version 1.98, 2010/04/08 07:05:38 |
version 1.238, 2018/06/25 16:54:59 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se> |
* Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv> |
|
* Copyright (c) 2011-2015, 2017, 2018 Ingo Schwarze <schwarze@openbsd.org> |
* |
* |
* Permission to use, copy, modify, and distribute this software for any |
* Permission to use, copy, modify, and distribute this software for any |
* purpose with or without fee is hereby granted, provided that the above |
* purpose with or without fee is hereby granted, provided that the above |
* copyright notice and this permission notice appear in all copies. |
* copyright notice and this permission notice appear in all copies. |
* |
* |
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES |
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHORS DISCLAIM ALL WARRANTIES |
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF |
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF |
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR |
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR |
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES |
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES |
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN |
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN |
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
*/ |
*/ |
#ifdef HAVE_CONFIG_H |
|
#include "config.h" |
#include "config.h" |
#endif |
|
|
|
#include <sys/types.h> |
#include <sys/types.h> |
|
|
#include <assert.h> |
#include <assert.h> |
#include <ctype.h> |
#include <ctype.h> |
#include <stdarg.h> |
#include <stdarg.h> |
|
#include <stddef.h> |
#include <stdio.h> |
#include <stdio.h> |
#include <stdint.h> |
#include <stdint.h> |
#include <stdlib.h> |
#include <stdlib.h> |
#include <string.h> |
#include <string.h> |
#include <unistd.h> |
#include <unistd.h> |
|
|
|
#include "mandoc_aux.h" |
|
#include "mandoc_ohash.h" |
|
#include "mandoc.h" |
|
#include "roff.h" |
#include "out.h" |
#include "out.h" |
#include "chars.h" |
|
#include "html.h" |
#include "html.h" |
|
#include "manconf.h" |
#include "main.h" |
#include "main.h" |
|
|
#define UNCONST(a) ((void *)(uintptr_t)(const void *)(a)) |
|
|
|
struct htmldata { |
struct htmldata { |
const char *name; |
const char *name; |
int flags; |
int flags; |
#define HTML_CLRLINE (1 << 0) |
#define HTML_NOSTACK (1 << 0) |
#define HTML_NOSTACK (1 << 1) |
#define HTML_AUTOCLOSE (1 << 1) |
#define HTML_AUTOCLOSE (1 << 2) /* Tag has auto-closure. */ |
#define HTML_NLBEFORE (1 << 2) |
|
#define HTML_NLBEGIN (1 << 3) |
|
#define HTML_NLEND (1 << 4) |
|
#define HTML_NLAFTER (1 << 5) |
|
#define HTML_NLAROUND (HTML_NLBEFORE | HTML_NLAFTER) |
|
#define HTML_NLINSIDE (HTML_NLBEGIN | HTML_NLEND) |
|
#define HTML_NLALL (HTML_NLAROUND | HTML_NLINSIDE) |
|
#define HTML_INDENT (1 << 6) |
|
#define HTML_NOINDENT (1 << 7) |
}; |
}; |
|
|
static const struct htmldata htmltags[TAG_MAX] = { |
static const struct htmldata htmltags[TAG_MAX] = { |
{"html", HTML_CLRLINE}, /* TAG_HTML */ |
{"html", HTML_NLALL}, |
{"head", HTML_CLRLINE}, /* TAG_HEAD */ |
{"head", HTML_NLALL | HTML_INDENT}, |
{"body", HTML_CLRLINE}, /* TAG_BODY */ |
{"body", HTML_NLALL}, |
{"meta", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_META */ |
{"meta", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, |
{"title", HTML_CLRLINE}, /* TAG_TITLE */ |
{"title", HTML_NLAROUND}, |
{"div", HTML_CLRLINE}, /* TAG_DIV */ |
{"div", HTML_NLAROUND}, |
{"h1", 0}, /* TAG_H1 */ |
{"div", 0}, |
{"h2", 0}, /* TAG_H2 */ |
{"h1", HTML_NLAROUND}, |
{"span", 0}, /* TAG_SPAN */ |
{"h2", HTML_NLAROUND}, |
{"link", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_LINK */ |
{"span", 0}, |
{"br", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_BR */ |
{"link", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, |
{"a", 0}, /* TAG_A */ |
{"br", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, |
{"table", HTML_CLRLINE}, /* TAG_TABLE */ |
{"a", 0}, |
{"col", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_COL */ |
{"table", HTML_NLALL | HTML_INDENT}, |
{"tr", HTML_CLRLINE}, /* TAG_TR */ |
{"tr", HTML_NLALL | HTML_INDENT}, |
{"td", HTML_CLRLINE}, /* TAG_TD */ |
{"td", HTML_NLAROUND}, |
{"li", HTML_CLRLINE}, /* TAG_LI */ |
{"li", HTML_NLAROUND | HTML_INDENT}, |
{"ul", HTML_CLRLINE}, /* TAG_UL */ |
{"ul", HTML_NLALL | HTML_INDENT}, |
{"ol", HTML_CLRLINE}, /* TAG_OL */ |
{"ol", HTML_NLALL | HTML_INDENT}, |
|
{"dl", HTML_NLALL | HTML_INDENT}, |
|
{"dt", HTML_NLAROUND}, |
|
{"dd", HTML_NLAROUND | HTML_INDENT}, |
|
{"pre", HTML_NLALL | HTML_NOINDENT}, |
|
{"var", 0}, |
|
{"cite", 0}, |
|
{"b", 0}, |
|
{"i", 0}, |
|
{"code", 0}, |
|
{"small", 0}, |
|
{"style", HTML_NLALL | HTML_INDENT}, |
|
{"math", HTML_NLALL | HTML_INDENT}, |
|
{"mrow", 0}, |
|
{"mi", 0}, |
|
{"mn", 0}, |
|
{"mo", 0}, |
|
{"msup", 0}, |
|
{"msub", 0}, |
|
{"msubsup", 0}, |
|
{"mfrac", 0}, |
|
{"msqrt", 0}, |
|
{"mfenced", 0}, |
|
{"mtable", 0}, |
|
{"mtr", 0}, |
|
{"mtd", 0}, |
|
{"munderover", 0}, |
|
{"munder", 0}, |
|
{"mover", 0}, |
}; |
}; |
|
|
static const char *const htmlfonts[HTMLFONT_MAX] = { |
/* Avoid duplicate HTML id= attributes. */ |
"roman", |
static struct ohash id_unique; |
"bold", |
|
"italic" |
|
}; |
|
|
|
static const char *const htmlattrs[ATTR_MAX] = { |
static void print_byte(struct html *, char); |
"http-equiv", |
static void print_endword(struct html *); |
"content", |
static void print_indent(struct html *); |
"name", |
static void print_word(struct html *, const char *); |
"rel", |
|
"href", |
|
"type", |
|
"media", |
|
"class", |
|
"style", |
|
"width", |
|
"valign", |
|
"target", |
|
"id", |
|
"summary", |
|
}; |
|
|
|
static void print_spec(struct html *, const char *, size_t); |
static void print_ctag(struct html *, struct tag *); |
static void print_res(struct html *, const char *, size_t); |
static int print_escape(struct html *, char); |
static void print_ctag(struct html *, enum htmltag); |
static int print_encode(struct html *, const char *, const char *, int); |
static void print_doctype(struct html *); |
static void print_href(struct html *, const char *, const char *, int); |
static void print_xmltype(struct html *); |
static void print_metaf(struct html *, enum mandoc_esc); |
static int print_encode(struct html *, const char *, int); |
|
static void print_metaf(struct html *, enum roffdeco); |
|
static void print_attr(struct html *, |
|
const char *, const char *); |
|
static void *ml_alloc(char *, enum htmltype); |
|
|
|
|
|
static void * |
void * |
ml_alloc(char *outopts, enum htmltype type) |
html_alloc(const struct manoutput *outopts) |
{ |
{ |
struct html *h; |
struct html *h; |
const char *toks[4]; |
|
char *v; |
|
|
|
toks[0] = "style"; |
h = mandoc_calloc(1, sizeof(struct html)); |
toks[1] = "man"; |
|
toks[2] = "includes"; |
|
toks[3] = NULL; |
|
|
|
h = calloc(1, sizeof(struct html)); |
h->tag = NULL; |
if (NULL == h) { |
h->style = outopts->style; |
perror(NULL); |
h->base_man = outopts->man; |
exit(EXIT_FAILURE); |
h->base_includes = outopts->includes; |
} |
if (outopts->fragment) |
|
h->oflags |= HTML_FRAGMENT; |
|
|
h->type = type; |
mandoc_ohash_init(&id_unique, 4, 0); |
h->tags.head = NULL; |
|
h->ords.head = NULL; |
|
h->symtab = chars_init(CHARS_HTML); |
|
|
|
while (outopts && *outopts) |
return h; |
switch (getsubopt(&outopts, UNCONST(toks), &v)) { |
|
case (0): |
|
h->style = v; |
|
break; |
|
case (1): |
|
h->base_man = v; |
|
break; |
|
case (2): |
|
h->base_includes = v; |
|
break; |
|
default: |
|
break; |
|
} |
|
|
|
return(h); |
|
} |
} |
|
|
void * |
|
html_alloc(char *outopts) |
|
{ |
|
|
|
return(ml_alloc(outopts, HTML_HTML_4_01_STRICT)); |
|
} |
|
|
|
|
|
void * |
|
xhtml_alloc(char *outopts) |
|
{ |
|
|
|
return(ml_alloc(outopts, HTML_XHTML_1_0_STRICT)); |
|
} |
|
|
|
|
|
void |
void |
html_free(void *p) |
html_free(void *p) |
{ |
{ |
struct tag *tag; |
struct tag *tag; |
struct ord *ord; |
|
struct html *h; |
struct html *h; |
|
char *cp; |
|
unsigned int slot; |
|
|
h = (struct html *)p; |
h = (struct html *)p; |
|
while ((tag = h->tag) != NULL) { |
while ((ord = h->ords.head) != NULL) { |
h->tag = tag->next; |
h->ords.head = ord->next; |
|
free(ord); |
|
} |
|
|
|
while ((tag = h->tags.head) != NULL) { |
|
h->tags.head = tag->next; |
|
free(tag); |
free(tag); |
} |
} |
|
|
if (h->symtab) |
|
chars_free(h->symtab); |
|
|
|
free(h); |
free(h); |
|
|
|
cp = ohash_first(&id_unique, &slot); |
|
while (cp != NULL) { |
|
free(cp); |
|
cp = ohash_next(&id_unique, &slot); |
|
} |
|
ohash_delete(&id_unique); |
} |
} |
|
|
|
|
void |
void |
print_gen_head(struct html *h) |
print_gen_head(struct html *h) |
{ |
{ |
struct htmlpair tag[4]; |
struct tag *t; |
|
|
tag[0].key = ATTR_HTTPEQUIV; |
print_otag(h, TAG_META, "?", "charset", "utf-8"); |
tag[0].val = "Content-Type"; |
if (h->style != NULL) { |
tag[1].key = ATTR_CONTENT; |
print_otag(h, TAG_LINK, "?h??", "rel", "stylesheet", |
tag[1].val = "text/html; charset=utf-8"; |
h->style, "type", "text/css", "media", "all"); |
print_otag(h, TAG_META, 2, tag); |
return; |
|
} |
|
|
tag[0].key = ATTR_NAME; |
/* |
tag[0].val = "resource-type"; |
* Print a minimal embedded style sheet. |
tag[1].key = ATTR_CONTENT; |
*/ |
tag[1].val = "document"; |
|
print_otag(h, TAG_META, 2, tag); |
|
|
|
if (h->style) { |
t = print_otag(h, TAG_STYLE, ""); |
tag[0].key = ATTR_REL; |
print_text(h, "table.head, table.foot { width: 100%; }"); |
tag[0].val = "stylesheet"; |
print_endline(h); |
tag[1].key = ATTR_HREF; |
print_text(h, "td.head-rtitle, td.foot-os { text-align: right; }"); |
tag[1].val = h->style; |
print_endline(h); |
tag[2].key = ATTR_TYPE; |
print_text(h, "td.head-vol { text-align: center; }"); |
tag[2].val = "text/css"; |
print_endline(h); |
tag[3].key = ATTR_MEDIA; |
print_text(h, "div.Pp { margin: 1ex 0ex; }"); |
tag[3].val = "all"; |
print_endline(h); |
print_otag(h, TAG_LINK, 4, tag); |
print_text(h, "div.Nd, div.Bf, div.Op { display: inline; }"); |
} |
print_endline(h); |
|
print_text(h, "span.Pa, span.Ad { font-style: italic; }"); |
|
print_endline(h); |
|
print_text(h, "span.Ms { font-weight: bold; }"); |
|
print_endline(h); |
|
print_text(h, "dl.Bl-diag "); |
|
print_byte(h, '>'); |
|
print_text(h, " dt { font-weight: bold; }"); |
|
print_endline(h); |
|
print_text(h, "code.Nm, code.Fl, code.Cm, code.Ic, " |
|
"code.In, code.Fd, code.Fn,"); |
|
print_endline(h); |
|
print_text(h, "code.Cd { font-weight: bold; " |
|
"font-family: inherit; }"); |
|
print_tagq(h, t); |
} |
} |
|
|
|
|
static void |
static void |
print_spec(struct html *h, const char *p, size_t len) |
print_metaf(struct html *h, enum mandoc_esc deco) |
{ |
{ |
const char *rhs; |
enum htmlfont font; |
size_t sz; |
|
|
|
rhs = chars_a2ascii(h->symtab, p, len, &sz); |
switch (deco) { |
|
case ESCAPE_FONTPREV: |
|
font = h->metal; |
|
break; |
|
case ESCAPE_FONTITALIC: |
|
font = HTMLFONT_ITALIC; |
|
break; |
|
case ESCAPE_FONTBOLD: |
|
font = HTMLFONT_BOLD; |
|
break; |
|
case ESCAPE_FONTBI: |
|
font = HTMLFONT_BI; |
|
break; |
|
case ESCAPE_FONT: |
|
case ESCAPE_FONTROMAN: |
|
font = HTMLFONT_NONE; |
|
break; |
|
default: |
|
abort(); |
|
} |
|
|
if (NULL == rhs) |
if (h->metaf) { |
return; |
print_tagq(h, h->metaf); |
fwrite(rhs, 1, sz, stdout); |
h->metaf = NULL; |
} |
} |
|
|
|
h->metal = h->metac; |
|
h->metac = font; |
|
|
static void |
switch (font) { |
print_res(struct html *h, const char *p, size_t len) |
case HTMLFONT_ITALIC: |
|
h->metaf = print_otag(h, TAG_I, ""); |
|
break; |
|
case HTMLFONT_BOLD: |
|
h->metaf = print_otag(h, TAG_B, ""); |
|
break; |
|
case HTMLFONT_BI: |
|
h->metaf = print_otag(h, TAG_B, ""); |
|
print_otag(h, TAG_I, ""); |
|
break; |
|
default: |
|
break; |
|
} |
|
} |
|
|
|
char * |
|
html_make_id(const struct roff_node *n, int unique) |
{ |
{ |
const char *rhs; |
const struct roff_node *nch; |
size_t sz; |
char *buf, *bufs, *cp; |
|
unsigned int slot; |
|
int suffix; |
|
|
rhs = chars_a2res(h->symtab, p, len, &sz); |
for (nch = n->child; nch != NULL; nch = nch->next) |
|
if (nch->type != ROFFT_TEXT) |
|
return NULL; |
|
|
if (NULL == rhs) |
buf = NULL; |
return; |
deroff(&buf, n); |
fwrite(rhs, 1, sz, stdout); |
if (buf == NULL) |
} |
return NULL; |
|
|
|
/* |
|
* In ID attributes, only use ASCII characters that are |
|
* permitted in URL-fragment strings according to the |
|
* explicit list at: |
|
* https://url.spec.whatwg.org/#url-fragment-string |
|
*/ |
|
|
struct tag * |
for (cp = buf; *cp != '\0'; cp++) |
print_ofont(struct html *h, enum htmlfont font) |
if (isalnum((unsigned char)*cp) == 0 && |
{ |
strchr("!$&'()*+,-./:;=?@_~", *cp) == NULL) |
struct htmlpair tag; |
*cp = '_'; |
|
|
h->metal = h->metac; |
if (unique == 0) |
h->metac = font; |
return buf; |
|
|
/* FIXME: DECO_ROMAN should just close out preexisting. */ |
/* Avoid duplicate HTML id= attributes. */ |
|
|
if (h->metaf && h->tags.head == h->metaf) |
bufs = NULL; |
print_tagq(h, h->metaf); |
suffix = 1; |
|
slot = ohash_qlookup(&id_unique, buf); |
PAIR_CLASS_INIT(&tag, htmlfonts[font]); |
cp = ohash_find(&id_unique, slot); |
h->metaf = print_otag(h, TAG_SPAN, 1, &tag); |
if (cp != NULL) { |
return(h->metaf); |
while (cp != NULL) { |
|
free(bufs); |
|
if (++suffix > 127) { |
|
free(buf); |
|
return NULL; |
|
} |
|
mandoc_asprintf(&bufs, "%s_%d", buf, suffix); |
|
slot = ohash_qlookup(&id_unique, bufs); |
|
cp = ohash_find(&id_unique, slot); |
|
} |
|
free(buf); |
|
buf = bufs; |
|
} |
|
ohash_insert(&id_unique, slot, buf); |
|
return buf; |
} |
} |
|
|
|
static int |
static void |
print_escape(struct html *h, char c) |
print_metaf(struct html *h, enum roffdeco deco) |
|
{ |
{ |
enum htmlfont font; |
|
|
|
switch (deco) { |
switch (c) { |
case (DECO_PREVIOUS): |
case '<': |
font = h->metal; |
print_word(h, "<"); |
break; |
break; |
case (DECO_ITALIC): |
case '>': |
font = HTMLFONT_ITALIC; |
print_word(h, ">"); |
break; |
break; |
case (DECO_BOLD): |
case '&': |
font = HTMLFONT_BOLD; |
print_word(h, "&"); |
break; |
break; |
case (DECO_ROMAN): |
case '"': |
font = HTMLFONT_NONE; |
print_word(h, """); |
break; |
break; |
|
case ASCII_NBRSP: |
|
print_word(h, " "); |
|
break; |
|
case ASCII_HYPH: |
|
print_byte(h, '-'); |
|
break; |
|
case ASCII_BREAK: |
|
break; |
default: |
default: |
abort(); |
return 0; |
/* NOTREACHED */ |
|
} |
} |
|
return 1; |
(void)print_ofont(h, font); |
|
} |
} |
|
|
|
|
static int |
static int |
print_encode(struct html *h, const char *p, int norecurse) |
print_encode(struct html *h, const char *p, const char *pend, int norecurse) |
{ |
{ |
size_t sz; |
char numbuf[16]; |
int len, nospace; |
struct tag *t; |
const char *seq; |
const char *seq; |
enum roffdeco deco; |
size_t sz; |
|
int c, len, breakline, nospace; |
|
enum mandoc_esc esc; |
|
static const char rejs[10] = { ' ', '\\', '<', '>', '&', '"', |
|
ASCII_NBRSP, ASCII_HYPH, ASCII_BREAK, '\0' }; |
|
|
|
if (pend == NULL) |
|
pend = strchr(p, '\0'); |
|
|
|
breakline = 0; |
nospace = 0; |
nospace = 0; |
|
|
for (; *p; p++) { |
while (p < pend) { |
sz = strcspn(p, "\\<>&"); |
if (HTML_SKIPCHAR & h->flags && '\\' != *p) { |
|
h->flags &= ~HTML_SKIPCHAR; |
|
p++; |
|
continue; |
|
} |
|
|
fwrite(p, 1, sz, stdout); |
for (sz = strcspn(p, rejs); sz-- && p < pend; p++) |
p += /* LINTED */ |
print_byte(h, *p); |
sz; |
|
|
|
if ('<' == *p) { |
if (breakline && |
printf("<"); |
(p >= pend || *p == ' ' || *p == ASCII_NBRSP)) { |
|
t = print_otag(h, TAG_DIV, ""); |
|
print_text(h, "\\~"); |
|
print_tagq(h, t); |
|
breakline = 0; |
|
while (p < pend && (*p == ' ' || *p == ASCII_NBRSP)) |
|
p++; |
continue; |
continue; |
} else if ('>' == *p) { |
} |
printf(">"); |
|
|
if (p >= pend) |
|
break; |
|
|
|
if (*p == ' ') { |
|
print_endword(h); |
|
p++; |
continue; |
continue; |
} else if ('&' == *p) { |
} |
printf("&"); |
|
|
if (print_escape(h, *p++)) |
continue; |
continue; |
} else if ('\0' == *p) |
|
|
esc = mandoc_escape(&p, &seq, &len); |
|
if (ESCAPE_ERROR == esc) |
break; |
break; |
|
|
seq = ++p; |
switch (esc) { |
len = a2roffdeco(&deco, &seq, &sz); |
case ESCAPE_FONT: |
|
case ESCAPE_FONTPREV: |
|
case ESCAPE_FONTBOLD: |
|
case ESCAPE_FONTITALIC: |
|
case ESCAPE_FONTBI: |
|
case ESCAPE_FONTROMAN: |
|
if (0 == norecurse) |
|
print_metaf(h, esc); |
|
continue; |
|
case ESCAPE_SKIPCHAR: |
|
h->flags |= HTML_SKIPCHAR; |
|
continue; |
|
default: |
|
break; |
|
} |
|
|
switch (deco) { |
if (h->flags & HTML_SKIPCHAR) { |
case (DECO_RESERVED): |
h->flags &= ~HTML_SKIPCHAR; |
print_res(h, seq, sz); |
continue; |
|
} |
|
|
|
switch (esc) { |
|
case ESCAPE_UNICODE: |
|
/* Skip past "u" header. */ |
|
c = mchars_num2uc(seq + 1, len - 1); |
break; |
break; |
case (DECO_SPECIAL): |
case ESCAPE_NUMBERED: |
print_spec(h, seq, sz); |
c = mchars_num2char(seq, len); |
|
if (c < 0) |
|
continue; |
break; |
break; |
case (DECO_PREVIOUS): |
case ESCAPE_SPECIAL: |
/* FALLTHROUGH */ |
c = mchars_spec2cp(seq, len); |
case (DECO_BOLD): |
if (c <= 0) |
/* FALLTHROUGH */ |
continue; |
case (DECO_ITALIC): |
|
/* FALLTHROUGH */ |
|
case (DECO_ROMAN): |
|
if (norecurse) |
|
break; |
|
print_metaf(h, deco); |
|
break; |
break; |
default: |
case ESCAPE_BREAK: |
|
breakline = 1; |
|
continue; |
|
case ESCAPE_NOSPACE: |
|
if ('\0' == *p) |
|
nospace = 1; |
|
continue; |
|
case ESCAPE_OVERSTRIKE: |
|
if (len == 0) |
|
continue; |
|
c = seq[len - 1]; |
break; |
break; |
|
default: |
|
continue; |
} |
} |
|
if ((c < 0x20 && c != 0x09) || |
p += len - 1; |
(c > 0x7E && c < 0xA0)) |
|
c = 0xFFFD; |
if (DECO_NOSPACE == deco && '\0' == *(p + 1)) |
if (c > 0x7E) { |
nospace = 1; |
(void)snprintf(numbuf, sizeof(numbuf), "&#x%.4X;", c); |
|
print_word(h, numbuf); |
|
} else if (print_escape(h, c) == 0) |
|
print_byte(h, c); |
} |
} |
|
|
return(nospace); |
return nospace; |
} |
} |
|
|
|
|
static void |
static void |
print_attr(struct html *h, const char *key, const char *val) |
print_href(struct html *h, const char *name, const char *sec, int man) |
{ |
{ |
printf(" %s=\"", key); |
const char *p, *pp; |
(void)print_encode(h, val, 1); |
|
putchar('\"'); |
pp = man ? h->base_man : h->base_includes; |
|
while ((p = strchr(pp, '%')) != NULL) { |
|
print_encode(h, pp, p, 1); |
|
if (man && p[1] == 'S') { |
|
if (sec == NULL) |
|
print_byte(h, '1'); |
|
else |
|
print_encode(h, sec, NULL, 1); |
|
} else if ((man && p[1] == 'N') || |
|
(man == 0 && p[1] == 'I')) |
|
print_encode(h, name, NULL, 1); |
|
else |
|
print_encode(h, p, p + 2, 1); |
|
pp = p + 2; |
|
} |
|
if (*pp != '\0') |
|
print_encode(h, pp, NULL, 1); |
} |
} |
|
|
|
|
struct tag * |
struct tag * |
print_otag(struct html *h, enum htmltag tag, |
print_otag(struct html *h, enum htmltag tag, const char *fmt, ...) |
int sz, const struct htmlpair *p) |
|
{ |
{ |
int i; |
va_list ap; |
struct tag *t; |
struct tag *t; |
|
const char *attr; |
|
char *arg1, *arg2; |
|
int tflags; |
|
|
/* Push this tags onto the stack of open scopes. */ |
tflags = htmltags[tag].flags; |
|
|
if ( ! (HTML_NOSTACK & htmltags[tag].flags)) { |
/* Push this tag onto the stack of open scopes. */ |
t = malloc(sizeof(struct tag)); |
|
if (NULL == t) { |
if ((tflags & HTML_NOSTACK) == 0) { |
perror(NULL); |
t = mandoc_malloc(sizeof(struct tag)); |
exit(EXIT_FAILURE); |
|
} |
|
t->tag = tag; |
t->tag = tag; |
t->next = h->tags.head; |
t->next = h->tag; |
h->tags.head = t; |
h->tag = t; |
} else |
} else |
t = NULL; |
t = NULL; |
|
|
if ( ! (HTML_NOSPACE & h->flags)) |
if (tflags & HTML_NLBEFORE) |
if ( ! (HTML_CLRLINE & htmltags[tag].flags)) |
print_endline(h); |
putchar(' '); |
if (h->col == 0) |
|
print_indent(h); |
|
else if ((h->flags & HTML_NOSPACE) == 0) { |
|
if (h->flags & HTML_KEEP) |
|
print_word(h, " "); |
|
else { |
|
if (h->flags & HTML_PREKEEP) |
|
h->flags |= HTML_KEEP; |
|
print_endword(h); |
|
} |
|
} |
|
|
|
if ( ! (h->flags & HTML_NONOSPACE)) |
|
h->flags &= ~HTML_NOSPACE; |
|
else |
|
h->flags |= HTML_NOSPACE; |
|
|
/* Print out the tag name and attributes. */ |
/* Print out the tag name and attributes. */ |
|
|
printf("<%s", htmltags[tag].name); |
print_byte(h, '<'); |
for (i = 0; i < sz; i++) |
print_word(h, htmltags[tag].name); |
print_attr(h, htmlattrs[p[i].key], p[i].val); |
|
|
|
/* Add non-overridable attributes. */ |
va_start(ap, fmt); |
|
|
if (TAG_HTML == tag && HTML_XHTML_1_0_STRICT == h->type) { |
while (*fmt != '\0') { |
print_attr(h, "xmlns", "http://www.w3.org/1999/xhtml"); |
|
print_attr(h, "xml:lang", "en"); |
|
print_attr(h, "lang", "en"); |
|
} |
|
|
|
/* Accomodate for XML "well-formed" singleton escaping. */ |
/* Parse attributes and arguments. */ |
|
|
if (HTML_AUTOCLOSE & htmltags[tag].flags) |
arg1 = va_arg(ap, char *); |
switch (h->type) { |
arg2 = NULL; |
case (HTML_XHTML_1_0_STRICT): |
switch (*fmt++) { |
putchar('/'); |
case 'c': |
|
attr = "class"; |
break; |
break; |
|
case 'h': |
|
attr = "href"; |
|
break; |
|
case 'i': |
|
attr = "id"; |
|
break; |
|
case 's': |
|
attr = "style"; |
|
arg2 = va_arg(ap, char *); |
|
break; |
|
case '?': |
|
attr = arg1; |
|
arg1 = va_arg(ap, char *); |
|
break; |
default: |
default: |
|
abort(); |
|
} |
|
if (*fmt == 'M') |
|
arg2 = va_arg(ap, char *); |
|
if (arg1 == NULL) |
|
continue; |
|
|
|
/* Print the attributes. */ |
|
|
|
print_byte(h, ' '); |
|
print_word(h, attr); |
|
print_byte(h, '='); |
|
print_byte(h, '"'); |
|
switch (*fmt) { |
|
case 'I': |
|
print_href(h, arg1, NULL, 0); |
|
fmt++; |
break; |
break; |
|
case 'M': |
|
print_href(h, arg1, arg2, 1); |
|
fmt++; |
|
break; |
|
case 'R': |
|
print_byte(h, '#'); |
|
print_encode(h, arg1, NULL, 1); |
|
fmt++; |
|
break; |
|
case 'T': |
|
print_encode(h, arg1, NULL, 1); |
|
print_word(h, "\" title=\""); |
|
print_encode(h, arg1, NULL, 1); |
|
fmt++; |
|
break; |
|
default: |
|
if (arg2 == NULL) |
|
print_encode(h, arg1, NULL, 1); |
|
else { |
|
print_word(h, arg1); |
|
print_byte(h, ':'); |
|
print_byte(h, ' '); |
|
print_word(h, arg2); |
|
print_byte(h, ';'); |
|
} |
|
break; |
} |
} |
|
print_byte(h, '"'); |
|
} |
|
va_end(ap); |
|
|
putchar('>'); |
/* Accommodate for "well-formed" singleton escaping. */ |
|
|
h->flags |= HTML_NOSPACE; |
if (HTML_AUTOCLOSE & htmltags[tag].flags) |
return(t); |
print_byte(h, '/'); |
} |
|
|
|
|
print_byte(h, '>'); |
|
|
static void |
if (tflags & HTML_NLBEGIN) |
print_ctag(struct html *h, enum htmltag tag) |
print_endline(h); |
{ |
else |
|
|
printf("</%s>", htmltags[tag].name); |
|
if (HTML_CLRLINE & htmltags[tag].flags) { |
|
h->flags |= HTML_NOSPACE; |
h->flags |= HTML_NOSPACE; |
putchar('\n'); |
|
} |
|
} |
|
|
|
|
if (tflags & HTML_INDENT) |
|
h->indent++; |
|
if (tflags & HTML_NOINDENT) |
|
h->noindent++; |
|
|
void |
return t; |
print_gen_decls(struct html *h) |
|
{ |
|
|
|
print_xmltype(h); |
|
print_doctype(h); |
|
} |
} |
|
|
|
|
static void |
static void |
print_xmltype(struct html *h) |
print_ctag(struct html *h, struct tag *tag) |
{ |
{ |
const char *decl; |
int tflags; |
|
|
switch (h->type) { |
/* |
case (HTML_XHTML_1_0_STRICT): |
* Remember to close out and nullify the current |
decl = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>"; |
* meta-font and table, if applicable. |
break; |
*/ |
default: |
if (tag == h->metaf) |
decl = NULL; |
h->metaf = NULL; |
break; |
if (tag == h->tblt) |
} |
h->tblt = NULL; |
|
|
if (NULL == decl) |
tflags = htmltags[tag->tag].flags; |
return; |
|
|
|
printf("%s\n", decl); |
if (tflags & HTML_INDENT) |
|
h->indent--; |
|
if (tflags & HTML_NOINDENT) |
|
h->noindent--; |
|
if (tflags & HTML_NLEND) |
|
print_endline(h); |
|
print_indent(h); |
|
print_byte(h, '<'); |
|
print_byte(h, '/'); |
|
print_word(h, htmltags[tag->tag].name); |
|
print_byte(h, '>'); |
|
if (tflags & HTML_NLAFTER) |
|
print_endline(h); |
|
|
|
h->tag = tag->next; |
|
free(tag); |
} |
} |
|
|
|
void |
|
print_gen_decls(struct html *h) |
|
{ |
|
print_word(h, "<!DOCTYPE html>"); |
|
print_endline(h); |
|
} |
|
|
static void |
void |
print_doctype(struct html *h) |
print_gen_comment(struct html *h, struct roff_node *n) |
{ |
{ |
const char *doctype; |
int wantblank; |
const char *dtd; |
|
const char *name; |
|
|
|
switch (h->type) { |
print_word(h, "<!-- This is an automatically generated file." |
case (HTML_HTML_4_01_STRICT): |
" Do not edit."); |
name = "HTML"; |
h->indent = 1; |
doctype = "-//W3C//DTD HTML 4.01//EN"; |
wantblank = 0; |
dtd = "http://www.w3.org/TR/html4/strict.dtd"; |
while (n != NULL && n->type == ROFFT_COMMENT) { |
break; |
if (strstr(n->string, "-->") == NULL && |
default: |
(wantblank || *n->string != '\0')) { |
name = "html"; |
print_endline(h); |
doctype = "-//W3C//DTD XHTML 1.0 Strict//EN"; |
print_indent(h); |
dtd = "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"; |
print_word(h, n->string); |
break; |
wantblank = *n->string != '\0'; |
|
} |
|
n = n->next; |
} |
} |
|
if (wantblank) |
printf("<!DOCTYPE %s PUBLIC \"%s\" \"%s\">\n", |
print_endline(h); |
name, doctype, dtd); |
print_word(h, " -->"); |
|
print_endline(h); |
|
h->indent = 0; |
} |
} |
|
|
|
|
void |
void |
print_text(struct html *h, const char *p) |
print_text(struct html *h, const char *word) |
{ |
{ |
|
if (h->col && (h->flags & HTML_NOSPACE) == 0) { |
|
if ( ! (HTML_KEEP & h->flags)) { |
|
if (HTML_PREKEEP & h->flags) |
|
h->flags |= HTML_KEEP; |
|
print_endword(h); |
|
} else |
|
print_word(h, " "); |
|
} |
|
|
if (*p && 0 == *(p + 1)) |
assert(NULL == h->metaf); |
switch (*p) { |
switch (h->metac) { |
case('.'): |
case HTMLFONT_ITALIC: |
/* FALLTHROUGH */ |
h->metaf = print_otag(h, TAG_I, ""); |
case(','): |
break; |
/* FALLTHROUGH */ |
case HTMLFONT_BOLD: |
case(';'): |
h->metaf = print_otag(h, TAG_B, ""); |
/* FALLTHROUGH */ |
break; |
case(':'): |
case HTMLFONT_BI: |
/* FALLTHROUGH */ |
h->metaf = print_otag(h, TAG_B, ""); |
case('?'): |
print_otag(h, TAG_I, ""); |
/* FALLTHROUGH */ |
break; |
case('!'): |
default: |
/* FALLTHROUGH */ |
print_indent(h); |
case(')'): |
break; |
/* FALLTHROUGH */ |
} |
case(']'): |
|
if ( ! (HTML_IGNDELIM & h->flags)) |
|
h->flags |= HTML_NOSPACE; |
|
break; |
|
default: |
|
break; |
|
} |
|
|
|
if ( ! (h->flags & HTML_NOSPACE)) |
assert(word); |
putchar(' '); |
if ( ! print_encode(h, word, NULL, 0)) { |
|
if ( ! (h->flags & HTML_NONOSPACE)) |
|
h->flags &= ~HTML_NOSPACE; |
|
h->flags &= ~HTML_NONEWLINE; |
|
} else |
|
h->flags |= HTML_NOSPACE | HTML_NONEWLINE; |
|
|
assert(p); |
if (h->metaf) { |
if ( ! print_encode(h, p, 0)) |
print_tagq(h, h->metaf); |
h->flags &= ~HTML_NOSPACE; |
h->metaf = NULL; |
|
} |
|
|
/* |
h->flags &= ~HTML_IGNDELIM; |
* Note that we don't process the pipe: the parser sees it as |
|
* punctuation, but we don't in terms of typography. |
|
*/ |
|
if (*p && 0 == *(p + 1)) |
|
switch (*p) { |
|
case('('): |
|
/* FALLTHROUGH */ |
|
case('['): |
|
h->flags |= HTML_NOSPACE; |
|
break; |
|
default: |
|
break; |
|
} |
|
} |
} |
|
|
|
|
void |
void |
print_tagq(struct html *h, const struct tag *until) |
print_tagq(struct html *h, const struct tag *until) |
{ |
{ |
struct tag *tag; |
struct tag *tag; |
|
|
while ((tag = h->tags.head) != NULL) { |
while ((tag = h->tag) != NULL) { |
if (tag == h->metaf) |
print_ctag(h, tag); |
h->metaf = NULL; |
|
print_ctag(h, tag->tag); |
|
h->tags.head = tag->next; |
|
free(tag); |
|
if (until && tag == until) |
if (until && tag == until) |
return; |
return; |
} |
} |
} |
} |
|
|
|
|
void |
void |
print_stagq(struct html *h, const struct tag *suntil) |
print_stagq(struct html *h, const struct tag *suntil) |
{ |
{ |
struct tag *tag; |
struct tag *tag; |
|
|
while ((tag = h->tags.head) != NULL) { |
while ((tag = h->tag) != NULL) { |
if (suntil && tag == suntil) |
if (suntil && tag == suntil) |
return; |
return; |
if (tag == h->metaf) |
print_ctag(h, tag); |
h->metaf = NULL; |
|
print_ctag(h, tag->tag); |
|
h->tags.head = tag->next; |
|
free(tag); |
|
} |
} |
} |
} |
|
|
|
|
void |
void |
bufinit(struct html *h) |
print_paragraph(struct html *h) |
{ |
{ |
|
struct tag *t; |
|
|
h->buf[0] = '\0'; |
t = print_otag(h, TAG_DIV, "c", "Pp"); |
h->buflen = 0; |
print_tagq(h, t); |
} |
} |
|
|
|
|
void |
/*********************************************************************** |
bufcat_style(struct html *h, const char *key, const char *val) |
* Low level output functions. |
{ |
* They implement line breaking using a short static buffer. |
|
***********************************************************************/ |
|
|
bufcat(h, key); |
/* |
bufncat(h, ":", 1); |
* Buffer one HTML output byte. |
bufcat(h, val); |
* If the buffer is full, flush and deactivate it and start a new line. |
bufncat(h, ";", 1); |
* If the buffer is inactive, print directly. |
} |
*/ |
|
static void |
|
print_byte(struct html *h, char c) |
void |
|
bufcat(struct html *h, const char *p) |
|
{ |
{ |
|
if ((h->flags & HTML_BUFFER) == 0) { |
|
putchar(c); |
|
h->col++; |
|
return; |
|
} |
|
|
bufncat(h, p, strlen(p)); |
if (h->col + h->bufcol < sizeof(h->buf)) { |
} |
h->buf[h->bufcol++] = c; |
|
return; |
|
} |
|
|
|
putchar('\n'); |
void |
h->col = 0; |
buffmt(struct html *h, const char *fmt, ...) |
print_indent(h); |
{ |
putchar(' '); |
va_list ap; |
putchar(' '); |
|
fwrite(h->buf, h->bufcol, 1, stdout); |
va_start(ap, fmt); |
putchar(c); |
(void)vsnprintf(h->buf + (int)h->buflen, |
h->col = (h->indent + 1) * 2 + h->bufcol + 1; |
BUFSIZ - h->buflen - 1, fmt, ap); |
h->bufcol = 0; |
va_end(ap); |
h->flags &= ~HTML_BUFFER; |
h->buflen = strlen(h->buf); |
|
} |
} |
|
|
|
/* |
|
* If something was printed on the current output line, end it. |
|
* Not to be called right after print_indent(). |
|
*/ |
void |
void |
bufncat(struct html *h, const char *p, size_t sz) |
print_endline(struct html *h) |
{ |
{ |
|
if (h->col == 0) |
|
return; |
|
|
if (h->buflen + sz > BUFSIZ - 1) |
if (h->bufcol) { |
sz = BUFSIZ - 1 - h->buflen; |
putchar(' '); |
|
fwrite(h->buf, h->bufcol, 1, stdout); |
(void)strncat(h->buf, p, sz); |
h->bufcol = 0; |
h->buflen += sz; |
} |
|
putchar('\n'); |
|
h->col = 0; |
|
h->flags |= HTML_NOSPACE; |
|
h->flags &= ~HTML_BUFFER; |
} |
} |
|
|
|
/* |
void |
* Flush the HTML output buffer. |
buffmt_includes(struct html *h, const char *name) |
* If it is inactive, activate it. |
|
*/ |
|
static void |
|
print_endword(struct html *h) |
{ |
{ |
const char *p, *pp; |
if (h->noindent) { |
|
print_byte(h, ' '); |
pp = h->base_includes; |
return; |
|
|
while (NULL != (p = strchr(pp, '%'))) { |
|
bufncat(h, pp, (size_t)(p - pp)); |
|
switch (*(p + 1)) { |
|
case('I'): |
|
bufcat(h, name); |
|
break; |
|
default: |
|
bufncat(h, p, 2); |
|
break; |
|
} |
|
pp = p + 2; |
|
} |
} |
if (pp) |
|
bufcat(h, pp); |
|
} |
|
|
|
|
if ((h->flags & HTML_BUFFER) == 0) { |
void |
h->col++; |
buffmt_man(struct html *h, |
h->flags |= HTML_BUFFER; |
const char *name, const char *sec) |
} else if (h->bufcol) { |
{ |
putchar(' '); |
const char *p, *pp; |
fwrite(h->buf, h->bufcol, 1, stdout); |
|
h->col += h->bufcol + 1; |
pp = h->base_man; |
|
|
|
/* LINTED */ |
|
while (NULL != (p = strchr(pp, '%'))) { |
|
bufncat(h, pp, (size_t)(p - pp)); |
|
switch (*(p + 1)) { |
|
case('S'): |
|
bufcat(h, sec ? sec : "1"); |
|
break; |
|
case('N'): |
|
buffmt(h, name); |
|
break; |
|
default: |
|
bufncat(h, p, 2); |
|
break; |
|
} |
|
pp = p + 2; |
|
} |
} |
if (pp) |
h->bufcol = 0; |
bufcat(h, pp); |
|
} |
} |
|
|
|
/* |
void |
* If at the beginning of a new output line, |
bufcat_su(struct html *h, const char *p, const struct roffsu *su) |
* perform indentation and mark the line as containing output. |
|
* Make sure to really produce some output right afterwards, |
|
* but do not use print_otag() for producing it. |
|
*/ |
|
static void |
|
print_indent(struct html *h) |
{ |
{ |
double v; |
size_t i; |
const char *u; |
|
|
|
v = su->scale; |
if (h->col) |
|
return; |
|
|
switch (su->unit) { |
if (h->noindent == 0) { |
case (SCALE_CM): |
h->col = h->indent * 2; |
u = "cm"; |
for (i = 0; i < h->col; i++) |
break; |
putchar(' '); |
case (SCALE_IN): |
|
u = "in"; |
|
break; |
|
case (SCALE_PC): |
|
u = "pc"; |
|
break; |
|
case (SCALE_PT): |
|
u = "pt"; |
|
break; |
|
case (SCALE_EM): |
|
u = "em"; |
|
break; |
|
case (SCALE_MM): |
|
if (0 == (v /= 100)) |
|
v = 1; |
|
u = "em"; |
|
break; |
|
case (SCALE_EN): |
|
u = "ex"; |
|
break; |
|
case (SCALE_BU): |
|
u = "ex"; |
|
break; |
|
case (SCALE_VS): |
|
u = "em"; |
|
break; |
|
default: |
|
u = "ex"; |
|
break; |
|
} |
} |
|
h->flags &= ~HTML_NOSPACE; |
if (su->pt) |
|
buffmt(h, "%s: %f%s;", p, v, u); |
|
else |
|
/* LINTED */ |
|
buffmt(h, "%s: %d%s;", p, (int)v, u); |
|
} |
} |
|
|
|
/* |
void |
* Print or buffer some characters |
html_idcat(char *dst, const char *src, int sz) |
* depending on the current HTML output buffer state. |
|
*/ |
|
static void |
|
print_word(struct html *h, const char *cp) |
{ |
{ |
int ssz; |
while (*cp != '\0') |
|
print_byte(h, *cp++); |
assert(sz); |
|
|
|
/* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */ |
|
|
|
for ( ; *dst != '\0' && sz; dst++, sz--) |
|
/* Jump to end. */ ; |
|
|
|
assert(sz > 2); |
|
|
|
/* We can't start with a number (bah). */ |
|
|
|
*dst++ = 'x'; |
|
*dst = '\0'; |
|
sz--; |
|
|
|
for ( ; *src != '\0' && sz > 1; src++) { |
|
ssz = snprintf(dst, (size_t)sz, "%.2x", *src); |
|
sz -= ssz; |
|
dst += ssz; |
|
} |
|
} |
} |