version 1.2, 2011/11/09 22:05:56 |
version 1.25, 2011/12/10 22:20:59 |
|
|
/* $Id$ */ |
/* $Id$ */ |
|
/* |
|
* Copyright (c) 2011 Kristaps Dzonsons <kristaps@bsd.lv> |
|
* |
|
* Permission to use, copy, modify, and distribute this software for any |
|
* purpose with or without fee is hereby granted, provided that the above |
|
* copyright notice and this permission notice appear in all copies. |
|
* |
|
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES |
|
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF |
|
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR |
|
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES |
|
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN |
|
* 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 <sys/param.h> |
|
#include <sys/wait.h> |
|
|
#include <assert.h> |
#include <assert.h> |
|
#include <ctype.h> |
|
#include <errno.h> |
|
#include <dirent.h> |
#include <fcntl.h> |
#include <fcntl.h> |
|
#include <limits.h> |
#include <regex.h> |
#include <regex.h> |
#include <stdio.h> |
#include <stdio.h> |
#include <stdarg.h> |
#include <stdarg.h> |
|
#include <stdint.h> |
#include <stdlib.h> |
#include <stdlib.h> |
#include <string.h> |
#include <string.h> |
|
#include <unistd.h> |
|
|
#include "apropos.h" |
#include "apropos_db.h" |
#include "mandoc.h" |
#include "mandoc.h" |
|
#include "mdoc.h" |
|
#include "man.h" |
|
#include "main.h" |
|
#include "manpath.h" |
|
|
/* |
#ifdef __linux__ |
* The page a request is trying to make. |
# include <db_185.h> |
*/ |
#else |
|
# include <db.h> |
|
#endif |
|
|
enum page { |
enum page { |
PAGE_INDEX, |
PAGE_INDEX, |
PAGE_SEARCH, |
PAGE_SEARCH, |
|
PAGE_SHOW, |
PAGE__MAX |
PAGE__MAX |
}; |
}; |
|
|
/* |
struct paths { |
* Key-value pair. |
char *name; |
* Both key and val are on the heap. |
char *path; |
*/ |
|
struct kval { |
|
char *key; |
|
char *val; |
|
}; |
}; |
|
|
/* |
/* |
* The media type, determined by suffix, of the requesting or responding |
* A query as passed to the search function. |
* context. |
|
*/ |
*/ |
enum media { |
struct query { |
MEDIA_HTML, |
const char *arch; /* architecture */ |
MEDIA__MAX |
const char *sec; /* manual section */ |
|
const char *expr; /* unparsed expression string */ |
|
int whatis; /* whether whatis mode */ |
|
int legacy; /* whether legacy mode */ |
}; |
}; |
|
|
/* |
|
* An HTTP request. |
|
*/ |
|
struct req { |
struct req { |
struct kval *fields; /* query fields */ |
struct query q; |
size_t fieldsz; |
struct paths *p; |
enum media media; |
size_t psz; |
enum page page; |
enum page page; |
}; |
}; |
|
|
#if 0 |
static int atou(const char *, unsigned *); |
static void html_printtext(const char *); |
static void catman(const char *); |
#endif |
static int cmp(const void *, const void *); |
static int kval_decode(char *); |
static void format(const char *); |
static void kval_parse(struct kval **, size_t *, char *); |
static void html_print(const char *); |
static void kval_free(struct kval *, size_t); |
static void html_putchar(char); |
|
static int http_decode(char *); |
|
static void http_parse(struct query *, char *); |
|
static int pathstop(DIR *); |
|
static void pathgen(DIR *, char *, struct req *); |
static void pg_index(const struct req *, char *); |
static void pg_index(const struct req *, char *); |
static void pg_search(const struct req *, char *); |
static void pg_search(const struct req *, char *); |
#if 0 |
static void pg_show(const struct req *, char *); |
static void pg_searchres(struct rec *, size_t, void *); |
static void resp_bad(void); |
#endif |
static void resp_baddb(void); |
|
static void resp_error400(void); |
|
static void resp_error404(const char *); |
|
static void resp_begin_html(int, const char *); |
|
static void resp_begin_http(int, const char *); |
|
static void resp_end_html(void); |
|
static void resp_index(const struct req *); |
|
static void resp_search(struct res *, size_t, void *); |
|
static void resp_searchform(const struct req *); |
|
|
|
static const char *progname; |
|
static const char *cache; |
|
static const char *host; |
|
|
static const char * const pages[PAGE__MAX] = { |
static const char * const pages[PAGE__MAX] = { |
"index", /* PAGE_INDEX */ |
"index", /* PAGE_INDEX */ |
"search", /* PAGE_SEARCH */ |
"search", /* PAGE_SEARCH */ |
|
"show", /* PAGE_SHOW */ |
}; |
}; |
|
|
static const char * const medias[MEDIA__MAX] = { |
/* |
"html", /* MEDIA_HTML */ |
* This is just OpenBSD's strtol(3) suggestion. |
}; |
* I use it instead of strtonum(3) for portability's sake. |
|
*/ |
#if 0 |
static int |
static void |
atou(const char *buf, unsigned *v) |
html_printtext(const char *p) |
|
{ |
{ |
char c; |
char *ep; |
|
long lval; |
|
|
while ('\0' != *p) |
errno = 0; |
switch ((c = *p++)) { |
lval = strtol(buf, &ep, 10); |
case ('"'): |
if (buf[0] == '\0' || *ep != '\0') |
printf(""e;"); |
return(0); |
break; |
if ((errno == ERANGE && (lval == LONG_MAX || |
case ('&'): |
lval == LONG_MIN)) || |
printf("&"); |
(lval > UINT_MAX || lval < 0)) |
break; |
return(0); |
case ('>'): |
|
printf(">"); |
*v = (unsigned int)lval; |
break; |
return(1); |
case ('<'): |
|
printf("<"); |
|
break; |
|
default: |
|
putchar((unsigned char)c); |
|
break; |
|
} |
|
} |
} |
#endif |
|
|
|
|
/* |
|
* Print a character, escaping HTML along the way. |
|
* This will pass non-ASCII straight to output: be warned! |
|
*/ |
static void |
static void |
kval_free(struct kval *p, size_t sz) |
html_putchar(char c) |
{ |
{ |
int i; |
|
|
|
for (i = 0; i < (int)sz; i++) { |
switch (c) { |
free(p[i].key); |
case ('"'): |
free(p[i].val); |
printf(""e;"); |
|
break; |
|
case ('&'): |
|
printf("&"); |
|
break; |
|
case ('>'): |
|
printf(">"); |
|
break; |
|
case ('<'): |
|
printf("<"); |
|
break; |
|
default: |
|
putchar((unsigned char)c); |
|
break; |
} |
} |
free(p); |
|
} |
} |
|
|
/* |
/* |
|
* Call through to html_putchar(). |
|
* Accepts NULL strings. |
|
*/ |
|
static void |
|
html_print(const char *p) |
|
{ |
|
|
|
if (NULL == p) |
|
return; |
|
while ('\0' != *p) |
|
html_putchar(*p++); |
|
} |
|
|
|
/* |
* Parse out key-value pairs from an HTTP request variable. |
* Parse out key-value pairs from an HTTP request variable. |
* This can be either a cookie or a POST/GET string. |
* This can be either a cookie or a POST/GET string, although man.cgi |
|
* uses only GET for simplicity. |
*/ |
*/ |
static void |
static void |
kval_parse(struct kval **kv, size_t *kvsz, char *p) |
http_parse(struct query *q, char *p) |
{ |
{ |
char *key, *val; |
char *key, *val; |
size_t sz, cur; |
size_t sz; |
|
int legacy; |
|
|
cur = 0; |
memset(q, 0, sizeof(struct query)); |
|
|
|
q->whatis = 1; |
|
legacy = -1; |
|
|
while (p && '\0' != *p) { |
while (p && '\0' != *p) { |
while (' ' == *p) |
while (' ' == *p) |
p++; |
p++; |
Line 152 kval_parse(struct kval **kv, size_t *kvsz, char *p) |
|
Line 226 kval_parse(struct kval **kv, size_t *kvsz, char *p) |
|
|
|
/* Just abort handling. */ |
/* Just abort handling. */ |
|
|
if ( ! kval_decode(key)) |
if ( ! http_decode(key)) |
return; |
break; |
if ( ! kval_decode(val)) |
if ( ! http_decode(val)) |
return; |
break; |
|
|
if (*kvsz + 1 >= cur) { |
if (0 == strcmp(key, "expr")) |
cur++; |
q->expr = val; |
*kv = mandoc_realloc |
else if (0 == strcmp(key, "query")) |
(*kv, cur * sizeof(struct kval)); |
q->expr = val; |
} |
else if (0 == strcmp(key, "sec")) |
|
q->sec = val; |
|
else if (0 == strcmp(key, "sektion")) |
|
q->sec = val; |
|
else if (0 == strcmp(key, "arch")) |
|
q->arch = val; |
|
else if (0 == strcmp(key, "apropos")) |
|
legacy = 0 == strcmp(val, "0"); |
|
else if (0 == strcmp(key, "op")) |
|
q->whatis = 0 == strcasecmp(val, "whatis"); |
|
} |
|
|
(*kv)[(int)*kvsz].key = mandoc_strdup(key); |
/* Test for old man.cgi compatibility mode. */ |
(*kv)[(int)*kvsz].val = mandoc_strdup(val); |
|
(*kvsz)++; |
if (legacy == 0) { |
|
q->whatis = 0; |
|
q->legacy = 1; |
|
} else if (legacy > 0) { |
|
q->legacy = 1; |
|
q->whatis = 1; |
} |
} |
|
|
|
/* |
|
* Section "0" means no section when in legacy mode. |
|
* For some man.cgi scripts, "default" arch is none. |
|
*/ |
|
|
|
if (q->legacy && NULL != q->sec) |
|
if (0 == strcmp(q->sec, "0")) |
|
q->sec = NULL; |
|
if (q->legacy && NULL != q->arch) |
|
if (0 == strcmp(q->arch, "default")) |
|
q->arch = NULL; |
} |
} |
|
|
/* |
/* |
* In-place HTTP-decode a string. The standard explanation is that this |
* HTTP-decode a string. The standard explanation is that this turns |
* turns "%4e+foo" into "n foo" in the regular way. This is done |
* "%4e+foo" into "n foo" in the regular way. This is done in-place |
* in-place over the allocated string. |
* over the allocated string. |
*/ |
*/ |
static int |
static int |
kval_decode(char *p) |
http_decode(char *p) |
{ |
{ |
char hex[3]; |
char hex[3]; |
int c; |
int c; |
Line 203 kval_decode(char *p) |
|
Line 304 kval_decode(char *p) |
|
return(1); |
return(1); |
} |
} |
|
|
|
static void |
|
resp_begin_http(int code, const char *msg) |
|
{ |
|
|
/* ARGSUSED */ |
if (200 != code) |
|
printf("Status: %d %s\n", code, msg); |
|
|
|
puts("Content-Type: text/html; charset=utf-8\n" |
|
"Cache-Control: no-cache\n" |
|
"Pragma: no-cache\n" |
|
""); |
|
|
|
fflush(stdout); |
|
} |
|
|
static void |
static void |
pg_index(const struct req *req, char *path) |
resp_begin_html(int code, const char *msg) |
{ |
{ |
|
|
|
resp_begin_http(code, msg); |
|
|
|
puts("<!DOCTYPE HTML PUBLIC " |
|
" \"-//W3C//DTD HTML 4.01//EN\"" |
|
" \"http://www.w3.org/TR/html4/strict.dtd\">\n" |
|
"<HTML>\n" |
|
"<HEAD>\n" |
|
"<META HTTP-EQUIV=\"Content-Type\"" |
|
" CONTENT=\"text/html; charset=utf-8\">\n" |
|
"<LINK REL=\"stylesheet\" HREF=\"/man.cgi.css\"" |
|
" TYPE=\"text/css\" media=\"all\">\n" |
|
"<TITLE>System Manpage Reference</TITLE>\n" |
|
"</HEAD>\n" |
|
"<BODY>\n" |
|
"<!-- Begin page content. //-->"); |
} |
} |
|
|
#if 0 |
|
static void |
static void |
pg_searchres(struct rec *recs, size_t sz, void *arg) |
resp_end_html(void) |
{ |
{ |
int i; |
|
const char *pg; |
|
|
|
if (NULL == (pg = getenv("SCRIPT_NAME"))) |
puts("</BODY>\n" |
pg = ""; |
"</HTML>"); |
|
} |
|
|
|
static void |
|
resp_searchform(const struct req *req) |
|
{ |
|
|
|
puts("<!-- Begin search form. //-->"); |
|
printf("<FORM ACTION=\""); |
|
html_print(progname); |
|
printf("/search.html\" METHOD=\"get\">\n"); |
|
printf("<FIELDSET>\n" |
|
"<LEGEND>Search Parameters</LEGEND>\n" |
|
"<INPUT TYPE=\"submit\" NAME=\"op\"" |
|
" VALUE=\"Whatis\"> or \n" |
|
"<INPUT TYPE=\"submit\" NAME=\"op\"" |
|
" VALUE=\"apropos\"> for manuals satisfying \n" |
|
"<INPUT TYPE=\"text\" NAME=\"expr\" VALUE=\""); |
|
html_print(req->q.expr ? req->q.expr : ""); |
|
printf("\">, section " |
|
"<INPUT TYPE=\"text\"" |
|
" SIZE=\"4\" NAME=\"sec\" VALUE=\""); |
|
html_print(req->q.sec ? req->q.sec : ""); |
|
printf("\">, arch " |
|
"<INPUT TYPE=\"text\"" |
|
" SIZE=\"8\" NAME=\"arch\" VALUE=\""); |
|
html_print(req->q.arch ? req->q.arch : ""); |
|
puts("\">.\n" |
|
"<INPUT TYPE=\"reset\" VALUE=\"Reset\">\n" |
|
"</FIELDSET>\n" |
|
"</FORM>"); |
|
puts("<!-- End search form. //-->"); |
|
} |
|
|
|
static void |
|
resp_index(const struct req *req) |
|
{ |
|
|
|
resp_begin_html(200, NULL); |
|
resp_searchform(req); |
|
resp_end_html(); |
|
} |
|
|
|
static void |
|
resp_error400(void) |
|
{ |
|
|
|
resp_begin_html(400, "Query Malformed"); |
|
printf("<H1>Malformed Query</H1>\n" |
|
"<P>\n" |
|
"The query your entered was malformed.\n" |
|
"Try again from the\n" |
|
"<A HREF=\"%s/index.html\">main page</A>.\n" |
|
"</P>", progname); |
|
resp_end_html(); |
|
} |
|
|
|
static void |
|
resp_error404(const char *page) |
|
{ |
|
|
|
resp_begin_html(404, "Not Found"); |
|
puts("<H1>Page Not Found</H1>\n" |
|
"<P>\n" |
|
"The page you're looking for, "); |
|
printf("<B>"); |
|
html_print(page); |
|
printf("</B>,\n" |
|
"could not be found.\n" |
|
"Try searching from the\n" |
|
"<A HREF=\"%s/index.html\">main page</A>.\n" |
|
"</P>", progname); |
|
resp_end_html(); |
|
} |
|
|
|
static void |
|
resp_bad(void) |
|
{ |
|
resp_begin_html(500, "Internal Server Error"); |
|
puts("<P>Generic badness happened.</P>"); |
|
resp_end_html(); |
|
} |
|
|
|
static void |
|
resp_baddb(void) |
|
{ |
|
|
|
resp_begin_html(500, "Internal Server Error"); |
|
puts("<P>Your database is broken.</P>"); |
|
resp_end_html(); |
|
} |
|
|
|
static void |
|
resp_search(struct res *r, size_t sz, void *arg) |
|
{ |
|
int i; |
|
const struct req *req; |
|
|
|
if (1 == sz) { |
|
/* |
|
* If we have just one result, then jump there now |
|
* without any delay. |
|
*/ |
|
puts("Status: 303 See Other"); |
|
printf("Location: http://%s%s/show/%u/%u.html\n", |
|
host, progname, |
|
r[0].volume, r[0].rec); |
|
puts("Content-Type: text/html; charset=utf-8\n"); |
|
return; |
|
} |
|
|
|
qsort(r, sz, sizeof(struct res), cmp); |
|
|
|
resp_begin_html(200, NULL); |
|
|
|
req = (const struct req *)arg; |
|
resp_searchform(req); |
|
|
|
if (0 == sz) { |
|
printf("<P>\n" |
|
"No %s results found.\n", |
|
req->q.whatis ? "whatis" : "apropos"); |
|
if (req->q.whatis) { |
|
printf("(Try <A HREF=\""); |
|
html_print(progname); |
|
printf("/search.html?op=apropos&expr="); |
|
html_print(req->q.expr ? req->q.expr : ""); |
|
printf("&sec="); |
|
html_print(req->q.sec ? req->q.sec : ""); |
|
printf("&arch="); |
|
html_print(req->q.arch ? req->q.arch : ""); |
|
puts("\">apropos</A>?)"); |
|
} |
|
puts("</P>"); |
|
resp_end_html(); |
|
return; |
|
} |
|
|
|
puts("<P></P>\n" |
|
"<TABLE>"); |
|
|
for (i = 0; i < (int)sz; i++) { |
for (i = 0; i < (int)sz; i++) { |
printf("<A HREF=\"%s/show/%u.html\">", |
printf("<TR>\n" |
pg, recs[i].rec); |
"<TD CLASS=\"title\">\n" |
html_printtext(recs[i].title); |
"<A HREF=\""); |
|
html_print(progname); |
|
printf("/show/0/%u/%u.html\">", r[i].volume, r[i].rec); |
|
html_print(r[i].title); |
putchar('('); |
putchar('('); |
html_printtext(recs[i].cat); |
html_print(r[i].cat); |
puts(")</A>"); |
if (r[i].arch && '\0' != *r[i].arch) { |
|
putchar('/'); |
|
html_print(r[i].arch); |
|
} |
|
printf(")</A>\n" |
|
"</TD>\n" |
|
"<TD CLASS=\"desc\">"); |
|
html_print(r[i].desc); |
|
puts("</TD>\n" |
|
"</TR>"); |
} |
} |
|
|
|
puts("</TABLE>"); |
|
resp_end_html(); |
} |
} |
#endif |
|
|
|
|
/* ARGSUSED */ |
static void |
static void |
pg_search(const struct req *req, char *path) |
pg_index(const struct req *req, char *path) |
{ |
{ |
|
|
|
resp_index(req); |
|
} |
|
|
|
static void |
|
catman(const char *file) |
|
{ |
|
FILE *f; |
|
size_t len; |
int i; |
int i; |
struct opts opt; |
char *p; |
|
int italic, bold; |
|
|
for (i = 0; i < (int)req->fieldsz; i++) |
if (NULL == (f = fopen(file, "r"))) { |
if (0 == strcmp(req->fields[i].key, "key")) |
resp_baddb(); |
break; |
return; |
|
} |
|
|
if (i == (int)req->fieldsz) |
resp_begin_http(200, NULL); |
|
puts("<!DOCTYPE HTML PUBLIC " |
|
" \"-//W3C//DTD HTML 4.01//EN\"" |
|
" \"http://www.w3.org/TR/html4/strict.dtd\">\n" |
|
"<HTML>\n" |
|
"<HEAD>\n" |
|
"<META HTTP-EQUIV=\"Content-Type\"" |
|
" CONTENT=\"text/html; charset=utf-8\">\n" |
|
"<LINK REL=\"stylesheet\" HREF=\"/catman.css\"" |
|
" TYPE=\"text/css\" media=\"all\">\n" |
|
"<TITLE>System Manpage Reference</TITLE>\n" |
|
"</HEAD>\n" |
|
"<BODY>\n" |
|
"<!-- Begin page content. //-->\n" |
|
"<PRE>"); |
|
|
|
while (NULL != (p = fgetln(f, &len))) { |
|
bold = italic = 0; |
|
for (i = 0; i < (int)len - 1; i++) { |
|
/* |
|
* This means that the catpage is out of state. |
|
* Ignore it and keep going (although the |
|
* catpage is bogus). |
|
*/ |
|
|
|
if ('\b' == p[i] || '\n' == p[i]) |
|
continue; |
|
|
|
/* |
|
* Print a regular character. |
|
* Close out any bold/italic scopes. |
|
* If we're in back-space mode, make sure we'll |
|
* have something to enter when we backspace. |
|
*/ |
|
|
|
if ('\b' != p[i + 1]) { |
|
if (italic) |
|
printf("</I>"); |
|
if (bold) |
|
printf("</B>"); |
|
italic = bold = 0; |
|
html_putchar(p[i]); |
|
continue; |
|
} else if (i + 2 >= (int)len) |
|
continue; |
|
|
|
/* Italic mode. */ |
|
|
|
if ('_' == p[i]) { |
|
if (bold) |
|
printf("</B>"); |
|
if ( ! italic) |
|
printf("<I>"); |
|
bold = 0; |
|
italic = 1; |
|
i += 2; |
|
html_putchar(p[i]); |
|
continue; |
|
} |
|
|
|
/* |
|
* Handle funny behaviour troff-isms. |
|
* These grok'd from the original man2html.c. |
|
*/ |
|
|
|
if (('+' == p[i] && 'o' == p[i + 2]) || |
|
('o' == p[i] && '+' == p[i + 2]) || |
|
('|' == p[i] && '=' == p[i + 2]) || |
|
('=' == p[i] && '|' == p[i + 2]) || |
|
('*' == p[i] && '=' == p[i + 2]) || |
|
('=' == p[i] && '*' == p[i + 2]) || |
|
('*' == p[i] && '|' == p[i + 2]) || |
|
('|' == p[i] && '*' == p[i + 2])) { |
|
if (italic) |
|
printf("</I>"); |
|
if (bold) |
|
printf("</B>"); |
|
italic = bold = 0; |
|
putchar('*'); |
|
i += 2; |
|
continue; |
|
} else if (('|' == p[i] && '-' == p[i + 2]) || |
|
('-' == p[i] && '|' == p[i + 1]) || |
|
('+' == p[i] && '-' == p[i + 1]) || |
|
('-' == p[i] && '+' == p[i + 1]) || |
|
('+' == p[i] && '|' == p[i + 1]) || |
|
('|' == p[i] && '+' == p[i + 1])) { |
|
if (italic) |
|
printf("</I>"); |
|
if (bold) |
|
printf("</B>"); |
|
italic = bold = 0; |
|
putchar('+'); |
|
i += 2; |
|
continue; |
|
} |
|
|
|
/* Bold mode. */ |
|
|
|
if (italic) |
|
printf("</I>"); |
|
if ( ! bold) |
|
printf("<B>"); |
|
bold = 1; |
|
italic = 0; |
|
i += 2; |
|
html_putchar(p[i]); |
|
} |
|
|
|
/* |
|
* Clean up the last character. |
|
* We can get to a newline; don't print that. |
|
*/ |
|
|
|
if (italic) |
|
printf("</I>"); |
|
if (bold) |
|
printf("</B>"); |
|
|
|
if (i == (int)len - 1 && '\n' != p[i]) |
|
html_putchar(p[i]); |
|
|
|
putchar('\n'); |
|
} |
|
|
|
puts("</PRE>\n" |
|
"</BODY>\n" |
|
"</HTML>"); |
|
|
|
fclose(f); |
|
} |
|
|
|
static void |
|
format(const char *file) |
|
{ |
|
struct mparse *mp; |
|
int fd; |
|
struct mdoc *mdoc; |
|
struct man *man; |
|
void *vp; |
|
enum mandoclevel rc; |
|
char opts[MAXPATHLEN + 128]; |
|
|
|
if (-1 == (fd = open(file, O_RDONLY, 0))) { |
|
resp_baddb(); |
return; |
return; |
|
} |
|
|
|
mp = mparse_alloc(MPARSE_AUTO, MANDOCLEVEL_FATAL, NULL, NULL); |
|
rc = mparse_readfd(mp, fd, file); |
|
close(fd); |
|
|
|
if (rc >= MANDOCLEVEL_FATAL) { |
|
resp_baddb(); |
|
return; |
|
} |
|
|
|
snprintf(opts, sizeof(opts), "style=/man.css," |
|
"man=%s/search.html?sec=%%S&expr=%%N," |
|
/*"includes=/cgi-bin/man.cgi/usr/include/%%I"*/, |
|
progname); |
|
|
|
mparse_result(mp, &mdoc, &man); |
|
vp = html_alloc(opts); |
|
|
|
if (NULL != mdoc) { |
|
resp_begin_http(200, NULL); |
|
html_mdoc(vp, mdoc); |
|
} else if (NULL != man) { |
|
resp_begin_http(200, NULL); |
|
html_man(vp, man); |
|
} else |
|
resp_baddb(); |
|
|
|
html_free(vp); |
|
mparse_free(mp); |
|
} |
|
|
|
static void |
|
pg_show(const struct req *req, char *path) |
|
{ |
|
struct manpaths ps; |
|
char *sub; |
|
char file[MAXPATHLEN]; |
|
const char *fn, *cp; |
|
int rc; |
|
unsigned int vol, rec, mr; |
|
DB *idx; |
|
DBT key, val; |
|
|
|
idx = NULL; |
|
|
|
/* Parse out mroot, volume, and record from the path. */ |
|
|
|
if (NULL == path || NULL == (sub = strchr(path, '/'))) { |
|
resp_error400(); |
|
return; |
|
} |
|
*sub++ = '\0'; |
|
if ( ! atou(path, &mr)) { |
|
resp_error400(); |
|
return; |
|
} |
|
path = sub; |
|
if (NULL == (sub = strchr(path, '/'))) { |
|
resp_error400(); |
|
return; |
|
} |
|
*sub++ = '\0'; |
|
if ( ! atou(path, &vol) || ! atou(sub, &rec)) { |
|
resp_error400(); |
|
return; |
|
} else if (mr >= (unsigned int)req->psz) { |
|
resp_error400(); |
|
return; |
|
} |
|
|
|
/* |
|
* Begin by chdir()ing into the root of the manpath. |
|
* This way we can pick up the database files, which are |
|
* relative to the manpath root. |
|
*/ |
|
|
|
if (-1 == chdir(req->p[(int)mr].path)) { |
|
perror(req->p[(int)mr].path); |
|
resp_baddb(); |
|
return; |
|
} |
|
|
|
memset(&ps, 0, sizeof(struct manpaths)); |
|
manpath_manconf("etc/catman.conf", &ps); |
|
|
|
if (vol >= (unsigned int)ps.sz) { |
|
resp_error400(); |
|
goto out; |
|
} |
|
|
|
strlcpy(file, ps.paths[vol], MAXPATHLEN); |
|
strlcat(file, "/mandoc.index", MAXPATHLEN); |
|
|
|
/* Open the index recno(3) database. */ |
|
|
|
idx = dbopen(file, O_RDONLY, 0, DB_RECNO, NULL); |
|
if (NULL == idx) { |
|
perror(file); |
|
resp_baddb(); |
|
goto out; |
|
} |
|
|
|
key.data = &rec; |
|
key.size = 4; |
|
|
|
if (0 != (rc = (*idx->get)(idx, &key, &val, 0))) { |
|
rc < 0 ? resp_baddb() : resp_error400(); |
|
goto out; |
|
} |
|
|
|
cp = (char *)val.data; |
|
|
|
if (NULL == (fn = memchr(cp, '\0', val.size))) |
|
resp_baddb(); |
|
else if (++fn - cp >= (int)val.size) |
|
resp_baddb(); |
|
else if (NULL == memchr(fn, '\0', val.size - (fn - cp))) |
|
resp_baddb(); |
|
else { |
|
if (0 == strcmp(cp, "cat")) |
|
catman(fn + 1); |
|
else |
|
format(fn + 1); |
|
} |
|
out: |
|
if (idx) |
|
(*idx->close)(idx); |
|
manpath_free(&ps); |
|
} |
|
|
|
static void |
|
pg_search(const struct req *req, char *path) |
|
{ |
|
size_t tt; |
|
struct manpaths ps; |
|
int i, sz, rc; |
|
const char *ep, *start; |
|
char **cp; |
|
struct opts opt; |
|
struct expr *expr; |
|
|
|
if (0 == req->psz) { |
|
resp_search(NULL, 0, (void *)req); |
|
return; |
|
} |
|
|
memset(&opt, 0, sizeof(struct opts)); |
memset(&opt, 0, sizeof(struct opts)); |
/*opt.types = TYPE_NAME | TYPE_DESC; |
|
apropos_search(&opt, req->fields[i].val, NULL, pg_searchres);*/ |
ep = req->q.expr; |
|
opt.arch = req->q.arch; |
|
opt.cat = req->q.sec; |
|
rc = -1; |
|
sz = 0; |
|
cp = NULL; |
|
|
|
/* |
|
* Begin by chdir()ing into the root of the manpath. |
|
* This way we can pick up the database files, which are |
|
* relative to the manpath root. |
|
*/ |
|
|
|
if (-1 == (chdir(req->p[0].path))) { |
|
perror(req->p[0].path); |
|
resp_search(NULL, 0, (void *)req); |
|
return; |
|
} |
|
|
|
memset(&ps, 0, sizeof(struct manpaths)); |
|
manpath_manconf("etc/catman.conf", &ps); |
|
|
|
/* |
|
* Poor man's tokenisation: just break apart by spaces. |
|
* Yes, this is half-ass. But it works for now. |
|
*/ |
|
|
|
while (ep && isspace((unsigned char)*ep)) |
|
ep++; |
|
|
|
while (ep && '\0' != *ep) { |
|
cp = mandoc_realloc(cp, (sz + 1) * sizeof(char *)); |
|
start = ep; |
|
while ('\0' != *ep && ! isspace((unsigned char)*ep)) |
|
ep++; |
|
cp[sz] = mandoc_malloc((ep - start) + 1); |
|
memcpy(cp[sz], start, ep - start); |
|
cp[sz++][ep - start] = '\0'; |
|
while (isspace((unsigned char)*ep)) |
|
ep++; |
|
} |
|
|
|
/* |
|
* Pump down into apropos backend. |
|
* The resp_search() function is called with the results. |
|
*/ |
|
|
|
expr = req->q.whatis ? |
|
termcomp(sz, cp, &tt) : exprcomp(sz, cp, &tt); |
|
|
|
if (NULL != expr) |
|
rc = apropos_search |
|
(ps.sz, ps.paths, &opt, |
|
expr, tt, (void *)req, resp_search); |
|
|
|
/* ...unless errors occured. */ |
|
|
|
if (0 == rc) |
|
resp_baddb(); |
|
else if (-1 == rc) |
|
resp_search(NULL, 0, (void *)req); |
|
|
|
for (i = 0; i < sz; i++) |
|
free(cp[i]); |
|
|
|
free(cp); |
|
exprfree(expr); |
|
manpath_free(&ps); |
} |
} |
|
|
int |
int |
main(void) |
main(void) |
{ |
{ |
int i; |
int i; |
|
char buf[MAXPATHLEN]; |
|
DIR *cwd; |
struct req req; |
struct req req; |
char *p; |
char *p, *path, *subpath; |
char *path, *subpath, *suffix; |
|
|
|
|
/* Scan our run-time environment. */ |
|
|
|
progname = getenv("SCRIPT_NAME"); |
|
if (NULL == progname) |
|
progname = ""; |
|
|
|
cache = getenv("CACHE_DIR"); |
|
if (NULL == cache) |
|
cache = "/cache/man.cgi"; |
|
|
|
host = getenv("HTTP_HOST"); |
|
if (NULL == host) |
|
host = "localhost"; |
|
|
|
/* |
|
* First we change directory into the cache directory so that |
|
* subsequent scanning for manpath directories is rooted |
|
* relative to the same position. |
|
*/ |
|
|
|
if (-1 == chdir(cache)) { |
|
perror(cache); |
|
resp_bad(); |
|
return(EXIT_FAILURE); |
|
} else if (NULL == (cwd = opendir(cache))) { |
|
perror(cache); |
|
resp_bad(); |
|
return(EXIT_FAILURE); |
|
} |
|
|
memset(&req, 0, sizeof(struct req)); |
memset(&req, 0, sizeof(struct req)); |
|
|
|
strlcpy(buf, ".", MAXPATHLEN); |
|
pathgen(cwd, buf, &req); |
|
closedir(cwd); |
|
|
|
/* Next parse out the query string. */ |
|
|
if (NULL != (p = getenv("QUERY_STRING"))) |
if (NULL != (p = getenv("QUERY_STRING"))) |
kval_parse(&req.fields, &req.fieldsz, p); |
http_parse(&req.q, p); |
|
|
suffix = subpath = path = NULL; |
/* |
|
* Now juggle paths to extract information. |
|
* We want to extract our filetype (the file suffix), the |
|
* initial path component, then the trailing component(s). |
|
* Start with leading subpath component. |
|
*/ |
|
|
req.media = MEDIA_HTML; |
subpath = path = NULL; |
req.page = PAGE__MAX; |
req.page = PAGE__MAX; |
|
|
if (NULL == (path = getenv("PATH_INFO")) || '\0' == *path) |
if (NULL == (path = getenv("PATH_INFO")) || '\0' == *path) |
req.page = PAGE_INDEX; |
req.page = PAGE_INDEX; |
|
|
if (NULL != path && '/' == *path && '\0' == *++path) |
if (NULL != path && '/' == *path && '\0' == *++path) |
req.page = PAGE_INDEX; |
req.page = PAGE_INDEX; |
|
|
if (NULL != path && NULL != (suffix = strrchr(path, '.'))) |
/* Strip file suffix. */ |
if (NULL != suffix && NULL == strchr(suffix, '/')) |
|
*suffix++ = '\0'; |
|
|
|
|
if (NULL != path && NULL != (p = strrchr(path, '.'))) |
|
if (NULL != p && NULL == strchr(p, '/')) |
|
*p++ = '\0'; |
|
|
|
/* Resolve subpath component. */ |
|
|
if (NULL != path && NULL != (subpath = strchr(path, '/'))) |
if (NULL != path && NULL != (subpath = strchr(path, '/'))) |
*subpath++ = '\0'; |
*subpath++ = '\0'; |
|
|
if (NULL != suffix && '\0' != *suffix) |
/* Map path into one we recognise. */ |
for (i = 0; i < (int)MEDIA__MAX; i++) |
|
if (0 == strcmp(medias[i], suffix)) { |
|
req.media = (enum media)i; |
|
break; |
|
} |
|
|
|
if (NULL != path && '\0' != *path) |
if (NULL != path && '\0' != *path) |
for (i = 0; i < (int)PAGE__MAX; i++) |
for (i = 0; i < (int)PAGE__MAX; i++) |
|
|
break; |
break; |
} |
} |
|
|
|
/* Route pages. */ |
|
|
switch (req.page) { |
switch (req.page) { |
case (PAGE_INDEX): |
case (PAGE_INDEX): |
pg_index(&req, subpath); |
pg_index(&req, subpath); |
|
|
case (PAGE_SEARCH): |
case (PAGE_SEARCH): |
pg_search(&req, subpath); |
pg_search(&req, subpath); |
break; |
break; |
|
case (PAGE_SHOW): |
|
pg_show(&req, subpath); |
|
break; |
default: |
default: |
/* Blah */ |
resp_error404(path); |
break; |
break; |
} |
} |
|
|
kval_free(req.fields, req.fieldsz); |
for (i = 0; i < (int)req.psz; i++) { |
|
free(req.p[i].path); |
|
free(req.p[i].name); |
|
} |
|
|
|
free(req.p); |
return(EXIT_SUCCESS); |
return(EXIT_SUCCESS); |
|
} |
|
|
|
static int |
|
cmp(const void *p1, const void *p2) |
|
{ |
|
|
|
return(strcasecmp(((const struct res *)p1)->title, |
|
((const struct res *)p2)->title)); |
|
} |
|
|
|
/* |
|
* Check to see if an "etc" path consists of a catman.conf file. If it |
|
* does, that means that the path contains a tree created by catman(8) |
|
* and should be used for indexing. |
|
*/ |
|
static int |
|
pathstop(DIR *dir) |
|
{ |
|
struct dirent *d; |
|
|
|
while (NULL != (d = readdir(dir))) |
|
if (DT_REG == d->d_type) |
|
if (0 == strcmp(d->d_name, "catman.conf")) |
|
return(1); |
|
|
|
return(0); |
|
} |
|
|
|
/* |
|
* Scan for indexable paths. |
|
* This adds all paths with "etc/catman.conf" to the buffer. |
|
*/ |
|
static void |
|
pathgen(DIR *dir, char *path, struct req *req) |
|
{ |
|
struct dirent *d; |
|
char *cp; |
|
DIR *cd; |
|
int rc; |
|
size_t sz, ssz; |
|
|
|
sz = strlcat(path, "/", MAXPATHLEN); |
|
if (sz >= MAXPATHLEN) { |
|
fprintf(stderr, "%s: Path too long", path); |
|
return; |
|
} |
|
|
|
/* |
|
* First, scan for the "etc" directory. |
|
* If it's found, then see if it should cause us to stop. This |
|
* happens when a catman.conf is found in the directory. |
|
*/ |
|
|
|
rc = 0; |
|
while (0 == rc && NULL != (d = readdir(dir))) { |
|
if (DT_DIR != d->d_type || strcmp(d->d_name, "etc")) |
|
continue; |
|
|
|
path[(int)sz] = '\0'; |
|
ssz = strlcat(path, d->d_name, MAXPATHLEN); |
|
|
|
if (ssz >= MAXPATHLEN) { |
|
fprintf(stderr, "%s: Path too long", path); |
|
return; |
|
} else if (NULL == (cd = opendir(path))) { |
|
perror(path); |
|
return; |
|
} |
|
|
|
rc = pathstop(cd); |
|
closedir(cd); |
|
} |
|
|
|
if (rc > 0) { |
|
/* This also strips the trailing slash. */ |
|
path[(int)sz - 1] = '\0'; |
|
req->p = mandoc_realloc |
|
(req->p, |
|
(req->psz + 1) * sizeof(struct paths)); |
|
req->p[(int)req->psz].path = mandoc_strdup(path); |
|
/* And this strips out the leading "./". */ |
|
req->p[(int)req->psz].name = |
|
cp = mandoc_strdup(path + 2); |
|
req->psz++; |
|
/* |
|
* The name is just the path with all the slashes taken |
|
* out of it. Simple but effective. |
|
*/ |
|
for ( ; '\0' != *cp; cp++) |
|
if ('/' == *cp) |
|
*cp = ' '; |
|
return; |
|
} |
|
|
|
/* |
|
* If no etc/catman.conf was found, recursively enter child |
|
* directory and continue scanning. |
|
*/ |
|
|
|
rewinddir(dir); |
|
while (NULL != (d = readdir(dir))) { |
|
if (DT_DIR != d->d_type || '.' == d->d_name[0]) |
|
continue; |
|
|
|
path[(int)sz] = '\0'; |
|
ssz = strlcat(path, d->d_name, MAXPATHLEN); |
|
|
|
if (ssz >= MAXPATHLEN) { |
|
fprintf(stderr, "%s: Path too long", path); |
|
return; |
|
} else if (NULL == (cd = opendir(path))) { |
|
perror(path); |
|
return; |
|
} |
|
|
|
pathgen(cd, path, req); |
|
closedir(cd); |
|
} |
} |
} |