]>
git.cameronkatri.com Git - mandoc.git/blob - html.c
1 /* $Id: html.c,v 1.139 2011/05/15 15:47:46 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 void print_num(struct html
*, const char *, size_t);
98 static void print_spec(struct html
*, const char *, size_t);
99 static void print_res(struct html
*, const char *, size_t);
100 static void print_ctag(struct html
*, enum htmltag
);
101 static void print_doctype(struct html
*);
102 static void print_xmltype(struct html
*);
103 static int print_encode(struct html
*, const char *, int);
104 static void print_metaf(struct html
*, enum mandoc_esc
);
105 static void print_attr(struct html
*,
106 const char *, const char *);
107 static void *ml_alloc(char *, enum htmltype
);
111 ml_alloc(char *outopts
, enum htmltype type
)
119 toks
[2] = "includes";
122 h
= mandoc_calloc(1, sizeof(struct html
));
126 h
->symtab
= mchars_alloc();
128 while (outopts
&& *outopts
)
129 switch (getsubopt(&outopts
, UNCONST(toks
), &v
)) {
137 h
->base_includes
= v
;
147 html_alloc(char *outopts
)
150 return(ml_alloc(outopts
, HTML_HTML_4_01_STRICT
));
155 xhtml_alloc(char *outopts
)
158 return(ml_alloc(outopts
, HTML_XHTML_1_0_STRICT
));
168 h
= (struct html
*)p
;
170 while ((tag
= h
->tags
.head
) != NULL
) {
171 h
->tags
.head
= tag
->next
;
176 mchars_free(h
->symtab
);
183 print_gen_head(struct html
*h
)
185 struct htmlpair tag
[4];
187 tag
[0].key
= ATTR_HTTPEQUIV
;
188 tag
[0].val
= "Content-Type";
189 tag
[1].key
= ATTR_CONTENT
;
190 tag
[1].val
= "text/html; charset=utf-8";
191 print_otag(h
, TAG_META
, 2, tag
);
193 tag
[0].key
= ATTR_NAME
;
194 tag
[0].val
= "resource-type";
195 tag
[1].key
= ATTR_CONTENT
;
196 tag
[1].val
= "document";
197 print_otag(h
, TAG_META
, 2, tag
);
200 tag
[0].key
= ATTR_REL
;
201 tag
[0].val
= "stylesheet";
202 tag
[1].key
= ATTR_HREF
;
203 tag
[1].val
= h
->style
;
204 tag
[2].key
= ATTR_TYPE
;
205 tag
[2].val
= "text/css";
206 tag
[3].key
= ATTR_MEDIA
;
208 print_otag(h
, TAG_LINK
, 4, tag
);
214 print_num(struct html
*h
, const char *p
, size_t len
)
218 if ('\0' != (c
= mchars_num2char(p
, len
)))
223 print_spec(struct html
*h
, const char *p
, size_t len
)
229 if ((cp
= mchars_spec2cp(h
->symtab
, p
, len
)) > 0) {
232 } else if (-1 == cp
&& 1 == len
) {
233 fwrite(p
, 1, len
, stdout
);
238 if (NULL
!= (rhs
= mchars_spec2str(h
->symtab
, p
, len
, &sz
)))
239 fwrite(rhs
, 1, sz
, stdout
);
244 print_res(struct html
*h
, const char *p
, size_t len
)
250 if ((cp
= mchars_res2cp(h
->symtab
, p
, len
)) > 0) {
256 if (NULL
!= (rhs
= mchars_res2str(h
->symtab
, p
, len
, &sz
)))
257 fwrite(rhs
, 1, sz
, stdout
);
262 print_metaf(struct html
*h
, enum mandoc_esc deco
)
267 case (ESCAPE_FONTPREV
):
270 case (ESCAPE_FONTITALIC
):
271 font
= HTMLFONT_ITALIC
;
273 case (ESCAPE_FONTBOLD
):
274 font
= HTMLFONT_BOLD
;
276 case (ESCAPE_FONTROMAN
):
277 font
= HTMLFONT_NONE
;
285 print_tagq(h
, h
->metaf
);
292 if (HTMLFONT_NONE
!= font
)
293 h
->metaf
= HTMLFONT_BOLD
== font
?
294 print_otag(h
, TAG_B
, 0, NULL
) :
295 print_otag(h
, TAG_I
, 0, NULL
);
299 html_strlen(const char *cp
)
305 * Account for escaped sequences within string length
306 * calculations. This follows the logic in term_strlen() as we
307 * must calculate the width of produced strings.
308 * Assume that characters are always width of "1". This is
309 * hacky, but it gets the job done for approximation of widths.
313 while (NULL
!= (p
= strchr(cp
, '\\'))) {
316 switch (mandoc_escape(&cp
, &seq
, &ssz
)) {
319 case (ESCAPE_NUMBERED
):
321 case (ESCAPE_PREDEF
):
323 case (ESCAPE_SPECIAL
):
332 return(sz
+ strlen(cp
));
336 print_encode(struct html
*h
, const char *p
, int norecurse
)
342 static const char rejs
[6] = { '\\', '<', '>', '&', ASCII_HYPH
, '\0' };
347 sz
= strcspn(p
, rejs
);
349 fwrite(p
, 1, sz
, stdout
);
372 esc
= mandoc_escape(&p
, &seq
, &len
);
373 if (ESCAPE_ERROR
== esc
)
377 case (ESCAPE_NUMBERED
):
378 print_num(h
, seq
, len
);
380 case (ESCAPE_PREDEF
):
381 print_res(h
, seq
, len
);
383 case (ESCAPE_SPECIAL
):
384 print_spec(h
, seq
, len
);
386 case (ESCAPE_FONTPREV
):
388 case (ESCAPE_FONTBOLD
):
390 case (ESCAPE_FONTITALIC
):
392 case (ESCAPE_FONTROMAN
):
397 case (ESCAPE_NOSPACE
):
411 print_attr(struct html
*h
, const char *key
, const char *val
)
413 printf(" %s=\"", key
);
414 (void)print_encode(h
, val
, 1);
420 print_otag(struct html
*h
, enum htmltag tag
,
421 int sz
, const struct htmlpair
*p
)
426 /* Push this tags onto the stack of open scopes. */
428 if ( ! (HTML_NOSTACK
& htmltags
[tag
].flags
)) {
429 t
= mandoc_malloc(sizeof(struct tag
));
431 t
->next
= h
->tags
.head
;
436 if ( ! (HTML_NOSPACE
& h
->flags
))
437 if ( ! (HTML_CLRLINE
& htmltags
[tag
].flags
)) {
439 if ( ! (HTML_KEEP
& h
->flags
)) {
440 if (HTML_PREKEEP
& h
->flags
)
441 h
->flags
|= HTML_KEEP
;
447 if ( ! (h
->flags
& HTML_NONOSPACE
))
448 h
->flags
&= ~HTML_NOSPACE
;
450 h
->flags
|= HTML_NOSPACE
;
452 /* Print out the tag name and attributes. */
454 printf("<%s", htmltags
[tag
].name
);
455 for (i
= 0; i
< sz
; i
++)
456 print_attr(h
, htmlattrs
[p
[i
].key
], p
[i
].val
);
458 /* Add non-overridable attributes. */
460 if (TAG_HTML
== tag
&& HTML_XHTML_1_0_STRICT
== h
->type
) {
461 print_attr(h
, "xmlns", "http://www.w3.org/1999/xhtml");
462 print_attr(h
, "xml:lang", "en");
463 print_attr(h
, "lang", "en");
466 /* Accommodate for XML "well-formed" singleton escaping. */
468 if (HTML_AUTOCLOSE
& htmltags
[tag
].flags
)
470 case (HTML_XHTML_1_0_STRICT
):
479 h
->flags
|= HTML_NOSPACE
;
481 if ((HTML_AUTOCLOSE
| HTML_CLRLINE
) & htmltags
[tag
].flags
)
489 print_ctag(struct html
*h
, enum htmltag tag
)
492 printf("</%s>", htmltags
[tag
].name
);
493 if (HTML_CLRLINE
& htmltags
[tag
].flags
) {
494 h
->flags
|= HTML_NOSPACE
;
501 print_gen_decls(struct html
*h
)
510 print_xmltype(struct html
*h
)
513 if (HTML_XHTML_1_0_STRICT
== h
->type
)
514 puts("<?xml version=\"1.0\" encoding=\"UTF-8\"?>");
519 print_doctype(struct html
*h
)
526 case (HTML_HTML_4_01_STRICT
):
528 doctype
= "-//W3C//DTD HTML 4.01//EN";
529 dtd
= "http://www.w3.org/TR/html4/strict.dtd";
533 doctype
= "-//W3C//DTD XHTML 1.0 Strict//EN";
534 dtd
= "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd";
538 printf("<!DOCTYPE %s PUBLIC \"%s\" \"%s\">\n",
543 print_text(struct html
*h
, const char *word
)
546 if ( ! (HTML_NOSPACE
& h
->flags
)) {
548 if ( ! (HTML_KEEP
& h
->flags
)) {
549 if (HTML_PREKEEP
& h
->flags
)
550 h
->flags
|= HTML_KEEP
;
556 assert(NULL
== h
->metaf
);
557 if (HTMLFONT_NONE
!= h
->metac
)
558 h
->metaf
= HTMLFONT_BOLD
== h
->metac
?
559 print_otag(h
, TAG_B
, 0, NULL
) :
560 print_otag(h
, TAG_I
, 0, NULL
);
563 if ( ! print_encode(h
, word
, 0))
564 if ( ! (h
->flags
& HTML_NONOSPACE
))
565 h
->flags
&= ~HTML_NOSPACE
;
568 print_tagq(h
, h
->metaf
);
572 h
->flags
&= ~HTML_IGNDELIM
;
577 print_tagq(struct html
*h
, const struct tag
*until
)
581 while ((tag
= h
->tags
.head
) != NULL
) {
583 * Remember to close out and nullify the current
584 * meta-font and table, if applicable.
590 print_ctag(h
, tag
->tag
);
591 h
->tags
.head
= tag
->next
;
593 if (until
&& tag
== until
)
600 print_stagq(struct html
*h
, const struct tag
*suntil
)
604 while ((tag
= h
->tags
.head
) != NULL
) {
605 if (suntil
&& tag
== suntil
)
608 * Remember to close out and nullify the current
609 * meta-font and table, if applicable.
615 print_ctag(h
, tag
->tag
);
616 h
->tags
.head
= tag
->next
;
623 bufinit(struct html
*h
)
632 bufcat_style(struct html
*h
, const char *key
, const char *val
)
643 bufcat(struct html
*h
, const char *p
)
646 bufncat(h
, p
, strlen(p
));
651 buffmt(struct html
*h
, const char *fmt
, ...)
656 (void)vsnprintf(h
->buf
+ (int)h
->buflen
,
657 BUFSIZ
- h
->buflen
- 1, fmt
, ap
);
659 h
->buflen
= strlen(h
->buf
);
664 bufncat(struct html
*h
, const char *p
, size_t sz
)
667 if (h
->buflen
+ sz
> BUFSIZ
- 1)
668 sz
= BUFSIZ
- 1 - h
->buflen
;
670 (void)strncat(h
->buf
, p
, sz
);
676 buffmt_includes(struct html
*h
, const char *name
)
680 pp
= h
->base_includes
;
682 while (NULL
!= (p
= strchr(pp
, '%'))) {
683 bufncat(h
, pp
, (size_t)(p
- pp
));
700 buffmt_man(struct html
*h
,
701 const char *name
, const char *sec
)
708 while (NULL
!= (p
= strchr(pp
, '%'))) {
709 bufncat(h
, pp
, (size_t)(p
- pp
));
712 bufcat(h
, sec
? sec
: "1");
729 bufcat_su(struct html
*h
, const char *p
, const struct roffsu
*su
)
772 * XXX: the CSS spec isn't clear as to which types accept
773 * integer or real numbers, so we just make them all decimals.
775 buffmt(h
, "%s: %.2f%s;", p
, v
, u
);
780 html_idcat(char *dst
, const char *src
, int sz
)
786 /* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
788 /* We can't start with a number (bah). */
800 for ( ; *dst
!= '\0' && sz
; dst
++, sz
--)
803 for ( ; *src
!= '\0' && sz
> 1; src
++) {
804 ssz
= snprintf(dst
, (size_t)sz
, "%.2x", *src
);