version 1.166, 2014/09/27 09:03:24 |
version 1.267, 2020/04/08 11:56:03 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011-2015, 2017-2020 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv> |
* |
* |
* 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. |
|
* |
|
* Common functions for mandoc(1) HTML formatters. |
|
* For use by individual formatters and by the main program. |
*/ |
*/ |
#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 "libmandoc.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 "main.h" |
#include "main.h" |
|
|
struct htmldata { |
struct htmldata { |
const char *name; |
const char *name; |
int flags; |
int flags; |
#define HTML_CLRLINE (1 << 0) |
#define HTML_INPHRASE (1 << 0) /* Can appear in phrasing context. */ |
#define HTML_NOSTACK (1 << 1) |
#define HTML_TOPHRASE (1 << 1) /* Establishes phrasing context. */ |
#define HTML_AUTOCLOSE (1 << 2) /* Tag has auto-closure. */ |
#define HTML_NOSTACK (1 << 2) /* Does not have an end tag. */ |
|
#define HTML_NLBEFORE (1 << 3) /* Output line break before opening. */ |
|
#define HTML_NLBEGIN (1 << 4) /* Output line break after opening. */ |
|
#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_NLINSIDE (HTML_NLBEGIN | HTML_NLEND) |
|
#define HTML_NLALL (HTML_NLAROUND | HTML_NLINSIDE) |
|
#define HTML_INDENT (1 << 7) /* Indent content by two spaces. */ |
|
#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_CLRLINE}, /* TAG_HTML */ |
{"html", HTML_NLALL}, |
{"head", HTML_CLRLINE}, /* TAG_HEAD */ |
{"head", HTML_NLALL | HTML_INDENT}, |
{"body", HTML_CLRLINE}, /* TAG_BODY */ |
{"meta", HTML_NOSTACK | HTML_NLALL}, |
{"meta", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_META */ |
{"link", HTML_NOSTACK | HTML_NLALL}, |
{"title", HTML_CLRLINE}, /* TAG_TITLE */ |
{"style", HTML_NLALL | HTML_INDENT}, |
{"div", HTML_CLRLINE}, /* TAG_DIV */ |
{"title", HTML_NLAROUND}, |
{"h1", 0}, /* TAG_H1 */ |
{"body", HTML_NLALL}, |
{"h2", 0}, /* TAG_H2 */ |
{"div", HTML_NLAROUND}, |
{"span", 0}, /* TAG_SPAN */ |
{"section", HTML_NLALL}, |
{"link", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_LINK */ |
{"table", HTML_NLALL | HTML_INDENT}, |
{"br", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_BR */ |
{"tr", HTML_NLALL | HTML_INDENT}, |
{"a", 0}, /* TAG_A */ |
{"td", HTML_NLAROUND}, |
{"table", HTML_CLRLINE}, /* TAG_TABLE */ |
{"li", HTML_NLAROUND | HTML_INDENT}, |
{"tbody", HTML_CLRLINE}, /* TAG_TBODY */ |
{"ul", HTML_NLALL | HTML_INDENT}, |
{"col", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_COL */ |
{"ol", HTML_NLALL | HTML_INDENT}, |
{"tr", HTML_CLRLINE}, /* TAG_TR */ |
{"dl", HTML_NLALL | HTML_INDENT}, |
{"td", HTML_CLRLINE}, /* TAG_TD */ |
{"dt", HTML_NLAROUND}, |
{"li", HTML_CLRLINE}, /* TAG_LI */ |
{"dd", HTML_NLAROUND | HTML_INDENT}, |
{"ul", HTML_CLRLINE}, /* TAG_UL */ |
{"h1", HTML_TOPHRASE | HTML_NLAROUND}, |
{"ol", HTML_CLRLINE}, /* TAG_OL */ |
{"h2", HTML_TOPHRASE | HTML_NLAROUND}, |
{"dl", HTML_CLRLINE}, /* TAG_DL */ |
{"p", HTML_TOPHRASE | HTML_NLAROUND | HTML_INDENT}, |
{"dt", HTML_CLRLINE}, /* TAG_DT */ |
{"pre", HTML_TOPHRASE | HTML_NLALL | HTML_NOINDENT}, |
{"dd", HTML_CLRLINE}, /* TAG_DD */ |
{"a", HTML_INPHRASE | HTML_TOPHRASE}, |
{"blockquote", HTML_CLRLINE}, /* TAG_BLOCKQUOTE */ |
{"b", HTML_INPHRASE | HTML_TOPHRASE}, |
{"p", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_P */ |
{"cite", HTML_INPHRASE | HTML_TOPHRASE}, |
{"pre", HTML_CLRLINE }, /* TAG_PRE */ |
{"code", HTML_INPHRASE | HTML_TOPHRASE}, |
{"b", 0 }, /* TAG_B */ |
{"i", HTML_INPHRASE | HTML_TOPHRASE}, |
{"i", 0 }, /* TAG_I */ |
{"small", HTML_INPHRASE | HTML_TOPHRASE}, |
{"code", 0 }, /* TAG_CODE */ |
{"span", HTML_INPHRASE | HTML_TOPHRASE}, |
{"small", 0 }, /* TAG_SMALL */ |
{"var", HTML_INPHRASE | HTML_TOPHRASE}, |
{"style", HTML_CLRLINE}, /* TAG_STYLE */ |
{"br", HTML_INPHRASE | HTML_NOSTACK | HTML_NLALL}, |
|
{"mark", HTML_INPHRASE }, |
|
{"math", HTML_INPHRASE | 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 htmlattrs[ATTR_MAX] = { |
/* Avoid duplicate HTML id= attributes. */ |
"name", /* ATTR_NAME */ |
static struct ohash id_unique; |
"rel", /* ATTR_REL */ |
|
"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 roffscales[SCALE_MAX] = { |
static void html_reset_internal(struct html *); |
"cm", /* SCALE_CM */ |
static void print_byte(struct html *, char); |
"in", /* SCALE_IN */ |
static void print_endword(struct html *); |
"pc", /* SCALE_PC */ |
static void print_indent(struct html *); |
"pt", /* SCALE_PT */ |
static void print_word(struct html *, const char *); |
"em", /* SCALE_EM */ |
|
"em", /* SCALE_MM */ |
|
"ex", /* SCALE_EN */ |
|
"ex", /* SCALE_BU */ |
|
"em", /* SCALE_VS */ |
|
"ex", /* SCALE_FS */ |
|
}; |
|
|
|
static void bufncat(struct html *, const char *, size_t); |
static void print_ctag(struct html *, struct tag *); |
static void print_ctag(struct html *, enum htmltag); |
static int print_escape(struct html *, char); |
static int print_escape(char); |
static int print_encode(struct html *, const char *, const char *, int); |
static int print_encode(struct html *, 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 *); |
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[5]; |
|
char *v; |
|
|
|
toks[0] = "style"; |
|
toks[1] = "man"; |
|
toks[2] = "includes"; |
|
toks[3] = "fragment"; |
|
toks[4] = NULL; |
|
|
|
h = mandoc_calloc(1, sizeof(struct html)); |
h = mandoc_calloc(1, sizeof(struct html)); |
|
|
h->type = type; |
h->tag = NULL; |
h->tags.head = NULL; |
h->style = outopts->style; |
h->symtab = mchars_alloc(); |
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; |
|
if (outopts->fragment) |
|
h->oflags |= HTML_FRAGMENT; |
|
if (outopts->toc) |
|
h->oflags |= HTML_TOC; |
|
|
while (outopts && *outopts) |
mandoc_ohash_init(&id_unique, 4, 0); |
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; |
|
case 3: |
|
h->oflags |= HTML_FRAGMENT; |
|
break; |
|
default: |
|
break; |
|
} |
|
|
|
return(h); |
return h; |
} |
} |
|
|
void * |
static void |
html_alloc(char *outopts) |
html_reset_internal(struct html *h) |
{ |
{ |
|
struct tag *tag; |
|
char *cp; |
|
unsigned int slot; |
|
|
return(ml_alloc(outopts, HTML_HTML_4_01_STRICT)); |
while ((tag = h->tag) != NULL) { |
|
h->tag = tag->next; |
|
free(tag); |
|
} |
|
cp = ohash_first(&id_unique, &slot); |
|
while (cp != NULL) { |
|
free(cp); |
|
cp = ohash_next(&id_unique, &slot); |
|
} |
|
ohash_delete(&id_unique); |
} |
} |
|
|
void * |
void |
xhtml_alloc(char *outopts) |
html_reset(void *p) |
{ |
{ |
|
html_reset_internal(p); |
return(ml_alloc(outopts, HTML_XHTML_1_0_STRICT)); |
mandoc_ohash_init(&id_unique, 4, 0); |
} |
} |
|
|
void |
void |
html_free(void *p) |
html_free(void *p) |
{ |
{ |
struct tag *tag; |
html_reset_internal(p); |
struct html *h; |
free(p); |
|
|
h = (struct html *)p; |
|
|
|
while ((tag = h->tags.head) != NULL) { |
|
h->tags.head = tag->next; |
|
free(tag); |
|
} |
|
|
|
if (h->symtab) |
|
mchars_free(h->symtab); |
|
|
|
free(h); |
|
} |
} |
|
|
void |
void |
print_gen_head(struct html *h) |
print_gen_head(struct html *h) |
{ |
{ |
struct htmlpair tag[4]; |
|
struct tag *t; |
struct tag *t; |
|
|
tag[0].key = ATTR_CHARSET; |
print_otag(h, TAG_META, "?", "charset", "utf-8"); |
tag[0].val = "utf-8"; |
if (h->style != NULL) { |
print_otag(h, TAG_META, 1, tag); |
print_otag(h, TAG_LINK, "?h??", "rel", "stylesheet", |
|
h->style, "type", "text/css", "media", "all"); |
|
return; |
|
} |
|
|
t = print_otag(h, TAG_STYLE, 0, NULL); |
/* |
print_text(h, "table.head, table.foot { width: 100%; }\n"); |
* Print a minimal embedded style sheet. |
print_tagq(h, t); |
*/ |
|
|
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, ".Nd, .Bf, .Op { display: inline; }"); |
tag[3].val = "all"; |
print_endline(h); |
print_otag(h, TAG_LINK, 4, tag); |
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); |
} |
} |
|
|
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: |
/* FALLTHROUGH */ |
font = ESCAPE_FONTROMAN; |
case ESCAPE_FONTROMAN: |
|
font = HTMLFONT_NONE; |
|
break; |
break; |
default: |
default: |
abort(); |
return 0; |
/* NOTREACHED */ |
|
} |
} |
|
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; |
h->metaf = print_otag(h, TAG_I, ""); |
|
|
switch (font) { |
|
case HTMLFONT_ITALIC: |
|
h->metaf = print_otag(h, TAG_I, 0, NULL); |
|
break; |
break; |
case HTMLFONT_BOLD: |
case ESCAPE_FONTBOLD: |
h->metaf = print_otag(h, TAG_B, 0, NULL); |
h->metaf = print_otag(h, TAG_B, ""); |
break; |
break; |
case HTMLFONT_BI: |
case ESCAPE_FONTBI: |
h->metaf = print_otag(h, TAG_B, 0, NULL); |
h->metaf = print_otag(h, TAG_B, ""); |
print_otag(h, TAG_I, 0, NULL); |
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 *this, *next; |
int skip, sz; |
int flags; |
|
|
/* |
this = h->tag; |
* Account for escaped sequences within string length |
for (;;) { |
* calculations. This follows the logic in term_strlen() as we |
next = this->next; |
* must calculate the width of produced strings. |
flags = htmltags[this->tag].flags; |
* Assume that characters are always width of "1". This is |
if (flags & (HTML_INPHRASE | HTML_TOPHRASE)) |
* hacky, but it gets the job done for approximation of widths. |
print_ctag(h, this); |
*/ |
if ((flags & HTML_INPHRASE) == 0) |
|
break; |
|
this = next; |
|
} |
|
} |
|
|
sz = 0; |
/* |
skip = 0; |
* ROFF_nf switches to no-fill mode, ROFF_fi to fill mode. |
while (1) { |
* TOKEN_NONE does not switch. The old mode is returned. |
rsz = strcspn(cp, "\\"); |
*/ |
if (rsz) { |
enum roff_tok |
cp += rsz; |
html_fillmode(struct html *h, enum roff_tok want) |
if (skip) { |
{ |
skip = 0; |
struct tag *t; |
rsz--; |
enum roff_tok had; |
} |
|
sz += rsz; |
for (t = h->tag; t != NULL; t = t->next) |
} |
if (t->tag == TAG_PRE) |
if ('\0' == *cp) |
|
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) { |
/* FALLTHROUGH */ |
case ROFF_fi: |
case ESCAPE_NUMBERED: |
print_tagq(h, t); |
/* FALLTHROUGH */ |
|
case ESCAPE_SPECIAL: |
|
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; |
|
case TOKEN_NONE: |
|
break; |
default: |
default: |
|
abort(); |
|
} |
|
} |
|
return had; |
|
} |
|
|
|
/* |
|
* Allocate a string to be used for the "id=" attribute of an HTML |
|
* element and/or as a segment identifier for a URI in an <a> element. |
|
* The function may fail and return NULL if the node lacks text data |
|
* to create the attribute from. |
|
* If the "unique" argument is 0, the caller is responsible for |
|
* free(3)ing the returned string after using it. |
|
* If the "unique" argument is non-zero, the "id_unique" ohash table |
|
* is used for de-duplication and owns the returned string, so the |
|
* caller must not free(3) it. In this case, it will be freed |
|
* automatically by html_reset() or html_free(). |
|
*/ |
|
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; |
|
|
|
if (n->tag != NULL) |
|
buf = mandoc_strdup(n->tag); |
|
else { |
|
switch (n->tok) { |
|
case MDOC_Sh: |
|
case MDOC_Ss: |
|
case MDOC_Sx: |
|
case MAN_SH: |
|
case MAN_SS: |
|
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; |
break; |
break; |
|
default: |
|
if (n->child == NULL || n->child->type != ROFFT_TEXT) |
|
return NULL; |
|
buf = mandoc_strdup(n->child->string); |
|
break; |
} |
} |
} |
} |
return(sz); |
|
|
/* |
|
* 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(char c) |
print_escape(struct html *h, char c) |
{ |
{ |
|
|
switch (c) { |
switch (c) { |
case '<': |
case '<': |
printf("<"); |
print_word(h, "<"); |
break; |
break; |
case '>': |
case '>': |
printf(">"); |
print_word(h, ">"); |
break; |
break; |
case '&': |
case '&': |
printf("&"); |
print_word(h, "&"); |
break; |
break; |
case '"': |
case '"': |
printf("""); |
print_word(h, """); |
break; |
break; |
case ASCII_NBRSP: |
case ASCII_NBRSP: |
putchar('-'); |
print_word(h, " "); |
break; |
break; |
case ASCII_HYPH: |
case ASCII_HYPH: |
putchar('-'); |
print_byte(h, '-'); |
/* FALLTHROUGH */ |
break; |
case ASCII_BREAK: |
case ASCII_BREAK: |
break; |
break; |
default: |
default: |
return(0); |
return 0; |
} |
} |
return(1); |
return 1; |
} |
} |
|
|
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 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) |
|
pend = strchr(p, '\0'); |
|
|
|
breakline = 0; |
nospace = 0; |
nospace = 0; |
|
|
while ('\0' != *p) { |
while (p < pend) { |
if (HTML_SKIPCHAR & h->flags && '\\' != *p) { |
if (HTML_SKIPCHAR & h->flags && '\\' != *p) { |
h->flags &= ~HTML_SKIPCHAR; |
h->flags &= ~HTML_SKIPCHAR; |
p++; |
p++; |
continue; |
continue; |
} |
} |
|
|
sz = strcspn(p, rejs); |
for (sz = strcspn(p, rejs); sz-- && p < pend; p++) |
|
print_byte(h, *p); |
|
|
fwrite(p, 1, sz, stdout); |
if (breakline && |
p += (int)sz; |
(p >= pend || *p == ' ' || *p == ASCII_NBRSP)) { |
|
print_otag(h, TAG_BR, ""); |
|
breakline = 0; |
|
while (p < pend && (*p == ' ' || *p == ASCII_NBRSP)) |
|
p++; |
|
continue; |
|
} |
|
|
if ('\0' == *p) |
if (p >= pend) |
break; |
break; |
|
|
if (print_escape(*p++)) |
if (*p == ' ') { |
|
print_endword(h); |
|
p++; |
continue; |
continue; |
|
} |
|
|
esc = mandoc_escape(&p, &seq, &len); |
if (print_escape(h, *p++)) |
if (ESCAPE_ERROR == esc) |
continue; |
break; |
|
|
|
|
esc = mandoc_escape(&p, &seq, &len); |
switch (esc) { |
switch (esc) { |
case ESCAPE_FONT: |
case ESCAPE_FONT: |
/* FALLTHROUGH */ |
|
case ESCAPE_FONTPREV: |
case ESCAPE_FONTPREV: |
/* FALLTHROUGH */ |
|
case ESCAPE_FONTBOLD: |
case ESCAPE_FONTBOLD: |
/* FALLTHROUGH */ |
|
case ESCAPE_FONTITALIC: |
case ESCAPE_FONTITALIC: |
/* FALLTHROUGH */ |
|
case ESCAPE_FONTBI: |
case ESCAPE_FONTBI: |
/* FALLTHROUGH */ |
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 415 print_encode(struct html *h, const char *p, int norecu |
|
Line 521 print_encode(struct html *h, const char *p, int norecu |
|
case ESCAPE_UNICODE: |
case ESCAPE_UNICODE: |
/* Skip past "u" header. */ |
/* Skip past "u" header. */ |
c = mchars_num2uc(seq + 1, len - 1); |
c = mchars_num2uc(seq + 1, len - 1); |
if ('\0' != c) |
|
printf("&#x%x;", c); |
|
break; |
break; |
case ESCAPE_NUMBERED: |
case ESCAPE_NUMBERED: |
c = mchars_num2char(seq, len); |
c = mchars_num2char(seq, len); |
if ( ! ('\0' == c || print_escape(c))) |
if (c < 0) |
putchar(c); |
continue; |
break; |
break; |
case ESCAPE_SPECIAL: |
case ESCAPE_SPECIAL: |
c = mchars_spec2cp(h->symtab, seq, len); |
c = mchars_spec2cp(seq, len); |
if (c > 0) |
if (c <= 0) |
printf("&#%d;", c); |
continue; |
else if (-1 == c && 1 == len && |
|
!print_escape(*seq)) |
|
putchar((int)*seq); |
|
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; |
|
continue; |
|
case ESCAPE_OVERSTRIKE: |
|
if (len == 0) |
|
continue; |
|
c = seq[len - 1]; |
break; |
break; |
default: |
default: |
break; |
continue; |
} |
} |
|
if ((c < 0x20 && c != 0x09) || |
|
(c > 0x7E && c < 0xA0)) |
|
c = 0xFFFD; |
|
if (c > 0x7E) { |
|
(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); |
struct stat sb; |
(void)print_encode(h, val, 1); |
const char *p, *pp; |
putchar('\"'); |
char *filename; |
|
|
|
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) { |
|
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 style_written, tflags; |
|
|
/* Push this tags onto the stack of open scopes. */ |
tflags = htmltags[tag].flags; |
|
|
if ( ! (HTML_NOSTACK & 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; |
|
} |
|
|
|
/* |
|
* Always wrap phrasing elements in a paragraph |
|
* unless already contained in some flow container; |
|
* never put them directly into a section. |
|
*/ |
|
|
|
} else if (tflags & HTML_TOPHRASE && h->tag->tag == TAG_SECTION) |
|
print_otag(h, TAG_P, "c", "Pp"); |
|
|
|
/* Push this tag onto the stack of open scopes. */ |
|
|
|
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->tags.head; |
t->next = h->tag; |
h->tags.head = t; |
t->refcnt = 0; |
|
t->closed = 0; |
|
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); |
/* Manage keeps! */ |
if (h->col == 0) |
if ( ! (HTML_KEEP & h->flags)) { |
print_indent(h); |
if (HTML_PREKEEP & h->flags) |
else if ((h->flags & HTML_NOSPACE) == 0) { |
h->flags |= HTML_KEEP; |
if (h->flags & HTML_KEEP) |
putchar(' '); |
print_word(h, " "); |
} else |
else { |
printf(" "); |
if (h->flags & HTML_PREKEEP) |
|
h->flags |= HTML_KEEP; |
|
print_endword(h); |
} |
} |
|
} |
|
|
if ( ! (h->flags & HTML_NONOSPACE)) |
if ( ! (h->flags & HTML_NONOSPACE)) |
h->flags &= ~HTML_NOSPACE; |
h->flags &= ~HTML_NOSPACE; |
Line 486 print_otag(struct html *h, enum htmltag tag, |
|
Line 665 print_otag(struct html *h, enum htmltag tag, |
|
|
|
/* 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); |
|
|
|
/* Accommodate for XML "well-formed" singleton escaping. */ |
va_start(ap, fmt); |
|
|
if (HTML_AUTOCLOSE & htmltags[tag].flags) |
while (*fmt != '\0' && *fmt != 's') { |
switch (h->type) { |
|
case HTML_XHTML_1_0_STRICT: |
/* Parse attributes and arguments. */ |
putchar('/'); |
|
|
arg1 = va_arg(ap, char *); |
|
arg2 = NULL; |
|
switch (*fmt++) { |
|
case 'c': |
|
attr = "class"; |
break; |
break; |
|
case 'h': |
|
attr = "href"; |
|
break; |
|
case 'i': |
|
attr = "id"; |
|
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; |
|
default: |
|
print_encode(h, arg1, NULL, 1); |
|
break; |
} |
} |
|
print_byte(h, '"'); |
|
} |
|
|
putchar('>'); |
style_written = 0; |
|
while (*fmt++ == 's') { |
|
arg1 = va_arg(ap, char *); |
|
arg2 = va_arg(ap, char *); |
|
if (arg2 == NULL) |
|
continue; |
|
print_byte(h, ' '); |
|
if (style_written == 0) { |
|
print_word(h, "style=\""); |
|
style_written = 1; |
|
} |
|
print_word(h, arg1); |
|
print_byte(h, ':'); |
|
print_byte(h, ' '); |
|
print_word(h, arg2); |
|
print_byte(h, ';'); |
|
} |
|
if (style_written) |
|
print_byte(h, '"'); |
|
|
h->flags |= HTML_NOSPACE; |
va_end(ap); |
|
|
if ((HTML_AUTOCLOSE | HTML_CLRLINE) & htmltags[tag].flags) |
/* Accommodate for "well-formed" singleton escaping. */ |
putchar('\n'); |
|
|
|
return(t); |
if (htmltags[tag].flags & HTML_NOSTACK) |
|
print_byte(h, '/'); |
|
|
|
print_byte(h, '>'); |
|
|
|
if (tflags & HTML_NLBEGIN) |
|
print_endline(h); |
|
else |
|
h->flags |= HTML_NOSPACE; |
|
|
|
if (tflags & HTML_INDENT) |
|
h->indent++; |
|
if (tflags & HTML_NOINDENT) |
|
h->noindent++; |
|
|
|
return t; |
} |
} |
|
|
|
/* |
|
* Print an element with an optional "id=" attribute. |
|
* If the element has phrasing content and an "id=" attribute, |
|
* also add a permalink: outside if it can be in phrasing context, |
|
* inside otherwise. |
|
*/ |
|
struct tag * |
|
print_otag_id(struct html *h, enum htmltag elemtype, const char *cattr, |
|
struct roff_node *n) |
|
{ |
|
struct roff_node *nch; |
|
struct tag *ret, *t; |
|
char *id, *href; |
|
|
|
ret = NULL; |
|
id = href = NULL; |
|
if (n->flags & NODE_ID) |
|
id = html_make_id(n, 1); |
|
if (n->flags & NODE_HREF) |
|
href = id == NULL ? html_make_id(n, 0) : id; |
|
if (href != NULL && htmltags[elemtype].flags & HTML_INPHRASE) |
|
ret = print_otag(h, TAG_A, "chR", "permalink", href); |
|
t = print_otag(h, elemtype, "ci", cattr, id); |
|
if (ret == NULL) { |
|
ret = t; |
|
if (href != NULL && (nch = n->child) != NULL) { |
|
/* man(7) is safe, it tags phrasing content only. */ |
|
if (n->tok > MDOC_MAX || |
|
htmltags[elemtype].flags & HTML_TOPHRASE) |
|
nch = NULL; |
|
else /* For mdoc(7), beware of nested blocks. */ |
|
while (nch != NULL && nch->type == ROFFT_TEXT) |
|
nch = nch->next; |
|
if (nch == NULL) |
|
print_otag(h, TAG_A, "chR", "permalink", href); |
|
} |
|
} |
|
if (id == NULL) |
|
free(href); |
|
return ret; |
|
} |
|
|
static void |
static void |
print_ctag(struct html *h, enum htmltag tag) |
print_ctag(struct html *h, struct tag *tag) |
{ |
{ |
|
int tflags; |
|
|
printf("</%s>", htmltags[tag].name); |
if (tag->closed == 0) { |
if (HTML_CLRLINE & htmltags[tag].flags) { |
tag->closed = 1; |
h->flags |= HTML_NOSPACE; |
if (tag == h->metaf) |
putchar('\n'); |
h->metaf = NULL; |
|
if (tag == h->tblt) |
|
h->tblt = NULL; |
|
|
|
tflags = htmltags[tag->tag].flags; |
|
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); |
} |
} |
|
if (tag->refcnt == 0) { |
|
h->tag = tag->next; |
|
free(tag); |
|
} |
} |
} |
|
|
void |
void |
print_gen_decls(struct html *h) |
print_gen_decls(struct html *h) |
{ |
{ |
|
print_word(h, "<!DOCTYPE html>"); |
|
print_endline(h); |
|
} |
|
|
puts("<!DOCTYPE html>"); |
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 |
void |
print_text(struct html *h, const char *word) |
print_text(struct html *h, const char *word) |
{ |
{ |
|
/* |
|
* Always wrap text in a paragraph unless already contained in |
|
* some flow container; never put it directly into a section. |
|
*/ |
|
|
if ( ! (HTML_NOSPACE & h->flags)) { |
if (h->tag->tag == TAG_SECTION) |
/* Manage keeps! */ |
print_otag(h, TAG_P, "c", "Pp"); |
|
|
|
/* Output whitespace before this text? */ |
|
|
|
if (h->col && (h->flags & HTML_NOSPACE) == 0) { |
if ( ! (HTML_KEEP & h->flags)) { |
if ( ! (HTML_KEEP & h->flags)) { |
if (HTML_PREKEEP & h->flags) |
if (HTML_PREKEEP & h->flags) |
h->flags |= HTML_KEEP; |
h->flags |= HTML_KEEP; |
putchar(' '); |
print_endword(h); |
} else |
} else |
printf(" "); |
print_word(h, " "); |
} |
} |
|
|
assert(NULL == h->metaf); |
/* |
switch (h->metac) { |
* Print the text, optionally surrounded by HTML whitespace, |
case HTMLFONT_ITALIC: |
* optionally manually switching fonts before and after. |
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(h->metaf == NULL); |
if ( ! print_encode(h, word, 0)) { |
print_metaf(h); |
|
print_indent(h); |
|
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; |
|
h->flags &= ~HTML_NONEWLINE; |
} else |
} else |
h->flags |= HTML_NOSPACE; |
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 577 print_text(struct html *h, const char *word) |
|
Line 923 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->tags.head) != NULL) { |
for (this = h->tag; this != NULL; this = next) { |
/* |
next = this == until ? NULL : this->next; |
* Remember to close out and nullify the current |
print_ctag(h, this); |
* meta-font and table, if applicable. |
|
*/ |
|
if (tag == h->metaf) |
|
h->metaf = NULL; |
|
if (tag == h->tblt) |
|
h->tblt = NULL; |
|
print_ctag(h, tag->tag); |
|
h->tags.head = tag->next; |
|
free(tag); |
|
if (until && tag == until) |
|
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->tags.head) != NULL) { |
for (this = h->tag; this != NULL; this = next) { |
if (suntil && tag == suntil) |
next = this->next; |
return; |
if (this == suntil || (next == suntil && |
/* |
(this->tag == TAG_P || this->tag == TAG_PRE))) |
* Remember to close out and nullify the current |
break; |
* meta-font and table, if applicable. |
print_ctag(h, this); |
*/ |
|
if (tag == h->metaf) |
|
h->metaf = NULL; |
|
if (tag == h->tblt) |
|
h->tblt = NULL; |
|
print_ctag(h, tag->tag); |
|
h->tags.head = tag->next; |
|
free(tag); |
|
} |
} |
} |
} |
|
|
void |
|
bufinit(struct html *h) |
|
{ |
|
|
|
h->buf[0] = '\0'; |
/*********************************************************************** |
h->buflen = 0; |
* Low level output functions. |
} |
* They implement line breaking using a short static buffer. |
|
***********************************************************************/ |
|
|
void |
/* |
bufcat_style(struct html *h, const char *key, const char *val) |
* Buffer one HTML output byte. |
|
* If the buffer is full, flush and deactivate it and start a new line. |
|
* If the buffer is inactive, print directly. |
|
*/ |
|
static void |
|
print_byte(struct html *h, char c) |
{ |
{ |
|
if ((h->flags & HTML_BUFFER) == 0) { |
|
putchar(c); |
|
h->col++; |
|
return; |
|
} |
|
|
bufcat(h, key); |
if (h->col + h->bufcol < sizeof(h->buf)) { |
bufcat(h, ":"); |
h->buf[h->bufcol++] = c; |
bufcat(h, val); |
return; |
bufcat(h, ";"); |
} |
} |
|
|
|
void |
putchar('\n'); |
bufcat(struct html *h, const char *p) |
h->col = 0; |
{ |
print_indent(h); |
|
putchar(' '); |
/* |
putchar(' '); |
* XXX This is broken and not easy to fix. |
fwrite(h->buf, h->bufcol, 1, stdout); |
* When using the -Oincludes option, buffmt_includes() |
putchar(c); |
* may pass in strings overrunning BUFSIZ, causing a crash. |
h->col = (h->indent + 1) * 2 + h->bufcol + 1; |
*/ |
h->bufcol = 0; |
|
h->flags &= ~HTML_BUFFER; |
h->buflen = strlcat(h->buf, p, BUFSIZ); |
|
assert(h->buflen < BUFSIZ); |
|
} |
} |
|
|
|
/* |
|
* If something was printed on the current output line, end it. |
|
* Not to be called right after print_indent(). |
|
*/ |
void |
void |
bufcat_fmt(struct html *h, const char *fmt, ...) |
print_endline(struct html *h) |
{ |
{ |
va_list ap; |
if (h->col == 0) |
|
return; |
|
|
va_start(ap, fmt); |
if (h->bufcol) { |
(void)vsnprintf(h->buf + (int)h->buflen, |
putchar(' '); |
BUFSIZ - h->buflen - 1, fmt, ap); |
fwrite(h->buf, h->bufcol, 1, stdout); |
va_end(ap); |
h->bufcol = 0; |
h->buflen = strlen(h->buf); |
} |
|
putchar('\n'); |
|
h->col = 0; |
|
h->flags |= HTML_NOSPACE; |
|
h->flags &= ~HTML_BUFFER; |
} |
} |
|
|
|
/* |
|
* Flush the HTML output buffer. |
|
* If it is inactive, activate it. |
|
*/ |
static void |
static void |
bufncat(struct html *h, const char *p, size_t sz) |
print_endword(struct html *h) |
{ |
{ |
|
if (h->noindent) { |
assert(h->buflen + sz + 1 < BUFSIZ); |
print_byte(h, ' '); |
strncat(h->buf, p, sz); |
return; |
h->buflen += sz; |
|
} |
|
|
|
void |
|
buffmt_includes(struct html *h, const char *name) |
|
{ |
|
const char *p, *pp; |
|
|
|
pp = h->base_includes; |
|
|
|
bufinit(h); |
|
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); |
|
} |
|
|
|
void |
if ((h->flags & HTML_BUFFER) == 0) { |
buffmt_man(struct html *h, const char *name, const char *sec) |
h->col++; |
{ |
h->flags |= HTML_BUFFER; |
const char *p, *pp; |
} else if (h->bufcol) { |
|
putchar(' '); |
pp = h->base_man; |
fwrite(h->buf, h->bufcol, 1, stdout); |
|
h->col += h->bufcol + 1; |
bufinit(h); |
|
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': |
|
bufcat_fmt(h, "%s", name); |
|
break; |
|
default: |
|
bufncat(h, p, 2); |
|
break; |
|
} |
|
pp = p + 2; |
|
} |
} |
if (pp) |
h->bufcol = 0; |
bufcat(h, pp); |
|
} |
} |
|
|
void |
/* |
bufcat_su(struct html *h, const char *p, const struct roffsu *su) |
* If at the beginning of a new output line, |
|
* 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; |
|
|
v = su->scale; |
if (h->col || h->noindent) |
if (SCALE_MM == su->unit && 0.0 == (v /= 100.0)) |
return; |
v = 1.0; |
|
else if (SCALE_BU == su->unit) |
|
v /= 24.0; |
|
|
|
bufcat_fmt(h, "%s: %.2f%s;", p, v, roffscales[su->unit]); |
h->col = h->indent * 2; |
|
for (i = 0; i < h->col; i++) |
|
putchar(' '); |
} |
} |
|
|
void |
/* |
bufcat_id(struct html *h, const char *src) |
* Print or buffer some characters |
|
* depending on the current HTML output buffer state. |
|
*/ |
|
static void |
|
print_word(struct html *h, const char *cp) |
{ |
{ |
|
while (*cp != '\0') |
/* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */ |
print_byte(h, *cp++); |
|
|
while ('\0' != *src) |
|
bufcat_fmt(h, "%.2x", *src++); |
|
} |
} |