X-Git-Url: http://sjero.net/git/?a=blobdiff_plain;f=src%2Fhttp.c;h=a4571ad7019b99180a1b44ffa1ffbd1d055e6d38;hb=d3007f1b3a5d033babe40bc4c56a899eb3b10bfa;hp=4813460f61baf36527a78083ddf36ea6e0c120eb;hpb=78aefb3bbf3afb0a3677419370ee197db389bfb7;p=wget
diff --git a/src/http.c b/src/http.c
index 4813460f..a4571ad7 100644
--- a/src/http.c
+++ b/src/http.c
@@ -1,5 +1,6 @@
/* HTTP support.
- Copyright (C) 1996-2006 Free Software Foundation, Inc.
+ Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003,
+ 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Wget.
@@ -16,17 +17,18 @@ 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 .
-In addition, as a special exception, the Free Software Foundation
-gives permission to link the code of its release of Wget with the
-OpenSSL project's "OpenSSL" library (or with modified versions of it
-that use the same license as the "OpenSSL" library), and distribute
-the linked executables. You must obey the GNU General Public License
-in all respects for all of the code used other than "OpenSSL". If you
-modify this file, you may extend this exception to your version of the
-file, but you are not obligated to do so. If you do not wish to do
-so, delete this exception statement from your version. */
+Additional permission under GNU GPL version 3 section 7
-#include
+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
#include
@@ -39,7 +41,7 @@ so, delete this exception statement from your version. */
#include
#include
-#include "wget.h"
+#include "hash.h"
#include "http.h"
#include "utils.h"
#include "url.h"
@@ -47,6 +49,7 @@ so, delete this exception statement from your version. */
#include "retr.h"
#include "connect.h"
#include "netrc.h"
+#include "iri.h"
#ifdef HAVE_SSL
# include "ssl.h"
#endif
@@ -66,6 +69,14 @@ so, delete this exception statement from your version. */
extern char *version_string;
+/* Forward decls. */
+static char *create_authorization_line (const char *, const char *,
+ const char *, const char *,
+ const char *, bool *);
+static char *basic_authentication_encode (const char *, const char *);
+static bool known_authentication_scheme_p (const char *, const char *);
+static void load_cookies (void);
+
#ifndef MIN
# define MIN(x, y) ((x) > (y) ? (y) : (x))
#endif
@@ -269,7 +280,7 @@ request_set_user_header (struct request *req, const char *header)
return;
BOUNDED_TO_ALLOCA (header, p, name);
++p;
- while (ISSPACE (*p))
+ while (c_isspace (*p))
++p;
request_set_header (req, xstrdup (name), (char *) p, rel_name);
}
@@ -373,6 +384,58 @@ request_free (struct request *req)
xfree (req);
}
+static struct hash_table *basic_authed_hosts;
+
+/* Find out if this host has issued a Basic challenge yet; if so, give
+ * it the username, password. A temporary measure until we can get
+ * proper authentication in place. */
+
+static bool
+maybe_send_basic_creds (const char *hostname, const char *user,
+ const char *passwd, struct request *req)
+{
+ bool do_challenge = false;
+
+ if (opt.auth_without_challenge)
+ {
+ DEBUGP(("Auth-without-challenge set, sending Basic credentials.\n"));
+ do_challenge = true;
+ }
+ else if (basic_authed_hosts
+ && hash_table_contains(basic_authed_hosts, hostname))
+ {
+ DEBUGP(("Found %s in basic_authed_hosts.\n", quote (hostname)));
+ do_challenge = true;
+ }
+ else
+ {
+ DEBUGP(("Host %s has not issued a general basic challenge.\n",
+ quote (hostname)));
+ }
+ if (do_challenge)
+ {
+ request_set_header (req, "Authorization",
+ basic_authentication_encode (user, passwd),
+ rel_value);
+ }
+ return do_challenge;
+}
+
+static void
+register_basic_auth_host (const char *hostname)
+{
+ if (!basic_authed_hosts)
+ {
+ basic_authed_hosts = make_nocase_string_hash_table (1);
+ }
+ if (!hash_table_contains(basic_authed_hosts, hostname))
+ {
+ hash_table_put (basic_authed_hosts, xstrdup(hostname), NULL);
+ DEBUGP(("Inserted %s into basic_authed_hosts\n", quote (hostname)));
+ }
+}
+
+
/* Send the contents of FILE_NAME to SOCK. Make sure that exactly
PROMISED_SIZE bytes are sent over the wire -- if the file is
longer, read only that much; if the file is shorter, report an error. */
@@ -599,9 +662,9 @@ resp_header_locate (const struct response *resp, const char *name, int start,
&& 0 == strncasecmp (b, name, name_len))
{
b += name_len + 1;
- while (b < e && ISSPACE (*b))
+ while (b < e && c_isspace (*b))
++b;
- while (b < e && ISSPACE (e[-1]))
+ while (b < e && c_isspace (e[-1]))
--e;
*begptr = b;
*endptr = e;
@@ -700,17 +763,17 @@ resp_status (const struct response *resp, char **message)
if (p < end && *p == '/')
{
++p;
- while (p < end && ISDIGIT (*p))
+ while (p < end && c_isdigit (*p))
++p;
if (p < end && *p == '.')
++p;
- while (p < end && ISDIGIT (*p))
+ while (p < end && c_isdigit (*p))
++p;
}
- while (p < end && ISSPACE (*p))
+ while (p < end && c_isspace (*p))
++p;
- if (end - p < 3 || !ISDIGIT (p[0]) || !ISDIGIT (p[1]) || !ISDIGIT (p[2]))
+ if (end - p < 3 || !c_isdigit (p[0]) || !c_isdigit (p[1]) || !c_isdigit (p[2]))
return -1;
status = 100 * (p[0] - '0') + 10 * (p[1] - '0') + (p[2] - '0');
@@ -718,9 +781,9 @@ resp_status (const struct response *resp, char **message)
if (message)
{
- while (p < end && ISSPACE (*p))
+ while (p < end && c_isspace (*p))
++p;
- while (p < end && ISSPACE (end[-1]))
+ while (p < end && c_isspace (end[-1]))
--end;
*message = strdupdelim (p, end);
}
@@ -748,7 +811,8 @@ print_response_line(const char *prefix, const char *b, const char *e)
{
char *copy;
BOUNDED_TO_ALLOCA(b, e, copy);
- logprintf (LOG_VERBOSE, "%s%s\n", prefix, escnonprint(copy));
+ logprintf (LOG_ALWAYS, "%s%s\n", prefix,
+ quotearg_style (escape_quoting_style, copy));
}
/* Print the server response, line by line, omitting the trailing CRLF
@@ -791,27 +855,30 @@ parse_content_range (const char *hdr, wgint *first_byte_ptr,
HTTP spec. */
if (*hdr == ':')
++hdr;
- while (ISSPACE (*hdr))
+ while (c_isspace (*hdr))
++hdr;
if (!*hdr)
return false;
}
- if (!ISDIGIT (*hdr))
+ if (!c_isdigit (*hdr))
return false;
- for (num = 0; ISDIGIT (*hdr); hdr++)
+ for (num = 0; c_isdigit (*hdr); hdr++)
num = 10 * num + (*hdr - '0');
- if (*hdr != '-' || !ISDIGIT (*(hdr + 1)))
+ if (*hdr != '-' || !c_isdigit (*(hdr + 1)))
return false;
*first_byte_ptr = num;
++hdr;
- for (num = 0; ISDIGIT (*hdr); hdr++)
+ for (num = 0; c_isdigit (*hdr); hdr++)
num = 10 * num + (*hdr - '0');
- if (*hdr != '/' || !ISDIGIT (*(hdr + 1)))
+ if (*hdr != '/' || !c_isdigit (*(hdr + 1)))
return false;
*last_byte_ptr = num;
++hdr;
- for (num = 0; ISDIGIT (*hdr); hdr++)
- num = 10 * num + (*hdr - '0');
+ if (*hdr == '*')
+ num = -1;
+ else
+ for (num = 0; c_isdigit (*hdr); hdr++)
+ num = 10 * num + (*hdr - '0');
*entity_length_ptr = num;
return true;
}
@@ -885,7 +952,7 @@ extract_param (const char **source, param_token *name, param_token *value,
{
const char *p = *source;
- while (ISSPACE (*p)) ++p;
+ while (c_isspace (*p)) ++p;
if (!*p)
{
*source = p;
@@ -894,11 +961,11 @@ extract_param (const char **source, param_token *name, param_token *value,
/* Extract name. */
name->b = p;
- while (*p && !ISSPACE (*p) && *p != '=' && *p != separator) ++p;
+ while (*p && !c_isspace (*p) && *p != '=' && *p != separator) ++p;
name->e = p;
if (name->b == name->e)
return false; /* empty name: error */
- while (ISSPACE (*p)) ++p;
+ while (c_isspace (*p)) ++p;
if (*p == separator || !*p) /* no value */
{
xzero (*value);
@@ -911,7 +978,7 @@ extract_param (const char **source, param_token *name, param_token *value,
/* *p is '=', extract value */
++p;
- while (ISSPACE (*p)) ++p;
+ while (c_isspace (*p)) ++p;
if (*p == '"') /* quoted */
{
value->b = ++p;
@@ -920,7 +987,7 @@ extract_param (const char **source, param_token *name, param_token *value,
return false;
value->e = p++;
/* Currently at closing quote; find the end of param. */
- while (ISSPACE (*p)) ++p;
+ while (c_isspace (*p)) ++p;
while (*p && *p != separator) ++p;
if (*p == separator)
++p;
@@ -933,7 +1000,7 @@ extract_param (const char **source, param_token *name, param_token *value,
value->b = p;
while (*p && *p != separator) ++p;
value->e = p;
- while (value->e != value->b && ISSPACE (value->e[-1]))
+ while (value->e != value->b && c_isspace (value->e[-1]))
--value->e;
if (*p == separator) ++p;
}
@@ -1230,10 +1297,15 @@ struct http_stat
char *remote_time; /* remote time-stamp string */
char *error; /* textual HTTP error */
int statcode; /* status code */
+ char *message; /* status message */
wgint rd_size; /* amount of data read from socket */
double dltime; /* time it took to download the data */
const char *referer; /* value of the referer header. */
char *local_file; /* local file name. */
+ bool existence_checked; /* true if we already checked for a file's
+ existence after having begun to download
+ (needed in gethttp for when connection is
+ interrupted/restarted. */
bool timestamp_checked; /* true if pre-download time-stamping checks
* have already been performed */
char *orig_file_name; /* name of file to compare for time-stamping
@@ -1252,6 +1324,7 @@ free_hstat (struct http_stat *hs)
xfree_null (hs->rderrmsg);
xfree_null (hs->local_file);
xfree_null (hs->orig_file_name);
+ xfree_null (hs->message);
/* Guard against being called twice. */
hs->newloc = NULL;
@@ -1259,16 +1332,9 @@ free_hstat (struct http_stat *hs)
hs->error = NULL;
}
-static char *create_authorization_line (const char *, const char *,
- const char *, const char *,
- const char *, bool *);
-static char *basic_authentication_encode (const char *, const char *);
-static bool known_authentication_scheme_p (const char *, const char *);
-static void load_cookies (void);
-
#define BEGINS_WITH(line, string_constant) \
(!strncasecmp (line, string_constant, sizeof (string_constant) - 1) \
- && (ISSPACE (line[sizeof (string_constant) - 1]) \
+ && (c_isspace (line[sizeof (string_constant) - 1]) \
|| !line[sizeof (string_constant) - 1]))
#define SET_USER_AGENT(req) do { \
@@ -1312,10 +1378,15 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
int sock = -1;
int flags;
- /* Set to 1 when the authorization has failed permanently and should
+ /* Set to 1 when the authorization has already been sent and should
not be tried again. */
bool auth_finished = false;
+ /* Set to 1 when just globally-set Basic authorization has been sent;
+ * should prevent further Basic negotiations, but not other
+ * mechanisms. */
+ bool basic_auth_finished = false;
+
/* Whether NTLM authentication is used for this request. */
bool ntlm_seen = false;
@@ -1373,6 +1444,7 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
hs->newloc = NULL;
hs->remote_time = NULL;
hs->error = NULL;
+ hs->message = NULL;
conn = u;
@@ -1421,66 +1493,13 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
user = user ? user : (opt.http_user ? opt.http_user : opt.user);
passwd = passwd ? passwd : (opt.http_passwd ? opt.http_passwd : opt.passwd);
- if (user && passwd)
+ if (user && passwd
+ && !u->user) /* We only do "site-wide" authentication with "global"
+ user/password values; URL user/password info overrides. */
{
- /* We have the username and the password, but haven't tried
- any authorization yet. Let's see if the "Basic" method
- works. If not, we'll come back here and construct a
- proper authorization method with the right challenges.
-
- If we didn't employ this kind of logic, every URL that
- requires authorization would have to be processed twice,
- which is very suboptimal and generates a bunch of false
- "unauthorized" errors in the server log.
-
- #### But this logic also has a serious problem when used
- with stronger authentications: we *first* transmit the
- username and the password in clear text, and *then* attempt a
- stronger authentication scheme. That cannot be right! We
- are only fortunate that almost everyone still uses the
- `Basic' scheme anyway.
-
- There should be an option to prevent this from happening, for
- those who use strong authentication schemes and value their
- passwords. */
- request_set_header (req, "Authorization",
- basic_authentication_encode (user, passwd),
- rel_value);
- }
-
- proxyauth = NULL;
- if (proxy)
- {
- char *proxy_user, *proxy_passwd;
- /* For normal username and password, URL components override
- command-line/wgetrc parameters. With proxy
- authentication, it's the reverse, because proxy URLs are
- normally the "permanent" ones, so command-line args
- should take precedence. */
- if (opt.proxy_user && opt.proxy_passwd)
- {
- proxy_user = opt.proxy_user;
- proxy_passwd = opt.proxy_passwd;
- }
- else
- {
- proxy_user = proxy->user;
- proxy_passwd = proxy->passwd;
- }
- /* #### This does not appear right. Can't the proxy request,
- say, `Digest' authentication? */
- if (proxy_user && proxy_passwd)
- proxyauth = basic_authentication_encode (proxy_user, proxy_passwd);
-
- /* If we're using a proxy, we will be connecting to the proxy
- server. */
- conn = proxy;
-
- /* Proxy authorization over SSL is handled below. */
-#ifdef HAVE_SSL
- if (u->scheme != SCHEME_HTTPS)
-#endif
- request_set_header (req, "Proxy-Authorization", proxyauth, rel_value);
+ /* If this is a host for which we've already received a Basic
+ * challenge, we'll go ahead and send Basic authentication creds. */
+ basic_auth_finished = maybe_send_basic_creds(u->host, user, passwd, req);
}
/* Generate the Host header, HOST:PORT. Take into account that:
@@ -1530,8 +1549,8 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
post_data_size = file_size (opt.post_file_name);
if (post_data_size == -1)
{
- logprintf (LOG_NOTQUIET, _("POST data file `%s' missing: %s\n"),
- opt.post_file_name, strerror (errno));
+ logprintf (LOG_NOTQUIET, _("POST data file %s missing: %s\n"),
+ quote (opt.post_file_name), strerror (errno));
post_data_size = 0;
}
}
@@ -1553,6 +1572,41 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
without authorization header fails. (Expected to happen at least
for the Digest authorization scheme.) */
+ proxyauth = NULL;
+ if (proxy)
+ {
+ char *proxy_user, *proxy_passwd;
+ /* For normal username and password, URL components override
+ command-line/wgetrc parameters. With proxy
+ authentication, it's the reverse, because proxy URLs are
+ normally the "permanent" ones, so command-line args
+ should take precedence. */
+ if (opt.proxy_user && opt.proxy_passwd)
+ {
+ proxy_user = opt.proxy_user;
+ proxy_passwd = opt.proxy_passwd;
+ }
+ else
+ {
+ proxy_user = proxy->user;
+ proxy_passwd = proxy->passwd;
+ }
+ /* #### This does not appear right. Can't the proxy request,
+ say, `Digest' authentication? */
+ if (proxy_user && proxy_passwd)
+ proxyauth = basic_authentication_encode (proxy_user, proxy_passwd);
+
+ /* If we're using a proxy, we will be connecting to the proxy
+ server. */
+ conn = proxy;
+
+ /* Proxy authorization over SSL is handled below. */
+#ifdef HAVE_SSL
+ if (u->scheme != SCHEME_HTTPS)
+#endif
+ request_set_header (req, "Proxy-Authorization", proxyauth, rel_value);
+ }
+
keep_alive = false;
/* Establish the connection. */
@@ -1580,7 +1634,8 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
sock = pconn.socket;
using_ssl = pconn.ssl;
logprintf (LOG_VERBOSE, _("Reusing existing connection to %s:%d.\n"),
- escnonprint (pconn.host), pconn.port);
+ quotearg_style (escape_quoting_style, pconn.host),
+ pconn.port);
DEBUGP (("Reusing fd %d.\n", sock));
if (pconn.authorized)
/* If the connection is already authorized, the "Basic"
@@ -1588,19 +1643,18 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
only hurts us. */
request_remove_header (req, "Authorization");
}
- }
-
- if (sock < 0)
- {
- /* In its current implementation, persistent_available_p will
- look up conn->host in some cases. If that lookup failed, we
- don't need to bother with connect_to_host. */
- if (host_lookup_failed)
+ else if (host_lookup_failed)
{
request_free (req);
+ logprintf(LOG_NOTQUIET,
+ _("%s: unable to resolve host address %s\n"),
+ exec_name, quote (relevant->host));
return HOSTERR;
}
+ }
+ if (sock < 0)
+ {
sock = connect_to_host (conn->host, conn->port);
if (sock == E_HOST)
{
@@ -1663,13 +1717,14 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
resp = resp_new (head);
statcode = resp_status (resp, &message);
+ hs->message = xstrdup (message);
resp_free (resp);
xfree (head);
if (statcode != 200)
{
failed_tunnel:
logprintf (LOG_NOTQUIET, _("Proxy tunneling failed: %s"),
- message ? escnonprint (message) : "?");
+ message ? quotearg_style (escape_quoting_style, message) : "?");
xfree_null (message);
return CONSSLERR;
}
@@ -1745,9 +1800,10 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
/* Check for status line. */
message = NULL;
statcode = resp_status (resp, &message);
+ hs->message = xstrdup (message);
if (!opt.server_response)
logprintf (LOG_VERBOSE, "%2d %s\n", statcode,
- message ? escnonprint (message) : "");
+ message ? quotearg_style (escape_quoting_style, message) : "");
else
{
logprintf (LOG_VERBOSE, "\n");
@@ -1771,14 +1827,15 @@ gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
}
/* TODO: perform this check only once. */
- if (file_exists_p (hs->local_file))
+ if (!hs->existence_checked && file_exists_p (hs->local_file))
{
- if (opt.noclobber)
+ if (opt.noclobber && !opt.output_document)
{
/* If opt.noclobber is turned on and file already exists, do not
- retrieve the file */
+ retrieve the file. But if the output_document was given, then this
+ test was already done and the file didn't exist. Hence the !opt.output_document */
logprintf (LOG_VERBOSE, _("\
-File `%s' already there; not retrieving.\n\n"), hs->local_file);
+File %s already there; not retrieving.\n\n"), quote (hs->local_file));
/* If the file is there, we suppose it's retrieved OK. */
*dt |= RETROKF;
@@ -1787,7 +1844,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
if (has_html_suffix_p (hs->local_file))
*dt |= TEXTHTML;
- return RETROK;
+ return RETRUNNEEDED;
}
else if (!ALLOW_CLOBBER)
{
@@ -1797,6 +1854,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
hs->local_file = unique;
}
}
+ hs->existence_checked = true;
/* Support timestamping */
/* TODO: move this code out of gethttp. */
@@ -1866,12 +1924,20 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
errno = 0;
parsed = str_to_wgint (hdrval, NULL, 10);
if (parsed == WGINT_MAX && errno == ERANGE)
- /* Out of range.
- #### If Content-Length is out of range, it most likely
- means that the file is larger than 2G and that we're
- compiled without LFS. In that case we should probably
- refuse to even attempt to download the file. */
- contlen = -1;
+ {
+ /* Out of range.
+ #### If Content-Length is out of range, it most likely
+ means that the file is larger than 2G and that we're
+ compiled without LFS. In that case we should probably
+ refuse to even attempt to download the file. */
+ contlen = -1;
+ }
+ else if (parsed < 0)
+ {
+ /* Negative Content-Length; nonsensical, so we can't
+ assume any information about the content to receive. */
+ contlen = -1;
+ }
else
contlen = parsed;
}
@@ -1919,16 +1985,13 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
}
if (!www_authenticate)
- /* If the authentication header is missing or
- unrecognized, there's no sense in retrying. */
- logputs (LOG_NOTQUIET, _("Unknown authentication scheme.\n"));
- else if (BEGINS_WITH (www_authenticate, "Basic"))
- /* If the authentication scheme is "Basic", which we send
- by default, there's no sense in retrying either. (This
- should be changed when we stop sending "Basic" data by
- default.) */
- ;
- else
+ {
+ /* If the authentication header is missing or
+ unrecognized, there's no sense in retrying. */
+ logputs (LOG_NOTQUIET, _("Unknown authentication scheme.\n"));
+ }
+ else if (!basic_auth_finished
+ || !BEGINS_WITH (www_authenticate, "Basic"))
{
char *pth;
pth = url_full_path (u);
@@ -1941,9 +2004,20 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
rel_value);
if (BEGINS_WITH (www_authenticate, "NTLM"))
ntlm_seen = true;
+ else if (!u->user && BEGINS_WITH (www_authenticate, "Basic"))
+ {
+ /* Need to register this host as using basic auth,
+ * so we automatically send creds next time. */
+ register_basic_auth_host (u->host);
+ }
xfree (pth);
goto retry_with_auth;
}
+ else
+ {
+ /* We already did Basic auth, and it failed. Gotta
+ * give up. */
+ }
}
logputs (LOG_NOTQUIET, _("Authorization failed.\n"));
request_free (req);
@@ -1972,9 +2046,16 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
char *tmp = strchr (type, ';');
if (tmp)
{
- while (tmp > type && ISSPACE (tmp[-1]))
+ /* sXXXav: only needed if IRI support is enabled */
+ char *tmp2 = tmp + 1;
+
+ while (tmp > type && c_isspace (tmp[-1]))
--tmp;
*tmp = '\0';
+
+ /* Try to get remote encoding if needed */
+ if (opt.enable_iri && !opt.encoding_remote)
+ /* xxx = */ parse_charset (tmp2);
}
}
hs->newloc = resp_header_strdup (resp, "Location");
@@ -2003,7 +2084,10 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
wgint first_byte_pos, last_byte_pos, entity_length;
if (parse_content_range (hdrval, &first_byte_pos, &last_byte_pos,
&entity_length))
- contrange = first_byte_pos;
+ {
+ contrange = first_byte_pos;
+ contlen = last_byte_pos - first_byte_pos + 1;
+ }
}
resp_free (resp);
@@ -2103,7 +2187,10 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
CLOSE_INVALIDATE (sock);
return RANGEERR;
}
- hs->contlen = contlen + contrange;
+ if (contlen == -1)
+ hs->contlen = -1;
+ else
+ hs->contlen = contlen + contrange;
if (opt.verbose)
{
@@ -2134,7 +2221,7 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
logputs (LOG_VERBOSE,
opt.ignore_length ? _("ignored") : _("unspecified"));
if (type)
- logprintf (LOG_VERBOSE, " [%s]\n", escnonprint (type));
+ logprintf (LOG_VERBOSE, " [%s]\n", quotearg_style (escape_quoting_style, type));
else
logputs (LOG_VERBOSE, "\n");
}
@@ -2203,8 +2290,8 @@ File `%s' already there; not retrieving.\n\n"), hs->local_file);
/* Print fetch message, if opt.verbose. */
if (opt.verbose)
{
- logprintf (LOG_NOTQUIET, _("Saving to: `%s'\n"),
- HYPHENP (hs->local_file) ? "STDOUT" : hs->local_file);
+ logprintf (LOG_NOTQUIET, _("Saving to: %s\n"),
+ HYPHENP (hs->local_file) ? quote ("STDOUT") : quote (hs->local_file));
}
/* This confuses the timestamping code that checks for file size.
@@ -2256,14 +2343,15 @@ http_loop (struct url *u, char **newloc, char **local_file, const char *referer,
{
int count;
bool got_head = false; /* used for time-stamping and filename detection */
+ bool time_came_from_head = false;
bool got_name = false;
char *tms;
const char *tmrate;
uerr_t err, ret = TRYLIMEXC;
time_t tmr = -1; /* remote time-stamp */
- wgint local_size = 0; /* the size of the local file */
struct http_stat hstat; /* HTTP status */
struct_stat st;
+ bool send_head_first = true;
/* Assert that no value for *LOCAL_FILE was passed. */
assert (local_file == NULL || *local_file == NULL);
@@ -2294,6 +2382,33 @@ http_loop (struct url *u, char **newloc, char **local_file, const char *referer,
hstat.local_file = xstrdup (opt.output_document);
got_name = true;
}
+ else if (!opt.content_disposition)
+ {
+ hstat.local_file = url_file_name (u);
+ got_name = true;
+ }
+
+ /* TODO: Ick! This code is now in both gethttp and http_loop, and is
+ * screaming for some refactoring. */
+ if (got_name && file_exists_p (hstat.local_file) && opt.noclobber && !opt.output_document)
+ {
+ /* If opt.noclobber is turned on and file already exists, do not
+ retrieve the file. But if the output_document was given, then this
+ test was already done and the file didn't exist. Hence the !opt.output_document */
+ logprintf (LOG_VERBOSE, _("\
+File %s already there; not retrieving.\n\n"),
+ quote (hstat.local_file));
+ /* If the file is there, we suppose it's retrieved OK. */
+ *dt |= RETROKF;
+
+ /* #### Bogusness alert. */
+ /* If its suffix is "html" or "htm" or similar, assume text/html. */
+ if (has_html_suffix_p (hstat.local_file))
+ *dt |= TEXTHTML;
+
+ ret = RETROK;
+ goto exit;
+ }
/* Reset the counter. */
count = 0;
@@ -2301,6 +2416,19 @@ http_loop (struct url *u, char **newloc, char **local_file, const char *referer,
/* Reset the document type. */
*dt = 0;
+ /* Skip preliminary HEAD request if we're not in spider mode AND
+ * if -O was given or HTTP Content-Disposition support is disabled. */
+ if (!opt.spider
+ && (got_name || !opt.content_disposition))
+ send_head_first = false;
+
+ /* Send preliminary HEAD request if -N is given and we have an existing
+ * destination file. */
+ if (opt.timestamping
+ && !opt.content_disposition
+ && file_exists_p (url_file_name (u)))
+ send_head_first = true;
+
/* THE loop */
do
{
@@ -2309,7 +2437,7 @@ http_loop (struct url *u, char **newloc, char **local_file, const char *referer,
sleep_between_retrievals (count);
/* Get the current time string. */
- tms = time_str (time (NULL));
+ tms = datetime_str (time (NULL));
if (opt.spider && !got_head)
logprintf (LOG_VERBOSE, _("\
@@ -2318,7 +2446,7 @@ Spider mode enabled. Check if remote file exists.\n"));
/* Print fetch message, if opt.verbose. */
if (opt.verbose)
{
- char *hurl = url_string (u, true);
+ char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
if (count > 1)
{
@@ -2342,8 +2470,7 @@ Spider mode enabled. Check if remote file exists.\n"));
/* Default document type is empty. However, if spider mode is
on or time-stamping is employed, HEAD_ONLY commands is
encoded within *dt. */
- if (((opt.spider || opt.timestamping) && !got_head)
- || (opt.always_rest && !got_name))
+ if (send_head_first && !got_head)
*dt |= HEAD_ONLY;
else
*dt &= ~HEAD_ONLY;
@@ -2379,12 +2506,12 @@ Spider mode enabled. Check if remote file exists.\n"));
err = gethttp (u, &hstat, dt, proxy);
/* Time? */
- tms = time_str (time (NULL));
+ tms = datetime_str (time (NULL));
/* Get the new location (with or without the redirection). */
if (hstat.newloc)
*newloc = xstrdup (hstat.newloc);
-
+
switch (err)
{
case HERR: case HEOF: case CONSOCKERR: case CONCLOSED:
@@ -2398,8 +2525,8 @@ Spider mode enabled. Check if remote file exists.\n"));
case FWRITEERR: case FOPENERR:
/* Another fatal error. */
logputs (LOG_VERBOSE, "\n");
- logprintf (LOG_NOTQUIET, _("Cannot write to `%s' (%s).\n"),
- hstat.local_file, strerror (errno));
+ logprintf (LOG_NOTQUIET, _("Cannot write to %s (%s).\n"),
+ quote (hstat.local_file), strerror (errno));
case HOSTERR: case CONIMPOSSIBLE: case PROXERR: case AUTHFAILED:
case SSLINITFAILED: case CONTNOTSUPPORTED:
/* Fatal errors just return from the function. */
@@ -2435,23 +2562,31 @@ Spider mode enabled. Check if remote file exists.\n"));
/* All possibilities should have been exhausted. */
abort ();
}
-
+
if (!(*dt & RETROKF))
{
char *hurl = NULL;
if (!opt.verbose)
{
/* #### Ugly ugly ugly! */
- hurl = url_string (u, true);
+ hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
logprintf (LOG_NONVERBOSE, "%s:\n", hurl);
}
+
+ /* Fall back to GET if HEAD fails with a 500 or 501 error code. */
+ if (*dt & HEAD_ONLY
+ && (hstat.statcode == 500 || hstat.statcode == 501))
+ {
+ got_head = true;
+ continue;
+ }
/* Maybe we should always keep track of broken links, not just in
* spider mode. */
- if (opt.spider)
+ else if (opt.spider)
{
/* #### Again: ugly ugly ugly! */
if (!hurl)
- hurl = url_string (u, true);
+ hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
nonexisting_url (hurl);
logprintf (LOG_NOTQUIET, _("\
Remote file does not exist -- broken link!!!\n"));
@@ -2459,7 +2594,8 @@ Remote file does not exist -- broken link!!!\n"));
else
{
logprintf (LOG_NOTQUIET, _("%s ERROR %d: %s.\n"),
- tms, hstat.statcode, escnonprint (hstat.error));
+ tms, hstat.statcode,
+ quotearg_style (escape_quoting_style, hstat.error));
}
logputs (LOG_VERBOSE, "\n");
ret = WRONGCODE;
@@ -2470,7 +2606,7 @@ Remote file does not exist -- broken link!!!\n"));
/* Did we get the time-stamp? */
if (!got_head)
{
- bool restart_loop = false;
+ got_head = true; /* no more time-stamping */
if (opt.timestamping && !hstat.remote_time)
{
@@ -2484,94 +2620,109 @@ Last-modified header missing -- time-stamps turned off.\n"));
if (tmr == (time_t) (-1))
logputs (LOG_VERBOSE, _("\
Last-modified header invalid -- time-stamp ignored.\n"));
+ if (*dt & HEAD_ONLY)
+ time_came_from_head = true;
}
- /* The time-stamping section. */
- if (opt.timestamping)
+ if (send_head_first)
{
- if (hstat.orig_file_name) /* Perform the following checks only
- if the file we're supposed to
- download already exists. */
+ /* The time-stamping section. */
+ if (opt.timestamping)
{
- if (hstat.remote_time &&
- tmr != (time_t) (-1))
+ if (hstat.orig_file_name) /* Perform the following
+ checks only if the file
+ we're supposed to
+ download already exists. */
{
- /* Now time-stamping can be used validly. Time-stamping
- means that if the sizes of the local and remote file
- match, and local file is newer than the remote file,
- it will not be retrieved. Otherwise, the normal
- download procedure is resumed. */
- if (hstat.orig_file_tstamp >= tmr)
+ if (hstat.remote_time &&
+ tmr != (time_t) (-1))
{
- if (hstat.contlen == -1
- || hstat.orig_file_size == hstat.contlen)
- {
- logprintf (LOG_VERBOSE, _("\
-Server file no newer than local file `%s' -- not retrieving.\n\n"),
- hstat.orig_file_name);
- ret = RETROK;
- goto exit;
- }
- else
+ /* Now time-stamping can be used validly.
+ Time-stamping means that if the sizes of
+ the local and remote file match, and local
+ file is newer than the remote file, it will
+ not be retrieved. Otherwise, the normal
+ download procedure is resumed. */
+ if (hstat.orig_file_tstamp >= tmr)
{
- logprintf (LOG_VERBOSE, _("\
+ if (hstat.contlen == -1
+ || hstat.orig_file_size == hstat.contlen)
+ {
+ logprintf (LOG_VERBOSE, _("\
+Server file no newer than local file %s -- not retrieving.\n\n"),
+ quote (hstat.orig_file_name));
+ ret = RETROK;
+ goto exit;
+ }
+ else
+ {
+ logprintf (LOG_VERBOSE, _("\
The sizes do not match (local %s) -- retrieving.\n"),
- number_to_static_string (local_size));
+ number_to_static_string (hstat.orig_file_size));
+ }
}
- }
- else
- logputs (LOG_VERBOSE,
- _("Remote file is newer, retrieving.\n"));
+ else
+ logputs (LOG_VERBOSE,
+ _("Remote file is newer, retrieving.\n"));
- logputs (LOG_VERBOSE, "\n");
+ logputs (LOG_VERBOSE, "\n");
+ }
}
+
+ /* free_hstat (&hstat); */
+ hstat.timestamp_checked = true;
}
- /* free_hstat (&hstat); */
- hstat.timestamp_checked = true;
- restart_loop = true;
- }
-
- if (opt.always_rest)
- {
- got_name = true;
- restart_loop = true;
- }
-
- if (opt.spider)
- {
- if (opt.recursive)
+ if (opt.spider)
{
- if (*dt & TEXTHTML)
+ bool finished = true;
+ if (opt.recursive)
{
- logputs (LOG_VERBOSE, _("\
+ if (*dt & TEXTHTML)
+ {
+ logputs (LOG_VERBOSE, _("\
Remote file exists and could contain links to other resources -- retrieving.\n\n"));
- restart_loop = true;
+ finished = false;
+ }
+ else
+ {
+ logprintf (LOG_VERBOSE, _("\
+Remote file exists but does not contain any link -- not retrieving.\n\n"));
+ ret = RETROK; /* RETRUNNEEDED is not for caller. */
+ }
}
- else
+ else
{
- logprintf (LOG_VERBOSE, _("\
-Remote file exists but does not contain any link -- not retrieving.\n\n"));
- ret = RETRUNNEEDED;
+ if (*dt & TEXTHTML)
+ {
+ logprintf (LOG_VERBOSE, _("\
+Remote file exists and could contain further links,\n\
+but recursion is disabled -- not retrieving.\n\n"));
+ }
+ else
+ {
+ logprintf (LOG_VERBOSE, _("\
+Remote file exists.\n\n"));
+ }
+ ret = RETROK; /* RETRUNNEEDED is not for caller. */
+ }
+
+ if (finished)
+ {
+ logprintf (LOG_NONVERBOSE,
+ _("%s URL:%s %2d %s\n"),
+ tms, u->url, hstat.statcode,
+ hstat.message ? quotearg_style (escape_quoting_style, hstat.message) : "");
goto exit;
}
}
- else
- {
- logprintf (LOG_VERBOSE, _("\
-Remote file exists but recursion is disabled -- not retrieving.\n\n"));
- ret = RETRUNNEEDED;
- goto exit;
- }
- }
-
- got_head = true; /* no more time-stamping */
- *dt &= ~HEAD_ONLY;
- count = 0; /* the retrieve count for HEAD is reset */
- if (restart_loop)
- continue;
- }
+ got_name = true;
+ *dt &= ~HEAD_ONLY;
+ count = 0; /* the retrieve count for HEAD is reset */
+ continue;
+ } /* send_head_first */
+ } /* !got_head */
if ((tmr != (time_t) (-1))
&& ((hstat.len == hstat.contlen) ||
@@ -2588,7 +2739,18 @@ Remote file exists but recursion is disabled -- not retrieving.\n\n"));
else
fl = hstat.local_file;
if (fl)
- touch (fl, tmr);
+ {
+ time_t newtmr = -1;
+ /* Reparse time header, in case it's changed. */
+ if (time_came_from_head
+ && hstat.remote_time && hstat.remote_time[0])
+ {
+ newtmr = http_atotm (hstat.remote_time);
+ if (newtmr != -1)
+ tmr = newtmr;
+ }
+ touch (fl, tmr);
+ }
}
/* End of time-stamping section. */
@@ -2600,8 +2762,8 @@ Remote file exists but recursion is disabled -- not retrieving.\n\n"));
if (*dt & RETROKF)
{
logprintf (LOG_VERBOSE,
- _("%s (%s) - `%s' saved [%s/%s]\n\n"),
- tms, tmrate, hstat.local_file,
+ _("%s (%s) - %s saved [%s/%s]\n\n"),
+ tms, tmrate, quote (hstat.local_file),
number_to_static_string (hstat.len),
number_to_static_string (hstat.contlen));
logprintf (LOG_NONVERBOSE,
@@ -2631,8 +2793,8 @@ Remote file exists but recursion is disabled -- not retrieving.\n\n"));
if (*dt & RETROKF)
{
logprintf (LOG_VERBOSE,
- _("%s (%s) - `%s' saved [%s]\n\n"),
- tms, tmrate, hstat.local_file,
+ _("%s (%s) - %s saved [%s]\n\n"),
+ tms, tmrate, quote (hstat.local_file),
number_to_static_string (hstat.len));
logprintf (LOG_NONVERBOSE,
"%s URL:%s [%s] -> \"%s\" [%d]\n",
@@ -2660,10 +2822,18 @@ Remote file exists but recursion is disabled -- not retrieving.\n\n"));
printwhat (count, opt.ntry);
continue;
}
- else
+ else if (hstat.len != hstat.restval)
/* Getting here would mean reading more data than
requested with content-length, which we never do. */
abort ();
+ else
+ {
+ /* Getting here probably means that the content-length was
+ * _less_ than the original, local size. We should probably
+ * truncate or re-read, or something. FIXME */
+ ret = RETROK;
+ goto exit;
+ }
}
else /* from now on hstat.res can only be -1 */
{
@@ -2713,11 +2883,11 @@ check_end (const char *p)
{
if (!p)
return false;
- while (ISSPACE (*p))
+ while (c_isspace (*p))
++p;
if (!*p
|| (p[0] == 'G' && p[1] == 'M' && p[2] == 'T')
- || ((p[0] == '+' || p[0] == '-') && ISDIGIT (p[1])))
+ || ((p[0] == '+' || p[0] == '-') && c_isdigit (p[1])))
return true;
else
return false;
@@ -2768,7 +2938,7 @@ http_atotm (const char *time_string)
Netscape cookie specification.) */
};
const char *oldlocale;
- int i;
+ size_t i;
time_t ret = (time_t) -1;
/* Solaris strptime fails to recognize English month names in
@@ -2833,7 +3003,7 @@ basic_authentication_encode (const char *user, const char *passwd)
}
#define SKIP_WS(x) do { \
- while (ISSPACE (*(x))) \
+ while (c_isspace (*(x))) \
++(x); \
} while (0)
@@ -2879,10 +3049,12 @@ digest_authentication_encode (const char *au, const char *user,
au += 6; /* skip over `Digest' */
while (extract_param (&au, &name, &value, ','))
{
- int i;
+ size_t i;
+ size_t namelen = name.e - name.b;
for (i = 0; i < countof (options); i++)
- if (name.e - name.b == strlen (options[i].name)
- && 0 == strncmp (name.b, options[i].name, name.e - name.b))
+ if (namelen == strlen (options[i].name)
+ && 0 == strncmp (name.b, options[i].name,
+ namelen))
{
*options[i].variable = strdupdelim (value.b, value.e);
break;
@@ -2962,10 +3134,11 @@ username=\"%s\", realm=\"%s\", nonce=\"%s\", uri=\"%s\", response=\"%s\"",
first argument and are followed by whitespace or terminating \0.
The comparison is case-insensitive. */
#define STARTS(literal, b, e) \
- ((e) - (b) >= STRSIZE (literal) \
+ ((e > b) \
+ && ((size_t) ((e) - (b))) >= STRSIZE (literal) \
&& 0 == strncasecmp (b, literal, STRSIZE (literal)) \
- && ((e) - (b) == STRSIZE (literal) \
- || ISSPACE (b[STRSIZE (literal)])))
+ && ((size_t) ((e) - (b)) == STRSIZE (literal) \
+ || c_isspace (b[STRSIZE (literal)])))
static bool
known_authentication_scheme_p (const char *hdrbeg, const char *hdrend)
@@ -2994,7 +3167,7 @@ create_authorization_line (const char *au, const char *user,
{
/* We are called only with known schemes, so we can dispatch on the
first letter. */
- switch (TOUPPER (*au))
+ switch (c_toupper (*au))
{
case 'B': /* Basic */
*finished = true;
@@ -3090,6 +3263,6 @@ test_parse_content_disposition()
#endif /* TESTING */
/*
- * vim: et ts=2 sw=2
+ * vim: et sts=2 sw=2 cino+={s
*/