/* HTTP support.
Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003,
- 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
+ 2004, 2005, 2006, 2007, 2008, 2009 Free Software Foundation, Inc.
This file is part of GNU Wget.
If PROXY is non-NULL, the connection will be made to the proxy
server, and u->url will be requested. */
static uerr_t
-gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy)
+gethttp (struct url *u, struct http_stat *hs, int *dt, struct url *proxy,
+ struct iri *iri)
{
struct request *req;
print_server_response (resp, " ");
}
+ /* Check for keep-alive related responses. */
+ if (!inhibit_keep_alive && contlen != -1)
+ {
+ if (resp_header_copy (resp, "Keep-Alive", NULL, 0))
+ keep_alive = true;
+ else if (resp_header_copy (resp, "Connection", hdrval, sizeof (hdrval)))
+ {
+ if (0 == strcasecmp (hdrval, "Keep-Alive"))
+ keep_alive = true;
+ }
+ }
+
+ if (keep_alive)
+ /* The server has promised that it will not close the connection
+ when we're done. This means that we can register it. */
+ register_persistent (conn->host, conn->port, sock, using_ssl);
+
+ if (statcode == HTTP_STATUS_UNAUTHORIZED)
+ {
+ /* Authorization is required. */
+ if (keep_alive && !head_only && skip_short_body (sock, contlen))
+ CLOSE_FINISH (sock);
+ else
+ CLOSE_INVALIDATE (sock);
+ pconn.authorized = false;
+ if (!auth_finished && (user && passwd))
+ {
+ /* IIS sends multiple copies of WWW-Authenticate, one with
+ the value "negotiate", and other(s) with data. Loop over
+ all the occurrences and pick the one we recognize. */
+ int wapos;
+ const char *wabeg, *waend;
+ char *www_authenticate = NULL;
+ for (wapos = 0;
+ (wapos = resp_header_locate (resp, "WWW-Authenticate", wapos,
+ &wabeg, &waend)) != -1;
+ ++wapos)
+ if (known_authentication_scheme_p (wabeg, waend))
+ {
+ BOUNDED_TO_ALLOCA (wabeg, waend, www_authenticate);
+ break;
+ }
+
+ 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 (!basic_auth_finished
+ || !BEGINS_WITH (www_authenticate, "Basic"))
+ {
+ char *pth;
+ pth = url_full_path (u);
+ request_set_header (req, "Authorization",
+ create_authorization_line (www_authenticate,
+ user, passwd,
+ request_method (req),
+ pth,
+ &auth_finished),
+ 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);
+ xfree_null (message);
+ resp_free (resp);
+ xfree (head);
+ 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);
+ xfree_null (message);
+ resp_free (resp);
+ xfree (head);
+ return AUTHFAILED;
+ }
+ else /* statcode != HTTP_STATUS_UNAUTHORIZED */
+ {
+ /* Kludge: if NTLM is used, mark the TCP connection as authorized. */
+ if (ntlm_seen)
+ pconn.authorized = true;
+ }
+
/* Determine the local filename if needed. Notice that if -O is used
* hstat.local_file is set by http_loop to the argument of -O. */
if (!hs->local_file)
hs->local_file = url_file_name (u);
}
}
-
+
/* TODO: perform this check only once. */
if (!hs->existence_checked && file_exists_p (hs->local_file))
{
if (has_html_suffix_p (hs->local_file))
*dt |= TEXTHTML;
+ xfree (head);
+ xfree_null (message);
return RETRUNNEEDED;
}
else if (!ALLOW_CLOBBER)
local_dot_orig_file_exists = true;
local_filename = filename_plus_orig_suffix;
}
- }
+ }
if (!local_dot_orig_file_exists)
/* Couldn't stat() <file>.orig, so try to stat() <file>. */
contlen = parsed;
}
- /* Check for keep-alive related responses. */
- if (!inhibit_keep_alive && contlen != -1)
- {
- if (resp_header_copy (resp, "Keep-Alive", NULL, 0))
- keep_alive = true;
- else if (resp_header_copy (resp, "Connection", hdrval, sizeof (hdrval)))
- {
- if (0 == strcasecmp (hdrval, "Keep-Alive"))
- keep_alive = true;
- }
- }
- if (keep_alive)
- /* The server has promised that it will not close the connection
- when we're done. This means that we can register it. */
- register_persistent (conn->host, conn->port, sock, using_ssl);
-
- if (statcode == HTTP_STATUS_UNAUTHORIZED)
- {
- /* Authorization is required. */
- if (keep_alive && !head_only && skip_short_body (sock, contlen))
- CLOSE_FINISH (sock);
- else
- CLOSE_INVALIDATE (sock);
- pconn.authorized = false;
- if (!auth_finished && (user && passwd))
- {
- /* IIS sends multiple copies of WWW-Authenticate, one with
- the value "negotiate", and other(s) with data. Loop over
- all the occurrences and pick the one we recognize. */
- int wapos;
- const char *wabeg, *waend;
- char *www_authenticate = NULL;
- for (wapos = 0;
- (wapos = resp_header_locate (resp, "WWW-Authenticate", wapos,
- &wabeg, &waend)) != -1;
- ++wapos)
- if (known_authentication_scheme_p (wabeg, waend))
- {
- BOUNDED_TO_ALLOCA (wabeg, waend, www_authenticate);
- break;
- }
-
- 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 (!basic_auth_finished
- || !BEGINS_WITH (www_authenticate, "Basic"))
- {
- char *pth;
- pth = url_full_path (u);
- request_set_header (req, "Authorization",
- create_authorization_line (www_authenticate,
- user, passwd,
- request_method (req),
- pth,
- &auth_finished),
- 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);
- return AUTHFAILED;
- }
- else /* statcode != HTTP_STATUS_UNAUTHORIZED */
- {
- /* Kludge: if NTLM is used, mark the TCP connection as authorized. */
- if (ntlm_seen)
- pconn.authorized = true;
- }
request_free (req);
hs->statcode = statcode;
char *tmp = strchr (type, ';');
if (tmp)
{
+ /* 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)
+ {
+ tmp = parse_charset (tmp2);
+ if (tmp)
+ set_content_encoding (iri, tmp);
+ }
}
}
hs->newloc = resp_header_strdup (resp, "Location");
else
CLOSE_INVALIDATE (sock);
xfree_null (type);
+ xfree (head);
return NEWLOCATION;
}
}
}
}
- if (statcode == HTTP_STATUS_RANGE_NOT_SATISFIABLE)
+ if (statcode == HTTP_STATUS_RANGE_NOT_SATISFIABLE
+ || (hs->restval > 0 && statcode == HTTP_STATUS_OK
+ && contrange == 0 && hs->restval >= contlen)
+ )
{
/* If `-c' is in use and the file has been fully downloaded (or
the remote file has shrunk), Wget effectively requests bytes
- after the end of file and the server response with 416. */
+ after the end of file and the server response with 416
+ (or 200 with a <= Content-Length. */
logputs (LOG_VERBOSE, _("\
\n The file is already fully retrieved; nothing to do.\n\n"));
/* In case the caller inspects. */
xfree_null (type);
CLOSE_INVALIDATE (sock); /* would be CLOSE_FINISH, but there
might be more bytes in the body. */
+ xfree (head);
return RETRUNNEEDED;
}
if ((contrange != 0 && contrange != hs->restval)
Bail out. */
xfree_null (type);
CLOSE_INVALIDATE (sock);
+ xfree (head);
return RANGEERR;
}
if (contlen == -1)
CLOSE_FINISH (sock);
else
CLOSE_INVALIDATE (sock);
+ xfree (head);
return RETRFINISHED;
}
_("%s has sprung into existence.\n"),
hs->local_file);
CLOSE_INVALIDATE (sock);
+ xfree (head);
return FOPEN_EXCL_ERR;
}
}
{
logprintf (LOG_NOTQUIET, "%s: %s\n", hs->local_file, strerror (errno));
CLOSE_INVALIDATE (sock);
+ xfree (head);
return FOPENERR;
}
}
retried, and retried, and retried, and... */
uerr_t
http_loop (struct url *u, char **newloc, char **local_file, const char *referer,
- int *dt, struct url *proxy)
+ int *dt, struct url *proxy, struct iri *iri)
{
int count;
bool got_head = false; /* used for time-stamping and filename detection */
uerr_t err, ret = TRYLIMEXC;
time_t tmr = -1; /* remote time-stamp */
struct http_stat hstat; /* HTTP status */
- struct_stat st;
+ struct_stat st;
bool send_head_first = true;
+ char *file_name;
/* Assert that no value for *LOCAL_FILE was passed. */
assert (local_file == NULL || *local_file == NULL);
-
+
/* Set LOCAL_FILE parameter. */
if (local_file && opt.output_document)
*local_file = HYPHENP (opt.output_document) ? NULL : xstrdup (opt.output_document);
-
+
/* Reset NEWLOC parameter. */
*newloc = NULL;
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"),
+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;
/* Reset the counter. */
count = 0;
-
+
/* 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
/* Send preliminary HEAD request if -N is given and we have an existing
* destination file. */
- if (opt.timestamping
+ file_name = url_file_name (u);
+ if (opt.timestamping
&& !opt.content_disposition
- && file_exists_p (url_file_name (u)))
+ && file_exists_p (file_name))
send_head_first = true;
+ xfree (file_name);
/* THE loop */
do
/* Increment the pass counter. */
++count;
sleep_between_retrievals (count);
-
+
/* Get the current time string. */
tms = datetime_str (time (NULL));
-
+
if (opt.spider && !got_head)
logprintf (LOG_VERBOSE, _("\
Spider mode enabled. Check if remote file exists.\n"));
if (opt.verbose)
{
char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
-
- if (count > 1)
+
+ if (count > 1)
{
char tmp[256];
sprintf (tmp, _("(try:%2d)"), count);
logprintf (LOG_NOTQUIET, "--%s-- %s %s\n",
tms, tmp, hurl);
}
- else
+ else
{
logprintf (LOG_NOTQUIET, "--%s-- %s\n",
tms, hurl);
}
-
+
#ifdef WINDOWS
ws_changetitle (hurl);
#endif
/* Default document type is empty. However, if spider mode is
on or time-stamping is employed, HEAD_ONLY commands is
encoded within *dt. */
- if (send_head_first && !got_head)
+ if (send_head_first && !got_head)
*dt |= HEAD_ONLY;
else
*dt &= ~HEAD_ONLY;
*dt &= ~SEND_NOCACHE;
/* Try fetching the document, or at least its head. */
- err = gethttp (u, &hstat, dt, proxy);
+ err = gethttp (u, &hstat, dt, proxy, iri);
/* Time? */
tms = datetime_str (time (NULL));
-
+
/* Get the new location (with or without the redirection). */
if (hstat.newloc)
*newloc = xstrdup (hstat.newloc);
hstat.statcode);
ret = WRONGCODE;
}
- else
+ else
{
ret = NEWLOCATION;
}
/* All possibilities should have been exhausted. */
abort ();
}
-
+
if (!(*dt & RETROKF))
{
char *hurl = NULL;
continue;
}
/* Maybe we should always keep track of broken links, not just in
- * spider mode. */
- else if (opt.spider)
+ * spider mode.
+ * Don't log error if it was UTF-8 encoded because we will try
+ * once unencoded. */
+ else if (opt.spider && !iri->utf8_encode)
{
/* #### Again: ugly ugly ugly! */
- if (!hurl)
+ if (!hurl)
hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
nonexisting_url (hurl);
logprintf (LOG_NOTQUIET, _("\
else
{
logprintf (LOG_NOTQUIET, _("%s ERROR %d: %s.\n"),
- tms, hstat.statcode,
+ tms, hstat.statcode,
quotearg_style (escape_quoting_style, hstat.error));
}
logputs (LOG_VERBOSE, "\n");
&& ((hstat.len == hstat.contlen) ||
((hstat.res == 0) && (hstat.contlen == -1))))
{
- /* #### This code repeats in http.c and ftp.c. Move it to a
- function! */
const char *fl = NULL;
- if (opt.output_document)
- {
- if (output_stream_regular)
- fl = opt.output_document;
- }
- else
- fl = hstat.local_file;
+ set_local_file (&fl, hstat.local_file);
if (fl)
{
time_t newtmr = -1;
{
if (*dt & RETROKF)
{
+ bool write_to_stdout = (opt.output_document && HYPHENP (opt.output_document));
+
logprintf (LOG_VERBOSE,
- _("%s (%s) - %s saved [%s/%s]\n\n"),
- tms, tmrate, quote (hstat.local_file),
+ write_to_stdout
+ ? _("%s (%s) - written to stdout %s[%s/%s]\n\n")
+ : _("%s (%s) - %s saved [%s/%s]\n\n"),
+ tms, tmrate,
+ write_to_stdout ? "" : quote (hstat.local_file),
number_to_static_string (hstat.len),
number_to_static_string (hstat.contlen));
logprintf (LOG_NONVERBOSE,
{
if (*dt & RETROKF)
{
+ bool write_to_stdout = (opt.output_document && HYPHENP (opt.output_document));
+
logprintf (LOG_VERBOSE,
- _("%s (%s) - %s saved [%s]\n\n"),
- tms, tmrate, quote (hstat.local_file),
+ 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 (hstat.local_file),
number_to_static_string (hstat.len));
logprintf (LOG_NONVERBOSE,
"%s URL:%s [%s] -> \"%s\" [%d]\n",
Netscape cookie specification.) */
};
const char *oldlocale;
+ char savedlocale[256];
size_t i;
time_t ret = (time_t) -1;
non-English locales, which we work around by temporarily setting
locale to C before invoking strptime. */
oldlocale = setlocale (LC_TIME, NULL);
+ if (oldlocale)
+ {
+ size_t l = strlen (oldlocale);
+ if (l >= sizeof savedlocale)
+ savedlocale[0] = '\0';
+ else
+ memcpy (savedlocale, oldlocale, l);
+ }
+ else savedlocale[0] = '\0';
+
setlocale (LC_TIME, "C");
for (i = 0; i < countof (time_formats); i++)
}
/* Restore the previous locale. */
- setlocale (LC_TIME, oldlocale);
+ if (savedlocale[0])
+ setlocale (LC_TIME, savedlocale);
return ret;
}
projects / wget / blobdiff