r69793 MediaWiki - Code Review archive

Repository:	MediaWiki
Revision:	< r69792‎ \| r69793 \| r69794 >
Date:	15:26, 23 July 2010
Author:	ashley
Status:	resolved (Comments)
Tags:
Comment:	committing my work on RSS extension. Now this extension should be totally self-contained. RSSCache.php, RSSFetch.php and RSSParse.php are MagpieRSS extension files by Kellan Elliott-McCrea, with some MediaWikification by me. Snoopy.class.php is Snoopy, with some coding style changes and whatnot by me. to-do: use memcached for caching instead of RSSCache.php
Modified paths:	/trunk/extensions/RSS/RSS.i18n.php (modified) (history) /trunk/extensions/RSS/RSS.php (modified) (history) /trunk/extensions/RSS/RSSCache.php (added) (history) /trunk/extensions/RSS/RSSFetch.php (added) (history) /trunk/extensions/RSS/RSSParse.php (added) (history) /trunk/extensions/RSS/Snoopy.class.php (added) (history) /trunk/extensions/RSS/magpierss (deleted) (history)

Diff [purge]

Index: trunk/extensions/RSS/RSSCache.php
—	—	@@ -0,0 +1,151 @@
	2	+<?php
	3	+/**
	4	+ * A simple, rolling (no GC) cache for RSS objects, keyed on URL.
	5	+ *
	6	+ * @file
	7	+ */
	8	+
	9	+class RSSCache {
	10	+ public $BASE_CACHE = './cache'; // where the cache files are stored
	11	+ public $MAX_AGE = 3600; // when are files stale, default one hour
	12	+ public $ERROR = ''; // accumulate error messages
	13	+
	14	+ function __construct( $base = '', $age = '' ) {
	15	+ if ( $base ) {
	16	+ $this->BASE_CACHE = $base;
	17	+ }
	18	+ if ( $age ) {
	19	+ $this->MAX_AGE = $age;
	20	+ }
	21	+
	22	+ // attempt to make the cache directory
	23	+ if ( !file_exists( $this->BASE_CACHE ) ) {
	24	+ $status = @mkdir( $this->BASE_CACHE, 0755 );
	25	+
	26	+ // if make failed
	27	+ if ( !$status ) {
	28	+ wfDebugLog(
	29	+ 'RSS',
	30	+ "Cache couldn't make dir '" . $this->BASE_CACHE . "'."
	31	+ );
	32	+ }
	33	+ }
	34	+ }
	35	+
	36	+ /**
	37	+ * Add an item to the cache, keyed on URL.
	38	+ * @param $url String: URL from which the RSS file was fetched
	39	+ * @param $rss Mixed: data to serialize
	40	+ */
	41	+ function set( $url, $rss ) {
	42	+ $this->ERROR = '';
	43	+ $cache_file = $this->file_name( $url );
	44	+ $fp = @fopen( $cache_file, 'w' );
	45	+
	46	+ if ( !$fp ) {
	47	+ wfDebugLog(
	48	+ 'RSS',
	49	+ "Cache unable to open file for writing: $cache_file"
	50	+ );
	51	+ return 0;
	52	+ }
	53	+
	54	+ $data = serialize( $rss );
	55	+ fwrite( $fp, $data );
	56	+ fclose( $fp );
	57	+
	58	+ return $cache_file;
	59	+ }
	60	+
	61	+ /**
	62	+ * Fetch an item from the cache.
	63	+ * @param $url String: URL from which the RSS file was fetched
	64	+ * @return Object or false: cached object on HIT, false on MISS
	65	+ */
	66	+ function get( $url ) {
	67	+ $this->ERROR = '';
	68	+ $cache_file = $this->file_name( $url );
	69	+
	70	+ if ( !file_exists( $cache_file ) ) {
	71	+ wfDebugLog(
	72	+ 'RSS',
	73	+ "Cache doesn't contain: $url (cache file: $cache_file)"
	74	+ );
	75	+ return 0;
	76	+ }
	77	+
	78	+ $fp = @fopen( $cache_file, 'r' );
	79	+ if ( !$fp ) {
	80	+ wfDebugLog(
	81	+ 'RSS',
	82	+ "Failed to open cache file for reading: $cache_file"
	83	+ );
	84	+ return 0;
	85	+ }
	86	+
	87	+ $filesize = filesize( $cache_file );
	88	+ if ( $filesize ) {
	89	+ $data = fread( $fp, filesize( $cache_file ) );
	90	+ $rss = unserialize( $data );
	91	+
	92	+ return $rss;
	93	+ }
	94	+
	95	+ return 0;
	96	+ }
	97	+
	98	+ /**
	99	+ * Check a URL for membership in the cache and whether the object is older
	100	+ * then MAX_AGE (ie. STALE)
	101	+ *
	102	+ * @param $url String: URL from which the RSS file was fetched
	103	+ * @return String: indicates whether there was a cache hit or not
	104	+ */
	105	+ function check_cache( $url ) {
	106	+ $this->ERROR = '';
	107	+ $filename = $this->file_name( $url );
	108	+
	109	+ if ( file_exists( $filename ) ) {
	110	+ // find how long ago the file was added to the cache
	111	+ // and whether that is longer then MAX_AGE
	112	+ $mtime = filemtime( $filename );
	113	+ $age = time() - $mtime;
	114	+ if ( $this->MAX_AGE > $age ) {
	115	+ // object exists and is current
	116	+ return 'HIT';
	117	+ } else {
	118	+ // object exists but is old
	119	+ return 'STALE';
	120	+ }
	121	+ } else {
	122	+ // object does not exist
	123	+ return 'MISS';
	124	+ }
	125	+ }
	126	+
	127	+ /**
	128	+ * @param $cache_key String: cache key, consisting of the URL + output enc.
	129	+ * @return Integer
	130	+ */
	131	+ function cache_age( $cache_key ) {
	132	+ $filename = $this->file_name( $url );
	133	+ if ( file_exists( $filename ) ) {
	134	+ $mtime = filemtime( $filename );
	135	+ $age = time() - $mtime;
	136	+ return $age;
	137	+ } else {
	138	+ return -1;
	139	+ }
	140	+ }
	141	+
	142	+ /**
	143	+ * Map URL to location in cache.
	144	+ * @param $url String: URL from which the RSS file was fetched
	145	+ * @return String: file name
	146	+ */
	147	+ function file_name( $url ) {
	148	+ $filename = md5( $url );
	149	+ return join( DIRECTORY_SEPARATOR, array( $this->BASE_CACHE, $filename ) );
	150	+ }
	151	+
	152	+}
\ No newline at end of file
Property changes on: trunk/extensions/RSS/RSSCache.php
___________________________________________________________________
Added: svn:eol-style
1	153	+ native
Index: trunk/extensions/RSS/Snoopy.class.php
—	—	@@ -0,0 +1,1241 @@
	2	+<?php
	3	+
	4	+/*************************************************
	5	+
	6	+Snoopy - the PHP net client
	7	+Author: Monte Ohrt <monte@ispi.net>
	8	+Copyright (c): 1999-2008 New Digital Group, all rights reserved
	9	+Version: 1.2.5-dev (revision 1.27)
	10	+Note: some coding style changes by Jack Phoenix <jack@countervandalism.net>
	11	+ var -> public, added some braces, double quotes -> single quotes, etc.
	12	+ also added the gzip support stuff from MagpieRSS' Snoopy to this ver
	13	+
	14	+ * This library is free software; you can redistribute it and/or
	15	+ * modify it under the terms of the GNU Lesser General Public
	16	+ * License as published by the Free Software Foundation; either
	17	+ * version 2.1 of the License, or (at your option) any later version.
	18	+ *
	19	+ * This library is distributed in the hope that it will be useful,
	20	+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
	21	+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	22	+ * Lesser General Public License for more details.
	23	+ *
	24	+ * You should have received a copy of the GNU Lesser General Public
	25	+ * License along with this library; if not, write to the Free Software
	26	+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
	27	+
	28	+You may contact the author of Snoopy by e-mail at:
	29	+monte@ohrt.com
	30	+
	31	+The latest version of Snoopy can be obtained from:
	32	+http://snoopy.sourceforge.net/
	33	+
	34	+*************************************************/
	35	+
	36	+class Snoopy {
	37	+ /** Public variables **/
	38	+
	39	+ /* user definable vars */
	40	+ public $host = 'www.php.net'; // host name we are connecting to
	41	+ public $port = 80; // port we are connecting to
	42	+ public $proxy_host = ''; // proxy host to use
	43	+ public $proxy_port = ''; // proxy port to use
	44	+ public $proxy_user = ''; // proxy user to use
	45	+ public $proxy_pass = ''; // proxy password to use
	46	+
	47	+ public $agent = 'Snoopy v1.2.5-dev'; // agent we masquerade as
	48	+ public $referer = ''; // referer info to pass
	49	+ public $cookies = array(); // array of cookies to pass
	50	+ // $cookies['username'] = 'joe';
	51	+ public $rawheaders = array(); // array of raw headers to send
	52	+ // $rawheaders['Content-type'] = 'text/html';
	53	+
	54	+ public $maxredirs = 5; // http redirection depth maximum. 0 = disallow
	55	+ public $lastredirectaddr = ''; // contains address of last redirected address
	56	+ public $offsiteok = true; // allows redirection off-site
	57	+ public $maxframes = 0; // frame content depth maximum. 0 = disallow
	58	+ public $expandlinks = true; // expand links to fully qualified URLs.
	59	+ // this only applies to fetchlinks()
	60	+ // submitlinks(), and submittext()
	61	+ public $passcookies = true; // pass set cookies back through redirects
	62	+ // NOTE: this currently does not respect
	63	+ // dates, domains or paths.
	64	+
	65	+ public $user = ''; // user for http authentication
	66	+ public $pass = ''; // password for http authentication
	67	+
	68	+ // http accept types
	69	+ public $accept = 'image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, /';
	70	+
	71	+ public $results = ''; // where the content is put
	72	+
	73	+ public $error = ''; // error messages sent here
	74	+ public $response_code = ''; // response code returned from server
	75	+ public $headers = array(); // headers returned from server sent here
	76	+ public $maxlength = 500000; // max return data length (body)
	77	+ public $read_timeout = 0; // timeout on read operations, in seconds
	78	+ // supported only since PHP 4 Beta 4
	79	+ // set to 0 to disallow timeouts
	80	+ public $timed_out = false; // if a read operation timed out
	81	+ public $status = 0; // http request status
	82	+
	83	+ public $temp_dir = '/tmp'; // temporary directory that the webserver
	84	+ // has permission to write to.
	85	+ // under Windows, this should be C:\temp
	86	+
	87	+ public $curl_path = '/usr/local/bin/curl';
	88	+ // Snoopy will use cURL for fetching
	89	+ // SSL content if a full system path to
	90	+ // the cURL binary is supplied here.
	91	+ // set to false if you do not have
	92	+ // cURL installed. See http://curl.haxx.se
	93	+ // for details on installing cURL.
	94	+ // Snoopy does not use the cURL
	95	+ // library functions built into php,
	96	+ // as these functions are not stable
	97	+ // as of this Snoopy release.
	98	+
	99	+ // send Accept-encoding: gzip?
	100	+ public $use_gzip = true;
	101	+
	102	+ /** Private variables **/
	103	+ var $_maxlinelen = 4096; // max line length (headers)
	104	+
	105	+ var $_httpmethod = 'GET'; // default http request method
	106	+ var $_httpversion = 'HTTP/1.0'; // default http request version
	107	+ var $_submit_method = 'POST'; // default submit method
	108	+ var $_submit_type = 'application/x-www-form-urlencoded'; // default submit type
	109	+ var $_mime_boundary = ''; // MIME boundary for multipart/form-data submit type
	110	+ var $_redirectaddr = false; // will be set if page fetched is a redirect
	111	+ var $_redirectdepth = 0; // increments on an http redirect
	112	+ var $_frameurls = array(); // frame src urls
	113	+ var $_framedepth = 0; // increments on frame depth
	114	+
	115	+ var $_isproxy = false; // set if using a proxy server
	116	+ var $_fp_timeout = 30; // timeout for socket connection
	117	+
	118	+ /======================================================================\
	119	+ Function: fetch
	120	+ Purpose: fetch the contents of a web page
	121	+ (and possibly other protocols in the
	122	+ future like ftp, nntp, gopher, etc.)
	123	+ Input: $URI the location of the page to fetch
	124	+ Output: $this->results the output text from the fetch
	125	+ \======================================================================/
	126	+ function fetch( $URI ) {
	127	+ //preg_match( "\|^([^:]+)://([^:/]+)(:[\d]+)(.)\|", $URI, $URI_PARTS );
	128	+ $URI_PARTS = parse_url( $URI );
	129	+ if ( !empty( $URI_PARTS['user'] ) ) {
	130	+ $this->user = $URI_PARTS['user'];
	131	+ }
	132	+ if ( !empty( $URI_PARTS['pass'] ) ) {
	133	+ $this->pass = $URI_PARTS['pass'];
	134	+ }
	135	+ if ( empty( $URI_PARTS['query'] ) ) {
	136	+ $URI_PARTS['query'] = '';
	137	+ }
	138	+ if ( empty( $URI_PARTS['path'] ) ) {
	139	+ $URI_PARTS['path'] = '';
	140	+ }
	141	+
	142	+ switch( strtolower( $URI_PARTS['scheme'] ) ) {
	143	+ case 'http':
	144	+ $this->host = $URI_PARTS['host'];
	145	+ if( !empty( $URI_PARTS['port'] ) ) {
	146	+ $this->port = $URI_PARTS['port'];
	147	+ }
	148	+ if( $this->_connect( $fp ) ) {
	149	+ if( $this->_isproxy ) {
	150	+ // using proxy, send entire URI
	151	+ $this->_httprequest( $URI, $fp, $URI, $this->_httpmethod );
	152	+ } else {
	153	+ $path = $URI_PARTS['path'] . ( isset( $URI_PARTS['query'] ) ? '?' . $URI_PARTS['query'] : '' );
	154	+ // no proxy, send only the path
	155	+ $this->_httprequest( $path, $fp, $URI, $this->_httpmethod );
	156	+ }
	157	+
	158	+ $this->_disconnect( $fp );
	159	+
	160	+ if( $this->_redirectaddr ) {
	161	+ /* url was redirected, check if we've hit the max depth */
	162	+ if( $this->maxredirs > $this->_redirectdepth ) {
	163	+ // only follow redirect if it's on this site, or offsiteok is true
	164	+ if( preg_match( "\|^http://" . preg_quote( $this->host ) . "\|i", $this->_redirectaddr ) \|\| $this->offsiteok )
	165	+ {
	166	+ /* follow the redirect */
	167	+ $this->_redirectdepth++;
	168	+ $this->lastredirectaddr = $this->_redirectaddr;
	169	+ $this->fetch( $this->_redirectaddr );
	170	+ }
	171	+ }
	172	+ }
	173	+
	174	+ if( $this->_framedepth < $this->maxframes && count( $this->_frameurls ) > 0 )
	175	+ {
	176	+ $frameurls = $this->_frameurls;
	177	+ $this->_frameurls = array();
	178	+
	179	+ while( list( , $frameurl ) = each( $frameurls ) ) {
	180	+ if( $this->_framedepth < $this->maxframes ) {
	181	+ $this->fetch( $frameurl );
	182	+ $this->_framedepth++;
	183	+ } else {
	184	+ break;
	185	+ }
	186	+ }
	187	+ }
	188	+ } else {
	189	+ return false;
	190	+ }
	191	+ return true;
	192	+ break;
	193	+ case 'https':
	194	+ if( !$this->curl_path ) {
	195	+ return false;
	196	+ }
	197	+ if( function_exists( 'is_executable' ) ) {
	198	+ if ( !is_executable( $this->curl_path ) ) {
	199	+ $this->error = "Bad curl ($this->curl_path), can't fetch HTTPS \n";
	200	+ return false;
	201	+ }
	202	+ }
	203	+ $this->host = $URI_PARTS['host'];
	204	+ if( !empty( $URI_PARTS['port'] ) ) {
	205	+ $this->port = $URI_PARTS['port'];
	206	+ }
	207	+ if( $this->_isproxy ) {
	208	+ // using proxy, send entire URI
	209	+ $this->_httpsrequest( $URI, $URI, $this->_httpmethod );
	210	+ } else {
	211	+ $path = $URI_PARTS['path'] . ( $URI_PARTS['query'] ? '?' . $URI_PARTS['query'] : '' );
	212	+ // no proxy, send only the path
	213	+ $this->_httpsrequest( $path, $URI, $this->_httpmethod );
	214	+ }
	215	+
	216	+ if( $this->_redirectaddr ) {
	217	+ /* url was redirected, check if we've hit the max depth */
	218	+ if( $this->maxredirs > $this->_redirectdepth ) {
	219	+ // only follow redirect if it's on this site, or offsiteok is true
	220	+ if( preg_match( "\|^http://" . preg_quote( $this->host ) . "\|i", $this->_redirectaddr ) \|\| $this->offsiteok )
	221	+ {
	222	+ /* follow the redirect */
	223	+ $this->_redirectdepth++;
	224	+ $this->lastredirectaddr = $this->_redirectaddr;
	225	+ $this->fetch( $this->_redirectaddr );
	226	+ }
	227	+ }
	228	+ }
	229	+
	230	+ if( $this->_framedepth < $this->maxframes && count( $this->_frameurls ) > 0 )
	231	+ {
	232	+ $frameurls = $this->_frameurls;
	233	+ $this->_frameurls = array();
	234	+
	235	+ while( list( , $frameurl ) = each( $frameurls ) ) {
	236	+ if( $this->_framedepth < $this->maxframes ) {
	237	+ $this->fetch( $frameurl );
	238	+ $this->_framedepth++;
	239	+ } else {
	240	+ break;
	241	+ }
	242	+ }
	243	+ }
	244	+ return true;
	245	+ break;
	246	+ default:
	247	+ // not a valid protocol
	248	+ $this->error = 'Invalid protocol "' . $URI_PARTS['scheme'] . '"\n';
	249	+ return false;
	250	+ break;
	251	+ }
	252	+ return true;
	253	+ }
	254	+
	255	+ /======================================================================\
	256	+ Function: submit
	257	+ Purpose: submit an HTTP form
	258	+ Input: $URI the location to post the data
	259	+ $formvars the formvars to use.
	260	+ format: $formvars['var'] = 'val';
	261	+ $formfiles an array of files to submit
	262	+ format: $formfiles['var'] = '/dir/filename.ext';
	263	+ Output: $this->results the text output from the post
	264	+ \======================================================================/
	265	+ function submit( $URI, $formvars = '', $formfiles = '' ) {
	266	+ unset( $postdata );
	267	+
	268	+ $postdata = $this->_prepare_post_body( $formvars, $formfiles );
	269	+
	270	+ $URI_PARTS = parse_url( $URI );
	271	+ if ( !empty( $URI_PARTS['user'] ) ) {
	272	+ $this->user = $URI_PARTS['user'];
	273	+ }
	274	+ if ( !empty( $URI_PARTS['pass'] ) ) {
	275	+ $this->pass = $URI_PARTS['pass'];
	276	+ }
	277	+ if ( empty( $URI_PARTS['query'] ) ) {
	278	+ $URI_PARTS['query'] = '';
	279	+ }
	280	+ if ( empty( $URI_PARTS['path'] ) ) {
	281	+ $URI_PARTS['path'] = '';
	282	+ }
	283	+
	284	+ switch( strtolower( $URI_PARTS['scheme'] ) ) {
	285	+ case 'http':
	286	+ $this->host = $URI_PARTS['host'];
	287	+ if( !empty( $URI_PARTS['port'] ) ) {
	288	+ $this->port = $URI_PARTS['port'];
	289	+ }
	290	+ if( $this->_connect( $fp ) ) {
	291	+ if( $this->_isproxy ) {
	292	+ // using proxy, send entire URI
	293	+ $this->_httprequest( $URI, $fp, $URI, $this->_submit_method, $this->_submit_type, $postdata );
	294	+ } else {
	295	+ $path = $URI_PARTS['path'] . ( $URI_PARTS['query'] ? '?' . $URI_PARTS['query'] : '' );
	296	+ // no proxy, send only the path
	297	+ $this->_httprequest(
	298	+ $path, $fp, $URI,
	299	+ $this->_submit_method,
	300	+ $this->_submit_type,
	301	+ $postdata
	302	+ );
	303	+ }
	304	+
	305	+ $this->_disconnect( $fp );
	306	+
	307	+ if( $this->_redirectaddr ) {
	308	+ /* url was redirected, check if we've hit the max depth */
	309	+ if( $this->maxredirs > $this->_redirectdepth ) {
	310	+ if( !preg_match( "\|^" . $URI_PARTS['scheme'] . "://\|", $this->_redirectaddr ) ) {
	311	+ $this->_redirectaddr = $this->_expandlinks( $this->_redirectaddr, $URI_PARTS['scheme'] . '://' . $URI_PARTS['host'] );
	312	+ }
	313	+
	314	+ // only follow redirect if it's on this site, or offsiteok is true
	315	+ if( preg_match( "\|^http://" . preg_quote( $this->host ) . "\|i", $this->_redirectaddr ) \|\| $this->offsiteok )
	316	+ {
	317	+ /* follow the redirect */
	318	+ $this->_redirectdepth++;
	319	+ $this->lastredirectaddr = $this->_redirectaddr;
	320	+ if( strpos( $this->_redirectaddr, '?' ) > 0 ) {
	321	+ $this->fetch( $this->_redirectaddr ); // the redirect has changed the request method from post to get
	322	+ } else {
	323	+ $this->submit( $this->_redirectaddr, $formvars, $formfiles );
	324	+ }
	325	+ }
	326	+ }
	327	+ }
	328	+
	329	+ if( $this->_framedepth < $this->maxframes && count( $this->_frameurls ) > 0 )
	330	+ {
	331	+ $frameurls = $this->_frameurls;
	332	+ $this->_frameurls = array();
	333	+
	334	+ while( list( , $frameurl ) = each( $frameurls ) ) {
	335	+ if( $this->_framedepth < $this->maxframes ) {
	336	+ $this->fetch( $frameurl );
	337	+ $this->_framedepth++;
	338	+ } else {
	339	+ break;
	340	+ }
	341	+ }
	342	+ }
	343	+ } else {
	344	+ return false;
	345	+ }
	346	+ return true;
	347	+ break;
	348	+ case 'https':
	349	+ if( !$this->curl_path ) {
	350	+ return false;
	351	+ }
	352	+ if( function_exists( 'is_executable' ) ) {
	353	+ if ( !is_executable( $this->curl_path ) ) {
	354	+ return false;
	355	+ }
	356	+ }
	357	+ $this->host = $URI_PARTS['host'];
	358	+ if( !empty( $URI_PARTS['port'] ) ) {
	359	+ $this->port = $URI_PARTS['port'];
	360	+ }
	361	+ if( $this->_isproxy ) {
	362	+ // using proxy, send entire URI
	363	+ $this->_httpsrequest(
	364	+ $URI,
	365	+ $URI,
	366	+ $this->_submit_method,
	367	+ $this->_submit_type,
	368	+ $postdata
	369	+ );
	370	+ } else {
	371	+ $path = $URI_PARTS['path'] . ( $URI_PARTS['query'] ? '?' . $URI_PARTS['query'] : '' );
	372	+ // no proxy, send only the path
	373	+ $this->_httpsrequest(
	374	+ $path,
	375	+ $URI,
	376	+ $this->_submit_method,
	377	+ $this->_submit_type,
	378	+ $postdata
	379	+ );
	380	+ }
	381	+
	382	+ if( $this->_redirectaddr ) {
	383	+ /* url was redirected, check if we've hit the max depth */
	384	+ if( $this->maxredirs > $this->_redirectdepth ) {
	385	+ if( !preg_match( "\|^" . $URI_PARTS['scheme'] . "://\|", $this->_redirectaddr ) ) {
	386	+ $this->_redirectaddr = $this->_expandlinks(
	387	+ $this->_redirectaddr,
	388	+ $URI_PARTS['scheme'] . '://' . $URI_PARTS['host']
	389	+ );
	390	+ }
	391	+
	392	+ // only follow redirect if it's on this site, or offsiteok is true
	393	+ if( preg_match( "\|^http://" . preg_quote( $this->host ) . "\|i", $this->_redirectaddr ) \|\| $this->offsiteok )
	394	+ {
	395	+ /* follow the redirect */
	396	+ $this->_redirectdepth++;
	397	+ $this->lastredirectaddr = $this->_redirectaddr;
	398	+ if( strpos( $this->_redirectaddr, '?' ) > 0 ) {
	399	+ $this->fetch( $this->_redirectaddr ); // the redirect has changed the request method from post to get
	400	+ } else {
	401	+ $this->submit( $this->_redirectaddr, $formvars, $formfiles );
	402	+ }
	403	+ }
	404	+ }
	405	+ }
	406	+
	407	+ if( $this->_framedepth < $this->maxframes && count( $this->_frameurls ) > 0 )
	408	+ {
	409	+ $frameurls = $this->_frameurls;
	410	+ $this->_frameurls = array();
	411	+
	412	+ while( list( , $frameurl ) = each( $frameurls ) ) {
	413	+ if( $this->_framedepth < $this->maxframes ) {
	414	+ $this->fetch( $frameurl );
	415	+ $this->_framedepth++;
	416	+ } else {
	417	+ break;
	418	+ }
	419	+ }
	420	+ }
	421	+ return true;
	422	+ break;
	423	+
	424	+ default:
	425	+ // not a valid protocol
	426	+ $this->error = 'Invalid protocol "' . $URI_PARTS['scheme'] . '"\n';
	427	+ return false;
	428	+ break;
	429	+ }
	430	+ return true;
	431	+ }
	432	+
	433	+ /======================================================================\
	434	+ Function: fetchlinks
	435	+ Purpose: fetch the links from a web page
	436	+ Input: $URI where you are fetching from
	437	+ Output: $this->results an array of the URLs
	438	+ \======================================================================/
	439	+ function fetchlinks( $URI ) {
	440	+ if ( $this->fetch( $URI ) ) {
	441	+ if( $this->lastredirectaddr ) {
	442	+ $URI = $this->lastredirectaddr;
	443	+ }
	444	+ if( is_array( $this->results ) ) {
	445	+ for( $x = 0; $x < count( $this->results ); $x++ ) {
	446	+ $this->results[$x] = $this->_striplinks( $this->results[$x] );
	447	+ }
	448	+ } else {
	449	+ $this->results = $this->_striplinks( $this->results );
	450	+ }
	451	+
	452	+ if( $this->expandlinks ) {
	453	+ $this->results = $this->_expandlinks( $this->results, $URI );
	454	+ }
	455	+ return true;
	456	+ } else {
	457	+ return false;
	458	+ }
	459	+ }
	460	+
	461	+ /======================================================================\
	462	+ Function: fetchform
	463	+ Purpose: fetch the form elements from a web page
	464	+ Input: $URI where you are fetching from
	465	+ Output: $this->results the resulting html form
	466	+ \======================================================================/
	467	+ function fetchform( $URI ) {
	468	+ if ( $this->fetch( $URI ) ) {
	469	+ if( is_array( $this->results ) ) {
	470	+ for( $x = 0; $x < count( $this->results ); $x++ ) {
	471	+ $this->results[$x] = $this->_stripform( $this->results[$x] );
	472	+ }
	473	+ } else {
	474	+ $this->results = $this->_stripform( $this->results );
	475	+ }
	476	+
	477	+ return true;
	478	+ } else {
	479	+ return false;
	480	+ }
	481	+ }
	482	+
	483	+
	484	+ /======================================================================\
	485	+ Function: fetchtext
	486	+ Purpose: fetch the text from a web page, stripping the links
	487	+ Input: $URI where you are fetching from
	488	+ Output: $this->results the text from the web page
	489	+ \======================================================================/
	490	+ function fetchtext( $URI ) {
	491	+ if( $this->fetch( $URI ) ) {
	492	+ if( is_array( $this->results ) ) {
	493	+ for( $x = 0; $x < count( $this->results ); $x++ ) {
	494	+ $this->results[$x] = $this->_striptext( $this->results[$x] );
	495	+ }
	496	+ } else {
	497	+ $this->results = $this->_striptext( $this->results );
	498	+ }
	499	+ return true;
	500	+ } else {
	501	+ return false;
	502	+ }
	503	+ }
	504	+
	505	+ /======================================================================\
	506	+ Function: submitlinks
	507	+ Purpose: grab links from a form submission
	508	+ Input: $URI where you are submitting from
	509	+ Output: $this->results an array of the links from the post
	510	+ \======================================================================/
	511	+ function submitlinks( $URI, $formvars = '', $formfiles = '' ) {
	512	+ if( $this->submit( $URI, $formvars, $formfiles ) ) {
	513	+ if( $this->lastredirectaddr ) {
	514	+ $URI = $this->lastredirectaddr;
	515	+ }
	516	+ if( is_array( $this->results ) ) {
	517	+ for( $x = 0; $x < count( $this->results ); $x++ ) {
	518	+ $this->results[$x] = $this->_striplinks( $this->results[$x] );
	519	+ if( $this->expandlinks ) {
	520	+ $this->results[$x] = $this->_expandlinks( $this->results[$x], $URI );
	521	+ }
	522	+ }
	523	+ } else {
	524	+ $this->results = $this->_striplinks( $this->results );
	525	+ if( $this->expandlinks ) {
	526	+ $this->results = $this->_expandlinks( $this->results, $URI );
	527	+ }
	528	+ }
	529	+ return true;
	530	+ } else {
	531	+ return false;
	532	+ }
	533	+ }
	534	+
	535	+ /======================================================================\
	536	+ Function: submittext
	537	+ Purpose: grab text from a form submission
	538	+ Input: $URI where you are submitting from
	539	+ Output: $this->results the text from the web page
	540	+ \======================================================================/
	541	+ function submittext( $URI, $formvars = '', $formfiles = '' ) {
	542	+ if( $this->submit( $URI, $formvars, $formfiles ) ) {
	543	+ if( $this->lastredirectaddr ) {
	544	+ $URI = $this->lastredirectaddr;
	545	+ }
	546	+ if( is_array( $this->results ) ) {
	547	+ for( $x = 0; $x < count( $this->results ); $x++ ) {
	548	+ $this->results[$x] = $this->_striptext( $this->results[$x] );
	549	+ if( $this->expandlinks ) {
	550	+ $this->results[$x] = $this->_expandlinks( $this->results[$x], $URI );
	551	+ }
	552	+ }
	553	+ } else {
	554	+ $this->results = $this->_striptext( $this->results );
	555	+ if( $this->expandlinks ) {
	556	+ $this->results = $this->_expandlinks( $this->results, $URI );
	557	+ }
	558	+ }
	559	+ return true;
	560	+ } else {
	561	+ return false;
	562	+ }
	563	+ }
	564	+
	565	+ /======================================================================\
	566	+ Function: set_submit_multipart
	567	+ Purpose: Set the form submission content type to
	568	+ multipart/form-data
	569	+ \======================================================================/
	570	+ function set_submit_multipart() {
	571	+ $this->_submit_type = 'multipart/form-data';
	572	+ }
	573	+
	574	+ /======================================================================\
	575	+ Function: set_submit_normal
	576	+ Purpose: Set the form submission content type to
	577	+ application/x-www-form-urlencoded
	578	+ \======================================================================/
	579	+ function set_submit_normal() {
	580	+ $this->_submit_type = 'application/x-www-form-urlencoded';
	581	+ }
	582	+
	583	+ /======================================================================\
	584	+ Private functions
	585	+ \======================================================================/
	586	+
	587	+ /======================================================================\
	588	+ Function: _striplinks
	589	+ Purpose: strip the hyperlinks from an html document
	590	+ Input: $document document to strip.
	591	+ Output: $match an array of the links
	592	+ \======================================================================/
	593	+ function _striplinks( $document ) {
	594	+ preg_match_all("'<\sa\s.?href\s=\s # find <a href=
	595	+ ([\"\'])? # find single or double quote
	596	+ (?(1) (.*?)\\1 \| ([^\s\>]+)) # if quote found, match up to next matching
	597	+ # quote, otherwise match up to next space
	598	+ 'isx", $document, $links
	599	+ );
	600	+
	601	+ // catenate the non-empty matches from the conditional subpattern
	602	+ while( list( $key, $val ) = each( $links[2] ) ) {
	603	+ if( !empty( $val ) ) {
	604	+ $match[] = $val;
	605	+ }
	606	+ }
	607	+
	608	+ while( list( $key, $val ) = each( $links[3] ) ) {
	609	+ if( !empty( $val ) ) {
	610	+ $match[] = $val;
	611	+ }
	612	+ }
	613	+
	614	+ // return the links
	615	+ return $match;
	616	+ }
	617	+
	618	+ /======================================================================\
	619	+ Function: _stripform
	620	+ Purpose: strip the form elements from an HTML document
	621	+ Input: $document document to strip.
	622	+ Output: $match an array of the links
	623	+ \======================================================================/
	624	+ function _stripform( $document ) {
	625	+ preg_match_all(
	626	+ "'<\/?(FORM\|INPUT\|SELECT\|TEXTAREA\|(OPTION))[^<>]>(?(2)(.(?=<\/?(option\|select)[^<>]>[\r\n])\|(?=[\r\n]))\|(?=[\r\n]))'Usi",
	627	+ $document,
	628	+ $elements
	629	+ );
	630	+
	631	+ // catenate the matches
	632	+ $match = implode( "\r\n", $elements[0] );
	633	+
	634	+ // return the links
	635	+ return $match;
	636	+ }
	637	+
	638	+ /======================================================================\
	639	+ Function: _striptext
	640	+ Purpose: strip the text from an html document
	641	+ Input: $document document to strip.
	642	+ Output: $text the resulting text
	643	+ \======================================================================/
	644	+ function _striptext( $document ) {
	645	+ // I didn't use preg eval (//e) since that is only available in PHP 4.0.
	646	+ // so, list your entities one by one here. I included some of the
	647	+ // more common ones.
	648	+ $search = array(
	649	+ "'<script[^>]?>.?</script>'si", // strip out JavaScript
	650	+ "'<[\/\!]?[^<>]?>'si", // strip out HTML tags
	651	+ "'([\r\n])[\s]+'", // strip out white space
	652	+ "'&(quot\|#34\|#034\|#x22);'i", // replace HTML entities
	653	+ "'&(amp\|#38\|#038\|#x26);'i", // added hexadecimal values
	654	+ "'&(lt\|#60\|#060\|#x3c);'i",
	655	+ "'&(gt\|#62\|#062\|#x3e);'i",
	656	+ "'&(nbsp\|#160\|#xa0);'i",
	657	+ "'&(iexcl\|#161);'i",
	658	+ "'&(cent\|#162);'i",
	659	+ "'&(pound\|#163);'i",
	660	+ "'&(copy\|#169);'i",
	661	+ "'&(reg\|#174);'i",
	662	+ "'&(deg\|#176);'i",
	663	+ "'&(#39\|#039\|#x27);'",
	664	+ "'&(euro\|#8364);'i", // Europe
	665	+ "'&a(uml\|UML);'", // German
	666	+ "'&o(uml\|UML);'",
	667	+ "'&u(uml\|UML);'",
	668	+ "'&A(uml\|UML);'",
	669	+ "'&O(uml\|UML);'",
	670	+ "'&U(uml\|UML);'",
	671	+ "'ß'i",
	672	+ );
	673	+ $replace = array(
	674	+ '',
	675	+ '',
	676	+ "\\1",
	677	+ "\"",
	678	+ '&',
	679	+ '<',
	680	+ '>',
	681	+ ' ',
	682	+ chr( 161 ),
	683	+ chr( 162 ),
	684	+ chr( 163 ),
	685	+ chr( 169 ),
	686	+ chr( 174 ),
	687	+ chr( 176 ),
	688	+ chr( 39 ),
	689	+ chr( 128 ),
	690	+ 'ä',
	691	+ 'ö',
	692	+ 'ü',
	693	+ 'Ä',
	694	+ 'Ö',
	695	+ 'Ü',
	696	+ 'ß',
	697	+ );
	698	+
	699	+ $text = preg_replace( $search, $replace, $document );
	700	+
	701	+ return $text;
	702	+ }
	703	+
	704	+ /======================================================================\
	705	+ Function: _expandlinks
	706	+ Purpose: expand each link into a fully qualified URL
	707	+ Input: $links the links to qualify
	708	+ $URI the full URI to get the base from
	709	+ Output: $expandedLinks the expanded links
	710	+ \======================================================================/
	711	+ function _expandlinks( $links, $URI ) {
	712	+ preg_match( "/^[^\?]+/", $URI, $match );
	713	+
	714	+ $match = preg_replace( "\|/[^\/\.]+\.[^\/\.]+$\|", '', $match[0] );
	715	+ $match = preg_replace( "\|/$\|", '', $match );
	716	+ $match_part = parse_url( $match );
	717	+ $match_root = $match_part['scheme'] . '://' . $match_part['host'];
	718	+
	719	+ $search = array(
	720	+ "\|^http://" . preg_quote( $this->host ) . "\|i",
	721	+ "\|^(\/)\|i",
	722	+ "\|^(?!http://)(?!mailto:)\|i",
	723	+ "\|/\./\|",
	724	+ "\|/[^\/]+/\.\./\|"
	725	+ );
	726	+
	727	+ $replace = array(
	728	+ '',
	729	+ $match_root . '/',
	730	+ $match . '/',
	731	+ '/',
	732	+ '/'
	733	+ );
	734	+
	735	+ $expandedLinks = preg_replace( $search, $replace, $links );
	736	+
	737	+ return $expandedLinks;
	738	+ }
	739	+
	740	+ /======================================================================\
	741	+ Function: _httprequest
	742	+ Purpose: go get the http data from the server
	743	+ Input: $url the url to fetch
	744	+ $fp the current open file pointer
	745	+ $URI the full URI
	746	+ $body body contents to send if any (POST)
	747	+ Output:
	748	+ \======================================================================/
	749	+ function _httprequest( $url, $fp, $URI, $http_method, $content_type = '', $body = '' ) {
	750	+ $cookie_headers = '';
	751	+ if( $this->passcookies && $this->_redirectaddr ) {
	752	+ $this->setcookies();
	753	+ }
	754	+
	755	+ $URI_PARTS = parse_url( $URI );
	756	+ if( empty( $url ) ) {
	757	+ $url = '/';
	758	+ }
	759	+ $headers = $http_method . ' ' . $url . ' ' . $this->_httpversion . "\r\n";
	760	+ if( !empty( $this->agent ) ) {
	761	+ $headers .= 'User-Agent: ' . $this->agent . "\r\n";
	762	+ }
	763	+ if( !empty( $this->host ) && !isset( $this->rawheaders['Host'] ) ) {
	764	+ $headers .= 'Host: ' . $this->host;
	765	+ if( !empty( $this->port ) ) {
	766	+ $headers .= ':' . $this->port;
	767	+ }
	768	+ $headers .= "\r\n";
	769	+ }
	770	+ if( !empty( $this->accept ) ) {
	771	+ $headers .= 'Accept: ' . $this->accept . "\r\n";
	772	+ }
	773	+
	774	+ if( $this->use_gzip ) {
	775	+ // make sure PHP was built with --with-zlib
	776	+ // and we can handle gzipp'ed data
	777	+ if ( function_exists( 'gzinflate' ) ) {
	778	+ $headers .= "Accept-encoding: gzip\r\n";
	779	+ } else {
	780	+ trigger_error(
	781	+ 'use_gzip is on, but PHP was built without zlib support.' .
	782	+ ' Requesting file(s) without gzip encoding.',
	783	+ E_USER_NOTICE
	784	+ );
	785	+ }
	786	+ }
	787	+
	788	+ if( !empty( $this->referer ) ) {
	789	+ $headers .= 'Referer: ' . $this->referer . "\r\n";
	790	+ }
	791	+ if( !empty( $this->cookies ) ) {
	792	+ if( !is_array( $this->cookies ) ) {
	793	+ $this->cookies = (array)$this->cookies;
	794	+ }
	795	+
	796	+ reset( $this->cookies );
	797	+ if ( count( $this->cookies ) > 0 ) {
	798	+ $cookie_headers .= 'Cookie: ';
	799	+ foreach ( $this->cookies as $cookieKey => $cookieVal ) {
	800	+ $cookie_headers .= $cookieKey . '=' . urlencode( $cookieVal ) . '; ';
	801	+ }
	802	+ $headers .= substr( $cookie_headers, 0, -2 ) . "\r\n";
	803	+ }
	804	+ }
	805	+ if( !empty( $this->rawheaders ) ) {
	806	+ if( !is_array( $this->rawheaders ) ) {
	807	+ $this->rawheaders = (array)$this->rawheaders;
	808	+ }
	809	+ while( list( $headerKey, $headerVal ) = each( $this->rawheaders ) ) {
	810	+ $headers .= $headerKey . ': ' . $headerVal . "\r\n";
	811	+ }
	812	+ }
	813	+ if( !empty( $content_type ) ) {
	814	+ $headers .= "Content-type: $content_type";
	815	+ if ( $content_type == 'multipart/form-data' ) {
	816	+ $headers .= '; boundary=' . $this->_mime_boundary;
	817	+ }
	818	+ $headers .= "\r\n";
	819	+ }
	820	+ if( !empty( $body ) ) {
	821	+ $headers .= 'Content-length: ' . strlen( $body ) . "\r\n";
	822	+ }
	823	+ if( !empty( $this->user ) \|\| !empty( $this->pass ) ) {
	824	+ $headers .= 'Authorization: Basic ' . base64_encode( $this->user . ':' . $this->pass ) . "\r\n";
	825	+ }
	826	+
	827	+ // add proxy auth headers
	828	+ if( !empty( $this->proxy_user ) ) {
	829	+ $headers .= 'Proxy-Authorization: ' . 'Basic ' . base64_encode( $this->proxy_user . ':' . $this->proxy_pass ) . "\r\n";
	830	+ }
	831	+
	832	+ $headers .= "\r\n";
	833	+
	834	+ // set the read timeout if needed
	835	+ if ( $this->read_timeout > 0 ) {
	836	+ socket_set_timeout( $fp, $this->read_timeout );
	837	+ }
	838	+ $this->timed_out = false;
	839	+
	840	+ fwrite( $fp, $headers . $body, strlen( $headers . $body ) );
	841	+
	842	+ $this->_redirectaddr = false;
	843	+ unset( $this->headers );
	844	+
	845	+ // content was returned gzip encoded?
	846	+ $is_gzipped = false;
	847	+
	848	+ while( $currentHeader = fgets( $fp, $this->_maxlinelen ) ) {
	849	+ if ( $this->read_timeout > 0 && $this->_check_timeout( $fp ) ) {
	850	+ $this->status = -100;
	851	+ return false;
	852	+ }
	853	+
	854	+ //if( $currentHeader == "\r\n" ) {
	855	+ if( preg_match( "/^\r?\n$/", $currentHeader ) ) {
	856	+ break;
	857	+ }
	858	+
	859	+ // if a header begins with Location: or URI:, set the redirect
	860	+ if( preg_match( "/^(Location:\|URI:)/i", $currentHeader ) ) {
	861	+ // get URL portion of the redirect
	862	+ preg_match( "/^(Location:\|URI:)[ ]+(.*)/i", chop( $currentHeader ), $matches );
	863	+ // look for :// in the Location header to see if hostname is included
	864	+ if( !preg_match( "\|\:\/\/\|", $matches[2] ) ) {
	865	+ // no host in the path, so prepend
	866	+ $this->_redirectaddr = $URI_PARTS['scheme'] . '://' . $this->host . ':' . $this->port;
	867	+ // eliminate double slash
	868	+ if( !preg_match( "\|^/\|", $matches[2] ) ) {
	869	+ $this->_redirectaddr .= '/' . $matches[2];
	870	+ } else {
	871	+ $this->_redirectaddr .= $matches[2];
	872	+ }
	873	+ } else {
	874	+ $this->_redirectaddr = $matches[2];
	875	+ }
	876	+ }
	877	+
	878	+ if( preg_match( "\|^HTTP/\|", $currentHeader ) ) {
	879	+ if( preg_match( "\|^HTTP/[^\s]\s(.?)\s\|", $currentHeader, $status ) ) {
	880	+ $this->status = $status[1];
	881	+ }
	882	+ $this->response_code = $currentHeader;
	883	+ }
	884	+
	885	+ if ( preg_match( "/Content-Encoding: gzip/", $currentHeader ) ) {
	886	+ $is_gzipped = true;
	887	+ }
	888	+
	889	+ $this->headers[] = $currentHeader;
	890	+ }
	891	+
	892	+ $results = '';
	893	+ do {
	894	+ $_data = fread( $fp, $this->maxlength );
	895	+ if ( strlen( $_data ) == 0 \|\| strlen( $results ) > $this->maxlength ) {
	896	+ break;
	897	+ }
	898	+ $results .= $_data;
	899	+ } while( true );
	900	+
	901	+ // gunzip
	902	+ if ( $is_gzipped ) {
	903	+ // per http://www.php.net/manual/en/function.gzencode.php
	904	+ $results = substr( $results, 10 );
	905	+ $results = gzinflate( $results );
	906	+ }
	907	+
	908	+ if ( $this->read_timeout > 0 && $this->_check_timeout( $fp ) ) {
	909	+ $this->status = -100;
	910	+ return false;
	911	+ }
	912	+
	913	+ // check if there is a a redirect meta tag
	914	+ if( preg_match( "'<meta[\s]http-equiv[^>]?content[\s]=[\s][\"\']?\d+;[\s]URL[\s]=[\s]([^\"\']?)[\"\']?>'i", $results, $match ) )
	915	+ {
	916	+ $this->_redirectaddr = $this->_expandlinks( $match[1], $URI );
	917	+ }
	918	+
	919	+ // have we hit our frame depth and is there frame src to fetch?
	920	+ if( ( $this->_framedepth < $this->maxframes ) && preg_match_all( "'<frame\s+.src[\s]=[\'\"]?([^\'\"\>]+)'i", $results, $match ) )
	921	+ {
	922	+ $this->results[] = $results;
	923	+ for( $x = 0; $x < count( $match[1] ); $x++ ) {
	924	+ $this->_frameurls[] = $this->_expandlinks( $match[1][$x], $URI_PARTS['scheme'] . '://' . $this->host );
	925	+ }
	926	+ } elseif( is_array( $this->results ) ) { // have we already fetched framed content?
	927	+ $this->results[] = $results;
	928	+ } else { // no framed content
	929	+ $this->results = $results;
	930	+ }
	931	+
	932	+ return true;
	933	+ }
	934	+
	935	+ /======================================================================\
	936	+ Function: _httpsrequest
	937	+ Purpose: go get the https data from the server using curl
	938	+ Input: $url the url to fetch
	939	+ $URI the full URI
	940	+ $body body contents to send if any (POST)
	941	+ Output:
	942	+ \======================================================================/
	943	+ function _httpsrequest( $url, $URI, $http_method, $content_type = '', $body = '' ) {
	944	+ if( $this->passcookies && $this->_redirectaddr ) {
	945	+ $this->setcookies();
	946	+ }
	947	+
	948	+ $headers = array();
	949	+
	950	+ $URI_PARTS = parse_url( $URI );
	951	+ if( empty( $url ) ) {
	952	+ $url = '/';
	953	+ }
	954	+ // GET ... header not needed for curl
	955	+ //$headers[] = $http_method." ".$url." ".$this->_httpversion;
	956	+ if( !empty( $this->agent ) ) {
	957	+ $headers[] = 'User-Agent: ' . $this->agent;
	958	+ }
	959	+ if( !empty( $this->host ) ) {
	960	+ if( !empty( $this->port ) ) {
	961	+ $headers[] = 'Host: ' . $this->host . ':' . $this->port;
	962	+ } else {
	963	+ $headers[] = 'Host: ' . $this->host;
	964	+ }
	965	+ }
	966	+ if( !empty( $this->accept ) ) {
	967	+ $headers[] = 'Accept: ' . $this->accept;
	968	+ }
	969	+ if( !empty( $this->referer ) ) {
	970	+ $headers[] = 'Referer: ' . $this->referer;
	971	+ }
	972	+ if( !empty( $this->cookies ) ) {
	973	+ if( !is_array( $this->cookies ) ) {
	974	+ $this->cookies = (array)$this->cookies;
	975	+ }
	976	+
	977	+ reset( $this->cookies );
	978	+ if ( count( $this->cookies ) > 0 ) {
	979	+ $cookie_str = 'Cookie: ';
	980	+ foreach ( $this->cookies as $cookieKey => $cookieVal ) {
	981	+ $cookie_str .= $cookieKey . '=' . urlencode( $cookieVal ) . '; ';
	982	+ }
	983	+ $headers[] = substr( $cookie_str, 0, -2 );
	984	+ }
	985	+ }
	986	+ if( !empty( $this->rawheaders ) ) {
	987	+ if( !is_array( $this->rawheaders ) ) {
	988	+ $this->rawheaders = (array)$this->rawheaders;
	989	+ }
	990	+ while( list( $headerKey, $headerVal ) = each( $this->rawheaders ) ) {
	991	+ $headers[] = $headerKey . ': ' . $headerVal;
	992	+ }
	993	+ }
	994	+ if( !empty( $content_type ) ) {
	995	+ if ( $content_type == 'multipart/form-data' ) {
	996	+ $headers[] = "Content-type: $content_type; boundary=" . $this->_mime_boundary;
	997	+ } else {
	998	+ $headers[] = "Content-type: $content_type";
	999	+ }
	1000	+ }
	1001	+ if( !empty( $body ) ) {
	1002	+ $headers[] = 'Content-length: ' . strlen( $body );
	1003	+ }
	1004	+ if( !empty( $this->user ) \|\| !empty( $this->pass ) ) {
	1005	+ $headers[] = 'Authorization: BASIC ' . base64_encode( $this->user . ':' . $this->pass );
	1006	+ }
	1007	+
	1008	+ for( $curr_header = 0; $curr_header < count( $headers ); $curr_header++ ) {
	1009	+ $cmdline_params .= " -H \"" . escapeshellcmd( $headers[$curr_header] ) . "\"";
	1010	+ }
	1011	+
	1012	+ if( !empty( $body ) ) {
	1013	+ $cmdline_params .= " -d \"" . escapeshellcmd( $body ) . "\"";
	1014	+ }
	1015	+
	1016	+ if( $this->read_timeout > 0 ) {
	1017	+ $cmdline_params .= ' -m ' . $this->read_timeout;
	1018	+ }
	1019	+
	1020	+ $headerfile = tempnam( $temp_dir, 'sno' );
	1021	+
	1022	+ exec(
	1023	+ $this->curl_path . " -k -D \"$headerfile\"" . $cmdline_params . " \"" . escapeshellcmd( $URI ) . "\"",
	1024	+ $results,
	1025	+ $return
	1026	+ );
	1027	+
	1028	+ if( $return ) {
	1029	+ $this->error = "Error: cURL could not retrieve the document, error $return.";
	1030	+ return false;
	1031	+ }
	1032	+
	1033	+ $results = implode( "\r\n", $results );
	1034	+
	1035	+ $result_headers = file( "$headerfile" );
	1036	+
	1037	+ $this->_redirectaddr = false;
	1038	+ unset( $this->headers );
	1039	+
	1040	+ for( $currentHeader = 0; $currentHeader < count( $result_headers ); $currentHeader++ ) {
	1041	+ // if a header begins with Location: or URI:, set the redirect
	1042	+ if( preg_match( "/^(Location: \|URI: )/i", $result_headers[$currentHeader] ) ) {
	1043	+ // get URL portion of the redirect
	1044	+ preg_match( "/^(Location: \|URI:)\s+(.*)/", chop( $result_headers[$currentHeader] ), $matches );
	1045	+ // look for :// in the Location header to see if hostname is included
	1046	+ if( !preg_match( "\|\:\/\/\|", $matches[2] ) ) {
	1047	+ // no host in the path, so prepend
	1048	+ $this->_redirectaddr = $URI_PARTS['scheme'] . '://' . $this->host . ':' . $this->port;
	1049	+ // eliminate double slash
	1050	+ if( !preg_match( "\|^/\|", $matches[2] ) ) {
	1051	+ $this->_redirectaddr .= '/' . $matches[2];
	1052	+ } else {
	1053	+ $this->_redirectaddr .= $matches[2];
	1054	+ }
	1055	+ } else {
	1056	+ $this->_redirectaddr = $matches[2];
	1057	+ }
	1058	+ }
	1059	+
	1060	+ if( preg_match( "\|^HTTP/\|", $result_headers[$currentHeader] ) ) {
	1061	+ $this->response_code = $result_headers[$currentHeader];
	1062	+ if( preg_match( "\|^HTTP/[^\s]\s(.?)\s\|", $this->response_code, $match ) ) {
	1063	+ $this->status = $match[1];
	1064	+ }
	1065	+ }
	1066	+
	1067	+ $this->headers[] = $result_headers[$currentHeader];
	1068	+ }
	1069	+
	1070	+ // check if there is a a redirect meta tag
	1071	+ if( preg_match( "'<meta[\s]http-equiv[^>]?content[\s]=[\s][\"\']?\d+;[\s]URL[\s]=[\s]([^\"\']?)[\"\']?>'i", $results, $match ) )
	1072	+ {
	1073	+ $this->_redirectaddr = $this->_expandlinks( $match[1], $URI );
	1074	+ }
	1075	+
	1076	+ // have we hit our frame depth and is there frame src to fetch?
	1077	+ if( ( $this->_framedepth < $this->maxframes ) && preg_match_all( "'<frame\s+.src[\s]=[\'\"]?([^\'\"\>]+)'i", $results, $match ) )
	1078	+ {
	1079	+ $this->results[] = $results;
	1080	+ for( $x = 0; $x < count( $match[1] ); $x++ ) {
	1081	+ $this->_frameurls[] = $this->_expandlinks( $match[1][$x], $URI_PARTS['scheme'] . '://' . $this->host );
	1082	+ }
	1083	+ } elseif( is_array( $this->results ) ) { // have we already fetched framed content?
	1084	+ $this->results[] = $results;
	1085	+ } else { // no framed content
	1086	+ $this->results = $results;
	1087	+ }
	1088	+
	1089	+ unlink( "$headerfile" );
	1090	+
	1091	+ return true;
	1092	+ }
	1093	+
	1094	+ /======================================================================\
	1095	+ Function: setcookies()
	1096	+ Purpose: set cookies for a redirection
	1097	+ \======================================================================/
	1098	+ function setcookies() {
	1099	+ for( $x = 0; $x < count( $this->headers ); $x++ ) {
	1100	+ if( preg_match( '/^set-cookie:[\s]+([^=]+)=([^;]+)/i', $this->headers[$x], $match ) ) {
	1101	+ $this->cookies[$match[1]] = urldecode( $match[2] );
	1102	+ }
	1103	+ }
	1104	+ }
	1105	+
	1106	+ /======================================================================\
	1107	+ Function: _check_timeout
	1108	+ Purpose: checks whether timeout has occurred
	1109	+ Input: $fp file pointer
	1110	+ \======================================================================/
	1111	+ function _check_timeout( $fp ) {
	1112	+ if ( $this->read_timeout > 0 ) {
	1113	+ $fp_status = socket_get_status( $fp );
	1114	+ if ( $fp_status['timed_out'] ) {
	1115	+ $this->timed_out = true;
	1116	+ return true;
	1117	+ }
	1118	+ }
	1119	+ return false;
	1120	+ }
	1121	+
	1122	+ /======================================================================\
	1123	+ Function: _connect
	1124	+ Purpose: make a socket connection
	1125	+ Input: $fp file pointer
	1126	+ \======================================================================/
	1127	+ function _connect( &$fp ) {
	1128	+ if( !empty( $this->proxy_host ) && !empty( $this->proxy_port ) ) {
	1129	+ $this->_isproxy = true;
	1130	+ $host = $this->proxy_host;
	1131	+ $port = $this->proxy_port;
	1132	+ } else {
	1133	+ $host = $this->host;
	1134	+ $port = $this->port;
	1135	+ }
	1136	+
	1137	+ $this->status = 0;
	1138	+
	1139	+ $fp = fsockopen( $host, $port, $errno, $errstr, $this->_fp_timeout );
	1140	+
	1141	+ if ( $fp ) {
	1142	+ // socket connection succeeded
	1143	+ return true;
	1144	+ } else {
	1145	+ // socket connection failed
	1146	+ $this->status = $errno;
	1147	+ switch( $errno ) {
	1148	+ case -3:
	1149	+ $this->error = 'socket creation failed (-3)';
	1150	+ case -4:
	1151	+ $this->error = 'dns lookup failure (-4)';
	1152	+ case -5:
	1153	+ $this->error = 'connection refused or timed out (-5)';
	1154	+ default:
	1155	+ $this->error = 'connection failed (' . $errno . ')';
	1156	+ }
	1157	+ return false;
	1158	+ }
	1159	+ }
	1160	+
	1161	+ /======================================================================\
	1162	+ Function: _disconnect
	1163	+ Purpose: disconnect a socket connection
	1164	+ Input: $fp file pointer
	1165	+ \======================================================================/
	1166	+ function _disconnect( $fp ) {
	1167	+ return( fclose( $fp ) );
	1168	+ }
	1169	+
	1170	+ /======================================================================\
	1171	+ Function: _prepare_post_body
	1172	+ Purpose: Prepare post body according to encoding type
	1173	+ Input: $formvars - form variables
	1174	+ $formfiles - form upload files
	1175	+ Output: post body
	1176	+ \======================================================================/
	1177	+ function _prepare_post_body( $formvars, $formfiles ) {
	1178	+ settype( $formvars, 'array' );
	1179	+ settype( $formfiles, 'array' );
	1180	+ $postdata = '';
	1181	+
	1182	+ if ( count( $formvars ) == 0 && count( $formfiles ) == 0 ) {
	1183	+ return;
	1184	+ }
	1185	+
	1186	+ switch ( $this->_submit_type ) {
	1187	+ case 'application/x-www-form-urlencoded':
	1188	+ reset( $formvars );
	1189	+ while( list( $key, $val ) = each( $formvars ) ) {
	1190	+ if ( is_array( $val ) \|\| is_object( $val ) ) {
	1191	+ while ( list( $cur_key, $cur_val ) = each( $val ) ) {
	1192	+ $postdata .= urlencode( $key ) . '[]=' . urlencode( $cur_val ) . '&';
	1193	+ }
	1194	+ } else {
	1195	+ $postdata .= urlencode( $key ) . '=' . urlencode( $val ) . '&';
	1196	+ }
	1197	+ }
	1198	+ break;
	1199	+
	1200	+ case 'multipart/form-data':
	1201	+ $this->_mime_boundary = 'Snoopy' . md5( uniqid( microtime() ) );
	1202	+
	1203	+ reset( $formvars );
	1204	+ while( list( $key, $val ) = each( $formvars ) ) {
	1205	+ if ( is_array( $val ) \|\| is_object( $val ) ) {
	1206	+ while ( list( $cur_key, $cur_val ) = each( $val ) ) {
	1207	+ $postdata .= '--' . $this->_mime_boundary . "\r\n";
	1208	+ $postdata .= "Content-Disposition: form-data; name=\"$key\[\]\"\r\n\r\n";
	1209	+ $postdata .= "$cur_val\r\n";
	1210	+ }
	1211	+ } else {
	1212	+ $postdata .= '--' . $this->_mime_boundary . "\r\n";
	1213	+ $postdata .= "Content-Disposition: form-data; name=\"$key\"\r\n\r\n";
	1214	+ $postdata .= "$val\r\n";
	1215	+ }
	1216	+ }
	1217	+
	1218	+ reset( $formfiles );
	1219	+ while ( list( $field_name, $file_names ) = each( $formfiles ) ) {
	1220	+ settype( $file_names, 'array' );
	1221	+ while ( list( , $file_name ) = each( $file_names ) ) {
	1222	+ if ( !is_readable( $file_name ) ) {
	1223	+ continue;
	1224	+ }
	1225	+
	1226	+ $fp = fopen( $file_name, 'r' );
	1227	+ $file_content = fread( $fp, filesize( $file_name ) );
	1228	+ fclose( $fp );
	1229	+ $base_name = basename( $file_name );
	1230	+
	1231	+ $postdata .= '--' . $this->_mime_boundary . "\r\n";
	1232	+ $postdata .= "Content-Disposition: form-data; name=\"$field_name\"; filename=\"$base_name\"\r\n\r\n";
	1233	+ $postdata .= "$file_content\r\n";
	1234	+ }
	1235	+ }
	1236	+ $postdata .= '--' . $this->_mime_boundary . "--\r\n";
	1237	+ break;
	1238	+ }
	1239	+
	1240	+ return $postdata;
	1241	+ }
	1242	+}
\ No newline at end of file
Property changes on: trunk/extensions/RSS/Snoopy.class.php
___________________________________________________________________
Added: svn:eol-style
1	1243	+ native
Index: trunk/extensions/RSS/RSSFetch.php
—	—	@@ -0,0 +1,256 @@
	2	+<?php
	3	+/**
	4	+ * A simple functional interface to fetching and parsing RSS files, via the
	5	+ * function fetch_rss().
	6	+ *
	7	+ * @file
	8	+ */
	9	+
	10	+/**
	11	+ * Globals - redefine these in your script to change the
	12	+ * behaviour of fetch_rss() currently, most options effect the cache
	13	+ *
	14	+ * $wgMagpieRSSCache - Should Magpie cache parsed RSS objects?
	15	+ *
	16	+ * $wgMagpieRSSCacheDirectory - Where should Magpie cache parsed RSS objects?
	17	+ * This should be a location that the webserver can write to. If this
	18	+ * directory does not already exist, Magpie will try to be smart and create it.
	19	+ * This will often fail for permissions reasons.
	20	+ *
	21	+ * $wgMagpieRSSCacheAge - How long to store cached RSS objects (in seconds)?.
	22	+ *
	23	+ * $wgMagpieRSSCacheFreshOnly - If remote fetch fails, throw an error
	24	+ * instead of returning stale object?
	25	+ */
	26	+
	27	+$MAGPIE_ERROR = '';
	28	+
	29	+/**
	30	+ * Return RSS object for the given URL, maintaining caching.
	31	+ *
	32	+ * NOTES ON CACHING:
	33	+ * If caching is on ($wgMagpieRSSCache) fetch_rss will first check the cache.
	34	+ *
	35	+ * NOTES ON RETRIEVING REMOTE FILES:
	36	+ * If conditional gets are on (MAGPIE_CONDITIONAL_GET_ON) fetch_rss will
	37	+ * return a cached object, and touch the cache object upon recieving a 304.
	38	+ *
	39	+ * NOTES ON FAILED REQUESTS:
	40	+ * If there is an HTTP error while fetching an RSS object, the cached version
	41	+ * will be returned, if it exists (and if $wgMagpieRSSCacheFreshOnly is off)
	42	+ *
	43	+ * @param $url String: URL of RSS file
	44	+ * @return parsed RSS object (see RSSParse)
	45	+ */
	46	+function fetch_rss( $url ) {
	47	+ global $wgMagpieRSSCache, $wgMagpieRSSCacheAge, $wgMagpieRSSCacheFreshOnly;
	48	+ global $wgMagpieRSSCacheDirectory, $wgMagpieRSSFetchTimeout;
	49	+ global $wgMagpieRSSOutputEncoding, $wgMagpieRSSInputEncoding;
	50	+ global $wgMagpieRSSDetectEncoding, $wgMagpieRSSUseGzip;
	51	+
	52	+ $wgMagpieRSSCache = true;
	53	+ $wgMagpieRSSCacheAge = 60 * 60; // one hour
	54	+ $wgMagpieRSSCacheFreshOnly = false;
	55	+ $wgMagpieRSSCacheDirectory = '/extensions/RSS/cache';
	56	+ $wgMagpieRSSOutputEncoding = 'ISO-8859-1';
	57	+ $wgMagpieRSSInputEncoding = null;
	58	+ $wgMagpieRSSDetectEncoding = true;
	59	+
	60	+ $wgMagpieRSSFetchTimeout = 5; // 5 second timeout
	61	+
	62	+ // use gzip encoding to fetch RSS files if supported?
	63	+ $wgMagpieRSSUseGzip = true;
	64	+
	65	+ if ( !isset( $url ) ) {
	66	+ wfDebugLog( 'RSS', 'fetch_rss (RSSFetch.php) called without a URL!' );
	67	+ return false;
	68	+ }
	69	+
	70	+ // if cache is disabled
	71	+ if ( !$wgMagpieRSSCache ) {
	72	+ // fetch file, and parse it
	73	+ $resp = _fetch_remote_file( $url );
	74	+ if ( $resp->status >= 200 && $resp->status < 300 ) {
	75	+ return _response_to_rss( $resp );
	76	+ } else {
	77	+ wfDebugLog( 'RSS', "Failed to fetch $url and cache is off" );
	78	+ return false;
	79	+ }
	80	+ } else { // else cache is ON
	81	+ // Flow
	82	+ // 1. check cache
	83	+ // 2. if there is a hit, make sure its fresh
	84	+ // 3. if cached obj fails freshness check, fetch remote
	85	+ // 4. if remote fails, return stale object, or error
	86	+ $cache = new RSSCache( $wgMagpieRSSCacheDirectory, $wgMagpieRSSCacheAge );
	87	+
	88	+ if ( $cache->ERROR ) {
	89	+ wfDebugLog(
	90	+ 'RSS',
	91	+ 'MagpieRSS: cache error on RSSFetch.php! Error msg: ' .
	92	+ $cache->ERROR
	93	+ );
	94	+ }
	95	+
	96	+ $cache_status = 0; // response of check_cache
	97	+ $request_headers = array(); // HTTP headers to send with fetch
	98	+ $rss = 0; // parsed RSS object
	99	+ $errormsg = 0; // errors, if any
	100	+
	101	+ // store parsed XML by desired output encoding
	102	+ // as character munging happens at parse time
	103	+ $cache_key = $url . $wgMagpieRSSOutputEncoding;
	104	+
	105	+ if ( !$cache->ERROR ) {
	106	+ // return cache HIT, MISS, or STALE
	107	+ $cache_status = $cache->check_cache( $cache_key );
	108	+ }
	109	+
	110	+ // if object cached, and cache is fresh, return cached obj
	111	+ if ( $cache_status == 'HIT' ) {
	112	+ $rss = $cache->get( $cache_key );
	113	+ if ( isset( $rss ) && $rss ) {
	114	+ // should be cache age
	115	+ $rss->from_cache = 1;
	116	+ wfDebugLog( 'RSS', 'MagpieRSS: Cache HIT' );
	117	+ return $rss;
	118	+ }
	119	+ }
	120	+
	121	+ // else attempt a conditional get
	122	+
	123	+ // setup headers
	124	+ if ( $cache_status == 'STALE' ) {
	125	+ $rss = $cache->get( $cache_key );
	126	+ if ( $rss && $rss->etag && $rss->last_modified ) {
	127	+ $request_headers['If-None-Match'] = $rss->etag;
	128	+ $request_headers['If-Last-Modified'] = $rss->last_modified;
	129	+ }
	130	+ }
	131	+
	132	+ $resp = _fetch_remote_file( $url, $request_headers );
	133	+
	134	+ if ( isset( $resp ) && $resp ) {
	135	+ if ( $resp->status == '304' ) {
	136	+ // we have the most current copy
	137	+ wfDebugLog( 'RSS', "Got 304 for $url" );
	138	+ // reset cache on 304 (at minutillo insistent prodding)
	139	+ $cache->set( $cache_key, $rss );
	140	+ return $rss;
	141	+ } elseif ( $resp->status >= 200 && $resp->status < 300 ) {
	142	+ $rss = _response_to_rss( $resp );
	143	+ if ( $rss ) {
	144	+ wfDebugLog( 'RSS', 'Fetch successful' );
	145	+ // add object to cache
	146	+ $cache->set( $cache_key, $rss );
	147	+ return $rss;
	148	+ }
	149	+ } else {
	150	+ $errormsg = "Failed to fetch $url ";
	151	+ if ( $resp->status == '-100' ) {
	152	+ global $wgMagpieRSSFetchTimeout;
	153	+ $errormsg .= '(Request timed out after ' . $wgMagpieRSSFetchTimeout . ' seconds)';
	154	+ } elseif ( $resp->error ) {
	155	+ // compensate for Snoopy's annoying habbit to tacking
	156	+ // on '\n'
	157	+ $http_error = substr( $resp->error, 0, -2 );
	158	+ $errormsg .= "(HTTP Error: $http_error)";
	159	+ } else {
	160	+ $errormsg .= '(HTTP Response: ' . $resp->response_code . ')';
	161	+ }
	162	+ }
	163	+ } else {
	164	+ $errormsg = 'Unable to retrieve RSS file for unknown reasons.';
	165	+ }
	166	+
	167	+ // else fetch failed
	168	+
	169	+ // attempt to return cached object
	170	+ if ( $rss ) {
	171	+ wfDebugLog( 'RSS', "Returning STALE object for $url" );
	172	+ return $rss;
	173	+ }
	174	+
	175	+ // else we totally failed
	176	+ $MAGPIE_ERROR = $errormsg;
	177	+ wfDebugLog(
	178	+ 'MagpieRSS (RSSFetch): we totally failed :-( Error message:' .
	179	+ $errormsg
	180	+ );
	181	+
	182	+ return false;
	183	+ } // end if ( !$wgMagpieRSSCache ) {
	184	+} // end fetch_rss()
	185	+
	186	+/**
	187	+ * Retrieve an arbitrary remote file.
	188	+ * @param $url String: URL of the remote file
	189	+ * @param $headers Array: headers to send along with the request
	190	+ * @return an HTTP response object (see Snoopy.class.php)
	191	+ */
	192	+function _fetch_remote_file( $url, $headers = '' ) {
	193	+ global $wgMagpieRSSFetchTimeout, $wgMagpieRSSUseGzip;
	194	+ // Snoopy is an HTTP client in PHP
	195	+ if ( !class_exists( 'Snoopy', false ) ) {
	196	+ require_once( dirname( __FILE__ ) . '/Snoopy.class.php' );
	197	+ }
	198	+ $client = new Snoopy();
	199	+ $client->agent = 'MagpieRSS/0.72 (+http://magpierss.sourceforge.net) / MediaWiki RSS extension';
	200	+ $client->read_timeout = $wgMagpieRSSFetchTimeout;
	201	+ $client->use_gzip = $wgMagpieRSSUseGzip;
	202	+ if ( is_array( $headers ) ) {
	203	+ $client->rawheaders = $headers;
	204	+ }
	205	+
	206	+ @$client->fetch( $url );
	207	+ return $client;
	208	+}
	209	+
	210	+/**
	211	+ * Parse an HTTP response object into an RSS object.
	212	+ * @param $resp Object: an HTTP response object (see Snoopy)
	213	+ * @return parsed RSS object (see RSSParse) or false
	214	+ */
	215	+function _response_to_rss( $resp ) {
	216	+ global $wgMagpieRSSOutputEncoding, $wgMagpieRSSInputEncoding, $wgMagpieRSSDetectEncoding;
	217	+ $rss = new MagpieRSS(
	218	+ $resp->results,
	219	+ $wgMagpieRSSOutputEncoding,
	220	+ $wgMagpieRSSInputEncoding,
	221	+ $wgMagpieRSSDetectEncoding
	222	+ );
	223	+
	224	+ // if RSS parsed successfully
	225	+ if ( $rss && !$rss->ERROR ) {
	226	+ // find Etag and Last-Modified
	227	+ foreach( $resp->headers as $h ) {
	228	+ // 2003-03-02 - Nicola Asuni (www.tecnick.com) - fixed bug "Undefined offset: 1"
	229	+ if ( strpos( $h, ': ' ) ) {
	230	+ list( $field, $val ) = explode( ': ', $h, 2 );
	231	+ } else {
	232	+ $field = $h;
	233	+ $val = '';
	234	+ }
	235	+
	236	+ if ( $field == 'ETag' ) {
	237	+ $rss->etag = $val;
	238	+ }
	239	+
	240	+ if ( $field == 'Last-Modified' ) {
	241	+ $rss->last_modified = $val;
	242	+ }
	243	+ }
	244	+
	245	+ return $rss;
	246	+ } else { // else construct error message
	247	+ $errormsg = 'Failed to parse RSS file.';
	248	+
	249	+ if ( $rss ) {
	250	+ $errormsg .= ' (' . $rss->ERROR . ')';
	251	+ }
	252	+ $MAGPIE_ERROR = $errormsg;
	253	+ wfDebugLog( 'RSS', 'error!' . $errormsg );
	254	+
	255	+ return false;
	256	+ } // end if ( $rss && !$rss->ERROR )
	257	+}
\ No newline at end of file
Property changes on: trunk/extensions/RSS/RSSFetch.php
___________________________________________________________________
Added: svn:eol-style
1	258	+ native
Index: trunk/extensions/RSS/RSSParse.php
—	—	@@ -0,0 +1,494 @@
	2	+<?php
	3	+/**
	4	+ * Hybrid parser, and object, takes RSS or Atom feed as a string and returns a
	5	+ * simple object.
	6	+ * Handles RSS 0.9x, RSS 2.0, RSS 1.0, and Atom 0.3
	7	+ *
	8	+ * @file
	9	+ * @see RSSFetch.php for a simpler interface with integrated caching support
	10	+ */
	11	+class MagpieRSS {
	12	+ public $parser;
	13	+
	14	+ public $current_item = array(); // item currently being parsed
	15	+ public $items = array(); // collection of parsed items
	16	+ public $channel = array(); // hash of channel fields
	17	+ public $textinput = array();
	18	+ public $image = array();
	19	+ public $feed_type;
	20	+ public $feed_version;
	21	+ public $encoding = ''; // output encoding of parsed rss
	22	+
	23	+ public $_source_encoding = ''; // only set if we have to parse xml prolog
	24	+
	25	+ public $ERROR = '';
	26	+ public $WARNING = '';
	27	+
	28	+ // define some constants
	29	+ public $_CONTENT_CONSTRUCTS = array( 'content', 'summary', 'info', 'title', 'tagline', 'copyright' );
	30	+ public $_KNOWN_ENCODINGS = array( 'UTF-8', 'US-ASCII', 'ISO-8859-1' );
	31	+
	32	+ // parser variables, useless if you're not a parser, treat as private
	33	+ public $stack = array(); // parser stack
	34	+ public $inchannel = false;
	35	+ public $initem = false;
	36	+ public $incontent = false; // if in Atom <content mode="xml"> field
	37	+ public $intextinput = false;
	38	+ public $inimage = false;
	39	+ public $current_namespace = false;
	40	+
	41	+ /**
	42	+ * Set up XML parser, parse source, and return populated RSS object..
	43	+ *
	44	+ * @param $source String: string containing the RSS to be parsed
	45	+ *
	46	+ * NOTE: Probably a good idea to leave the encoding options alone unless
	47	+ * you know what you're doing as PHP's character set support is
	48	+ * a little weird.
	49	+ *
	50	+ * NOTE: A lot of this is unnecessary but harmless with PHP5
	51	+ *
	52	+ *
	53	+ * @param $output_encoding String: output the parsed RSS in this character
	54	+ * set defaults to ISO-8859-1 as this is PHP's
	55	+ * default.
	56	+ *
	57	+ * NOTE: might be changed to UTF-8 in future
	58	+ * versions.
	59	+ *
	60	+ * @param $input_encoding String: the character set of the incoming RSS source.
	61	+ * Leave blank and Magpie will try to figure it
	62	+ * out.
	63	+ *
	64	+ * @param $detect_encoding Boolean: if false, Magpie won't attempt to
	65	+ * detect source encoding. (caveat emptor)
	66	+ */
	67	+ function __construct( $source, $output_encoding = 'ISO-8859-1',
	68	+ $input_encoding = null, $detect_encoding = true )
	69	+ {
	70	+ # if PHP xml isn't compiled in, die
	71	+ if ( !function_exists( 'xml_parser_create' ) ) {
	72	+ $this->error(
	73	+ "Failed to load PHP's XML Extension. " .
	74	+ 'http://www.php.net/manual/en/ref.xml.php',
	75	+ E_USER_ERROR
	76	+ );
	77	+ }
	78	+
	79	+ list( $parser, $source ) = $this->create_parser(
	80	+ $source,
	81	+ $output_encoding,
	82	+ $input_encoding,
	83	+ $detect_encoding
	84	+ );
	85	+
	86	+ if ( !is_resource( $parser ) ) {
	87	+ $this->error(
	88	+ "Failed to create an instance of PHP's XML parser. " .
	89	+ 'http://www.php.net/manual/en/ref.xml.php',
	90	+ E_USER_ERROR
	91	+ );
	92	+ }
	93	+
	94	+ $this->parser = $parser;
	95	+
	96	+ # pass in parser, and a reference to this object
	97	+ # setup handlers
	98	+ xml_set_object( $this->parser, $this );
	99	+ xml_set_element_handler(
	100	+ $this->parser,
	101	+ 'feed_start_element',
	102	+ 'feed_end_element'
	103	+ );
	104	+
	105	+ xml_set_character_data_handler( $this->parser, 'feed_cdata' );
	106	+
	107	+ $status = xml_parse( $this->parser, $source );
	108	+
	109	+ if ( !$status ) {
	110	+ $errorcode = xml_get_error_code( $this->parser );
	111	+ if ( $errorcode != XML_ERROR_NONE ) {
	112	+ $xml_error = xml_error_string( $errorcode );
	113	+ $error_line = xml_get_current_line_number( $this->parser );
	114	+ $error_col = xml_get_current_column_number( $this->parser );
	115	+ $errormsg = "$xml_error at line $error_line, column $error_col";
	116	+
	117	+ $this->error( $errormsg );
	118	+ }
	119	+ }
	120	+
	121	+ xml_parser_free( $this->parser );
	122	+
	123	+ $this->normalize();
	124	+ }
	125	+
	126	+ function feed_start_element( $p, $element, &$attrs ) {
	127	+ $el = $element = strtolower( $element );
	128	+ $attrs = array_change_key_case( $attrs, CASE_LOWER );
	129	+
	130	+ // check for a namespace, and split if found
	131	+ $ns = false;
	132	+ if ( strpos( $element, ':' ) ) {
	133	+ list( $ns, $el ) = explode( ':', $element, 2 );
	134	+ }
	135	+ if ( $ns && $ns != 'rdf' ) {
	136	+ $this->current_namespace = $ns;
	137	+ }
	138	+
	139	+ // if feed type isn't set, then this is first element of feed
	140	+ // identify feed from root element
	141	+ if ( !isset( $this->feed_type ) ) {
	142	+ if ( $el == 'rdf' ) {
	143	+ $this->feed_type = 'RSS';
	144	+ $this->feed_version = '1.0';
	145	+ } elseif ( $el == 'rss' ) {
	146	+ $this->feed_type = 'RSS';
	147	+ $this->feed_version = $attrs['version'];
	148	+ } elseif ( $el == 'feed' ) {
	149	+ $this->feed_type = 'Atom';
	150	+ $this->feed_version = $attrs['version'];
	151	+ $this->inchannel = true;
	152	+ }
	153	+ return;
	154	+ }
	155	+
	156	+ if ( $el == 'channel' ) {
	157	+ $this->inchannel = true;
	158	+ } elseif ( $el == 'item' \|\| $el == 'entry' ) {
	159	+ $this->initem = true;
	160	+ if ( isset( $attrs['rdf:about'] ) ) {
	161	+ $this->current_item['about'] = $attrs['rdf:about'];
	162	+ }
	163	+ }
	164	+
	165	+ // if we're in the default namespace of an RSS feed,
	166	+ // record textinput or image fields
	167	+ elseif (
	168	+ $this->feed_type == 'RSS' &&
	169	+ $this->current_namespace == '' &&
	170	+ $el == 'textinput' )
	171	+ {
	172	+ $this->intextinput = true;
	173	+ } elseif (
	174	+ $this->feed_type == 'RSS' &&
	175	+ $this->current_namespace == '' &&
	176	+ $el == 'image' )
	177	+ {
	178	+ $this->inimage = true;
	179	+ }
	180	+
	181	+ // handle Atom content constructs
	182	+ elseif ( $this->feed_type == 'Atom' && in_array( $el, $this->_CONTENT_CONSTRUCTS ) ) {
	183	+ // avoid clashing w/ RSS mod_content
	184	+ if ( $el == 'content' ) {
	185	+ $el = 'atom_content';
	186	+ }
	187	+
	188	+ $this->incontent = $el;
	189	+ }
	190	+
	191	+ // if inside an Atom content construct (e.g. content or summary) field treat tags as text
	192	+ elseif ( $this->feed_type == 'Atom' && $this->incontent ) {
	193	+ // if tags are inlined, then flatten
	194	+ $attrs_str = join(
	195	+ ' ',
	196	+ array_map(
	197	+ array( 'MagpieRSS', 'mapAttributes' ),
	198	+ array_keys( $attrs ),
	199	+ array_values( $attrs )
	200	+ )
	201	+ );
	202	+
	203	+ $this->append_content( "<$element $attrs_str>" );
	204	+
	205	+ array_unshift( $this->stack, $el );
	206	+ }
	207	+
	208	+ // Atom support many links per containging element.
	209	+ // Magpie treats link elements of type rel='alternate'
	210	+ // as being equivalent to RSS's simple link element.
	211	+ elseif ( $this->feed_type == 'Atom' && $el == 'link' ) {
	212	+ if ( isset( $attrs['rel'] ) && $attrs['rel'] == 'alternate' ) {
	213	+ $link_el = 'link';
	214	+ } else {
	215	+ $link_el = 'link_' . $attrs['rel'];
	216	+ }
	217	+
	218	+ $this->append( $link_el, $attrs['href'] );
	219	+ } else { // set stack[0] to current element
	220	+ array_unshift( $this->stack, $el );
	221	+ }
	222	+ }
	223	+
	224	+ function feed_cdata( $p, $text ) {
	225	+ if ( $this->feed_type == 'Atom' && $this->incontent ) {
	226	+ $this->append_content( $text );
	227	+ } else {
	228	+ $current_el = join( '_', array_reverse( $this->stack ) );
	229	+ $this->append( $current_el, $text );
	230	+ }
	231	+ }
	232	+
	233	+ function feed_end_element( $p, $el ) {
	234	+ $el = strtolower( $el );
	235	+
	236	+ if ( $el == 'item' \|\| $el == 'entry' ) {
	237	+ $this->items[] = $this->current_item;
	238	+ $this->current_item = array();
	239	+ $this->initem = false;
	240	+ } elseif ( $this->feed_type == 'RSS' && $this->current_namespace == '' && $el == 'textinput' ) {
	241	+ $this->intextinput = false;
	242	+ } elseif ( $this->feed_type == 'RSS' && $this->current_namespace == '' && $el == 'image' ) {
	243	+ $this->inimage = false;
	244	+ } elseif ( $this->feed_type == 'Atom' && in_array( $el, $this->_CONTENT_CONSTRUCTS ) ) {
	245	+ $this->incontent = false;
	246	+ } elseif ( $el == 'channel' \|\| $el == 'feed' ) {
	247	+ $this->inchannel = false;
	248	+ } elseif ( $this->feed_type == 'Atom' && $this->incontent ) {
	249	+ // balance tags properly
	250	+ // note: I don't think this is actually neccessary
	251	+ if ( $this->stack[0] == $el ) {
	252	+ $this->append_content( "</$el>" );
	253	+ } else {
	254	+ $this->append_content( "<$el />" );
	255	+ }
	256	+
	257	+ array_shift( $this->stack );
	258	+ } else {
	259	+ array_shift( $this->stack );
	260	+ }
	261	+
	262	+ $this->current_namespace = false;
	263	+ }
	264	+
	265	+ function concat( &$str1, $str2 = '' ) {
	266	+ if ( !isset( $str1 ) ) {
	267	+ $str1 = '';
	268	+ }
	269	+ $str1 .= $str2;
	270	+ }
	271	+
	272	+ function append_content( $text ) {
	273	+ if ( $this->initem ) {
	274	+ $this->concat( $this->current_item[$this->incontent], $text );
	275	+ } elseif ( $this->inchannel ) {
	276	+ $this->concat( $this->channel[$this->incontent], $text );
	277	+ }
	278	+ }
	279	+
	280	+ // smart append - field and namespace aware
	281	+ function append( $el, $text ) {
	282	+ if ( !$el ) {
	283	+ return;
	284	+ }
	285	+ if ( $this->current_namespace ) {
	286	+ if ( $this->initem ) {
	287	+ $this->concat(
	288	+ $this->current_item[$this->current_namespace][$el], $text
	289	+ );
	290	+ } elseif ( $this->inchannel ) {
	291	+ $this->concat(
	292	+ $this->channel[$this->current_namespace][$el], $text
	293	+ );
	294	+ } elseif ( $this->intextinput ) {
	295	+ $this->concat(
	296	+ $this->textinput[$this->current_namespace][$el], $text
	297	+ );
	298	+ } elseif ( $this->inimage ) {
	299	+ $this->concat(
	300	+ $this->image[$this->current_namespace][$el], $text
	301	+ );
	302	+ }
	303	+ } else {
	304	+ if ( $this->initem ) {
	305	+ $this->concat(
	306	+ $this->current_item[$el], $text
	307	+ );
	308	+ } elseif ( $this->intextinput ) {
	309	+ $this->concat(
	310	+ $this->textinput[$el], $text
	311	+ );
	312	+ } elseif ( $this->inimage ) {
	313	+ $this->concat(
	314	+ $this->image[$el], $text
	315	+ );
	316	+ } elseif ( $this->inchannel ) {
	317	+ $this->concat(
	318	+ $this->channel[$el], $text
	319	+ );
	320	+ }
	321	+ }
	322	+ }
	323	+
	324	+ function normalize() {
	325	+ // if atom populate rss fields
	326	+ if ( $this->is_atom() ) {
	327	+ $this->channel['description'] = $this->channel['tagline'];
	328	+ for ( $i = 0; $i < count( $this->items ); $i++ ) {
	329	+ $item = $this->items[$i];
	330	+ if ( isset( $item['summary'] ) ) {
	331	+ $item['description'] = $item['summary'];
	332	+ }
	333	+ if ( isset( $item['atom_content'] ) ) {
	334	+ $item['content']['encoded'] = $item['atom_content'];
	335	+ }
	336	+
	337	+ $atom_date = ( isset( $item['issued'] ) ) ? $item['issued'] : $item['modified'];
	338	+ if ( $atom_date ) {
	339	+ $epoch = @$this->parse_w3cdtf( $atom_date );
	340	+ if ( $epoch && $epoch > 0 ) {
	341	+ $item['date_timestamp'] = $epoch;
	342	+ }
	343	+ }
	344	+
	345	+ $this->items[$i] = $item;
	346	+ }
	347	+ } elseif ( $this->is_rss() ) {
	348	+ $this->channel['tagline'] = $this->channel['description'];
	349	+ for ( $i = 0; $i < count( $this->items ); $i++ ) {
	350	+ $item = $this->items[$i];
	351	+ if ( isset( $item['description'] ) ) {
	352	+ $item['summary'] = $item['description'];
	353	+ }
	354	+ if ( isset( $item['content']['encoded'] ) ) {
	355	+ $item['atom_content'] = $item['content']['encoded'];
	356	+ }
	357	+
	358	+ if ( $this->is_rss() == '1.0' && isset( $item['dc']['date'] ) ) {
	359	+ $epoch = @$this->parse_w3cdtf( $item['dc']['date'] );
	360	+ if ( $epoch && $epoch > 0 ) {
	361	+ $item['date_timestamp'] = $epoch;
	362	+ }
	363	+ } elseif ( isset( $item['pubdate'] ) ) {
	364	+ $epoch = @strtotime( $item['pubdate'] );
	365	+ if ( $epoch > 0 ) {
	366	+ $item['date_timestamp'] = $epoch;
	367	+ }
	368	+ }
	369	+
	370	+ $this->items[$i] = $item;
	371	+ }
	372	+ }
	373	+ }
	374	+
	375	+ function is_rss() {
	376	+ if ( $this->feed_type == 'RSS' ) {
	377	+ return $this->feed_version;
	378	+ } else {
	379	+ return false;
	380	+ }
	381	+ }
	382	+
	383	+ function is_atom() {
	384	+ if ( $this->feed_type == 'Atom' ) {
	385	+ return $this->feed_version;
	386	+ } else {
	387	+ return false;
	388	+ }
	389	+ }
	390	+
	391	+ /**
	392	+ * Instantiate an XML parser.
	393	+ * @return XML parser, and possibly re-encoded source
	394	+ */
	395	+ function create_parser( $source, $out_enc, $in_enc, $detect ) {
	396	+ // by default PHP5 does a fine job of detecting input encodings
	397	+ if( !$detect && $in_enc ) {
	398	+ $parser = xml_parser_create( $in_enc );
	399	+ } else {
	400	+ $parser = xml_parser_create( '' );
	401	+ }
	402	+ if ( $out_enc ) {
	403	+ $this->encoding = $out_enc;
	404	+ xml_parser_set_option(
	405	+ $parser,
	406	+ XML_OPTION_TARGET_ENCODING,
	407	+ $out_enc
	408	+ );
	409	+ }
	410	+
	411	+ return array( $parser, $source );
	412	+ }
	413	+
	414	+ /**
	415	+ * Checks if $enc is an encoding type supported by MagpieRSS.
	416	+ * @param $enc String: encoding name
	417	+ * @return String or false
	418	+ */
	419	+ function known_encoding( $enc ) {
	420	+ $enc = strtoupper( $enc );
	421	+ if ( in_array( $enc, $this->_KNOWN_ENCODINGS ) ) {
	422	+ return $enc;
	423	+ } else {
	424	+ return false;
	425	+ }
	426	+ }
	427	+
	428	+ function error( $errormsg, $lvl = E_USER_WARNING ) {
	429	+ // append PHP's error message if track_errors is enabled
	430	+ if ( isset( $php_errormsg ) ) {
	431	+ $errormsg .= " ($php_errormsg)";
	432	+ }
	433	+
	434	+ $notices = E_USER_NOTICE\|E_NOTICE;
	435	+ if ( $lvl&$notices ) {
	436	+ $this->WARNING = $errormsg;
	437	+ } else {
	438	+ $this->ERROR = $errormsg;
	439	+ }
	440	+ }
	441	+
	442	+ /**
	443	+ * Parse a W3CDTF date into unix epoch.
	444	+ * This used to be in its own file.
	445	+ * @note http://www.w3.org/TR/NOTE-datetime
	446	+ * @param $date_str String: date string to parse
	447	+ * @return Integer
	448	+ */
	449	+ public static function parse_w3cdtf( $date_str ) {
	450	+ // regex to match wc3dtf
	451	+ $pat = "/(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2})(:(\d{2}))?(?:([-+])(\d{2}):?(\d{2})\|(Z))?/";
	452	+
	453	+ if ( preg_match( $pat, $date_str, $match ) ) {
	454	+ list( $year, $month, $day, $hours, $minutes, $seconds ) =
	455	+ array( $match[1], $match[2], $match[3], $match[4], $match[5], $match[6] );
	456	+
	457	+ // calculate epoch for current date assuming GMT
	458	+ $epoch = gmmktime( $hours, $minutes, $seconds, $month, $day, $year );
	459	+
	460	+ $offset = 0;
	461	+ if ( $match[10] == 'Z' ) {
	462	+ // zulu time, aka GMT
	463	+ } else {
	464	+ list( $tz_mod, $tz_hour, $tz_min ) =
	465	+ array( $match[8], $match[9], $match[10] );
	466	+
	467	+ // zero out the variables
	468	+ if ( !$tz_hour ) {
	469	+ $tz_hour = 0;
	470	+ }
	471	+ if ( !$tz_min ) {
	472	+ $tz_min = 0;
	473	+ }
	474	+
	475	+ $offset_secs = ( ( $tz_hour * 60 ) + $tz_min ) * 60;
	476	+
	477	+ // is timezone ahead of GMT? then subtract offset
	478	+ if ( $tz_mod == '+' ) {
	479	+ $offset_secs = $offset_secs * -1;
	480	+ }
	481	+
	482	+ $offset = $offset_secs;
	483	+ }
	484	+ $epoch = $epoch + $offset;
	485	+ return $epoch;
	486	+ } else {
	487	+ return -1;
	488	+ }
	489	+ }
	490	+
	491	+ public static function mapAttributes( $k, $v ) {
	492	+ return "$k=\"$v\"";
	493	+ }
	494	+
	495	+} // end class MagpieRSS
\ No newline at end of file
Property changes on: trunk/extensions/RSS/RSSParse.php
___________________________________________________________________
Added: svn:eol-style
1	496	+ native
Index: trunk/extensions/RSS/RSS.i18n.php
—	—	@@ -1,6 +1,6 @@
2	2	<?php
3	3	/**
4		~~- * Internationalisation file for extension RSS.~~
	4	+ * Internationalization file for RSS extension.
5	5	*
6	6	* @file
7	7	* @ingroup Extensions
—	—	@@ -8,7 +8,35 @@
9	9
10	10	$messages = array();
11	11
	12	+/** English
	13	+ * @author Łukasz Garczewski (TOR) <tor@wikia-inc.com>
	14	+ */
12	15	$messages['en'] = array(
13	16	'rss-desc' => 'Displays an RSS feed on a wiki page',
	17	+ 'rss-error' => 'Failed to load RSS feed from $1: $2',
	18	+ 'rss-empty' => 'Failed to load RSS feed from $1!',
14	19	);
15	20
	21	+/** Finnish (Suomi)
	22	+ * @author Jack Phoenix <jack@countervandalism.net>
	23	+ */
	24	+$messages['fi'] = array(
	25	+ 'rss-error' => 'RSS-syötteen lataaminen osoitteesta $1 epäonnistui: $2',
	26	+ 'rss-empty' => 'RSS-syötteen lataaminen osoitteesta $1 epäonnistui!',
	27	+);
	28	+
	29	+/** Dutch (Nederlands)
	30	+ * @author Mitchel Corstjens
	31	+ */
	32	+$messages['nl'] = array(
	33	+ 'rss-error' => 'Kon RSS feed van $1 niet laden, fout: $2',
	34	+ 'rss-empty' => 'Kon RSS feed van $1 niet laden!',
	35	+);
	36	+
	37	+/** Polish (Polski)
	38	+ * @author Łukasz Garczewski (TOR) <tor@wikia-inc.com>
	39	+ */
	40	+$messages['pl'] = array(
	41	+ 'rss-error' => 'Nie udało się odczytać kanału $1: $2',
	42	+ 'rss-empty' => 'Nie udało się odczytać kanału $1!',
	43	+);
\ No newline at end of file
Index: trunk/extensions/RSS/RSS.php
—	—	@@ -1,61 +1,57 @@
2	2	<?php
3		-
4	3	/**
5		~~- * RSS-Feed MediaWiki extension.~~
6		~~- * @link http://www.mediawiki.org/wiki/Extension:RSS Documentation~~
	4	+ * RSS-Feed MediaWiki extension
7	5	*
8		~~- * @file RSS.php~~
	6	+ * @file
9	7	* @ingroup Extensions
10		- *
11		~~- * TODO: replace all @ by wfSurpressWarnings and wfResumeWarnings~~
	8	+ * @version 1.7
	9	+ * @author mutante, Daniel Kinzler, Rdb, Mafs, Alxndr, Chris Reigrut, K001
	10	+ * @author Kellan Elliott-McCrea <kellan@protest.net> -- author of MagpieRSS
	11	+ * @author Jeroen De Dauw
	12	+ * @author Jack Phoenix <jack@countervandalism.net>
	13	+ * @copyright © Kellan Elliott-McCrea <kellan@protest.net>
	14	+ * @copyright © mutante, Daniel Kinzler, Rdb, Mafs, Alxndr, Chris Reigrut, K001
	15	+ * @link http://www.mediawiki.org/wiki/Extension:RSS Documentation
12	16	*/
13	17
14		~~-if ( !defined( 'MEDIAWIKI' ) ) {~~
	18	+if( !defined( 'MEDIAWIKI' ) ) {
15	19	die( "This is not a valid entry point.\n" );
16	20	}
17	21
18		~~-define( 'RSS_VERSION', '1.7' );~~
19		-
	22	+// Extension credits that will show up on Special:Version
20	23	$wgExtensionCredits['parserhook'][] = array(
21		~~- 'path' => __FILE__,~~
22	24	'name' => 'RSS feed',
23	25	'author' => array(
	26	+ 'Kellan Elliott-McCrea',
24	27	'mutante',
25		~~- 'Duesentrieb',~~
	28	+ 'Daniel Kinzler',
26	29	'Rdb',
27	30	'Mafs',
28	31	'Alxndr',
29	32	'Wikinaut',
30		~~- 'Cmreigrut',~~
	33	+ 'Chris Reigrut',
31	34	'K001',
32		~~- '[http://www.mediawiki.org/wiki/User:Jeroen_De_Dauw Jeroen De Dauw]'~~
	35	+ 'Jack Phoenix',
	36	+ 'Jeroen De Dauw'
33	37	),
34		~~- 'version' => RSS_VERSION,~~
	38	+ 'version' => '1.7',
35	39	'url' => 'http://www.mediawiki.org/wiki/Extension:RSS',
	40	+ 'description' => 'Displays an RSS feed on a wiki page',
36	41	'descriptionmsg' => 'rss-desc',
37	42	);
38	43
39		~~-$dir = dirname( __FILE__ );~~
40		~~-$wgExtensionMessagesFiles['RSS'] = "$dir/RSS.i18n.php";~~
	44	+// Internationalization file and autoloadable classes
	45	+$dir = dirname( __FILE__ ) . '/';
	46	+$wgExtensionMessagesFiles['RSS'] = $dir . 'RSS.i18n.php';
	47	+$wgAutoloadClasses['MagpieRSS'] = $dir . 'RSSParse.php';
	48	+$wgAutoloadClasses['RSSCache'] = $dir . 'RSSCache.php';
41	49
42		~~-define( 'MAGPIE_OUTPUT_ENCODING', 'UTF-8' );~~
	50	+$wgHooks['ParserFirstCallInit'][] = 'wfRssExtension';
43	51
44		~~-# change this according to your magpie installation!~~
45		~~-require_once( dirname( __FILE__ ) . '/magpierss/rss_fetch.inc' );~~
46		-
47		~~-// Avoid unstubbing $wgParser too early on modern (1.12+) MW versions, as per r35980~~
48		~~-if ( defined( 'MW_SUPPORTS_PARSERFIRSTCALLINIT' ) ) {~~
49		~~- $wgHooks['ParserFirstCallInit'][] = 'wfRssExtension';~~
50		~~-} else {~~
51		~~- $wgExtensionFunctions[] = 'wfRssExtension';~~
52		-}
53		-
54	52	# Extension hook callback function
55		~~-function wfRssExtension() {~~
56		~~- global $wgParser;~~
57		-
	53	+function wfRssExtension( &$parser ) {
58	54	# Install parser hook for <rss> tags
59		~~- $wgParser->setHook( 'rss', 'renderRss' );~~
	55	+ $parser->setHook( 'rss', 'renderRss' );
60	56	return true;
61	57	}
62	58
—	—	@@ -66,90 +62,137 @@
67	63	// Kill parser cache
68	64	$wgParser->disableCache();
69	65
70		~~- if ( !$input ) return ''; # if <rss>-section is empty, return nothing~~
	66	+ if ( !$input ) {
	67	+ return ''; # if <rss>-section is empty, return nothing
	68	+ }
71	69
72		~~- # Parse fields in rss-section~~
	70	+ # Parse fields in rss section
73	71	$fields = explode( '\|', $input );
74		~~- $url = @$fields[0];~~
	72	+ wfSuppressWarnings();
	73	+ $url = $fields[0];
	74	+ wfRestoreWarnings();
75	75
76	76	$args = array();
77	77	for ( $i = 1; $i < sizeof( $fields ); $i++ ) {
78	78	$f = $fields[$i];
79	79
80		~~- if ( strpos( $f, '=' ) === false ) $args[strtolower( trim( $f ) )] = false;~~
81		~~- else {~~
	80	+ if ( strpos( $f, '=' ) === false ) {
	81	+ $args[strtolower( trim( $f ) )] = false;
	82	+ } else {
82	83	list( $k, $v ) = explode( '=', $f, 2 );
83		~~- if ( trim( $v ) == false ) $args[strtolower( trim( $k ) )] = false;~~
84		~~- else $args[strtolower( trim( $k ) )] = trim( $v );~~
	84	+ if ( trim( $v ) == false ) {
	85	+ $args[strtolower( trim( $k ) )] = false;
	86	+ } else {
	87	+ $args[strtolower( trim( $k ) )] = trim( $v );
	88	+ }
85	89	}
86	90	}
87	91
88		~~- # Get charset from argument-array~~
89		~~- $charset = @$args['charset'];~~
90		~~- if ( !$charset ) $charset = $wgOutputEncoding;~~
	92	+ # Get charset from argument array
	93	+ wfSuppressWarnings();
	94	+ $charset = $args['charset'];
	95	+ wfRestoreWarnings();
	96	+ if( !$charset ) {
	97	+ $charset = $wgOutputEncoding;
	98	+ }
	99	+
91	100	# Get max number of headlines from argument-array
92		~~- $maxheads = @$args['max'];~~
	101	+ wfSuppressWarnings();
	102	+ $maxheads = $args['max'];
	103	+ wfRestoreWarnings();
93	104	$headcnt = 0;
94	105
95		~~- # Get short-flag from argument-array~~
	106	+ # Get short flag from argument array
96	107	# If short is set, no description text is printed
97		~~- if ( isset( $args['short'] ) ) $short = true; else $short = false;~~
98		~~- # Get reverse-flag from argument-array~~
99		~~- if ( isset( $args['reverse'] ) ) $reverse = true; else $reverse = false;~~
	108	+ if( isset( $args['short'] ) ) {
	109	+ $short = true;
	110	+ } else {
	111	+ $short = false;
	112	+ }
100	113
101		~~- # Get date format from argument-array~~
102		~~- if ( isset( $args["date"] ) ) {~~
103		~~- $date = @$args["date"];~~
104		~~- if ( $date == '' )~~
	114	+ # Get reverse flag from argument array
	115	+ if( isset( $args['reverse'] ) ) {
	116	+ $reverse = true;
	117	+ } else {
	118	+ $reverse = false;
	119	+ }
	120	+
	121	+ # Get date format from argument array
	122	+ if ( isset( $args['date'] ) ) {
	123	+ wfSuppressWarnings();
	124	+ $date = $args['date'];
	125	+ wfRestoreWarnings();
	126	+ if ( $date == '' ) {
105	127	$date = 'd M Y H:i';
	128	+ }
	129	+ } else {
	130	+ $date = false;
106	131	}
107		~~- else~~
108		~~- $date = false;~~
109	132
110	133	# Get highlight terms from argument array
111		~~- $rssHighlight = @$args['highlight'];~~
	134	+ wfSuppressWarnings();
	135	+ $rssHighlight = $args['highlight'];
	136	+ wfRestoreWarnings();
112	137	$rssHighlight = str_replace( ' ', ' ', $rssHighlight );
113	138	$rssHighlight = explode( ' ', trim( $rssHighlight ) );
114	139
115		~~- # Get filter terms from argument-array~~
116		~~- $rssFilter = @$args['filter'];~~
	140	+ # Get filter terms from argument array
	141	+ wfSuppressWarnings();
	142	+ $rssFilter = $args['filter'];
	143	+ wfRestoreWarnings();
117	144	$rssFilter = str_replace( ' ', ' ', $rssFilter );
118	145	$rssFilter = explode( ' ', trim( $rssFilter ) );
119	146
120	147	# Filterout terms
121		~~- $rssFilterout = @$args['filterout'];~~
	148	+ wfSuppressWarnings();
	149	+ $rssFilterout = $args['filterout'];
	150	+ wfRestoreWarnings();
122	151	$rssFilterout = str_replace( ' ', ' ', $rssFilterout );
123	152	$rssFilterout = explode( ' ', trim( $rssFilterout ) );
124	153
125	154	# Fetch RSS. May be cached locally.
126		~~- # Refer to the documentation of magpie for details.~~
127		~~- $rss = @fetch_rss( $url );~~
	155	+ # Refer to the documentation of MagpieRSS for details.
	156	+ if ( !function_exists( 'fetch_rss' ) ) {
	157	+ include( dirname( __FILE__ ) . '/RSSFetch.php' ); // provides fetch_rss() function
	158	+ }
	159	+ wfSuppressWarnings();
	160	+ $rss = fetch_rss( $url );
	161	+ wfRestoreWarnings();
128	162
129	163	# Check for errors.
	164	+ if ( empty( $rss ) ) {
	165	+ wfLoadExtensionMessages( 'RSS' );
	166	+ return wfMsg( 'rss-empty', $url );
	167	+ }
	168	+
130	169	if ( $rss->ERROR ) {
131		~~- return "<div>Failed to load RSS feed from $url: " . $rss->ERROR . "</div>"; # localize…~~
	170	+ wfLoadExtensionMessages( 'RSS' );
	171	+ return '<div>' . wfMsg( 'rss-error', $url, $rss->ERROR ) . '</div>';
132	172	}
133	173
134	174	if ( !is_array( $rss->items ) ) {
135		~~- return "<div>Failed to load RSS feed from $url!</div>"; # localize…~~
	175	+ wfLoadExtensionMessages( 'RSS' );
	176	+ return '<div>' . wfMsg( 'rss-empty', $url ) . '</div>';
136	177	}
137	178
138	179	# Build title line
139		~~- # $title = iconv($charset, $wgOutputEncoding, $rss->channel['title']);~~
140		~~- # if( $rss->channel['link'] ) $title = "<a href='".$rss->channel['link']."'>$title</a>";~~
	180	+ #$title = iconv( $charset, $wgOutputEncoding, $rss->channel['title'] );
	181	+ #if( $rss->channel['link'] ) $title = "<a href='" . $rss->channel['link'] . "'>$title</a>";
141	182
142	183	$output = '';
143		~~- if ( $reverse ) $rss->items = array_reverse( $rss->items );~~
	184	+ if( $reverse ) {
	185	+ $rss->items = array_reverse( $rss->items );
	186	+ }
144	187	$description = false;
145	188	foreach ( $rss->items as $item ) {
146		~~- if ( $item['description'] ) {~~
	189	+ if ( isset( $item['description'] ) && $item['description'] ) {
147	190	$description = true;
148	191	break;
149	192	}
150	193	}
151	194
152	195	# Build items
153		~~- if ( !$short and $description ) { # full item list~~
	196	+ if ( !$short && $description ) { # full item list
154	197	$output .= '<dl>';
155	198
156	199	foreach ( $rss->items as $item ) {
—	—	@@ -171,7 +214,7 @@
172	215	# Build description text if desired
173	216	if ( $item['description'] ) {
174	217	$text = trim( iconv( $charset, $wgOutputEncoding, $item['description'] ) );
175		~~- # Avoid pre-tags~~
	218	+ # Avoid <pre> tags
176	219	$text = str_replace( "\r", ' ', $text );
177	220	$text = str_replace( "\n", ' ', $text );
178	221	$text = str_replace( "\t", ' ', $text );
—	—	@@ -180,23 +223,29 @@
181	224	$d_text = wfRssFilter( $text, $rssFilter );
182	225	$d_text = wfRssFilterout( $text, $rssFilterout );
183	226	$text = wfRssHighlight( $text, $rssHighlight );
184		~~- $display = $d_text or $d_title;~~
	227	+ $display = $d_text \|\| $d_title;
185	228	} else {
186	229	$text = '';
187	230	$display = $d_title;
188	231	}
189	232	if ( $display ) {
190		~~- $output .= "<dt><a href='$href'><b>$title</b></a></dt>";~~
191		~~- if ( $date ) $output .= " ($pubdate)";~~
192		~~- if ( $text ) $output .= "<dd>$text <b>[<a href='$href'>?</a>]</b></dd>";~~
	233	+ $output.= "<dt><a href='$href'><b>$title</b></a></dt>";
	234	+ if ( $date ) {
	235	+ $output .= " ($pubdate)";
	236	+ }
	237	+ if ( $text ) {
	238	+ $output .= "<dd>$text <b>[<a href='$href'>?</a>]</b></dd>";
	239	+ }
193	240	}
194	241	# Cut off output when maxheads is reached:
195		~~- if ( ++$headcnt == $maxheads ) break;~~
	242	+ if ( ++$headcnt == $maxheads ) {
	243	+ break;
	244	+ }
196	245	}
197	246
198	247	$output .= '</dl>';
199	248	} else { # short item list
200		~~- # # HACKY HACKY HACKY~~
	249	+ ## HACKY HACKY HACKY
201	250	$output .= '<ul>';
202	251	$displayed = array();
203	252	foreach ( $rss->items as $item ) {
—	—	@@ -205,26 +254,29 @@
206	255	$d_title = wfRssFilter( $title, $rssFilter ) && wfRssFilterout( $title, $rssFilterout );
207	256	$title = wfRssHighlight( $title, $rssHighlight );
208	257	if ( $date ) {
209		~~- $pubdate = trim( iconv( $charset, $wgOutputEncoding, $item['pubdate'] ) );~~
	258	+ $pubdate = isset( $item['pubdate'] ) ? trim( iconv( $charset, $wgOutputEncoding, $item['pubdate'] ) ) : '';
210	259	if ( $pubdate == '' ) {
211	260	$pubdate = trim( iconv( $charset, $wgOutputEncoding, $item['dc']['date'] ) );
212		~~- }~~
	261	+ }
213	262	$pubdate = date( $date, strtotime( $pubdate ) );
214	263	}
	264	+
215	265	if ( $d_title && !in_array( $title, $displayed ) ) {
216	266	// Add date to ouput if specified
217	267	$output .= '<li><a href="' . $href . '" title="' . $title . '">' . $title . '</a>';
218		~~- if ( $date ) {~~
	268	+ if( $date ) {
219	269	$output .= " ($pubdate)";
220	270	}
221	271	$output .= '</li>';
222	272
223	273	$displayed[] = $title;
224	274	# Cut off output when maxheads is reached:
225		~~- if ( ++$headcnt == $maxheads ) break;~~
	275	+ if ( ++$headcnt == $maxheads ) {
	276	+ break;
	277	+ }
226	278	}
227	279	}
228		~~- $output .= '</ul>';~~
	280	+ $output.= '</ul>';
229	281	}
230	282
231	283	return $output;
—	—	@@ -233,7 +285,7 @@
234	286	function wfRssFilter( $text, $rssFilter ) {
235	287	$display = true;
236	288	if ( is_array( $rssFilter ) ) {
237		~~- foreach ( $rssFilter as $term ) {~~
	289	+ foreach( $rssFilter as $term ) {
238	290	if ( $term ) {
239	291	$display = false;
240	292	if ( preg_match( "\|$term\|i", $text, $a ) ) {
—	—	@@ -241,7 +293,9 @@
242	294	return $display;
243	295	}
244	296	}
245		~~- if ( $display ) break;~~
	297	+ if ( $display ) {
	298	+ break;
	299	+ }
246	300	}
247	301	}
248	302	return $display;
—	—	@@ -275,11 +329,13 @@
276	330	$count_color = count( $color );
277	331
278	332	if ( is_array( $rssHighlight ) ) {
279		~~- foreach ( $rssHighlight as $term ) {~~
	333	+ foreach( $rssHighlight as $term ) {
280	334	if ( $term ) {
281	335	$text = preg_replace( "\|\b(\w?" . $term . "\w?)\b\|i", "$starttag" . "_" . $i . "\\1$endtag", $text );
282	336	$i++;
283		~~- if ( $i == $count_color ) $i = 0;~~
	337	+ if ( $i == $count_color ) {
	338	+ $i = 0;
	339	+ }
284	340	}
285	341	}
286	342	}
—	—	@@ -291,5 +347,4 @@
292	348	}
293	349
294	350	return $text;
295		-}
296		~~-# PHP closing tag intentionally left blank~~
\ No newline at end of file
	351	+}
\ No newline at end of file

Comments

#Comment by 😂 (talk | contribs) 00:30, 22 January 2011

Full review was on r75046, which resolved some issues since this commit.

Status & tagging log

00:30, 22 January 2011 😂 (talk | contribs) changed the status of r69793 [removed: new added: resolved]