]>
git.cameronkatri.com Git - mandoc.git/blob - html.c
1 /* $Id: html.c,v 1.148 2011/07/04 09:42:38 kristaps Exp $ */
3 * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
4 * Copyright (c) 2011 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 "libmandoc.h"
42 #define HTML_CLRLINE (1 << 0)
43 #define HTML_NOSTACK (1 << 1)
44 #define HTML_AUTOCLOSE (1 << 2) /* Tag has auto-closure. */
47 static const struct htmldata htmltags
[TAG_MAX
] = {
48 {"html", HTML_CLRLINE
}, /* TAG_HTML */
49 {"head", HTML_CLRLINE
}, /* TAG_HEAD */
50 {"body", HTML_CLRLINE
}, /* TAG_BODY */
51 {"meta", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_META */
52 {"title", HTML_CLRLINE
}, /* TAG_TITLE */
53 {"div", HTML_CLRLINE
}, /* TAG_DIV */
54 {"h1", 0}, /* TAG_H1 */
55 {"h2", 0}, /* TAG_H2 */
56 {"span", 0}, /* TAG_SPAN */
57 {"link", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_LINK */
58 {"br", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_BR */
60 {"table", HTML_CLRLINE
}, /* TAG_TABLE */
61 {"tbody", HTML_CLRLINE
}, /* TAG_TBODY */
62 {"col", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_COL */
63 {"tr", HTML_CLRLINE
}, /* TAG_TR */
64 {"td", HTML_CLRLINE
}, /* TAG_TD */
65 {"li", HTML_CLRLINE
}, /* TAG_LI */
66 {"ul", HTML_CLRLINE
}, /* TAG_UL */
67 {"ol", HTML_CLRLINE
}, /* TAG_OL */
68 {"dl", HTML_CLRLINE
}, /* TAG_DL */
69 {"dt", HTML_CLRLINE
}, /* TAG_DT */
70 {"dd", HTML_CLRLINE
}, /* TAG_DD */
71 {"blockquote", HTML_CLRLINE
}, /* TAG_BLOCKQUOTE */
72 {"p", HTML_CLRLINE
| HTML_NOSTACK
| HTML_AUTOCLOSE
}, /* TAG_P */
73 {"pre", HTML_CLRLINE
}, /* TAG_PRE */
74 {"b", 0 }, /* TAG_B */
75 {"i", 0 }, /* TAG_I */
76 {"code", 0 }, /* TAG_CODE */
77 {"small", 0 }, /* TAG_SMALL */
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_encode(struct html
*, const char *, int);
113 static void print_metaf(struct html
*, enum mandoc_esc
);
114 static void print_attr(struct html
*, const char *, const char *);
115 static void *ml_alloc(char *, enum htmltype
);
118 ml_alloc(char *outopts
, enum htmltype type
)
126 toks
[2] = "includes";
129 h
= mandoc_calloc(1, sizeof(struct html
));
133 h
->symtab
= mchars_alloc();
135 while (outopts
&& *outopts
)
136 switch (getsubopt(&outopts
, UNCONST(toks
), &v
)) {
144 h
->base_includes
= v
;
154 html_alloc(char *outopts
)
157 return(ml_alloc(outopts
, HTML_HTML_4_01_STRICT
));
162 xhtml_alloc(char *outopts
)
165 return(ml_alloc(outopts
, HTML_XHTML_1_0_STRICT
));
175 h
= (struct html
*)p
;
177 while ((tag
= h
->tags
.head
) != NULL
) {
178 h
->tags
.head
= tag
->next
;
183 mchars_free(h
->symtab
);
190 print_gen_head(struct html
*h
)
192 struct htmlpair tag
[4];
194 tag
[0].key
= ATTR_HTTPEQUIV
;
195 tag
[0].val
= "Content-Type";
196 tag
[1].key
= ATTR_CONTENT
;
197 tag
[1].val
= "text/html; charset=utf-8";
198 print_otag(h
, TAG_META
, 2, tag
);
200 tag
[0].key
= ATTR_NAME
;
201 tag
[0].val
= "resource-type";
202 tag
[1].key
= ATTR_CONTENT
;
203 tag
[1].val
= "document";
204 print_otag(h
, TAG_META
, 2, tag
);
207 tag
[0].key
= ATTR_REL
;
208 tag
[0].val
= "stylesheet";
209 tag
[1].key
= ATTR_HREF
;
210 tag
[1].val
= h
->style
;
211 tag
[2].key
= ATTR_TYPE
;
212 tag
[2].val
= "text/css";
213 tag
[3].key
= ATTR_MEDIA
;
215 print_otag(h
, TAG_LINK
, 4, tag
);
220 print_metaf(struct html
*h
, enum mandoc_esc deco
)
225 case (ESCAPE_FONTPREV
):
228 case (ESCAPE_FONTITALIC
):
229 font
= HTMLFONT_ITALIC
;
231 case (ESCAPE_FONTBOLD
):
232 font
= HTMLFONT_BOLD
;
236 case (ESCAPE_FONTROMAN
):
237 font
= HTMLFONT_NONE
;
245 print_tagq(h
, h
->metaf
);
252 if (HTMLFONT_NONE
!= font
)
253 h
->metaf
= HTMLFONT_BOLD
== font
?
254 print_otag(h
, TAG_B
, 0, NULL
) :
255 print_otag(h
, TAG_I
, 0, NULL
);
259 html_strlen(const char *cp
)
265 * Account for escaped sequences within string length
266 * calculations. This follows the logic in term_strlen() as we
267 * must calculate the width of produced strings.
268 * Assume that characters are always width of "1". This is
269 * hacky, but it gets the job done for approximation of widths.
273 while (NULL
!= (p
= strchr(cp
, '\\'))) {
276 switch (mandoc_escape(&cp
, &seq
, &ssz
)) {
279 case (ESCAPE_UNICODE
):
281 case (ESCAPE_NUMBERED
):
283 case (ESCAPE_SPECIAL
):
292 return(sz
+ strlen(cp
));
296 print_encode(struct html
*h
, const char *p
, int norecurse
)
302 static const char rejs
[6] = { '\\', '<', '>', '&', ASCII_HYPH
, '\0' };
307 sz
= strcspn(p
, rejs
);
309 fwrite(p
, 1, sz
, stdout
);
332 esc
= mandoc_escape(&p
, &seq
, &len
);
333 if (ESCAPE_ERROR
== esc
)
337 case (ESCAPE_UNICODE
):
338 /* Skip passed "u" header. */
339 c
= mchars_num2uc(seq
+ 1, len
- 1);
343 case (ESCAPE_NUMBERED
):
344 c
= mchars_num2char(seq
, len
);
348 case (ESCAPE_SPECIAL
):
349 c
= mchars_spec2cp(h
->symtab
, seq
, len
);
352 else if (-1 == c
&& 1 == len
)
357 case (ESCAPE_FONTPREV
):
359 case (ESCAPE_FONTBOLD
):
361 case (ESCAPE_FONTITALIC
):
363 case (ESCAPE_FONTROMAN
):
368 case (ESCAPE_NOSPACE
):
382 print_attr(struct html
*h
, const char *key
, const char *val
)
384 printf(" %s=\"", key
);
385 (void)print_encode(h
, val
, 1);
391 print_otag(struct html
*h
, enum htmltag tag
,
392 int sz
, const struct htmlpair
*p
)
397 /* Push this tags onto the stack of open scopes. */
399 if ( ! (HTML_NOSTACK
& htmltags
[tag
].flags
)) {
400 t
= mandoc_malloc(sizeof(struct tag
));
402 t
->next
= h
->tags
.head
;
407 if ( ! (HTML_NOSPACE
& h
->flags
))
408 if ( ! (HTML_CLRLINE
& htmltags
[tag
].flags
)) {
410 if ( ! (HTML_KEEP
& h
->flags
)) {
411 if (HTML_PREKEEP
& h
->flags
)
412 h
->flags
|= HTML_KEEP
;
418 if ( ! (h
->flags
& HTML_NONOSPACE
))
419 h
->flags
&= ~HTML_NOSPACE
;
421 h
->flags
|= HTML_NOSPACE
;
423 /* Print out the tag name and attributes. */
425 printf("<%s", htmltags
[tag
].name
);
426 for (i
= 0; i
< sz
; i
++)
427 print_attr(h
, htmlattrs
[p
[i
].key
], p
[i
].val
);
429 /* Add non-overridable attributes. */
431 if (TAG_HTML
== tag
&& HTML_XHTML_1_0_STRICT
== h
->type
) {
432 print_attr(h
, "xmlns", "http://www.w3.org/1999/xhtml");
433 print_attr(h
, "xml:lang", "en");
434 print_attr(h
, "lang", "en");
437 /* Accommodate for XML "well-formed" singleton escaping. */
439 if (HTML_AUTOCLOSE
& htmltags
[tag
].flags
)
441 case (HTML_XHTML_1_0_STRICT
):
450 h
->flags
|= HTML_NOSPACE
;
452 if ((HTML_AUTOCLOSE
| HTML_CLRLINE
) & htmltags
[tag
].flags
)
460 print_ctag(struct html
*h
, enum htmltag tag
)
463 printf("</%s>", htmltags
[tag
].name
);
464 if (HTML_CLRLINE
& htmltags
[tag
].flags
) {
465 h
->flags
|= HTML_NOSPACE
;
471 print_gen_decls(struct html
*h
)
478 case (HTML_HTML_4_01_STRICT
):
480 doctype
= "-//W3C//DTD HTML 4.01//EN";
481 dtd
= "http://www.w3.org/TR/html4/strict.dtd";
484 puts("<?xml version=\"1.0\" encoding=\"UTF-8\"?>");
486 doctype
= "-//W3C//DTD XHTML 1.0 Strict//EN";
487 dtd
= "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd";
491 printf("<!DOCTYPE %s PUBLIC \"%s\" \"%s\">\n",
496 print_text(struct html
*h
, const char *word
)
499 if ( ! (HTML_NOSPACE
& h
->flags
)) {
501 if ( ! (HTML_KEEP
& h
->flags
)) {
502 if (HTML_PREKEEP
& h
->flags
)
503 h
->flags
|= HTML_KEEP
;
509 assert(NULL
== h
->metaf
);
510 if (HTMLFONT_NONE
!= h
->metac
)
511 h
->metaf
= HTMLFONT_BOLD
== h
->metac
?
512 print_otag(h
, TAG_B
, 0, NULL
) :
513 print_otag(h
, TAG_I
, 0, NULL
);
516 if ( ! print_encode(h
, word
, 0))
517 if ( ! (h
->flags
& HTML_NONOSPACE
))
518 h
->flags
&= ~HTML_NOSPACE
;
521 print_tagq(h
, h
->metaf
);
525 h
->flags
&= ~HTML_IGNDELIM
;
530 print_tagq(struct html
*h
, const struct tag
*until
)
534 while ((tag
= h
->tags
.head
) != NULL
) {
536 * Remember to close out and nullify the current
537 * meta-font and table, if applicable.
543 print_ctag(h
, tag
->tag
);
544 h
->tags
.head
= tag
->next
;
546 if (until
&& tag
== until
)
553 print_stagq(struct html
*h
, const struct tag
*suntil
)
557 while ((tag
= h
->tags
.head
) != NULL
) {
558 if (suntil
&& tag
== suntil
)
561 * Remember to close out and nullify the current
562 * meta-font and table, if applicable.
568 print_ctag(h
, tag
->tag
);
569 h
->tags
.head
= tag
->next
;
575 bufinit(struct html
*h
)
583 bufcat_style(struct html
*h
, const char *key
, const char *val
)
593 bufcat(struct html
*h
, const char *p
)
596 h
->buflen
= strlcat(h
->buf
, p
, BUFSIZ
);
597 assert(h
->buflen
< BUFSIZ
);
601 bufcat_fmt(struct html
*h
, const char *fmt
, ...)
606 (void)vsnprintf(h
->buf
+ (int)h
->buflen
,
607 BUFSIZ
- h
->buflen
- 1, fmt
, ap
);
609 h
->buflen
= strlen(h
->buf
);
613 bufncat(struct html
*h
, const char *p
, size_t sz
)
616 assert(h
->buflen
+ sz
+ 1 < BUFSIZ
);
617 strncat(h
->buf
, p
, sz
);
622 buffmt_includes(struct html
*h
, const char *name
)
626 pp
= h
->base_includes
;
629 while (NULL
!= (p
= strchr(pp
, '%'))) {
630 bufncat(h
, pp
, (size_t)(p
- pp
));
646 buffmt_man(struct html
*h
,
647 const char *name
, const char *sec
)
654 while (NULL
!= (p
= strchr(pp
, '%'))) {
655 bufncat(h
, pp
, (size_t)(p
- pp
));
658 bufcat(h
, sec
? sec
: "1");
674 bufcat_su(struct html
*h
, const char *p
, const struct roffsu
*su
)
679 if (SCALE_MM
== su
->unit
&& 0.0 == (v
/= 100.0))
682 bufcat_fmt(h
, "%s: %.2f%s;", p
, v
, roffscales
[su
->unit
]);
686 bufcat_id(struct html
*h
, const char *src
)
689 /* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
692 bufcat_fmt(h
, "%.2x", *src
++);