1 /*- 2 * Copyright (c) 1998 Dag-Erling Co�dan Sm�rgrav 3 * All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer 10 * in this position and unchanged. 11 * 2. Redistributions in binary form must reproduce the above copyright 12 * notice, this list of conditions and the following disclaimer in the 13 * documentation and/or other materials provided with the distribution. 14 * 3. The name of the author may not be used to endorse or promote products 15 * derived from this software without specific prior written permission 16 * 17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR 18 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 19 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 20 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, 21 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 22 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 23 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 24 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 25 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 26 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 27 * 28 * $FreeBSD$ 29 */ 30 31 #include <sys/param.h> 32 #include <sys/errno.h> 33 34 #include <ctype.h> 35 #include <stdio.h> 36 #include <stdlib.h> 37 #include <string.h> 38 39 #include "fetch.h" 40 #include "common.h" 41 42 43 int fetchLastErrCode; 44 int fetchTimeout; 45 46 47 /*** Local data **************************************************************/ 48 49 /* 50 * Error messages for parser errors 51 */ 52 #define URL_MALFORMED 1 53 #define URL_BAD_SCHEME 2 54 #define URL_BAD_PORT 3 55 static struct fetcherr _url_errlist[] = { 56 { URL_MALFORMED, FETCH_URL, "Malformed URL" }, 57 { URL_BAD_SCHEME, FETCH_URL, "Invalid URL scheme" }, 58 { URL_BAD_PORT, FETCH_URL, "Invalid server port" }, 59 { -1, FETCH_UNKNOWN, "Unknown parser error" } 60 }; 61 62 63 /*** Public API **************************************************************/ 64 65 /* 66 * Select the appropriate protocol for the URL scheme, and return a 67 * read-only stream connected to the document referenced by the URL. 68 */ 69 FILE * 70 fetchGet(struct url *URL, char *flags) 71 { 72 if (strcasecmp(URL->scheme, "file") == 0) 73 return fetchGetFile(URL, flags); 74 else if (strcasecmp(URL->scheme, "http") == 0) 75 return fetchGetHTTP(URL, flags); 76 else if (strcasecmp(URL->scheme, "ftp") == 0) 77 return fetchGetFTP(URL, flags); 78 else { 79 _url_seterr(URL_BAD_SCHEME); 80 return NULL; 81 } 82 } 83 84 /* 85 * Select the appropriate protocol for the URL scheme, and return a 86 * write-only stream connected to the document referenced by the URL. 87 */ 88 FILE * 89 fetchPut(struct url *URL, char *flags) 90 { 91 if (strcasecmp(URL->scheme, "file") == 0) 92 return fetchPutFile(URL, flags); 93 else if (strcasecmp(URL->scheme, "http") == 0) 94 return fetchPutHTTP(URL, flags); 95 else if (strcasecmp(URL->scheme, "ftp") == 0) 96 return fetchPutFTP(URL, flags); 97 else { 98 _url_seterr(URL_BAD_SCHEME); 99 return NULL; 100 } 101 } 102 103 /* 104 * Select the appropriate protocol for the URL scheme, and return the 105 * size of the document referenced by the URL if it exists. 106 */ 107 int 108 fetchStat(struct url *URL, struct url_stat *us, char *flags) 109 { 110 if (strcasecmp(URL->scheme, "file") == 0) 111 return fetchStatFile(URL, us, flags); 112 else if (strcasecmp(URL->scheme, "http") == 0) 113 return fetchStatHTTP(URL, us, flags); 114 else if (strcasecmp(URL->scheme, "ftp") == 0) 115 return fetchStatFTP(URL, us, flags); 116 else { 117 _url_seterr(URL_BAD_SCHEME); 118 return -1; 119 } 120 } 121 122 /* 123 * Select the appropriate protocol for the URL scheme, and return a 124 * list of files in the directory pointed to by the URL. 125 */ 126 struct url_ent * 127 fetchList(struct url *URL, char *flags) 128 { 129 if (strcasecmp(URL->scheme, "file") == 0) 130 return fetchListFile(URL, flags); 131 else if (strcasecmp(URL->scheme, "http") == 0) 132 return fetchListHTTP(URL, flags); 133 else if (strcasecmp(URL->scheme, "ftp") == 0) 134 return fetchListFTP(URL, flags); 135 else { 136 _url_seterr(URL_BAD_SCHEME); 137 return NULL; 138 } 139 } 140 141 /* 142 * Attempt to parse the given URL; if successful, call fetchGet(). 143 */ 144 FILE * 145 fetchGetURL(char *URL, char *flags) 146 { 147 struct url *u; 148 FILE *f; 149 150 if ((u = fetchParseURL(URL)) == NULL) 151 return NULL; 152 153 f = fetchGet(u, flags); 154 155 free(u); 156 return f; 157 } 158 159 160 /* 161 * Attempt to parse the given URL; if successful, call fetchPut(). 162 */ 163 FILE * 164 fetchPutURL(char *URL, char *flags) 165 { 166 struct url *u; 167 FILE *f; 168 169 if ((u = fetchParseURL(URL)) == NULL) 170 return NULL; 171 172 f = fetchPut(u, flags); 173 174 free(u); 175 return f; 176 } 177 178 /* 179 * Attempt to parse the given URL; if successful, call fetchStat(). 180 */ 181 int 182 fetchStatURL(char *URL, struct url_stat *us, char *flags) 183 { 184 struct url *u; 185 int s; 186 187 if ((u = fetchParseURL(URL)) == NULL) 188 return -1; 189 190 s = fetchStat(u, us, flags); 191 192 free(u); 193 return s; 194 } 195 196 /* 197 * Attempt to parse the given URL; if successful, call fetchList(). 198 */ 199 struct url_ent * 200 fetchListURL(char *URL, char *flags) 201 { 202 struct url *u; 203 struct url_ent *ue; 204 205 if ((u = fetchParseURL(URL)) == NULL) 206 return NULL; 207 208 ue = fetchList(u, flags); 209 210 free(u); 211 return ue; 212 } 213 214 /* 215 * Split an URL into components. URL syntax is: 216 * method:[//[user[:pwd]@]host[:port]]/[document] 217 * This almost, but not quite, RFC1738 URL syntax. 218 */ 219 struct url * 220 fetchParseURL(char *URL) 221 { 222 char *p, *q; 223 struct url *u; 224 int i; 225 226 /* allocate struct url */ 227 if ((u = calloc(1, sizeof(struct url))) == NULL) { 228 errno = ENOMEM; 229 _fetch_syserr(); 230 return NULL; 231 } 232 233 /* scheme name */ 234 for (i = 0; *URL && (*URL != ':'); URL++) 235 if (i < URL_SCHEMELEN) 236 u->scheme[i++] = *URL; 237 if (!URL[0] || (URL[1] != '/')) { 238 _url_seterr(URL_BAD_SCHEME); 239 goto ouch; 240 } 241 else URL++; 242 if (URL[1] != '/') { 243 p = URL; 244 goto nohost; 245 } 246 else URL += 2; 247 248 p = strpbrk(URL, "/@"); 249 if (p && *p == '@') { 250 /* username */ 251 for (q = URL, i = 0; (*q != ':') && (*q != '@'); q++) 252 if (i < URL_USERLEN) 253 u->user[i++] = *q; 254 255 /* password */ 256 if (*q == ':') 257 for (q++, i = 0; (*q != ':') && (*q != '@'); q++) 258 if (i < URL_PWDLEN) 259 u->pwd[i++] = *q; 260 261 p++; 262 } else p = URL; 263 264 /* hostname */ 265 for (i = 0; *p && (*p != '/') && (*p != ':'); p++) 266 if (i < MAXHOSTNAMELEN) 267 u->host[i++] = *p; 268 269 /* port */ 270 if (*p == ':') { 271 for (q = ++p; *q && (*q != '/'); q++) 272 if (isdigit(*q)) 273 u->port = u->port * 10 + (*q - '0'); 274 else { 275 /* invalid port */ 276 _url_seterr(URL_BAD_PORT); 277 goto ouch; 278 } 279 while (*p && (*p != '/')) 280 p++; 281 } 282 283 nohost: 284 /* document */ 285 if (*p) { 286 struct url *t; 287 t = realloc(u, sizeof(*u)+strlen(p)-1); 288 if (t == NULL) { 289 errno = ENOMEM; 290 _fetch_syserr(); 291 goto ouch; 292 } 293 u = t; 294 strcpy(u->doc, p); 295 } else { 296 u->doc[0] = '/'; 297 u->doc[1] = 0; 298 } 299 300 DEBUG(fprintf(stderr, 301 "scheme: [\033[1m%s\033[m]\n" 302 "user: [\033[1m%s\033[m]\n" 303 "password: [\033[1m%s\033[m]\n" 304 "host: [\033[1m%s\033[m]\n" 305 "port: [\033[1m%d\033[m]\n" 306 "document: [\033[1m%s\033[m]\n", 307 u->scheme, u->user, u->pwd, 308 u->host, u->port, u->doc)); 309 310 return u; 311 312 ouch: 313 free(u); 314 return NULL; 315 } 316