PHP rss feed parser help

Figment

I am looking for a bit of help to better understand a script so i can easily adapt it now and in the future.

I am using this PHP script in a page to parse an rss feed:

<?php
/*
Class RSSParser: 2 October 2002
Original Author: Duncan Gough
Single Feed Hack: By Gregg 't3h GeeK' van der Sluys, http://geek.scorpiorising.ca 12-13-03. 
*/

class RSSParser	{

    var $title			= "";
    var $link 			= "";
    var $description 	= "";
    var $inside_item 	= false;


    var $all_rss_urls = array(
							"http://www.spoiltchild.com/new/rss.xml"	=> "Spoiltchild.com",
				);

	function startElement( $parser, $name, $attrs='' ){
		global $current_tag;

		$current_tag = $name;

		if( $current_tag == "ITEM" )
			$this->inside_item = true;

	} // endfunc startElement

	function endElement( $parser, $tagName, $attrs='' ){
		global $current_tag;

    	if ( $tagName == "ITEM" ) {

			printf( "\t<div class='newsitem'><h3><a href='%s' target='_blank'> %s</a></h3>\n", trim( $this->link ), htmlspecialchars( trim( $this->title ) ) );
    		printf( "\t<p>%s </p></div>\n", htmlspecialchars( trim( $this->description ) ) );

    		$this->title = "";
    		$this->description = "";
    		$this->link = "";
    		$this->inside_item = false;

    	}

	} // endfunc endElement

	function characterData( $parser, $data ){
		global $current_tag;

		if( $this->inside_item ){
			switch($current_tag){

				case "TITLE":
					$this->title .= $data;
					break;
				case "DESCRIPTION":
					$this->description .= $data;
					break;
				case "LINK":
					$this->link .= $data;
					break;

				default:
					break;

			} // endswitch

		} // end if

	} // endfunc characterData

	function parse_results( $xml_parser, $rss_parser, $file )	{

		xml_set_object( $xml_parser, &$rss_parser );
		xml_set_element_handler( $xml_parser, "startElement", "endElement" );
		xml_set_character_data_handler( $xml_parser, "characterData" );

		$fp = fopen("$file","r") or die( "Error reading XML file, $file" );

		while ($data = fread($fp, 4096))	{

			// parse the data
			xml_parse( $xml_parser, $data, feof($fp) ) or die( sprintf( "XML error: %s at line %d", xml_error_string( xml_get_error_code($xml_parser) ), xml_get_current_line_number( $xml_parser ) ) );

		} // endwhile

		fclose($fp);

		xml_parser_free( $xml_parser );

	} // endfunc parse_results

	function show_title( $rss_url ){
					?>
						<small></small>
					<?
	} // endfunc show_title

} // endclass RSSParser

global $rss_url;

// Set a default feed
if( $rss_url == "" )
	$rss_url = "http://www.spoiltchild.com/new/rss.xml";

$xml_parser = xml_parser_create();
$rss_parser = new RSSParser();

$rss_parser->show_title( $rss_url );
$rss_parser->parse_results( $xml_parser, &$rss_parser, $rss_url );

?>

you can see this script in action here: http://www.spoiltchild.com/new/news_test.php

And the source xml feed here: http://www.spoiltchild.com/new/rss.xml

My question is how to i pick out the individual elements and display them, such as the publish date and image and how can i take the URL of each news piece and add it to a 'More' link at the end of the description?

At the moment it seems to only pull and print each node (%s) in order. How do i add to what it pulls from the xml and change the order?

thanks

Ph3n0m

If you look carefully you can see the pattern


printf( "\t<div class='newsitem'><h3><a href='%s' target='_blank'> %s</a></h3>\n", trim( $this->link ), htmlspecialchars( trim( $this->title ) ) );

the first %s refers to trime($this->link)
the second %s refers to htmlspecialchars (trim($this->table))

and in this line

printf( "\t<p>%s </p></div>\n", htmlspecialchars( trim( $this->description ) ) );

the %s refers to htmlspecialchars( trim( $this->description ) )


Therefore using that logic you can define whatever you want as %s by changing the subsequent reference using the <tag> in the xml

for example I the headline to link to the story with the date in brackets beside that and underneath I want a shortened description


printf( "\t<div class='newsitem'><h3><a href='%s' target='_blank'> %s</a> (%s)</h3>\n", trim( $this->link ), htmlspecialchars(trim($this->title)), trim( $this->date ) );


and now for the description
printf( "\t<br>%s </p></div>\n", htmlspecialchars( trim( $this->description ) ) );

or something similiar to that


AND dont forget to declare any extra vars you want at the start of the script

i.e.

var $date = "";

and

$this->date = "";

Figment

Cool, thanks. I got it now.

next thing...
the date is coming in as "Tue, 21 Dec 2004 07:02:00 EDT"
Is there any way to drop the last few characters do i end up with "Tue, 21 Dec 2004" ?

Thanks for your help

dangerman

just use substring

http://ie.php.net/manual/en/function.substr.php



so for Tue, 21 Dec 2004 i think it would be

$date = substr($date, 0, 16);

Figment

Sorry, my knowledge of php is below basic. Where would i place
$pubDate = substr($pubDate, 0, 16); in that code?

A few attempts gave me errors.

Ph3n0m

if you are using the following


printf( "\t<div class='newsitem'><h3><a href='%s' target='_blank'> %s</a> (%s)</h3>\n", trim( $this->link ), htmlspecialchars(trim($this->title)), trim( $this->date ) );


the just use this version of it (i think this will work)


printf( "\t<div class='newsitem'><h3><a href='%s' target='_blank'> %s</a> (%s)</h3>\n", trim( $this->link ), htmlspecialchars(trim($this->title)), substr(trim( $this->date ), 0, 16) );

Figment

That did the job brilliant. Thank you.

dahamsta

Unless you're doing it to educate yourself, you'd be a lot better off using Magpie. It's been around for an age so it's as mature as you get, and it'll parse just about anything.

adam

Figment

I looked at Magpie first but it was overkill for what i needed.
I also needed a solution with less of an install footprint (none) so the script i used above was ideal.

Thanks