=================================================================== RCS file: /cvs/mandoc/html.c,v retrieving revision 1.153 retrieving revision 1.158 diff -u -p -r1.153 -r1.158 --- mandoc/html.c 2014/01/05 19:10:56 1.153 +++ mandoc/html.c 2014/07/22 22:41:35 1.158 @@ -1,7 +1,7 @@ -/* $Id: html.c,v 1.153 2014/01/05 19:10:56 joerg Exp $ */ +/* $Id: html.c,v 1.158 2014/07/22 22:41:35 schwarze Exp $ */ /* * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons - * Copyright (c) 2011, 2012, 2013 Ingo Schwarze + * 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 @@ -31,6 +31,7 @@ #include #include "mandoc.h" +#include "mandoc_aux.h" #include "libmandoc.h" #include "out.h" #include "html.h" @@ -114,6 +115,7 @@ static void print_metaf(struct html *, enum mandoc_es 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) { @@ -135,16 +137,16 @@ ml_alloc(char *outopts, enum htmltype type) 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): + case 3: h->oflags |= HTML_FRAGMENT; break; default: @@ -161,7 +163,6 @@ html_alloc(char *outopts) return(ml_alloc(outopts, HTML_HTML_4_01_STRICT)); } - void * xhtml_alloc(char *outopts) { @@ -169,7 +170,6 @@ xhtml_alloc(char *outopts) return(ml_alloc(outopts, HTML_XHTML_1_0_STRICT)); } - void html_free(void *p) { @@ -179,17 +179,16 @@ 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) mchars_free(h->symtab); free(h); } - void print_gen_head(struct html *h) { @@ -226,21 +225,21 @@ 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_FONTBI): + case ESCAPE_FONTBI: font = HTMLFONT_BI; break; - case (ESCAPE_FONT): + case ESCAPE_FONT: /* FALLTHROUGH */ - case (ESCAPE_FONTROMAN): + case ESCAPE_FONTROMAN: font = HTMLFONT_NONE; break; default: @@ -257,13 +256,13 @@ print_metaf(struct html *h, enum mandoc_esc deco) h->metac = font; switch (font) { - case (HTMLFONT_ITALIC): + case HTMLFONT_ITALIC: h->metaf = print_otag(h, TAG_I, 0, NULL); break; - case (HTMLFONT_BOLD): + case HTMLFONT_BOLD: h->metaf = print_otag(h, TAG_B, 0, NULL); break; - case (HTMLFONT_BI): + case HTMLFONT_BI: h->metaf = print_otag(h, TAG_B, 0, NULL); print_otag(h, TAG_I, 0, NULL); break; @@ -302,19 +301,19 @@ html_strlen(const char *cp) break; cp++; switch (mandoc_escape(&cp, NULL, NULL)) { - case (ESCAPE_ERROR): + case ESCAPE_ERROR: return(sz); - case (ESCAPE_UNICODE): + case ESCAPE_UNICODE: /* FALLTHROUGH */ - case (ESCAPE_NUMBERED): + case ESCAPE_NUMBERED: /* FALLTHROUGH */ - case (ESCAPE_SPECIAL): + case ESCAPE_SPECIAL: if (skip) skip = 0; else sz++; break; - case (ESCAPE_SKIPCHAR): + case ESCAPE_SKIPCHAR: skip = 1; break; default: @@ -331,7 +330,8 @@ print_encode(struct html *h, const char *p, int norecu 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; @@ -351,18 +351,26 @@ print_encode(struct html *h, const char *p, int norecu break; switch (*p++) { - case ('<'): + case '<': printf("<"); continue; - case ('>'): + case '>': printf(">"); continue; - case ('&'): + case '&': printf("&"); continue; - case (ASCII_HYPH): + case '"': + printf("""); + continue; + case ASCII_NBRSP: putchar('-'); continue; + case ASCII_HYPH: + putchar('-'); + /* FALLTHROUGH */ + case ASCII_BREAK: + continue; default: break; } @@ -372,21 +380,21 @@ print_encode(struct html *h, const char *p, int norecu break; 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_FONTROMAN): + case ESCAPE_FONTROMAN: if (0 == norecurse) print_metaf(h, esc); continue; - case (ESCAPE_SKIPCHAR): + case ESCAPE_SKIPCHAR: h->flags |= HTML_SKIPCHAR; continue; default: @@ -399,25 +407,25 @@ print_encode(struct html *h, const char *p, int norecu } switch (esc) { - case (ESCAPE_UNICODE): + 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): + case ESCAPE_NUMBERED: c = mchars_num2char(seq, len); if ('\0' != c) putchar(c); break; - case (ESCAPE_SPECIAL): + 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): + case ESCAPE_NOSPACE: if ('\0' == *p) nospace = 1; break; @@ -429,7 +437,6 @@ 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) { @@ -438,9 +445,8 @@ 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; @@ -490,7 +496,7 @@ print_otag(struct html *h, enum htmltag tag, if (HTML_AUTOCLOSE & htmltags[tag].flags) switch (h->type) { - case (HTML_XHTML_1_0_STRICT): + case HTML_XHTML_1_0_STRICT: putchar('/'); break; default: @@ -507,16 +513,15 @@ 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 @@ -527,7 +532,7 @@ print_gen_decls(struct html *h) const char *name; switch (h->type) { - case (HTML_HTML_4_01_STRICT): + case HTML_HTML_4_01_STRICT: name = "HTML"; doctype = "-//W3C//DTD HTML 4.01//EN"; dtd = "http://www.w3.org/TR/html4/strict.dtd"; @@ -540,8 +545,8 @@ print_gen_decls(struct html *h) break; } - printf("\n", - name, doctype, dtd); + printf("\n", + name, doctype, dtd); } void @@ -560,13 +565,13 @@ print_text(struct html *h, const char *word) assert(NULL == h->metaf); switch (h->metac) { - case (HTMLFONT_ITALIC): + case HTMLFONT_ITALIC: h->metaf = print_otag(h, TAG_I, 0, NULL); break; - case (HTMLFONT_BOLD): + case HTMLFONT_BOLD: h->metaf = print_otag(h, TAG_B, 0, NULL); break; - case (HTMLFONT_BI): + case HTMLFONT_BI: h->metaf = print_otag(h, TAG_B, 0, NULL); print_otag(h, TAG_I, 0, NULL); break; @@ -589,14 +594,13 @@ 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. */ @@ -612,7 +616,6 @@ print_tagq(struct html *h, const struct tag *until) } } - void print_stagq(struct html *h, const struct tag *suntil) { @@ -621,7 +624,7 @@ 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. */ @@ -657,6 +660,12 @@ void bufcat(struct html *h, const char *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); } @@ -667,8 +676,8 @@ 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); } @@ -688,12 +697,12 @@ 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: @@ -707,21 +716,20 @@ buffmt_includes(struct html *h, const char *name) } 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; - + 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'): + case 'N': bufcat_fmt(h, "%s", name); break; default: