]>
git.cameronkatri.com Git - mandoc.git/blob - html.c
1 /* $Id: html.c,v 1.164 2014/09/27 08:54:34 kristaps 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 */
79 static const char *const htmlattrs
[ATTR_MAX
] = {
80 "http-equiv", /* ATTR_HTTPEQUIV */
81 "content", /* ATTR_CONTENT */
82 "name", /* ATTR_NAME */
84 "href", /* ATTR_HREF */
85 "type", /* ATTR_TYPE */
86 "media", /* ATTR_MEDIA */
87 "class", /* ATTR_CLASS */
88 "style", /* ATTR_STYLE */
89 "width", /* ATTR_WIDTH */
91 "summary", /* ATTR_SUMMARY */
92 "align", /* ATTR_ALIGN */
93 "colspan", /* ATTR_COLSPAN */
94 "charset", /* ATTR_CHARSET */
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_CHARSET
;
198 tag
[0].val
= "utf-8";
199 print_otag(h
, TAG_META
, 1, tag
);
202 tag
[0].key
= ATTR_REL
;
203 tag
[0].val
= "stylesheet";
204 tag
[1].key
= ATTR_HREF
;
205 tag
[1].val
= h
->style
;
206 tag
[2].key
= ATTR_TYPE
;
207 tag
[2].val
= "text/css";
208 tag
[3].key
= ATTR_MEDIA
;
210 print_otag(h
, TAG_LINK
, 4, tag
);
215 print_metaf(struct html
*h
, enum mandoc_esc deco
)
220 case ESCAPE_FONTPREV
:
223 case ESCAPE_FONTITALIC
:
224 font
= HTMLFONT_ITALIC
;
226 case ESCAPE_FONTBOLD
:
227 font
= HTMLFONT_BOLD
;
234 case ESCAPE_FONTROMAN
:
235 font
= HTMLFONT_NONE
;
243 print_tagq(h
, h
->metaf
);
251 case HTMLFONT_ITALIC
:
252 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
255 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
258 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
259 print_otag(h
, TAG_I
, 0, NULL
);
267 html_strlen(const char *cp
)
273 * Account for escaped sequences within string length
274 * calculations. This follows the logic in term_strlen() as we
275 * must calculate the width of produced strings.
276 * Assume that characters are always width of "1". This is
277 * hacky, but it gets the job done for approximation of widths.
283 rsz
= strcspn(cp
, "\\");
295 switch (mandoc_escape(&cp
, NULL
, NULL
)) {
300 case ESCAPE_NUMBERED
:
308 case ESCAPE_SKIPCHAR
:
350 print_encode(struct html
*h
, const char *p
, int norecurse
)
356 static const char rejs
[9] = { '\\', '<', '>', '&', '"',
357 ASCII_NBRSP
, ASCII_HYPH
, ASCII_BREAK
, '\0' };
362 if (HTML_SKIPCHAR
& h
->flags
&& '\\' != *p
) {
363 h
->flags
&= ~HTML_SKIPCHAR
;
368 sz
= strcspn(p
, rejs
);
370 fwrite(p
, 1, sz
, stdout
);
376 if (print_escape(*p
++))
379 esc
= mandoc_escape(&p
, &seq
, &len
);
380 if (ESCAPE_ERROR
== esc
)
386 case ESCAPE_FONTPREV
:
388 case ESCAPE_FONTBOLD
:
390 case ESCAPE_FONTITALIC
:
394 case ESCAPE_FONTROMAN
:
398 case ESCAPE_SKIPCHAR
:
399 h
->flags
|= HTML_SKIPCHAR
;
405 if (h
->flags
& HTML_SKIPCHAR
) {
406 h
->flags
&= ~HTML_SKIPCHAR
;
412 /* Skip past "u" header. */
413 c
= mchars_num2uc(seq
+ 1, len
- 1);
417 case ESCAPE_NUMBERED
:
418 c
= mchars_num2char(seq
, len
);
419 if ( ! ('\0' == c
|| print_escape(c
)))
423 c
= mchars_spec2cp(h
->symtab
, seq
, len
);
426 else if (-1 == c
&& 1 == len
&&
443 print_attr(struct html
*h
, const char *key
, const char *val
)
445 printf(" %s=\"", key
);
446 (void)print_encode(h
, val
, 1);
451 print_otag(struct html
*h
, enum htmltag tag
,
452 int sz
, const struct htmlpair
*p
)
457 /* Push this tags onto the stack of open scopes. */
459 if ( ! (HTML_NOSTACK
& htmltags
[tag
].flags
)) {
460 t
= mandoc_malloc(sizeof(struct tag
));
462 t
->next
= h
->tags
.head
;
467 if ( ! (HTML_NOSPACE
& h
->flags
))
468 if ( ! (HTML_CLRLINE
& htmltags
[tag
].flags
)) {
470 if ( ! (HTML_KEEP
& h
->flags
)) {
471 if (HTML_PREKEEP
& h
->flags
)
472 h
->flags
|= HTML_KEEP
;
478 if ( ! (h
->flags
& HTML_NONOSPACE
))
479 h
->flags
&= ~HTML_NOSPACE
;
481 h
->flags
|= HTML_NOSPACE
;
483 /* Print out the tag name and attributes. */
485 printf("<%s", htmltags
[tag
].name
);
486 for (i
= 0; i
< sz
; i
++)
487 print_attr(h
, htmlattrs
[p
[i
].key
], p
[i
].val
);
489 /* Accommodate for XML "well-formed" singleton escaping. */
491 if (HTML_AUTOCLOSE
& htmltags
[tag
].flags
)
493 case HTML_XHTML_1_0_STRICT
:
502 h
->flags
|= HTML_NOSPACE
;
504 if ((HTML_AUTOCLOSE
| HTML_CLRLINE
) & htmltags
[tag
].flags
)
511 print_ctag(struct html
*h
, enum htmltag tag
)
514 printf("</%s>", htmltags
[tag
].name
);
515 if (HTML_CLRLINE
& htmltags
[tag
].flags
) {
516 h
->flags
|= HTML_NOSPACE
;
522 print_gen_decls(struct html
*h
)
525 puts("<!DOCTYPE html>");
529 print_text(struct html
*h
, const char *word
)
532 if ( ! (HTML_NOSPACE
& h
->flags
)) {
534 if ( ! (HTML_KEEP
& h
->flags
)) {
535 if (HTML_PREKEEP
& h
->flags
)
536 h
->flags
|= HTML_KEEP
;
542 assert(NULL
== h
->metaf
);
544 case HTMLFONT_ITALIC
:
545 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
548 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
551 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
552 print_otag(h
, TAG_I
, 0, NULL
);
559 if ( ! print_encode(h
, word
, 0)) {
560 if ( ! (h
->flags
& HTML_NONOSPACE
))
561 h
->flags
&= ~HTML_NOSPACE
;
563 h
->flags
|= HTML_NOSPACE
;
566 print_tagq(h
, h
->metaf
);
570 h
->flags
&= ~HTML_IGNDELIM
;
574 print_tagq(struct html
*h
, const struct tag
*until
)
578 while ((tag
= h
->tags
.head
) != NULL
) {
580 * Remember to close out and nullify the current
581 * meta-font and table, if applicable.
587 print_ctag(h
, tag
->tag
);
588 h
->tags
.head
= tag
->next
;
590 if (until
&& tag
== until
)
596 print_stagq(struct html
*h
, const struct tag
*suntil
)
600 while ((tag
= h
->tags
.head
) != NULL
) {
601 if (suntil
&& tag
== suntil
)
604 * Remember to close out and nullify the current
605 * meta-font and table, if applicable.
611 print_ctag(h
, tag
->tag
);
612 h
->tags
.head
= tag
->next
;
618 bufinit(struct html
*h
)
626 bufcat_style(struct html
*h
, const char *key
, const char *val
)
636 bufcat(struct html
*h
, const char *p
)
640 * XXX This is broken and not easy to fix.
641 * When using the -Oincludes option, buffmt_includes()
642 * may pass in strings overrunning BUFSIZ, causing a crash.
645 h
->buflen
= strlcat(h
->buf
, p
, BUFSIZ
);
646 assert(h
->buflen
< BUFSIZ
);
650 bufcat_fmt(struct html
*h
, const char *fmt
, ...)
655 (void)vsnprintf(h
->buf
+ (int)h
->buflen
,
656 BUFSIZ
- h
->buflen
- 1, fmt
, ap
);
658 h
->buflen
= strlen(h
->buf
);
662 bufncat(struct html
*h
, const char *p
, size_t sz
)
665 assert(h
->buflen
+ sz
+ 1 < BUFSIZ
);
666 strncat(h
->buf
, p
, sz
);
671 buffmt_includes(struct html
*h
, const char *name
)
675 pp
= h
->base_includes
;
678 while (NULL
!= (p
= strchr(pp
, '%'))) {
679 bufncat(h
, pp
, (size_t)(p
- pp
));
695 buffmt_man(struct html
*h
, const char *name
, const char *sec
)
702 while (NULL
!= (p
= strchr(pp
, '%'))) {
703 bufncat(h
, pp
, (size_t)(p
- pp
));
706 bufcat(h
, sec
? sec
: "1");
709 bufcat_fmt(h
, "%s", name
);
722 bufcat_su(struct html
*h
, const char *p
, const struct roffsu
*su
)
727 if (SCALE_MM
== su
->unit
&& 0.0 == (v
/= 100.0))
729 else if (SCALE_BU
== su
->unit
)
732 bufcat_fmt(h
, "%s: %.2f%s;", p
, v
, roffscales
[su
->unit
]);
736 bufcat_id(struct html
*h
, const char *src
)
739 /* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
742 bufcat_fmt(h
, "%.2x", *src
++);