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
  29 #ifndef _WIN32
  30 #include <sys/time.h>
  31 #include <unistd.h>
  32 #include <sys/socket.h>
  33 #include <netdb.h>
  34 #include <netinet/in.h>
  35 #endif
  36
  37 #include <sys/types.h>
  38 #include <fcntl.h>
  39 #include <errno.h>
  40 #include "gaim.h"
  41 #include "proxy.h"
  42
  43 #ifdef _WIN32
  44 #include "win32dep.h"
  45 #endif
  46
  47 gchar *strip_html(gchar *text)
  48 {
  49         int i, j, k;
  50         int visible = 1;
  51         gchar *text2 = g_strdup(text);
  52
  53         for (i = 0, j = 0; text2[i]; i++) {
  54                 if (text2[i] == '<') {
  55                         k = i + 1;
  56                         while (text2[k]) {
  57                                 if (text2[k] == '<') {
  58                                         visible = 1;
  59                                         break;
  60                                 }
  61                                 if (text2[k] == '>') {
  62                                         visible = 0;
  63                                         break;
  64                                 }
  65                                 k++;
  66                         }
  67                 } else if (text2[i] == '>' && !visible) {
  68                         visible = 1;
  69                         continue;
  70                 }
  71                 if (visible) {
  72                         text2[j++] = text2[i];
  73                 }
  74         }
  75         text2[j] = '\0';
  76         return text2;
  77 }
  78
  79 struct g_url *parse_url(char *url)
  80 {
  81         struct g_url *test = g_new0(struct g_url, 1);
  82         char scan_info[255];
  83         char port[5];
  84         int f;
  85
  86         if (strstr(url, "http://"))
  87                 g_snprintf(scan_info, sizeof(scan_info),
  88                            "http://%%[A-Za-z0-9.]:%%[0-9]/%%[A-Za-z0-9.~_-/&%%?=+]");
  89         else
  90                 g_snprintf(scan_info, sizeof(scan_info),
  91                            "%%[A-Za-z0-9.]:%%[0-9]/%%[A-Za-z0-9.~_-/&%%?=+^]");
  92         f = sscanf(url, scan_info, test->address, port, test->page);
  93         if (f == 1) {
  94                 if (strstr(url, "http://"))
  95                         g_snprintf(scan_info, sizeof(scan_info),
  96                                    "http://%%[A-Za-z0-9.]/%%[A-Za-z0-9.~_-/&%%?=+^]");
  97                 else
  98                         g_snprintf(scan_info, sizeof(scan_info),
  99                                    "%%[A-Za-z0-9.]/%%[A-Za-z0-9.~_-/&%%?=+^]");
 100                 f = sscanf(url, scan_info, test->address, test->page);
 101                 g_snprintf(port, sizeof(test->port), "80");
 102                 port[2] = 0;
 103         }
 104         if (f == 1) {
 105                 if (strstr(url, "http://"))
 106                         g_snprintf(scan_info, sizeof(scan_info), "http://%%[A-Za-z0-9.]");
 107                 else
 108                         g_snprintf(scan_info, sizeof(scan_info), "%%[A-Za-z0-9.]");
 109                 f = sscanf(url, scan_info, test->address);
 110                 g_snprintf(test->page, sizeof(test->page), "%c", '\0');
 111         }
 112
 113         sscanf(port, "%d", &test->port);
 114         return test;
 115 }
 116
 117 struct grab_url_data {
 118         void (* callback)(gpointer, char *);
 119         gpointer data;
 120         struct g_url *website;
 121         char *url;
 122         gboolean full;
 123
 124         int inpa;
 125
 126         gboolean sentreq;
 127         gboolean newline;
 128         gboolean startsaving;
 129         char *webdata;
 130         int len;
 131 };
 132
 133 static void grab_url_callback(gpointer dat, gint sock, GaimInputCondition cond)
 134 {
 135         struct grab_url_data *gunk = dat;
 136         char data;
 137
 138         if (sock == -1) {
 139                 gunk->callback(gunk->data, NULL);
 140                 g_free(gunk->website);
 141                 g_free(gunk->url);
 142                 g_free(gunk);
 143                 return;
 144         }
 145
 146         if (!gunk->sentreq) {
 147                 char buf[256];
 148
 149                 g_snprintf(buf, sizeof(buf), "GET %s%s HTTP/1.0\r\n\r\n", gunk->full ? "" : "/",
 150                            gunk->full ? gunk->url : gunk->website->page);
 151                 debug_printf("Request: %s\n", buf);
 152
 153                 write(sock, buf, strlen(buf));
 154                 fcntl(sock, F_SETFL, O_NONBLOCK);
 155                 gunk->sentreq = TRUE;
 156                 gunk->inpa = gaim_input_add(sock, GAIM_INPUT_READ, grab_url_callback, dat);
 157                 return;
 158         }
 159
 160         if (read(sock, &data, 1) > 0 || errno == EWOULDBLOCK) {
 161                 if (errno == EWOULDBLOCK) {
 162                         errno = 0;
 163                         return;
 164                 }
 165                 if (!gunk->startsaving) {
 166                         if (data == '\r')
 167                                 return;
 168                         if (data == '\n') {
 169                                 if (gunk->newline)
 170                                         gunk->startsaving = TRUE;
 171                                 else
 172                                         gunk->newline = TRUE;
 173                                 return;
 174                         }
 175                         gunk->newline = FALSE;
 176                 } else {
 177                         gunk->len++;
 178                         gunk->webdata = g_realloc(gunk->webdata, gunk->len);
 179                         gunk->webdata[gunk->len - 1] = data;
 180                 }
 181         } else if (errno != ETIMEDOUT) {
 182                 gunk->webdata = g_realloc(gunk->webdata, gunk->len + 1);
 183                 gunk->webdata[gunk->len] = 0;
 184
 185                 debug_printf(_("Received: '%s'\n"), gunk->webdata);
 186
 187                 gaim_input_remove(gunk->inpa);
 188                 close(sock);
 189                 gunk->callback(gunk->data, gunk->webdata);
 190                 if (gunk->webdata)
 191                         g_free(gunk->webdata);
 192                 g_free(gunk->website);
 193                 g_free(gunk->url);
 194                 g_free(gunk);
 195         } else {
 196                 gaim_input_remove(gunk->inpa);
 197                 close(sock);
 198                 gunk->callback(gunk->data, NULL);
 199                 if (gunk->webdata)
 200                         g_free(gunk->webdata);
 201                 g_free(gunk->website);
 202                 g_free(gunk->url);
 203                 g_free(gunk);
 204         }
 205 }
 206
 207 void grab_url(char *url, gboolean full, void callback(gpointer, char *), gpointer data)
 208 {
 209         int sock;
 210         struct grab_url_data *gunk = g_new0(struct grab_url_data, 1);
 211
 212         gunk->callback = callback;
 213         gunk->data = data;
 214         gunk->url = g_strdup(url);
 215         gunk->website = parse_url(url);
 216         gunk->full = full;
 217
 218         if ((sock = proxy_connect(gunk->website->address, gunk->website->port,
 219                                   grab_url_callback, gunk)) < 0) {
 220                 g_free(gunk->website);
 221                 g_free(gunk->url);
 222                 g_free(gunk);
 223                 callback(data, g_strdup(_("g003: Error opening connection.\n")));
 224         }
 225 }