]>
git.cameronkatri.com Git - mandoc.git/blob - html.c
1 /* $Id: html.c,v 1.155 2014/03/23 11:25:26 schwarze Exp $ */
3 * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
4 * Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org>
6 * Permission to use, copy, modify, and distribute this software for any
7 * purpose with or without fee is hereby granted, provided that the above
8 * copyright notice and this permission notice appear in all copies.
10 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
11 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
12 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
13 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
14 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
15 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
16 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
22 #include <sys/types.h>
34 #include "mandoc_aux.h"
35 #include "libmandoc.h"
43 #define HTML_CLRLINE (1 << 0)
44 #define HTML_NOSTACK (1 << 1)
45 #define HTML_AUTOCLOSE (1 << 2) /* Tag has auto-closure. */
48 static const struct htmldata htmltags
[TAG_MAX
] = {
49 {"html", HTML_CLRLINE
}, /* TAG_HTML */
50 {"head", HTML_CLRLINE
}, /* TAG_HEAD */
51 {"body", HTML_CLRLINE
}, /* TAG_BODY */
52 {"meta", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_META */
53 {"title", HTML_CLRLINE
}, /* TAG_TITLE */
54 {"div", HTML_CLRLINE
}, /* TAG_DIV */
55 {"h1", 0}, /* TAG_H1 */
56 {"h2", 0}, /* TAG_H2 */
57 {"span", 0}, /* TAG_SPAN */
58 {"link", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_LINK */
59 {"br", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_BR */
61 {"table", HTML_CLRLINE
}, /* TAG_TABLE */
62 {"tbody", HTML_CLRLINE
}, /* TAG_TBODY */
63 {"col", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_COL */
64 {"tr", HTML_CLRLINE
}, /* TAG_TR */
65 {"td", HTML_CLRLINE
}, /* TAG_TD */
66 {"li", HTML_CLRLINE
}, /* TAG_LI */
67 {"ul", HTML_CLRLINE
}, /* TAG_UL */
68 {"ol", HTML_CLRLINE
}, /* TAG_OL */
69 {"dl", HTML_CLRLINE
}, /* TAG_DL */
70 {"dt", HTML_CLRLINE
}, /* TAG_DT */
71 {"dd", HTML_CLRLINE
}, /* TAG_DD */
72 {"blockquote", HTML_CLRLINE
}, /* TAG_BLOCKQUOTE */
73 {"p", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_P */
74 {"pre", HTML_CLRLINE
}, /* TAG_PRE */
75 {"b", 0 }, /* TAG_B */
76 {"i", 0 }, /* TAG_I */
77 {"code", 0 }, /* TAG_CODE */
78 {"small", 0 }, /* TAG_SMALL */
81 static const char *const htmlattrs
[ATTR_MAX
] = {
82 "http-equiv", /* ATTR_HTTPEQUIV */
83 "content", /* ATTR_CONTENT */
84 "name", /* ATTR_NAME */
86 "href", /* ATTR_HREF */
87 "type", /* ATTR_TYPE */
88 "media", /* ATTR_MEDIA */
89 "class", /* ATTR_CLASS */
90 "style", /* ATTR_STYLE */
91 "width", /* ATTR_WIDTH */
93 "summary", /* ATTR_SUMMARY */
94 "align", /* ATTR_ALIGN */
95 "colspan", /* ATTR_COLSPAN */
98 static const char *const roffscales
[SCALE_MAX
] = {
111 static void bufncat(struct html
*, const char *, size_t);
112 static void print_ctag(struct html
*, enum htmltag
);
113 static int print_encode(struct html
*, const char *, int);
114 static void print_metaf(struct html
*, enum mandoc_esc
);
115 static void print_attr(struct html
*, const char *, const char *);
116 static void *ml_alloc(char *, enum htmltype
);
119 ml_alloc(char *outopts
, enum htmltype type
)
127 toks
[2] = "includes";
128 toks
[3] = "fragment";
131 h
= mandoc_calloc(1, sizeof(struct html
));
135 h
->symtab
= mchars_alloc();
137 while (outopts
&& *outopts
)
138 switch (getsubopt(&outopts
, UNCONST(toks
), &v
)) {
146 h
->base_includes
= v
;
149 h
->oflags
|= HTML_FRAGMENT
;
159 html_alloc(char *outopts
)
162 return(ml_alloc(outopts
, HTML_HTML_4_01_STRICT
));
167 xhtml_alloc(char *outopts
)
170 return(ml_alloc(outopts
, HTML_XHTML_1_0_STRICT
));
180 h
= (struct html
*)p
;
182 while ((tag
= h
->tags
.head
) != NULL
) {
183 h
->tags
.head
= tag
->next
;
188 mchars_free(h
->symtab
);
195 print_gen_head(struct html
*h
)
197 struct htmlpair tag
[4];
199 tag
[0].key
= ATTR_HTTPEQUIV
;
200 tag
[0].val
= "Content-Type";
201 tag
[1].key
= ATTR_CONTENT
;
202 tag
[1].val
= "text/html; charset=utf-8";
203 print_otag(h
, TAG_META
, 2, tag
);
205 tag
[0].key
= ATTR_NAME
;
206 tag
[0].val
= "resource-type";
207 tag
[1].key
= ATTR_CONTENT
;
208 tag
[1].val
= "document";
209 print_otag(h
, TAG_META
, 2, tag
);
212 tag
[0].key
= ATTR_REL
;
213 tag
[0].val
= "stylesheet";
214 tag
[1].key
= ATTR_HREF
;
215 tag
[1].val
= h
->style
;
216 tag
[2].key
= ATTR_TYPE
;
217 tag
[2].val
= "text/css";
218 tag
[3].key
= ATTR_MEDIA
;
220 print_otag(h
, TAG_LINK
, 4, tag
);
225 print_metaf(struct html
*h
, enum mandoc_esc deco
)
230 case (ESCAPE_FONTPREV
):
233 case (ESCAPE_FONTITALIC
):
234 font
= HTMLFONT_ITALIC
;
236 case (ESCAPE_FONTBOLD
):
237 font
= HTMLFONT_BOLD
;
239 case (ESCAPE_FONTBI
):
244 case (ESCAPE_FONTROMAN
):
245 font
= HTMLFONT_NONE
;
253 print_tagq(h
, h
->metaf
);
261 case (HTMLFONT_ITALIC
):
262 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
264 case (HTMLFONT_BOLD
):
265 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
268 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
269 print_otag(h
, TAG_I
, 0, NULL
);
277 html_strlen(const char *cp
)
283 * Account for escaped sequences within string length
284 * calculations. This follows the logic in term_strlen() as we
285 * must calculate the width of produced strings.
286 * Assume that characters are always width of "1". This is
287 * hacky, but it gets the job done for approximation of widths.
293 rsz
= strcspn(cp
, "\\");
305 switch (mandoc_escape(&cp
, NULL
, NULL
)) {
308 case (ESCAPE_UNICODE
):
310 case (ESCAPE_NUMBERED
):
312 case (ESCAPE_SPECIAL
):
318 case (ESCAPE_SKIPCHAR
):
329 print_encode(struct html
*h
, const char *p
, int norecurse
)
335 static const char rejs
[8] = { '\\', '<', '>', '&',
336 ASCII_NBRSP
, ASCII_HYPH
, ASCII_BREAK
, '\0' };
341 if (HTML_SKIPCHAR
& h
->flags
&& '\\' != *p
) {
342 h
->flags
&= ~HTML_SKIPCHAR
;
347 sz
= strcspn(p
, rejs
);
349 fwrite(p
, 1, sz
, stdout
);
377 esc
= mandoc_escape(&p
, &seq
, &len
);
378 if (ESCAPE_ERROR
== esc
)
384 case (ESCAPE_FONTPREV
):
386 case (ESCAPE_FONTBOLD
):
388 case (ESCAPE_FONTITALIC
):
390 case (ESCAPE_FONTBI
):
392 case (ESCAPE_FONTROMAN
):
396 case (ESCAPE_SKIPCHAR
):
397 h
->flags
|= HTML_SKIPCHAR
;
403 if (h
->flags
& HTML_SKIPCHAR
) {
404 h
->flags
&= ~HTML_SKIPCHAR
;
409 case (ESCAPE_UNICODE
):
410 /* Skip passed "u" header. */
411 c
= mchars_num2uc(seq
+ 1, len
- 1);
415 case (ESCAPE_NUMBERED
):
416 c
= mchars_num2char(seq
, len
);
420 case (ESCAPE_SPECIAL
):
421 c
= mchars_spec2cp(h
->symtab
, seq
, len
);
424 else if (-1 == c
&& 1 == len
)
427 case (ESCAPE_NOSPACE
):
441 print_attr(struct html
*h
, const char *key
, const char *val
)
443 printf(" %s=\"", key
);
444 (void)print_encode(h
, val
, 1);
450 print_otag(struct html
*h
, enum htmltag tag
,
451 int sz
, const struct htmlpair
*p
)
456 /* Push this tags onto the stack of open scopes. */
458 if ( ! (HTML_NOSTACK
& htmltags
[tag
].flags
)) {
459 t
= mandoc_malloc(sizeof(struct tag
));
461 t
->next
= h
->tags
.head
;
466 if ( ! (HTML_NOSPACE
& h
->flags
))
467 if ( ! (HTML_CLRLINE
& htmltags
[tag
].flags
)) {
469 if ( ! (HTML_KEEP
& h
->flags
)) {
470 if (HTML_PREKEEP
& h
->flags
)
471 h
->flags
|= HTML_KEEP
;
477 if ( ! (h
->flags
& HTML_NONOSPACE
))
478 h
->flags
&= ~HTML_NOSPACE
;
480 h
->flags
|= HTML_NOSPACE
;
482 /* Print out the tag name and attributes. */
484 printf("<%s", htmltags
[tag
].name
);
485 for (i
= 0; i
< sz
; i
++)
486 print_attr(h
, htmlattrs
[p
[i
].key
], p
[i
].val
);
488 /* Add non-overridable attributes. */
490 if (TAG_HTML
== tag
&& HTML_XHTML_1_0_STRICT
== h
->type
) {
491 print_attr(h
, "xmlns", "http://www.w3.org/1999/xhtml");
492 print_attr(h
, "xml:lang", "en");
493 print_attr(h
, "lang", "en");
496 /* Accommodate for XML "well-formed" singleton escaping. */
498 if (HTML_AUTOCLOSE
& htmltags
[tag
].flags
)
500 case (HTML_XHTML_1_0_STRICT
):
509 h
->flags
|= HTML_NOSPACE
;
511 if ((HTML_AUTOCLOSE
| HTML_CLRLINE
) & htmltags
[tag
].flags
)
519 print_ctag(struct html
*h
, enum htmltag tag
)
522 printf("</%s>", htmltags
[tag
].name
);
523 if (HTML_CLRLINE
& htmltags
[tag
].flags
) {
524 h
->flags
|= HTML_NOSPACE
;
530 print_gen_decls(struct html
*h
)
537 case (HTML_HTML_4_01_STRICT
):
539 doctype
= "-//W3C//DTD HTML 4.01//EN";
540 dtd
= "http://www.w3.org/TR/html4/strict.dtd";
543 puts("<?xml version=\"1.0\" encoding=\"UTF-8\"?>");
545 doctype
= "-//W3C//DTD XHTML 1.0 Strict//EN";
546 dtd
= "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd";
550 printf("<!DOCTYPE %s PUBLIC \"%s\" \"%s\">\n",
555 print_text(struct html
*h
, const char *word
)
558 if ( ! (HTML_NOSPACE
& h
->flags
)) {
560 if ( ! (HTML_KEEP
& h
->flags
)) {
561 if (HTML_PREKEEP
& h
->flags
)
562 h
->flags
|= HTML_KEEP
;
568 assert(NULL
== h
->metaf
);
570 case (HTMLFONT_ITALIC
):
571 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
573 case (HTMLFONT_BOLD
):
574 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
577 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
578 print_otag(h
, TAG_I
, 0, NULL
);
585 if ( ! print_encode(h
, word
, 0)) {
586 if ( ! (h
->flags
& HTML_NONOSPACE
))
587 h
->flags
&= ~HTML_NOSPACE
;
589 h
->flags
|= HTML_NOSPACE
;
592 print_tagq(h
, h
->metaf
);
596 h
->flags
&= ~HTML_IGNDELIM
;
601 print_tagq(struct html
*h
, const struct tag
*until
)
605 while ((tag
= h
->tags
.head
) != NULL
) {
607 * Remember to close out and nullify the current
608 * meta-font and table, if applicable.
614 print_ctag(h
, tag
->tag
);
615 h
->tags
.head
= tag
->next
;
617 if (until
&& tag
== until
)
624 print_stagq(struct html
*h
, const struct tag
*suntil
)
628 while ((tag
= h
->tags
.head
) != NULL
) {
629 if (suntil
&& tag
== suntil
)
632 * Remember to close out and nullify the current
633 * meta-font and table, if applicable.
639 print_ctag(h
, tag
->tag
);
640 h
->tags
.head
= tag
->next
;
646 bufinit(struct html
*h
)
654 bufcat_style(struct html
*h
, const char *key
, const char *val
)
664 bufcat(struct html
*h
, const char *p
)
667 h
->buflen
= strlcat(h
->buf
, p
, BUFSIZ
);
668 assert(h
->buflen
< BUFSIZ
);
672 bufcat_fmt(struct html
*h
, const char *fmt
, ...)
677 (void)vsnprintf(h
->buf
+ (int)h
->buflen
,
678 BUFSIZ
- h
->buflen
- 1, fmt
, ap
);
680 h
->buflen
= strlen(h
->buf
);
684 bufncat(struct html
*h
, const char *p
, size_t sz
)
687 assert(h
->buflen
+ sz
+ 1 < BUFSIZ
);
688 strncat(h
->buf
, p
, sz
);
693 buffmt_includes(struct html
*h
, const char *name
)
697 pp
= h
->base_includes
;
700 while (NULL
!= (p
= strchr(pp
, '%'))) {
701 bufncat(h
, pp
, (size_t)(p
- pp
));
717 buffmt_man(struct html
*h
,
718 const char *name
, const char *sec
)
725 while (NULL
!= (p
= strchr(pp
, '%'))) {
726 bufncat(h
, pp
, (size_t)(p
- pp
));
729 bufcat(h
, sec
? sec
: "1");
732 bufcat_fmt(h
, "%s", name
);
745 bufcat_su(struct html
*h
, const char *p
, const struct roffsu
*su
)
750 if (SCALE_MM
== su
->unit
&& 0.0 == (v
/= 100.0))
753 bufcat_fmt(h
, "%s: %.2f%s;", p
, v
, roffscales
[su
->unit
]);
757 bufcat_id(struct html
*h
, const char *src
)
760 /* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
763 bufcat_fmt(h
, "%.2x", *src
++);