Annotation of src/usr.bin/mandoc/html.c, Revision 1.3
1.3 ! schwarze 1: /* $Id: html.c,v 1.2 2009/10/27 21:40:07 schwarze Exp $ */
1.1 schwarze 2: /*
3: * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
4: *
5: * Permission to use, copy, modify, and distribute this software for any
6: * purpose with or without fee is hereby granted, provided that the above
7: * copyright notice and this permission notice appear in all copies.
8: *
9: * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
10: * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
11: * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
12: * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
13: * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
14: * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
15: * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
16: */
17: #include <sys/types.h>
18:
19: #include <assert.h>
1.3 ! schwarze 20: #include <ctype.h>
1.1 schwarze 21: #include <stdio.h>
22: #include <stdarg.h>
23: #include <stdint.h>
24: #include <stdlib.h>
25: #include <string.h>
26: #include <unistd.h>
27:
28: #include "out.h"
29: #include "chars.h"
30: #include "html.h"
31: #include "main.h"
32:
33: #define UNCONST(a) ((void *)(uintptr_t)(const void *)(a))
34:
35: #define DOCTYPE "-//W3C//DTD HTML 4.01//EN"
36: #define DTD "http://www.w3.org/TR/html4/strict.dtd"
37:
38: struct htmldata {
39: const char *name;
40: int flags;
41: #define HTML_CLRLINE (1 << 0)
42: #define HTML_NOSTACK (1 << 1)
43: };
44:
45: static const struct htmldata htmltags[TAG_MAX] = {
46: {"html", HTML_CLRLINE}, /* TAG_HTML */
47: {"head", HTML_CLRLINE}, /* TAG_HEAD */
48: {"body", HTML_CLRLINE}, /* TAG_BODY */
49: {"meta", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_META */
50: {"title", HTML_CLRLINE}, /* TAG_TITLE */
51: {"div", HTML_CLRLINE}, /* TAG_DIV */
52: {"h1", 0}, /* TAG_H1 */
53: {"h2", 0}, /* TAG_H2 */
54: {"p", HTML_CLRLINE}, /* TAG_P */
55: {"span", 0}, /* TAG_SPAN */
56: {"link", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_LINK */
57: {"br", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_LINK */
58: {"a", 0}, /* TAG_A */
59: {"table", HTML_CLRLINE}, /* TAG_TABLE */
60: {"col", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_COL */
61: {"tr", HTML_CLRLINE}, /* TAG_TR */
62: {"td", HTML_CLRLINE}, /* TAG_TD */
63: {"li", HTML_CLRLINE}, /* TAG_LI */
64: {"ul", HTML_CLRLINE}, /* TAG_UL */
65: {"ol", HTML_CLRLINE}, /* TAG_OL */
66: {"base", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_BASE */
67: };
68:
69: static const char *const htmlattrs[ATTR_MAX] = {
70: "http-equiv",
71: "content",
72: "name",
73: "rel",
74: "href",
75: "type",
76: "media",
77: "class",
78: "style",
79: "width",
80: "valign",
81: "target",
82: "id",
1.3 ! schwarze 83: "summary",
1.1 schwarze 84: };
85:
86: void *
87: html_alloc(char *outopts)
88: {
89: struct html *h;
90: const char *toks[4];
91: char *v;
92:
93: toks[0] = "style";
94: toks[1] = "man";
95: toks[2] = "includes";
96: toks[3] = NULL;
97:
1.3 ! schwarze 98: h = calloc(1, sizeof(struct html));
! 99: if (NULL == h) {
! 100: perror(NULL);
! 101: exit(EXIT_FAILURE);
! 102: }
1.1 schwarze 103:
1.2 schwarze 104: h->tags.head = NULL;
105: h->ords.head = NULL;
1.3 ! schwarze 106: h->symtab = chars_init(CHARS_HTML);
1.1 schwarze 107:
108: while (outopts && *outopts)
109: switch (getsubopt(&outopts, UNCONST(toks), &v)) {
110: case (0):
111: h->style = v;
112: break;
113: case (1):
114: h->base_man = v;
115: break;
116: case (2):
117: h->base_includes = v;
118: break;
119: default:
120: break;
121: }
122:
123: return(h);
124: }
125:
126:
127: void
128: html_free(void *p)
129: {
130: struct tag *tag;
131: struct ord *ord;
132: struct html *h;
133:
134: h = (struct html *)p;
135:
1.2 schwarze 136: while ((ord = h->ords.head) != NULL) {
137: h->ords.head = ord->next;
1.1 schwarze 138: free(ord);
139: }
140:
1.2 schwarze 141: while ((tag = h->tags.head) != NULL) {
142: h->tags.head = tag->next;
1.1 schwarze 143: free(tag);
144: }
145:
146: if (h->symtab)
147: chars_free(h->symtab);
148:
149: free(h);
150: }
151:
152:
153: void
154: print_gen_head(struct html *h)
155: {
156: struct htmlpair tag[4];
157:
158: tag[0].key = ATTR_HTTPEQUIV;
159: tag[0].val = "Content-Type";
160: tag[1].key = ATTR_CONTENT;
161: tag[1].val = "text/html; charset=utf-8";
162: print_otag(h, TAG_META, 2, tag);
163:
164: tag[0].key = ATTR_NAME;
165: tag[0].val = "resource-type";
166: tag[1].key = ATTR_CONTENT;
167: tag[1].val = "document";
168: print_otag(h, TAG_META, 2, tag);
169:
170: if (h->style) {
171: tag[0].key = ATTR_REL;
172: tag[0].val = "stylesheet";
173: tag[1].key = ATTR_HREF;
174: tag[1].val = h->style;
175: tag[2].key = ATTR_TYPE;
176: tag[2].val = "text/css";
177: tag[3].key = ATTR_MEDIA;
178: tag[3].val = "all";
179: print_otag(h, TAG_LINK, 4, tag);
180: }
181: }
182:
183:
184: static void
185: print_spec(struct html *h, const char *p, int len)
186: {
187: const char *rhs;
188: int i;
189: size_t sz;
190:
191: rhs = chars_a2ascii(h->symtab, p, (size_t)len, &sz);
192:
193: if (NULL == rhs)
194: return;
195: for (i = 0; i < (int)sz; i++)
196: putchar(rhs[i]);
197: }
198:
199:
200: static void
201: print_res(struct html *h, const char *p, int len)
202: {
203: const char *rhs;
204: int i;
205: size_t sz;
206:
207: rhs = chars_a2res(h->symtab, p, (size_t)len, &sz);
208:
209: if (NULL == rhs)
210: return;
211: for (i = 0; i < (int)sz; i++)
212: putchar(rhs[i]);
213: }
214:
215:
216: static void
217: print_escape(struct html *h, const char **p)
218: {
219: int j, type;
220: const char *wp;
221:
222: wp = *p;
223: type = 1;
224:
225: if (0 == *(++wp)) {
226: *p = wp;
227: return;
228: }
229:
230: if ('(' == *wp) {
231: wp++;
232: if (0 == *wp || 0 == *(wp + 1)) {
233: *p = 0 == *wp ? wp : wp + 1;
234: return;
235: }
236:
237: print_spec(h, wp, 2);
238: *p = ++wp;
239: return;
240:
241: } else if ('*' == *wp) {
242: if (0 == *(++wp)) {
243: *p = wp;
244: return;
245: }
246:
247: switch (*wp) {
248: case ('('):
249: wp++;
250: if (0 == *wp || 0 == *(wp + 1)) {
251: *p = 0 == *wp ? wp : wp + 1;
252: return;
253: }
254:
255: print_res(h, wp, 2);
256: *p = ++wp;
257: return;
258: case ('['):
259: type = 0;
260: break;
261: default:
262: print_res(h, wp, 1);
263: *p = wp;
264: return;
265: }
266:
267: } else if ('f' == *wp) {
268: if (0 == *(++wp)) {
269: *p = wp;
270: return;
271: }
272:
273: switch (*wp) {
274: case ('B'):
275: /* TODO */
276: break;
277: case ('I'):
278: /* TODO */
279: break;
280: case ('P'):
281: /* FALLTHROUGH */
282: case ('R'):
283: /* TODO */
284: break;
285: default:
286: break;
287: }
288:
289: *p = wp;
290: return;
291:
292: } else if ('[' != *wp) {
293: print_spec(h, wp, 1);
294: *p = wp;
295: return;
296: }
297:
298: wp++;
299: for (j = 0; *wp && ']' != *wp; wp++, j++)
300: /* Loop... */ ;
301:
302: if (0 == *wp) {
303: *p = wp;
304: return;
305: }
306:
307: if (type)
308: print_spec(h, wp - j, j);
309: else
310: print_res(h, wp - j, j);
311:
312: *p = wp;
313: }
314:
315:
316: static void
317: print_encode(struct html *h, const char *p)
318: {
319:
320: for (; *p; p++) {
321: if ('\\' == *p) {
322: print_escape(h, &p);
323: continue;
324: }
325: switch (*p) {
326: case ('<'):
327: printf("<");
328: break;
329: case ('>'):
330: printf(">");
331: break;
332: case ('&'):
333: printf("&");
334: break;
335: default:
336: putchar(*p);
337: break;
338: }
339: }
340: }
341:
342:
343: struct tag *
344: print_otag(struct html *h, enum htmltag tag,
345: int sz, const struct htmlpair *p)
346: {
347: int i;
348: struct tag *t;
349:
350: if ( ! (HTML_NOSTACK & htmltags[tag].flags)) {
1.3 ! schwarze 351: t = malloc(sizeof(struct tag));
! 352: if (NULL == t) {
! 353: perror(NULL);
! 354: exit(EXIT_FAILURE);
! 355: }
1.1 schwarze 356: t->tag = tag;
1.2 schwarze 357: t->next = h->tags.head;
358: h->tags.head = t;
1.1 schwarze 359: } else
360: t = NULL;
361:
362: if ( ! (HTML_NOSPACE & h->flags))
363: if ( ! (HTML_CLRLINE & htmltags[tag].flags))
364: printf(" ");
365:
366: printf("<%s", htmltags[tag].name);
367: for (i = 0; i < sz; i++) {
368: printf(" %s=\"", htmlattrs[p[i].key]);
369: assert(p->val);
370: print_encode(h, p[i].val);
371: printf("\"");
372: }
373: printf(">");
374:
375: h->flags |= HTML_NOSPACE;
376: if (HTML_CLRLINE & htmltags[tag].flags)
377: h->flags |= HTML_NEWLINE;
378: else
379: h->flags &= ~HTML_NEWLINE;
380:
381: return(t);
382: }
383:
384:
385: /* ARGSUSED */
386: static void
387: print_ctag(struct html *h, enum htmltag tag)
388: {
389:
390: printf("</%s>", htmltags[tag].name);
1.3 ! schwarze 391: if (HTML_CLRLINE & htmltags[tag].flags) {
1.1 schwarze 392: h->flags |= HTML_NOSPACE;
393: h->flags |= HTML_NEWLINE;
1.3 ! schwarze 394: printf("\n");
! 395: } else
1.1 schwarze 396: h->flags &= ~HTML_NEWLINE;
397: }
398:
399:
400: /* ARGSUSED */
401: void
402: print_gen_doctype(struct html *h)
403: {
404:
405: printf("<!DOCTYPE HTML PUBLIC \"%s\" \"%s\">", DOCTYPE, DTD);
406: }
407:
408:
409: void
410: print_text(struct html *h, const char *p)
411: {
412:
413: if (*p && 0 == *(p + 1))
414: switch (*p) {
415: case('.'):
416: /* FALLTHROUGH */
417: case(','):
418: /* FALLTHROUGH */
419: case(';'):
420: /* FALLTHROUGH */
421: case(':'):
422: /* FALLTHROUGH */
423: case('?'):
424: /* FALLTHROUGH */
425: case('!'):
426: /* FALLTHROUGH */
427: case(')'):
428: /* FALLTHROUGH */
429: case(']'):
430: /* FALLTHROUGH */
431: case('}'):
432: if ( ! (HTML_IGNDELIM & h->flags))
433: h->flags |= HTML_NOSPACE;
434: break;
435: default:
436: break;
437: }
438:
439: if ( ! (h->flags & HTML_NOSPACE))
440: printf(" ");
441:
442: h->flags &= ~HTML_NOSPACE;
443: h->flags &= ~HTML_NEWLINE;
444:
445: if (p)
446: print_encode(h, p);
447:
448: if (*p && 0 == *(p + 1))
449: switch (*p) {
450: case('('):
451: /* FALLTHROUGH */
452: case('['):
453: /* FALLTHROUGH */
454: case('{'):
455: h->flags |= HTML_NOSPACE;
456: break;
457: default:
458: break;
459: }
460: }
461:
462:
463: void
464: print_tagq(struct html *h, const struct tag *until)
465: {
466: struct tag *tag;
467:
1.2 schwarze 468: while ((tag = h->tags.head) != NULL) {
1.1 schwarze 469: print_ctag(h, tag->tag);
1.2 schwarze 470: h->tags.head = tag->next;
1.1 schwarze 471: free(tag);
472: if (until && tag == until)
473: return;
474: }
475: }
476:
477:
478: void
479: print_stagq(struct html *h, const struct tag *suntil)
480: {
481: struct tag *tag;
482:
1.2 schwarze 483: while ((tag = h->tags.head) != NULL) {
1.1 schwarze 484: if (suntil && tag == suntil)
485: return;
486: print_ctag(h, tag->tag);
1.2 schwarze 487: h->tags.head = tag->next;
1.1 schwarze 488: free(tag);
489: }
490: }
491:
492:
493: void
494: bufinit(struct html *h)
495: {
496:
497: h->buf[0] = '\0';
498: h->buflen = 0;
499: }
500:
501:
502: void
503: bufcat_style(struct html *h, const char *key, const char *val)
504: {
505:
506: bufcat(h, key);
507: bufncat(h, ":", 1);
508: bufcat(h, val);
509: bufncat(h, ";", 1);
510: }
511:
512:
513: void
514: bufcat(struct html *h, const char *p)
515: {
516:
517: bufncat(h, p, strlen(p));
518: }
519:
520:
521: void
522: buffmt(struct html *h, const char *fmt, ...)
523: {
524: va_list ap;
525:
526: va_start(ap, fmt);
527: (void)vsnprintf(h->buf + (int)h->buflen,
528: BUFSIZ - h->buflen - 1, fmt, ap);
529: va_end(ap);
530: h->buflen = strlen(h->buf);
531: }
532:
533:
534: void
535: bufncat(struct html *h, const char *p, size_t sz)
536: {
537:
538: if (h->buflen + sz > BUFSIZ - 1)
539: sz = BUFSIZ - 1 - h->buflen;
540:
541: (void)strncat(h->buf, p, sz);
542: h->buflen += sz;
543: }
544:
545:
546: void
547: buffmt_includes(struct html *h, const char *name)
548: {
549: const char *p, *pp;
550:
551: pp = h->base_includes;
552:
553: while (NULL != (p = strchr(pp, '%'))) {
554: bufncat(h, pp, (size_t)(p - pp));
555: switch (*(p + 1)) {
556: case('I'):
557: bufcat(h, name);
558: break;
559: default:
560: bufncat(h, p, 2);
561: break;
562: }
563: pp = p + 2;
564: }
565: if (pp)
566: bufcat(h, pp);
567: }
568:
569:
570: void
571: buffmt_man(struct html *h,
572: const char *name, const char *sec)
573: {
574: const char *p, *pp;
575:
576: pp = h->base_man;
577:
578: /* LINTED */
579: while (NULL != (p = strchr(pp, '%'))) {
580: bufncat(h, pp, (size_t)(p - pp));
581: switch (*(p + 1)) {
582: case('S'):
583: bufcat(h, sec ? sec : "1");
584: break;
585: case('N'):
586: buffmt(h, name);
587: break;
588: default:
589: bufncat(h, p, 2);
590: break;
591: }
592: pp = p + 2;
593: }
594: if (pp)
595: bufcat(h, pp);
596: }
597:
598:
599: void
600: bufcat_su(struct html *h, const char *p, const struct roffsu *su)
601: {
602: double v;
603: const char *u;
604:
605: v = su->scale;
606:
607: switch (su->unit) {
608: case (SCALE_CM):
609: u = "cm";
610: break;
611: case (SCALE_IN):
612: u = "in";
613: break;
614: case (SCALE_PC):
615: u = "pc";
616: break;
617: case (SCALE_PT):
618: u = "pt";
619: break;
620: case (SCALE_EM):
621: u = "em";
622: break;
623: case (SCALE_MM):
624: if (0 == (v /= 100))
625: v = 1;
626: u = "em";
627: break;
628: case (SCALE_EN):
629: u = "ex";
630: break;
631: case (SCALE_BU):
632: u = "ex";
633: break;
634: case (SCALE_VS):
635: u = "em";
636: break;
637: default:
638: u = "ex";
639: break;
640: }
641:
642: if (su->pt)
643: buffmt(h, "%s: %f%s;", p, v, u);
644: else
645: /* LINTED */
646: buffmt(h, "%s: %d%s;", p, (int)v, u);
647: }
648:
1.3 ! schwarze 649:
! 650: void
! 651: html_idcat(char *dst, const char *src, int sz)
! 652: {
! 653: int ssz;
! 654:
! 655: assert(sz);
! 656:
! 657: /* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
! 658:
! 659: for ( ; *dst != '\0' && sz; dst++, sz--)
! 660: /* Jump to end. */ ;
! 661:
! 662: assert(sz > 2);
! 663:
! 664: /* We can't start with a number (bah). */
! 665:
! 666: *dst++ = 'x';
! 667: *dst = '\0';
! 668: sz--;
! 669:
! 670: for ( ; *src != '\0' && sz > 1; src++) {
! 671: ssz = snprintf(dst, (size_t)sz, "%.2x", *src);
! 672: sz -= ssz;
! 673: dst += ssz;
! 674: }
! 675: }