]>
git.cameronkatri.com Git - mandoc.git/blob - html.c
1 /* $Id: html.c,v 1.161 2014/08/13 15:25:22 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.
20 #include <sys/types.h>
32 #include "mandoc_aux.h"
33 #include "libmandoc.h"
41 #define HTML_CLRLINE (1 << 0)
42 #define HTML_NOSTACK (1 << 1)
43 #define HTML_AUTOCLOSE (1 << 2) /* Tag has auto-closure. */
46 static const struct htmldata htmltags
[TAG_MAX
] = {
47 {"html", HTML_CLRLINE
}, /* TAG_HTML */
48 {"head", HTML_CLRLINE
}, /* TAG_HEAD */
49 {"body", HTML_CLRLINE
}, /* TAG_BODY */
50 {"meta", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_META */
51 {"title", HTML_CLRLINE
}, /* TAG_TITLE */
52 {"div", HTML_CLRLINE
}, /* TAG_DIV */
53 {"h1", 0}, /* TAG_H1 */
54 {"h2", 0}, /* TAG_H2 */
55 {"span", 0}, /* TAG_SPAN */
56 {"link", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_LINK */
57 {"br", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_BR */
59 {"table", HTML_CLRLINE
}, /* TAG_TABLE */
60 {"tbody", HTML_CLRLINE
}, /* TAG_TBODY */
61 {"col", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_COL */
62 {"tr", HTML_CLRLINE
}, /* TAG_TR */
63 {"td", HTML_CLRLINE
}, /* TAG_TD */
64 {"li", HTML_CLRLINE
}, /* TAG_LI */
65 {"ul", HTML_CLRLINE
}, /* TAG_UL */
66 {"ol", HTML_CLRLINE
}, /* TAG_OL */
67 {"dl", HTML_CLRLINE
}, /* TAG_DL */
68 {"dt", HTML_CLRLINE
}, /* TAG_DT */
69 {"dd", HTML_CLRLINE
}, /* TAG_DD */
70 {"blockquote", HTML_CLRLINE
}, /* TAG_BLOCKQUOTE */
71 {"p", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_P */
72 {"pre", HTML_CLRLINE
}, /* TAG_PRE */
73 {"b", 0 }, /* TAG_B */
74 {"i", 0 }, /* TAG_I */
75 {"code", 0 }, /* TAG_CODE */
76 {"small", 0 }, /* TAG_SMALL */
77 {"em", 0 }, /* TAG_EM */
80 static const char *const htmlattrs
[ATTR_MAX
] = {
81 "http-equiv", /* ATTR_HTTPEQUIV */
82 "content", /* ATTR_CONTENT */
83 "name", /* ATTR_NAME */
85 "href", /* ATTR_HREF */
86 "type", /* ATTR_TYPE */
87 "media", /* ATTR_MEDIA */
88 "class", /* ATTR_CLASS */
89 "style", /* ATTR_STYLE */
90 "width", /* ATTR_WIDTH */
92 "summary", /* ATTR_SUMMARY */
93 "align", /* ATTR_ALIGN */
94 "colspan", /* ATTR_COLSPAN */
97 static const char *const roffscales
[SCALE_MAX
] = {
110 static void bufncat(struct html
*, const char *, size_t);
111 static void print_ctag(struct html
*, enum htmltag
);
112 static int print_escape(char);
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
);
120 ml_alloc(char *outopts
, enum htmltype type
)
128 toks
[2] = "includes";
129 toks
[3] = "fragment";
132 h
= mandoc_calloc(1, sizeof(struct html
));
136 h
->symtab
= mchars_alloc();
138 while (outopts
&& *outopts
)
139 switch (getsubopt(&outopts
, UNCONST(toks
), &v
)) {
147 h
->base_includes
= v
;
150 h
->oflags
|= HTML_FRAGMENT
;
160 html_alloc(char *outopts
)
163 return(ml_alloc(outopts
, HTML_HTML_4_01_STRICT
));
167 xhtml_alloc(char *outopts
)
170 return(ml_alloc(outopts
, HTML_XHTML_1_0_STRICT
));
179 h
= (struct html
*)p
;
181 while ((tag
= h
->tags
.head
) != NULL
) {
182 h
->tags
.head
= tag
->next
;
187 mchars_free(h
->symtab
);
193 print_gen_head(struct html
*h
)
195 struct htmlpair tag
[4];
197 tag
[0].key
= ATTR_HTTPEQUIV
;
198 tag
[0].val
= "Content-Type";
199 tag
[1].key
= ATTR_CONTENT
;
200 tag
[1].val
= "text/html; charset=utf-8";
201 print_otag(h
, TAG_META
, 2, tag
);
203 tag
[0].key
= ATTR_NAME
;
204 tag
[0].val
= "resource-type";
205 tag
[1].key
= ATTR_CONTENT
;
206 tag
[1].val
= "document";
207 print_otag(h
, TAG_META
, 2, tag
);
210 tag
[0].key
= ATTR_REL
;
211 tag
[0].val
= "stylesheet";
212 tag
[1].key
= ATTR_HREF
;
213 tag
[1].val
= h
->style
;
214 tag
[2].key
= ATTR_TYPE
;
215 tag
[2].val
= "text/css";
216 tag
[3].key
= ATTR_MEDIA
;
218 print_otag(h
, TAG_LINK
, 4, tag
);
223 print_metaf(struct html
*h
, enum mandoc_esc deco
)
228 case ESCAPE_FONTPREV
:
231 case ESCAPE_FONTITALIC
:
232 font
= HTMLFONT_ITALIC
;
234 case ESCAPE_FONTBOLD
:
235 font
= HTMLFONT_BOLD
;
242 case ESCAPE_FONTROMAN
:
243 font
= HTMLFONT_NONE
;
251 print_tagq(h
, h
->metaf
);
259 case HTMLFONT_ITALIC
:
260 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
263 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
266 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
267 print_otag(h
, TAG_I
, 0, NULL
);
275 html_strlen(const char *cp
)
281 * Account for escaped sequences within string length
282 * calculations. This follows the logic in term_strlen() as we
283 * must calculate the width of produced strings.
284 * Assume that characters are always width of "1". This is
285 * hacky, but it gets the job done for approximation of widths.
291 rsz
= strcspn(cp
, "\\");
303 switch (mandoc_escape(&cp
, NULL
, NULL
)) {
308 case ESCAPE_NUMBERED
:
316 case ESCAPE_SKIPCHAR
:
358 print_encode(struct html
*h
, const char *p
, int norecurse
)
364 static const char rejs
[9] = { '\\', '<', '>', '&', '"',
365 ASCII_NBRSP
, ASCII_HYPH
, ASCII_BREAK
, '\0' };
370 if (HTML_SKIPCHAR
& h
->flags
&& '\\' != *p
) {
371 h
->flags
&= ~HTML_SKIPCHAR
;
376 sz
= strcspn(p
, rejs
);
378 fwrite(p
, 1, sz
, stdout
);
384 if (print_escape(*p
++))
387 esc
= mandoc_escape(&p
, &seq
, &len
);
388 if (ESCAPE_ERROR
== esc
)
394 case ESCAPE_FONTPREV
:
396 case ESCAPE_FONTBOLD
:
398 case ESCAPE_FONTITALIC
:
402 case ESCAPE_FONTROMAN
:
406 case ESCAPE_SKIPCHAR
:
407 h
->flags
|= HTML_SKIPCHAR
;
413 if (h
->flags
& HTML_SKIPCHAR
) {
414 h
->flags
&= ~HTML_SKIPCHAR
;
420 /* Skip past "u" header. */
421 c
= mchars_num2uc(seq
+ 1, len
- 1);
425 case ESCAPE_NUMBERED
:
426 c
= mchars_num2char(seq
, len
);
427 if ( ! ('\0' == c
|| print_escape(c
)))
431 c
= mchars_spec2cp(h
->symtab
, seq
, len
);
434 else if (-1 == c
&& 1 == len
&&
451 print_attr(struct html
*h
, const char *key
, const char *val
)
453 printf(" %s=\"", key
);
454 (void)print_encode(h
, val
, 1);
459 print_otag(struct html
*h
, enum htmltag tag
,
460 int sz
, const struct htmlpair
*p
)
465 /* Push this tags onto the stack of open scopes. */
467 if ( ! (HTML_NOSTACK
& htmltags
[tag
].flags
)) {
468 t
= mandoc_malloc(sizeof(struct tag
));
470 t
->next
= h
->tags
.head
;
475 if ( ! (HTML_NOSPACE
& h
->flags
))
476 if ( ! (HTML_CLRLINE
& htmltags
[tag
].flags
)) {
478 if ( ! (HTML_KEEP
& h
->flags
)) {
479 if (HTML_PREKEEP
& h
->flags
)
480 h
->flags
|= HTML_KEEP
;
486 if ( ! (h
->flags
& HTML_NONOSPACE
))
487 h
->flags
&= ~HTML_NOSPACE
;
489 h
->flags
|= HTML_NOSPACE
;
491 /* Print out the tag name and attributes. */
493 printf("<%s", htmltags
[tag
].name
);
494 for (i
= 0; i
< sz
; i
++)
495 print_attr(h
, htmlattrs
[p
[i
].key
], p
[i
].val
);
497 /* Add non-overridable attributes. */
499 if (TAG_HTML
== tag
&& HTML_XHTML_1_0_STRICT
== h
->type
) {
500 print_attr(h
, "xmlns", "http://www.w3.org/1999/xhtml");
501 print_attr(h
, "xml:lang", "en");
502 print_attr(h
, "lang", "en");
505 /* Accommodate for XML "well-formed" singleton escaping. */
507 if (HTML_AUTOCLOSE
& htmltags
[tag
].flags
)
509 case HTML_XHTML_1_0_STRICT
:
518 h
->flags
|= HTML_NOSPACE
;
520 if ((HTML_AUTOCLOSE
| HTML_CLRLINE
) & htmltags
[tag
].flags
)
527 print_ctag(struct html
*h
, enum htmltag tag
)
530 printf("</%s>", htmltags
[tag
].name
);
531 if (HTML_CLRLINE
& htmltags
[tag
].flags
) {
532 h
->flags
|= HTML_NOSPACE
;
538 print_gen_decls(struct html
*h
)
545 case HTML_HTML_4_01_STRICT
:
547 doctype
= "-//W3C//DTD HTML 4.01//EN";
548 dtd
= "http://www.w3.org/TR/html4/strict.dtd";
551 puts("<?xml version=\"1.0\" encoding=\"UTF-8\"?>");
553 doctype
= "-//W3C//DTD XHTML 1.0 Strict//EN";
554 dtd
= "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd";
558 printf("<!DOCTYPE %s PUBLIC \"%s\" \"%s\">\n",
563 print_text(struct html
*h
, const char *word
)
566 if ( ! (HTML_NOSPACE
& h
->flags
)) {
568 if ( ! (HTML_KEEP
& h
->flags
)) {
569 if (HTML_PREKEEP
& h
->flags
)
570 h
->flags
|= HTML_KEEP
;
576 assert(NULL
== h
->metaf
);
578 case HTMLFONT_ITALIC
:
579 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
582 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
585 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
586 print_otag(h
, TAG_I
, 0, NULL
);
593 if ( ! print_encode(h
, word
, 0)) {
594 if ( ! (h
->flags
& HTML_NONOSPACE
))
595 h
->flags
&= ~HTML_NOSPACE
;
597 h
->flags
|= HTML_NOSPACE
;
600 print_tagq(h
, h
->metaf
);
604 h
->flags
&= ~HTML_IGNDELIM
;
608 print_tagq(struct html
*h
, const struct tag
*until
)
612 while ((tag
= h
->tags
.head
) != NULL
) {
614 * Remember to close out and nullify the current
615 * meta-font and table, if applicable.
621 print_ctag(h
, tag
->tag
);
622 h
->tags
.head
= tag
->next
;
624 if (until
&& tag
== until
)
630 print_stagq(struct html
*h
, const struct tag
*suntil
)
634 while ((tag
= h
->tags
.head
) != NULL
) {
635 if (suntil
&& tag
== suntil
)
638 * Remember to close out and nullify the current
639 * meta-font and table, if applicable.
645 print_ctag(h
, tag
->tag
);
646 h
->tags
.head
= tag
->next
;
652 bufinit(struct html
*h
)
660 bufcat_style(struct html
*h
, const char *key
, const char *val
)
670 bufcat(struct html
*h
, const char *p
)
674 * XXX This is broken and not easy to fix.
675 * When using the -Oincludes option, buffmt_includes()
676 * may pass in strings overrunning BUFSIZ, causing a crash.
679 h
->buflen
= strlcat(h
->buf
, p
, BUFSIZ
);
680 assert(h
->buflen
< BUFSIZ
);
684 bufcat_fmt(struct html
*h
, const char *fmt
, ...)
689 (void)vsnprintf(h
->buf
+ (int)h
->buflen
,
690 BUFSIZ
- h
->buflen
- 1, fmt
, ap
);
692 h
->buflen
= strlen(h
->buf
);
696 bufncat(struct html
*h
, const char *p
, size_t sz
)
699 assert(h
->buflen
+ sz
+ 1 < BUFSIZ
);
700 strncat(h
->buf
, p
, sz
);
705 buffmt_includes(struct html
*h
, const char *name
)
709 pp
= h
->base_includes
;
712 while (NULL
!= (p
= strchr(pp
, '%'))) {
713 bufncat(h
, pp
, (size_t)(p
- pp
));
729 buffmt_man(struct html
*h
, const char *name
, const char *sec
)
736 while (NULL
!= (p
= strchr(pp
, '%'))) {
737 bufncat(h
, pp
, (size_t)(p
- pp
));
740 bufcat(h
, sec
? sec
: "1");
743 bufcat_fmt(h
, "%s", name
);
756 bufcat_su(struct html
*h
, const char *p
, const struct roffsu
*su
)
761 if (SCALE_MM
== su
->unit
&& 0.0 == (v
/= 100.0))
764 bufcat_fmt(h
, "%s: %.2f%s;", p
, v
, roffscales
[su
->unit
]);
768 bufcat_id(struct html
*h
, const char *src
)
771 /* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
774 bufcat_fmt(h
, "%.2x", *src
++);