=================================================================== RCS file: /cvs/mandoc/html.c,v retrieving revision 1.148 retrieving revision 1.155 diff -u -p -r1.148 -r1.155 --- mandoc/html.c 2011/07/04 09:42:38 1.148 +++ mandoc/html.c 2014/03/23 11:25:26 1.155 @@ -1,7 +1,7 @@ -/* $Id: html.c,v 1.148 2011/07/04 09:42:38 kristaps Exp $ */ +/* $Id: html.c,v 1.155 2014/03/23 11:25:26 schwarze Exp $ */ /* * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons - * Copyright (c) 2011 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" @@ -118,13 +119,14 @@ 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)); @@ -143,6 +145,9 @@ ml_alloc(char *outopts, enum htmltype type) case (2): h->base_includes = v; break; + case (3): + h->oflags |= HTML_FRAGMENT; + break; default: break; } @@ -231,6 +236,9 @@ print_metaf(struct html *h, enum mandoc_esc deco) case (ESCAPE_FONTBOLD): font = HTMLFONT_BOLD; break; + case (ESCAPE_FONTBI): + font = HTMLFONT_BI; + break; case (ESCAPE_FONT): /* FALLTHROUGH */ case (ESCAPE_FONTROMAN): @@ -249,17 +257,27 @@ 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) { - int ssz, sz; - const char *seq, *p; + size_t rsz; + int skip, sz; /* * Account for escaped sequences within string length @@ -270,10 +288,21 @@ 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_UNICODE): @@ -281,15 +310,19 @@ html_strlen(const char *cp) 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 @@ -299,11 +332,18 @@ 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[8] = { '\\', '<', '>', '&', + 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); @@ -322,8 +362,13 @@ print_encode(struct html *h, const char *p, int norecu case ('&'): printf("&"); continue; + case (ASCII_NBRSP): + putchar('-'); + continue; case (ASCII_HYPH): putchar('-'); + /* FALLTHROUGH */ + case (ASCII_BREAK): continue; default: break; @@ -334,6 +379,33 @@ print_encode(struct html *h, const char *p, int norecu 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 (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); @@ -352,19 +424,6 @@ print_encode(struct html *h, const char *p, int norecu else if (-1 == c && 1 == len) putchar((int)*seq); break; - case (ESCAPE_FONT): - /* FALLTHROUGH */ - 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): if ('\0' == *p) nospace = 1; @@ -507,15 +566,27 @@ 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); @@ -658,7 +729,7 @@ buffmt_man(struct html *h, bufcat(h, sec ? sec : "1"); break; case('N'): - bufcat_fmt(h, name); + bufcat_fmt(h, "%s", name); break; default: bufncat(h, p, 2);