1 /* $Id: html.c,v 1.159 2014/07/23 15:00:08 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_escape(char);
114 static int print_encode(struct html
*, const char *, int);
115 static void print_metaf(struct html
*, enum mandoc_esc
);
116 static void print_attr(struct html
*, const char *, const char *);
117 static void *ml_alloc(char *, enum htmltype
);
121 ml_alloc(char *outopts
, enum htmltype type
)
129 toks
[2] = "includes";
130 toks
[3] = "fragment";
133 h
= mandoc_calloc(1, sizeof(struct html
));
137 h
->symtab
= mchars_alloc();
139 while (outopts
&& *outopts
)
140 switch (getsubopt(&outopts
, UNCONST(toks
), &v
)) {
148 h
->base_includes
= v
;
151 h
->oflags
|= HTML_FRAGMENT
;
161 html_alloc(char *outopts
)
164 return(ml_alloc(outopts
, HTML_HTML_4_01_STRICT
));
168 xhtml_alloc(char *outopts
)
171 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
);
194 print_gen_head(struct html
*h
)
196 struct htmlpair tag
[4];
198 tag
[0].key
= ATTR_HTTPEQUIV
;
199 tag
[0].val
= "Content-Type";
200 tag
[1].key
= ATTR_CONTENT
;
201 tag
[1].val
= "text/html; charset=utf-8";
202 print_otag(h
, TAG_META
, 2, tag
);
204 tag
[0].key
= ATTR_NAME
;
205 tag
[0].val
= "resource-type";
206 tag
[1].key
= ATTR_CONTENT
;
207 tag
[1].val
= "document";
208 print_otag(h
, TAG_META
, 2, tag
);
211 tag
[0].key
= ATTR_REL
;
212 tag
[0].val
= "stylesheet";
213 tag
[1].key
= ATTR_HREF
;
214 tag
[1].val
= h
->style
;
215 tag
[2].key
= ATTR_TYPE
;
216 tag
[2].val
= "text/css";
217 tag
[3].key
= ATTR_MEDIA
;
219 print_otag(h
, TAG_LINK
, 4, tag
);
224 print_metaf(struct html
*h
, enum mandoc_esc deco
)
229 case ESCAPE_FONTPREV
:
232 case ESCAPE_FONTITALIC
:
233 font
= HTMLFONT_ITALIC
;
235 case ESCAPE_FONTBOLD
:
236 font
= HTMLFONT_BOLD
;
243 case ESCAPE_FONTROMAN
:
244 font
= HTMLFONT_NONE
;
252 print_tagq(h
, h
->metaf
);
260 case HTMLFONT_ITALIC
:
261 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
264 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
267 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
268 print_otag(h
, TAG_I
, 0, NULL
);
276 html_strlen(const char *cp
)
282 * Account for escaped sequences within string length
283 * calculations. This follows the logic in term_strlen() as we
284 * must calculate the width of produced strings.
285 * Assume that characters are always width of "1". This is
286 * hacky, but it gets the job done for approximation of widths.
292 rsz
= strcspn(cp
, "\\");
304 switch (mandoc_escape(&cp
, NULL
, NULL
)) {
309 case ESCAPE_NUMBERED
:
317 case ESCAPE_SKIPCHAR
:
359 print_encode(struct html
*h
, const char *p
, int norecurse
)
365 static const char rejs
[9] = { '\\', '<', '>', '&', '"',
366 ASCII_NBRSP
, ASCII_HYPH
, ASCII_BREAK
, '\0' };
371 if (HTML_SKIPCHAR
& h
->flags
&& '\\' != *p
) {
372 h
->flags
&= ~HTML_SKIPCHAR
;
377 sz
= strcspn(p
, rejs
);
379 fwrite(p
, 1, sz
, stdout
);
385 if (print_escape(*p
++))
388 esc
= mandoc_escape(&p
, &seq
, &len
);
389 if (ESCAPE_ERROR
== esc
)
395 case ESCAPE_FONTPREV
:
397 case ESCAPE_FONTBOLD
:
399 case ESCAPE_FONTITALIC
:
403 case ESCAPE_FONTROMAN
:
407 case ESCAPE_SKIPCHAR
:
408 h
->flags
|= HTML_SKIPCHAR
;
414 if (h
->flags
& HTML_SKIPCHAR
) {
415 h
->flags
&= ~HTML_SKIPCHAR
;
421 /* Skip past "u" header. */
422 c
= mchars_num2uc(seq
+ 1, len
- 1);
426 case ESCAPE_NUMBERED
:
427 c
= mchars_num2char(seq
, len
);
428 if ( ! ('\0' == c
|| print_escape(c
)))
432 c
= mchars_spec2cp(h
->symtab
, seq
, len
);
435 else if (-1 == c
&& 1 == len
&&
452 print_attr(struct html
*h
, const char *key
, const char *val
)
454 printf(" %s=\"", key
);
455 (void)print_encode(h
, val
, 1);
460 print_otag(struct html
*h
, enum htmltag tag
,
461 int sz
, const struct htmlpair
*p
)
466 /* Push this tags onto the stack of open scopes. */
468 if ( ! (HTML_NOSTACK
& htmltags
[tag
].flags
)) {
469 t
= mandoc_malloc(sizeof(struct tag
));
471 t
->next
= h
->tags
.head
;
476 if ( ! (HTML_NOSPACE
& h
->flags
))
477 if ( ! (HTML_CLRLINE
& htmltags
[tag
].flags
)) {
479 if ( ! (HTML_KEEP
& h
->flags
)) {
480 if (HTML_PREKEEP
& h
->flags
)
481 h
->flags
|= HTML_KEEP
;
487 if ( ! (h
->flags
& HTML_NONOSPACE
))
488 h
->flags
&= ~HTML_NOSPACE
;
490 h
->flags
|= HTML_NOSPACE
;
492 /* Print out the tag name and attributes. */
494 printf("<%s", htmltags
[tag
].name
);
495 for (i
= 0; i
< sz
; i
++)
496 print_attr(h
, htmlattrs
[p
[i
].key
], p
[i
].val
);
498 /* Add non-overridable attributes. */
500 if (TAG_HTML
== tag
&& HTML_XHTML_1_0_STRICT
== h
->type
) {
501 print_attr(h
, "xmlns", "http://www.w3.org/1999/xhtml");
502 print_attr(h
, "xml:lang", "en");
503 print_attr(h
, "lang", "en");
506 /* Accommodate for XML "well-formed" singleton escaping. */
508 if (HTML_AUTOCLOSE
& htmltags
[tag
].flags
)
510 case HTML_XHTML_1_0_STRICT
:
519 h
->flags
|= HTML_NOSPACE
;
521 if ((HTML_AUTOCLOSE
| HTML_CLRLINE
) & htmltags
[tag
].flags
)
528 print_ctag(struct html
*h
, enum htmltag tag
)
531 printf("</%s>", htmltags
[tag
].name
);
532 if (HTML_CLRLINE
& htmltags
[tag
].flags
) {
533 h
->flags
|= HTML_NOSPACE
;
539 print_gen_decls(struct html
*h
)
546 case HTML_HTML_4_01_STRICT
:
548 doctype
= "-//W3C//DTD HTML 4.01//EN";
549 dtd
= "http://www.w3.org/TR/html4/strict.dtd";
552 puts("<?xml version=\"1.0\" encoding=\"UTF-8\"?>");
554 doctype
= "-//W3C//DTD XHTML 1.0 Strict//EN";
555 dtd
= "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd";
559 printf("<!DOCTYPE %s PUBLIC \"%s\" \"%s\">\n",
564 print_text(struct html
*h
, const char *word
)
567 if ( ! (HTML_NOSPACE
& h
->flags
)) {
569 if ( ! (HTML_KEEP
& h
->flags
)) {
570 if (HTML_PREKEEP
& h
->flags
)
571 h
->flags
|= HTML_KEEP
;
577 assert(NULL
== h
->metaf
);
579 case HTMLFONT_ITALIC
:
580 h
->metaf
= print_otag(h
, TAG_I
, 0, NULL
);
583 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
586 h
->metaf
= print_otag(h
, TAG_B
, 0, NULL
);
587 print_otag(h
, TAG_I
, 0, NULL
);
594 if ( ! print_encode(h
, word
, 0)) {
595 if ( ! (h
->flags
& HTML_NONOSPACE
))
596 h
->flags
&= ~HTML_NOSPACE
;
598 h
->flags
|= HTML_NOSPACE
;
601 print_tagq(h
, h
->metaf
);
605 h
->flags
&= ~HTML_IGNDELIM
;
609 print_tagq(struct html
*h
, const struct tag
*until
)
613 while ((tag
= h
->tags
.head
) != NULL
) {
615 * Remember to close out and nullify the current
616 * meta-font and table, if applicable.
622 print_ctag(h
, tag
->tag
);
623 h
->tags
.head
= tag
->next
;
625 if (until
&& tag
== until
)
631 print_stagq(struct html
*h
, const struct tag
*suntil
)
635 while ((tag
= h
->tags
.head
) != NULL
) {
636 if (suntil
&& tag
== suntil
)
639 * Remember to close out and nullify the current
640 * meta-font and table, if applicable.
646 print_ctag(h
, tag
->tag
);
647 h
->tags
.head
= tag
->next
;
653 bufinit(struct html
*h
)
661 bufcat_style(struct html
*h
, const char *key
, const char *val
)
671 bufcat(struct html
*h
, const char *p
)
675 * XXX This is broken and not easy to fix.
676 * When using the -Oincludes option, buffmt_includes()
677 * may pass in strings overrunning BUFSIZ, causing a crash.
680 h
->buflen
= strlcat(h
->buf
, p
, BUFSIZ
);
681 assert(h
->buflen
< BUFSIZ
);
685 bufcat_fmt(struct html
*h
, const char *fmt
, ...)
690 (void)vsnprintf(h
->buf
+ (int)h
->buflen
,
691 BUFSIZ
- h
->buflen
- 1, fmt
, ap
);
693 h
->buflen
= strlen(h
->buf
);
697 bufncat(struct html
*h
, const char *p
, size_t sz
)
700 assert(h
->buflen
+ sz
+ 1 < BUFSIZ
);
701 strncat(h
->buf
, p
, sz
);
706 buffmt_includes(struct html
*h
, const char *name
)
710 pp
= h
->base_includes
;
713 while (NULL
!= (p
= strchr(pp
, '%'))) {
714 bufncat(h
, pp
, (size_t)(p
- pp
));
730 buffmt_man(struct html
*h
, const char *name
, const char *sec
)
737 while (NULL
!= (p
= strchr(pp
, '%'))) {
738 bufncat(h
, pp
, (size_t)(p
- pp
));
741 bufcat(h
, sec
? sec
: "1");
744 bufcat_fmt(h
, "%s", name
);
757 bufcat_su(struct html
*h
, const char *p
, const struct roffsu
*su
)
762 if (SCALE_MM
== su
->unit
&& 0.0 == (v
/= 100.0))
765 bufcat_fmt(h
, "%s: %.2f%s;", p
, v
, roffscales
[su
->unit
]);
769 bufcat_id(struct html
*h
, const char *src
)
772 /* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
775 bufcat_fmt(h
, "%.2x", *src
++);