=================================================================== RCS file: /cvs/mandoc/html.c,v retrieving revision 1.52 retrieving revision 1.246 diff -u -p -r1.52 -r1.246 --- mandoc/html.c 2009/09/24 09:50:31 1.52 +++ mandoc/html.c 2018/12/15 19:30:26 1.246 @@ -1,487 +1,928 @@ -/* $Id: html.c,v 1.52 2009/09/24 09:50:31 kristaps Exp $ */ +/* $Id: html.c,v 1.246 2018/12/15 19:30:26 schwarze Exp $ */ /* - * Copyright (c) 2008, 2009 Kristaps Dzonsons + * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons + * Copyright (c) 2011-2015, 2017, 2018 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above * 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 - * 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 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ +#include "config.h" + #include -#include +#include #include -#include +#include +#include +#include #include +#include #include #include #include -#include "chars.h" +#include "mandoc_aux.h" +#include "mandoc_ohash.h" +#include "mandoc.h" +#include "roff.h" +#include "out.h" #include "html.h" +#include "manconf.h" +#include "main.h" -#define DOCTYPE "-//W3C//DTD HTML 4.01//EN" -#define DTD "http://www.w3.org/TR/html4/strict.dtd" - struct htmldata { - char *name; + const char *name; int flags; -#define HTML_CLRLINE (1 << 0) -#define HTML_NOSTACK (1 << 1) +#define HTML_NOSTACK (1 << 0) +#define HTML_AUTOCLOSE (1 << 1) +#define HTML_NLBEFORE (1 << 2) +#define HTML_NLBEGIN (1 << 3) +#define HTML_NLEND (1 << 4) +#define HTML_NLAFTER (1 << 5) +#define HTML_NLAROUND (HTML_NLBEFORE | HTML_NLAFTER) +#define HTML_NLINSIDE (HTML_NLBEGIN | HTML_NLEND) +#define HTML_NLALL (HTML_NLAROUND | HTML_NLINSIDE) +#define HTML_INDENT (1 << 6) +#define HTML_NOINDENT (1 << 7) }; static const struct htmldata htmltags[TAG_MAX] = { - {"html", HTML_CLRLINE}, /* TAG_HTML */ - {"head", HTML_CLRLINE}, /* TAG_HEAD */ - {"body", HTML_CLRLINE}, /* TAG_BODY */ - {"meta", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_META */ - {"title", HTML_CLRLINE}, /* TAG_TITLE */ - {"div", HTML_CLRLINE}, /* TAG_DIV */ - {"h1", 0}, /* TAG_H1 */ - {"h2", 0}, /* TAG_H2 */ - {"p", HTML_CLRLINE}, /* TAG_P */ - {"span", 0}, /* TAG_SPAN */ - {"link", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_LINK */ - {"br", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_LINK */ - {"a", 0}, /* TAG_A */ - {"table", HTML_CLRLINE}, /* TAG_TABLE */ - {"col", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_COL */ - {"tr", HTML_CLRLINE}, /* TAG_TR */ - {"td", HTML_CLRLINE}, /* TAG_TD */ - {"li", HTML_CLRLINE}, /* TAG_LI */ - {"ul", HTML_CLRLINE}, /* TAG_UL */ - {"ol", HTML_CLRLINE}, /* TAG_OL */ - {"base", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_BASE */ + {"html", HTML_NLALL}, + {"head", HTML_NLALL | HTML_INDENT}, + {"body", HTML_NLALL}, + {"meta", HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL}, + {"title", HTML_NLAROUND}, + {"div", HTML_NLAROUND}, + {"div", 0}, + {"h1", HTML_NLAROUND}, + {"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}, + {"tr", HTML_NLALL | HTML_INDENT}, + {"td", HTML_NLAROUND}, + {"li", HTML_NLAROUND | HTML_INDENT}, + {"ul", HTML_NLALL | HTML_INDENT}, + {"ol", HTML_NLALL | HTML_INDENT}, + {"dl", HTML_NLALL | HTML_INDENT}, + {"dt", HTML_NLAROUND}, + {"dd", HTML_NLAROUND | HTML_INDENT}, + {"pre", HTML_NLALL | HTML_NOINDENT}, + {"var", 0}, + {"cite", 0}, + {"b", 0}, + {"i", 0}, + {"code", 0}, + {"small", 0}, + {"style", HTML_NLALL | HTML_INDENT}, + {"math", HTML_NLALL | HTML_INDENT}, + {"mrow", 0}, + {"mi", 0}, + {"mn", 0}, + {"mo", 0}, + {"msup", 0}, + {"msub", 0}, + {"msubsup", 0}, + {"mfrac", 0}, + {"msqrt", 0}, + {"mfenced", 0}, + {"mtable", 0}, + {"mtr", 0}, + {"mtd", 0}, + {"munderover", 0}, + {"munder", 0}, + {"mover", 0}, }; -static const char *const htmlattrs[ATTR_MAX] = { - "http-equiv", - "content", - "name", - "rel", - "href", - "type", - "media", - "class", - "style", - "width", - "valign", -}; +/* Avoid duplicate HTML id= attributes. */ +static struct ohash id_unique; -#ifdef __linux__ -extern int getsubopt(char **, char * const *, char **); -#endif +static void print_byte(struct html *, char); +static void print_endword(struct html *); +static void print_indent(struct html *); +static void print_word(struct html *, const char *); +static void print_ctag(struct html *, struct tag *); +static int print_escape(struct html *, char); +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_metaf(struct html *, enum mandoc_esc); + + void * -html_alloc(char *outopts) +html_alloc(const struct manoutput *outopts) { struct html *h; - char *toks[3], *v; - toks[0] = "style"; - toks[1] = "base"; - toks[2] = NULL; + h = mandoc_calloc(1, sizeof(struct html)); - if (NULL == (h = calloc(1, sizeof(struct html)))) - return(NULL); + h->tag = NULL; + h->style = outopts->style; + 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; - SLIST_INIT(&h->tags); - SLIST_INIT(&h->ords); + mandoc_ohash_init(&id_unique, 4, 0); - if (NULL == (h->symtab = chars_init(CHARS_HTML))) { - free(h); - return(NULL); - } - - while (outopts && *outopts) - switch (getsubopt(&outopts, toks, &v)) { - case (0): - h->style = v; - break; - case (1): - h->base = v; - break; - default: - break; - } - - return(h); + return h; } - void html_free(void *p) { struct tag *tag; - struct ord *ord; struct html *h; + char *cp; + unsigned int slot; h = (struct html *)p; - - while ( ! SLIST_EMPTY(&h->ords)) { - ord = SLIST_FIRST(&h->ords); - SLIST_REMOVE_HEAD(&h->ords, entry); - free(ord); - } - - while ( ! SLIST_EMPTY(&h->tags)) { - tag = SLIST_FIRST(&h->tags); - SLIST_REMOVE_HEAD(&h->tags, entry); + while ((tag = h->tag) != NULL) { + h->tag = tag->next; free(tag); } - - if (h->symtab) - chars_free(h->symtab); free(h); + + cp = ohash_first(&id_unique, &slot); + while (cp != NULL) { + free(cp); + cp = ohash_next(&id_unique, &slot); + } + ohash_delete(&id_unique); } - void print_gen_head(struct html *h) { - struct htmlpair tag[4]; + struct tag *t; - tag[0].key = ATTR_HTTPEQUIV; - tag[0].val = "Content-Type"; - tag[1].key = ATTR_CONTENT; - tag[1].val = "text/html; charset=utf-8"; - print_otag(h, TAG_META, 2, tag); + 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; + } - tag[0].key = ATTR_NAME; - tag[0].val = "resource-type"; - tag[1].key = ATTR_CONTENT; - tag[1].val = "document"; - print_otag(h, TAG_META, 2, tag); + /* + * Print a minimal embedded style sheet. + */ - if (h->style) { - tag[0].key = ATTR_REL; - tag[0].val = "stylesheet"; - tag[1].key = ATTR_HREF; - tag[1].val = h->style; - tag[2].key = ATTR_TYPE; - tag[2].val = "text/css"; - tag[3].key = ATTR_MEDIA; - tag[3].val = "all"; - print_otag(h, TAG_LINK, 4, tag); + t = print_otag(h, TAG_STYLE, ""); + print_text(h, "table.head, table.foot { width: 100%; }"); + print_endline(h); + print_text(h, "td.head-rtitle, td.foot-os { text-align: right; }"); + print_endline(h); + print_text(h, "td.head-vol { text-align: center; }"); + print_endline(h); + print_text(h, "div.Pp { margin: 1ex 0ex; }"); + print_endline(h); + print_text(h, "div.Nd, div.Bf, div.Op { display: inline; }"); + print_endline(h); + print_text(h, "span.Pa, span.Ad { font-style: italic; }"); + print_endline(h); + print_text(h, "span.Ms { font-weight: bold; }"); + print_endline(h); + print_text(h, "dl.Bl-diag "); + print_byte(h, '>'); + print_text(h, " dt { font-weight: bold; }"); + print_endline(h); + print_text(h, "code.Nm, code.Fl, code.Cm, code.Ic, " + "code.In, code.Fd, code.Fn,"); + print_endline(h); + print_text(h, "code.Cd { font-weight: bold; " + "font-family: inherit; }"); + print_tagq(h, t); +} + +static void +print_metaf(struct html *h, enum mandoc_esc deco) +{ + enum htmlfont font; + + switch (deco) { + case ESCAPE_FONTPREV: + font = h->metal; + break; + case ESCAPE_FONTITALIC: + font = HTMLFONT_ITALIC; + break; + case ESCAPE_FONTBOLD: + font = HTMLFONT_BOLD; + break; + case ESCAPE_FONTBI: + font = HTMLFONT_BI; + break; + case ESCAPE_FONTCW: + font = HTMLFONT_CW; + break; + case ESCAPE_FONT: + case ESCAPE_FONTROMAN: + font = HTMLFONT_NONE; + break; + default: + abort(); } - if (h->base) { - tag[0].key = ATTR_HREF; - tag[1].val = h->base; - print_otag(h, TAG_BASE, 1, tag); + if (h->metaf) { + print_tagq(h, h->metaf); + h->metaf = NULL; } -} + h->metal = h->metac; + h->metac = font; -static void -print_spec(struct html *h, const char *p, int len) + switch (font) { + case HTMLFONT_ITALIC: + h->metaf = print_otag(h, TAG_I, ""); + break; + case HTMLFONT_BOLD: + h->metaf = print_otag(h, TAG_B, ""); + break; + case HTMLFONT_BI: + h->metaf = print_otag(h, TAG_B, ""); + print_otag(h, TAG_I, ""); + break; + case HTMLFONT_CW: + h->metaf = print_otag(h, TAG_SPAN, "c", "Li"); + break; + default: + break; + } +} + +char * +html_make_id(const struct roff_node *n, int unique) { - const char *rhs; - int i; - size_t sz; + const struct roff_node *nch; + char *buf, *bufs, *cp; + unsigned int slot; + int suffix; - rhs = chars_a2ascii(h->symtab, p, (size_t)len, &sz); + for (nch = n->child; nch != NULL; nch = nch->next) + if (nch->type != ROFFT_TEXT) + return NULL; - if (NULL == rhs) - return; - for (i = 0; i < (int)sz; i++) - putchar(rhs[i]); -} + 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 + */ -static void -print_res(struct html *h, const char *p, int len) -{ - const char *rhs; - int i; - size_t sz; + for (cp = buf; *cp != '\0'; cp++) + if (isalnum((unsigned char)*cp) == 0 && + strchr("!$&'()*+,-./:;=?@_~", *cp) == NULL) + *cp = '_'; - rhs = chars_a2res(h->symtab, p, (size_t)len, &sz); + if (unique == 0) + return buf; - if (NULL == rhs) - return; - for (i = 0; i < (int)sz; i++) - putchar(rhs[i]); + /* 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 +print_escape(struct html *h, char c) +{ -static void -print_escape(struct html *h, const char **p) + switch (c) { + case '<': + print_word(h, "<"); + break; + case '>': + print_word(h, ">"); + break; + case '&': + print_word(h, "&"); + break; + case '"': + print_word(h, """); + break; + case ASCII_NBRSP: + print_word(h, " "); + break; + case ASCII_HYPH: + print_byte(h, '-'); + break; + case ASCII_BREAK: + break; + default: + return 0; + } + return 1; +} + +static int +print_encode(struct html *h, const char *p, const char *pend, int norecurse) { - int j, type; - const char *wp; + char numbuf[16]; + const char *seq; + size_t sz; + int c, len, breakline, nospace; + enum mandoc_esc esc; + static const char rejs[10] = { ' ', '\\', '<', '>', '&', '"', + ASCII_NBRSP, ASCII_HYPH, ASCII_BREAK, '\0' }; - wp = *p; - type = 1; + if (pend == NULL) + pend = strchr(p, '\0'); - if (0 == *(++wp)) { - *p = wp; - return; - } + breakline = 0; + nospace = 0; - if ('(' == *wp) { - wp++; - if (0 == *wp || 0 == *(wp + 1)) { - *p = 0 == *wp ? wp : wp + 1; - return; + while (p < pend) { + if (HTML_SKIPCHAR & h->flags && '\\' != *p) { + h->flags &= ~HTML_SKIPCHAR; + p++; + continue; } - print_spec(h, wp, 2); - *p = ++wp; - return; + for (sz = strcspn(p, rejs); sz-- && p < pend; p++) + print_byte(h, *p); - } else if ('*' == *wp) { - if (0 == *(++wp)) { - *p = wp; - return; + if (breakline && + (p >= pend || *p == ' ' || *p == ASCII_NBRSP)) { + print_otag(h, TAG_BR, ""); + breakline = 0; + while (p < pend && (*p == ' ' || *p == ASCII_NBRSP)) + p++; + continue; } - switch (*wp) { - case ('('): - wp++; - if (0 == *wp || 0 == *(wp + 1)) { - *p = 0 == *wp ? wp : wp + 1; - return; - } - - print_res(h, wp, 2); - *p = ++wp; - return; - case ('['): - type = 0; + if (p >= pend) break; + + if (*p == ' ') { + print_endword(h); + p++; + continue; + } + + if (print_escape(h, *p++)) + continue; + + esc = mandoc_escape(&p, &seq, &len); + switch (esc) { + case ESCAPE_FONT: + case ESCAPE_FONTPREV: + case ESCAPE_FONTBOLD: + case ESCAPE_FONTITALIC: + case ESCAPE_FONTBI: + case ESCAPE_FONTCW: + case ESCAPE_FONTROMAN: + if (0 == norecurse) { + h->flags |= HTML_NOSPACE; + print_metaf(h, esc); + h->flags &= ~HTML_NOSPACE; + } + continue; + case ESCAPE_SKIPCHAR: + h->flags |= HTML_SKIPCHAR; + continue; + case ESCAPE_ERROR: + continue; default: - print_res(h, wp, 1); - *p = wp; - return; + break; } - - } else if ('f' == *wp) { - if (0 == *(++wp)) { - *p = wp; - return; + + if (h->flags & HTML_SKIPCHAR) { + h->flags &= ~HTML_SKIPCHAR; + continue; } - switch (*wp) { - case ('B'): - /* TODO */ + switch (esc) { + case ESCAPE_UNICODE: + /* Skip past "u" header. */ + c = mchars_num2uc(seq + 1, len - 1); break; - case ('I'): - /* TODO */ + case ESCAPE_NUMBERED: + c = mchars_num2char(seq, len); + if (c < 0) + continue; break; - case ('P'): - /* FALLTHROUGH */ - case ('R'): - /* TODO */ + case ESCAPE_SPECIAL: + c = mchars_spec2cp(seq, len); + if (c <= 0) + continue; break; - default: + case ESCAPE_UNDEF: + c = *seq; break; + case ESCAPE_DEVICE: + print_word(h, "html"); + continue; + case ESCAPE_BREAK: + breakline = 1; + continue; + case ESCAPE_NOSPACE: + if ('\0' == *p) + nospace = 1; + continue; + case ESCAPE_OVERSTRIKE: + if (len == 0) + continue; + c = seq[len - 1]; + break; + default: + 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); + } - *p = wp; - return; + return nospace; +} - } else if ('[' != *wp) { - print_spec(h, wp, 1); - *p = wp; - return; +static void +print_href(struct html *h, const char *name, const char *sec, int man) +{ + struct stat sb; + const char *p, *pp; + 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); +} - wp++; - for (j = 0; *wp && ']' != *wp; wp++, j++) - /* Loop... */ ; +struct tag * +print_otag(struct html *h, enum htmltag tag, const char *fmt, ...) +{ + va_list ap; + struct tag *t; + const char *attr; + char *arg1, *arg2; + int style_written, tflags; - if (0 == *wp) { - *p = wp; - return; + tflags = htmltags[tag].flags; + + /* Push this tag onto the stack of open scopes. */ + + if ((tflags & HTML_NOSTACK) == 0) { + t = mandoc_malloc(sizeof(struct tag)); + t->tag = tag; + t->next = h->tag; + h->tag = t; + } else + t = NULL; + + if (tflags & HTML_NLBEFORE) + print_endline(h); + if (h->col == 0) + print_indent(h); + else if ((h->flags & HTML_NOSPACE) == 0) { + if (h->flags & HTML_KEEP) + print_word(h, " "); + else { + if (h->flags & HTML_PREKEEP) + h->flags |= HTML_KEEP; + print_endword(h); + } } - if (type) - print_spec(h, wp - j, j); + if ( ! (h->flags & HTML_NONOSPACE)) + h->flags &= ~HTML_NOSPACE; else - print_res(h, wp - j, j); + h->flags |= HTML_NOSPACE; - *p = wp; -} + /* Print out the tag name and attributes. */ + print_byte(h, '<'); + print_word(h, htmltags[tag].name); -static void -print_encode(struct html *h, const char *p) -{ + va_start(ap, fmt); - for (; *p; p++) { - if ('\\' == *p) { - print_escape(h, &p); - continue; + while (*fmt != '\0' && *fmt != 's') { + + /* Parse attributes and arguments. */ + + arg1 = va_arg(ap, char *); + arg2 = NULL; + switch (*fmt++) { + case 'c': + attr = "class"; + break; + case 'h': + attr = "href"; + break; + case 'i': + attr = "id"; + break; + case '?': + attr = arg1; + arg1 = va_arg(ap, char *); + break; + default: + abort(); } - switch (*p) { - case ('<'): - printf("<"); + 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; - case ('>'): - printf(">"); + case 'M': + print_href(h, arg1, arg2, 1); + fmt++; break; - case ('&'): - printf("&"); + case 'R': + print_byte(h, '#'); + print_encode(h, arg1, NULL, 1); + fmt++; break; + case 'T': + print_encode(h, arg1, NULL, 1); + print_word(h, "\" title=\""); + print_encode(h, arg1, NULL, 1); + fmt++; + break; default: - putchar(*p); + print_encode(h, arg1, NULL, 1); break; } + print_byte(h, '"'); } -} + 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, '"'); -struct tag * -print_otag(struct html *h, enum htmltag tag, - int sz, const struct htmlpair *p) -{ - int i; - struct tag *t; + va_end(ap); - if ( ! (HTML_NOSTACK & htmltags[tag].flags)) { - if (NULL == (t = malloc(sizeof(struct tag)))) - err(EXIT_FAILURE, "malloc"); - t->tag = tag; - SLIST_INSERT_HEAD(&h->tags, t, entry); - } else - t = NULL; + /* Accommodate for "well-formed" singleton escaping. */ - if ( ! (HTML_NOSPACE & h->flags)) - if ( ! (HTML_CLRLINE & htmltags[tag].flags)) - printf(" "); + if (HTML_AUTOCLOSE & htmltags[tag].flags) + print_byte(h, '/'); - printf("<%s", htmltags[tag].name); - for (i = 0; i < sz; i++) { - printf(" %s=\"", htmlattrs[p[i].key]); - assert(p->val); - print_encode(h, p[i].val); - printf("\""); - } - printf(">"); + print_byte(h, '>'); - h->flags |= HTML_NOSPACE; - if (HTML_CLRLINE & htmltags[tag].flags) - h->flags |= HTML_NEWLINE; + if (tflags & HTML_NLBEGIN) + print_endline(h); else - h->flags &= ~HTML_NEWLINE; + h->flags |= HTML_NOSPACE; - return(t); + if (tflags & HTML_INDENT) + h->indent++; + if (tflags & HTML_NOINDENT) + h->noindent++; + + return t; } - -/* ARGSUSED */ static void -print_ctag(struct html *h, enum htmltag tag) +print_ctag(struct html *h, struct tag *tag) { - - printf("", htmltags[tag].name); - if (HTML_CLRLINE & htmltags[tag].flags) - h->flags |= HTML_NOSPACE; - if (HTML_CLRLINE & htmltags[tag].flags) - h->flags |= HTML_NEWLINE; - else - h->flags &= ~HTML_NEWLINE; -} + int tflags; + /* + * Remember to close out and nullify the current + * meta-font and table, if applicable. + */ + if (tag == h->metaf) + h->metaf = NULL; + if (tag == h->tblt) + h->tblt = NULL; -/* ARGSUSED */ + 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); + + h->tag = tag->next; + free(tag); +} + void -print_gen_doctype(struct html *h) +print_gen_decls(struct html *h) { - - printf("", DOCTYPE, DTD); + print_word(h, ""); + print_endline(h); } - void -print_text(struct html *h, const char *p) +print_gen_comment(struct html *h, struct roff_node *n) { + int wantblank; - if (*p && 0 == *(p + 1)) - switch (*p) { - case('.'): - /* FALLTHROUGH */ - 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; + print_word(h, "") == 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; +} - if ( ! (h->flags & HTML_NOSPACE)) - printf(" "); +void +print_text(struct html *h, const char *word) +{ + if (h->col && (h->flags & HTML_NOSPACE) == 0) { + if ( ! (HTML_KEEP & h->flags)) { + if (HTML_PREKEEP & h->flags) + h->flags |= HTML_KEEP; + print_endword(h); + } else + print_word(h, " "); + } - h->flags &= ~HTML_NOSPACE; - h->flags &= ~HTML_NEWLINE; + assert(NULL == h->metaf); + switch (h->metac) { + case HTMLFONT_ITALIC: + h->metaf = print_otag(h, TAG_I, ""); + break; + case HTMLFONT_BOLD: + h->metaf = print_otag(h, TAG_B, ""); + break; + case HTMLFONT_BI: + h->metaf = print_otag(h, TAG_B, ""); + print_otag(h, TAG_I, ""); + break; + case HTMLFONT_CW: + h->metaf = print_otag(h, TAG_SPAN, "c", "Li"); + break; + default: + print_indent(h); + break; + } - if (p) - print_encode(h, p); + assert(word); + if ( ! print_encode(h, word, NULL, 0)) { + if ( ! (h->flags & HTML_NONOSPACE)) + h->flags &= ~HTML_NOSPACE; + h->flags &= ~HTML_NONEWLINE; + } else + h->flags |= HTML_NOSPACE | HTML_NONEWLINE; - if (*p && 0 == *(p + 1)) - switch (*p) { - case('('): - /* FALLTHROUGH */ - case('['): - /* FALLTHROUGH */ - case('{'): - h->flags |= HTML_NOSPACE; - break; - default: - break; - } + if (h->metaf) { + print_tagq(h, h->metaf); + h->metaf = NULL; + } + + h->flags &= ~HTML_IGNDELIM; } - void print_tagq(struct html *h, const struct tag *until) { struct tag *tag; - while ( ! SLIST_EMPTY(&h->tags)) { - tag = SLIST_FIRST(&h->tags); - print_ctag(h, tag->tag); - SLIST_REMOVE_HEAD(&h->tags, entry); - free(tag); + while ((tag = h->tag) != NULL) { + print_ctag(h, tag); if (until && tag == until) return; } } - void print_stagq(struct html *h, const struct tag *suntil) { struct tag *tag; - while ( ! SLIST_EMPTY(&h->tags)) { - tag = SLIST_FIRST(&h->tags); + while ((tag = h->tag) != NULL) { if (suntil && tag == suntil) return; - print_ctag(h, tag->tag); - SLIST_REMOVE_HEAD(&h->tags, entry); - free(tag); + print_ctag(h, tag); } +} + +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. + * They implement line breaking using a short static buffer. + ***********************************************************************/ + +/* + * 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; + } + + if (h->col + h->bufcol < sizeof(h->buf)) { + h->buf[h->bufcol++] = c; + return; + } + + putchar('\n'); + h->col = 0; + print_indent(h); + putchar(' '); + putchar(' '); + fwrite(h->buf, h->bufcol, 1, stdout); + putchar(c); + h->col = (h->indent + 1) * 2 + h->bufcol + 1; + h->bufcol = 0; + h->flags &= ~HTML_BUFFER; +} + +/* + * If something was printed on the current output line, end it. + * Not to be called right after print_indent(). + */ +void +print_endline(struct html *h) +{ + if (h->col == 0) + return; + + if (h->bufcol) { + putchar(' '); + fwrite(h->buf, h->bufcol, 1, stdout); + h->bufcol = 0; + } + 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 +print_endword(struct html *h) +{ + if (h->noindent) { + print_byte(h, ' '); + return; + } + + if ((h->flags & HTML_BUFFER) == 0) { + h->col++; + h->flags |= HTML_BUFFER; + } else if (h->bufcol) { + putchar(' '); + fwrite(h->buf, h->bufcol, 1, stdout); + h->col += h->bufcol + 1; + } + h->bufcol = 0; +} + +/* + * 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) +{ + size_t i; + + if (h->col) + return; + + if (h->noindent == 0) { + h->col = h->indent * 2; + for (i = 0; i < h->col; i++) + putchar(' '); + } + h->flags &= ~HTML_NOSPACE; +} + +/* + * 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') + print_byte(h, *cp++); }