=================================================================== RCS file: /cvs/mandoc/html.c,v retrieving revision 1.132 retrieving revision 1.175 diff -u -p -r1.132 -r1.175 --- mandoc/html.c 2011/04/09 15:29:40 1.132 +++ mandoc/html.c 2014/10/10 08:44:24 1.175 @@ -1,7 +1,7 @@ -/* $Id: html.c,v 1.132 2011/04/09 15:29:40 kristaps Exp $ */ +/* $Id: html.c,v 1.175 2014/10/10 08:44:24 kristaps Exp $ */ /* - * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons - * Copyright (c) 2011 Ingo Schwarze + * Copyright (c) 2008, 2009, 2010, 2011, 2014 Kristaps Dzonsons + * Copyright (c) 2011, 2012, 2013, 2014 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 @@ -15,9 +15,7 @@ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ -#ifdef HAVE_CONFIG_H #include "config.h" -#endif #include @@ -31,6 +29,8 @@ #include #include "mandoc.h" +#include "mandoc_aux.h" +#include "libmandoc.h" #include "out.h" #include "html.h" #include "main.h" @@ -68,17 +68,31 @@ static const struct htmldata htmltags[TAG_MAX] = { {"dt", HTML_CLRLINE}, /* TAG_DT */ {"dd", HTML_CLRLINE}, /* TAG_DD */ {"blockquote", HTML_CLRLINE}, /* TAG_BLOCKQUOTE */ - {"p", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_P */ {"pre", HTML_CLRLINE }, /* TAG_PRE */ {"b", 0 }, /* TAG_B */ {"i", 0 }, /* TAG_I */ {"code", 0 }, /* TAG_CODE */ {"small", 0 }, /* TAG_SMALL */ + {"style", HTML_CLRLINE}, /* TAG_STYLE */ + {"math", HTML_CLRLINE}, /* TAG_MATH */ + {"mrow", 0}, /* TAG_MROW */ + {"mi", 0}, /* TAG_MI */ + {"mo", 0}, /* TAG_MO */ + {"msup", 0}, /* TAG_MSUP */ + {"msub", 0}, /* TAG_MSUB */ + {"msubsup", 0}, /* TAG_MSUBSUP */ + {"mfrac", 0}, /* TAG_MFRAC */ + {"msqrt", 0}, /* TAG_MSQRT */ + {"mfenced", 0}, /* TAG_MFENCED */ + {"mtable", 0}, /* TAG_MTABLE */ + {"mtr", 0}, /* TAG_MTR */ + {"mtd", 0}, /* TAG_MTD */ + {"munderover", 0}, /* TAG_MUNDEROVER */ + {"munder", 0}, /* TAG_MUNDER*/ + {"mover", 0}, /* TAG_MOVER*/ }; static const char *const htmlattrs[ATTR_MAX] = { - "http-equiv", /* ATTR_HTTPEQUIV */ - "content", /* ATTR_CONTENT */ "name", /* ATTR_NAME */ "rel", /* ATTR_REL */ "href", /* ATTR_HREF */ @@ -86,55 +100,68 @@ static const char *const htmlattrs[ATTR_MAX] = { "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 */ + "open", /* ATTR_OPEN */ + "close", /* ATTR_CLOSE */ + "mathvariant", /* ATTR_MATHVARIANT */ }; -static void print_num(struct html *, const char *, size_t); -static void print_spec(struct html *, const char *, size_t); -static void print_res(struct html *, const char *, size_t); -static void print_ctag(struct html *, enum htmltag); -static void print_doctype(struct html *); -static void print_xmltype(struct html *); -static int print_encode(struct html *, const char *, int); -static void print_metaf(struct html *, enum mandoc_esc); -static void print_attr(struct html *, - const char *, const char *); -static void *ml_alloc(char *, enum htmltype); +static const char *const roffscales[SCALE_MAX] = { + "cm", /* SCALE_CM */ + "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 bufncat(struct html *, const char *, size_t); +static void print_ctag(struct html *, enum htmltag); +static int print_escape(char); +static int print_encode(struct html *, const char *, int); +static void print_metaf(struct html *, enum mandoc_esc); +static void print_attr(struct html *, const char *, const char *); +static void *ml_alloc(char *); + static void * -ml_alloc(char *outopts, enum htmltype type) +ml_alloc(char *outopts) { struct html *h; - const char *toks[4]; + const char *toks[5]; char *v; toks[0] = "style"; toks[1] = "man"; toks[2] = "includes"; - toks[3] = NULL; + toks[3] = "fragment"; + toks[4] = NULL; h = mandoc_calloc(1, sizeof(struct html)); - h->type = type; h->tags.head = NULL; - h->symtab = chars_init(CHARS_HTML); + h->symtab = mchars_alloc(); while (outopts && *outopts) switch (getsubopt(&outopts, UNCONST(toks), &v)) { - case (0): + case 0: h->style = v; break; - case (1): + case 1: h->base_man = v; break; - case (2): + case 2: h->base_includes = v; break; + case 3: + h->oflags |= HTML_FRAGMENT; + break; default: break; } @@ -146,18 +173,16 @@ void * html_alloc(char *outopts) { - return(ml_alloc(outopts, HTML_HTML_4_01_STRICT)); + return(ml_alloc(outopts)); } - void * xhtml_alloc(char *outopts) { - return(ml_alloc(outopts, HTML_XHTML_1_0_STRICT)); + return(ml_alloc(outopts)); } - void html_free(void *p) { @@ -167,33 +192,37 @@ html_free(void *p) h = (struct html *)p; while ((tag = h->tags.head) != NULL) { - h->tags.head = tag->next; + h->tags.head = tag->next; free(tag); } - + if (h->symtab) - chars_free(h->symtab); + mchars_free(h->symtab); free(h); } - 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); + tag[0].key = ATTR_CHARSET; + tag[0].val = "utf-8"; + print_otag(h, TAG_META, 1, tag); - 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 default style-sheet. + */ + t = print_otag(h, TAG_STYLE, 0, NULL); + print_text(h, "table.head, table.foot { width: 100%; }\n" + "td.head-rtitle, td.foot-os { text-align: right; }\n" + "td.head-vol { text-align: center; }\n" + "table.foot td { width: 50%; }\n" + "table.head td { width: 33%; }\n" + "div.spacer { margin: 1em 0; }\n"); + print_tagq(h, t); if (h->style) { tag[0].key = ATTR_REL; @@ -208,72 +237,27 @@ print_gen_head(struct html *h) } } -/* ARGSUSED */ static void -print_num(struct html *h, const char *p, size_t len) -{ - const char *rhs; - - rhs = chars_num2char(p, len); - if (rhs) - putchar((int)*rhs); -} - -static void -print_spec(struct html *h, const char *p, size_t len) -{ - int cp; - const char *rhs; - size_t sz; - - if ((cp = chars_spec2cp(h->symtab, p, len)) > 0) { - printf("&#%d;", cp); - return; - } else if (-1 == cp && 1 == len) { - fwrite(p, 1, len, stdout); - return; - } else if (-1 == cp) - return; - - if (NULL != (rhs = chars_spec2str(h->symtab, p, len, &sz))) - fwrite(rhs, 1, sz, stdout); -} - - -static void -print_res(struct html *h, const char *p, size_t len) -{ - int cp; - const char *rhs; - size_t sz; - - if ((cp = chars_res2cp(h->symtab, p, len)) > 0) { - printf("&#%d;", cp); - return; - } else if (-1 == cp) - return; - - if (NULL != (rhs = chars_res2str(h->symtab, p, len, &sz))) - fwrite(rhs, 1, sz, stdout); -} - - -static void print_metaf(struct html *h, enum mandoc_esc deco) { enum htmlfont font; switch (deco) { - case (ESCAPE_FONTPREV): + case ESCAPE_FONTPREV: font = h->metal; break; - case (ESCAPE_FONTITALIC): + case ESCAPE_FONTITALIC: font = HTMLFONT_ITALIC; break; - case (ESCAPE_FONTBOLD): + case ESCAPE_FONTBOLD: font = HTMLFONT_BOLD; break; - case (ESCAPE_FONTROMAN): + case ESCAPE_FONTBI: + font = HTMLFONT_BI; + break; + case ESCAPE_FONT: + /* FALLTHROUGH */ + case ESCAPE_FONTROMAN: font = HTMLFONT_NONE; break; default: @@ -289,25 +273,124 @@ print_metaf(struct html *h, enum mandoc_esc deco) h->metal = h->metac; h->metac = font; - if (HTMLFONT_NONE != font) - h->metaf = HTMLFONT_BOLD == font ? - print_otag(h, TAG_B, 0, NULL) : - print_otag(h, TAG_I, 0, NULL); + switch (font) { + case HTMLFONT_ITALIC: + h->metaf = print_otag(h, TAG_I, 0, NULL); + break; + case HTMLFONT_BOLD: + h->metaf = print_otag(h, TAG_B, 0, NULL); + break; + case HTMLFONT_BI: + h->metaf = print_otag(h, TAG_B, 0, NULL); + print_otag(h, TAG_I, 0, NULL); + break; + default: + break; + } } +int +html_strlen(const char *cp) +{ + size_t rsz; + int skip, sz; + /* + * Account for escaped sequences within string length + * calculations. This follows the logic in term_strlen() as we + * must calculate the width of produced strings. + * Assume that characters are always width of "1". This is + * hacky, but it gets the job done for approximation of widths. + */ + + sz = 0; + skip = 0; + while (1) { + rsz = strcspn(cp, "\\"); + if (rsz) { + cp += rsz; + if (skip) { + skip = 0; + rsz--; + } + sz += rsz; + } + if ('\0' == *cp) + break; + cp++; + switch (mandoc_escape(&cp, NULL, NULL)) { + case ESCAPE_ERROR: + return(sz); + case ESCAPE_UNICODE: + /* FALLTHROUGH */ + case ESCAPE_NUMBERED: + /* FALLTHROUGH */ + case ESCAPE_SPECIAL: + if (skip) + skip = 0; + else + sz++; + break; + case ESCAPE_SKIPCHAR: + skip = 1; + break; + default: + break; + } + } + return(sz); +} + static int +print_escape(char c) +{ + + switch (c) { + case '<': + printf("<"); + break; + case '>': + printf(">"); + break; + case '&': + printf("&"); + break; + case '"': + printf("""); + break; + case ASCII_NBRSP: + putchar('-'); + break; + case ASCII_HYPH: + putchar('-'); + /* FALLTHROUGH */ + case ASCII_BREAK: + break; + default: + return(0); + } + return(1); +} + +static int print_encode(struct html *h, const char *p, int norecurse) { size_t sz; - int len, nospace; + int c, len, nospace; const char *seq; enum mandoc_esc esc; - static const char rejs[6] = { '\\', '<', '>', '&', ASCII_HYPH, '\0' }; + static const char rejs[9] = { '\\', '<', '>', '&', '"', + ASCII_NBRSP, ASCII_HYPH, ASCII_BREAK, '\0' }; nospace = 0; while ('\0' != *p) { + if (HTML_SKIPCHAR & h->flags && '\\' != *p) { + h->flags &= ~HTML_SKIPCHAR; + p++; + continue; + } + sz = strcspn(p, rejs); fwrite(p, 1, sz, stdout); @@ -316,49 +399,61 @@ print_encode(struct html *h, const char *p, int norecu if ('\0' == *p) break; - switch (*p++) { - case ('<'): - printf("<"); + if (print_escape(*p++)) continue; - case ('>'): - printf(">"); + + esc = mandoc_escape(&p, &seq, &len); + if (ESCAPE_ERROR == esc) + break; + + switch (esc) { + case ESCAPE_FONT: + /* FALLTHROUGH */ + case ESCAPE_FONTPREV: + /* FALLTHROUGH */ + case ESCAPE_FONTBOLD: + /* FALLTHROUGH */ + case ESCAPE_FONTITALIC: + /* FALLTHROUGH */ + case ESCAPE_FONTBI: + /* FALLTHROUGH */ + case ESCAPE_FONTROMAN: + if (0 == norecurse) + print_metaf(h, esc); continue; - case ('&'): - printf("&"); + case ESCAPE_SKIPCHAR: + h->flags |= HTML_SKIPCHAR; continue; - case (ASCII_HYPH): - putchar('-'); - continue; default: break; } - esc = mandoc_escape(&p, &seq, &len); - if (ESCAPE_ERROR == esc) - break; + if (h->flags & HTML_SKIPCHAR) { + h->flags &= ~HTML_SKIPCHAR; + continue; + } switch (esc) { - case (ESCAPE_NUMBERED): - print_num(h, seq, len); + case ESCAPE_UNICODE: + /* Skip past "u" header. */ + c = mchars_num2uc(seq + 1, len - 1); + if ('\0' != c) + printf("&#x%x;", c); break; - case (ESCAPE_PREDEF): - print_res(h, seq, len); + case ESCAPE_NUMBERED: + c = mchars_num2char(seq, len); + if ( ! ('\0' == c || print_escape(c))) + putchar(c); break; - case (ESCAPE_SPECIAL): - print_spec(h, seq, len); + case ESCAPE_SPECIAL: + c = mchars_spec2cp(h->symtab, seq, len); + if (c > 0) + printf("&#%d;", c); + else if (-1 == c && 1 == len && + !print_escape(*seq)) + putchar((int)*seq); break; - case (ESCAPE_FONTPREV): - /* FALLTHROUGH */ - case (ESCAPE_FONTBOLD): - /* FALLTHROUGH */ - case (ESCAPE_FONTITALIC): - /* FALLTHROUGH */ - case (ESCAPE_FONTROMAN): - if (norecurse) - break; - print_metaf(h, esc); - break; - case (ESCAPE_NOSPACE): + case ESCAPE_NOSPACE: if ('\0' == *p) nospace = 1; break; @@ -370,7 +465,6 @@ print_encode(struct html *h, const char *p, int norecu return(nospace); } - static void print_attr(struct html *h, const char *key, const char *val) { @@ -379,9 +473,8 @@ print_attr(struct html *h, const char *key, const char putchar('\"'); } - struct tag * -print_otag(struct html *h, enum htmltag tag, +print_otag(struct html *h, enum htmltag tag, int sz, const struct htmlpair *p) { int i; @@ -419,24 +512,10 @@ print_otag(struct html *h, enum htmltag tag, for (i = 0; i < sz; i++) print_attr(h, htmlattrs[p[i].key], p[i].val); - /* Add non-overridable attributes. */ + /* Accommodate for "well-formed" singleton escaping. */ - if (TAG_HTML == tag && HTML_XHTML_1_0_STRICT == h->type) { - print_attr(h, "xmlns", "http://www.w3.org/1999/xhtml"); - print_attr(h, "xml:lang", "en"); - print_attr(h, "lang", "en"); - } - - /* Accomodate for XML "well-formed" singleton escaping. */ - if (HTML_AUTOCLOSE & htmltags[tag].flags) - switch (h->type) { - case (HTML_XHTML_1_0_STRICT): - putchar('/'); - break; - default: - break; - } + putchar('/'); putchar('>'); @@ -448,61 +527,24 @@ print_otag(struct html *h, enum htmltag tag, return(t); } - static void print_ctag(struct html *h, enum htmltag tag) { - + printf("", htmltags[tag].name); if (HTML_CLRLINE & htmltags[tag].flags) { h->flags |= HTML_NOSPACE; putchar('\n'); - } + } } - void print_gen_decls(struct html *h) { - print_xmltype(h); - print_doctype(h); + puts(""); } - -static void -print_xmltype(struct html *h) -{ - - if (HTML_XHTML_1_0_STRICT == h->type) - puts(""); -} - - -static void -print_doctype(struct html *h) -{ - const char *doctype; - const char *dtd; - const char *name; - - switch (h->type) { - case (HTML_HTML_4_01_STRICT): - name = "HTML"; - doctype = "-//W3C//DTD HTML 4.01//EN"; - dtd = "http://www.w3.org/TR/html4/strict.dtd"; - break; - default: - name = "html"; - doctype = "-//W3C//DTD XHTML 1.0 Strict//EN"; - dtd = "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"; - break; - } - - printf("\n", - name, doctype, dtd); -} - void print_text(struct html *h, const char *word) { @@ -518,15 +560,27 @@ print_text(struct html *h, const char *word) } assert(NULL == h->metaf); - if (HTMLFONT_NONE != h->metac) - h->metaf = HTMLFONT_BOLD == h->metac ? - print_otag(h, TAG_B, 0, NULL) : - print_otag(h, TAG_I, 0, NULL); + switch (h->metac) { + case HTMLFONT_ITALIC: + h->metaf = print_otag(h, TAG_I, 0, NULL); + break; + case HTMLFONT_BOLD: + h->metaf = print_otag(h, TAG_B, 0, NULL); + break; + case HTMLFONT_BI: + h->metaf = print_otag(h, TAG_B, 0, NULL); + print_otag(h, TAG_I, 0, NULL); + break; + default: + break; + } assert(word); - if ( ! print_encode(h, word, 0)) + if ( ! print_encode(h, word, 0)) { if ( ! (h->flags & HTML_NONOSPACE)) h->flags &= ~HTML_NOSPACE; + } else + h->flags |= HTML_NOSPACE; if (h->metaf) { print_tagq(h, h->metaf); @@ -536,14 +590,13 @@ print_text(struct html *h, const char *word) h->flags &= ~HTML_IGNDELIM; } - void print_tagq(struct html *h, const struct tag *until) { struct tag *tag; while ((tag = h->tags.head) != NULL) { - /* + /* * Remember to close out and nullify the current * meta-font and table, if applicable. */ @@ -559,7 +612,6 @@ print_tagq(struct html *h, const struct tag *until) } } - void print_stagq(struct html *h, const struct tag *suntil) { @@ -568,7 +620,7 @@ print_stagq(struct html *h, const struct tag *suntil) while ((tag = h->tags.head) != NULL) { if (suntil && tag == suntil) return; - /* + /* * Remember to close out and nullify the current * meta-font and table, if applicable. */ @@ -582,7 +634,18 @@ print_stagq(struct html *h, const struct tag *suntil) } } +void +print_paragraph(struct html *h) +{ + struct tag *t; + struct htmlpair tag; + PAIR_CLASS_INIT(&tag, "spacer"); + t = print_otag(h, TAG_DIV, 1, &tag); + print_tagq(h, t); +} + + void bufinit(struct html *h) { @@ -591,62 +654,63 @@ bufinit(struct html *h) h->buflen = 0; } - void bufcat_style(struct html *h, const char *key, const char *val) { bufcat(h, key); - bufncat(h, ":", 1); + bufcat(h, ":"); bufcat(h, val); - bufncat(h, ";", 1); + bufcat(h, ";"); } - void bufcat(struct html *h, const char *p) { - bufncat(h, p, strlen(p)); + /* + * XXX This is broken and not easy to fix. + * When using the -Oincludes option, buffmt_includes() + * may pass in strings overrunning BUFSIZ, causing a crash. + */ + + h->buflen = strlcat(h->buf, p, BUFSIZ); + assert(h->buflen < BUFSIZ); } - void -buffmt(struct html *h, const char *fmt, ...) +bufcat_fmt(struct html *h, const char *fmt, ...) { va_list ap; va_start(ap, fmt); - (void)vsnprintf(h->buf + (int)h->buflen, - BUFSIZ - h->buflen - 1, fmt, ap); + (void)vsnprintf(h->buf + (int)h->buflen, + BUFSIZ - h->buflen - 1, fmt, ap); va_end(ap); h->buflen = strlen(h->buf); } - -void +static void bufncat(struct html *h, const char *p, size_t sz) { - if (h->buflen + sz > BUFSIZ - 1) - sz = BUFSIZ - 1 - h->buflen; - - (void)strncat(h->buf, p, sz); + assert(h->buflen + sz + 1 < BUFSIZ); + strncat(h->buf, p, sz); 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'): + case'I': bufcat(h, name); break; default: @@ -659,24 +723,22 @@ buffmt_includes(struct html *h, const char *name) bufcat(h, pp); } - void -buffmt_man(struct html *h, - const char *name, const char *sec) +buffmt_man(struct html *h, const char *name, const char *sec) { const char *p, *pp; pp = h->base_man; - - /* LINTED */ + + bufinit(h); while (NULL != (p = strchr(pp, '%'))) { bufncat(h, pp, (size_t)(p - pp)); switch (*(p + 1)) { - case('S'): + case 'S': bufcat(h, sec ? sec : "1"); break; - case('N'): - buffmt(h, name); + case 'N': + bufcat_fmt(h, "%s", name); break; default: bufncat(h, p, 2); @@ -688,85 +750,26 @@ buffmt_man(struct html *h, bufcat(h, pp); } - void bufcat_su(struct html *h, const char *p, const struct roffsu *su) { double v; - const char *u; v = su->scale; + if (SCALE_MM == su->unit && 0.0 == (v /= 100.0)) + v = 1.0; + else if (SCALE_BU == su->unit) + v /= 24.0; - switch (su->unit) { - case (SCALE_CM): - u = "cm"; - break; - case (SCALE_IN): - u = "in"; - break; - case (SCALE_PC): - u = "pc"; - break; - case (SCALE_PT): - u = "pt"; - break; - case (SCALE_EM): - u = "em"; - break; - case (SCALE_MM): - if (0 == (v /= 100)) - v = 1; - u = "em"; - break; - case (SCALE_EN): - u = "ex"; - break; - case (SCALE_BU): - u = "ex"; - break; - case (SCALE_VS): - u = "em"; - break; - default: - u = "ex"; - break; - } - - /* - * XXX: the CSS spec isn't clear as to which types accept - * integer or real numbers, so we just make them all decimals. - */ - buffmt(h, "%s: %.2f%s;", p, v, u); + bufcat_fmt(h, "%s: %.2f%s;", p, v, roffscales[su->unit]); } - void -html_idcat(char *dst, const char *src, int sz) +bufcat_id(struct html *h, const char *src) { - int ssz; - assert(sz > 2); - /* Cf. . */ - /* We can't start with a number (bah). */ - - if ('#' == *dst) { - dst++; - sz--; - } - if ('\0' == *dst) { - *dst++ = 'x'; - *dst = '\0'; - sz--; - } - - for ( ; *dst != '\0' && sz; dst++, sz--) - /* Jump to end. */ ; - - for ( ; *src != '\0' && sz > 1; src++) { - ssz = snprintf(dst, (size_t)sz, "%.2x", *src); - sz -= ssz; - dst += ssz; - } + while ('\0' != *src) + bufcat_fmt(h, "%.2x", *src++); }