extern int numurls;
-/* Create a new, empty request. At least request_set_method must be
- called before the request can be used. */
+/* Create a new, empty request. Set the request's method and its
+ arguments. METHOD should be a literal string (or it should outlive
+ the request) because it will not be freed. ARG will be freed by
+ request_free. */
static struct request *
-request_new (void)
+request_new (const char *method, char *arg)
{
struct request *req = xnew0 (struct request);
req->hcapacity = 8;
req->headers = xnew_array (struct request_header, req->hcapacity);
- return req;
-}
-
-/* Set the request's method and its arguments. METH should be a
- literal string (or it should outlive the request) because it will
- not be freed. ARG will be freed by request_free. */
-
-static void
-request_set_method (struct request *req, const char *meth, char *arg)
-{
- req->method = meth;
+ req->method = method;
req->arg = arg;
+ return req;
}
/* Return the method string passed with the last call to
also be written to that file. */
static int
-post_file (int sock, const char *file_name, wgint promised_size, FILE *warc_tmp)
+body_file_send (int sock, const char *file_name, wgint promised_size, FILE *warc_tmp)
{
static char chunk[8192];
wgint written = 0;
int write_error;
FILE *fp;
- DEBUGP (("[writing POST file %s ... ", file_name));
+ DEBUGP (("[writing BODY file %s ... ", file_name));
fp = fopen (file_name, "rb");
if (!fp)
!opt.http_keep_alive || opt.ignore_length;
/* Headers sent when using POST. */
- wgint post_data_size = 0;
+ wgint body_data_size = 0;
bool host_lookup_failed = false;
conn = u;
/* Prepare the request to send. */
-
- req = request_new ();
{
char *meth_arg;
const char *meth = "GET";
if (head_only)
meth = "HEAD";
- else if (opt.post_file_name || opt.post_data)
- meth = "POST";
+ else if (opt.method)
+ meth = opt.method;
/* Use the full path, i.e. one that includes the leading slash and
the query string. E.g. if u->path is "foo/bar" and u->query is
"param=value", full_path will be "/foo/bar?param=value". */
meth_arg = xstrdup (u->url);
else
meth_arg = url_full_path (u);
- request_set_method (req, meth, meth_arg);
+ req = request_new (meth, meth_arg);
}
request_set_header (req, "Referer", (char *) hs->referer, rel_none);
}
}
- if (opt.post_data || opt.post_file_name)
+ if (opt.method)
{
- request_set_header (req, "Content-Type",
- "application/x-www-form-urlencoded", rel_none);
- if (opt.post_data)
- post_data_size = strlen (opt.post_data);
- else
+
+ if (opt.body_data || opt.body_file)
{
- post_data_size = file_size (opt.post_file_name);
- if (post_data_size == -1)
+ request_set_header (req, "Content-Type",
+ "application/x-www-form-urlencoded", rel_none);
+
+ if (opt.body_data)
+ body_data_size = strlen (opt.body_data);
+ else
{
- logprintf (LOG_NOTQUIET, _("POST data file %s missing: %s\n"),
- quote (opt.post_file_name), strerror (errno));
- return FILEBADFILE;
+ body_data_size = file_size (opt.body_file);
+ if (body_data_size == -1)
+ {
+ logprintf (LOG_NOTQUIET, _("BODY data file %s missing: %s\n"),
+ quote (opt.body_file), strerror (errno));
+ return FILEBADFILE;
+ }
}
+ request_set_header (req, "Content-Length",
+ xstrdup (number_to_static_string (body_data_size)),
+ rel_value);
}
- request_set_header (req, "Content-Length",
- xstrdup (number_to_static_string (post_data_size)),
- rel_value);
}
retry_with_auth:
{
/* When requesting SSL URLs through proxies, use the
CONNECT method to request passthrough. */
- struct request *connreq = request_new ();
- request_set_method (connreq, "CONNECT",
+ struct request *connreq = request_new ("CONNECT",
aprintf ("%s:%d", u->host, u->port));
SET_USER_AGENT (connreq);
if (proxyauth)
if (write_error >= 0)
{
- if (opt.post_data)
+ if (opt.body_data)
{
- DEBUGP (("[POST data: %s]\n", opt.post_data));
- write_error = fd_write (sock, opt.post_data, post_data_size, -1);
+ DEBUGP (("[BODY data: %s]\n", opt.body_data));
+ write_error = fd_write (sock, opt.body_data, body_data_size, -1);
if (write_error >= 0 && warc_tmp != NULL)
{
/* Remember end of headers / start of payload. */
warc_payload_offset = ftello (warc_tmp);
/* Write a copy of the data to the WARC record. */
- int warc_tmp_written = fwrite (opt.post_data, 1, post_data_size, warc_tmp);
- if (warc_tmp_written != post_data_size)
+ int warc_tmp_written = fwrite (opt.body_data, 1, body_data_size, warc_tmp);
+ if (warc_tmp_written != body_data_size)
write_error = -2;
}
- }
- else if (opt.post_file_name && post_data_size != 0)
+ }
+ else if (opt.body_file && body_data_size != 0)
{
if (warc_tmp != NULL)
- /* Remember end of headers / start of payload. */
+ /* Remember end of headers / start of payload */
warc_payload_offset = ftello (warc_tmp);
- write_error = post_file (sock, opt.post_file_name, post_data_size, warc_tmp);
+ write_error = body_file_send (sock, opt.body_file, body_data_size, warc_tmp);
}
}
/* From RFC2616: The status codes 303 and 307 have
been added for servers that wish to make unambiguously
clear which kind of reaction is expected of the client.
-
+
A 307 should be redirected using the same method,
in other words, a POST should be preserved and not
- converted to a GET in that case. */
- if (statcode == HTTP_STATUS_TEMPORARY_REDIRECT)
- return NEWLOCATION_KEEP_POST;
+ converted to a GET in that case.
+
+ With strict adherence to RFC2616, POST requests are not
+ converted to a GET request on 301 Permanent Redirect
+ or 302 Temporary Redirect.
+
+ A switch may be provided later based on the HTTPbis draft
+ that allows clients to convert POST requests to GET
+ requests on 301 and 302 response codes. */
+ switch (statcode)
+ {
+ case HTTP_STATUS_TEMPORARY_REDIRECT:
+ return NEWLOCATION_KEEP_POST;
+ break;
+ case HTTP_STATUS_MOVED_PERMANENTLY:
+ if (opt.method && strcasecmp (opt.method, "post") != 0)
+ return NEWLOCATION_KEEP_POST;
+ break;
+ case HTTP_STATUS_MOVED_TEMPORARILY:
+ if (opt.method && strcasecmp (opt.method, "post") != 0)
+ return NEWLOCATION_KEEP_POST;
+ break;
+ default:
+ return NEWLOCATION;
+ break;
+ }
return NEWLOCATION;
}
}