static void
modify_param_value (param_token *value, int encoding_type )
{
- if (RFC2231_ENCODING == encoding_type)
+ if (encoding_type == RFC2231_ENCODING)
{
const char *delim = memrchr (value->b, '\'', value->e - value->b);
if ( delim != NULL )
filename=\"foo bar\"", the first call to this function will return
the token named "attachment" and no value, and the second call will
return the token named "filename" and value "foo bar". The third
- call will return false, indicating no more valid tokens. */
+ call will return false, indicating no more valid tokens.
+
+ is_url_encoded is an out parameter. If not NULL, a boolean value will be
+ stored into it, letting the caller know whether or not the extracted value is
+ URL-encoded. The caller can then decode it with url_unescape(), which however
+ performs decoding in-place. URL-encoding is used by RFC 2231 to support
+ non-US-ASCII characters in HTTP header values. */
bool
extract_param (const char **source, param_token *name, param_token *value,
- char separator)
+ char separator, bool *is_url_encoded)
{
const char *p = *source;
+ int param_type;
+ if (is_url_encoded)
+ *is_url_encoded = false; /* initializing the out parameter */
while (c_isspace (*p)) ++p;
if (!*p)
}
*source = p;
- int param_type = modify_param_name(name);
- if (NOT_RFC2231 != param_type)
+ param_type = modify_param_name(name);
+ if (param_type != NOT_RFC2231)
{
+ if (param_type == RFC2231_ENCODING && is_url_encoded)
+ *is_url_encoded = true;
modify_param_value(value, param_type);
}
return true;
/* Appends the string represented by VALUE to FILENAME */
static void
-append_value_to_filename (char **filename, param_token const * const value)
+append_value_to_filename (char **filename, param_token const * const value,
+ bool is_url_encoded)
{
int original_length = strlen(*filename);
int new_length = strlen(*filename) + (value->e - value->b);
*filename = xrealloc (*filename, new_length+1);
memcpy (*filename + original_length, value->b, (value->e - value->b));
(*filename)[new_length] = '\0';
+ if (is_url_encoded)
+ url_unescape (*filename + original_length);
}
#undef MAX
{
param_token name, value;
*filename = NULL;
- while (extract_param (&hdr, &name, &value, ';'))
+ bool is_url_encoded = false;
+ for ( ; extract_param (&hdr, &name, &value, ';', &is_url_encoded);
+ is_url_encoded = false)
{
int isFilename = BOUNDED_EQUAL_NO_CASE ( name.b, name.e, "filename" );
if ( isFilename && value.b != NULL)
continue;
if (*filename)
- append_value_to_filename (filename, &value);
+ append_value_to_filename (filename, &value, is_url_encoded);
else
- *filename = strdupdelim (value.b, value.e);
+ {
+ *filename = strdupdelim (value.b, value.e);
+ if (is_url_encoded)
+ url_unescape (*filename);
+ }
}
}
/* Download the response body and write it to fp.
If we are working on a WARC file, we simultaneously write the
response body to warc_tmp. */
- hs->res = fd_read_body (sock, fp, contlen != -1 ? contlen : 0,
+ hs->res = fd_read_body (hs->local_file, sock, fp, contlen != -1 ? contlen : 0,
hs->restval, &hs->rd_size, &hs->len, &hs->dltime,
flags, warc_tmp);
if (hs->res >= 0)
But if we are writing a WARC file we are: we like to keep everyting. */
if (warc_enabled)
{
- int err;
+ int _err;
type = resp_header_strdup (resp, "Content-Type");
- err = read_response_body (hs, sock, NULL, contlen, 0,
+ _err = read_response_body (hs, sock, NULL, contlen, 0,
chunked_transfer_encoding,
u->url, warc_timestamp_str,
warc_request_uuid, warc_ip, type,
statcode, head);
xfree_null (type);
- if (err != RETRFINISHED || hs->res < 0)
+ if (_err != RETRFINISHED || hs->res < 0)
{
CLOSE_INVALIDATE (sock);
request_free (req);
xfree_null (message);
resp_free (resp);
xfree (head);
- return err;
+ return _err;
}
else
CLOSE_FINISH (sock);
tmp = parse_charset (tmp2);
if (tmp)
set_content_encoding (iri, tmp);
+ xfree_null(tmp);
}
}
}
if (H_20X (statcode))
*dt |= RETROKF;
+ if (statcode == HTTP_STATUS_NO_CONTENT)
+ {
+ /* 204 response has no body (RFC 2616, 4.3) */
+
+ /* In case the caller cares to look... */
+ hs->len = 0;
+ hs->res = 0;
+ hs->restval = 0;
+
+ CLOSE_FINISH (sock);
+ xfree_null (type);
+ xfree (head);
+
+ return RETRFINISHED;
+ }
+
/* Return if redirected. */
if (H_REDIRECTED (statcode) || statcode == HTTP_STATUS_MULTIPLE_CHOICES)
{
But if we are writing a WARC file we are: we like to keep everyting. */
if (warc_enabled)
{
- int err = read_response_body (hs, sock, NULL, contlen, 0,
+ int _err = read_response_body (hs, sock, NULL, contlen, 0,
chunked_transfer_encoding,
u->url, warc_timestamp_str,
warc_request_uuid, warc_ip, type,
statcode, head);
- if (err != RETRFINISHED || hs->res < 0)
+ if (_err != RETRFINISHED || hs->res < 0)
{
CLOSE_INVALIDATE (sock);
xfree_null (type);
xfree (head);
- return err;
+ return _err;
}
else
CLOSE_FINISH (sock);
{
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;
default:
return NEWLOCATION;
- break;
}
return NEWLOCATION;
}
But if we are writing a WARC file we are: we like to keep everyting. */
if (warc_enabled)
{
- int err = read_response_body (hs, sock, NULL, contlen, 0,
+ int _err = read_response_body (hs, sock, NULL, contlen, 0,
chunked_transfer_encoding,
u->url, warc_timestamp_str,
warc_request_uuid, warc_ip, type,
statcode, head);
- if (err != RETRFINISHED || hs->res < 0)
+ if (_err != RETRFINISHED || hs->res < 0)
{
CLOSE_INVALIDATE (sock);
xfree (head);
xfree_null (type);
- return err;
+ return _err;
}
else
CLOSE_FINISH (sock);
fp = output_stream;
/* Print fetch message, if opt.verbose. */
- if (opt.verbose)
- {
- logprintf (LOG_NOTQUIET, _("Saving to: %s\n"),
+ logprintf (LOG_VERBOSE, _("Saving to: %s\n"),
HYPHENP (hs->local_file) ? quote ("STDOUT") : quote (hs->local_file));
- }
err = read_response_body (hs, sock, fp, contlen, contrange,
/* Decide whether or not to restart. */
if (force_full_retrieve)
hstat.restval = hstat.len;
+ else if (opt.start_pos >= 0)
+ hstat.restval = opt.start_pos;
else if (opt.always_rest
&& got_name
&& stat (hstat.local_file, &st) == 0
realm = opaque = nonce = algorithm = qop = NULL;
au += 6; /* skip over `Digest' */
- while (extract_param (&au, &name, &value, ','))
+ while (extract_param (&au, &name, &value, ',', NULL))
{
size_t i;
size_t namelen = name.e - name.b;
#ifdef TESTING
const char *
-test_parse_content_disposition()
+test_parse_content_disposition(void)
{
- int i;
- struct {
- char *hdrval;
- char *filename;
+ unsigned i;
+ static const struct {
+ const char *hdrval;
+ const char *filename;
bool result;
} test_array[] = {
{ "filename=\"file.ext\"", "file.ext", true },
{ "attachement; filename*0=\"hello\"; filename*1=\"world.txt\"", "helloworld.txt", true },
};
- for (i = 0; i < sizeof(test_array)/sizeof(test_array[0]); ++i)
+ for (i = 0; i < countof(test_array); ++i)
{
char *filename;
bool res;
projects / wget / blobdiff