/* 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.
You should have received a copy of the GNU General Public License
along with Wget. If not, see <http://www.gnu.org/licenses/>.
-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 <config.h>
+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 <time.h>
#include <locale.h>
-#include "wget.h"
#include "hash.h"
#include "http.h"
#include "utils.h"
extern char *version_string;
/* Forward decls. */
+struct http_stat;
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 ensure_extension (struct http_stat *, const char *, int *);
static void load_cookies (void);
#ifndef MIN
#define TEXTHTML_S "text/html"
#define TEXTXHTML_S "application/xhtml+xml"
+#define TEXTCSS_S "text/css"
/* Some status code validation macros: */
#define H_20X(x) (((x) >= 200) && ((x) < 300))
int hcount, hcapacity;
};
+extern int numurls;
+
/* Create a new, empty request. At least request_set_method must be
called before the request can be used. */
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);
}
* it the username, password. A temporary measure until we can get
* proper authentication in place. */
-static int
+static bool
maybe_send_basic_creds (const char *hostname, const char *user,
const char *passwd, struct request *req)
{
- int did_challenge = 0;
+ bool do_challenge = false;
- if (basic_authed_hosts
+ 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", hostname));
- request_set_header (req, "Authorization",
- basic_authentication_encode (user, passwd),
- rel_value);
- did_challenge = 1;
+ 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",
- hostname));
+ DEBUGP(("Host %s has not issued a general basic challenge.\n",
+ quote (hostname)));
}
- return did_challenge;
+ if (do_challenge)
+ {
+ request_set_header (req, "Authorization",
+ basic_authentication_encode (user, passwd),
+ rel_value);
+ }
+ return do_challenge;
}
static void
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", hostname));
+ DEBUGP(("Inserted %s into basic_authed_hosts\n", quote (hostname)));
}
}
&& 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;
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');
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);
}
{
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
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;
}
{
const char *p = *source;
- while (ISSPACE (*p)) ++p;
+ while (c_isspace (*p)) ++p;
if (!*p)
{
*source = p;
/* 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);
/* *p is '=', extract value */
++p;
- while (ISSPACE (*p)) ++p;
+ while (c_isspace (*p)) ++p;
if (*p == '"') /* quoted */
{
value->b = ++p;
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;
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;
}
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
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;
#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 { \
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;
hs->newloc = NULL;
hs->remote_time = NULL;
hs->error = NULL;
+ hs->message = NULL;
conn = u;
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
- && !u->user) /* We only do "site-wide" authentication with "global"
- user/password values; URL user/password info overrides. */
+ /* We only do "site-wide" authentication with "global" user/password
+ * values unless --auth-no-challange has been requested; URL user/password
+ * info overrides. */
+ if (user && passwd && (!u->user || opt.auth_without_challenge))
{
/* 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);
}
- 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);
- }
-
/* Generate the Host header, HOST:PORT. Take into account that:
- Broken server-side software often doesn't recognize the PORT
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;
}
}
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. */
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"
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)
{
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;
}
/* 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");
hs->local_file = url_file_name (u);
}
}
-
+
/* 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;
if (has_html_suffix_p (hs->local_file))
*dt |= TEXTHTML;
- return RETROK;
+ return RETRUNNEEDED;
}
else if (!ALLOW_CLOBBER)
{
hs->local_file = unique;
}
}
+ hs->existence_checked = true;
/* Support timestamping */
/* TODO: move this code out of gethttp. */
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>. */
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)
+ {
+ tmp = parse_charset (tmp2);
+ if (tmp)
+ set_content_encoding (iri, tmp);
+ }
}
}
hs->newloc = resp_header_strdup (resp, "Location");
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);
else
*dt &= ~TEXTHTML;
- if (opt.html_extension && (*dt & TEXTHTML))
- /* -E / --html-extension / html_extension = on was specified, and this is a
- text/html file. If some case-insensitive variation on ".htm[l]" isn't
- already the file's suffix, tack on ".html". */
- {
- char *last_period_in_local_filename = strrchr (hs->local_file, '.');
+ if (type &&
+ 0 == strncasecmp (type, TEXTCSS_S, strlen (TEXTCSS_S)))
+ *dt |= TEXTCSS;
+ else
+ *dt &= ~TEXTCSS;
- if (last_period_in_local_filename == NULL
- || !(0 == strcasecmp (last_period_in_local_filename, ".htm")
- || 0 == strcasecmp (last_period_in_local_filename, ".html")))
+ if (opt.html_extension)
+ {
+ if (*dt & TEXTHTML)
+ /* -E / --html-extension / html_extension = on was specified,
+ and this is a text/html file. If some case-insensitive
+ variation on ".htm[l]" isn't already the file's suffix,
+ tack on ".html". */
{
- int local_filename_len = strlen (hs->local_file);
- /* Resize the local file, allowing for ".html" preceded by
- optional ".NUMBER". */
- hs->local_file = xrealloc (hs->local_file,
- local_filename_len + 24 + sizeof (".html"));
- strcpy(hs->local_file + local_filename_len, ".html");
- /* If clobbering is not allowed and the file, as named,
- exists, tack on ".NUMBER.html" instead. */
- if (!ALLOW_CLOBBER && file_exists_p (hs->local_file))
- {
- int ext_num = 1;
- do
- sprintf (hs->local_file + local_filename_len,
- ".%d.html", ext_num++);
- while (file_exists_p (hs->local_file));
- }
- *dt |= ADDED_HTML_EXTENSION;
+ ensure_extension (hs, ".html", dt);
+ }
+ else if (*dt & TEXTCSS)
+ {
+ ensure_extension (hs, ".css", dt);
}
}
- 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. */
CLOSE_INVALIDATE (sock);
return RANGEERR;
}
- hs->contlen = contlen + contrange;
+ if (contlen == -1)
+ hs->contlen = -1;
+ else
+ hs->contlen = contlen + contrange;
if (opt.verbose)
{
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");
}
/* 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.
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 */
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;
+ struct_stat st;
bool send_head_first = true;
/* 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;
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;
-
+
/* 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
+ if (opt.timestamping
&& !opt.content_disposition
&& file_exists_p (url_file_name (u)))
send_head_first = true;
-
+
/* THE loop */
do
{
/* Increment the pass counter. */
++count;
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, _("\
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 = 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);
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. */
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, escnonprint (hstat.error));
+ tms, hstat.statcode,
+ quotearg_style (escape_quoting_style, hstat.error));
}
logputs (LOG_VERBOSE, "\n");
ret = WRONGCODE;
}
/* Did we get the time-stamp? */
- if (send_head_first && !got_head)
+ if (!got_head)
{
- bool restart_loop = false;
+ got_head = true; /* no more time-stamping */
if (opt.timestamping && !hstat.remote_time)
{
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)
+ /* 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, _("\
-Server file no newer than local file `%s' -- not retrieving.\n\n"),
- hstat.orig_file_name);
- ret = RETROK;
- goto exit;
- }
- else
- {
- 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.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;
- }
- }
- if (send_head_first)
- {
got_name = true;
- restart_loop = true;
- }
-
- got_head = true; /* no more time-stamping */
- *dt &= ~HEAD_ONLY;
- count = 0; /* the retrieve count for HEAD is reset */
-
- if (restart_loop)
- continue;
- }
+ *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) ||
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,
number_to_static_string (hstat.contlen),
hstat.local_file, count);
}
- ++opt.numurls;
+ ++numurls;
total_downloaded_bytes += hstat.len;
/* Remember that we downloaded the file for later ".orig" code. */
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",
tms, u->url, number_to_static_string (hstat.len),
hstat.local_file, count);
}
- ++opt.numurls;
+ ++numurls;
total_downloaded_bytes += hstat.len;
/* Remember that we downloaded the file for later ".orig" code. */
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 */
{
{
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;
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
}
#define SKIP_WS(x) do { \
- while (ISSPACE (*(x))) \
+ while (c_isspace (*(x))) \
++(x); \
} while (0)
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;
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)
{
/* 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;
cookie_jar_delete (wget_cookie_jar);
}
+void
+ensure_extension (struct http_stat *hs, const char *ext, int *dt)
+{
+ char *last_period_in_local_filename = strrchr (hs->local_file, '.');
+ char shortext[8];
+ int len = strlen (ext);
+ if (len == 5)
+ {
+ strncpy (shortext, ext, len - 1);
+ shortext[len - 2] = '\0';
+ }
+
+ if (last_period_in_local_filename == NULL
+ || !(0 == strcasecmp (last_period_in_local_filename, shortext)
+ || 0 == strcasecmp (last_period_in_local_filename, ext)))
+ {
+ int local_filename_len = strlen (hs->local_file);
+ /* Resize the local file, allowing for ".html" preceded by
+ optional ".NUMBER". */
+ hs->local_file = xrealloc (hs->local_file,
+ local_filename_len + 24 + len);
+ strcpy (hs->local_file + local_filename_len, ext);
+ /* If clobbering is not allowed and the file, as named,
+ exists, tack on ".NUMBER.html" instead. */
+ if (!ALLOW_CLOBBER && file_exists_p (hs->local_file))
+ {
+ int ext_num = 1;
+ do
+ sprintf (hs->local_file + local_filename_len,
+ ".%d%s", ext_num++, ext);
+ while (file_exists_p (hs->local_file));
+ }
+ *dt |= ADDED_HTML_EXTENSION;
+ }
+}
+
#ifdef TESTING