Subversion Repositories wimsdev

Rev

Rev 8185 | Rev 11128 | Go to most recent revision | Details | Compare with Previous | Last modification | View Log | RSS feed

Rev Author Line No. Line
10 reyssat 1
/*    Copyright (C) 1998-2003 XIAO, Gang of Universite de Nice - Sophia Antipolis
2
 *
3
 *  This program is free software; you can redistribute it and/or modify
4
 *  it under the terms of the GNU General Public License as published by
5
 *  the Free Software Foundation; either version 2 of the License, or
6
 *  (at your option) any later version.
7
 *
8
 *  This program is distributed in the hope that it will be useful,
9
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
10
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
11
 *  GNU General Public License for more details.
12
 *
13
 *  You should have received a copy of the GNU General Public License
14
 *  along with this program; if not, write to the Free Software
15
 *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
16
 */
17
 
18
 /* This is a small program used simply to fetch web pages.
19
  * No fancy functionalities such as link redirection or site sucking is
20
  * present.
21
  * Page fetched can only be sent to stdout. */
22
 
23
#include <netdb.h>
24
#include <sys/socket.h>
25
#include <netinet/in.h>
26
 
8185 bpr 27
#include "../includes.h"
28
#include "../wimsdef.h"
29
 
10 reyssat 30
char *cheater1="User-Agent: WIMS-webget";
31
char *cheater2="Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, image/png, */*\r\n\
32
Accept-Encoding: gzip\r\n\
33
Accept-Language: en, fr, it, de, es\r\n\
34
Accept-Charset: iso-8859-1,*,utf-8";
35
char pbuf[4096], tbuf[4096];
36
char tfname[1024];
37
char *tmpdir="/tmp";
38
int soc, port, https;
39
int charcnt;
40
FILE *outf;
41
 
42
void errorquit(char *msg)
43
{
44
    fprintf(stderr,"%s: %s\n",msg,strerror(errno)); exit(1);
45
}
46
 
7676 bpr 47
/* Points to the end of the word */
10 reyssat 48
char *find_word_end(char *p)
49
{
50
    int i;
51
    for(i=0;!isspace(*p) && *p!=0 && i<MAX_LINELEN; p++,i++);
52
    return p;
53
}
54
 
7676 bpr 55
/* Strips leading spaces */
10 reyssat 56
char *find_word_start(char *p)
57
{
58
    int i;
59
    for(i=0; isspace(*p) && i<MAX_LINELEN; p++,i++);
60
    return p;
61
}
62
 
7676 bpr 63
/* Secured execution */
10 reyssat 64
void secure(char *host)
65
{
66
    char *p1, *p2, *p3, buf[MAX_LINELEN+1];
67
    long int l;
68
    FILE *f;
69
 
70
    p1=getenv("w_module"); if(p1==NULL || *p1==0) return;
71
    p1=getenv("untrust"); if(p1==NULL || *p1==0) return;
72
    f=fopen("webget.sites","r"); if(f==NULL) return;
73
    l=fread(buf,1,MAX_LINELEN,f); fclose(f);
74
    if(l<=0 || l>MAX_LINELEN) return;
75
    buf[l]=0;
76
    for(p1=find_word_start(buf);*p1;p1=find_word_start(p2)) {
7676 bpr 77
      p2=find_word_end(p1); if(*p2) *p2++=0;
78
      p3=strstr(host,p1); if(p3==NULL) continue;
79
      if((p3==host || *(p3-1)=='.') && *(p3+strlen(p1))==0) return;
10 reyssat 80
    }
81
    exit(1);  /* unauthorized sites refused. */
82
}
83
 
7676 bpr 84
/* open a TCP/IP socket with host/port
85
 * returns the file descriptor for the socket */
10 reyssat 86
int net_connect(char *host)
87
{
88
    struct hostent *hp;
89
    struct sockaddr_in sin;
90
    int soc;
91
 
92
    secure(host);
93
    if(!(hp = gethostbyname(host))) errorquit("unknown host.");
94
    if((soc = socket(hp->h_addrtype,SOCK_STREAM,0))<0)
95
      errorquit("socket() error");
96
    memmove(&sin.sin_addr,hp->h_addr,hp->h_length);
97
    sin.sin_port=htons(port);
98
    sin.sin_family = hp->h_addrtype;
99
    if(connect(soc,(struct sockaddr *)&sin,sizeof(sin))<0) {
7676 bpr 100
      close(soc); errorquit("connect() error");
10 reyssat 101
    }
102
    return soc;
103
}
104
 
105
int gethttps(char *host)
106
{
107
    char buf[65536];
108
    char *tp;
7676 bpr 109
 
110
    tp=getenv("tmp_dir"); if(tp!=NULL && *tp!=0) tmpdir=tp;
10 reyssat 111
    snprintf(tfname,sizeof(tfname),"%s/https.tmp",tmpdir);
112
    snprintf(buf,sizeof(buf),"\
113
mkdir -p %s\n\
114
openssl s_client -connect %s:%d -quiet 2>/dev/null >%s <<@\n\
115
%s\n\
116
@\n", tmpdir,host,port,tfname,tbuf);
3840 kbelabas 117
    if (system(buf))
118
      errorquit("system() error");
10 reyssat 119
    return open(tfname,O_RDONLY);
120
}
121
 
122
int main(int argc, char *argv[])
123
{
124
    char *parm, *pt, *p1, *p2, *p3, *p4, *dp, *pre;
125
    char nbuf[4096], *pp1, *pp2;
126
    char c;
7676 bpr 127
 
10 reyssat 128
    parm=getenv("wims_exec_parm");
129
    if(parm==NULL || *parm==0) errorquit("no_parameter");
130
    snprintf(pbuf,sizeof(pbuf),"%s",parm);
131
    p1=find_word_start(pbuf); p2=find_word_end(p1);
11124 georgesk 132
    if(*p2!=0) *p2++=0;
133
    https=0;
10 reyssat 134
    outf=stdout; pp1=getenv("w_webget_output");
135
    pp2=getenv("session_dir");
136
    if(pp1!=NULL && strstr(pp1,"..")==NULL && isalnum(*pp1) && pp2!=NULL) {
7676 bpr 137
      snprintf(nbuf,sizeof(nbuf),"%s/%s",pp2,pp1);
138
      outf=fopen(nbuf,"w"); if(outf==NULL) outf=stdout;
10 reyssat 139
    }
140
    dp=getenv("w_webget_option");
7676 bpr 141
    if(dp!=NULL && strstr(dp,"direct")!=NULL) { /* direct get */
142
      p1=getenv("w_webget_host");
143
      p2=getenv("w_webget_port");
144
      if(p1==NULL || p2==NULL) errorquit("incomplete_request");
145
      port=atoi(p2);
146
      soc=net_connect(p1); if(soc==-1) return 1;
147
      c=' '; for(p3=parm; *p3; p3++) {
148
          if(*p3=='\n' && c!='\r') (void)write(soc,"\r",1);
149
          (void)write(soc,p3,1); c=*p3;
150
      }
151
      (void)write(soc,"\r\n\r\n",4);
152
      pt=getenv("w_module");
153
      if(pt==NULL || *pt==0 || strncmp(pt,"adm/",4)==0 ) {  /* File to post? */
154
          pt=getenv("w_webget_post"); if(pt!=NULL && *pt!=0) {
155
            FILE *f;
156
            char buf[4096];
157
            size_t l;
158
            f=fopen(pt,"r"); if(f!=NULL) {
159
                do {
160
                  l=fread(buf,1,sizeof(buf),f);
161
                  if(l>0 && l<=sizeof(buf)) (void)write(soc,buf,l);
162
                } while(l==sizeof(buf));
163
                fclose(f);
164
            }
165
          }
166
      }
167
      if(strstr(dp,"normalread")!=NULL) goto read;
168
      charcnt=0;
169
      while(read(soc,pbuf,1)>0 && charcnt<10240) {
170
          fputc(pbuf[0],outf); charcnt++;
171
      }
172
      close(soc);
173
      return 0;
10 reyssat 174
    }
175
    if(strncasecmp(p1,"http://",strlen("http://"))==0) p1+=strlen("http://");
176
    else if(strncasecmp(p1,"https://",strlen("https://"))==0) {
7676 bpr 177
      https=1; p1+=strlen("https://");
10 reyssat 178
    }
179
    p3=strchr(p1,'/'); if(p3==NULL) p3="";
180
    else {*p3++=0; while(*p3=='/') p3++;}
181
    if(strncasecmp(p3,"http://",strlen("http://"))==0 ||
182
       strncasecmp(p3,"https://",strlen("https://"))==0) pre="";
183
    else pre="/";
184
    snprintf(tbuf,sizeof(tbuf),"GET %s%s HTTP/1.0\r\n%s\r\n\
185
Host: %s\r\n\
186
%s\r\n\r\n",
7676 bpr 187
           pre,p3,cheater1,p1,cheater2);
10 reyssat 188
    p4=strchr(p1,':'); if(p4==NULL) {
7676 bpr 189
      if(https) port=443; else port=80;
10 reyssat 190
    }
191
    else {*p4++=0; port=atoi(p4);}
192
    if(https) {
7676 bpr 193
      soc=gethttps(p1); goto read;
10 reyssat 194
    }
195
    soc=net_connect(p1);
3840 kbelabas 196
    (void)write(soc,tbuf,strlen(tbuf));
7676 bpr 197
/* header */
10 reyssat 198
    read: if(soc==-1) return 1;
199
    c=-1;
200
    while(read(soc,pbuf,1)>0) {
7676 bpr 201
      if(pbuf[0]=='\r') continue;
202
      fputc(pbuf[0],stderr);
203
      if((c=='\n') && (pbuf[0]=='\n')) break; else c=pbuf[0];
10 reyssat 204
    }
7676 bpr 205
/* body */
10 reyssat 206
    charcnt=0;
207
    while(read(soc,pbuf,1)>0 && charcnt<MAX_WEBGETFLEN) {
7676 bpr 208
      fputc(pbuf[0],outf); charcnt++;
10 reyssat 209
    }
210
    close(soc);
211
    if(outf!=stdout) fclose(outf);
212
    if(https) unlink(tfname);
213
    return 0;
214
}
215