version 1.204, 2017/01/29 14:02:41 |
version 1.258, 2019/09/01 15:12:19 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011-2015, 2017 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2011-2015, 2017-2019 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 |
|
|
#include "config.h" |
#include "config.h" |
|
|
#include <sys/types.h> |
#include <sys/types.h> |
|
#include <sys/stat.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.h" |
|
#include "mandoc_aux.h" |
#include "mandoc_aux.h" |
|
#include "mandoc_ohash.h" |
|
#include "mandoc.h" |
|
#include "roff.h" |
#include "out.h" |
#include "out.h" |
#include "html.h" |
#include "html.h" |
#include "manconf.h" |
#include "manconf.h" |
|
|
struct htmldata { |
struct htmldata { |
const char *name; |
const char *name; |
int flags; |
int flags; |
#define HTML_NOSTACK (1 << 0) |
#define HTML_INPHRASE (1 << 0) /* Can appear in phrasing context. */ |
#define HTML_AUTOCLOSE (1 << 1) |
#define HTML_TOPHRASE (1 << 1) /* Establishes phrasing context. */ |
#define HTML_NLBEFORE (1 << 2) |
#define HTML_NOSTACK (1 << 2) /* Does not have an end tag. */ |
#define HTML_NLBEGIN (1 << 3) |
#define HTML_NLBEFORE (1 << 3) /* Output line break before opening. */ |
#define HTML_NLEND (1 << 4) |
#define HTML_NLBEGIN (1 << 4) /* Output line break after opening. */ |
#define HTML_NLAFTER (1 << 5) |
#define HTML_NLEND (1 << 5) /* Output line break before closing. */ |
|
#define HTML_NLAFTER (1 << 6) /* Output line break after closing. */ |
#define HTML_NLAROUND (HTML_NLBEFORE | HTML_NLAFTER) |
#define HTML_NLAROUND (HTML_NLBEFORE | HTML_NLAFTER) |
#define HTML_NLINSIDE (HTML_NLBEGIN | HTML_NLEND) |
#define HTML_NLINSIDE (HTML_NLBEGIN | HTML_NLEND) |
#define HTML_NLALL (HTML_NLAROUND | HTML_NLINSIDE) |
#define HTML_NLALL (HTML_NLAROUND | HTML_NLINSIDE) |
#define HTML_INDENT (1 << 6) |
#define HTML_INDENT (1 << 7) /* Indent content by two spaces. */ |
#define HTML_NOINDENT (1 << 7) |
#define HTML_NOINDENT (1 << 8) /* Exception: never indent content. */ |
}; |
}; |
|
|
static const struct htmldata htmltags[TAG_MAX] = { |
static const struct htmldata htmltags[TAG_MAX] = { |
{"html", HTML_NLALL}, |
{"html", HTML_NLALL}, |
{"head", HTML_NLALL | HTML_INDENT}, |
{"head", HTML_NLALL | HTML_INDENT}, |
{"body", HTML_NLALL}, |
{"meta", HTML_NOSTACK | HTML_NLALL}, |
{"meta", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, |
{"link", HTML_NOSTACK | HTML_NLALL}, |
|
{"style", HTML_NLALL | HTML_INDENT}, |
{"title", HTML_NLAROUND}, |
{"title", HTML_NLAROUND}, |
|
{"body", HTML_NLALL}, |
{"div", HTML_NLAROUND}, |
{"div", HTML_NLAROUND}, |
{"h1", HTML_NLAROUND}, |
{"section", HTML_NLALL}, |
{"h2", HTML_NLAROUND}, |
|
{"span", 0}, |
|
{"link", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, |
|
{"br", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, |
|
{"a", 0}, |
|
{"table", HTML_NLALL | HTML_INDENT}, |
{"table", HTML_NLALL | HTML_INDENT}, |
{"tbody", HTML_NLALL | HTML_INDENT}, |
|
{"col", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, |
|
{"tr", HTML_NLALL | HTML_INDENT}, |
{"tr", HTML_NLALL | HTML_INDENT}, |
{"td", HTML_NLAROUND}, |
{"td", HTML_NLAROUND}, |
{"li", HTML_NLAROUND | HTML_INDENT}, |
{"li", HTML_NLAROUND | HTML_INDENT}, |
Line 75 static const struct htmldata htmltags[TAG_MAX] = { |
|
Line 75 static const struct htmldata htmltags[TAG_MAX] = { |
|
{"dl", HTML_NLALL | HTML_INDENT}, |
{"dl", HTML_NLALL | HTML_INDENT}, |
{"dt", HTML_NLAROUND}, |
{"dt", HTML_NLAROUND}, |
{"dd", HTML_NLAROUND | HTML_INDENT}, |
{"dd", HTML_NLAROUND | HTML_INDENT}, |
{"pre", HTML_NLALL | HTML_NOINDENT}, |
{"h1", HTML_TOPHRASE | HTML_NLAROUND}, |
{"b", 0}, |
{"h2", HTML_TOPHRASE | HTML_NLAROUND}, |
{"i", 0}, |
{"p", HTML_TOPHRASE | HTML_NLAROUND | HTML_INDENT}, |
{"code", 0}, |
{"pre", HTML_TOPHRASE | HTML_NLALL | HTML_NOINDENT}, |
{"small", 0}, |
{"a", HTML_INPHRASE | HTML_TOPHRASE}, |
{"style", HTML_NLALL | HTML_INDENT}, |
{"b", HTML_INPHRASE | HTML_TOPHRASE}, |
{"math", HTML_NLALL | HTML_INDENT}, |
{"cite", HTML_INPHRASE | HTML_TOPHRASE}, |
|
{"code", HTML_INPHRASE | HTML_TOPHRASE}, |
|
{"i", HTML_INPHRASE | HTML_TOPHRASE}, |
|
{"small", HTML_INPHRASE | HTML_TOPHRASE}, |
|
{"span", HTML_INPHRASE | HTML_TOPHRASE}, |
|
{"var", HTML_INPHRASE | HTML_TOPHRASE}, |
|
{"br", HTML_INPHRASE | HTML_NOSTACK | HTML_NLALL}, |
|
{"math", HTML_INPHRASE | HTML_NLALL | HTML_INDENT}, |
{"mrow", 0}, |
{"mrow", 0}, |
{"mi", 0}, |
{"mi", 0}, |
|
{"mn", 0}, |
{"mo", 0}, |
{"mo", 0}, |
{"msup", 0}, |
{"msup", 0}, |
{"msub", 0}, |
{"msub", 0}, |
Line 99 static const struct htmldata htmltags[TAG_MAX] = { |
|
Line 107 static const struct htmldata htmltags[TAG_MAX] = { |
|
{"mover", 0}, |
{"mover", 0}, |
}; |
}; |
|
|
static const char *const roffscales[SCALE_MAX] = { |
/* Avoid duplicate HTML id= attributes. */ |
"cm", /* SCALE_CM */ |
static struct ohash id_unique; |
"in", /* SCALE_IN */ |
|
"pc", /* SCALE_PC */ |
|
"pt", /* SCALE_PT */ |
|
"em", /* SCALE_EM */ |
|
"em", /* SCALE_MM */ |
|
"ex", /* SCALE_EN */ |
|
"ex", /* SCALE_BU */ |
|
"em", /* SCALE_VS */ |
|
"ex", /* SCALE_FS */ |
|
}; |
|
|
|
static void a2width(const char *, struct roffsu *); |
static void html_reset_internal(struct html *); |
static void print_byte(struct html *, char); |
static void print_byte(struct html *, char); |
static void print_endword(struct html *); |
static void print_endword(struct html *); |
static void print_indent(struct html *); |
static void print_indent(struct html *); |
Line 122 static void print_ctag(struct html *, struct tag *); |
|
Line 120 static void print_ctag(struct html *, struct tag *); |
|
static int print_escape(struct html *, char); |
static int print_escape(struct html *, char); |
static int print_encode(struct html *, const char *, const char *, int); |
static int print_encode(struct html *, const char *, const char *, int); |
static void print_href(struct html *, const char *, const char *, int); |
static void print_href(struct html *, const char *, const char *, int); |
static void print_metaf(struct html *, enum mandoc_esc); |
static void print_metaf(struct html *); |
|
|
|
|
void * |
void * |
Line 134 html_alloc(const struct manoutput *outopts) |
|
Line 132 html_alloc(const struct manoutput *outopts) |
|
|
|
h->tag = NULL; |
h->tag = NULL; |
h->style = outopts->style; |
h->style = outopts->style; |
h->base_man = outopts->man; |
if ((h->base_man1 = outopts->man) == NULL) |
|
h->base_man2 = NULL; |
|
else if ((h->base_man2 = strchr(h->base_man1, ';')) != NULL) |
|
*h->base_man2++ = '\0'; |
h->base_includes = outopts->includes; |
h->base_includes = outopts->includes; |
if (outopts->fragment) |
if (outopts->fragment) |
h->oflags |= HTML_FRAGMENT; |
h->oflags |= HTML_FRAGMENT; |
|
if (outopts->toc) |
|
h->oflags |= HTML_TOC; |
|
|
|
mandoc_ohash_init(&id_unique, 4, 0); |
|
|
return h; |
return h; |
} |
} |
|
|
void |
static void |
html_free(void *p) |
html_reset_internal(struct html *h) |
{ |
{ |
struct tag *tag; |
struct tag *tag; |
struct html *h; |
char *cp; |
|
unsigned int slot; |
|
|
h = (struct html *)p; |
|
|
|
while ((tag = h->tag) != NULL) { |
while ((tag = h->tag) != NULL) { |
h->tag = tag->next; |
h->tag = tag->next; |
free(tag); |
free(tag); |
} |
} |
|
cp = ohash_first(&id_unique, &slot); |
|
while (cp != NULL) { |
|
free(cp); |
|
cp = ohash_next(&id_unique, &slot); |
|
} |
|
ohash_delete(&id_unique); |
|
} |
|
|
free(h); |
void |
|
html_reset(void *p) |
|
{ |
|
html_reset_internal(p); |
|
mandoc_ohash_init(&id_unique, 4, 0); |
} |
} |
|
|
void |
void |
|
html_free(void *p) |
|
{ |
|
html_reset_internal(p); |
|
free(p); |
|
} |
|
|
|
void |
print_gen_head(struct html *h) |
print_gen_head(struct html *h) |
{ |
{ |
struct tag *t; |
struct tag *t; |
|
|
print_otag(h, TAG_META, "?", "charset", "utf-8"); |
print_otag(h, TAG_META, "?", "charset", "utf-8"); |
|
if (h->style != NULL) { |
|
print_otag(h, TAG_LINK, "?h??", "rel", "stylesheet", |
|
h->style, "type", "text/css", "media", "all"); |
|
return; |
|
} |
|
|
/* |
/* |
* Print a default style-sheet. |
* Print a minimal embedded style sheet. |
*/ |
*/ |
|
|
t = print_otag(h, TAG_STYLE, ""); |
t = print_otag(h, TAG_STYLE, ""); |
Line 176 print_gen_head(struct html *h) |
|
Line 203 print_gen_head(struct html *h) |
|
print_endline(h); |
print_endline(h); |
print_text(h, "td.head-vol { text-align: center; }"); |
print_text(h, "td.head-vol { text-align: center; }"); |
print_endline(h); |
print_endline(h); |
print_text(h, "div.Pp { margin: 1ex 0ex; }"); |
print_text(h, ".Nd, .Bf, .Op { display: inline; }"); |
|
print_endline(h); |
|
print_text(h, ".Pa, .Ad { font-style: italic; }"); |
|
print_endline(h); |
|
print_text(h, ".Ms { font-weight: bold; }"); |
|
print_endline(h); |
|
print_text(h, ".Bl-diag "); |
|
print_byte(h, '>'); |
|
print_text(h, " dt { font-weight: bold; }"); |
|
print_endline(h); |
|
print_text(h, "code.Nm, .Fl, .Cm, .Ic, code.In, .Fd, .Fn, .Cd " |
|
"{ font-weight: bold; font-family: inherit; }"); |
print_tagq(h, t); |
print_tagq(h, t); |
|
|
if (h->style) |
|
print_otag(h, TAG_LINK, "?h??", "rel", "stylesheet", |
|
h->style, "type", "text/css", "media", "all"); |
|
} |
} |
|
|
static void |
int |
print_metaf(struct html *h, enum mandoc_esc deco) |
html_setfont(struct html *h, enum mandoc_esc font) |
{ |
{ |
enum htmlfont font; |
switch (font) { |
|
|
switch (deco) { |
|
case ESCAPE_FONTPREV: |
case ESCAPE_FONTPREV: |
font = h->metal; |
font = h->metal; |
break; |
break; |
case ESCAPE_FONTITALIC: |
case ESCAPE_FONTITALIC: |
font = HTMLFONT_ITALIC; |
|
break; |
|
case ESCAPE_FONTBOLD: |
case ESCAPE_FONTBOLD: |
font = HTMLFONT_BOLD; |
|
break; |
|
case ESCAPE_FONTBI: |
case ESCAPE_FONTBI: |
font = HTMLFONT_BI; |
case ESCAPE_FONTCW: |
|
case ESCAPE_FONTROMAN: |
break; |
break; |
case ESCAPE_FONT: |
case ESCAPE_FONT: |
case ESCAPE_FONTROMAN: |
font = ESCAPE_FONTROMAN; |
font = HTMLFONT_NONE; |
|
break; |
break; |
default: |
default: |
abort(); |
return 0; |
} |
} |
|
h->metal = h->metac; |
|
h->metac = font; |
|
return 1; |
|
} |
|
|
|
static void |
|
print_metaf(struct html *h) |
|
{ |
if (h->metaf) { |
if (h->metaf) { |
print_tagq(h, h->metaf); |
print_tagq(h, h->metaf); |
h->metaf = NULL; |
h->metaf = NULL; |
} |
} |
|
switch (h->metac) { |
h->metal = h->metac; |
case ESCAPE_FONTITALIC: |
h->metac = font; |
|
|
|
switch (font) { |
|
case HTMLFONT_ITALIC: |
|
h->metaf = print_otag(h, TAG_I, ""); |
h->metaf = print_otag(h, TAG_I, ""); |
break; |
break; |
case HTMLFONT_BOLD: |
case ESCAPE_FONTBOLD: |
h->metaf = print_otag(h, TAG_B, ""); |
h->metaf = print_otag(h, TAG_B, ""); |
break; |
break; |
case HTMLFONT_BI: |
case ESCAPE_FONTBI: |
h->metaf = print_otag(h, TAG_B, ""); |
h->metaf = print_otag(h, TAG_B, ""); |
print_otag(h, TAG_I, ""); |
print_otag(h, TAG_I, ""); |
break; |
break; |
|
case ESCAPE_FONTCW: |
|
h->metaf = print_otag(h, TAG_SPAN, "c", "Li"); |
|
break; |
default: |
default: |
break; |
break; |
} |
} |
} |
} |
|
|
int |
void |
html_strlen(const char *cp) |
html_close_paragraph(struct html *h) |
{ |
{ |
size_t rsz; |
struct tag *t; |
int skip, sz; |
|
|
|
/* |
for (t = h->tag; t != NULL && t->closed == 0; t = t->next) { |
* Account for escaped sequences within string length |
switch(t->tag) { |
* calculations. This follows the logic in term_strlen() as we |
case TAG_P: |
* must calculate the width of produced strings. |
case TAG_PRE: |
* Assume that characters are always width of "1". This is |
print_tagq(h, t); |
* hacky, but it gets the job done for approximation of widths. |
break; |
*/ |
case TAG_A: |
|
print_tagq(h, t); |
sz = 0; |
continue; |
skip = 0; |
default: |
while (1) { |
continue; |
rsz = strcspn(cp, "\\"); |
|
if (rsz) { |
|
cp += rsz; |
|
if (skip) { |
|
skip = 0; |
|
rsz--; |
|
} |
|
sz += rsz; |
|
} |
} |
if ('\0' == *cp) |
break; |
|
} |
|
} |
|
|
|
/* |
|
* ROFF_nf switches to no-fill mode, ROFF_fi to fill mode. |
|
* TOKEN_NONE does not switch. The old mode is returned. |
|
*/ |
|
enum roff_tok |
|
html_fillmode(struct html *h, enum roff_tok want) |
|
{ |
|
struct tag *t; |
|
enum roff_tok had; |
|
|
|
for (t = h->tag; t != NULL; t = t->next) |
|
if (t->tag == TAG_PRE) |
break; |
break; |
cp++; |
|
switch (mandoc_escape(&cp, NULL, NULL)) { |
had = t == NULL ? ROFF_fi : ROFF_nf; |
case ESCAPE_ERROR: |
|
return sz; |
if (want != had) { |
case ESCAPE_UNICODE: |
switch (want) { |
case ESCAPE_NUMBERED: |
case ROFF_fi: |
case ESCAPE_SPECIAL: |
print_tagq(h, t); |
case ESCAPE_OVERSTRIKE: |
|
if (skip) |
|
skip = 0; |
|
else |
|
sz++; |
|
break; |
break; |
case ESCAPE_SKIPCHAR: |
case ROFF_nf: |
skip = 1; |
html_close_paragraph(h); |
|
print_otag(h, TAG_PRE, ""); |
break; |
break; |
default: |
case TOKEN_NONE: |
break; |
break; |
|
default: |
|
abort(); |
} |
} |
} |
} |
return sz; |
return had; |
} |
} |
|
|
|
char * |
|
html_make_id(const struct roff_node *n, int unique) |
|
{ |
|
const struct roff_node *nch; |
|
char *buf, *bufs, *cp; |
|
unsigned int slot; |
|
int suffix; |
|
|
|
for (nch = n->child; nch != NULL; nch = nch->next) |
|
if (nch->type != ROFFT_TEXT) |
|
return NULL; |
|
|
|
buf = NULL; |
|
deroff(&buf, n); |
|
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 |
|
*/ |
|
|
|
for (cp = buf; *cp != '\0'; cp++) |
|
if (isalnum((unsigned char)*cp) == 0 && |
|
strchr("!$&'()*+,-./:;=?@_~", *cp) == NULL) |
|
*cp = '_'; |
|
|
|
if (unique == 0) |
|
return buf; |
|
|
|
/* Avoid duplicate HTML id= attributes. */ |
|
|
|
bufs = NULL; |
|
suffix = 1; |
|
slot = ohash_qlookup(&id_unique, buf); |
|
cp = ohash_find(&id_unique, slot); |
|
if (cp != NULL) { |
|
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 int |
print_escape(struct html *h, char c) |
print_escape(struct html *h, char c) |
{ |
{ |
|
|
print_encode(struct html *h, const char *p, const char *pend, int norecurse) |
print_encode(struct html *h, const char *p, const char *pend, int norecurse) |
{ |
{ |
char numbuf[16]; |
char numbuf[16]; |
size_t sz; |
|
int c, len, nospace; |
|
const char *seq; |
const char *seq; |
|
size_t sz; |
|
int c, len, breakline, nospace; |
enum mandoc_esc esc; |
enum mandoc_esc esc; |
static const char rejs[9] = { '\\', '<', '>', '&', '"', |
static const char rejs[10] = { ' ', '\\', '<', '>', '&', '"', |
ASCII_NBRSP, ASCII_HYPH, ASCII_BREAK, '\0' }; |
ASCII_NBRSP, ASCII_HYPH, ASCII_BREAK, '\0' }; |
|
|
if (pend == NULL) |
if (pend == NULL) |
pend = strchr(p, '\0'); |
pend = strchr(p, '\0'); |
|
|
|
breakline = 0; |
nospace = 0; |
nospace = 0; |
|
|
while (p < pend) { |
while (p < pend) { |
Line 340 print_encode(struct html *h, const char *p, const char |
|
Line 435 print_encode(struct html *h, const char *p, const char |
|
} |
} |
|
|
for (sz = strcspn(p, rejs); sz-- && p < pend; p++) |
for (sz = strcspn(p, rejs); sz-- && p < pend; p++) |
if (*p == ' ') |
print_byte(h, *p); |
print_endword(h); |
|
else |
|
print_byte(h, *p); |
|
|
|
|
if (breakline && |
|
(p >= pend || *p == ' ' || *p == ASCII_NBRSP)) { |
|
print_otag(h, TAG_BR, ""); |
|
breakline = 0; |
|
while (p < pend && (*p == ' ' || *p == ASCII_NBRSP)) |
|
p++; |
|
continue; |
|
} |
|
|
if (p >= pend) |
if (p >= pend) |
break; |
break; |
|
|
|
if (*p == ' ') { |
|
print_endword(h); |
|
p++; |
|
continue; |
|
} |
|
|
if (print_escape(h, *p++)) |
if (print_escape(h, *p++)) |
continue; |
continue; |
|
|
esc = mandoc_escape(&p, &seq, &len); |
esc = mandoc_escape(&p, &seq, &len); |
if (ESCAPE_ERROR == esc) |
|
break; |
|
|
|
switch (esc) { |
switch (esc) { |
case ESCAPE_FONT: |
case ESCAPE_FONT: |
case ESCAPE_FONTPREV: |
case ESCAPE_FONTPREV: |
case ESCAPE_FONTBOLD: |
case ESCAPE_FONTBOLD: |
case ESCAPE_FONTITALIC: |
case ESCAPE_FONTITALIC: |
case ESCAPE_FONTBI: |
case ESCAPE_FONTBI: |
|
case ESCAPE_FONTCW: |
case ESCAPE_FONTROMAN: |
case ESCAPE_FONTROMAN: |
if (0 == norecurse) |
if (0 == norecurse) { |
print_metaf(h, esc); |
h->flags |= HTML_NOSPACE; |
|
if (html_setfont(h, esc)) |
|
print_metaf(h); |
|
h->flags &= ~HTML_NOSPACE; |
|
} |
continue; |
continue; |
case ESCAPE_SKIPCHAR: |
case ESCAPE_SKIPCHAR: |
h->flags |= HTML_SKIPCHAR; |
h->flags |= HTML_SKIPCHAR; |
continue; |
continue; |
|
case ESCAPE_ERROR: |
|
continue; |
default: |
default: |
break; |
break; |
} |
} |
Line 392 print_encode(struct html *h, const char *p, const char |
|
Line 503 print_encode(struct html *h, const char *p, const char |
|
if (c <= 0) |
if (c <= 0) |
continue; |
continue; |
break; |
break; |
|
case ESCAPE_UNDEF: |
|
c = *seq; |
|
break; |
|
case ESCAPE_DEVICE: |
|
print_word(h, "html"); |
|
continue; |
|
case ESCAPE_BREAK: |
|
breakline = 1; |
|
continue; |
case ESCAPE_NOSPACE: |
case ESCAPE_NOSPACE: |
if ('\0' == *p) |
if ('\0' == *p) |
nospace = 1; |
nospace = 1; |
Line 408 print_encode(struct html *h, const char *p, const char |
|
Line 528 print_encode(struct html *h, const char *p, const char |
|
(c > 0x7E && c < 0xA0)) |
(c > 0x7E && c < 0xA0)) |
c = 0xFFFD; |
c = 0xFFFD; |
if (c > 0x7E) { |
if (c > 0x7E) { |
(void)snprintf(numbuf, sizeof(numbuf), "&#%d;", c); |
(void)snprintf(numbuf, sizeof(numbuf), "&#x%.4X;", c); |
print_word(h, numbuf); |
print_word(h, numbuf); |
} else if (print_escape(h, c) == 0) |
} else if (print_escape(h, c) == 0) |
print_byte(h, c); |
print_byte(h, c); |
Line 420 print_encode(struct html *h, const char *p, const char |
|
Line 540 print_encode(struct html *h, const char *p, const char |
|
static void |
static void |
print_href(struct html *h, const char *name, const char *sec, int man) |
print_href(struct html *h, const char *name, const char *sec, int man) |
{ |
{ |
|
struct stat sb; |
const char *p, *pp; |
const char *p, *pp; |
|
char *filename; |
|
|
pp = man ? h->base_man : h->base_includes; |
if (man) { |
|
pp = h->base_man1; |
|
if (h->base_man2 != NULL) { |
|
mandoc_asprintf(&filename, "%s.%s", name, sec); |
|
if (stat(filename, &sb) == -1) |
|
pp = h->base_man2; |
|
free(filename); |
|
} |
|
} else |
|
pp = h->base_includes; |
|
|
while ((p = strchr(pp, '%')) != NULL) { |
while ((p = strchr(pp, '%')) != NULL) { |
print_encode(h, pp, p, 1); |
print_encode(h, pp, p, 1); |
if (man && p[1] == 'S') { |
if (man && p[1] == 'S') { |
|
|
print_otag(struct html *h, enum htmltag tag, const char *fmt, ...) |
print_otag(struct html *h, enum htmltag tag, const char *fmt, ...) |
{ |
{ |
va_list ap; |
va_list ap; |
struct roffsu mysu, *su; |
|
char numbuf[16]; |
|
struct tag *t; |
struct tag *t; |
const char *attr; |
const char *attr; |
char *arg1, *arg2; |
char *arg1, *arg2; |
double v; |
int style_written, tflags; |
int i, have_style, tflags; |
|
|
|
tflags = htmltags[tag].flags; |
tflags = htmltags[tag].flags; |
|
|
|
/* Flow content is not allowed in phrasing context. */ |
|
|
|
if ((tflags & HTML_INPHRASE) == 0) { |
|
for (t = h->tag; t != NULL; t = t->next) { |
|
if (t->closed) |
|
continue; |
|
assert((htmltags[t->tag].flags & HTML_TOPHRASE) == 0); |
|
break; |
|
} |
|
} |
|
|
/* Push this tag onto the stack of open scopes. */ |
/* Push this tag onto the stack of open scopes. */ |
|
|
if ((tflags & HTML_NOSTACK) == 0) { |
if ((tflags & HTML_NOSTACK) == 0) { |
t = mandoc_malloc(sizeof(struct tag)); |
t = mandoc_malloc(sizeof(struct tag)); |
t->tag = tag; |
t->tag = tag; |
t->next = h->tag; |
t->next = h->tag; |
|
t->refcnt = 0; |
|
t->closed = 0; |
h->tag = t; |
h->tag = t; |
} else |
} else |
t = NULL; |
t = NULL; |
Line 471 print_otag(struct html *h, enum htmltag tag, const cha |
|
Line 613 print_otag(struct html *h, enum htmltag tag, const cha |
|
print_indent(h); |
print_indent(h); |
else if ((h->flags & HTML_NOSPACE) == 0) { |
else if ((h->flags & HTML_NOSPACE) == 0) { |
if (h->flags & HTML_KEEP) |
if (h->flags & HTML_KEEP) |
print_word(h, " "); |
print_word(h, " "); |
else { |
else { |
if (h->flags & HTML_PREKEEP) |
if (h->flags & HTML_PREKEEP) |
h->flags |= HTML_KEEP; |
h->flags |= HTML_KEEP; |
Line 491 print_otag(struct html *h, enum htmltag tag, const cha |
|
Line 633 print_otag(struct html *h, enum htmltag tag, const cha |
|
|
|
va_start(ap, fmt); |
va_start(ap, fmt); |
|
|
have_style = 0; |
while (*fmt != '\0' && *fmt != 's') { |
while (*fmt != '\0') { |
|
if (*fmt == 's') { |
|
have_style = 1; |
|
fmt++; |
|
break; |
|
} |
|
|
|
/* Parse a non-style attribute and its arguments. */ |
/* Parse attributes and arguments. */ |
|
|
arg1 = va_arg(ap, char *); |
arg1 = va_arg(ap, char *); |
|
arg2 = NULL; |
switch (*fmt++) { |
switch (*fmt++) { |
case 'c': |
case 'c': |
attr = "class"; |
attr = "class"; |
Line 519 print_otag(struct html *h, enum htmltag tag, const cha |
|
Line 656 print_otag(struct html *h, enum htmltag tag, const cha |
|
default: |
default: |
abort(); |
abort(); |
} |
} |
arg2 = NULL; |
|
if (*fmt == 'M') |
if (*fmt == 'M') |
arg2 = va_arg(ap, char *); |
arg2 = va_arg(ap, char *); |
if (arg1 == NULL) |
if (arg1 == NULL) |
continue; |
continue; |
|
|
/* Print the non-style attributes. */ |
/* Print the attributes. */ |
|
|
print_byte(h, ' '); |
print_byte(h, ' '); |
print_word(h, attr); |
print_word(h, attr); |
print_byte(h, '='); |
print_byte(h, '='); |
print_byte(h, '"'); |
print_byte(h, '"'); |
switch (*fmt) { |
switch (*fmt) { |
case 'M': |
|
print_href(h, arg1, arg2, 1); |
|
fmt++; |
|
break; |
|
case 'I': |
case 'I': |
print_href(h, arg1, NULL, 0); |
print_href(h, arg1, NULL, 0); |
fmt++; |
fmt++; |
break; |
break; |
|
case 'M': |
|
print_href(h, arg1, arg2, 1); |
|
fmt++; |
|
break; |
case 'R': |
case 'R': |
print_byte(h, '#'); |
print_byte(h, '#'); |
|
print_encode(h, arg1, NULL, 1); |
fmt++; |
fmt++; |
/* FALLTHROUGH */ |
break; |
default: |
default: |
print_encode(h, arg1, NULL, 1); |
print_encode(h, arg1, NULL, 1); |
break; |
break; |
Line 551 print_otag(struct html *h, enum htmltag tag, const cha |
|
Line 688 print_otag(struct html *h, enum htmltag tag, const cha |
|
print_byte(h, '"'); |
print_byte(h, '"'); |
} |
} |
|
|
/* Print out styles. */ |
style_written = 0; |
|
while (*fmt++ == 's') { |
while (*fmt != '\0') { |
arg1 = va_arg(ap, char *); |
arg1 = NULL; |
arg2 = va_arg(ap, char *); |
su = NULL; |
if (arg2 == NULL) |
|
|
/* First letter: input argument type. */ |
|
|
|
switch (*fmt++) { |
|
case 'h': |
|
i = va_arg(ap, int); |
|
su = &mysu; |
|
SCALE_HS_INIT(su, i); |
|
break; |
|
case 's': |
|
arg1 = va_arg(ap, char *); |
|
break; |
|
case 'u': |
|
su = va_arg(ap, struct roffsu *); |
|
break; |
|
case 'v': |
|
i = va_arg(ap, int); |
|
su = &mysu; |
|
SCALE_VS_INIT(su, i); |
|
break; |
|
case 'w': |
|
case 'W': |
|
if ((arg2 = va_arg(ap, char *)) == NULL) |
|
break; |
|
su = &mysu; |
|
a2width(arg2, su); |
|
if (fmt[-1] == 'W') |
|
su->scale *= -1.0; |
|
break; |
|
default: |
|
abort(); |
|
} |
|
|
|
/* Second letter: style name. */ |
|
|
|
switch (*fmt++) { |
|
case 'b': |
|
attr = "margin-bottom"; |
|
break; |
|
case 'h': |
|
attr = "height"; |
|
break; |
|
case 'i': |
|
attr = "text-indent"; |
|
break; |
|
case 'l': |
|
attr = "margin-left"; |
|
break; |
|
case 't': |
|
attr = "margin-top"; |
|
break; |
|
case 'w': |
|
attr = "width"; |
|
break; |
|
case 'W': |
|
attr = "min-width"; |
|
break; |
|
case '?': |
|
attr = arg1; |
|
arg1 = va_arg(ap, char *); |
|
break; |
|
default: |
|
abort(); |
|
} |
|
if (su == NULL && arg1 == NULL) |
|
continue; |
continue; |
|
print_byte(h, ' '); |
if (have_style == 1) |
if (style_written == 0) { |
print_word(h, " style=\""); |
print_word(h, "style=\""); |
else |
style_written = 1; |
print_byte(h, ' '); |
} |
print_word(h, attr); |
print_word(h, arg1); |
print_byte(h, ':'); |
print_byte(h, ':'); |
print_byte(h, ' '); |
print_byte(h, ' '); |
if (su != NULL) { |
print_word(h, arg2); |
v = su->scale; |
|
if (su->unit == SCALE_MM && (v /= 100.0) == 0.0) |
|
v = 1.0; |
|
else if (su->unit == SCALE_BU) |
|
v /= 24.0; |
|
(void)snprintf(numbuf, sizeof(numbuf), "%.2f", v); |
|
print_word(h, numbuf); |
|
print_word(h, roffscales[su->unit]); |
|
} else |
|
print_word(h, arg1); |
|
print_byte(h, ';'); |
print_byte(h, ';'); |
have_style = 2; |
|
} |
} |
if (have_style == 2) |
if (style_written) |
print_byte(h, '"'); |
print_byte(h, '"'); |
|
|
va_end(ap); |
va_end(ap); |
|
|
/* Accommodate for "well-formed" singleton escaping. */ |
/* Accommodate for "well-formed" singleton escaping. */ |
|
|
if (HTML_AUTOCLOSE & htmltags[tag].flags) |
if (htmltags[tag].flags & HTML_NOSTACK) |
print_byte(h, '/'); |
print_byte(h, '/'); |
|
|
print_byte(h, '>'); |
print_byte(h, '>'); |
Line 674 print_ctag(struct html *h, struct tag *tag) |
|
Line 735 print_ctag(struct html *h, struct tag *tag) |
|
{ |
{ |
int tflags; |
int tflags; |
|
|
/* |
if (tag->closed == 0) { |
* Remember to close out and nullify the current |
tag->closed = 1; |
* meta-font and table, if applicable. |
if (tag == h->metaf) |
*/ |
h->metaf = NULL; |
if (tag == h->metaf) |
if (tag == h->tblt) |
h->metaf = NULL; |
h->tblt = NULL; |
if (tag == h->tblt) |
|
h->tblt = NULL; |
|
|
|
tflags = htmltags[tag->tag].flags; |
tflags = htmltags[tag->tag].flags; |
|
if (tflags & HTML_INDENT) |
if (tflags & HTML_INDENT) |
h->indent--; |
h->indent--; |
if (tflags & HTML_NOINDENT) |
if (tflags & HTML_NOINDENT) |
h->noindent--; |
h->noindent--; |
if (tflags & HTML_NLEND) |
if (tflags & HTML_NLEND) |
print_endline(h); |
print_endline(h); |
print_indent(h); |
print_indent(h); |
print_byte(h, '<'); |
print_byte(h, '<'); |
print_byte(h, '/'); |
print_byte(h, '/'); |
print_word(h, htmltags[tag->tag].name); |
print_word(h, htmltags[tag->tag].name); |
print_byte(h, '>'); |
print_byte(h, '>'); |
if (tflags & HTML_NLAFTER) |
if (tflags & HTML_NLAFTER) |
print_endline(h); |
print_endline(h); |
} |
|
if (tag->refcnt == 0) { |
h->tag = tag->next; |
h->tag = tag->next; |
free(tag); |
free(tag); |
|
} |
} |
} |
|
|
void |
void |
Line 711 print_gen_decls(struct html *h) |
|
Line 771 print_gen_decls(struct html *h) |
|
} |
} |
|
|
void |
void |
|
print_gen_comment(struct html *h, struct roff_node *n) |
|
{ |
|
int wantblank; |
|
|
|
print_word(h, "<!-- This is an automatically generated file." |
|
" Do not edit."); |
|
h->indent = 1; |
|
wantblank = 0; |
|
while (n != NULL && n->type == ROFFT_COMMENT) { |
|
if (strstr(n->string, "-->") == NULL && |
|
(wantblank || *n->string != '\0')) { |
|
print_endline(h); |
|
print_indent(h); |
|
print_word(h, n->string); |
|
wantblank = *n->string != '\0'; |
|
} |
|
n = n->next; |
|
} |
|
if (wantblank) |
|
print_endline(h); |
|
print_word(h, " -->"); |
|
print_endline(h); |
|
h->indent = 0; |
|
} |
|
|
|
void |
print_text(struct html *h, const char *word) |
print_text(struct html *h, const char *word) |
{ |
{ |
if (h->col && (h->flags & HTML_NOSPACE) == 0) { |
if (h->col && (h->flags & HTML_NOSPACE) == 0) { |
Line 719 print_text(struct html *h, const char *word) |
|
Line 805 print_text(struct html *h, const char *word) |
|
h->flags |= HTML_KEEP; |
h->flags |= HTML_KEEP; |
print_endword(h); |
print_endword(h); |
} else |
} else |
print_word(h, " "); |
print_word(h, " "); |
} |
} |
|
|
assert(NULL == h->metaf); |
assert(h->metaf == NULL); |
switch (h->metac) { |
print_metaf(h); |
case HTMLFONT_ITALIC: |
print_indent(h); |
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: |
|
print_indent(h); |
|
break; |
|
} |
|
|
|
assert(word); |
|
if ( ! print_encode(h, word, NULL, 0)) { |
if ( ! print_encode(h, word, NULL, 0)) { |
if ( ! (h->flags & HTML_NONOSPACE)) |
if ( ! (h->flags & HTML_NONOSPACE)) |
h->flags &= ~HTML_NOSPACE; |
h->flags &= ~HTML_NOSPACE; |
Line 747 print_text(struct html *h, const char *word) |
|
Line 818 print_text(struct html *h, const char *word) |
|
} else |
} else |
h->flags |= HTML_NOSPACE | HTML_NONEWLINE; |
h->flags |= HTML_NOSPACE | HTML_NONEWLINE; |
|
|
if (h->metaf) { |
if (h->metaf != NULL) { |
print_tagq(h, h->metaf); |
print_tagq(h, h->metaf); |
h->metaf = NULL; |
h->metaf = NULL; |
} |
} |
Line 758 print_text(struct html *h, const char *word) |
|
Line 829 print_text(struct html *h, const char *word) |
|
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 *this, *next; |
|
|
while ((tag = h->tag) != NULL) { |
for (this = h->tag; this != NULL; this = next) { |
print_ctag(h, tag); |
next = this == until ? NULL : this->next; |
if (until && tag == until) |
print_ctag(h, this); |
return; |
|
} |
} |
} |
} |
|
|
|
/* |
|
* Close out all open elements up to but excluding suntil. |
|
* Note that a paragraph just inside stays open together with it |
|
* because paragraphs include subsequent phrasing content. |
|
*/ |
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 *this, *next; |
|
|
while ((tag = h->tag) != NULL) { |
for (this = h->tag; this != NULL; this = next) { |
if (suntil && tag == suntil) |
next = this->next; |
return; |
if (this == suntil || (next == suntil && |
print_ctag(h, tag); |
(this->tag == TAG_P || this->tag == TAG_PRE))) |
|
break; |
|
print_ctag(h, this); |
} |
} |
} |
} |
|
|
void |
|
print_paragraph(struct html *h) |
|
{ |
|
struct tag *t; |
|
|
|
t = print_otag(h, TAG_DIV, "c", "Pp"); |
|
print_tagq(h, t); |
|
} |
|
|
|
|
|
/*********************************************************************** |
/*********************************************************************** |
* Low level output functions. |
* Low level output functions. |
* They implement line breaking using a short static buffer. |
* They implement line breaking using a short static buffer. |
Line 900 print_word(struct html *h, const char *cp) |
|
Line 968 print_word(struct html *h, const char *cp) |
|
{ |
{ |
while (*cp != '\0') |
while (*cp != '\0') |
print_byte(h, *cp++); |
print_byte(h, *cp++); |
} |
|
|
|
/* |
|
* Calculate the scaling unit passed in a `-width' argument. This uses |
|
* either a native scaling unit (e.g., 1i, 2m) or the string length of |
|
* the value. |
|
*/ |
|
static void |
|
a2width(const char *p, struct roffsu *su) |
|
{ |
|
if (a2roffsu(p, su, SCALE_MAX) < 2) { |
|
su->unit = SCALE_EN; |
|
su->scale = html_strlen(p); |
|
} else if (su->scale < 0.0) |
|
su->scale = 0.0; |
|
} |
} |