version 1.108, 2010/07/21 20:35:03 |
version 1.166, 2014/09/27 09:03:24 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv> |
|
* Copyright (c) 2011, 2012, 2013, 2014 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 |
|
|
* 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 <unistd.h> |
#include <unistd.h> |
|
|
#include "mandoc.h" |
#include "mandoc.h" |
|
#include "mandoc_aux.h" |
|
#include "libmandoc.h" |
#include "out.h" |
#include "out.h" |
#include "chars.h" |
|
#include "html.h" |
#include "html.h" |
#include "main.h" |
#include "main.h" |
|
|
Line 57 static const struct htmldata htmltags[TAG_MAX] = { |
|
Line 57 static const struct htmldata htmltags[TAG_MAX] = { |
|
{"br", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_BR */ |
{"br", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_BR */ |
{"a", 0}, /* TAG_A */ |
{"a", 0}, /* TAG_A */ |
{"table", HTML_CLRLINE}, /* TAG_TABLE */ |
{"table", HTML_CLRLINE}, /* TAG_TABLE */ |
|
{"tbody", HTML_CLRLINE}, /* TAG_TBODY */ |
{"col", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_COL */ |
{"col", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_COL */ |
{"tr", HTML_CLRLINE}, /* TAG_TR */ |
{"tr", HTML_CLRLINE}, /* TAG_TR */ |
{"td", HTML_CLRLINE}, /* TAG_TD */ |
{"td", HTML_CLRLINE}, /* TAG_TD */ |
{"li", HTML_CLRLINE}, /* TAG_LI */ |
{"li", HTML_CLRLINE}, /* TAG_LI */ |
{"ul", HTML_CLRLINE}, /* TAG_UL */ |
{"ul", HTML_CLRLINE}, /* TAG_UL */ |
{"ol", HTML_CLRLINE}, /* TAG_OL */ |
{"ol", HTML_CLRLINE}, /* TAG_OL */ |
|
{"dl", HTML_CLRLINE}, /* TAG_DL */ |
|
{"dt", HTML_CLRLINE}, /* TAG_DT */ |
|
{"dd", HTML_CLRLINE}, /* TAG_DD */ |
|
{"blockquote", HTML_CLRLINE}, /* TAG_BLOCKQUOTE */ |
|
{"p", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_P */ |
|
{"pre", HTML_CLRLINE }, /* TAG_PRE */ |
|
{"b", 0 }, /* TAG_B */ |
|
{"i", 0 }, /* TAG_I */ |
|
{"code", 0 }, /* TAG_CODE */ |
|
{"small", 0 }, /* TAG_SMALL */ |
|
{"style", HTML_CLRLINE}, /* TAG_STYLE */ |
}; |
}; |
|
|
static const char *const htmlfonts[HTMLFONT_MAX] = { |
static const char *const htmlattrs[ATTR_MAX] = { |
"roman", |
"name", /* ATTR_NAME */ |
"bold", |
"rel", /* ATTR_REL */ |
"italic" |
"href", /* ATTR_HREF */ |
|
"type", /* ATTR_TYPE */ |
|
"media", /* ATTR_MEDIA */ |
|
"class", /* ATTR_CLASS */ |
|
"style", /* ATTR_STYLE */ |
|
"width", /* ATTR_WIDTH */ |
|
"id", /* ATTR_ID */ |
|
"summary", /* ATTR_SUMMARY */ |
|
"align", /* ATTR_ALIGN */ |
|
"colspan", /* ATTR_COLSPAN */ |
|
"charset", /* ATTR_CHARSET */ |
}; |
}; |
|
|
static const char *const htmlattrs[ATTR_MAX] = { |
static const char *const roffscales[SCALE_MAX] = { |
"http-equiv", |
"cm", /* SCALE_CM */ |
"content", |
"in", /* SCALE_IN */ |
"name", |
"pc", /* SCALE_PC */ |
"rel", |
"pt", /* SCALE_PT */ |
"href", |
"em", /* SCALE_EM */ |
"type", |
"em", /* SCALE_MM */ |
"media", |
"ex", /* SCALE_EN */ |
"class", |
"ex", /* SCALE_BU */ |
"style", |
"em", /* SCALE_VS */ |
"width", |
"ex", /* SCALE_FS */ |
"valign", |
|
"target", |
|
"id", |
|
"summary", |
|
}; |
}; |
|
|
static void print_spec(struct html *, enum roffdeco, |
static void bufncat(struct html *, const char *, size_t); |
const char *, size_t); |
static void print_ctag(struct html *, enum htmltag); |
static void print_res(struct html *, const char *, size_t); |
static int print_escape(char); |
static void print_ctag(struct html *, enum htmltag); |
static int print_encode(struct html *, const char *, int); |
static void print_doctype(struct html *); |
static void print_metaf(struct html *, enum mandoc_esc); |
static void print_xmltype(struct html *); |
static void print_attr(struct html *, const char *, const char *); |
static int print_encode(struct html *, const char *, int); |
static void *ml_alloc(char *, enum htmltype); |
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 * |
static void * |
ml_alloc(char *outopts, enum htmltype type) |
ml_alloc(char *outopts, enum htmltype type) |
{ |
{ |
struct html *h; |
struct html *h; |
const char *toks[4]; |
const char *toks[5]; |
char *v; |
char *v; |
|
|
toks[0] = "style"; |
toks[0] = "style"; |
toks[1] = "man"; |
toks[1] = "man"; |
toks[2] = "includes"; |
toks[2] = "includes"; |
toks[3] = NULL; |
toks[3] = "fragment"; |
|
toks[4] = NULL; |
|
|
h = calloc(1, sizeof(struct html)); |
h = mandoc_calloc(1, sizeof(struct html)); |
if (NULL == h) { |
|
perror(NULL); |
|
exit(EXIT_FAILURE); |
|
} |
|
|
|
h->type = type; |
h->type = type; |
h->tags.head = NULL; |
h->tags.head = NULL; |
h->ords.head = NULL; |
h->symtab = mchars_alloc(); |
h->symtab = chars_init(CHARS_HTML); |
|
|
|
while (outopts && *outopts) |
while (outopts && *outopts) |
switch (getsubopt(&outopts, UNCONST(toks), &v)) { |
switch (getsubopt(&outopts, UNCONST(toks), &v)) { |
case (0): |
case 0: |
h->style = v; |
h->style = v; |
break; |
break; |
case (1): |
case 1: |
h->base_man = v; |
h->base_man = v; |
break; |
break; |
case (2): |
case 2: |
h->base_includes = v; |
h->base_includes = v; |
break; |
break; |
|
case 3: |
|
h->oflags |= HTML_FRAGMENT; |
|
break; |
default: |
default: |
break; |
break; |
} |
} |
Line 149 html_alloc(char *outopts) |
|
Line 162 html_alloc(char *outopts) |
|
return(ml_alloc(outopts, HTML_HTML_4_01_STRICT)); |
return(ml_alloc(outopts, HTML_HTML_4_01_STRICT)); |
} |
} |
|
|
|
|
void * |
void * |
xhtml_alloc(char *outopts) |
xhtml_alloc(char *outopts) |
{ |
{ |
Line 157 xhtml_alloc(char *outopts) |
|
Line 169 xhtml_alloc(char *outopts) |
|
return(ml_alloc(outopts, HTML_XHTML_1_0_STRICT)); |
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; |
|
|
h = (struct html *)p; |
h = (struct html *)p; |
|
|
while ((ord = h->ords.head) != NULL) { |
|
h->ords.head = ord->next; |
|
free(ord); |
|
} |
|
|
|
while ((tag = h->tags.head) != NULL) { |
while ((tag = h->tags.head) != NULL) { |
h->tags.head = tag->next; |
h->tags.head = tag->next; |
free(tag); |
free(tag); |
} |
} |
|
|
if (h->symtab) |
if (h->symtab) |
chars_free(h->symtab); |
mchars_free(h->symtab); |
|
|
free(h); |
free(h); |
} |
} |
|
|
|
|
void |
void |
print_gen_head(struct html *h) |
print_gen_head(struct html *h) |
{ |
{ |
struct htmlpair tag[4]; |
struct htmlpair tag[4]; |
|
struct tag *t; |
|
|
tag[0].key = ATTR_HTTPEQUIV; |
tag[0].key = ATTR_CHARSET; |
tag[0].val = "Content-Type"; |
tag[0].val = "utf-8"; |
tag[1].key = ATTR_CONTENT; |
print_otag(h, TAG_META, 1, tag); |
tag[1].val = "text/html; charset=utf-8"; |
|
print_otag(h, TAG_META, 2, tag); |
|
|
|
tag[0].key = ATTR_NAME; |
t = print_otag(h, TAG_STYLE, 0, NULL); |
tag[0].val = "resource-type"; |
print_text(h, "table.head, table.foot { width: 100%; }\n"); |
tag[1].key = ATTR_CONTENT; |
print_tagq(h, t); |
tag[1].val = "document"; |
|
print_otag(h, TAG_META, 2, tag); |
|
|
|
if (h->style) { |
if (h->style) { |
tag[0].key = ATTR_REL; |
tag[0].key = ATTR_REL; |
Line 214 print_gen_head(struct html *h) |
|
Line 215 print_gen_head(struct html *h) |
|
} |
} |
} |
} |
|
|
|
|
static void |
static void |
print_spec(struct html *h, enum roffdeco d, const char *p, size_t len) |
print_metaf(struct html *h, enum mandoc_esc deco) |
{ |
{ |
int cp; |
enum htmlfont font; |
const char *rhs; |
|
size_t sz; |
|
|
|
if ((cp = chars_spec2cp(h->symtab, p, len)) > 0) { |
switch (deco) { |
printf("&#%d;", cp); |
case ESCAPE_FONTPREV: |
return; |
font = h->metal; |
} else if (-1 == cp && DECO_SSPECIAL == d) { |
break; |
fwrite(p, 1, len, stdout); |
case ESCAPE_FONTITALIC: |
return; |
font = HTMLFONT_ITALIC; |
} else if (-1 == cp) |
break; |
return; |
case ESCAPE_FONTBOLD: |
|
font = HTMLFONT_BOLD; |
|
break; |
|
case ESCAPE_FONTBI: |
|
font = HTMLFONT_BI; |
|
break; |
|
case ESCAPE_FONT: |
|
/* FALLTHROUGH */ |
|
case ESCAPE_FONTROMAN: |
|
font = HTMLFONT_NONE; |
|
break; |
|
default: |
|
abort(); |
|
/* NOTREACHED */ |
|
} |
|
|
if (NULL != (rhs = chars_spec2str(h->symtab, p, len, &sz))) |
if (h->metaf) { |
fwrite(rhs, 1, sz, stdout); |
print_tagq(h, h->metaf); |
} |
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, 0, NULL); |
int cp; |
break; |
const char *rhs; |
case HTMLFONT_BOLD: |
size_t sz; |
h->metaf = print_otag(h, TAG_B, 0, NULL); |
|
break; |
if ((cp = chars_res2cp(h->symtab, p, len)) > 0) { |
case HTMLFONT_BI: |
printf("&#%d;", cp); |
h->metaf = print_otag(h, TAG_B, 0, NULL); |
return; |
print_otag(h, TAG_I, 0, NULL); |
} else if (-1 == cp) |
break; |
return; |
default: |
|
break; |
if (NULL != (rhs = chars_res2str(h->symtab, p, len, &sz))) |
} |
fwrite(rhs, 1, sz, stdout); |
|
} |
} |
|
|
|
int |
struct tag * |
html_strlen(const char *cp) |
print_ofont(struct html *h, enum htmlfont font) |
|
{ |
{ |
struct htmlpair tag; |
size_t rsz; |
|
int skip, sz; |
|
|
h->metal = h->metac; |
/* |
h->metac = font; |
* Account for escaped sequences within string length |
|
* calculations. This follows the logic in term_strlen() as we |
|
* must calculate the width of produced strings. |
|
* Assume that characters are always width of "1". This is |
|
* hacky, but it gets the job done for approximation of widths. |
|
*/ |
|
|
/* FIXME: DECO_ROMAN should just close out preexisting. */ |
sz = 0; |
|
skip = 0; |
if (h->metaf && h->tags.head == h->metaf) |
while (1) { |
print_tagq(h, h->metaf); |
rsz = strcspn(cp, "\\"); |
|
if (rsz) { |
PAIR_CLASS_INIT(&tag, htmlfonts[font]); |
cp += rsz; |
h->metaf = print_otag(h, TAG_SPAN, 1, &tag); |
if (skip) { |
return(h->metaf); |
skip = 0; |
|
rsz--; |
|
} |
|
sz += rsz; |
|
} |
|
if ('\0' == *cp) |
|
break; |
|
cp++; |
|
switch (mandoc_escape(&cp, NULL, NULL)) { |
|
case ESCAPE_ERROR: |
|
return(sz); |
|
case ESCAPE_UNICODE: |
|
/* FALLTHROUGH */ |
|
case ESCAPE_NUMBERED: |
|
/* FALLTHROUGH */ |
|
case ESCAPE_SPECIAL: |
|
if (skip) |
|
skip = 0; |
|
else |
|
sz++; |
|
break; |
|
case ESCAPE_SKIPCHAR: |
|
skip = 1; |
|
break; |
|
default: |
|
break; |
|
} |
|
} |
|
return(sz); |
} |
} |
|
|
|
static int |
static void |
print_escape(char c) |
print_metaf(struct html *h, enum roffdeco deco) |
|
{ |
{ |
enum htmlfont font; |
|
|
|
switch (deco) { |
switch (c) { |
case (DECO_PREVIOUS): |
case '<': |
font = h->metal; |
printf("<"); |
break; |
break; |
case (DECO_ITALIC): |
case '>': |
font = HTMLFONT_ITALIC; |
printf(">"); |
break; |
break; |
case (DECO_BOLD): |
case '&': |
font = HTMLFONT_BOLD; |
printf("&"); |
break; |
break; |
case (DECO_ROMAN): |
case '"': |
font = HTMLFONT_NONE; |
printf("""); |
break; |
break; |
|
case ASCII_NBRSP: |
|
putchar('-'); |
|
break; |
|
case ASCII_HYPH: |
|
putchar('-'); |
|
/* FALLTHROUGH */ |
|
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, int norecurse) |
{ |
{ |
size_t sz; |
size_t sz; |
int len, nospace; |
int c, len, nospace; |
const char *seq; |
const char *seq; |
enum roffdeco deco; |
enum mandoc_esc esc; |
static const char rejs[6] = { '\\', '<', '>', '&', ASCII_HYPH, '\0' }; |
static const char rejs[9] = { '\\', '<', '>', '&', '"', |
|
ASCII_NBRSP, ASCII_HYPH, ASCII_BREAK, '\0' }; |
|
|
nospace = 0; |
nospace = 0; |
|
|
for (; *p; p++) { |
while ('\0' != *p) { |
|
if (HTML_SKIPCHAR & h->flags && '\\' != *p) { |
|
h->flags &= ~HTML_SKIPCHAR; |
|
p++; |
|
continue; |
|
} |
|
|
sz = strcspn(p, rejs); |
sz = strcspn(p, rejs); |
|
|
fwrite(p, 1, sz, stdout); |
fwrite(p, 1, sz, stdout); |
p += /* LINTED */ |
p += (int)sz; |
sz; |
|
|
|
if ('<' == *p) { |
if ('\0' == *p) |
printf("<"); |
|
continue; |
|
} else if ('>' == *p) { |
|
printf(">"); |
|
continue; |
|
} else if ('&' == *p) { |
|
printf("&"); |
|
continue; |
|
} else if (ASCII_HYPH == *p) { |
|
/* |
|
* Note: "soft hyphens" aren't graphically |
|
* displayed when not breaking the text; we want |
|
* them to be displayed. |
|
*/ |
|
/*printf("­");*/ |
|
putchar('-'); |
|
continue; |
|
} else if ('\0' == *p) |
|
break; |
break; |
|
|
seq = ++p; |
if (print_escape(*p++)) |
len = a2roffdeco(&deco, &seq, &sz); |
continue; |
|
|
switch (deco) { |
esc = mandoc_escape(&p, &seq, &len); |
case (DECO_RESERVED): |
if (ESCAPE_ERROR == esc) |
print_res(h, seq, sz); |
|
break; |
break; |
case (DECO_SSPECIAL): |
|
|
switch (esc) { |
|
case ESCAPE_FONT: |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case (DECO_SPECIAL): |
case ESCAPE_FONTPREV: |
print_spec(h, deco, seq, sz); |
|
break; |
|
case (DECO_PREVIOUS): |
|
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case (DECO_BOLD): |
case ESCAPE_FONTBOLD: |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case (DECO_ITALIC): |
case ESCAPE_FONTITALIC: |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case (DECO_ROMAN): |
case ESCAPE_FONTBI: |
if (norecurse) |
/* FALLTHROUGH */ |
break; |
case ESCAPE_FONTROMAN: |
print_metaf(h, deco); |
if (0 == norecurse) |
break; |
print_metaf(h, esc); |
|
continue; |
|
case ESCAPE_SKIPCHAR: |
|
h->flags |= HTML_SKIPCHAR; |
|
continue; |
default: |
default: |
break; |
break; |
} |
} |
|
|
p += len - 1; |
if (h->flags & HTML_SKIPCHAR) { |
|
h->flags &= ~HTML_SKIPCHAR; |
|
continue; |
|
} |
|
|
if (DECO_NOSPACE == deco && '\0' == *(p + 1)) |
switch (esc) { |
nospace = 1; |
case ESCAPE_UNICODE: |
|
/* Skip past "u" header. */ |
|
c = mchars_num2uc(seq + 1, len - 1); |
|
if ('\0' != c) |
|
printf("&#x%x;", c); |
|
break; |
|
case ESCAPE_NUMBERED: |
|
c = mchars_num2char(seq, len); |
|
if ( ! ('\0' == c || print_escape(c))) |
|
putchar(c); |
|
break; |
|
case ESCAPE_SPECIAL: |
|
c = mchars_spec2cp(h->symtab, seq, len); |
|
if (c > 0) |
|
printf("&#%d;", c); |
|
else if (-1 == c && 1 == len && |
|
!print_escape(*seq)) |
|
putchar((int)*seq); |
|
break; |
|
case ESCAPE_NOSPACE: |
|
if ('\0' == *p) |
|
nospace = 1; |
|
break; |
|
default: |
|
break; |
|
} |
} |
} |
|
|
return(nospace); |
return(nospace); |
} |
} |
|
|
|
|
static void |
static void |
print_attr(struct html *h, const char *key, const char *val) |
print_attr(struct html *h, const char *key, const char *val) |
{ |
{ |
Line 384 print_attr(struct html *h, const char *key, const char |
|
Line 451 print_attr(struct html *h, const char *key, const char |
|
putchar('\"'); |
putchar('\"'); |
} |
} |
|
|
|
|
struct tag * |
struct tag * |
print_otag(struct html *h, enum htmltag tag, |
print_otag(struct html *h, enum htmltag tag, |
int sz, const struct htmlpair *p) |
int sz, const struct htmlpair *p) |
{ |
{ |
int i; |
int i; |
Line 395 print_otag(struct html *h, enum htmltag tag, |
|
Line 461 print_otag(struct html *h, enum htmltag tag, |
|
/* Push this tags onto the stack of open scopes. */ |
/* Push this tags onto the stack of open scopes. */ |
|
|
if ( ! (HTML_NOSTACK & htmltags[tag].flags)) { |
if ( ! (HTML_NOSTACK & htmltags[tag].flags)) { |
t = malloc(sizeof(struct tag)); |
t = mandoc_malloc(sizeof(struct tag)); |
if (NULL == t) { |
|
perror(NULL); |
|
exit(EXIT_FAILURE); |
|
} |
|
t->tag = tag; |
t->tag = tag; |
t->next = h->tags.head; |
t->next = h->tags.head; |
h->tags.head = t; |
h->tags.head = t; |
Line 417 print_otag(struct html *h, enum htmltag tag, |
|
Line 479 print_otag(struct html *h, enum htmltag tag, |
|
printf(" "); |
printf(" "); |
} |
} |
|
|
|
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); |
printf("<%s", htmltags[tag].name); |
for (i = 0; i < sz; i++) |
for (i = 0; i < sz; i++) |
print_attr(h, htmlattrs[p[i].key], p[i].val); |
print_attr(h, htmlattrs[p[i].key], p[i].val); |
|
|
/* Add non-overridable attributes. */ |
/* Accommodate for XML "well-formed" singleton escaping. */ |
|
|
if (TAG_HTML == tag && HTML_XHTML_1_0_STRICT == h->type) { |
|
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. */ |
|
|
|
if (HTML_AUTOCLOSE & htmltags[tag].flags) |
if (HTML_AUTOCLOSE & htmltags[tag].flags) |
switch (h->type) { |
switch (h->type) { |
case (HTML_XHTML_1_0_STRICT): |
case HTML_XHTML_1_0_STRICT: |
putchar('/'); |
putchar('/'); |
break; |
break; |
default: |
default: |
Line 445 print_otag(struct html *h, enum htmltag tag, |
|
Line 504 print_otag(struct html *h, enum htmltag tag, |
|
putchar('>'); |
putchar('>'); |
|
|
h->flags |= HTML_NOSPACE; |
h->flags |= HTML_NOSPACE; |
|
|
|
if ((HTML_AUTOCLOSE | HTML_CLRLINE) & htmltags[tag].flags) |
|
putchar('\n'); |
|
|
return(t); |
return(t); |
} |
} |
|
|
|
|
static void |
static void |
print_ctag(struct html *h, enum htmltag tag) |
print_ctag(struct html *h, enum htmltag tag) |
{ |
{ |
|
|
printf("</%s>", htmltags[tag].name); |
printf("</%s>", htmltags[tag].name); |
if (HTML_CLRLINE & htmltags[tag].flags) { |
if (HTML_CLRLINE & htmltags[tag].flags) { |
h->flags |= HTML_NOSPACE; |
h->flags |= HTML_NOSPACE; |
putchar('\n'); |
putchar('\n'); |
} |
} |
} |
} |
|
|
|
|
void |
void |
print_gen_decls(struct html *h) |
print_gen_decls(struct html *h) |
{ |
{ |
|
|
print_xmltype(h); |
puts("<!DOCTYPE html>"); |
print_doctype(h); |
|
} |
} |
|
|
|
|
static void |
|
print_xmltype(struct html *h) |
|
{ |
|
|
|
if (HTML_XHTML_1_0_STRICT == h->type) |
|
printf("<?xml version=\"1.0\" encoding=\"UTF-8\"?>"); |
|
} |
|
|
|
|
|
static void |
|
print_doctype(struct html *h) |
|
{ |
|
const char *doctype; |
|
const char *dtd; |
|
const char *name; |
|
|
|
switch (h->type) { |
|
case (HTML_HTML_4_01_STRICT): |
|
name = "HTML"; |
|
doctype = "-//W3C//DTD HTML 4.01//EN"; |
|
dtd = "http://www.w3.org/TR/html4/strict.dtd"; |
|
break; |
|
default: |
|
name = "html"; |
|
doctype = "-//W3C//DTD XHTML 1.0 Strict//EN"; |
|
dtd = "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"; |
|
break; |
|
} |
|
|
|
printf("<!DOCTYPE %s PUBLIC \"%s\" \"%s\">\n", |
|
name, doctype, dtd); |
|
} |
|
|
|
|
|
void |
void |
print_text(struct html *h, const char *word) |
print_text(struct html *h, const char *word) |
{ |
{ |
|
|
if (word[0] && '\0' == word[1]) |
|
switch (word[0]) { |
|
case('.'): |
|
/* FALLTHROUGH */ |
|
case(','): |
|
/* FALLTHROUGH */ |
|
case(';'): |
|
/* FALLTHROUGH */ |
|
case(':'): |
|
/* FALLTHROUGH */ |
|
case('?'): |
|
/* FALLTHROUGH */ |
|
case('!'): |
|
/* FALLTHROUGH */ |
|
case(')'): |
|
/* FALLTHROUGH */ |
|
case(']'): |
|
if ( ! (HTML_IGNDELIM & h->flags)) |
|
h->flags |= HTML_NOSPACE; |
|
break; |
|
default: |
|
break; |
|
} |
|
|
|
if ( ! (HTML_NOSPACE & h->flags)) { |
if ( ! (HTML_NOSPACE & h->flags)) { |
/* Manage keeps! */ |
/* Manage keeps! */ |
if ( ! (HTML_KEEP & h->flags)) { |
if ( ! (HTML_KEEP & h->flags)) { |
Line 542 print_text(struct html *h, const char *word) |
|
Line 543 print_text(struct html *h, const char *word) |
|
printf(" "); |
printf(" "); |
} |
} |
|
|
|
assert(NULL == h->metaf); |
|
switch (h->metac) { |
|
case HTMLFONT_ITALIC: |
|
h->metaf = print_otag(h, TAG_I, 0, NULL); |
|
break; |
|
case HTMLFONT_BOLD: |
|
h->metaf = print_otag(h, TAG_B, 0, NULL); |
|
break; |
|
case HTMLFONT_BI: |
|
h->metaf = print_otag(h, TAG_B, 0, NULL); |
|
print_otag(h, TAG_I, 0, NULL); |
|
break; |
|
default: |
|
break; |
|
} |
|
|
assert(word); |
assert(word); |
if ( ! print_encode(h, word, 0)) |
if ( ! print_encode(h, word, 0)) { |
h->flags &= ~HTML_NOSPACE; |
if ( ! (h->flags & HTML_NONOSPACE)) |
|
h->flags &= ~HTML_NOSPACE; |
|
} else |
|
h->flags |= HTML_NOSPACE; |
|
|
/* |
if (h->metaf) { |
* Note that we don't process the pipe: the parser sees it as |
print_tagq(h, h->metaf); |
* punctuation, but we don't in terms of typography. |
h->metaf = NULL; |
*/ |
} |
if (word[0] && '\0' == word[1]) |
|
switch (word[0]) { |
h->flags &= ~HTML_IGNDELIM; |
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->tags.head) != NULL) { |
|
/* |
|
* Remember to close out and nullify the current |
|
* meta-font and table, if applicable. |
|
*/ |
if (tag == h->metaf) |
if (tag == h->metaf) |
h->metaf = NULL; |
h->metaf = NULL; |
|
if (tag == h->tblt) |
|
h->tblt = NULL; |
print_ctag(h, tag->tag); |
print_ctag(h, tag->tag); |
h->tags.head = tag->next; |
h->tags.head = tag->next; |
free(tag); |
free(tag); |
Line 579 print_tagq(struct html *h, const struct tag *until) |
|
Line 596 print_tagq(struct html *h, const struct tag *until) |
|
} |
} |
} |
} |
|
|
|
|
void |
void |
print_stagq(struct html *h, const struct tag *suntil) |
print_stagq(struct html *h, const struct tag *suntil) |
{ |
{ |
Line 588 print_stagq(struct html *h, const struct tag *suntil) |
|
Line 604 print_stagq(struct html *h, const struct tag *suntil) |
|
while ((tag = h->tags.head) != NULL) { |
while ((tag = h->tags.head) != NULL) { |
if (suntil && tag == suntil) |
if (suntil && tag == suntil) |
return; |
return; |
|
/* |
|
* Remember to close out and nullify the current |
|
* meta-font and table, if applicable. |
|
*/ |
if (tag == h->metaf) |
if (tag == h->metaf) |
h->metaf = NULL; |
h->metaf = NULL; |
|
if (tag == h->tblt) |
|
h->tblt = NULL; |
print_ctag(h, tag->tag); |
print_ctag(h, tag->tag); |
h->tags.head = tag->next; |
h->tags.head = tag->next; |
free(tag); |
free(tag); |
} |
} |
} |
} |
|
|
|
|
void |
void |
bufinit(struct html *h) |
bufinit(struct html *h) |
{ |
{ |
Line 605 bufinit(struct html *h) |
|
Line 626 bufinit(struct html *h) |
|
h->buflen = 0; |
h->buflen = 0; |
} |
} |
|
|
|
|
void |
void |
bufcat_style(struct html *h, const char *key, const char *val) |
bufcat_style(struct html *h, const char *key, const char *val) |
{ |
{ |
|
|
bufcat(h, key); |
bufcat(h, key); |
bufncat(h, ":", 1); |
bufcat(h, ":"); |
bufcat(h, val); |
bufcat(h, val); |
bufncat(h, ";", 1); |
bufcat(h, ";"); |
} |
} |
|
|
|
|
void |
void |
bufcat(struct html *h, const char *p) |
bufcat(struct html *h, const char *p) |
{ |
{ |
|
|
bufncat(h, p, strlen(p)); |
/* |
|
* XXX This is broken and not easy to fix. |
|
* When using the -Oincludes option, buffmt_includes() |
|
* may pass in strings overrunning BUFSIZ, causing a crash. |
|
*/ |
|
|
|
h->buflen = strlcat(h->buf, p, BUFSIZ); |
|
assert(h->buflen < BUFSIZ); |
} |
} |
|
|
|
|
void |
void |
buffmt(struct html *h, const char *fmt, ...) |
bufcat_fmt(struct html *h, const char *fmt, ...) |
{ |
{ |
va_list ap; |
va_list ap; |
|
|
va_start(ap, fmt); |
va_start(ap, fmt); |
(void)vsnprintf(h->buf + (int)h->buflen, |
(void)vsnprintf(h->buf + (int)h->buflen, |
BUFSIZ - h->buflen - 1, fmt, ap); |
BUFSIZ - h->buflen - 1, fmt, ap); |
va_end(ap); |
va_end(ap); |
h->buflen = strlen(h->buf); |
h->buflen = strlen(h->buf); |
} |
} |
|
|
|
static void |
void |
|
bufncat(struct html *h, const char *p, size_t sz) |
bufncat(struct html *h, const char *p, size_t sz) |
{ |
{ |
|
|
if (h->buflen + sz > BUFSIZ - 1) |
assert(h->buflen + sz + 1 < BUFSIZ); |
sz = BUFSIZ - 1 - h->buflen; |
strncat(h->buf, p, sz); |
|
|
(void)strncat(h->buf, p, sz); |
|
h->buflen += sz; |
h->buflen += sz; |
} |
} |
|
|
|
|
void |
void |
buffmt_includes(struct html *h, const char *name) |
buffmt_includes(struct html *h, const char *name) |
{ |
{ |
const char *p, *pp; |
const char *p, *pp; |
|
|
pp = h->base_includes; |
pp = h->base_includes; |
|
|
|
bufinit(h); |
while (NULL != (p = strchr(pp, '%'))) { |
while (NULL != (p = strchr(pp, '%'))) { |
bufncat(h, pp, (size_t)(p - pp)); |
bufncat(h, pp, (size_t)(p - pp)); |
switch (*(p + 1)) { |
switch (*(p + 1)) { |
case('I'): |
case'I': |
bufcat(h, name); |
bufcat(h, name); |
break; |
break; |
default: |
default: |
Line 673 buffmt_includes(struct html *h, const char *name) |
|
Line 695 buffmt_includes(struct html *h, const char *name) |
|
bufcat(h, pp); |
bufcat(h, pp); |
} |
} |
|
|
|
|
void |
void |
buffmt_man(struct html *h, |
buffmt_man(struct html *h, const char *name, const char *sec) |
const char *name, const char *sec) |
|
{ |
{ |
const char *p, *pp; |
const char *p, *pp; |
|
|
pp = h->base_man; |
pp = h->base_man; |
|
|
/* LINTED */ |
bufinit(h); |
while (NULL != (p = strchr(pp, '%'))) { |
while (NULL != (p = strchr(pp, '%'))) { |
bufncat(h, pp, (size_t)(p - pp)); |
bufncat(h, pp, (size_t)(p - pp)); |
switch (*(p + 1)) { |
switch (*(p + 1)) { |
case('S'): |
case 'S': |
bufcat(h, sec ? sec : "1"); |
bufcat(h, sec ? sec : "1"); |
break; |
break; |
case('N'): |
case 'N': |
buffmt(h, name); |
bufcat_fmt(h, "%s", name); |
break; |
break; |
default: |
default: |
bufncat(h, p, 2); |
bufncat(h, p, 2); |
Line 702 buffmt_man(struct html *h, |
|
Line 722 buffmt_man(struct html *h, |
|
bufcat(h, pp); |
bufcat(h, pp); |
} |
} |
|
|
|
|
void |
void |
bufcat_su(struct html *h, const char *p, const struct roffsu *su) |
bufcat_su(struct html *h, const char *p, const struct roffsu *su) |
{ |
{ |
double v; |
double v; |
const char *u; |
|
|
|
v = su->scale; |
v = su->scale; |
|
if (SCALE_MM == su->unit && 0.0 == (v /= 100.0)) |
|
v = 1.0; |
|
else if (SCALE_BU == su->unit) |
|
v /= 24.0; |
|
|
switch (su->unit) { |
bufcat_fmt(h, "%s: %.2f%s;", p, v, roffscales[su->unit]); |
case (SCALE_CM): |
|
u = "cm"; |
|
break; |
|
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; |
|
} |
|
|
|
/* |
|
* XXX: the CSS spec isn't clear as to which types accept |
|
* integer or real numbers, so we just make them all decimals. |
|
*/ |
|
buffmt(h, "%s: %.2f%s;", p, v, u); |
|
} |
} |
|
|
|
|
void |
void |
html_idcat(char *dst, const char *src, int sz) |
bufcat_id(struct html *h, const char *src) |
{ |
{ |
int ssz; |
|
|
|
assert(sz); |
|
|
|
/* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */ |
/* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */ |
|
|
for ( ; *dst != '\0' && sz; dst++, sz--) |
while ('\0' != *src) |
/* Jump to end. */ ; |
bufcat_fmt(h, "%.2x", *src++); |
|
|
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; |
|
} |
|
} |
} |