=================================================================== RCS file: /cvs/mandoc/html.c,v retrieving revision 1.138 retrieving revision 1.151 diff -u -p -r1.138 -r1.151 --- mandoc/html.c 2011/05/14 16:28:23 1.138 +++ mandoc/html.c 2012/05/31 22:29:13 1.151 @@ -1,7 +1,7 @@ -/* $Id: html.c,v 1.138 2011/05/14 16:28:23 kristaps Exp $ */ +/* $Id: html.c,v 1.151 2012/05/31 22:29:13 schwarze Exp $ */ /* * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons - * Copyright (c) 2011 Ingo Schwarze + * Copyright (c) 2011, 2012 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 @@ -94,30 +94,38 @@ static const char *const htmlattrs[ATTR_MAX] = { "colspan", /* ATTR_COLSPAN */ }; -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_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 *outopts, enum htmltype type) { 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)); @@ -136,6 +144,9 @@ ml_alloc(char *outopts, enum htmltype type) case (2): h->base_includes = v; break; + case (3): + h->oflags |= HTML_FRAGMENT; + break; default: break; } @@ -209,56 +220,7 @@ print_gen_head(struct html *h) } } -/* ARGSUSED */ static void -print_num(struct html *h, const char *p, size_t len) -{ - char c; - - if ('\0' != (c = mchars_num2char(p, len))) - putchar((int)c); -} - -static void -print_spec(struct html *h, const char *p, size_t len) -{ - int cp; - const char *rhs; - size_t sz; - - if ((cp = mchars_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 = mchars_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 = mchars_res2cp(h->symtab, p, len)) > 0) { - printf("&#%d;", cp); - return; - } else if (-1 == cp) - return; - - if (NULL != (rhs = mchars_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; @@ -273,6 +235,8 @@ print_metaf(struct html *h, enum mandoc_esc deco) case (ESCAPE_FONTBOLD): font = HTMLFONT_BOLD; break; + case (ESCAPE_FONT): + /* FALLTHROUGH */ case (ESCAPE_FONTROMAN): font = HTMLFONT_NONE; break; @@ -298,8 +262,8 @@ print_metaf(struct html *h, enum mandoc_esc deco) int html_strlen(const char *cp) { - int ssz, sz; - const char *seq, *p; + size_t rsz; + int skip, sz; /* * Account for escaped sequences within string length @@ -310,32 +274,48 @@ html_strlen(const char *cp) */ sz = 0; - while (NULL != (p = strchr(cp, '\\'))) { - sz += (int)(p - cp); - ++cp; - switch (mandoc_escape(&cp, &seq, &ssz)) { + 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_PREDEF): - sz++; - break; + case (ESCAPE_UNICODE): + /* FALLTHROUGH */ + case (ESCAPE_NUMBERED): + /* FALLTHROUGH */ case (ESCAPE_SPECIAL): - sz++; + if (skip) + skip = 0; + else + sz++; break; + case (ESCAPE_SKIPCHAR): + skip = 1; + break; default: break; } } - - assert(sz >= 0); - return(sz + strlen(cp)); + return(sz); } 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' }; @@ -343,6 +323,12 @@ print_encode(struct html *h, const char *p, int norecu 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); @@ -373,15 +359,8 @@ print_encode(struct html *h, const char *p, int norecu break; switch (esc) { - case (ESCAPE_NUMBERED): - print_num(h, seq, len); - break; - case (ESCAPE_PREDEF): - print_res(h, seq, len); - break; - case (ESCAPE_SPECIAL): - print_spec(h, seq, len); - break; + case (ESCAPE_FONT): + /* FALLTHROUGH */ case (ESCAPE_FONTPREV): /* FALLTHROUGH */ case (ESCAPE_FONTBOLD): @@ -389,10 +368,40 @@ print_encode(struct html *h, const char *p, int norecu case (ESCAPE_FONTITALIC): /* FALLTHROUGH */ case (ESCAPE_FONTROMAN): - if (norecurse) - break; - print_metaf(h, esc); + if (0 == norecurse) + print_metaf(h, esc); + continue; + case (ESCAPE_SKIPCHAR): + h->flags |= HTML_SKIPCHAR; + continue; + default: break; + } + + if (h->flags & HTML_SKIPCHAR) { + h->flags &= ~HTML_SKIPCHAR; + continue; + } + + switch (esc) { + case (ESCAPE_UNICODE): + /* Skip passed "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; @@ -495,28 +504,9 @@ print_ctag(struct html *h, enum htmltag tag) } } - void print_gen_decls(struct html *h) { - - print_xmltype(h); - print_doctype(h); -} - - -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; @@ -528,6 +518,7 @@ print_doctype(struct html *h) 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"; @@ -559,9 +550,11 @@ print_text(struct html *h, const char *word) print_otag(h, TAG_I, 0, NULL); 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); @@ -617,7 +610,6 @@ print_stagq(struct html *h, const struct tag *suntil) } } - void bufinit(struct html *h) { @@ -626,28 +618,26 @@ 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)); + 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; @@ -658,19 +648,15 @@ buffmt(struct html *h, const char *fmt, ...) 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) { @@ -678,6 +664,7 @@ buffmt_includes(struct html *h, const char *name) pp = h->base_includes; + bufinit(h); while (NULL != (p = strchr(pp, '%'))) { bufncat(h, pp, (size_t)(p - pp)); switch (*(p + 1)) { @@ -694,7 +681,6 @@ buffmt_includes(struct html *h, const char *name) bufcat(h, pp); } - void buffmt_man(struct html *h, const char *name, const char *sec) @@ -703,7 +689,7 @@ buffmt_man(struct html *h, pp = h->base_man; - /* LINTED */ + bufinit(h); while (NULL != (p = strchr(pp, '%'))) { bufncat(h, pp, (size_t)(p - pp)); switch (*(p + 1)) { @@ -711,7 +697,7 @@ buffmt_man(struct html *h, bufcat(h, sec ? sec : "1"); break; case('N'): - buffmt(h, name); + bufcat_fmt(h, name); break; default: bufncat(h, p, 2); @@ -723,85 +709,24 @@ 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; - 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++); }