static int
raptor_www_file_fetch(raptor_www* www)
{
char *filename;
FILE *fh;
unsigned char *uri_string = raptor_uri_as_string(www->uri);
#if defined(HAVE_UNISTD_H) && defined(HAVE_SYS_STAT_H)
struct stat buf;
#endif
www->status_code = 200;
filename = raptor_uri_uri_string_to_filename(uri_string);
if(!filename) {
raptor_www_error(www, "Not a file: URI");
return 1;
}
#if defined(HAVE_UNISTD_H) && defined(HAVE_SYS_STAT_H)
if(!stat(filename, &buf) && S_ISDIR(buf.st_mode)) {
raptor_www_error(www, "Cannot read from a directory '%s'", filename);
RAPTOR_FREE(cstring, filename);
www->status_code = 404;
return 1;
}
#endif
fh = fopen(filename, "rb");
if(!fh) {
raptor_www_error(www, "file '%s' open failed - %s",
filename, strerror(errno));
RAPTOR_FREE(cstring, filename);
www->status_code = (errno == EACCES) ? 403: 404;
www->failed = 1;
return www->failed;
}
raptor_www_file_handle_fetch(www, fh);
fclose(fh);
RAPTOR_FREE(cstring, filename);
return www->failed;
}
int
raptor_www_curl_fetch(raptor_www *www)
{
struct curl_slist *slist=NULL;
if(www->proxy)
curl_easy_setopt(www->curl_handle, CURLOPT_PROXY, www->proxy);
if(www->user_agent)
curl_easy_setopt(www->curl_handle, CURLOPT_USERAGENT, www->user_agent);
if(www->http_accept)
slist=curl_slist_append(slist, (const char*)www->http_accept);
/* ALWAYS disable curl default "Pragma: no-cache" */
slist=curl_slist_append(slist, "Pragma:");
if(www->cache_control)
slist=curl_slist_append(slist, (const char*)www->cache_control);
if(slist)
curl_easy_setopt(www->curl_handle, CURLOPT_HTTPHEADER, slist);
/* specify URL to get */
curl_easy_setopt(www->curl_handle, CURLOPT_URL,
raptor_uri_as_string(www->uri));
if(curl_easy_perform(www->curl_handle)) {
/* failed */
www->failed=1;
raptor_www_error(www, "Resolving URI failed: %s", www->error_buffer);
} else {
long lstatus;
#ifndef CURLINFO_RESPONSE_CODE
#define CURLINFO_RESPONSE_CODE CURLINFO_HTTP_CODE
#endif
/* Requires pointer to a long */
if(curl_easy_getinfo(www->curl_handle, CURLINFO_RESPONSE_CODE, &lstatus) == CURLE_OK)
www->status_code=lstatus;
}
if(slist)
curl_slist_free_all(slist);
return www->failed;
}
/**
* raptor_www_fetch:
* @www: WWW object
* @uri: URI to read from
*
* Start a WWW content retrieval for the given URI, returning data via the write_bytes handler.
*
* Return value: non-0 on failure.
**/
int
raptor_www_fetch(raptor_www *www, raptor_uri *uri)
{
int status = 1;
www->uri = raptor_new_uri_for_retrieval(uri);
www->locator.uri = uri;
www->locator.line= -1;
www->locator.column= -1;
if(www->uri_filter)
if(www->uri_filter(www->uri_filter_user_data, uri))
return status;
#ifdef RAPTOR_WWW_NONE
status = raptor_www_file_fetch(www);
#else
if(raptor_uri_uri_string_is_file_uri(raptor_uri_as_string(www->uri)))
status = raptor_www_file_fetch(www);
else {
#ifdef RAPTOR_WWW_LIBCURL
status = raptor_www_curl_fetch(www);
#endif
#ifdef RAPTOR_WWW_LIBXML
status = raptor_www_libxml_fetch(www);
#endif
#ifdef RAPTOR_WWW_LIBFETCH
status = raptor_www_libfetch_fetch(www);
#endif
}
#endif
if(!status && www->status_code && www->status_code != 200){
raptor_www_error(www, "Resolving URI failed with HTTP status %d",
www->status_code);
status = 1;
}
www->failed = status;
return www->failed;
}
int
raptor_www_libfetch_fetch(raptor_www *www)
{
FILE *stream;
if(www->proxy) {
setenv("HTTP_PROXY", www->proxy, 0);
setenv("FTP_PROXY", www->proxy, 0);
}
if(www->user_agent)
setenv("HTTP_USER_AGENT", www->user_agent, 0);
stream = fetchXGetURL((const char*)raptor_uri_as_string(www->uri), NULL, NULL);
if(!stream) {
www->failed = 1;
raptor_www_error(www, "%s", fetchLastErrString);
return 1;
}
/* fetch does not give us access to this */
www->status_code = 200;
while(!feof(stream)) {
size_t len = fread(www->buffer, 1, RAPTOR_WWW_BUFFER_SIZE, stream);
www->total_bytes += len;
if(www->write_bytes)
www->write_bytes(www, www->write_bytes_userdata, www->buffer, len, 1);
if(len < RAPTOR_WWW_BUFFER_SIZE)
break;
}
fclose(stream);
return www->failed;
}