=================================================================== RCS file: /cvs/mandoc/html.c,v retrieving revision 1.32 retrieving revision 1.148 diff -u -p -r1.32 -r1.148 --- mandoc/html.c 2009/09/17 08:21:42 1.32 +++ mandoc/html.c 2011/07/04 09:42:38 1.148 @@ -1,6 +1,7 @@ -/* $Id: html.c,v 1.32 2009/09/17 08:21:42 kristaps Exp $ */ +/* $Id: html.c,v 1.148 2011/07/04 09:42:38 kristaps Exp $ */ /* - * Copyright (c) 2008, 2009 Kristaps Dzonsons + * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons + * Copyright (c) 2011 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 @@ -14,318 +15,157 @@ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ -#include +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif +#include + #include -#include +#include +#include #include +#include #include +#include +#include -#include "chars.h" -#include "mdoc.h" -#include "man.h" +#include "mandoc.h" +#include "libmandoc.h" +#include "out.h" +#include "html.h" +#include "main.h" -#define DOCTYPE "-//W3C//DTD HTML 4.01//EN" -#define DTD "http://www.w3.org/TR/html4/strict.dtd" - -enum htmltag { - TAG_HTML, - TAG_HEAD, - TAG_BODY, - TAG_META, - TAG_TITLE, - TAG_DIV, - TAG_H1, - TAG_H2, - TAG_P, - TAG_SPAN, - TAG_LINK, - TAG_BR, - TAG_A, - TAG_MAX -}; - -enum htmlattr { - ATTR_HTTPEQUIV, - ATTR_CONTENT, - ATTR_NAME, - ATTR_REL, - ATTR_HREF, - ATTR_TYPE, - ATTR_MEDIA, - ATTR_CLASS, - ATTR_MAX -}; - struct htmldata { - char *name; + const char *name; int flags; #define HTML_CLRLINE (1 << 0) #define HTML_NOSTACK (1 << 1) +#define HTML_AUTOCLOSE (1 << 2) /* Tag has auto-closure. */ }; static const struct htmldata htmltags[TAG_MAX] = { {"html", HTML_CLRLINE}, /* TAG_HTML */ {"head", HTML_CLRLINE}, /* TAG_HEAD */ {"body", HTML_CLRLINE}, /* TAG_BODY */ - {"meta", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_META */ - {"title", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_TITLE */ + {"meta", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_META */ + {"title", HTML_CLRLINE}, /* TAG_TITLE */ {"div", HTML_CLRLINE}, /* TAG_DIV */ {"h1", 0}, /* TAG_H1 */ {"h2", 0}, /* TAG_H2 */ - {"p", HTML_CLRLINE}, /* TAG_P */ {"span", 0}, /* TAG_SPAN */ - {"link", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_LINK */ - {"br", HTML_CLRLINE | HTML_NOSTACK}, /* TAG_LINK */ + {"link", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_LINK */ + {"br", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_BR */ {"a", 0}, /* TAG_A */ + {"table", HTML_CLRLINE}, /* TAG_TABLE */ + {"tbody", HTML_CLRLINE}, /* TAG_TBODY */ + {"col", HTML_CLRLINE | HTML_NOSTACK | HTML_AUTOCLOSE}, /* TAG_COL */ + {"tr", HTML_CLRLINE}, /* TAG_TR */ + {"td", HTML_CLRLINE}, /* TAG_TD */ + {"li", HTML_CLRLINE}, /* TAG_LI */ + {"ul", HTML_CLRLINE}, /* TAG_UL */ + {"ol", HTML_CLRLINE}, /* TAG_OL */ + {"dl", HTML_CLRLINE}, /* TAG_DL */ + {"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 */ }; -static const char *const htmlattrs[ATTR_MAX] = { - "http-equiv", - "content", - "name", - "rel", - "href", - "type", - "media", - "class" +static const char *const htmlattrs[ATTR_MAX] = { + "http-equiv", /* ATTR_HTTPEQUIV */ + "content", /* ATTR_CONTENT */ + "name", /* ATTR_NAME */ + "rel", /* ATTR_REL */ + "href", /* ATTR_HREF */ + "type", /* ATTR_TYPE */ + "media", /* ATTR_MEDIA */ + "class", /* ATTR_CLASS */ + "style", /* ATTR_STYLE */ + "width", /* ATTR_WIDTH */ + "id", /* ATTR_ID */ + "summary", /* ATTR_SUMMARY */ + "align", /* ATTR_ALIGN */ + "colspan", /* ATTR_COLSPAN */ }; -struct htmlpair { - enum htmlattr key; - char *val; +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 */ }; -struct tag { - enum htmltag tag; - SLIST_ENTRY(tag) entry; -}; +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); -SLIST_HEAD(tagq, tag); +static void * +ml_alloc(char *outopts, enum htmltype type) +{ + struct html *h; + const char *toks[4]; + char *v; -struct html { - int flags; -#define HTML_NOSPACE (1 << 0) -#define HTML_NEWLINE (1 << 1) - struct tagq stack; - void *symtab; -}; + toks[0] = "style"; + toks[1] = "man"; + toks[2] = "includes"; + toks[3] = NULL; -#define MDOC_ARGS const struct mdoc_meta *m, \ - const struct mdoc_node *n, \ - struct html *h -#define MAN_ARGS const struct man_meta *m, \ - const struct man_node *n, \ - struct html *h -struct htmlmdoc { - int (*pre)(MDOC_ARGS); - void (*post)(MDOC_ARGS); -}; + h = mandoc_calloc(1, sizeof(struct html)); -static void print_gen_doctype(struct html *); -static void print_gen_head(struct html *); -static void print_mdoc(MDOC_ARGS); -static void print_mdoc_head(MDOC_ARGS); -static void print_mdoc_title(MDOC_ARGS); -static void print_mdoc_node(MDOC_ARGS); -static void print_man(MAN_ARGS); -static void print_man_head(MAN_ARGS); -static void print_man_body(MAN_ARGS); -static struct tag *print_otag(struct html *, enum htmltag, - int, const struct htmlpair *); -static void print_tagq(struct html *, const struct tag *); -static void print_stagq(struct html *, const struct tag *); -static void print_ctag(struct html *, enum htmltag); -static void print_encode(struct html *, const char *); -static void print_escape(struct html *, const char **); -static void print_text(struct html *, const char *); -static void print_res(struct html *, const char *, int); -static void print_spec(struct html *, const char *, int); -static int mdoc_root_pre(MDOC_ARGS); + h->type = type; + h->tags.head = NULL; + h->symtab = mchars_alloc(); -static int mdoc_ar_pre(MDOC_ARGS); -static int mdoc_fl_pre(MDOC_ARGS); -static int mdoc_nd_pre(MDOC_ARGS); -static int mdoc_nm_pre(MDOC_ARGS); -static int mdoc_ns_pre(MDOC_ARGS); -static int mdoc_op_pre(MDOC_ARGS); -static void mdoc_op_post(MDOC_ARGS); -static int mdoc_pp_pre(MDOC_ARGS); -static int mdoc_sh_pre(MDOC_ARGS); -static int mdoc_ss_pre(MDOC_ARGS); -static int mdoc_xr_pre(MDOC_ARGS); + while (outopts && *outopts) + switch (getsubopt(&outopts, UNCONST(toks), &v)) { + case (0): + h->style = v; + break; + case (1): + h->base_man = v; + break; + case (2): + h->base_includes = v; + break; + default: + break; + } -static const struct htmlmdoc mdocs[MDOC_MAX] = { - {NULL, NULL}, /* Ap */ - {NULL, NULL}, /* Dd */ - {NULL, NULL}, /* Dt */ - {NULL, NULL}, /* Os */ - {mdoc_sh_pre, NULL }, /* Sh */ - {mdoc_ss_pre, NULL }, /* Ss */ - {mdoc_pp_pre, NULL}, /* Pp */ - {NULL, NULL}, /* D1 */ - {NULL, NULL}, /* Dl */ - {NULL, NULL}, /* Bd */ - {NULL, NULL}, /* Ed */ - {NULL, NULL}, /* Bl */ - {NULL, NULL}, /* El */ - {NULL, NULL}, /* It */ - {NULL, NULL}, /* Ad */ - {NULL, NULL}, /* An */ - {mdoc_ar_pre, NULL}, /* Ar */ - {NULL, NULL}, /* Cd */ - {NULL, NULL}, /* Cm */ - {NULL, NULL}, /* Dv */ - {NULL, NULL}, /* Er */ - {NULL, NULL}, /* Ev */ - {NULL, NULL}, /* Ex */ - {NULL, NULL}, /* Fa */ - {NULL, NULL}, /* Fd */ - {mdoc_fl_pre, NULL}, /* Fl */ - {NULL, NULL}, /* Fn */ - {NULL, NULL}, /* Ft */ - {NULL, NULL}, /* Ic */ - {NULL, NULL}, /* In */ - {NULL, NULL}, /* Li */ - {mdoc_nd_pre, NULL}, /* Nd */ - {mdoc_nm_pre, NULL}, /* Nm */ - {mdoc_op_pre, mdoc_op_post}, /* Op */ - {NULL, NULL}, /* Ot */ - {NULL, NULL}, /* Pa */ - {NULL, NULL}, /* Rv */ - {NULL, NULL}, /* St */ - {NULL, NULL}, /* Va */ - {NULL, NULL}, /* Vt */ - {mdoc_xr_pre, NULL}, /* Xr */ - {NULL, NULL}, /* %A */ - {NULL, NULL}, /* %B */ - {NULL, NULL}, /* %D */ - {NULL, NULL}, /* %I */ - {NULL, NULL}, /* %J */ - {NULL, NULL}, /* %N */ - {NULL, NULL}, /* %O */ - {NULL, NULL}, /* %P */ - {NULL, NULL}, /* %R */ - {NULL, NULL}, /* %T */ - {NULL, NULL}, /* %V */ - {NULL, NULL}, /* Ac */ - {NULL, NULL}, /* Ao */ - {NULL, NULL}, /* Aq */ - {NULL, NULL}, /* At */ - {NULL, NULL}, /* Bc */ - {NULL, NULL}, /* Bf */ - {NULL, NULL}, /* Bo */ - {NULL, NULL}, /* Bq */ - {NULL, NULL}, /* Bsx */ - {NULL, NULL}, /* Bx */ - {NULL, NULL}, /* Db */ - {NULL, NULL}, /* Dc */ - {NULL, NULL}, /* Do */ - {NULL, NULL}, /* Dq */ - {NULL, NULL}, /* Ec */ - {NULL, NULL}, /* Ef */ - {NULL, NULL}, /* Em */ - {NULL, NULL}, /* Eo */ - {NULL, NULL}, /* Fx */ - {NULL, NULL}, /* Ms */ - {NULL, NULL}, /* No */ - {mdoc_ns_pre, NULL}, /* Ns */ - {NULL, NULL}, /* Nx */ - {NULL, NULL}, /* Ox */ - {NULL, NULL}, /* Pc */ - {NULL, NULL}, /* Pf */ - {NULL, NULL}, /* Po */ - {NULL, NULL}, /* Pq */ - {NULL, NULL}, /* Qc */ - {NULL, NULL}, /* Ql */ - {NULL, NULL}, /* Qo */ - {NULL, NULL}, /* Qq */ - {NULL, NULL}, /* Re */ - {NULL, NULL}, /* Rs */ - {NULL, NULL}, /* Sc */ - {NULL, NULL}, /* So */ - {NULL, NULL}, /* Sq */ - {NULL, NULL}, /* Sm */ - {NULL, NULL}, /* Sx */ - {NULL, NULL}, /* Sy */ - {NULL, NULL}, /* Tn */ - {NULL, NULL}, /* Ux */ - {NULL, NULL}, /* Xc */ - {NULL, NULL}, /* Xo */ - {NULL, NULL}, /* Fo */ - {NULL, NULL}, /* Fc */ - {NULL, NULL}, /* Oo */ - {NULL, NULL}, /* Oc */ - {NULL, NULL}, /* Bk */ - {NULL, NULL}, /* Ek */ - {NULL, NULL}, /* Bt */ - {NULL, NULL}, /* Hf */ - {NULL, NULL}, /* Fr */ - {NULL, NULL}, /* Ud */ - {NULL, NULL}, /* Lb */ - {NULL, NULL}, /* Lp */ - {NULL, NULL}, /* Lk */ - {NULL, NULL}, /* Mt */ - {NULL, NULL}, /* Brq */ - {NULL, NULL}, /* Bro */ - {NULL, NULL}, /* Brc */ - {NULL, NULL}, /* %C */ - {NULL, NULL}, /* Es */ - {NULL, NULL}, /* En */ - {NULL, NULL}, /* Dx */ - {NULL, NULL}, /* %Q */ - {NULL, NULL}, /* br */ - {NULL, NULL}, /* sp */ -}; - -void -html_mdoc(void *arg, const struct mdoc *m) -{ - struct html *h; - struct tag *t; - - h = (struct html *)arg; - - print_gen_doctype(h); - t = print_otag(h, TAG_HTML, 0, NULL); - print_mdoc(mdoc_meta(m), mdoc_node(m), h); - print_tagq(h, t); - - printf("\n"); + return(h); } -void -html_man(void *arg, const struct man *m) +void * +html_alloc(char *outopts) { - struct html *h; - struct tag *t; - h = (struct html *)arg; - - print_gen_doctype(h); - t = print_otag(h, TAG_HTML, 0, NULL); - print_man(man_meta(m), man_node(m), h); - print_tagq(h, t); - - printf("\n"); + return(ml_alloc(outopts, HTML_HTML_4_01_STRICT)); } + void * -html_alloc(void) +xhtml_alloc(char *outopts) { - struct html *h; - if (NULL == (h = calloc(1, sizeof(struct html)))) - return(NULL); - - SLIST_INIT(&h->stack); - if (NULL == (h->symtab = chars_init(CHARS_HTML))) { - free(h); - return(NULL); - } - return(h); + return(ml_alloc(outopts, HTML_XHTML_1_0_STRICT)); } + void html_free(void *p) { @@ -334,420 +174,374 @@ html_free(void *p) h = (struct html *)p; - while ( ! SLIST_EMPTY(&h->stack)) { - tag = SLIST_FIRST(&h->stack); - SLIST_REMOVE_HEAD(&h->stack, entry); + while ((tag = h->tags.head) != NULL) { + h->tags.head = tag->next; free(tag); } + + if (h->symtab) + mchars_free(h->symtab); + free(h); } -static void -print_mdoc(MDOC_ARGS) -{ - struct tag *t; - t = print_otag(h, TAG_HEAD, 0, NULL); - print_mdoc_head(m, n, h); - print_tagq(h, t); - - t = print_otag(h, TAG_BODY, 0, NULL); - print_mdoc_title(m, n, h); - print_mdoc_node(m, n, h); - print_tagq(h, t); -} - -static void +void print_gen_head(struct html *h) { - struct htmlpair meta0[2]; - struct htmlpair meta1[2]; - struct htmlpair link[4]; + struct htmlpair tag[4]; - meta0[0].key = ATTR_HTTPEQUIV; - meta0[0].val = "Content-Type"; - meta0[1].key = ATTR_CONTENT; - meta0[1].val = "text/html; charest-utf-8"; + 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); - meta1[0].key = ATTR_NAME; - meta1[0].val = "resource-type"; - meta1[1].key = ATTR_CONTENT; - meta1[1].val = "document"; + 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); - link[0].key = ATTR_REL; - link[0].val = "stylesheet"; - link[1].key = ATTR_HREF; - link[1].val = "style.css"; /* XXX */ - link[2].key = ATTR_TYPE; - link[2].val = "text/css"; - link[3].key = ATTR_MEDIA; - link[3].val = "all"; - - print_otag(h, TAG_META, 2, meta0); - print_otag(h, TAG_META, 2, meta1); - print_otag(h, TAG_LINK, 4, link); + if (h->style) { + tag[0].key = ATTR_REL; + tag[0].val = "stylesheet"; + tag[1].key = ATTR_HREF; + tag[1].val = h->style; + tag[2].key = ATTR_TYPE; + tag[2].val = "text/css"; + tag[3].key = ATTR_MEDIA; + tag[3].val = "all"; + print_otag(h, TAG_LINK, 4, tag); + } } -/* ARGSUSED */ static void -print_mdoc_head(MDOC_ARGS) +print_metaf(struct html *h, enum mandoc_esc deco) { + enum htmlfont font; - print_gen_head(h); - print_otag(h, TAG_TITLE, 0, NULL); - print_encode(h, m->title); -} - -/* ARGSUSED */ -static void -print_mdoc_title(MDOC_ARGS) -{ - - /* TODO */ -} - -static void -print_mdoc_node(MDOC_ARGS) -{ - int child; - struct tag *t; - - child = 1; - t = SLIST_FIRST(&h->stack); - - switch (n->type) { - case (MDOC_ROOT): - child = mdoc_root_pre(m, n, h); + switch (deco) { + case (ESCAPE_FONTPREV): + font = h->metal; break; - case (MDOC_TEXT): - print_text(h, n->string); + case (ESCAPE_FONTITALIC): + font = HTMLFONT_ITALIC; break; - default: - if (mdocs[n->tok].pre) - child = (*mdocs[n->tok].pre)(m, n, h); + case (ESCAPE_FONTBOLD): + font = HTMLFONT_BOLD; break; - } - - if (child && n->child) - print_mdoc_node(m, n->child, h); - - print_stagq(h, t); - - switch (n->type) { - case (MDOC_ROOT): + case (ESCAPE_FONT): + /* FALLTHROUGH */ + case (ESCAPE_FONTROMAN): + font = HTMLFONT_NONE; break; - case (MDOC_TEXT): - break; default: - if (mdocs[n->tok].post) - (*mdocs[n->tok].post)(m, n, h); - break; + abort(); + /* NOTREACHED */ } - if (n->next) - print_mdoc_node(m, n->next, h); -} + if (h->metaf) { + print_tagq(h, h->metaf); + h->metaf = NULL; + } -static void -print_man(MAN_ARGS) -{ - struct tag *t; + h->metal = h->metac; + h->metac = font; - t = print_otag(h, TAG_HEAD, 0, NULL); - print_man_head(m, n, h); - print_tagq(h, t); - - t = print_otag(h, TAG_BODY, 0, NULL); - print_man_body(m, n, h); - print_tagq(h, t); + if (HTMLFONT_NONE != font) + h->metaf = HTMLFONT_BOLD == font ? + print_otag(h, TAG_B, 0, NULL) : + print_otag(h, TAG_I, 0, NULL); } -/* ARGSUSED */ -static void -print_man_head(MAN_ARGS) +int +html_strlen(const char *cp) { + int ssz, sz; + const char *seq, *p; - print_gen_head(h); - print_otag(h, TAG_TITLE, 0, NULL); - print_encode(h, m->title); -} + /* + * 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. + */ -/* ARGSUSED */ -static void -print_man_body(MAN_ARGS) -{ + sz = 0; + while (NULL != (p = strchr(cp, '\\'))) { + sz += (int)(p - cp); + ++cp; + switch (mandoc_escape(&cp, &seq, &ssz)) { + case (ESCAPE_ERROR): + return(sz); + case (ESCAPE_UNICODE): + /* FALLTHROUGH */ + case (ESCAPE_NUMBERED): + /* FALLTHROUGH */ + case (ESCAPE_SPECIAL): + sz++; + break; + default: + break; + } + } - /* TODO */ + assert(sz >= 0); + return(sz + strlen(cp)); } -static void -print_spec(struct html *h, const char *p, int len) +static int +print_encode(struct html *h, const char *p, int norecurse) { - const char *rhs; - int i; size_t sz; + int c, len, nospace; + const char *seq; + enum mandoc_esc esc; + static const char rejs[6] = { '\\', '<', '>', '&', ASCII_HYPH, '\0' }; - rhs = chars_a2ascii(h->symtab, p, (size_t)len, &sz); + nospace = 0; - if (NULL == rhs) - return; - for (i = 0; i < (int)sz; i++) - putchar(rhs[i]); -} + while ('\0' != *p) { + sz = strcspn(p, rejs); -static void -print_res(struct html *h, const char *p, int len) -{ - const char *rhs; - int i; - size_t sz; + fwrite(p, 1, sz, stdout); + p += (int)sz; - rhs = chars_a2res(h->symtab, p, (size_t)len, &sz); + if ('\0' == *p) + break; - if (NULL == rhs) - return; - for (i = 0; i < (int)sz; i++) - putchar(rhs[i]); -} - -static void -print_escape(struct html *h, const char **p) -{ - int j, type; - const char *wp; - - wp = *p; - type = 1; - - if (0 == *(++wp)) { - *p = wp; - return; - } - - if ('(' == *wp) { - wp++; - if (0 == *wp || 0 == *(wp + 1)) { - *p = 0 == *wp ? wp : wp + 1; - return; + switch (*p++) { + case ('<'): + printf("<"); + continue; + case ('>'): + printf(">"); + continue; + case ('&'): + printf("&"); + continue; + case (ASCII_HYPH): + putchar('-'); + continue; + default: + break; } - print_spec(h, wp, 2); - *p = ++wp; - return; - - } else if ('*' == *wp) { - if (0 == *(++wp)) { - *p = wp; - return; - } - - switch (*wp) { - case ('('): - wp++; - if (0 == *wp || 0 == *(wp + 1)) { - *p = 0 == *wp ? wp : wp + 1; - return; - } - - print_res(h, wp, 2); - *p = ++wp; - return; - case ('['): - type = 0; + esc = mandoc_escape(&p, &seq, &len); + if (ESCAPE_ERROR == esc) break; - default: - print_res(h, wp, 1); - *p = wp; - return; - } - - } else if ('f' == *wp) { - if (0 == *(++wp)) { - *p = wp; - return; - } - switch (*wp) { - case ('B'): - /* TODO */ + 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 ('I'): - /* TODO */ + case (ESCAPE_NUMBERED): + c = mchars_num2char(seq, len); + if ('\0' != c) + putchar(c); break; - case ('P'): + 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_FONT): /* FALLTHROUGH */ - case ('R'): - /* TODO */ + 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; + break; default: break; } - - *p = wp; - return; - - } else if ('[' != *wp) { - print_spec(h, wp, 1); - *p = wp; - return; } - wp++; - for (j = 0; *wp && ']' != *wp; wp++, j++) - /* Loop... */ ; - - if (0 == *wp) { - *p = wp; - return; - } - - if (type) - print_spec(h, wp - j, j); - else - print_res(h, wp - j, j); - - *p = wp; + return(nospace); } static void -print_encode(struct html *h, const char *p) +print_attr(struct html *h, const char *key, const char *val) { - - for (; *p; p++) { - if ('\\' != *p) { - putchar(*p); - continue; - } - print_escape(h, &p); - } + printf(" %s=\"", key); + (void)print_encode(h, val, 1); + putchar('\"'); } -static struct tag * +struct tag * print_otag(struct html *h, enum htmltag tag, int sz, const struct htmlpair *p) { int i; struct tag *t; + /* Push this tags onto the stack of open scopes. */ + if ( ! (HTML_NOSTACK & htmltags[tag].flags)) { - if (NULL == (t = malloc(sizeof(struct tag)))) - err(EXIT_FAILURE, "malloc"); + t = mandoc_malloc(sizeof(struct tag)); t->tag = tag; - SLIST_INSERT_HEAD(&h->stack, t, entry); + t->next = h->tags.head; + h->tags.head = t; } else t = NULL; if ( ! (HTML_NOSPACE & h->flags)) - if ( ! (HTML_CLRLINE & htmltags[tag].flags)) - printf(" "); + if ( ! (HTML_CLRLINE & htmltags[tag].flags)) { + /* Manage keeps! */ + if ( ! (HTML_KEEP & h->flags)) { + if (HTML_PREKEEP & h->flags) + h->flags |= HTML_KEEP; + putchar(' '); + } else + printf(" "); + } + if ( ! (h->flags & HTML_NONOSPACE)) + h->flags &= ~HTML_NOSPACE; + else + h->flags |= HTML_NOSPACE; + + /* Print out the tag name and attributes. */ + printf("<%s", htmltags[tag].name); - for (i = 0; i < sz; i++) { - printf(" %s=\"", htmlattrs[p[i].key]); - assert(p->val); - print_encode(h, p[i].val); - printf("\""); + for (i = 0; i < sz; i++) + print_attr(h, htmlattrs[p[i].key], p[i].val); + + /* Add non-overridable attributes. */ + + 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"); } - printf(">"); + /* 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('>'); + h->flags |= HTML_NOSPACE; - if (HTML_CLRLINE & htmltags[tag].flags) - h->flags |= HTML_NEWLINE; - else - h->flags &= ~HTML_NEWLINE; + if ((HTML_AUTOCLOSE | HTML_CLRLINE) & htmltags[tag].flags) + putchar('\n'); + return(t); } -/* ARGSUSED */ static void print_ctag(struct html *h, enum htmltag tag) { printf("", htmltags[tag].name); - if (HTML_CLRLINE & htmltags[tag].flags) + if (HTML_CLRLINE & htmltags[tag].flags) { h->flags |= HTML_NOSPACE; - if (HTML_CLRLINE & htmltags[tag].flags) - h->flags |= HTML_NEWLINE; - else - h->flags &= ~HTML_NEWLINE; + putchar('\n'); + } } - -/* ARGSUSED */ -static void -print_gen_doctype(struct html *h) +void +print_gen_decls(struct html *h) { - - printf("\n", DOCTYPE, DTD); -} + 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; + } -static void -print_text(struct html *h, const char *p) + printf("\n", + name, doctype, dtd); +} + +void +print_text(struct html *h, const char *word) { - if (*p && 0 == *(p + 1)) - switch (*p) { - case('.'): - /* FALLTHROUGH */ - case(','): - /* FALLTHROUGH */ - case(';'): - /* FALLTHROUGH */ - case(':'): - /* FALLTHROUGH */ - case('?'): - /* FALLTHROUGH */ - case('!'): - /* FALLTHROUGH */ - case(')'): - /* FALLTHROUGH */ - case(']'): - /* FALLTHROUGH */ - case('}'): - h->flags |= HTML_NOSPACE; - break; - default: - break; - } + if ( ! (HTML_NOSPACE & h->flags)) { + /* Manage keeps! */ + if ( ! (HTML_KEEP & h->flags)) { + if (HTML_PREKEEP & h->flags) + h->flags |= HTML_KEEP; + putchar(' '); + } else + printf(" "); + } - if ( ! (h->flags & HTML_NOSPACE)) - printf(" "); + 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); - h->flags &= ~HTML_NOSPACE; - h->flags &= ~HTML_NEWLINE; + assert(word); + if ( ! print_encode(h, word, 0)) + if ( ! (h->flags & HTML_NONOSPACE)) + h->flags &= ~HTML_NOSPACE; - if (p) - print_encode(h, p); + if (h->metaf) { + print_tagq(h, h->metaf); + h->metaf = NULL; + } - if (*p && 0 == *(p + 1)) - switch (*p) { - case('('): - /* FALLTHROUGH */ - case('['): - /* FALLTHROUGH */ - case('{'): - h->flags |= HTML_NOSPACE; - break; - default: - break; - } + h->flags &= ~HTML_IGNDELIM; } -static void +void print_tagq(struct html *h, const struct tag *until) { struct tag *tag; - while ( ! SLIST_EMPTY(&h->stack)) { - tag = SLIST_FIRST(&h->stack); + while ((tag = h->tags.head) != NULL) { + /* + * Remember to close out and nullify the current + * meta-font and table, if applicable. + */ + if (tag == h->metaf) + h->metaf = NULL; + if (tag == h->tblt) + h->tblt = NULL; print_ctag(h, tag->tag); - SLIST_REMOVE_HEAD(&h->stack, entry); + h->tags.head = tag->next; free(tag); if (until && tag == until) return; @@ -755,191 +549,145 @@ print_tagq(struct html *h, const struct tag *until) } -static void +void print_stagq(struct html *h, const struct tag *suntil) { struct tag *tag; - while ( ! SLIST_EMPTY(&h->stack)) { - tag = SLIST_FIRST(&h->stack); + 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. + */ + if (tag == h->metaf) + h->metaf = NULL; + if (tag == h->tblt) + h->tblt = NULL; print_ctag(h, tag->tag); - SLIST_REMOVE_HEAD(&h->stack, entry); + h->tags.head = tag->next; free(tag); } } - -/* ARGSUSED */ -static int -mdoc_root_pre(MDOC_ARGS) +void +bufinit(struct html *h) { - struct htmlpair tag; - tag.key = ATTR_CLASS; - tag.val = "body"; - - print_otag(h, TAG_DIV, 1, &tag); - return(1); + h->buf[0] = '\0'; + h->buflen = 0; } - -/* ARGSUSED */ -static int -mdoc_ss_pre(MDOC_ARGS) +void +bufcat_style(struct html *h, const char *key, const char *val) { - if (MDOC_BODY == n->type) - print_otag(h, TAG_P, 0, NULL); - if (MDOC_HEAD == n->type) - print_otag(h, TAG_H2, 0, NULL); - return(1); + bufcat(h, key); + bufcat(h, ":"); + bufcat(h, val); + bufcat(h, ";"); } - -/* ARGSUSED */ -static int -mdoc_fl_pre(MDOC_ARGS) +void +bufcat(struct html *h, const char *p) { - struct htmlpair tag; - tag.key = ATTR_CLASS; - tag.val = "flag"; - - print_otag(h, TAG_SPAN, 1, &tag); - print_text(h, "\\-"); - h->flags |= HTML_NOSPACE; - return(1); + h->buflen = strlcat(h->buf, p, BUFSIZ); + assert(h->buflen < BUFSIZ); } - -/* ARGSUSED */ -static int -mdoc_pp_pre(MDOC_ARGS) +void +bufcat_fmt(struct html *h, const char *fmt, ...) { + va_list ap; - print_otag(h, TAG_BR, 0, NULL); - print_otag(h, TAG_BR, 0, NULL); - return(0); + va_start(ap, fmt); + (void)vsnprintf(h->buf + (int)h->buflen, + BUFSIZ - h->buflen - 1, fmt, ap); + va_end(ap); + h->buflen = strlen(h->buf); } - -/* ARGSUSED */ -static int -mdoc_nd_pre(MDOC_ARGS) +static void +bufncat(struct html *h, const char *p, size_t sz) { - if (MDOC_BODY == n->type) - print_text(h, "\\(en"); - return(1); + assert(h->buflen + sz + 1 < BUFSIZ); + strncat(h->buf, p, sz); + h->buflen += sz; } - -/* ARGSUSED */ -static int -mdoc_op_pre(MDOC_ARGS) +void +buffmt_includes(struct html *h, const char *name) { + const char *p, *pp; - if (MDOC_BODY == n->type) { - print_text(h, "\\(lB"); - h->flags |= HTML_NOSPACE; + pp = h->base_includes; + + bufinit(h); + while (NULL != (p = strchr(pp, '%'))) { + bufncat(h, pp, (size_t)(p - pp)); + switch (*(p + 1)) { + case('I'): + bufcat(h, name); + break; + default: + bufncat(h, p, 2); + break; + } + pp = p + 2; } - return(1); + if (pp) + bufcat(h, pp); } - -/* ARGSUSED */ -static void -mdoc_op_post(MDOC_ARGS) +void +buffmt_man(struct html *h, + const char *name, const char *sec) { + const char *p, *pp; - if (MDOC_BODY != n->type) - return; - h->flags |= HTML_NOSPACE; - print_text(h, "\\(rB"); + pp = h->base_man; + + bufinit(h); + while (NULL != (p = strchr(pp, '%'))) { + bufncat(h, pp, (size_t)(p - pp)); + switch (*(p + 1)) { + case('S'): + bufcat(h, sec ? sec : "1"); + break; + case('N'): + bufcat_fmt(h, name); + break; + default: + bufncat(h, p, 2); + break; + } + pp = p + 2; + } + if (pp) + bufcat(h, pp); } - -static int -mdoc_nm_pre(MDOC_ARGS) +void +bufcat_su(struct html *h, const char *p, const struct roffsu *su) { - struct htmlpair class; + double v; - if ( ! (HTML_NEWLINE & h->flags)) - if (SEC_SYNOPSIS == n->sec) - print_otag(h, TAG_BR, 0, NULL); + v = su->scale; + if (SCALE_MM == su->unit && 0.0 == (v /= 100.0)) + v = 1.0; - class.key = ATTR_CLASS; - class.val = "name"; - - print_otag(h, TAG_SPAN, 1, &class); - if (NULL == n->child) - print_text(h, m->name); - - return(1); + bufcat_fmt(h, "%s: %.2f%s;", p, v, roffscales[su->unit]); } - -/* ARGSUSED */ -static int -mdoc_sh_pre(MDOC_ARGS) +void +bufcat_id(struct html *h, const char *src) { - if (MDOC_BODY == n->type) - print_otag(h, TAG_P, 0, NULL); - if (MDOC_HEAD == n->type) - print_otag(h, TAG_H1, 0, NULL); - return(1); -} + /* Cf. . */ - -/* ARGSUSED */ -static int -mdoc_xr_pre(MDOC_ARGS) -{ - struct htmlpair tag; - - tag.key = ATTR_HREF; - tag.val = "#"; /* TODO */ - - print_otag(h, TAG_A, 1, &tag); - - n = n->child; - print_text(h, n->string); - if (NULL == (n = n->next)) - return(0); - - h->flags |= HTML_NOSPACE; - print_text(h, "("); - h->flags |= HTML_NOSPACE; - print_text(h, n->string); - h->flags |= HTML_NOSPACE; - print_text(h, ")"); - - return(0); -} - - -/* ARGSUSED */ -static int -mdoc_ns_pre(MDOC_ARGS) -{ - - h->flags |= HTML_NOSPACE; - return(1); -} - - -/* ARGSUSED */ -static int -mdoc_ar_pre(MDOC_ARGS) -{ - struct htmlpair tag; - - tag.key = ATTR_CLASS; - tag.val = "arg"; - - print_otag(h, TAG_SPAN, 1, &tag); - return(1); + while ('\0' != *src) + bufcat_fmt(h, "%.2x", *src++); }