version 1.1, 2011/11/09 01:24:23 |
version 1.55, 2014/07/09 09:27:01 |
|
|
/* $Id$ */ |
/* $Id$ */ |
#include <assert.h> |
/* |
|
* Copyright (c) 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv> |
|
* Copyright (c) 2014 Ingo Schwarze <schwarze@usta.de> |
|
* |
|
* 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 <ctype.h> |
#include <fcntl.h> |
#include <fcntl.h> |
#include <regex.h> |
#include <limits.h> |
#include <stdio.h> |
#include <stdio.h> |
#include <stdarg.h> |
|
#include <stdlib.h> |
#include <stdlib.h> |
#include <string.h> |
#include <string.h> |
|
#include <unistd.h> |
|
|
#include "apropos.h" |
#if defined(__sun) |
|
/* for stat() */ |
|
#include <fcntl.h> |
|
#include <sys/types.h> |
|
#include <sys/stat.h> |
|
#endif |
|
|
#include "mandoc.h" |
#include "mandoc.h" |
|
#include "mandoc_aux.h" |
|
#include "main.h" |
|
#include "manpath.h" |
|
#include "mansearch.h" |
|
|
/* |
|
* The page a request is trying to make. |
|
*/ |
|
enum page { |
enum page { |
PAGE_INDEX, |
PAGE_INDEX, |
PAGE_SEARCH, |
PAGE_SEARCH, |
|
PAGE_SHOW, |
PAGE__MAX |
PAGE__MAX |
}; |
}; |
|
|
/* |
/* |
* Key-value pair. |
* A query as passed to the search function. |
* Both key and val are on the heap. |
|
*/ |
*/ |
struct kval { |
struct query { |
char *key; |
const char *manroot; /* manual root directory */ |
char *val; |
const char *arch; /* architecture */ |
|
const char *sec; /* manual section */ |
|
const char *expr; /* unparsed expression string */ |
|
int legacy; /* whether legacy mode */ |
}; |
}; |
|
|
/* |
|
* The media type, determined by suffix, of the requesting or responding |
|
* context. |
|
*/ |
|
enum media { |
|
MEDIA_HTML, |
|
MEDIA__MAX |
|
}; |
|
|
|
/* |
|
* An HTTP request. |
|
*/ |
|
struct req { |
struct req { |
struct kval *fields; /* query fields */ |
struct query q; |
size_t fieldsz; |
char **p; /* array of available manroots */ |
enum media media; |
size_t psz; |
enum page page; |
enum page page; |
}; |
}; |
|
|
static void html_printtext(const char *); |
static void catman(const struct req *, const char *); |
static int kval_decode(char *); |
static int cmp(const void *, const void *); |
static void kval_parse(struct kval **, size_t *, char *); |
static void format(const struct req *, const char *); |
static void kval_free(struct kval *, size_t); |
static void html_print(const char *); |
|
static void html_printquery(const struct req *); |
|
static void html_putchar(char); |
|
static int http_decode(char *); |
|
static void http_parse(struct req *, char *); |
|
static void http_print(const char *); |
|
static void http_putchar(char); |
|
static void http_printquery(const struct req *); |
|
static void pathgen(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 *); |
static void pg_searchres(struct rec *, size_t, void *); |
static void pg_show(const struct req *, char *); |
|
static void resp_bad(void); |
|
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(const struct req *, |
|
struct manpage *, size_t); |
|
static void resp_searchform(const struct req *); |
|
|
|
static const char *progname; /* cgi script name */ |
|
static const char *cache; /* cache directory */ |
|
static const char *css; /* css directory */ |
|
static const char *host; /* hostname */ |
|
|
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 */ |
* Print a character, escaping HTML along the way. |
}; |
* This will pass non-ASCII straight to output: be warned! |
|
*/ |
|
static void |
|
html_putchar(char c) |
|
{ |
|
|
|
switch (c) { |
|
case ('"'): |
|
printf(""e;"); |
|
break; |
|
case ('&'): |
|
printf("&"); |
|
break; |
|
case ('>'): |
|
printf(">"); |
|
break; |
|
case ('<'): |
|
printf("<"); |
|
break; |
|
default: |
|
putchar((unsigned char)c); |
|
break; |
|
} |
|
} |
static void |
static void |
html_printtext(const char *p) |
http_printquery(const struct req *req) |
{ |
{ |
char c; |
|
|
|
while ('\0' != *p) |
if (NULL != req->q.manroot) { |
switch ((c = *p++)) { |
printf("&manpath="); |
case ('"'): |
http_print(req->q.manroot); |
printf(""e;"); |
} |
break; |
if (NULL != req->q.sec) { |
case ('&'): |
printf("&sec="); |
printf("&"); |
http_print(req->q.sec); |
break; |
} |
case ('>'): |
if (NULL != req->q.arch) { |
printf(">"); |
printf("&arch="); |
break; |
http_print(req->q.arch); |
case ('<'): |
} |
printf("<"); |
if (NULL != req->q.expr) { |
break; |
printf("&expr="); |
default: |
http_print(req->q.expr ? req->q.expr : ""); |
putchar((unsigned char)c); |
} |
break; |
|
} |
|
} |
} |
|
|
|
|
static void |
static void |
kval_free(struct kval *p, size_t sz) |
html_printquery(const struct req *req) |
{ |
{ |
int i; |
|
|
|
for (i = 0; i < (int)sz; i++) { |
if (NULL != req->q.manroot) { |
free(p[i].key); |
printf("&manpath="); |
free(p[i].val); |
html_print(req->q.manroot); |
} |
} |
free(p); |
if (NULL != req->q.sec) { |
|
printf("&sec="); |
|
html_print(req->q.sec); |
|
} |
|
if (NULL != req->q.arch) { |
|
printf("&arch="); |
|
html_print(req->q.arch); |
|
} |
|
if (NULL != req->q.expr) { |
|
printf("&expr="); |
|
html_print(req->q.expr ? req->q.expr : ""); |
|
} |
} |
} |
|
|
|
static void |
|
http_print(const char *p) |
|
{ |
|
|
|
if (NULL == p) |
|
return; |
|
while ('\0' != *p) |
|
http_putchar(*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 req *req, char *p) |
{ |
{ |
char *key, *val; |
char *key, *val; |
size_t sz, cur; |
int legacy; |
|
|
cur = 0; |
memset(&req->q, 0, sizeof(struct query)); |
|
req->q.manroot = req->p[0]; |
|
|
while (p && '\0' != *p) { |
legacy = -1; |
while (' ' == *p) |
while ('\0' != *p) { |
p++; |
|
|
|
key = p; |
key = p; |
val = NULL; |
val = NULL; |
|
|
if (NULL != (p = strchr(p, '='))) { |
p += (int)strcspn(p, ";&"); |
|
if ('\0' != *p) |
*p++ = '\0'; |
*p++ = '\0'; |
val = p; |
if (NULL != (val = strchr(key, '='))) |
|
*val++ = '\0'; |
|
|
sz = strcspn(p, ";&"); |
if ('\0' == *key || NULL == val || '\0' == *val) |
/* LINTED */ |
continue; |
p += sz; |
|
|
|
if ('\0' != *p) |
/* Just abort handling. */ |
*p++ = '\0'; |
|
} else { |
|
p = key; |
|
sz = strcspn(p, ";&"); |
|
/* LINTED */ |
|
p += sz; |
|
|
|
if ('\0' != *p) |
if ( ! http_decode(key)) |
p++; |
break; |
continue; |
if (NULL != val && ! http_decode(val)) |
} |
break; |
|
|
if ('\0' == *key || '\0' == *val) |
if (0 == strcmp(key, "expr")) |
continue; |
req->q.expr = val; |
|
else if (0 == strcmp(key, "query")) |
|
req->q.expr = val; |
|
else if (0 == strcmp(key, "sec")) |
|
req->q.sec = val; |
|
else if (0 == strcmp(key, "sektion")) |
|
req->q.sec = val; |
|
else if (0 == strcmp(key, "arch")) |
|
req->q.arch = val; |
|
else if (0 == strcmp(key, "manpath")) |
|
req->q.manroot = val; |
|
else if (0 == strcmp(key, "apropos")) |
|
legacy = 0 == strcmp(val, "0"); |
|
} |
|
|
/* Just abort handling. */ |
/* Test for old man.cgi compatibility mode. */ |
|
|
if ( ! kval_decode(key)) |
req->q.legacy = legacy > 0; |
return; |
|
if ( ! kval_decode(val)) |
|
return; |
|
|
|
if (*kvsz + 1 >= cur) { |
/* |
cur++; |
* Section "0" means no section when in legacy mode. |
*kv = mandoc_realloc |
* For some man.cgi scripts, "default" arch is none. |
(*kv, cur * sizeof(struct kval)); |
*/ |
} |
|
|
|
(*kv)[(int)*kvsz].key = mandoc_strdup(key); |
if (req->q.legacy && NULL != req->q.sec) |
(*kv)[(int)*kvsz].val = mandoc_strdup(val); |
if (0 == strcmp(req->q.sec, "0")) |
(*kvsz)++; |
req->q.sec = NULL; |
|
if (req->q.legacy && NULL != req->q.arch) |
|
if (0 == strcmp(req->q.arch, "default")) |
|
req->q.arch = NULL; |
|
} |
|
|
|
static void |
|
http_putchar(char c) |
|
{ |
|
|
|
if (isalnum((unsigned char)c)) { |
|
putchar((unsigned char)c); |
|
return; |
|
} else if (' ' == c) { |
|
putchar('+'); |
|
return; |
} |
} |
|
printf("%%%.2x", c); |
} |
} |
|
|
/* |
/* |
* 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 197 kval_decode(char *p) |
|
Line 314 kval_decode(char *p) |
|
return(1); |
return(1); |
} |
} |
|
|
|
static void |
|
resp_begin_http(int code, const char *msg) |
|
{ |
|
|
|
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 |
|
resp_begin_html(int code, const char *msg) |
|
{ |
|
|
|
resp_begin_http(code, msg); |
|
|
|
printf("<!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=\"%s/man-cgi.css\"" |
|
" TYPE=\"text/css\" media=\"all\">\n" |
|
"<LINK REL=\"stylesheet\" HREF=\"%s/man.css\"" |
|
" TYPE=\"text/css\" media=\"all\">\n" |
|
"<TITLE>System Manpage Reference</TITLE>\n" |
|
"</HEAD>\n" |
|
"<BODY>\n" |
|
"<!-- Begin page content. //-->\n", css, css); |
|
} |
|
|
|
static void |
|
resp_end_html(void) |
|
{ |
|
|
|
puts("</BODY>\n" |
|
"</HTML>"); |
|
} |
|
|
|
static void |
|
resp_searchform(const struct req *req) |
|
{ |
|
int i; |
|
|
|
puts("<!-- Begin search form. //-->"); |
|
printf("<DIV ID=\"mancgi\">\n" |
|
"<FORM ACTION=\"%s/search\" METHOD=\"get\">\n" |
|
"<FIELDSET>\n" |
|
"<LEGEND>Search Parameters</LEGEND>\n" |
|
"<INPUT TYPE=\"submit\" " |
|
" VALUE=\"Search\"> for manuals satisfying \n" |
|
"<INPUT TYPE=\"text\" NAME=\"expr\" VALUE=\"", |
|
progname); |
|
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 : ""); |
|
printf("\">"); |
|
if (req->psz > 1) { |
|
puts(", <SELECT NAME=\"manpath\">"); |
|
for (i = 0; i < (int)req->psz; i++) { |
|
printf("<OPTION "); |
|
if (NULL == req->q.manroot ? 0 == i : |
|
0 == strcmp(req->q.manroot, req->p[i])) |
|
printf("SELECTED=\"selected\" "); |
|
printf("VALUE=\""); |
|
html_print(req->p[i]); |
|
printf("\">"); |
|
html_print(req->p[i]); |
|
puts("</OPTION>"); |
|
} |
|
puts("</SELECT>"); |
|
} |
|
puts(".\n" |
|
"<INPUT TYPE=\"reset\" VALUE=\"Reset\">\n" |
|
"</FIELDSET>\n" |
|
"</FORM>\n" |
|
"</DIV>"); |
|
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(const struct req *req, struct manpage *r, size_t sz) |
|
{ |
|
size_t i; |
|
|
|
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/%s/%s?", |
|
host, progname, req->q.manroot, r[0].file); |
|
http_printquery(req); |
|
puts("\n" |
|
"Content-Type: text/html; charset=utf-8\n"); |
|
return; |
|
} |
|
|
|
resp_begin_html(200, NULL); |
|
resp_searchform(req); |
|
|
|
puts("<DIV CLASS=\"results\">"); |
|
|
|
if (0 == sz) { |
|
puts("<P>\n" |
|
"No results found.\n" |
|
"</P>\n" |
|
"</DIV>"); |
|
resp_end_html(); |
|
return; |
|
} |
|
|
|
qsort(r, sz, sizeof(struct manpage), cmp); |
|
|
|
puts("<TABLE>"); |
|
|
|
for (i = 0; i < sz; i++) { |
|
printf("<TR>\n" |
|
"<TD CLASS=\"title\">\n" |
|
"<A HREF=\"%s/show/%s/%s?", |
|
progname, req->q.manroot, r[i].file); |
|
html_printquery(req); |
|
printf("\">"); |
|
html_print(r[i].names); |
|
printf("</A>\n" |
|
"</TD>\n" |
|
"<TD CLASS=\"desc\">"); |
|
html_print(r[i].output); |
|
puts("</TD>\n" |
|
"</TR>"); |
|
} |
|
|
|
puts("</TABLE>\n" |
|
"</DIV>"); |
|
resp_end_html(); |
|
} |
|
|
/* ARGSUSED */ |
/* ARGSUSED */ |
static void |
static void |
pg_index(const struct req *req, char *path) |
pg_index(const struct req *req, char *path) |
{ |
{ |
|
|
|
resp_index(req); |
} |
} |
|
|
static void |
static void |
pg_searchres(struct rec *recs, size_t sz, void *arg) |
catman(const struct req *req, const char *file) |
{ |
{ |
|
FILE *f; |
|
size_t len; |
int i; |
int i; |
const char *pg; |
char *p; |
|
int italic, bold; |
|
|
if (NULL == (pg = getenv("SCRIPT_NAME"))) |
if (NULL == (f = fopen(file, "r"))) { |
pg = ""; |
resp_baddb(); |
|
return; |
|
} |
|
|
for (i = 0; i < (int)sz; i++) { |
resp_begin_html(200, NULL); |
printf("<A HREF=\"%s/show/%u.html\">", |
resp_searchform(req); |
pg, recs[i].rec); |
puts("<DIV CLASS=\"catman\">\n" |
html_printtext(recs[i].title); |
"<PRE>"); |
putchar('('); |
|
html_printtext(recs[i].cat); |
while (NULL != (p = fgetln(f, &len))) { |
puts(")</A>"); |
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" |
|
"</DIV>\n" |
|
"</BODY>\n" |
|
"</HTML>"); |
|
|
|
fclose(f); |
} |
} |
|
|
static void |
static void |
|
format(const struct req *req, const char *file) |
|
{ |
|
struct mparse *mp; |
|
int fd; |
|
struct mdoc *mdoc; |
|
struct man *man; |
|
void *vp; |
|
enum mandoclevel rc; |
|
char opts[PATH_MAX + 128]; |
|
|
|
if (-1 == (fd = open(file, O_RDONLY, 0))) { |
|
resp_baddb(); |
|
return; |
|
} |
|
|
|
mp = mparse_alloc(MPARSE_SO, MANDOCLEVEL_FATAL, NULL, NULL); |
|
rc = mparse_readfd(mp, fd, file); |
|
close(fd); |
|
|
|
if (rc >= MANDOCLEVEL_FATAL) { |
|
resp_baddb(); |
|
return; |
|
} |
|
|
|
snprintf(opts, sizeof(opts), |
|
"fragment,man=%s/search?sec=%%S&expr=Nm~^%%N$", |
|
progname); |
|
|
|
mparse_result(mp, &mdoc, &man, NULL); |
|
if (NULL == man && NULL == mdoc) { |
|
resp_baddb(); |
|
mparse_free(mp); |
|
return; |
|
} |
|
|
|
resp_begin_html(200, NULL); |
|
resp_searchform(req); |
|
|
|
vp = html_alloc(opts); |
|
|
|
if (NULL != mdoc) |
|
html_mdoc(vp, mdoc); |
|
else |
|
html_man(vp, man); |
|
|
|
puts("</BODY>\n" |
|
"</HTML>"); |
|
|
|
html_free(vp); |
|
mparse_free(mp); |
|
} |
|
|
|
static void |
|
pg_show(const struct req *req, char *path) |
|
{ |
|
char *sub; |
|
|
|
if (NULL == path || NULL == (sub = strchr(path, '/'))) { |
|
resp_error400(); |
|
return; |
|
} |
|
*sub++ = '\0'; |
|
|
|
/* |
|
* Begin by chdir()ing into the manroot. |
|
* This way we can pick up the database files, which are |
|
* relative to the manpath root. |
|
*/ |
|
|
|
if (-1 == chdir(path)) { |
|
perror(path); |
|
resp_baddb(); |
|
return; |
|
} |
|
|
|
if ('c' == *sub) |
|
catman(req, sub); |
|
else |
|
format(req, sub); |
|
} |
|
|
|
static void |
pg_search(const struct req *req, char *path) |
pg_search(const struct req *req, char *path) |
{ |
{ |
int i; |
struct mansearch search; |
struct opts opt; |
struct manpaths paths; |
|
struct manpage *res; |
|
char **cp; |
|
const char *ep, *start; |
|
size_t ressz; |
|
int i, sz; |
|
|
for (i = 0; i < (int)req->fieldsz; i++) |
/* |
if (0 == strcmp(req->fields[i].key, "key")) |
* Begin by chdir()ing into the root of the manpath. |
break; |
* This way we can pick up the database files, which are |
|
* relative to the manpath root. |
|
*/ |
|
|
if (i == (int)req->fieldsz) |
if (-1 == (chdir(req->q.manroot))) { |
|
perror(req->q.manroot); |
|
resp_search(req, NULL, 0); |
return; |
return; |
|
} |
|
|
memset(&opt, 0, sizeof(struct opts)); |
search.arch = req->q.arch; |
opt.types = TYPE_NAME | TYPE_DESC; |
search.sec = req->q.sec; |
apropos_search(&opt, req->fields[i].val, NULL, pg_searchres); |
search.deftype = TYPE_Nm | TYPE_Nd; |
|
search.flags = 0; |
|
|
|
paths.sz = 1; |
|
paths.paths = mandoc_malloc(sizeof(char *)); |
|
paths.paths[0] = mandoc_strdup("."); |
|
|
|
/* |
|
* Poor man's tokenisation: just break apart by spaces. |
|
* Yes, this is half-ass. But it works for now. |
|
*/ |
|
|
|
ep = req->q.expr; |
|
while (ep && isspace((unsigned char)*ep)) |
|
ep++; |
|
|
|
sz = 0; |
|
cp = NULL; |
|
while (ep && '\0' != *ep) { |
|
cp = mandoc_reallocarray(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++; |
|
} |
|
|
|
if (mansearch(&search, &paths, sz, cp, "Nd", &res, &ressz)) |
|
resp_search(req, res, ressz); |
|
else |
|
resp_baddb(); |
|
|
|
for (i = 0; i < sz; i++) |
|
free(cp[i]); |
|
free(cp); |
|
|
|
for (i = 0; i < (int)ressz; i++) { |
|
free(res[i].file); |
|
free(res[i].names); |
|
free(res[i].output); |
|
} |
|
free(res); |
|
|
|
free(paths.paths[0]); |
|
free(paths.paths); |
} |
} |
|
|
int |
int |
|
|
{ |
{ |
int i; |
int i; |
struct req req; |
struct req req; |
char *p; |
char *p, *path, *subpath; |
char *path, *subpath, *suffix; |
|
|
|
|
/* Scan our run-time environment. */ |
|
|
|
if (NULL == (cache = getenv("CACHE_DIR"))) |
|
cache = "/cache/man.cgi"; |
|
|
|
if (NULL == (progname = getenv("SCRIPT_NAME"))) |
|
progname = ""; |
|
|
|
if (NULL == (css = getenv("CSS_DIR"))) |
|
css = ""; |
|
|
|
if (NULL == (host = getenv("HTTP_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); |
|
} |
|
|
memset(&req, 0, sizeof(struct req)); |
memset(&req, 0, sizeof(struct req)); |
|
pathgen(&req); |
|
|
|
/* 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, 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, '.'))) |
/* Resolve subpath component. */ |
if (NULL != suffix && NULL == strchr(suffix, '/')) |
|
*suffix++ = '\0'; |
|
|
|
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]); |
|
free(req.p); |
return(EXIT_SUCCESS); |
return(EXIT_SUCCESS); |
|
} |
|
|
|
static int |
|
cmp(const void *p1, const void *p2) |
|
{ |
|
|
|
return(strcasecmp(((const struct manpage *)p1)->names, |
|
((const struct manpage *)p2)->names)); |
|
} |
|
|
|
/* |
|
* Scan for indexable paths. |
|
*/ |
|
static void |
|
pathgen(struct req *req) |
|
{ |
|
FILE *fp; |
|
char *dp; |
|
size_t dpsz; |
|
|
|
if (NULL == (fp = fopen("manpath.conf", "r"))) |
|
return; |
|
|
|
while (NULL != (dp = fgetln(fp, &dpsz))) { |
|
if ('\n' == dp[dpsz - 1]) |
|
dpsz--; |
|
req->p = mandoc_realloc(req->p, |
|
(req->psz + 1) * sizeof(char *)); |
|
req->p[req->psz++] = mandoc_strndup(dp, dpsz); |
|
} |
} |
} |