src/html.c

   1 /*
   2  * gaim
   3  *
   4  * Copyright (C) 1998-1999, Mark Spencer <markster@marko.net>
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  11  * This program is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License
  17  * along with this program; if not, write to the Free Software
  18  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  19  *
  20  */
  21
  22 #ifdef HAVE_CONFIG_H
  23 #include <config.h>
  24 #endif
  25 #include <string.h>
  26 #include <stdio.h>
  27 #include <stdlib.h>
  28 #include <sys/time.h>
  29 #include <unistd.h>
  30 #include "gaim.h"
  31 #include <sys/types.h>
  32 #include <sys/socket.h>
  33 #include <netdb.h>
  34 #include <netinet/in.h>
  35 #include <fcntl.h>
  36 #include <errno.h>
  37 #include "proxy.h"
  38
  39 struct g_url {
  40         char address[255];
  41         int port;
  42         char page[255];
  43 };
  44
  45 gchar *strip_html(gchar *text)
  46 {
  47         int i, j, k;
  48         int visible = 1;
  49         gchar *text2 = g_strdup(text);
  50
  51         for (i = 0, j = 0; text2[i]; i++) {
  52                 if (text2[i] == '<') {
  53                         k = i + 1;
  54                         while (text2[k]) {
  55                                 if (text2[k] == '<') {
  56                                         visible = 1;
  57                                         break;
  58                                 }
  59                                 if (text2[k] == '>') {
  60                                         visible = 0;
  61                                         break;
  62                                 }
  63                                 k++;
  64                         }
  65                 } else if (text2[i] == '>' && !visible) {
  66                         visible = 1;
  67                         continue;
  68                 }
  69                 if (visible) {
  70                         text2[j++] = text2[i];
  71                 }
  72         }
  73         text2[j] = '\0';
  74         return text2;
  75 }
  76
  77 static struct g_url *parse_url(char *url)
  78 {
  79         struct g_url *test = g_new0(struct g_url, 1);
  80         char scan_info[255];
  81         char port[5];
  82         int f;
  83
  84         if (strstr(url, "http://"))
  85                 g_snprintf(scan_info, sizeof(scan_info),
  86                            "http://%%[A-Za-z0-9.]:%%[0-9]/%%[A-Za-z0-9.~_-/&%%?=+]");
  87         else
  88                 g_snprintf(scan_info, sizeof(scan_info),
  89                            "%%[A-Za-z0-9.]:%%[0-9]/%%[A-Za-z0-9.~_-/&%%?=+^]");
  90         f = sscanf(url, scan_info, test->address, port, test->page);
  91         if (f == 1) {
  92                 if (strstr(url, "http://"))
  93                         g_snprintf(scan_info, sizeof(scan_info),
  94                                    "http://%%[A-Za-z0-9.]/%%[A-Za-z0-9.~_-/&%%?=+^]");
  95                 else
  96                         g_snprintf(scan_info, sizeof(scan_info),
  97                                    "%%[A-Za-z0-9.]/%%[A-Za-z0-9.~_-/&%%?=+^]");
  98                 f = sscanf(url, scan_info, test->address, test->page);
  99                 g_snprintf(port, sizeof(test->port), "80");
 100                 port[2] = 0;
 101         }
 102         if (f == 1) {
 103                 if (strstr(url, "http://"))
 104                         g_snprintf(scan_info, sizeof(scan_info), "http://%%[A-Za-z0-9.]");
 105                 else
 106                         g_snprintf(scan_info, sizeof(scan_info), "%%[A-Za-z0-9.]");
 107                 f = sscanf(url, scan_info, test->address);
 108                 g_snprintf(test->page, sizeof(test->page), "%c", '\0');
 109         }
 110
 111         sscanf(port, "%d", &test->port);
 112         return test;
 113 }
 114
 115 struct grab_url_data {
 116         void (* callback)(gpointer, char *);
 117         gpointer data;
 118         struct g_url *website;
 119         char *url;
 120         gboolean full;
 121
 122         int inpa;
 123
 124         gboolean sentreq;
 125         gboolean newline;
 126         gboolean startsaving;
 127         char *webdata;
 128         int len;
 129 };
 130
 131 static void grab_url_callback(gpointer dat, gint sock, GaimInputCondition cond)
 132 {
 133         struct grab_url_data *gunk = dat;
 134         char data;
 135
 136         if (sock == -1) {
 137                 gunk->callback(gunk->data, NULL);
 138                 g_free(gunk->website);
 139                 g_free(gunk->url);
 140                 g_free(gunk);
 141                 return;
 142         }
 143
 144         if (!gunk->sentreq) {
 145                 char buf[256];
 146                 g_snprintf(buf, sizeof(buf), "GET %s%s HTTP/1.0\r\n\r\n", gunk->full ? "" : "/",
 147                            gunk->full ? gunk->url : gunk->website->page);
 148                 debug_printf("Request: %s\n", buf);
 149                 write(sock, buf, strlen(buf));
 150                 fcntl(sock, F_SETFL, O_NONBLOCK);
 151                 gunk->sentreq = TRUE;
 152                 gunk->inpa = gaim_input_add(sock, GAIM_INPUT_READ, grab_url_callback, dat);
 153                 return;
 154         }
 155
 156         if (read(sock, &data, 1) > 0 || errno == EWOULDBLOCK) {
 157                 if (errno == EWOULDBLOCK) {
 158                         errno = 0;
 159                         return;
 160                 }
 161
 162                 if (!gunk->startsaving) {
 163                         if (data == '\r')
 164                                 return;
 165                         if (data == '\n') {
 166                                 if (gunk->newline)
 167                                         gunk->startsaving = TRUE;
 168                                 else
 169                                         gunk->newline = TRUE;
 170                                 return;
 171                         }
 172                         gunk->newline = FALSE;
 173                 } else {
 174                         gunk->len++;
 175                         gunk->webdata = g_realloc(gunk->webdata, gunk->len);
 176                         gunk->webdata[gunk->len - 1] = data;
 177                 }
 178         } else if (errno != ETIMEDOUT) {
 179
 180                 gunk->webdata = g_realloc(gunk->webdata, gunk->len + 1);
 181                 gunk->webdata[gunk->len] = 0;
 182
 183                 debug_printf(_("Receieved: '%s'\n"), gunk->webdata);
 184
 185                 gaim_input_remove(gunk->inpa);
 186                 close(sock);
 187                 gunk->callback(gunk->data, gunk->webdata);
 188                 if (gunk->webdata)
 189                         g_free(gunk->webdata);
 190                 g_free(gunk->website);
 191                 g_free(gunk->url);
 192                 g_free(gunk);
 193         } else {
 194                 gaim_input_remove(gunk->inpa);
 195                 close(sock);
 196                 gunk->callback(gunk->data, NULL);
 197                 if (gunk->webdata)
 198                         g_free(gunk->webdata);
 199                 g_free(gunk->website);
 200                 g_free(gunk->url);
 201                 g_free(gunk);
 202         }
 203 }
 204
 205 void grab_url(char *url, gboolean full, void callback(gpointer, char *), gpointer data)
 206 {
 207         int sock;
 208         struct grab_url_data *gunk = g_new0(struct grab_url_data, 1);
 209
 210         gunk->callback = callback;
 211         gunk->data = data;
 212         gunk->url = g_strdup(url);
 213         gunk->website = parse_url(url);
 214         gunk->full = full;
 215
 216         if ((sock = proxy_connect(gunk->website->address, gunk->website->port,
 217                                   grab_url_callback, gunk)) < 0) {
 218                 g_free(gunk->website);
 219                 g_free(gunk->url);
 220                 g_free(gunk);
 221                 callback(data, g_strdup(_("g003: Error opening connection.\n")));
 222         }
 223 }