ */
531 case 339: /* small ligature OE œ */
534 case 352: /* capital S w/caron Š */
535 case 353: /* small S w/caron š */
536 case 376: /* cap Y w/ diaeres Ÿ */
538 case 710: /* circumflex accent ˆ */
541 case 732: /* small tilde ˜ */
544 case 8194: /* en space   */
545 case 8195: /* em space   */
546 case 8201: /* thin space   */
549 case 8204: /* zero width non-joiner ‌ */
550 case 8205: /* zero width joiner ‍ */
551 case 8206: /* l-t-r mark ‎ */
552 case 8207: /* r-t-l mark &rlm */
554 case 8211: /* en dash – */
557 case 8212: /* em dash — */
560 case 8216: /* l single quot mark ‘ */
563 case 8217: /* r single quot mark ’ */
566 case 8218: /* single low-9 quot ‚ */
569 case 8220: /* l double quot mark “ */
572 case 8221: /* r double quot mark ” */
575 case 8222: /* double low-9 quot „ */
578 case 8224: /* dagger † */
579 case 8225: /* double dagger ‡ */
581 case 8230: /* horizontal ellipsis … */
584 case 8240: /* per mile ‰ */
587 case 8249: /* l-pointing angle quot ‹ */
590 case 8250: /* r-pointing angle quot › */
593 case 8364: /* euro € */
596 case 8482: /* trademark ™ */
603 html_append_str(parser, symb, -1);
604 parser->state = HTML_NORMAL;
610 html_append_str(parser, symbol_name, -1);
613 static void html_get_parenthesis(HTMLParser *parser, gchar *buf, gint len)
618 g_return_if_fail(*parser->bufp == '<');
620 /* ignore comment / CSS / script stuff */
621 if (!strncmp(parser->bufp, "<!--", 4)) {
623 while ((p = strstr(parser->bufp, "-->")) == NULL)
624 if (html_read_line(parser) == HTML_EOF) return;
625 parser->bufp = p + 3;
628 if (!g_ascii_strncasecmp(parser->bufp, "<style", 6)) {
630 while ((p = strcasestr(parser->bufp, "</style>")) == NULL)
631 if (html_read_line(parser) == HTML_EOF) return;
632 parser->bufp = p + 8;
635 if (!g_ascii_strncasecmp(parser->bufp, "<script", 7)) {
637 while ((p = strcasestr(parser->bufp, "</script>")) == NULL)
638 if (html_read_line(parser) == HTML_EOF) return;
639 parser->bufp = p + 9;
644 while ((p = strchr(parser->bufp, '>')) == NULL)
645 if (html_read_line(parser) == HTML_EOF) return;
647 strncpy2(buf, parser->bufp, MIN(p - parser->bufp + 1, len));
649 parser->bufp = p + 1;