📄 rss_parse_magpie.inc.php
字号:
function concat (&$str1, $str2="") {
if (!isset($str1) ) {
$str1="";
}
$str1 .= $str2;
}
function append_content($text) {
if ( $this->initem ) {
$this->concat( $this->current_item[ $this->incontent ], $text );
}
elseif ( $this->inchannel ) {
$this->concat( $this->channel[ $this->incontent ], $text );
}
}
// smart append - field and namespace aware
function append($el, $text) {
if (!$el) {
return;
}
if ( $this->current_namespace )
{
if ( $this->initem ) {
$this->concat(
$this->current_item[ $this->current_namespace ][ $el ], $text);
}
elseif ($this->inchannel) {
$this->concat(
$this->channel[ $this->current_namespace][ $el ], $text );
}
elseif ($this->intextinput) {
$this->concat(
$this->textinput[ $this->current_namespace][ $el ], $text );
}
elseif ($this->inimage) {
$this->concat(
$this->image[ $this->current_namespace ][ $el ], $text );
}
}
else {
if ( $this->initem ) {
$this->concat(
$this->current_item[ $el ], $text);
}
elseif ($this->intextinput) {
$this->concat(
$this->textinput[ $el ], $text );
}
elseif ($this->inimage) {
$this->concat(
$this->image[ $el ], $text );
}
elseif ($this->inchannel) {
$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'];
$atom_date = (isset($item['issued']) ) ? $item['issued'] : $item['modified'];
if ( $atom_date ) {
$epoch = @parse_w3cdtf($item['modified']);
if ($epoch and $epoch > 0) {
$item['date_timestamp'] = $epoch;
}
}
$this->items[$i] = $item;
}
}
elseif ( $this->is_rss() ) {
$this->channel['tagline'] = $this->channel['description'];
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'];
if ( $this->is_rss() == '1.0' and isset($item['dc']['date']) ) {
$epoch = @parse_w3cdtf($item['dc']['date']);
if ($epoch and $epoch > 0) {
$item['date_timestamp'] = $epoch;
}
}
elseif ( isset($item['pubdate']) ) {
$epoch = @strtotime($item['pubdate']);
if ($epoch > 0) {
$item['date_timestamp'] = $epoch;
}
}
$this->items[$i] = $item;
}
}
}
function is_rss () {
if ( $this->feed_type == RSS ) {
return $this->feed_version;
}
else {
return false;
}
}
function is_atom() {
if ( $this->feed_type == ATOM ) {
return $this->feed_version;
}
else {
return false;
}
}
/**
* return XML parser, and possibly re-encoded source
*
*/
function create_parser($source, $out_enc, $in_enc, $detect) {
if ( substr(phpversion(),0,1) == 5) {
$parser = $this->php5_create_parser($in_enc, $detect);
}
else {
list($parser, $source) = $this->php4_create_parser($source, $in_enc, $detect);
}
if ($out_enc) {
$this->encoding = $out_enc;
xml_parser_set_option($parser, XML_OPTION_TARGET_ENCODING, $out_enc);
}
return array($parser, $source);
}
/**
* Instantiate an XML parser under PHP5
*
* PHP5 will do a fine job of detecting input encoding
* if passed an empty string as the encoding.
*
* All hail libxml2!
*
*/
function php5_create_parser($in_enc, $detect) {
// by default php5 does a fine job of detecting input encodings
if(!$detect && $in_enc) {
return xml_parser_create($in_enc);
}
else {
return xml_parser_create('');
}
}
/**
* Instaniate an XML parser under PHP4
*
* Unfortunately PHP4's support for character encodings
* and especially XML and character encodings sucks. As
* long as the documents you parse only contain characters
* from the ISO-8859-1 character set (a superset of ASCII,
* and a subset of UTF-8) you're fine. However once you
* step out of that comfy little world things get mad, bad,
* and dangerous to know.
*
* The following code is based on SJM's work with FoF
* @see http://minutillo.com/steve/weblog/2004/6/17/php-xml-and-character-encodings-a-tale-of-sadness-rage-and-data-loss
*
*/
function php4_create_parser($source, $in_enc, $detect) {
if ( !$detect ) {
return array(xml_parser_create($in_enc), $source);
}
if (!$in_enc) {
if (preg_match('/<?xml.*encoding=[\'"](.*?)[\'"].*?>/m', $source, $m)) {
$in_enc = strtoupper($m[1]);
$this->source_encoding = $in_enc;
}
else {
$in_enc = 'UTF-8';
}
}
if ($this->known_encoding($in_enc)) {
return array(xml_parser_create($in_enc), $source);
}
/*
// the dectected encoding is not one of the simple encodings PHP knows
// attempt to use the iconv extension to
// cast the XML to a known encoding
// @see http://php.net/iconv
if (function_exists('iconv')) {
$encoded_source = iconv($in_enc,'UTF-8', $source);
if ($encoded_source) {
return array(xml_parser_create('UTF-8'), $encoded_source);
}
}
// iconv didn't work, try mb_convert_encoding
// @see http://php.net/mbstring
if(function_exists('mb_convert_encoding')) {
$encoded_source = iconv($source, 'UTF-8', $in_enc );
if ($encoded_source) {
return array(xml_parser_create('UTF-8'), $encoded_source);
}
}
// else
$this->error("Feed is in an unsupported character encoding. ($in_enc) " .
"You may see strange artifacts, and mangled characters.",
E_USER_NOTICE);
*/
return array(xml_parser_create(), $source);
}
function known_encoding($enc) {
$enc = strtoupper($enc);
if ( in_array($enc, $this->_KNOWN_ENCODINGS) ) {
return $enc;
}
else {
return false;
}
}
function error ($errormsg, $lvl=E_USER_WARNING) {
// append PHP's error message if track_errors enabled
if ( $php_errormsg ) {
$errormsg .= " ($php_errormsg)";
}
if ( MAGPIE_DEBUG ) {
trigger_error( $errormsg, $lvl);
}
else {
error_log( $errormsg, 0);
}
$notices = E_USER_NOTICE|E_NOTICE;
if ( $lvl&$notices ) {
$this->WARNING = $errormsg;
} else {
$this->ERROR = $errormsg;
}
}
} // end class RSS
function map_attrs($k, $v) {
return "$k=\"$v\"";
}
function parse_w3cdtf ( $date_str ) {
# regex to match wc3dtf
$pat = "/(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2})(:(\d{2}))?(?:([-+])(\d{2}):?(\d{2})|(Z))?/";
if ( preg_match( $pat, $date_str, $match ) ) {
list( $year, $month, $day, $hours, $minutes, $seconds) =
array( $match[1], $match[2], $match[3], $match[4], $match[5], $match[6]);
# calc epoch for current date assuming GMT
$epoch = gmmktime( $hours, $minutes, $seconds, $month, $day, $year);
$offset = 0;
if ( $match[10] == 'Z' ) {
# zulu time, aka GMT
}
else {
list( $tz_mod, $tz_hour, $tz_min ) =
array( $match[8], $match[9], $match[10]);
# zero out the variables
if ( ! $tz_hour ) { $tz_hour = 0; }
if ( ! $tz_min ) { $tz_min = 0; }
$offset_secs = (($tz_hour*60)+$tz_min)*60;
# is timezone ahead of GMT? then subtract offset
#
if ( $tz_mod == '+' ) {
$offset_secs = $offset_secs * -1;
}
$offset = $offset_secs;
}
$epoch = $epoch + $offset;
return $epoch;
}
else {
return -1;
}
}
?>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -