lib/libfetch/fetch.c

   1 /*-
   2  * Copyright (c) 1998-2004 Dag-Erling Coïdan Smørgrav
   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: src/lib/libfetch/fetch.c,v 1.42 2008/12/17 18:00:18 murray Exp $
  29  * $DragonFly: src/lib/libfetch/fetch.c,v 1.3 2007/08/05 21:48:12 swildner Exp $
  30  */
  31
  32 #include <sys/param.h>
  33 #include <sys/errno.h>
  34
  35 #include <ctype.h>
  36 #include <stdio.h>
  37 #include <stdlib.h>
  38 #include <string.h>
  39
  40 #include "fetch.h"
  41 #include "common.h"
  42
  43 auth_t   fetchAuthMethod;
  44 int      fetchLastErrCode;
  45 char     fetchLastErrString[MAXERRSTRING];
  46 int      fetchTimeout;
  47 int      fetchRestartCalls = 1;
  48 int      fetchDebug;
  49
  50
  51 /*** Local data **************************************************************/
  52
  53 /*
  54  * Error messages for parser errors
  55  */
  56 #define URL_MALFORMED           1
  57 #define URL_BAD_SCHEME          2
  58 #define URL_BAD_PORT            3
  59 static struct fetcherr url_errlist[] = {
  60         { URL_MALFORMED,        FETCH_URL,      "Malformed URL" },
  61         { URL_BAD_SCHEME,       FETCH_URL,      "Invalid URL scheme" },
  62         { URL_BAD_PORT,         FETCH_URL,      "Invalid server port" },
  63         { -1,                   FETCH_UNKNOWN,  "Unknown parser error" }
  64 };
  65
  66
  67 /*** Public API **************************************************************/
  68
  69 /*
  70  * Select the appropriate protocol for the URL scheme, and return a
  71  * read-only stream connected to the document referenced by the URL.
  72  * Also fill out the struct url_stat.
  73  */
  74 FILE *
  75 fetchXGet(struct url *URL, struct url_stat *us, const char *flags)
  76 {
  77
  78         if (us != NULL) {
  79                 us->size = -1;
  80                 us->atime = us->mtime = 0;
  81         }
  82         if (strcasecmp(URL->scheme, SCHEME_FILE) == 0)
  83                 return (fetchXGetFile(URL, us, flags));
  84         else if (strcasecmp(URL->scheme, SCHEME_FTP) == 0)
  85                 return (fetchXGetFTP(URL, us, flags));
  86         else if (strcasecmp(URL->scheme, SCHEME_HTTP) == 0)
  87                 return (fetchXGetHTTP(URL, us, flags));
  88         else if (strcasecmp(URL->scheme, SCHEME_HTTPS) == 0)
  89                 return (fetchXGetHTTP(URL, us, flags));
  90         url_seterr(URL_BAD_SCHEME);
  91         return (NULL);
  92 }
  93
  94 /*
  95  * Select the appropriate protocol for the URL scheme, and return a
  96  * read-only stream connected to the document referenced by the URL.
  97  */
  98 FILE *
  99 fetchGet(struct url *URL, const char *flags)
 100 {
 101         return (fetchXGet(URL, NULL, flags));
 102 }
 103
 104 /*
 105  * Select the appropriate protocol for the URL scheme, and return a
 106  * write-only stream connected to the document referenced by the URL.
 107  */
 108 FILE *
 109 fetchPut(struct url *URL, const char *flags)
 110 {
 111
 112         if (strcasecmp(URL->scheme, SCHEME_FILE) == 0)
 113                 return (fetchPutFile(URL, flags));
 114         else if (strcasecmp(URL->scheme, SCHEME_FTP) == 0)
 115                 return (fetchPutFTP(URL, flags));
 116         else if (strcasecmp(URL->scheme, SCHEME_HTTP) == 0)
 117                 return (fetchPutHTTP(URL, flags));
 118         else if (strcasecmp(URL->scheme, SCHEME_HTTPS) == 0)
 119                 return (fetchPutHTTP(URL, flags));
 120         url_seterr(URL_BAD_SCHEME);
 121         return (NULL);
 122 }
 123
 124 /*
 125  * Select the appropriate protocol for the URL scheme, and return the
 126  * size of the document referenced by the URL if it exists.
 127  */
 128 int
 129 fetchStat(struct url *URL, struct url_stat *us, const char *flags)
 130 {
 131
 132         if (us != NULL) {
 133                 us->size = -1;
 134                 us->atime = us->mtime = 0;
 135         }
 136         if (strcasecmp(URL->scheme, SCHEME_FILE) == 0)
 137                 return (fetchStatFile(URL, us, flags));
 138         else if (strcasecmp(URL->scheme, SCHEME_FTP) == 0)
 139                 return (fetchStatFTP(URL, us, flags));
 140         else if (strcasecmp(URL->scheme, SCHEME_HTTP) == 0)
 141                 return (fetchStatHTTP(URL, us, flags));
 142         else if (strcasecmp(URL->scheme, SCHEME_HTTPS) == 0)
 143                 return (fetchStatHTTP(URL, us, flags));
 144         url_seterr(URL_BAD_SCHEME);
 145         return (-1);
 146 }
 147
 148 /*
 149  * Select the appropriate protocol for the URL scheme, and return a
 150  * list of files in the directory pointed to by the URL.
 151  */
 152 struct url_ent *
 153 fetchList(struct url *URL, const char *flags)
 154 {
 155
 156         if (strcasecmp(URL->scheme, SCHEME_FILE) == 0)
 157                 return (fetchListFile(URL, flags));
 158         else if (strcasecmp(URL->scheme, SCHEME_FTP) == 0)
 159                 return (fetchListFTP(URL, flags));
 160         else if (strcasecmp(URL->scheme, SCHEME_HTTP) == 0)
 161                 return (fetchListHTTP(URL, flags));
 162         else if (strcasecmp(URL->scheme, SCHEME_HTTPS) == 0)
 163                 return (fetchListHTTP(URL, flags));
 164         url_seterr(URL_BAD_SCHEME);
 165         return (NULL);
 166 }
 167
 168 /*
 169  * Attempt to parse the given URL; if successful, call fetchXGet().
 170  */
 171 FILE *
 172 fetchXGetURL(const char *URL, struct url_stat *us, const char *flags)
 173 {
 174         struct url *u;
 175         FILE *f;
 176
 177         if ((u = fetchParseURL(URL)) == NULL)
 178                 return (NULL);
 179
 180         f = fetchXGet(u, us, flags);
 181
 182         fetchFreeURL(u);
 183         return (f);
 184 }
 185
 186 /*
 187  * Attempt to parse the given URL; if successful, call fetchGet().
 188  */
 189 FILE *
 190 fetchGetURL(const char *URL, const char *flags)
 191 {
 192         return (fetchXGetURL(URL, NULL, flags));
 193 }
 194
 195 /*
 196  * Attempt to parse the given URL; if successful, call fetchPut().
 197  */
 198 FILE *
 199 fetchPutURL(const char *URL, const char *flags)
 200 {
 201         struct url *u;
 202         FILE *f;
 203
 204         if ((u = fetchParseURL(URL)) == NULL)
 205                 return (NULL);
 206
 207         f = fetchPut(u, flags);
 208
 209         fetchFreeURL(u);
 210         return (f);
 211 }
 212
 213 /*
 214  * Attempt to parse the given URL; if successful, call fetchStat().
 215  */
 216 int
 217 fetchStatURL(const char *URL, struct url_stat *us, const char *flags)
 218 {
 219         struct url *u;
 220         int s;
 221
 222         if ((u = fetchParseURL(URL)) == NULL)
 223                 return (-1);
 224
 225         s = fetchStat(u, us, flags);
 226
 227         fetchFreeURL(u);
 228         return (s);
 229 }
 230
 231 /*
 232  * Attempt to parse the given URL; if successful, call fetchList().
 233  */
 234 struct url_ent *
 235 fetchListURL(const char *URL, const char *flags)
 236 {
 237         struct url *u;
 238         struct url_ent *ue;
 239
 240         if ((u = fetchParseURL(URL)) == NULL)
 241                 return (NULL);
 242
 243         ue = fetchList(u, flags);
 244
 245         fetchFreeURL(u);
 246         return (ue);
 247 }
 248
 249 /*
 250  * Make a URL
 251  */
 252 struct url *
 253 fetchMakeURL(const char *scheme, const char *host, int port, const char *doc,
 254     const char *user, const char *pwd)
 255 {
 256         struct url *u;
 257
 258         if (!scheme || (!host && !doc)) {
 259                 url_seterr(URL_MALFORMED);
 260                 return (NULL);
 261         }
 262
 263         if (port < 0 || port > 65535) {
 264                 url_seterr(URL_BAD_PORT);
 265                 return (NULL);
 266         }
 267
 268         /* allocate struct url */
 269         if ((u = calloc(1, sizeof(*u))) == NULL) {
 270                 fetch_syserr();
 271                 return (NULL);
 272         }
 273
 274         if ((u->doc = strdup(doc ? doc : "/")) == NULL) {
 275                 fetch_syserr();
 276                 free(u);
 277                 return (NULL);
 278         }
 279
 280 #define seturl(x) snprintf(u->x, sizeof(u->x), "%s", x)
 281         seturl(scheme);
 282         seturl(host);
 283         seturl(user);
 284         seturl(pwd);
 285 #undef seturl
 286         u->port = port;
 287
 288         return (u);
 289 }
 290
 291 /*
 292  * Split an URL into components. URL syntax is:
 293  * [method:/][/[user[:pwd]@]host[:port]/][document]
 294  * This almost, but not quite, RFC1738 URL syntax.
 295  */
 296 struct url *
 297 fetchParseURL(const char *URL)
 298 {
 299         char *doc;
 300         const char *p, *q;
 301         struct url *u;
 302         int i;
 303
 304         /* allocate struct url */
 305         if ((u = calloc(1, sizeof(*u))) == NULL) {
 306                 fetch_syserr();
 307                 return (NULL);
 308         }
 309
 310         /* scheme name */
 311         if ((p = strstr(URL, ":/"))) {
 312                 snprintf(u->scheme, URL_SCHEMELEN+1,
 313                     "%.*s", (int)(p - URL), URL);
 314                 URL = ++p;
 315                 /*
 316                  * Only one slash: no host, leave slash as part of document
 317                  * Two slashes: host follows, strip slashes
 318                  */
 319                 if (URL[1] == '/')
 320                         URL = (p += 2);
 321         } else {
 322                 p = URL;
 323         }
 324         if (!*URL || *URL == '/' || *URL == '.' ||
 325             (u->scheme[0] == '\0' &&
 326                 strchr(URL, '/') == NULL && strchr(URL, ':') == NULL))
 327                 goto nohost;
 328
 329         p = strpbrk(URL, "/@");
 330         if (p && *p == '@') {
 331                 /* username */
 332                 for (q = URL, i = 0; (*q != ':') && (*q != '@'); q++)
 333                         if (i < URL_USERLEN)
 334                                 u->user[i++] = *q;
 335
 336                 /* password */
 337                 if (*q == ':')
 338                         for (q++, i = 0; (*q != ':') && (*q != '@'); q++)
 339                                 if (i < URL_PWDLEN)
 340                                         u->pwd[i++] = *q;
 341
 342                 p++;
 343         } else {
 344                 p = URL;
 345         }
 346
 347         /* hostname */
 348 #ifdef INET6
 349         if (*p == '[' && (q = strchr(p + 1, ']')) != NULL &&
 350             (*++q == '\0' || *q == '/' || *q == ':')) {
 351                 if ((i = q - p - 2) > MAXHOSTNAMELEN)
 352                         i = MAXHOSTNAMELEN;
 353                 strncpy(u->host, ++p, i);
 354                 p = q;
 355         } else
 356 #endif
 357                 for (i = 0; *p && (*p != '/') && (*p != ':'); p++)
 358                         if (i < MAXHOSTNAMELEN)
 359                                 u->host[i++] = *p;
 360
 361         /* port */
 362         if (*p == ':') {
 363                 for (q = ++p; *q && (*q != '/'); q++)
 364                         if (isdigit((unsigned char)*q))
 365                                 u->port = u->port * 10 + (*q - '0');
 366                         else {
 367                                 /* invalid port */
 368                                 url_seterr(URL_BAD_PORT);
 369                                 goto ouch;
 370                         }
 371                 p = q;
 372         }
 373
 374 nohost:
 375         /* document */
 376         if (!*p)
 377                 p = "/";
 378
 379         if (strcasecmp(u->scheme, SCHEME_HTTP) == 0 ||
 380             strcasecmp(u->scheme, SCHEME_HTTPS) == 0) {
 381                 const char hexnums[] = "0123456789abcdef";
 382
 383                 /* percent-escape whitespace. */
 384                 if ((doc = malloc(strlen(p) * 3 + 1)) == NULL) {
 385                         fetch_syserr();
 386                         goto ouch;
 387                 }
 388                 u->doc = doc;
 389                 while (*p != '\0') {
 390                         if (!isspace((unsigned char)*p)) {
 391                                 *doc++ = *p++;
 392                         } else {
 393                                 *doc++ = '%';
 394                                 *doc++ = hexnums[((unsigned int)*p) >> 4];
 395                                 *doc++ = hexnums[((unsigned int)*p) & 0xf];
 396                                 p++;
 397                         }
 398                 }
 399                 *doc = '\0';
 400         } else if ((u->doc = strdup(p)) == NULL) {
 401                 fetch_syserr();
 402                 goto ouch;
 403         }
 404
 405         DEBUG(fprintf(stderr,
 406                   "scheme:   [%s]\n"
 407                   "user:     [%s]\n"
 408                   "password: [%s]\n"
 409                   "host:     [%s]\n"
 410                   "port:     [%d]\n"
 411                   "document: [%s]\n",
 412                   u->scheme, u->user, u->pwd,
 413                   u->host, u->port, u->doc));
 414
 415         return (u);
 416
 417 ouch:
 418         free(u);
 419         return (NULL);
 420 }
 421
 422 /*
 423  * Free a URL
 424  */
 425 void
 426 fetchFreeURL(struct url *u)
 427 {
 428         free(u->doc);
 429         free(u);
 430 }