=================================================================== RCS file: /cvs/mandoc/html.c,v retrieving revision 1.158 retrieving revision 1.179 diff -u -p -r1.158 -r1.179 --- mandoc/html.c 2014/07/22 22:41:35 1.158 +++ mandoc/html.c 2014/10/27 16:29:06 1.179 @@ -1,6 +1,6 @@ -/* $Id: html.c,v 1.158 2014/07/22 22:41:35 schwarze Exp $ */ +/* $Id: html.c,v 1.179 2014/10/27 16:29:06 schwarze Exp $ */ /* - * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons + * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons * Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any @@ -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 @@ -70,17 +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 */ @@ -88,11 +100,12 @@ 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 const char *const roffscales[SCALE_MAX] = { @@ -110,14 +123,15 @@ static const char *const roffscales[SCALE_MAX] = { 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 *, enum htmltype); +static void *ml_alloc(char *); static void * -ml_alloc(char *outopts, enum htmltype type) +ml_alloc(char *outopts) { struct html *h; const char *toks[5]; @@ -131,7 +145,6 @@ ml_alloc(char *outopts, enum htmltype type) h = mandoc_calloc(1, sizeof(struct html)); - h->type = type; h->tags.head = NULL; h->symtab = mchars_alloc(); @@ -160,14 +173,14 @@ 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 @@ -193,18 +206,23 @@ 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; @@ -324,6 +342,37 @@ html_strlen(const char *cp) } 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; @@ -350,30 +399,8 @@ 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(">"); - continue; - case '&': - printf("&"); - continue; - case '"': - printf("""); - continue; - case ASCII_NBRSP: - putchar('-'); - continue; - case ASCII_HYPH: - putchar('-'); - /* FALLTHROUGH */ - case ASCII_BREAK: - continue; - default: - break; - } esc = mandoc_escape(&p, &seq, &len); if (ESCAPE_ERROR == esc) @@ -408,30 +435,30 @@ print_encode(struct html *h, const char *p, int norecu switch (esc) { case ESCAPE_UNICODE: - /* Skip passed "u" header. */ + /* Skip past "u" header. */ c = mchars_num2uc(seq + 1, len - 1); - if ('\0' != c) - printf("&#x%x;", c); break; case ESCAPE_NUMBERED: c = mchars_num2char(seq, len); - if ('\0' != c) - putchar(c); break; case ESCAPE_SPECIAL: c = mchars_spec2cp(h->symtab, seq, len); - if (c > 0) - printf("&#%d;", c); - else if (-1 == c && 1 == len) - putchar((int)*seq); break; case ESCAPE_NOSPACE: if ('\0' == *p) nospace = 1; - break; + continue; default: - break; + continue; } + if (c <= 0) + continue; + if (c < 0x20 || (c > 0x7E && c < 0xA0)) + c = 0xFFFD; + if (c > 0x7E) + printf("&#%d;", c); + else if ( ! print_escape(c)) + putchar(c); } return(nospace); @@ -484,24 +511,10 @@ 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"); - } - - /* Accommodate 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('>'); @@ -527,26 +540,8 @@ print_ctag(struct html *h, enum htmltag tag) void print_gen_decls(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: - puts(""); - 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); + puts(""); } void @@ -639,6 +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) { @@ -750,6 +757,8 @@ bufcat_su(struct html *h, const char *p, const struct 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; bufcat_fmt(h, "%s: %.2f%s;", p, v, roffscales[su->unit]); }