moparisthebest/wget
1
0
Fork 0
mirror of https://github.com/moparisthebest/wget synced 2024-07-03 16:38:41 -04:00
Code Issues Releases Wiki Activity
1ad97cc7ca
wget/src/ftp.c
Giuseppe Scrivano 3a62674582 Fix some memory leaks.
2011-03-11 15:25:58 +01:00

2297 lines
72 KiB
C
Raw Blame History

/* File Transfer Protocol support.
Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation,
Inc.
This file is part of GNU Wget.
GNU Wget is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
(at your option) any later version.
GNU Wget is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with Wget. If not, see <http://www.gnu.org/licenses/>.
Additional permission under GNU GPL version 3 section 7
If you modify this program, or any covered work, by linking or
combining it with the OpenSSL project's OpenSSL library (or a
modified version of that library), containing parts covered by the
terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
grants you additional permission to convey the resulting work.
Corresponding Source for a non-source form of such a combination
shall include the source code for the parts of OpenSSL used as well
as that of the covered work. */
#include "wget.h"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <strings.h>
#include <unistd.h>
#include <assert.h>
#include <errno.h>
#include <time.h>
#include "utils.h"
#include "url.h"
#include "retr.h"
#include "ftp.h"
#include "connect.h"
#include "host.h"
#include "netrc.h"
#include "convert.h" /* for downloaded_file */
#include "recur.h" /* for INFINITE_RECURSION */
#ifdef __VMS
# include "vms.h"
#endif /* def __VMS */
/* File where the "ls -al" listing will be saved. */
#ifdef MSDOS
#define LIST_FILENAME "_listing"
#else
#define LIST_FILENAME ".listing"
#endif
typedef struct
{
int st; /* connection status */
int cmd; /* command code */
int csock; /* control connection socket */
double dltime; /* time of the download in msecs */
enum stype rs; /* remote system reported by ftp server */
char *id; /* initial directory */
char *target; /* target file name */
struct url *proxy; /* FTWK-style proxy */
} ccon;
extern int numurls;
/* Look for regexp "( *[0-9]+ *byte" (literal parenthesis) anywhere in
the string S, and return the number converted to wgint, if found, 0
otherwise. */
static wgint
ftp_expected_bytes (const char *s)
{
wgint res;
while (1)
{
while (*s && *s != '(')
++s;
if (!*s)
return 0;
++s; /* skip the '(' */
res = str_to_wgint (s, (char **) &s, 10);
if (!*s)
return 0;
while (*s && c_isspace (*s))
++s;
if (!*s)
return 0;
if (c_tolower (*s) != 'b')
continue;
if (strncasecmp (s, "byte", 4))
continue;
else
break;
}
return res;
}
#ifdef ENABLE_IPV6
/*
* This function sets up a passive data connection with the FTP server.
* It is merely a wrapper around ftp_epsv, ftp_lpsv and ftp_pasv.
*/
static uerr_t
ftp_do_pasv (int csock, ip_address *addr, int *port)
{
uerr_t err;
/* We need to determine the address family and need to call
getpeername, so while we're at it, store the address to ADDR.
ftp_pasv and ftp_lpsv can simply override it. */
if (!socket_ip_address (csock, addr, ENDPOINT_PEER))
abort ();
/* If our control connection is over IPv6, then we first try EPSV and then
* LPSV if the former is not supported. If the control connection is over
* IPv4, we simply issue the good old PASV request. */
switch (addr->family)
{
case AF_INET:
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> PASV ... ");
err = ftp_pasv (csock, addr, port);
break;
case AF_INET6:
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> EPSV ... ");
err = ftp_epsv (csock, addr, port);
/* If EPSV is not supported try LPSV */
if (err == FTPNOPASV)
{
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> LPSV ... ");
err = ftp_lpsv (csock, addr, port);
}
break;
default:
abort ();
}
return err;
}
/*
* This function sets up an active data connection with the FTP server.
* It is merely a wrapper around ftp_eprt, ftp_lprt and ftp_port.
*/
static uerr_t
ftp_do_port (int csock, int *local_sock)
{
uerr_t err;
ip_address cip;
if (!socket_ip_address (csock, &cip, ENDPOINT_PEER))
abort ();
/* If our control connection is over IPv6, then we first try EPRT and then
* LPRT if the former is not supported. If the control connection is over
* IPv4, we simply issue the good old PORT request. */
switch (cip.family)
{
case AF_INET:
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> PORT ... ");
err = ftp_port (csock, local_sock);
break;
case AF_INET6:
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> EPRT ... ");
err = ftp_eprt (csock, local_sock);
/* If EPRT is not supported try LPRT */
if (err == FTPPORTERR)
{
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> LPRT ... ");
err = ftp_lprt (csock, local_sock);
}
break;
default:
abort ();
}
return err;
}
#else
static uerr_t
ftp_do_pasv (int csock, ip_address *addr, int *port)
{
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> PASV ... ");
return ftp_pasv (csock, addr, port);
}
static uerr_t
ftp_do_port (int csock, int *local_sock)
{
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> PORT ... ");
return ftp_port (csock, local_sock);
}
#endif
static void
print_length (wgint size, wgint start, bool authoritative)
{
logprintf (LOG_VERBOSE, _("Length: %s"), number_to_static_string (size));
if (size >= 1024)
logprintf (LOG_VERBOSE, " (%s)", human_readable (size));
if (start > 0)
{
if (size - start >= 1024)
logprintf (LOG_VERBOSE, _(", %s (%s) remaining"),
number_to_static_string (size - start),
human_readable (size - start));
else
logprintf (LOG_VERBOSE, _(", %s remaining"),
number_to_static_string (size - start));
}
logputs (LOG_VERBOSE, !authoritative ? _(" (unauthoritative)\n") : "\n");
}
static uerr_t ftp_get_listing (struct url *, ccon *, struct fileinfo **);
/* Retrieves a file with denoted parameters through opening an FTP
connection to the server. It always closes the data connection,
and closes the control connection in case of error. */
static uerr_t
getftp (struct url *u, wgint passed_expected_bytes, wgint *qtyread,
wgint restval, ccon *con, int count)
{
int csock, dtsock, local_sock, res;
uerr_t err = RETROK; /* appease the compiler */
FILE *fp;
char *user, *passwd, *respline;
char *tms;
const char *tmrate;
int cmd = con->cmd;
bool pasv_mode_open = false;
wgint expected_bytes = 0;
bool got_expected_bytes = false;
bool rest_failed = false;
int flags;
wgint rd_size;
char type_char;
assert (con != NULL);
assert (con->target != NULL);
/* Debug-check of the sanity of the request by making sure that LIST
and RETR are never both requested (since we can handle only one
at a time. */
assert (!((cmd & DO_LIST) && (cmd & DO_RETR)));
/* Make sure that at least *something* is requested. */
assert ((cmd & (DO_LIST | DO_CWD | DO_RETR | DO_LOGIN)) != 0);
*qtyread = restval;
user = u->user;
passwd = u->passwd;
search_netrc (u->host, (const char **)&user, (const char **)&passwd, 1);
user = user ? user : (opt.ftp_user ? opt.ftp_user : opt.user);
if (!user) user = "anonymous";
passwd = passwd ? passwd : (opt.ftp_passwd ? opt.ftp_passwd : opt.passwd);
if (!passwd) passwd = "-wget@";
dtsock = -1;
local_sock = -1;
con->dltime = 0;
if (!(cmd & DO_LOGIN))
csock = con->csock;
else /* cmd & DO_LOGIN */
{
char *host = con->proxy ? con->proxy->host : u->host;
int port = con->proxy ? con->proxy->port : u->port;
char *logname = user;
if (con->proxy)
{
/* If proxy is in use, log in as username@target-site. */
logname = concat_strings (user, "@", u->host, (char *) 0);
}
/* Login to the server: */
/* First: Establish the control connection. */
csock = connect_to_host (host, port);
if (csock == E_HOST)
{
if (con->proxy)
xfree (logname);
return HOSTERR;
}
else if (csock < 0)
{
if (con->proxy)
xfree (logname);
return (retryable_socket_connect_error (errno)
? CONERROR : CONIMPOSSIBLE);
}
if (cmd & LEAVE_PENDING)
con->csock = csock;
else
con->csock = -1;
/* Second: Login with proper USER/PASS sequence. */
logprintf (LOG_VERBOSE, _("Logging in as %s ... "),
quotearg_style (escape_quoting_style, user));
if (opt.server_response)
logputs (LOG_ALWAYS, "\n");
err = ftp_login (csock, logname, passwd);
if (con->proxy)
xfree (logname);
/* FTPRERR, FTPSRVERR, WRITEFAILED, FTPLOGREFUSED, FTPLOGINC */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPSRVERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("Error in server greeting.\n"));
fd_close (csock);
con->csock = -1;
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPLOGREFUSED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("The server refuses login.\n"));
fd_close (csock);
con->csock = -1;
return FTPLOGREFUSED;
case FTPLOGINC:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("Login incorrect.\n"));
fd_close (csock);
con->csock = -1;
return FTPLOGINC;
case FTPOK:
if (!opt.server_response)
logputs (LOG_VERBOSE, _("Logged in!\n"));
break;
default:
abort ();
}
/* Third: Get the system type */
if (!opt.server_response)
logprintf (LOG_VERBOSE, "==> SYST ... ");
err = ftp_syst (csock, &con->rs);
/* FTPRERR */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPSRVERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Server error, can't determine system type.\n"));
break;
case FTPOK:
/* Everything is OK. */
break;
default:
abort ();
}
if (!opt.server_response && err != FTPSRVERR)
logputs (LOG_VERBOSE, _("done. "));
/* Fourth: Find the initial ftp directory */
if (!opt.server_response)
logprintf (LOG_VERBOSE, "==> PWD ... ");
err = ftp_pwd (csock, &con->id);
/* FTPRERR */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPSRVERR :
/* PWD unsupported -- assume "/". */
xfree_null (con->id);
con->id = xstrdup ("/");
break;
case FTPOK:
/* Everything is OK. */
break;
default:
abort ();
}
#if 0
/* 2004-09-17 SMS.
Don't help me out. Please.
A reasonably recent VMS FTP server will cope just fine with
UNIX file specifications. This code just spoils things.
Discarding the device name, for example, is not a wise move.
This code was disabled but left in as an example of what not
to do.
*/
/* VMS will report something like "PUB$DEVICE:[INITIAL.FOLDER]".
Convert it to "/INITIAL/FOLDER" */
if (con->rs == ST_VMS)
{
char *path = strchr (con->id, '[');
char *pathend = path ? strchr (path + 1, ']') : NULL;
if (!path || !pathend)
DEBUGP (("Initial VMS directory not in the form [...]!\n"));
else
{
char *idir = con->id;
DEBUGP (("Preprocessing the initial VMS directory\n"));
DEBUGP ((" old = '%s'\n", con->id));
/* We do the conversion in-place by copying the stuff
between [ and ] to the beginning, and changing dots
to slashes at the same time. */
*idir++ = '/';
for (++path; path < pathend; path++, idir++)
*idir = *path == '.' ? '/' : *path;
*idir = '\0';
DEBUGP ((" new = '%s'\n\n", con->id));
}
}
#endif /* 0 */
if (!opt.server_response)
logputs (LOG_VERBOSE, _("done.\n"));
/* Fifth: Set the FTP type. */
type_char = ftp_process_type (u->params);
if (!opt.server_response)
logprintf (LOG_VERBOSE, "==> TYPE %c ... ", type_char);
err = ftp_type (csock, type_char);
/* FTPRERR, WRITEFAILED, FTPUNKNOWNTYPE */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPUNKNOWNTYPE:
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET,
_("Unknown type `%c', closing control connection.\n"),
type_char);
fd_close (csock);
con->csock = -1;
return err;
case FTPOK:
/* Everything is OK. */
break;
default:
abort ();
}
if (!opt.server_response)
logputs (LOG_VERBOSE, _("done. "));
} /* do login */
if (cmd & DO_CWD)
{
if (!*u->dir)
logputs (LOG_VERBOSE, _("==> CWD not needed.\n"));
else
{
char *targ = NULL;
int cwd_count;
int cwd_end;
int cwd_start;
char *target = u->dir;
DEBUGP (("changing working directory\n"));
/* Change working directory. To change to a non-absolute
Unix directory, we need to prepend initial directory
(con->id) to it. Absolute directories "just work".
A relative directory is one that does not begin with '/'
and, on non-Unix OS'es, one that doesn't begin with
"[a-z]:".
This is not done for OS400, which doesn't use
"/"-delimited directories, nor does it support directory
hierarchies. "CWD foo" followed by "CWD bar" leaves us
in "bar", not in "foo/bar", as would be customary
elsewhere. */
/* 2004-09-20 SMS.
Why is this wise even on UNIX? It certainly fouls VMS.
See below for a more reliable, more universal method.
*/
/* 2008-04-22 MJC.
I'm not crazy about it either. I'm informed it's useful
for misconfigured servers that have some dirs in the path
with +x but -r, but this method is not RFC-conformant. I
understand the need to deal with crappy server
configurations, but it's far better to use the canonical
method first, and fall back to kludges second.
*/
if (target[0] != '/'
&& !(con->rs != ST_UNIX
&& c_isalpha (target[0])
&& target[1] == ':')
&& (con->rs != ST_OS400)
&& (con->rs != ST_VMS))
{
int idlen = strlen (con->id);
char *ntarget, *p;
/* Strip trailing slash(es) from con->id. */
while (idlen > 0 && con->id[idlen - 1] == '/')
--idlen;
p = ntarget = (char *)alloca (idlen + 1 + strlen (u->dir) + 1);
memcpy (p, con->id, idlen);
p += idlen;
*p++ = '/';
strcpy (p, target);
DEBUGP (("Prepended initial PWD to relative path:\n"));
DEBUGP ((" pwd: '%s'\n old: '%s'\n new: '%s'\n",
con->id, target, ntarget));
target = ntarget;
}
#if 0
/* 2004-09-17 SMS.
Don't help me out. Please.
A reasonably recent VMS FTP server will cope just fine with
UNIX file specifications. This code just spoils things.
Discarding the device name, for example, is not a wise
move.
This code was disabled but left in as an example of what
not to do.
*/
/* If the FTP host runs VMS, we will have to convert the absolute
directory path in UNIX notation to absolute directory path in
VMS notation as VMS FTP servers do not like UNIX notation of
absolute paths. "VMS notation" is [dir.subdir.subsubdir]. */
if (con->rs == ST_VMS)
{
char *tmpp;
char *ntarget = (char *)alloca (strlen (target) + 2);
/* We use a converted initial dir, so directories in
TARGET will be separated with slashes, something like
"/INITIAL/FOLDER/DIR/SUBDIR". Convert that to
"[INITIAL.FOLDER.DIR.SUBDIR]". */
strcpy (ntarget, target);
assert (*ntarget == '/');
*ntarget = '[';
for (tmpp = ntarget + 1; *tmpp; tmpp++)
if (*tmpp == '/')
*tmpp = '.';
*tmpp++ = ']';
*tmpp = '\0';
DEBUGP (("Changed file name to VMS syntax:\n"));
DEBUGP ((" Unix: '%s'\n VMS: '%s'\n", target, ntarget));
target = ntarget;
}
#endif /* 0 */
/* 2004-09-20 SMS.
A relative directory is relative to the initial directory.
Thus, what _is_ useful on VMS (and probably elsewhere) is
to CWD to the initial directory (ideally, whatever the
server reports, _exactly_, NOT badly UNIX-ixed), and then
CWD to the (new) relative directory. This should probably
be restructured as a function, called once or twice, but
I'm lazy enough to take the badly indented loop short-cut
for now.
*/
/* Decide on one pass (absolute) or two (relative).
The VMS restriction may be relaxed when the squirrely code
above is reformed.
*/
if ((con->rs == ST_VMS) && (target[0] != '/'))
{
cwd_start = 0;
DEBUGP (("Using two-step CWD for relative path.\n"));
}
else
{
/* Go straight to the target. */
cwd_start = 1;
}
/* At least one VMS FTP server (TCPware V5.6-2) can switch to
a UNIX emulation mode when given a UNIX-like directory
specification (like "a/b/c"). If allowed to continue this
way, LIST interpretation will be confused, because the
system type (SYST response) will not be re-checked, and
future UNIX-format directory listings (for multiple URLs or
"-r") will be horribly misinterpreted.
The cheap and nasty work-around is to do a "CWD []" after a
UNIX-like directory specification is used. (A single-level
directory is harmless.) This puts the TCPware server back
into VMS mode, and does no harm on other servers.
Unlike the rest of this block, this particular behavior
_is_ VMS-specific, so it gets its own VMS test.
*/
if ((con->rs == ST_VMS) && (strchr( target, '/') != NULL))
{
cwd_end = 3;
DEBUGP (("Using extra \"CWD []\" step for VMS server.\n"));
}
else
{
cwd_end = 2;
}
/* 2004-09-20 SMS. */
/* Sorry about the deviant indenting. Laziness. */
for (cwd_count = cwd_start; cwd_count < cwd_end; cwd_count++)
{
switch (cwd_count)
{
case 0:
/* Step one (optional): Go to the initial directory,
exactly as reported by the server.
*/
targ = con->id;
break;
case 1:
/* Step two: Go to the target directory. (Absolute or
relative will work now.)
*/
targ = target;
break;
case 2:
/* Step three (optional): "CWD []" to restore server
VMS-ness.
*/
targ = "[]";
break;
default:
/* Can't happen. */
assert (1);
}
if (!opt.server_response)
logprintf (LOG_VERBOSE, "==> CWD (%d) %s ... ", cwd_count,
quotearg_style (escape_quoting_style, target));
err = ftp_cwd (csock, targ);
/* FTPRERR, WRITEFAILED, FTPNSFOD */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPNSFOD:
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET, _("No such directory %s.\n\n"),
quote (u->dir));
fd_close (csock);
con->csock = -1;
return err;
case FTPOK:
break;
default:
abort ();
}
if (!opt.server_response)
logputs (LOG_VERBOSE, _("done.\n"));
} /* for */
/* 2004-09-20 SMS. */
/* End of deviant indenting. */
} /* else */
}
else /* do not CWD */
logputs (LOG_VERBOSE, _("==> CWD not required.\n"));
if ((cmd & DO_RETR) && passed_expected_bytes == 0)
{
if (opt.verbose)
{
if (!opt.server_response)
logprintf (LOG_VERBOSE, "==> SIZE %s ... ",
quotearg_style (escape_quoting_style, u->file));
}
err = ftp_size (csock, u->file, &expected_bytes);
/* FTPRERR */
switch (err)
{
case FTPRERR:
case FTPSRVERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPOK:
got_expected_bytes = true;
/* Everything is OK. */
break;
default:
abort ();
}
if (!opt.server_response)
logprintf (LOG_VERBOSE, expected_bytes ? "%s\n" : _("done.\n"),
number_to_static_string (expected_bytes));
}
if (cmd & DO_RETR && restval > 0 && restval == expected_bytes)
{
/* Server confirms that file has length restval. We should stop now.
Some servers (f.e. NcFTPd) return error when receive REST 0 */
logputs (LOG_VERBOSE, _("File has already been retrieved.\n"));
fd_close (csock);
con->csock = -1;
return RETRFINISHED;
}
/* If anything is to be retrieved, PORT (or PASV) must be sent. */
if (cmd & (DO_LIST | DO_RETR))
{
if (opt.ftp_pasv)
{
ip_address passive_addr;
int passive_port;
err = ftp_do_pasv (csock, &passive_addr, &passive_port);
/* FTPRERR, WRITEFAILED, FTPNOPASV, FTPINVPASV */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
return err;
case FTPNOPASV:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("Cannot initiate PASV transfer.\n"));
break;
case FTPINVPASV:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("Cannot parse PASV response.\n"));
break;
case FTPOK:
break;
default:
abort ();
} /* switch (err) */
if (err==FTPOK)
{
DEBUGP (("trying to connect to %s port %d\n",
print_address (&passive_addr), passive_port));
dtsock = connect_to_ip (&passive_addr, passive_port, NULL);
if (dtsock < 0)
{
int save_errno = errno;
fd_close (csock);
con->csock = -1;
logprintf (LOG_VERBOSE, _("couldn't connect to %s port %d: %s\n"),
print_address (&passive_addr), passive_port,
strerror (save_errno));
return (retryable_socket_connect_error (save_errno)
? CONERROR : CONIMPOSSIBLE);
}
pasv_mode_open = true; /* Flag to avoid accept port */
if (!opt.server_response)
logputs (LOG_VERBOSE, _("done. "));
} /* err==FTP_OK */
}
if (!pasv_mode_open) /* Try to use a port command if PASV failed */
{
err = ftp_do_port (csock, &local_sock);
/* FTPRERR, WRITEFAILED, bindport (FTPSYSERR), HOSTERR,
FTPPORTERR */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case CONSOCKERR:
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET, "socket: %s\n", strerror (errno));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case FTPSYSERR:
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET, _("Bind error (%s).\n"),
strerror (errno));
fd_close (dtsock);
return err;
case FTPPORTERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("Invalid PORT.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case FTPOK:
break;
default:
abort ();
} /* port switch */
if (!opt.server_response)
logputs (LOG_VERBOSE, _("done. "));
} /* dtsock == -1 */
} /* cmd & (DO_LIST | DO_RETR) */
/* Restart if needed. */
if (restval && (cmd & DO_RETR))
{
if (!opt.server_response)
logprintf (LOG_VERBOSE, "==> REST %s ... ",
number_to_static_string (restval));
err = ftp_rest (csock, restval);
/* FTPRERR, WRITEFAILED, FTPRESTFAIL */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case FTPRESTFAIL:
logputs (LOG_VERBOSE, _("\nREST failed, starting from scratch.\n"));
rest_failed = true;
break;
case FTPOK:
break;
default:
abort ();
}
if (err != FTPRESTFAIL && !opt.server_response)
logputs (LOG_VERBOSE, _("done. "));
} /* restval && cmd & DO_RETR */
if (cmd & DO_RETR)
{
/* If we're in spider mode, don't really retrieve anything except
the directory listing and verify whether the given "file" exists. */
if (opt.spider)
{
bool exists = false;
uerr_t res;
struct fileinfo *f;
res = ftp_get_listing (u, con, &f);
/* Set the DO_RETR command flag again, because it gets unset when
calling ftp_get_listing() and would otherwise cause an assertion
failure earlier on when this function gets repeatedly called
(e.g., when recursing). */
con->cmd |= DO_RETR;
if (res == RETROK)
{
while (f)
{
if (!strcmp (f->name, u->file))
{
exists = true;
break;
}
f = f->next;
}
if (exists)
{
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET, _("File %s exists.\n"),
quote (u->file));
}
else
{
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET, _("No such file %s.\n"),
quote (u->file));
}
}
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return RETRFINISHED;
}
if (opt.verbose)
{
if (!opt.server_response)
{
if (restval)
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_VERBOSE, "==> RETR %s ... ",
quotearg_style (escape_quoting_style, u->file));
}
}
err = ftp_retr (csock, u->file);
/* FTPRERR, WRITEFAILED, FTPNSFOD */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case FTPNSFOD:
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET, _("No such file %s.\n\n"),
quote (u->file));
fd_close (dtsock);
fd_close (local_sock);
return err;
case FTPOK:
break;
default:
abort ();
}
if (!opt.server_response)
logputs (LOG_VERBOSE, _("done.\n"));
if (! got_expected_bytes)
expected_bytes = ftp_expected_bytes (ftp_last_respline);
} /* do retrieve */
if (cmd & DO_LIST)
{
if (!opt.server_response)
logputs (LOG_VERBOSE, "==> LIST ... ");
/* As Maciej W. Rozycki (macro@ds2.pg.gda.pl) says, `LIST'
without arguments is better than `LIST .'; confirmed by
RFC959. */
err = ftp_list (csock, NULL, con->rs);
/* FTPRERR, WRITEFAILED */
switch (err)
{
case FTPRERR:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET, _("\
Error in server response, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case WRITEFAILED:
logputs (LOG_VERBOSE, "\n");
logputs (LOG_NOTQUIET,
_("Write failed, closing control connection.\n"));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return err;
case FTPNSFOD:
logputs (LOG_VERBOSE, "\n");
logprintf (LOG_NOTQUIET, _("No such file or directory %s.\n\n"),
quote ("."));
fd_close (dtsock);
fd_close (local_sock);
return err;
case FTPOK:
break;
default:
abort ();
}
if (!opt.server_response)
logputs (LOG_VERBOSE, _("done.\n"));
if (! got_expected_bytes)
expected_bytes = ftp_expected_bytes (ftp_last_respline);
} /* cmd & DO_LIST */
if (!(cmd & (DO_LIST | DO_RETR)) || (opt.spider && !(cmd & DO_LIST)))
return RETRFINISHED;
/* Some FTP servers return the total length of file after REST
command, others just return the remaining size. */
if (passed_expected_bytes && restval && expected_bytes
&& (expected_bytes == passed_expected_bytes - restval))
{
DEBUGP (("Lying FTP server found, adjusting.\n"));
expected_bytes = passed_expected_bytes;
}
/* If no transmission was required, then everything is OK. */
if (!pasv_mode_open) /* we are not using pasive mode so we need
to accept */
{
/* Wait for the server to connect to the address we're waiting
at. */
dtsock = accept_connection (local_sock);
if (dtsock < 0)
{
logprintf (LOG_NOTQUIET, "accept: %s\n", strerror (errno));
return CONERROR;
}
}
/* Open the file -- if output_stream is set, use it instead. */
/* 2005-04-17 SMS.
Note that having the output_stream ("-O") file opened in main()
(main.c) rather limits the ability in VMS to open the file
differently for ASCII versus binary FTP here. (Of course, doing it
there allows a open failure to be detected immediately, without first
connecting to the server.)
*/
if (!output_stream || con->cmd & DO_LIST)
{
/* On VMS, alter the name as required. */
#ifdef __VMS
char *targ;
targ = ods_conform (con->target);
if (targ != con->target)
{
xfree (con->target);
con->target = targ;
}
#endif /* def __VMS */
mkalldirs (con->target);
if (opt.backups)
rotate_backups (con->target);
/* 2005-04-15 SMS.
For VMS, define common fopen() optional arguments, and a handy macro
for use as a variable "binary" flag.
Elsewhere, define a constant "binary" flag.
Isn't it nice to have distinct text and binary file types?
*/
# define BIN_TYPE_TRANSFER (type_char != 'A')
#ifdef __VMS
# define FOPEN_OPT_ARGS "fop=sqo", "acc", acc_cb, &open_id
# define FOPEN_OPT_ARGS_BIN "ctx=bin,stm", "rfm=fix", "mrs=512" FOPEN_OPT_ARGS
# define BIN_TYPE_FILE (BIN_TYPE_TRANSFER && (opt.ftp_stmlf == 0))
#else /* def __VMS */
# define BIN_TYPE_FILE 1
#endif /* def __VMS [else] */
if (restval && !(con->cmd & DO_LIST))
{
#ifdef __VMS
int open_id;
if (BIN_TYPE_FILE)
{
open_id = 3;
fp = fopen (con->target, "ab", FOPEN_OPT_ARGS_BIN);
}
else
{
open_id = 4;
fp = fopen (con->target, "a", FOPEN_OPT_ARGS);
}
#else /* def __VMS */
fp = fopen (con->target, "ab");
#endif /* def __VMS [else] */
}
else if (opt.noclobber || opt.always_rest || opt.timestamping || opt.dirstruct
|| opt.output_document || count > 0)
{
if (opt.unlink && file_exists_p (con->target))
{
int res = unlink (con->target);
if (res < 0)
{
logprintf (LOG_NOTQUIET, "%s: %s\n", con->target,
strerror (errno));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return UNLINKERR;
}
}
#ifdef __VMS
int open_id;
if (BIN_TYPE_FILE)
{
open_id = 5;
fp = fopen (con->target, "wb", FOPEN_OPT_ARGS_BIN);
}
else
{
open_id = 6;
fp = fopen (con->target, "w", FOPEN_OPT_ARGS);
}
#else /* def __VMS */
fp = fopen (con->target, "wb");
#endif /* def __VMS [else] */
}
else
{
fp = fopen_excl (con->target, true);
if (!fp && errno == EEXIST)
{
/* We cannot just invent a new name and use it (which is
what functions like unique_create typically do)
because we told the user we'd use this name.
Instead, return and retry the download. */
logprintf (LOG_NOTQUIET, _("%s has sprung into existence.\n"),
con->target);
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return FOPEN_EXCL_ERR;
}
}
if (!fp)
{
logprintf (LOG_NOTQUIET, "%s: %s\n", con->target, strerror (errno));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
fd_close (local_sock);
return FOPENERR;
}
}
else
fp = output_stream;
if (passed_expected_bytes)
{
print_length (passed_expected_bytes, restval, true);
expected_bytes = passed_expected_bytes;
/* for fd_read_body's progress bar */
}
else if (expected_bytes)
print_length (expected_bytes, restval, false);
/* Get the contents of the document. */
flags = 0;
if (restval && rest_failed)
flags |= rb_skip_startpos;
rd_size = 0;
res = fd_read_body (dtsock, fp,
expected_bytes ? expected_bytes - restval : 0,
restval, &rd_size, qtyread, &con->dltime, flags);
tms = datetime_str (time (NULL));
tmrate = retr_rate (rd_size, con->dltime);
total_download_time += con->dltime;
fd_close (local_sock);
/* Close the local file. */
if (!output_stream || con->cmd & DO_LIST)
fclose (fp);
/* If fd_read_body couldn't write to fp, bail out. */
if (res == -2)
{
logprintf (LOG_NOTQUIET, _("%s: %s, closing control connection.\n"),
con->target, strerror (errno));
fd_close (csock);
con->csock = -1;
fd_close (dtsock);
return FWRITEERR;
}
else if (res == -1)
{
logprintf (LOG_NOTQUIET, _("%s (%s) - Data connection: %s; "),
tms, tmrate, fd_errstr (dtsock));
if (opt.server_response)
logputs (LOG_ALWAYS, "\n");
}
fd_close (dtsock);
/* Get the server to tell us if everything is retrieved. */
err = ftp_response (csock, &respline);
if (err != FTPOK)
{
/* The control connection is decidedly closed. Print the time
only if it hasn't already been printed. */
if (res != -1)
logprintf (LOG_NOTQUIET, "%s (%s) - ", tms, tmrate);
logputs (LOG_NOTQUIET, _("Control connection closed.\n"));
/* If there is an error on the control connection, close it, but
return FTPRETRINT, since there is a possibility that the
whole file was retrieved nevertheless (but that is for
ftp_loop_internal to decide). */
fd_close (csock);
con->csock = -1;
return FTPRETRINT;
} /* err != FTPOK */
/* If retrieval failed for any reason, return FTPRETRINT, but do not
close socket, since the control connection is still alive. If
there is something wrong with the control connection, it will
become apparent later. */
if (*respline != '2')
{
xfree (respline);
if (res != -1)
logprintf (LOG_NOTQUIET, "%s (%s) - ", tms, tmrate);
logputs (LOG_NOTQUIET, _("Data transfer aborted.\n"));
return FTPRETRINT;
}
xfree (respline);
if (res == -1)
{
/* What now? The data connection was erroneous, whereas the
response says everything is OK. We shall play it safe. */
return FTPRETRINT;
}
if (!(cmd & LEAVE_PENDING))
{
/* Closing the socket is faster than sending 'QUIT' and the
effect is the same. */
fd_close (csock);
con->csock = -1;
}
/* If it was a listing, and opt.server_response is true,
print it out. */
if (opt.server_response && (con->cmd & DO_LIST))
{
/* 2005-02-25 SMS.
Much of this work may already have been done, but repeating it should
do no damage beyond wasting time.
*/
/* On VMS, alter the name as required. */
#ifdef __VMS
char *targ;
targ = ods_conform( con->target);
if (targ != con->target)
{
xfree( con->target);
con->target = targ;
}
#endif /* def __VMS */
mkalldirs (con->target);
fp = fopen (con->target, "r");
if (!fp)
logprintf (LOG_ALWAYS, "%s: %s\n", con->target, strerror (errno));
else
{
char *line;
/* The lines are being read with read_whole_line because of
no-buffering on opt.lfile. */
while ((line = read_whole_line (fp)) != NULL)
{
char *p = strchr (line, '\0');
while (p > line && (p[-1] == '\n' || p[-1] == '\r'))
*--p = '\0';
logprintf (LOG_ALWAYS, "%s\n",
quotearg_style (escape_quoting_style, line));
xfree (line);
}
fclose (fp);
}
} /* con->cmd & DO_LIST && server_response */
return RETRFINISHED;
}
/* A one-file FTP loop. This is the part where FTP retrieval is
retried, and retried, and retried, and...
This loop either gets commands from con, or (if ON_YOUR_OWN is
set), makes them up to retrieve the file given by the URL. */
static uerr_t
ftp_loop_internal (struct url *u, struct fileinfo *f, ccon *con, char **local_file)
{
int count, orig_lp;
wgint restval, len = 0, qtyread = 0;
char *tms, *locf;
const char *tmrate = NULL;
uerr_t err;
struct_stat st;
/* Get the target, and set the name for the message accordingly. */
if ((f == NULL) && (con->target))
{
/* Explicit file (like ".listing"). */
locf = con->target;
}
else
{
/* URL-derived file. Consider "-O file" name. */
con->target = url_file_name (u, NULL);
if (!opt.output_document)
locf = con->target;
else
locf = opt.output_document;
}
/* If the output_document was given, then this check was already done and
the file didn't exist. Hence the !opt.output_document */
if (opt.noclobber && !opt.output_document && file_exists_p (con->target))
{
logprintf (LOG_VERBOSE,
_("File %s already there; not retrieving.\n"), quote (con->target));
/* If the file is there, we suppose it's retrieved OK. */
return RETROK;
}
/* Remove it if it's a link. */
remove_link (con->target);
count = 0;
if (con->st & ON_YOUR_OWN)
con->st = ON_YOUR_OWN;
orig_lp = con->cmd & LEAVE_PENDING ? 1 : 0;
/* THE loop. */
do
{
/* Increment the pass counter. */
++count;
sleep_between_retrievals (count);
if (con->st & ON_YOUR_OWN)
{
con->cmd = 0;
con->cmd |= (DO_RETR | LEAVE_PENDING);
if (con->csock != -1)
con->cmd &= ~ (DO_LOGIN | DO_CWD);
else
con->cmd |= (DO_LOGIN | DO_CWD);
}
else /* not on your own */
{
if (con->csock != -1)
con->cmd &= ~DO_LOGIN;
else
con->cmd |= DO_LOGIN;
if (con->st & DONE_CWD)
con->cmd &= ~DO_CWD;
else
con->cmd |= DO_CWD;
}
/* Decide whether or not to restart. */
if (con->cmd & DO_LIST)
restval = 0;
else if (opt.always_rest
&& stat (locf, &st) == 0
&& S_ISREG (st.st_mode))
/* When -c is used, continue from on-disk size. (Can't use
hstat.len even if count>1 because we don't want a failed
first attempt to clobber existing data.) */
restval = st.st_size;
else if (count > 1)
restval = qtyread; /* start where the previous run left off */
else
restval = 0;
/* Get the current time string. */
tms = datetime_str (time (NULL));
/* Print fetch message, if opt.verbose. */
if (opt.verbose)
{
char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
char tmp[256];
strcpy (tmp, " ");
if (count > 1)
sprintf (tmp, _("(try:%2d)"), count);
logprintf (LOG_VERBOSE, "--%s-- %s\n %s => %s\n",
tms, hurl, tmp, quote (locf));
#ifdef WINDOWS
ws_changetitle (hurl);
#endif
xfree (hurl);
}
/* Send getftp the proper length, if fileinfo was provided. */
if (f && f->type != FT_SYMLINK)
len = f->size;
else
len = 0;
err = getftp (u, len, &qtyread, restval, con, count);
if (con->csock == -1)
con->st &= ~DONE_CWD;
else
con->st |= DONE_CWD;
switch (err)
{
case HOSTERR: case CONIMPOSSIBLE: case FWRITEERR: case FOPENERR:
case FTPNSFOD: case FTPLOGINC: case FTPNOPASV: case CONTNOTSUPPORTED:
case UNLINKERR:
/* Fatal errors, give up. */
return err;
case CONSOCKERR: case CONERROR: case FTPSRVERR: case FTPRERR:
case WRITEFAILED: case FTPUNKNOWNTYPE: case FTPSYSERR:
case FTPPORTERR: case FTPLOGREFUSED: case FTPINVPASV:
case FOPEN_EXCL_ERR:
printwhat (count, opt.ntry);
/* non-fatal errors */
if (err == FOPEN_EXCL_ERR)
{
/* Re-determine the file name. */
xfree_null (con->target);
con->target = url_file_name (u, NULL);
locf = con->target;
}
continue;
case FTPRETRINT:
/* If the control connection was closed, the retrieval
will be considered OK if f->size == len. */
if (!f || qtyread != f->size)
{
printwhat (count, opt.ntry);
continue;
}
break;
case RETRFINISHED:
/* Great! */
break;
default:
/* Not as great. */
abort ();
}
tms = datetime_str (time (NULL));
if (!opt.spider)
tmrate = retr_rate (qtyread - restval, con->dltime);
/* If we get out of the switch above without continue'ing, we've
successfully downloaded a file. Remember this fact. */
downloaded_file (FILE_DOWNLOADED_NORMALLY, locf);
if (con->st & ON_YOUR_OWN)
{
fd_close (con->csock);
con->csock = -1;
}
if (!opt.spider)
{
bool write_to_stdout = (opt.output_document && HYPHENP (opt.output_document));
logprintf (LOG_VERBOSE,
write_to_stdout
? _("%s (%s) - written to stdout %s[%s]\n\n")
: _("%s (%s) - %s saved [%s]\n\n"),
tms, tmrate,
write_to_stdout ? "" : quote (locf),
number_to_static_string (qtyread));
}
if (!opt.verbose && !opt.quiet)
{
/* Need to hide the password from the URL. The `if' is here
so that we don't do the needless allocation every
time. */
char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
logprintf (LOG_NONVERBOSE, "%s URL: %s [%s] -> \"%s\" [%d]\n",
tms, hurl, number_to_static_string (qtyread), locf, count);
xfree (hurl);
}
if ((con->cmd & DO_LIST))
/* This is a directory listing file. */
{
if (!opt.remove_listing)
/* --dont-remove-listing was specified, so do count this towards the
number of bytes and files downloaded. */
{
total_downloaded_bytes += qtyread;
numurls++;
}
/* Deletion of listing files is not controlled by --delete-after, but
by the more specific option --dont-remove-listing, and the code
to do this deletion is in another function. */
}
else if (!opt.spider)
/* This is not a directory listing file. */
{
/* Unlike directory listing files, don't pretend normal files weren't
downloaded if they're going to be deleted. People seeding proxies,
for instance, may want to know how many bytes and files they've
downloaded through it. */
total_downloaded_bytes += qtyread;
numurls++;
if (opt.delete_after && !input_file_url (opt.input_filename))
{
DEBUGP (("\
Removing file due to --delete-after in ftp_loop_internal():\n"));
logprintf (LOG_VERBOSE, _("Removing %s.\n"), locf);
if (unlink (locf))
logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
}
}
/* Restore the original leave-pendingness. */
if (orig_lp)
con->cmd |= LEAVE_PENDING;
else
con->cmd &= ~LEAVE_PENDING;
if (local_file)
*local_file = xstrdup (locf);
return RETROK;
} while (!opt.ntry || (count < opt.ntry));
if (con->csock != -1 && (con->st & ON_YOUR_OWN))
{
fd_close (con->csock);
con->csock = -1;
}
return TRYLIMEXC;
}
/* Return the directory listing in a reusable format. The directory
is specifed in u->dir. */
static uerr_t
ftp_get_listing (struct url *u, ccon *con, struct fileinfo **f)
{
uerr_t err;
char *uf; /* url file name */
char *lf; /* list file name */
char *old_target = con->target;
con->st &= ~ON_YOUR_OWN;
con->cmd |= (DO_LIST | LEAVE_PENDING);
con->cmd &= ~DO_RETR;
/* Find the listing file name. We do it by taking the file name of
the URL and replacing the last component with the listing file
name. */
uf = url_file_name (u, NULL);
lf = file_merge (uf, LIST_FILENAME);
xfree (uf);
DEBUGP ((_("Using %s as listing tmp file.\n"), quote (lf)));
con->target = xstrdup (lf);
xfree (lf);
err = ftp_loop_internal (u, NULL, con, NULL);
lf = xstrdup (con->target);
xfree (con->target);
con->target = old_target;
if (err == RETROK)
{
*f = ftp_parse_ls (lf, con->rs);
if (opt.remove_listing)
{
if (unlink (lf))
logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
else
logprintf (LOG_VERBOSE, _("Removed %s.\n"), quote (lf));
}
}
else
*f = NULL;
xfree (lf);
con->cmd &= ~DO_LIST;
return err;
}
static uerr_t ftp_retrieve_dirs (struct url *, struct fileinfo *, ccon *);
static uerr_t ftp_retrieve_glob (struct url *, ccon *, int);
static struct fileinfo *delelement (struct fileinfo *, struct fileinfo **);
static void freefileinfo (struct fileinfo *f);
/* Retrieve a list of files given in struct fileinfo linked list. If
a file is a symbolic link, do not retrieve it, but rather try to
set up a similar link on the local disk, if the symlinks are
supported.
If opt.recursive is set, after all files have been retrieved,
ftp_retrieve_dirs will be called to retrieve the directories. */
static uerr_t
ftp_retrieve_list (struct url *u, struct fileinfo *f, ccon *con)
{
static int depth = 0;
uerr_t err;
struct fileinfo *orig;
wgint local_size;
time_t tml;
bool dlthis; /* Download this (file). */
const char *actual_target = NULL;
/* Increase the depth. */
++depth;
if (opt.reclevel != INFINITE_RECURSION && depth > opt.reclevel)
{
DEBUGP ((_("Recursion depth %d exceeded max. depth %d.\n"),
depth, opt.reclevel));
--depth;
return RECLEVELEXC;
}
assert (f != NULL);
orig = f;
con->st &= ~ON_YOUR_OWN;
if (!(con->st & DONE_CWD))
con->cmd |= DO_CWD;
else
con->cmd &= ~DO_CWD;
con->cmd |= (DO_RETR | LEAVE_PENDING);
if (con->csock < 0)
con->cmd |= DO_LOGIN;
else
con->cmd &= ~DO_LOGIN;
err = RETROK; /* in case it's not used */
while (f)
{
char *old_target, *ofile;
if (opt.quota && total_downloaded_bytes > opt.quota)
{
--depth;
return QUOTEXC;
}
old_target = con->target;
ofile = xstrdup (u->file);
url_set_file (u, f->name);
con->target = url_file_name (u, NULL);
err = RETROK;
dlthis = true;
if (opt.timestamping && f->type == FT_PLAINFILE)
{
struct_stat st;
/* If conversion of HTML files retrieved via FTP is ever implemented,
we'll need to stat() <file>.orig here when -K has been specified.
I'm not implementing it now since files on an FTP server are much
more likely than files on an HTTP server to legitimately have a
.orig suffix. */
if (!stat (con->target, &st))
{
bool eq_size;
bool cor_val;
/* Else, get it from the file. */
local_size = st.st_size;
tml = st.st_mtime;
#ifdef WINDOWS
/* Modification time granularity is 2 seconds for Windows, so
increase local time by 1 second for later comparison. */
tml++;
#endif
/* Compare file sizes only for servers that tell us correct
values. Assume sizes being equal for servers that lie
about file size. */
cor_val = (con->rs == ST_UNIX || con->rs == ST_WINNT);
eq_size = cor_val ? (local_size == f->size) : true;
if (f->tstamp <= tml && eq_size)
{
/* Remote file is older, file sizes can be compared and
are both equal. */
logprintf (LOG_VERBOSE, _("\
Remote file no newer than local file %s -- not retrieving.\n"), quote (con->target));
dlthis = false;
}
else if (eq_size)
{
/* Remote file is newer or sizes cannot be matched */
logprintf (LOG_VERBOSE, _("\
Remote file is newer than local file %s -- retrieving.\n\n"),
quote (con->target));
}
else
{
/* Sizes do not match */
logprintf (LOG_VERBOSE, _("\
The sizes do not match (local %s) -- retrieving.\n\n"),
number_to_static_string (local_size));
}
}
} /* opt.timestamping && f->type == FT_PLAINFILE */
switch (f->type)
{
case FT_SYMLINK:
/* If opt.retr_symlinks is defined, we treat symlinks as
if they were normal files. There is currently no way
to distinguish whether they might be directories, and
follow them. */
if (!opt.retr_symlinks)
{
#ifdef HAVE_SYMLINK
if (!f->linkto)
logputs (LOG_NOTQUIET,
_("Invalid name of the symlink, skipping.\n"));
else
{
struct_stat st;
/* Check whether we already have the correct
symbolic link. */
int rc = lstat (con->target, &st);
if (rc == 0)
{
size_t len = strlen (f->linkto) + 1;
if (S_ISLNK (st.st_mode))
{
char *link_target = (char *)alloca (len);
size_t n = readlink (con->target, link_target, len);
if ((n == len - 1)
&& (memcmp (link_target, f->linkto, n) == 0))
{
logprintf (LOG_VERBOSE, _("\
Already have correct symlink %s -> %s\n\n"),
quote (con->target),
quote (f->linkto));
dlthis = false;
break;
}
}
}
logprintf (LOG_VERBOSE, _("Creating symlink %s -> %s\n"),
quote (con->target), quote (f->linkto));
/* Unlink before creating symlink! */
unlink (con->target);
if (symlink (f->linkto, con->target) == -1)
logprintf (LOG_NOTQUIET, "symlink: %s\n", strerror (errno));
logputs (LOG_VERBOSE, "\n");
} /* have f->linkto */
#else /* not HAVE_SYMLINK */
logprintf (LOG_NOTQUIET,
_("Symlinks not supported, skipping symlink %s.\n"),
quote (con->target));
#endif /* not HAVE_SYMLINK */
}
else /* opt.retr_symlinks */
{
if (dlthis)
err = ftp_loop_internal (u, f, con, NULL);
} /* opt.retr_symlinks */
break;
case FT_DIRECTORY:
if (!opt.recursive)
logprintf (LOG_NOTQUIET, _("Skipping directory %s.\n"),
quote (f->name));
break;
case FT_PLAINFILE:
/* Call the retrieve loop. */
if (dlthis)
err = ftp_loop_internal (u, f, con, NULL);
break;
case FT_UNKNOWN:
logprintf (LOG_NOTQUIET, _("%s: unknown/unsupported file type.\n"),
quote (f->name));
break;
} /* switch */
/* 2004-12-15 SMS.
* Set permissions _before_ setting the times, as setting the
* permissions changes the modified-time, at least on VMS.
* Also, use the opt.output_document name here, too, as
* appropriate. (Do the test once, and save the result.)
*/
set_local_file (&actual_target, con->target);
/* If downloading a plain file, set valid (non-zero) permissions. */
if (dlthis && (actual_target != NULL) && (f->type == FT_PLAINFILE))
{
if (f->perms)
chmod (actual_target, f->perms);
else
DEBUGP (("Unrecognized permissions for %s.\n", actual_target));
}
/* Set the time-stamp information to the local file. Symlinks
are not to be stamped because it sets the stamp on the
original. :( */
if (actual_target != NULL)
{
if (opt.useservertimestamps
&& !(f->type == FT_SYMLINK && !opt.retr_symlinks)
&& f->tstamp != -1
&& dlthis
&& file_exists_p (con->target))
{
touch (actual_target, f->tstamp);
}
else if (f->tstamp == -1)
logprintf (LOG_NOTQUIET, _("%s: corrupt time-stamp.\n"),
actual_target);
}
xfree (con->target);
con->target = old_target;
url_set_file (u, ofile);
xfree (ofile);
/* Break on fatals. */
if (err == QUOTEXC || err == HOSTERR || err == FWRITEERR)
break;
con->cmd &= ~ (DO_CWD | DO_LOGIN);
f = f->next;
}
/* We do not want to call ftp_retrieve_dirs here */
if (opt.recursive &&
!(opt.reclevel != INFINITE_RECURSION && depth >= opt.reclevel))
err = ftp_retrieve_dirs (u, orig, con);
else if (opt.recursive)
DEBUGP ((_("Will not retrieve dirs since depth is %d (max %d).\n"),
depth, opt.reclevel));
--depth;
return err;
}
/* Retrieve the directories given in a file list. This function works
by simply going through the linked list and calling
ftp_retrieve_glob on each directory entry. The function knows
about excluded directories. */
static uerr_t
ftp_retrieve_dirs (struct url *u, struct fileinfo *f, ccon *con)
{
char *container = NULL;
int container_size = 0;
for (; f; f = f->next)
{
int size;
char *odir, *newdir;
if (opt.quota && total_downloaded_bytes > opt.quota)
break;
if (f->type != FT_DIRECTORY)
continue;
/* Allocate u->dir off stack, but reallocate only if a larger
string is needed. It's a pity there's no "realloca" for an
item on the bottom of the stack. */
size = strlen (u->dir) + 1 + strlen (f->name) + 1;
if (size > container_size)
container = (char *)alloca (size);
newdir = container;
odir = u->dir;
if (*odir == '\0'
|| (*odir == '/' && *(odir + 1) == '\0'))
/* If ODIR is empty or just "/", simply append f->name to
ODIR. (In the former case, to preserve u->dir being
relative; in the latter case, to avoid double slash.) */
sprintf (newdir, "%s%s", odir, f->name);
else
/* Else, use a separator. */
sprintf (newdir, "%s/%s", odir, f->name);
DEBUGP (("Composing new CWD relative to the initial directory.\n"));
DEBUGP ((" odir = '%s'\n f->name = '%s'\n newdir = '%s'\n\n",
odir, f->name, newdir));
if (!accdir (newdir))
{
logprintf (LOG_VERBOSE, _("\
Not descending to %s as it is excluded/not-included.\n"),
quote (newdir));
continue;
}
con->st &= ~DONE_CWD;
odir = xstrdup (u->dir); /* because url_set_dir will free
u->dir. */
url_set_dir (u, newdir);
ftp_retrieve_glob (u, con, GLOB_GETALL);
url_set_dir (u, odir);
xfree (odir);
/* Set the time-stamp? */
}
if (opt.quota && total_downloaded_bytes > opt.quota)
return QUOTEXC;
else
return RETROK;
}
/* Return true if S has a leading '/' or contains '../' */
static bool
has_insecure_name_p (const char *s)
{
if (*s == '/')
return true;
if (strstr (s, "../") != 0)
return true;
return false;
}
/* A near-top-level function to retrieve the files in a directory.
The function calls ftp_get_listing, to get a linked list of files.
Then it weeds out the file names that do not match the pattern.
ftp_retrieve_list is called with this updated list as an argument.
If the argument ACTION is GLOB_GETONE, just download the file (but
first get the listing, so that the time-stamp is heeded); if it's
GLOB_GLOBALL, use globbing; if it's GLOB_GETALL, download the whole
directory. */
static uerr_t
ftp_retrieve_glob (struct url *u, ccon *con, int action)
{
struct fileinfo *f, *start;
uerr_t res;
con->cmd |= LEAVE_PENDING;
res = ftp_get_listing (u, con, &start);
if (res != RETROK)
return res;
/* First: weed out that do not conform the global rules given in
opt.accepts and opt.rejects. */
if (opt.accepts || opt.rejects)
{
f = start;
while (f)
{
if (f->type != FT_DIRECTORY && !acceptable (f->name))
{
logprintf (LOG_VERBOSE, _("Rejecting %s.\n"),
quote (f->name));
f = delelement (f, &start);
}
else
f = f->next;
}
}
/* Remove all files with possible harmful names */
f = start;
while (f)
{
if (has_insecure_name_p (f->name))
{
logprintf (LOG_VERBOSE, _("Rejecting %s.\n"),
quote (f->name));
f = delelement (f, &start);
}
else
f = f->next;
}
/* Now weed out the files that do not match our globbing pattern.
If we are dealing with a globbing pattern, that is. */
if (*u->file)
{
if (action == GLOB_GLOBALL)
{
int (*matcher) (const char *, const char *, int)
= opt.ignore_case ? fnmatch_nocase : fnmatch;
int matchres = 0;
f = start;
while (f)
{
matchres = matcher (u->file, f->name, 0);
if (matchres == -1)
{
logprintf (LOG_NOTQUIET, _("Error matching %s against %s: %s\n"),
u->file, quotearg_style (escape_quoting_style, f->name),
strerror (errno));
break;
}
if (matchres == FNM_NOMATCH)
f = delelement (f, &start); /* delete the element from the list */
else
f = f->next; /* leave the element in the list */
}
if (matchres == -1)
{
freefileinfo (start);
return RETRBADPATTERN;
}
}
else if (action == GLOB_GETONE)
{
#ifdef __VMS
/* 2009-09-09 SMS.
* Odd-ball compiler ("HP C V7.3-009 on OpenVMS Alpha V7.3-2")
* bug causes spurious %CC-E-BADCONDIT complaint with this
* "?:" statement. (Different linkage attributes for strcmp()
* and strcasecmp().) Converting to "if" changes the
* complaint to %CC-W-PTRMISMATCH on "cmp = strcmp;". Adding
* the senseless type cast clears the complaint, and looks
* harmless.
*/
int (*cmp) (const char *, const char *)
= opt.ignore_case ? strcasecmp : (int (*)())strcmp;
#else /* def __VMS */
int (*cmp) (const char *, const char *)
= opt.ignore_case ? strcasecmp : strcmp;
#endif /* def __VMS [else] */
f = start;
while (f)
{
if (0 != cmp(u->file, f->name))
f = delelement (f, &start);
else
f = f->next;
}
}
}
if (start)
{
/* Just get everything. */
ftp_retrieve_list (u, start, con);
}
else
{
if (action == GLOB_GLOBALL)
{
/* No luck. */
/* #### This message SUCKS. We should see what was the
reason that nothing was retrieved. */
logprintf (LOG_VERBOSE, _("No matches on pattern %s.\n"),
quote (u->file));
}
else if (action == GLOB_GETONE) /* GLOB_GETONE or GLOB_GETALL */
{
/* Let's try retrieving it anyway. */
con->st |= ON_YOUR_OWN;
res = ftp_loop_internal (u, NULL, con, NULL);
return res;
}
/* If action == GLOB_GETALL, and the file list is empty, there's
no point in trying to download anything or in complaining about
it. (An empty directory should not cause complaints.)
*/
}
freefileinfo (start);
if (opt.quota && total_downloaded_bytes > opt.quota)
return QUOTEXC;
else
/* #### Should we return `res' here? */
return RETROK;
}
/* The wrapper that calls an appropriate routine according to contents
of URL. Inherently, its capabilities are limited on what can be
encoded into a URL. */
uerr_t
ftp_loop (struct url *u, char **local_file, int *dt, struct url *proxy,
bool recursive, bool glob)
{
ccon con; /* FTP connection */
uerr_t res;
*dt = 0;
xzero (con);
con.csock = -1;
con.st = ON_YOUR_OWN;
con.rs = ST_UNIX;
con.id = NULL;
con.proxy = proxy;
/* If the file name is empty, the user probably wants a directory
index. We'll provide one, properly HTML-ized. Unless
opt.htmlify is 0, of course. :-) */
if (!*u->file && !recursive)
{
struct fileinfo *f;
res = ftp_get_listing (u, &con, &f);
if (res == RETROK)
{
if (opt.htmlify && !opt.spider)
{
char *filename = (opt.output_document
? xstrdup (opt.output_document)
: (con.target ? xstrdup (con.target)
: url_file_name (u, NULL)));
res = ftp_index (filename, u, f);
if (res == FTPOK && opt.verbose)
{
if (!opt.output_document)
{
struct_stat st;
wgint sz;
if (stat (filename, &st) == 0)
sz = st.st_size;
else
sz = -1;
logprintf (LOG_NOTQUIET,
_("Wrote HTML-ized index to %s [%s].\n"),
quote (filename), number_to_static_string (sz));
}
else
logprintf (LOG_NOTQUIET,
_("Wrote HTML-ized index to %s.\n"),
quote (filename));
}
xfree (filename);
}
freefileinfo (f);
}
}
else
{
bool ispattern = false;
if (glob)
{
/* Treat the URL as a pattern if the file name part of the
URL path contains wildcards. (Don't check for u->file
because it is unescaped and therefore doesn't leave users
the option to escape literal '*' as %2A.) */
char *file_part = strrchr (u->path, '/');
if (!file_part)
file_part = u->path;
ispattern = has_wildcards_p (file_part);
}
if (ispattern || recursive || opt.timestamping)
{
/* ftp_retrieve_glob is a catch-all function that gets called
if we need globbing, time-stamping or recursion. Its
third argument is just what we really need. */
res = ftp_retrieve_glob (u, &con,
ispattern ? GLOB_GLOBALL : GLOB_GETONE);
}
else
res = ftp_loop_internal (u, NULL, &con, local_file);
}
if (res == FTPOK)
res = RETROK;
if (res == RETROK)
*dt |= RETROKF;
/* If a connection was left, quench it. */
if (con.csock != -1)
fd_close (con.csock);
xfree_null (con.id);
con.id = NULL;
xfree_null (con.target);
con.target = NULL;
return res;
}
/* Delete an element from the fileinfo linked list. Returns the
address of the next element, or NULL if the list is exhausted. It
can modify the start of the list. */
static struct fileinfo *
delelement (struct fileinfo *f, struct fileinfo **start)
{
struct fileinfo *prev = f->prev;
struct fileinfo *next = f->next;
xfree (f->name);
xfree_null (f->linkto);
xfree (f);
if (next)
next->prev = prev;
if (prev)
prev->next = next;
else
*start = next;
return next;
}
/* Free the fileinfo linked list of files. */
static void
freefileinfo (struct fileinfo *f)
{
while (f)
{
struct fileinfo *next = f->next;
xfree (f->name);
if (f->linkto)
xfree (f->linkto);
xfree (f);
f = next;
}
}
Reference in New Issue View Git Blame Copy Permalink