magpie: add latest dev. build of 0.8x from cvs
This commit is contained in:
parent
1f8274d3c1
commit
489ffe5a21
1 changed files with 434 additions and 134 deletions
|
@ -5,7 +5,7 @@
|
|||
* File: rss_parse.inc - parse an RSS or Atom feed
|
||||
* return as a simple object.
|
||||
*
|
||||
* Handles RSS 0.9x, RSS 2.0, RSS 1.0, and Atom 0.3
|
||||
* Handles RSS 0.9x, RSS 2.0, RSS 1.0, Atom 0.3, and Atom 1.0
|
||||
*
|
||||
* The lastest version of MagpieRSS can be obtained from:
|
||||
* http://magpierss.sourceforge.net
|
||||
|
@ -15,7 +15,7 @@
|
|||
* magpierss-general@lists.sourceforge.net
|
||||
*
|
||||
* @author Kellan Elliott-McCrea <kellan@protest.net>
|
||||
* @version 0.7a
|
||||
* @version 0.8
|
||||
* @license GPL
|
||||
*
|
||||
*/
|
||||
|
@ -50,19 +50,26 @@ class MagpieRSS {
|
|||
|
||||
// define some constants
|
||||
|
||||
var $_CONTENT_CONSTRUCTS = array('content', 'summary', 'info', 'title', 'tagline', 'copyright');
|
||||
var $_ATOM_CONTENT_CONSTRUCTS = array(
|
||||
'content', 'summary', 'title', /* common */
|
||||
'info', 'tagline', 'copyright', /* Atom 0.3 */
|
||||
'rights', 'subtitle', /* Atom 1.0 */
|
||||
);
|
||||
var $_XHTML_CONTENT_CONSTRUCTS = array('body', 'div');
|
||||
var $_KNOWN_ENCODINGS = array('UTF-8', 'US-ASCII', 'ISO-8859-1');
|
||||
|
||||
// parser variables, useless if you're not a parser, treat as private
|
||||
var $stack = array(); // parser stack
|
||||
var $inchannel = false;
|
||||
var $initem = false;
|
||||
var $incontent = false; // if in Atom <content mode="xml"> field
|
||||
|
||||
var $incontent = array(); // non-empty if in namespaced XML content field
|
||||
var $exclude_top = false; // true when Atom 1.0 type="xhtml"
|
||||
|
||||
var $intextinput = false;
|
||||
var $inimage = false;
|
||||
var $current_namespace = false;
|
||||
|
||||
|
||||
/**
|
||||
* Set up XML parser, parse source, and return populated RSS object..
|
||||
*
|
||||
|
@ -122,8 +129,6 @@ class MagpieRSS {
|
|||
xml_set_element_handler($this->parser,
|
||||
'feed_start_element', 'feed_end_element' );
|
||||
|
||||
$source = preg_replace("/&(?![^ ][^ ]*;)/","&",$source);
|
||||
|
||||
xml_set_character_data_handler( $this->parser, 'feed_cdata' );
|
||||
|
||||
$status = xml_parse( $this->parser, $source );
|
||||
|
@ -150,13 +155,16 @@ class MagpieRSS {
|
|||
$attrs = array_change_key_case($attrs, CASE_LOWER);
|
||||
|
||||
// check for a namespace, and split if found
|
||||
$ns = false;
|
||||
if ( strpos( $element, ':' ) ) {
|
||||
list($ns, $el) = split( ':', $element, 2);
|
||||
}
|
||||
if ( $ns and $ns != 'rdf' ) {
|
||||
$this->current_namespace = $ns;
|
||||
}
|
||||
// Don't munge content tags
|
||||
if ( empty($this->incontent) ) {
|
||||
$ns = false;
|
||||
if ( strpos( $element, ':' ) ) {
|
||||
list($ns, $el) = split( ':', $element, 2);
|
||||
}
|
||||
if ( $ns and $ns != 'rdf' ) {
|
||||
$this->current_namespace = $ns;
|
||||
}
|
||||
}
|
||||
|
||||
# if feed type isn't set, then this is first element of feed
|
||||
# identify feed from root element
|
||||
|
@ -172,16 +180,39 @@ class MagpieRSS {
|
|||
}
|
||||
elseif ( $el == 'feed' ) {
|
||||
$this->feed_type = ATOM;
|
||||
$this->feed_version = $attrs['version'];
|
||||
if ($attrs['xmlns'] == 'http://www.w3.org/2005/Atom') { // Atom 1.0
|
||||
$this->feed_version = '1.0';
|
||||
}
|
||||
else { // Atom 0.3, probably.
|
||||
$this->feed_version = $attrs['version'];
|
||||
}
|
||||
$this->inchannel = true;
|
||||
}
|
||||
return;
|
||||
}
|
||||
|
||||
if ( $el == 'channel' )
|
||||
// if we're inside a namespaced content construct, treat tags as text
|
||||
if ( !empty($this->incontent) )
|
||||
{
|
||||
if ((count($this->incontent) > 1) or !$this->exclude_top) {
|
||||
// if tags are inlined, then flatten
|
||||
$attrs_str = join(' ',
|
||||
array_map('map_attrs',
|
||||
array_keys($attrs),
|
||||
array_values($attrs) )
|
||||
);
|
||||
|
||||
if (strlen($attrs_str) > 0) { $attrs_str = ' '.$attrs_str; }
|
||||
|
||||
$this->append_content( "<{$element}{$attrs_str}>" );
|
||||
}
|
||||
array_push($this->incontent, $el); // stack for parsing content XML
|
||||
}
|
||||
|
||||
elseif ( $el == 'channel' ) {
|
||||
$this->inchannel = true;
|
||||
}
|
||||
|
||||
elseif ($el == 'item' or $el == 'entry' )
|
||||
{
|
||||
$this->initem = true;
|
||||
|
@ -208,60 +239,84 @@ class MagpieRSS {
|
|||
$this->inimage = true;
|
||||
}
|
||||
|
||||
# handle atom content constructs
|
||||
elseif ( $this->feed_type == ATOM and in_array($el, $this->_CONTENT_CONSTRUCTS) )
|
||||
{
|
||||
// avoid clashing w/ RSS mod_content
|
||||
if ($el == 'content' ) {
|
||||
$el = 'atom_content';
|
||||
}
|
||||
|
||||
$this->incontent = $el;
|
||||
|
||||
|
||||
}
|
||||
|
||||
// if inside an Atom content construct (e.g. content or summary) field treat tags as text
|
||||
elseif ($this->feed_type == ATOM and $this->incontent )
|
||||
{
|
||||
// if tags are inlined, then flatten
|
||||
$attrs_str = join(' ',
|
||||
array_map('map_attrs',
|
||||
array_keys($attrs),
|
||||
array_values($attrs) ) );
|
||||
|
||||
$this->append_content( "<$element $attrs_str>" );
|
||||
|
||||
array_unshift( $this->stack, $el );
|
||||
}
|
||||
|
||||
// Atom support many links per containging element.
|
||||
// Magpie treats link elements of type rel='alternate'
|
||||
// as being equivalent to RSS's simple link element.
|
||||
//
|
||||
elseif ($this->feed_type == ATOM and $el == 'link' )
|
||||
{
|
||||
if ( isset($attrs['rel']) and $attrs['rel'] == 'alternate' )
|
||||
{
|
||||
$link_el = 'link';
|
||||
}
|
||||
else {
|
||||
$link_el = 'link_' . $attrs['rel'];
|
||||
}
|
||||
|
||||
$this->append($link_el, $attrs['href']);
|
||||
}
|
||||
// set stack[0] to current element
|
||||
else {
|
||||
array_unshift($this->stack, $el);
|
||||
}
|
||||
// Atom support many links per containing element.
|
||||
// Magpie treats link elements of type rel='alternate'
|
||||
// as being equivalent to RSS's simple link element.
|
||||
|
||||
$atom_link = false;
|
||||
if ($this->feed_type == ATOM and $el == 'link') {
|
||||
$atom_link = true;
|
||||
if (isset($attrs['rel']) and $attrs['rel'] != 'alternate') {
|
||||
$el = $el . "_" . $attrs['rel']; // pseudo-element names for Atom link elements
|
||||
}
|
||||
}
|
||||
# handle atom content constructs
|
||||
elseif ( $this->feed_type == ATOM and in_array($el, $this->_ATOM_CONTENT_CONSTRUCTS) )
|
||||
{
|
||||
// avoid clashing w/ RSS mod_content
|
||||
if ($el == 'content' ) {
|
||||
$el = 'atom_content';
|
||||
}
|
||||
|
||||
// assume that everything accepts namespaced XML
|
||||
// (that will pass through some non-validating feeds;
|
||||
// but so what? this isn't a validating parser)
|
||||
$this->incontent = array();
|
||||
array_push($this->incontent, $el); // start a stack
|
||||
|
||||
if ( isset($attrs['type']) and trim(strtolower($attrs['type']))=='xhtml') {
|
||||
$this->exclude_top = true;
|
||||
} else {
|
||||
$this->exclude_top = false;
|
||||
}
|
||||
}
|
||||
# Handle inline XHTML body elements --CWJ
|
||||
elseif (($this->current_namespace=='xhtml' or
|
||||
(isset($attrs['xmlns']) and $attrs['xmlns'] == 'http://www.w3.org/1999/xhtml'))
|
||||
and in_array($el, $this->_XHTML_CONTENT_CONSTRUCTS) )
|
||||
{
|
||||
$this->current_namespace = 'xhtml';
|
||||
$this->incontent = array();
|
||||
array_push($this->incontent, $el); // start a stack
|
||||
$this->exclude_top = false;
|
||||
}
|
||||
|
||||
array_unshift($this->stack, $el);
|
||||
$elpath = join('_', array_reverse($this->stack));
|
||||
|
||||
$n = $this->element_count($elpath);
|
||||
$this->element_count($elpath, $n+1);
|
||||
|
||||
if ($n > 0) {
|
||||
array_shift($this->stack);
|
||||
array_unshift($this->stack, $el.'#'.($n+1));
|
||||
$elpath = join('_', array_reverse($this->stack));
|
||||
}
|
||||
|
||||
// this makes the baby Jesus cry, but we can't do it in normalize()
|
||||
// because we've made the element name for Atom links unpredictable
|
||||
// by tacking on the relation to the end. -CWJ
|
||||
if ($atom_link and isset($attrs['href'])) {
|
||||
$this->append($elpath, $attrs['href']);
|
||||
}
|
||||
|
||||
// add attributes
|
||||
if (count($attrs) > 0) {
|
||||
$this->append($elpath.'@', join(',', array_keys($attrs)));
|
||||
foreach ($attrs as $attr => $value) {
|
||||
$this->append($elpath.'@'.$attr, $value);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
|
||||
function feed_cdata ($p, $text) {
|
||||
if ($this->feed_type == ATOM and $this->incontent)
|
||||
{
|
||||
|
||||
if ($this->incontent) {
|
||||
$this->append_content( $text );
|
||||
}
|
||||
else {
|
||||
|
@ -273,13 +328,40 @@ class MagpieRSS {
|
|||
function feed_end_element ($p, $el) {
|
||||
$el = strtolower($el);
|
||||
|
||||
if ( $el == 'item' or $el == 'entry' )
|
||||
if ( $this->incontent ) {
|
||||
$opener = array_pop($this->incontent);
|
||||
|
||||
// Don't get bamboozled by namespace voodoo
|
||||
if (strpos($el, ':')) { list($ns, $closer) = split(':', $el); }
|
||||
else { $ns = false; $closer = $el; }
|
||||
|
||||
// Don't get bamboozled by our munging of <atom:content>, either
|
||||
if ($this->feed_type == ATOM and $closer == 'content') {
|
||||
$closer = 'atom_content';
|
||||
}
|
||||
|
||||
// balance tags properly
|
||||
// note: i don't think this is actually neccessary
|
||||
if ($opener != $closer) {
|
||||
array_push($this->incontent, $opener);
|
||||
$this->append_content("<$el />");
|
||||
} elseif ($this->incontent) { // are we in the content construct still?
|
||||
if ((count($this->incontent) > 1) or !$this->exclude_top) {
|
||||
$this->append_content("</$el>");
|
||||
}
|
||||
} else { // shift the opening of the content construct off the normal stack
|
||||
array_shift( $this->stack );
|
||||
}
|
||||
}
|
||||
elseif ( $el == 'item' or $el == 'entry' )
|
||||
{
|
||||
$this->items[] = $this->current_item;
|
||||
$this->current_item = array();
|
||||
$this->initem = false;
|
||||
|
||||
$this->current_category = 0;
|
||||
}
|
||||
elseif ($this->feed_type == RSS and $this->current_namespace == '' and $el == 'textinput' )
|
||||
elseif ($this->feed_type == RSS and $this->current_namespace == '' and $el == 'textinput' )
|
||||
{
|
||||
$this->intextinput = false;
|
||||
}
|
||||
|
@ -287,33 +369,18 @@ class MagpieRSS {
|
|||
{
|
||||
$this->inimage = false;
|
||||
}
|
||||
elseif ($this->feed_type == ATOM and in_array($el, $this->_CONTENT_CONSTRUCTS) )
|
||||
{
|
||||
$this->incontent = false;
|
||||
}
|
||||
elseif ($el == 'channel' or $el == 'feed' )
|
||||
{
|
||||
$this->inchannel = false;
|
||||
}
|
||||
elseif ($this->feed_type == ATOM and $this->incontent ) {
|
||||
// balance tags properly
|
||||
// note: i don't think this is actually neccessary
|
||||
if ( $this->stack[0] == $el )
|
||||
{
|
||||
$this->append_content("</$el>");
|
||||
}
|
||||
else {
|
||||
$this->append_content("<$el />");
|
||||
}
|
||||
|
||||
array_shift( $this->stack );
|
||||
}
|
||||
else {
|
||||
array_shift( $this->stack );
|
||||
array_shift( $this->stack );
|
||||
}
|
||||
|
||||
if ( !$this->incontent ) { // Don't munge the namespace after finishing with elements in namespaced content constructs -CWJ
|
||||
$this->current_namespace = false;
|
||||
}
|
||||
}
|
||||
|
||||
function concat (&$str1, $str2="") {
|
||||
if (!isset($str1) ) {
|
||||
|
@ -322,16 +389,22 @@ class MagpieRSS {
|
|||
$str1 .= $str2;
|
||||
}
|
||||
|
||||
|
||||
|
||||
function append_content($text) {
|
||||
if ( $this->initem ) {
|
||||
$this->concat( $this->current_item[ $this->incontent ], $text );
|
||||
if ( $this->initem ) {
|
||||
if ($this->current_namespace) {
|
||||
$this->concat( $this->current_item[$this->current_namespace][ reset($this->incontent) ], $text );
|
||||
} else {
|
||||
$this->concat( $this->current_item[ reset($this->incontent) ], $text );
|
||||
}
|
||||
elseif ( $this->inchannel ) {
|
||||
$this->concat( $this->channel[ $this->incontent ], $text );
|
||||
}
|
||||
elseif ( $this->inchannel ) {
|
||||
if ($this->current_namespace) {
|
||||
$this->concat( $this->channel[$this->current_namespace][ reset($this->incontent) ], $text );
|
||||
} else {
|
||||
$this->concat( $this->channel[ reset($this->incontent) ], $text );
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// smart append - field and namespace aware
|
||||
function append($el, $text) {
|
||||
|
@ -341,13 +414,13 @@ class MagpieRSS {
|
|||
if ( $this->current_namespace )
|
||||
{
|
||||
if ( $this->initem ) {
|
||||
$this->concat(
|
||||
$this->current_item[ $this->current_namespace ][ $el ], $text);
|
||||
$this->concat(
|
||||
$this->current_item[ $this->current_namespace ][ $el ], $text);
|
||||
}
|
||||
elseif ($this->inchannel) {
|
||||
$this->concat(
|
||||
$this->channel[ $this->current_namespace][ $el ], $text );
|
||||
}
|
||||
$this->concat(
|
||||
$this->channel[ $this->current_namespace][ $el ], $text );
|
||||
}
|
||||
elseif ($this->intextinput) {
|
||||
$this->concat(
|
||||
$this->textinput[ $this->current_namespace][ $el ], $text );
|
||||
|
@ -359,8 +432,8 @@ class MagpieRSS {
|
|||
}
|
||||
else {
|
||||
if ( $this->initem ) {
|
||||
$this->concat(
|
||||
$this->current_item[ $el ], $text);
|
||||
$this->concat(
|
||||
$this->current_item[ $el ], $text);
|
||||
}
|
||||
elseif ($this->intextinput) {
|
||||
$this->concat(
|
||||
|
@ -371,44 +444,267 @@ class MagpieRSS {
|
|||
$this->image[ $el ], $text );
|
||||
}
|
||||
elseif ($this->inchannel) {
|
||||
$this->concat(
|
||||
$this->channel[ $el ], $text );
|
||||
$this->concat(
|
||||
$this->channel[ $el ], $text );
|
||||
}
|
||||
|
||||
}
|
||||
}
|
||||
|
||||
function normalize () {
|
||||
// if atom populate rss fields
|
||||
if ( $this->is_atom() ) {
|
||||
$this->channel['description'] = $this->channel['tagline'];
|
||||
for ( $i = 0; $i < count($this->items); $i++) {
|
||||
$item = $this->items[$i];
|
||||
if ( isset($item['summary']) )
|
||||
$item['description'] = $item['summary'];
|
||||
if ( isset($item['atom_content']))
|
||||
$item['content']['encoded'] = $item['atom_content'];
|
||||
// smart count - field and namespace aware
|
||||
function element_count ($el, $set = NULL) {
|
||||
if (!$el) {
|
||||
return;
|
||||
}
|
||||
if ( $this->current_namespace )
|
||||
{
|
||||
if ( $this->initem ) {
|
||||
if (!is_null($set)) { $this->current_item[ $this->current_namespace ][ $el.'#' ] = $set; }
|
||||
$ret = (isset($this->current_item[ $this->current_namespace ][ $el.'#' ]) ?
|
||||
$this->current_item[ $this->current_namespace ][ $el.'#' ] : 0);
|
||||
}
|
||||
elseif ($this->inchannel) {
|
||||
if (!is_null($set)) { $this->channel[ $this->current_namespace ][ $el.'#' ] = $set; }
|
||||
$ret = (isset($this->channel[ $this->current_namespace][ $el.'#' ]) ?
|
||||
$this->channel[ $this->current_namespace][ $el.'#' ] : 0);
|
||||
}
|
||||
}
|
||||
else {
|
||||
if ( $this->initem ) {
|
||||
if (!is_null($set)) { $this->current_item[ $el.'#' ] = $set; }
|
||||
$ret = (isset($this->current_item[ $el.'#' ]) ?
|
||||
$this->current_item[ $el.'#' ] : 0);
|
||||
}
|
||||
elseif ($this->inchannel) {
|
||||
if (!is_null($set)) {$this->channel[ $el.'#' ] = $set; }
|
||||
$ret = (isset($this->channel[ $el.'#' ]) ?
|
||||
$this->channel[ $el.'#' ] : 0);
|
||||
}
|
||||
}
|
||||
return $ret;
|
||||
}
|
||||
|
||||
$atom_date = (isset($item['issued']) ) ? $item['issued'] : $item['modified'];
|
||||
if ( $atom_date ) {
|
||||
$epoch = @parse_w3cdtf($atom_date);
|
||||
if ($epoch and $epoch > 0) {
|
||||
$item['date_timestamp'] = $epoch;
|
||||
}
|
||||
function normalize_enclosure (&$source, $from, &$dest, $to, $i) {
|
||||
$id_from = $this->element_id($from, $i);
|
||||
$id_to = $this->element_id($to, $i);
|
||||
if (isset($source["{$id_from}@"])) {
|
||||
foreach (explode(',', $source["{$id_from}@"]) as $attr) {
|
||||
if ($from=='link_enclosure' and $attr=='href') { // from Atom
|
||||
$dest["{$id_to}@url"] = $source["{$id_from}@{$attr}"];
|
||||
$dest["{$id_to}"] = $source["{$id_from}@{$attr}"];
|
||||
}
|
||||
elseif ($from=='enclosure' and $attr=='url') { // from RSS
|
||||
$dest["{$id_to}@href"] = $source["{$id_from}@{$attr}"];
|
||||
$dest["{$id_to}"] = $source["{$id_from}@{$attr}"];
|
||||
}
|
||||
else {
|
||||
$dest["{$id_to}@{$attr}"] = $source["{$id_from}@{$attr}"];
|
||||
}
|
||||
|
||||
$this->items[$i] = $item;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
function normalize_atom_person (&$source, $person, &$dest, $to, $i) {
|
||||
$id = $this->element_id($person, $i);
|
||||
$id_to = $this->element_id($to, $i);
|
||||
|
||||
// Atom 0.3 <=> Atom 1.0
|
||||
if ($this->feed_version >= 1.0) { $used = 'uri'; $norm = 'url'; }
|
||||
else { $used = 'url'; $norm = 'uri'; }
|
||||
|
||||
if (isset($source["{$id}_{$used}"])) {
|
||||
$dest["{$id_to}_{$norm}"] = $source["{$id}_{$used}"];
|
||||
}
|
||||
|
||||
// Atom to RSS 2.0 and Dublin Core
|
||||
// RSS 2.0 person strings should be valid e-mail addresses if possible.
|
||||
if (isset($source["{$id}_email"])) {
|
||||
$rss_author = $source["{$id}_email"];
|
||||
}
|
||||
if (isset($source["{$id}_name"])) {
|
||||
$rss_author = $source["{$id}_name"]
|
||||
. (isset($rss_author) ? " <$rss_author>" : '');
|
||||
}
|
||||
if (isset($rss_author)) {
|
||||
$source[$id] = $rss_author; // goes to top-level author or contributor
|
||||
$dest[$id_to] = $rss_author; // goes to dc:creator or dc:contributor
|
||||
}
|
||||
}
|
||||
|
||||
// Normalize Atom 1.0 and RSS 2.0 categories to Dublin Core...
|
||||
function normalize_category (&$source, $from, &$dest, $to, $i) {
|
||||
$cat_id = $this->element_id($from, $i);
|
||||
$dc_id = $this->element_id($to, $i);
|
||||
|
||||
// first normalize category elements: Atom 1.0 <=> RSS 2.0
|
||||
if ( isset($source["{$cat_id}@term"]) ) { // category identifier
|
||||
$source[$cat_id] = $source["{$cat_id}@term"];
|
||||
} elseif ( $this->feed_type == RSS ) {
|
||||
$source["{$cat_id}@term"] = $source[$cat_id];
|
||||
}
|
||||
|
||||
if ( isset($source["{$cat_id}@scheme"]) ) { // URI to taxonomy
|
||||
$source["{$cat_id}@domain"] = $source["{$cat_id}@scheme"];
|
||||
} elseif ( isset($source["{$cat_id}@domain"]) ) {
|
||||
$source["{$cat_id}@scheme"] = $source["{$cat_id}@domain"];
|
||||
}
|
||||
|
||||
// Now put the identifier into dc:subject
|
||||
$dest[$dc_id] = $source[$cat_id];
|
||||
}
|
||||
|
||||
// ... or vice versa
|
||||
function normalize_dc_subject (&$source, $from, &$dest, $to, $i) {
|
||||
$dc_id = $this->element_id($from, $i);
|
||||
$cat_id = $this->element_id($to, $i);
|
||||
|
||||
$dest[$cat_id] = $source[$dc_id]; // RSS 2.0
|
||||
$dest["{$cat_id}@term"] = $source[$dc_id]; // Atom 1.0
|
||||
}
|
||||
|
||||
// simplify the logic for normalize(). Makes sure that count of elements and
|
||||
// each of multiple elements is normalized properly. If you need to mess
|
||||
// with things like attributes or change formats or the like, pass it a
|
||||
// callback to handle each element.
|
||||
function normalize_element (&$source, $from, &$dest, $to, $via = NULL) {
|
||||
if (isset($source[$from]) or isset($source["{$from}#"])) {
|
||||
if (isset($source["{$from}#"])) {
|
||||
$n = $source["{$from}#"];
|
||||
$dest["{$to}#"] = $source["{$from}#"];
|
||||
}
|
||||
else { $n = 1; }
|
||||
|
||||
for ($i = 1; $i <= $n; $i++) {
|
||||
if (isset($via)) { // custom callback for ninja attacks
|
||||
$this->{$via}($source, $from, $dest, $to, $i);
|
||||
}
|
||||
else { // just make it the same
|
||||
$from_id = $this->element_id($from, $i);
|
||||
$to_id = $this->element_id($to, $i);
|
||||
$dest[$to_id] = $source[$from_id];
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
function normalize () {
|
||||
// if atom populate rss fields and normalize 0.3 and 1.0 feeds
|
||||
if ( $this->is_atom() ) {
|
||||
// Atom 1.0 elements <=> Atom 0.3 elements (Thanks, o brilliant wordsmiths of the Atom 1.0 standard!)
|
||||
if ($this->feed_version < 1.0) {
|
||||
$this->normalize_element($this->channel, 'tagline', $this->channel, 'subtitle');
|
||||
$this->normalize_element($this->channel, 'copyright', $this->channel, 'rights');
|
||||
$this->normalize_element($this->channel, 'modified', $this->channel, 'updated');
|
||||
} else {
|
||||
$this->normalize_element($this->channel, 'subtitle', $this->channel, 'tagline');
|
||||
$this->normalize_element($this->channel, 'rights', $this->channel, 'copyright');
|
||||
$this->normalize_element($this->channel, 'updated', $this->channel, 'modified');
|
||||
}
|
||||
$this->normalize_element($this->channel, 'author', $this->channel['dc'], 'creator', 'normalize_atom_person');
|
||||
$this->normalize_element($this->channel, 'contributor', $this->channel['dc'], 'contributor', 'normalize_atom_person');
|
||||
|
||||
// Atom elements to RSS elements
|
||||
$this->normalize_element($this->channel, 'subtitle', $this->channel, 'description');
|
||||
|
||||
if ( isset($this->channel['logo']) ) {
|
||||
$this->normalize_element($this->channel, 'logo', $this->image, 'url');
|
||||
$this->normalize_element($this->channel, 'link', $this->image, 'link');
|
||||
$this->normalize_element($this->channel, 'title', $this->image, 'title');
|
||||
}
|
||||
|
||||
for ( $i = 0; $i < count($this->items); $i++) {
|
||||
$item = $this->items[$i];
|
||||
|
||||
// Atom 1.0 elements <=> Atom 0.3 elements
|
||||
if ($this->feed_version < 1.0) {
|
||||
$this->normalize_element($item, 'modified', $item, 'updated');
|
||||
$this->normalize_element($item, 'issued', $item, 'published');
|
||||
} else {
|
||||
$this->normalize_element($item, 'updated', $item, 'modified');
|
||||
$this->normalize_element($item, 'published', $item, 'issued');
|
||||
}
|
||||
|
||||
// "If an atom:entry element does not contain
|
||||
// atom:author elements, then the atom:author elements
|
||||
// of the contained atom:source element are considered
|
||||
// to apply. In an Atom Feed Document, the atom:author
|
||||
// elements of the containing atom:feed element are
|
||||
// considered to apply to the entry if there are no
|
||||
// atom:author elements in the locations described
|
||||
// above." <http://atompub.org/2005/08/17/draft-ietf-atompub-format-11.html#rfc.section.4.2.1>
|
||||
if (!isset($item["author#"])) {
|
||||
if (isset($item["source_author#"])) { // from aggregation source
|
||||
$source = $item;
|
||||
$author = "source_author";
|
||||
} elseif (isset($this->channel["author#"])) { // from containing feed
|
||||
$source = $this->channel;
|
||||
$author = "author";
|
||||
}
|
||||
|
||||
$item["author#"] = $source["{$author}#"];
|
||||
for ($au = 1; $au <= $item["author#"]; $au++) {
|
||||
$id_to = $this->element_id('author', $au);
|
||||
$id_from = $this->element_id($author, $au);
|
||||
|
||||
$item[$id_to] = $source[$id_from];
|
||||
foreach (array('name', 'email', 'uri', 'url') as $what) {
|
||||
if (isset($source["{$id_from}_{$what}"])) {
|
||||
$item["{$id_to}_{$what}"] = $source["{$id_from}_{$what}"];
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// Atom elements to RSS elements
|
||||
$this->normalize_element($item, 'author', $item['dc'], 'creator', 'normalize_atom_person');
|
||||
$this->normalize_element($item, 'contributor', $item['dc'], 'contributor', 'normalize_atom_person');
|
||||
$this->normalize_element($item, 'summary', $item, 'description');
|
||||
$this->normalize_element($item, 'atom_content', $item['content'], 'encoded');
|
||||
$this->normalize_element($item, 'link_enclosure', $item, 'enclosure', 'normalize_enclosure');
|
||||
|
||||
// Categories
|
||||
if ( isset($item['category#']) ) { // Atom 1.0 categories to dc:subject and RSS 2.0 categories
|
||||
$this->normalize_element($item, 'category', $item['dc'], 'subject', 'normalize_category');
|
||||
}
|
||||
elseif ( isset($item['dc']['subject#']) ) { // dc:subject to Atom 1.0 and RSS 2.0 categories
|
||||
$this->normalize_element($item['dc'], 'subject', $item, 'category', 'normalize_dc_subject');
|
||||
}
|
||||
|
||||
// Normalized item timestamp
|
||||
$atom_date = (isset($item['published']) ) ? $item['published'] : $item['updated'];
|
||||
if ( $atom_date ) {
|
||||
$epoch = @parse_w3cdtf($atom_date);
|
||||
if ($epoch and $epoch > 0) {
|
||||
$item['date_timestamp'] = $epoch;
|
||||
}
|
||||
}
|
||||
|
||||
$this->items[$i] = $item;
|
||||
}
|
||||
}
|
||||
elseif ( $this->is_rss() ) {
|
||||
$this->channel['tagline'] = $this->channel['description'];
|
||||
// RSS elements to Atom elements
|
||||
$this->normalize_element($this->channel, 'description', $this->channel, 'tagline'); // Atom 0.3
|
||||
$this->normalize_element($this->channel, 'description', $this->channel, 'subtitle'); // Atom 1.0 (yay wordsmithing!)
|
||||
$this->normalize_element($this->image, 'url', $this->channel, 'logo');
|
||||
|
||||
for ( $i = 0; $i < count($this->items); $i++) {
|
||||
$item = $this->items[$i];
|
||||
if ( isset($item['description']))
|
||||
$item['summary'] = $item['description'];
|
||||
if ( isset($item['content']['encoded'] ) )
|
||||
$item['atom_content'] = $item['content']['encoded'];
|
||||
|
||||
// RSS elements to Atom elements
|
||||
$this->normalize_element($item, 'description', $item, 'summary');
|
||||
$this->normalize_element($item['content'], 'encoded', $item, 'atom_content');
|
||||
$this->normalize_element($item, 'enclosure', $item, 'link_enclosure', 'normalize_enclosure');
|
||||
|
||||
// Categories
|
||||
if ( isset($item['category#']) ) { // RSS 2.0 categories to dc:subject and Atom 1.0 categories
|
||||
$this->normalize_element($item, 'category', $item['dc'], 'subject', 'normalize_category');
|
||||
}
|
||||
elseif ( isset($item['dc']['subject#']) ) { // dc:subject to Atom 1.0 and RSS 2.0 categories
|
||||
$this->normalize_element($item['dc'], 'subject', $item, 'category', 'normalize_dc_subject');
|
||||
}
|
||||
|
||||
// Normalized item timestamp
|
||||
if ( $this->is_rss() == '1.0' and isset($item['dc']['date']) ) {
|
||||
$epoch = @parse_w3cdtf($item['dc']['date']);
|
||||
if ($epoch and $epoch > 0) {
|
||||
|
@ -578,7 +874,11 @@ class MagpieRSS {
|
|||
}
|
||||
}
|
||||
|
||||
|
||||
// magic ID function for multiple elemenets.
|
||||
// can be called as static MagpieRSS::element_id()
|
||||
function element_id ($el, $counter) {
|
||||
return $el . (($counter > 1) ? '#'.$counter : '');
|
||||
}
|
||||
} // end class RSS
|
||||
|
||||
function map_attrs($k, $v) {
|
||||
|
@ -589,18 +889,18 @@ function map_attrs($k, $v) {
|
|||
// courtesy, Ryan Currie, ryan@digibliss.com
|
||||
|
||||
if (!function_exists('array_change_key_case')) {
|
||||
define("CASE_UPPER",1);
|
||||
define("CASE_LOWER",0);
|
||||
define("CASE_UPPER",1);
|
||||
define("CASE_LOWER",0);
|
||||
|
||||
|
||||
function array_change_key_case($array,$case=CASE_LOWER) {
|
||||
if ($case=CASE_LOWER) $cmd=strtolower;
|
||||
elseif ($case=CASE_UPPER) $cmd=strtoupper;
|
||||
function array_change_key_case($array,$case=CASE_LOWER) {
|
||||
if ($case==CASE_LOWER) $cmd='strtolower';
|
||||
elseif ($case==CASE_UPPER) $cmd='strtoupper';
|
||||
foreach($array as $key=>$value) {
|
||||
$output[$cmd($key)]=$value;
|
||||
}
|
||||
return $output;
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
|
Loading…
Reference in a new issue