Help with PHP XML Display

[h3x3n]

Hi all

I have this XML file containing news articles, and I wish to use PHP to parse and then display requested articles.

PROBLEM

The problem is the articles do not display correctly, in particular:

-$articleContent contains data from BOTH articles
-when using showRecentArticles(), Article ID 1 shows up twice, the first time containing both article content from Articles ID 1 and 2, and the second time containing article content from Articles 1,2, then 1 and 2 again.

I hope I can get help fixing this code, and also welcome all suggestions pertaining to rewriting code to improve performance.

INSTRUCTIONS

newsHandler.php (the PHP code)

<?php

//CONFIGURATION SECTION
$xmlSource = "news.xml";	//Location of your news XML file
$noRecentArticles = 2;		//Number of recent articles to be shown
$articleFormat = '

<table class="newsholder">
	<tr>
		<td style="width:100px" align="center" colspan="1" rowspan="3">
			<div class="bold">[AUTHORNAME]</div>
			<div>[AUTHORRANK]</div>
			<img src="[AUTHORPICT]" alt="" />
		</td>
		<td style="width:450px;" colspan="1" rowspan="1">
			<div class="newstitle">[TITLE]</div>
			<ins><hr class="newsDivider" /></ins>
		</td>
	</tr>
	<tr>
		<td style="width:450px;" colspan="1" rowspan="1">
			<div class="newscontent">
			[CONTENT]
			</div>
		</td>
	</tr>
  	<tr>
    	<td style="width:450px;">
		<div class="date">[DATE]</div>
		</td>
	</tr>
</table>
<ins><img src="img/area.jpg" width="550" height="25" alt="" /></ins>
				 ';

//GLOBAL CONSTANTS SECTION
define("OPTION_UNDEFINED", 0);
define("OPTION_GET_ARTICLE", 1);
define("OPTION_GET_LATEST_ARTICLE_ID", 2);

//GLOBAL VARIABLES SECTION
$requestedArticle = 0;
$foundArticle = 0;
$resultString = "";
$currentNode = "";
$handlerMode = 0;
$articleId = 0;
$articleTitle = "";
$articleDate = "";
$articleAuthorName = "";
$articleAuthorRank = "";
$articleAuthorPict = "";
$articleContent = "";

//FRONT-END SECTION
function showArticle($articleId) {
	getArticle($articleId);
	transformArticle();
	print $GLOBALS["resultString"];
}

function showRecentArticles() {
	$temp = getLatestArticleId();
	$GLOBALS["resultString"] = "";
	for ($i = $temp; $i > ($temp - $GLOBALS["noRecentArticles"]); $i--) {
		//Get and display that specific article
		getArticle($i);
		transformArticle();
		print $GLOBALS["resultString"];
	}
}

//BACK-END SECTION
function getArticle($articleId) {
	//Set global variables
	$GLOBALS["requestedArticle"] = $articleId;
	$GLOBALS["handlerMode"] = OPTION_GET_ARTICLE;

//Parse XML
$file = fopen($GLOBALS["xmlSource"], "r") or die ("Unable to open XML file");
$xmlParser = xml_parser_create();
xml_parser_set_option($xmlParser, XML_OPTION_CASE_FOLDING, false);
xml_parser_set_option($xmlParser, XML_OPTION_SKIP_WHITE, true);
xml_set_element_handler($xmlParser, "startElement", "endElement");
xml_set_character_data_handler($xmlParser, "cData");
while ($s = fread($file, 4096)) {
	xml_parse($xmlParser, $s, feof($file));
}
xml_parser_free($xmlParser);

//Reset global variables
$GLOBALS["requestedArticle"] = 0;
$GLOBALS["handlerMode"] = OPTION_UNDEFINED;
}

function getLatestArticleId() {
	//Set global variables
	$GLOBALS["requestedArticle"] = 0;
	$GLOBALS["handlerMode"] = OPTION_GET_LATEST_ARTICLE_ID;

//Parse XML
$file = fopen($GLOBALS["xmlSource"], "r") or die ("Unable to open XML file");
$xmlParser = xml_parser_create();
xml_parser_set_option($xmlParser, XML_OPTION_CASE_FOLDING, false);
xml_parser_set_option($xmlParser, XML_OPTION_SKIP_WHITE, true);
xml_set_element_handler($xmlParser, "startElement", "endElement");
xml_set_character_data_handler($xmlParser, "cData");
while ($s = fread($file, 4096)) {
	xml_parse($xmlParser, $s, feof($file));
}
xml_parser_free($xmlParser);
$result = $GLOBALS["requestedArticle"];

//Reset global variables
$GLOBALS["requestedArticle"] = 0;
$GLOBALS["handlerMode"] = OPTION_UNDEFINED;

//Return result
return $result;
}

function transformArticle() {
	$result = $GLOBALS["articleFormat"];

//Replace entities with values
$result = str_replace('[ID]', $GLOBALS["articleId"], $result);
$result = str_replace('[TITLE]', $GLOBALS["articleTitle"], $result);
$result = str_replace('[DATE]', $GLOBALS["articleDate"], $result);
$result = str_replace('[AUTHORNAME]', $GLOBALS["articleAuthorName"], $result);
$result = str_replace('[AUTHORRANK]', $GLOBALS["articleAuthorRank"], $result);
$result = str_replace('[AUTHORPICT]', $GLOBALS["articleAuthorPict"], $result);
$result = str_replace('[CONTENT]', $GLOBALS["articleContent"], $result);

$GLOBALS["resultString"] = $result;
}

function startElement($parser, $name, $attribs) {
	$GLOBALS["currentNode"] = $name;

if ($GLOBALS["handlerMode"] == OPTION_GET_ARTICLE) {
	if ($GLOBALS["currentNode"] == "article") {
		$GLOBALS["articleId"] = $attribs["id"];
		if (strcmp($GLOBALS["articleId"], $GLOBALS["requestedArticle"]) == 0) {
			$GLOBALS["foundArticle"] == 1;
		}
	}
	else if ($GLOBALS["currentsNode"] == "a") {
		$GLOBALS["articleContent"] .= "<a class=\"newsLink\" href=\"" . $attribs["href"] . "\">";
	}
	elseif ($GLOBALS["currentNode"] == "link") {
		$GLOBALS["articleContent"] .= "<ins id=\"" . $attribs["id"] . "\" style=\"" . $attribs["style"] . "\" onmouseover=\"" . $attribs["onmouseover"]  . "\" onmouseout=\"" . $attribs["onmouseout"] . "\" onclick=\"" . $attribs["onclick"] . "\">";
	}
	elseif ($GLOBALS["currentNode"] == "br") {
		$GLOBALS["articleContent"] .= "<ins><br /></ins>";
	}
	elseif ($GLOBALS["currentNode"] == "img") {
		$GLOBALS["articleContent"] .= "<img src=\"" . $attribs['src'] . "\" alt=\"" . $attribs['alt'] . "\" style=\"" . $attribs['style'] . "\" />";
	}
}
elseif ($GLOBALS["handlerMode"] == OPTION_GET_LATEST_ARTICLE_ID) {
	if ($GLOBALS["currentNode"] == "article") {
		$GLOBALS["articleId"] = (int) $attribs["id"];
		if ($GLOBALS["articleId"] > $GLOBALS["requestedArticle"]) {
			$GLOBALS["requestedArticle"] = $GLOBALS["articleId"];
		}
	}
}
else {
	die ("Undefined option for function startElement()");
}
}

function endElement($parser, $name) {
	if ($GLOBALS["currentNode"] == "article") {
		//Reset article information
		$GLOBALS["articleTitle"] = "";
		$GLOBALS["articleDate"] = "";
		$GLOBALS["articleAuthorName"] = "";
		$GLOBALS["articleAuthorRank"] = "";
		$GLOBALS["articleAuthorPict"] = "";
		$GLOBALS["articleContent"] = "";
	}
	if ($GLOBALS["currentNode"] == "a") {
		$GLOBALS["articleContent"] .= "</a>";
		$GLOBALS["currentNode"] = "content";
	}
	elseif ($GLOBALS["currentNode"] == "br") {
		$GLOBALS["currentNode"] = "content";
	}
	elseif ($GLOBALS["currentNode"] == "link") {
		$GLOBALS["articleContent"] .= "</ins>";
		$GLOBALS["currentNode"] = "content";
	}
	else {
		//Reset global variables
		$GLOBALS["foundArticle"] = 0;
		$GLOBALS["currentNode"] = "";
	}
}

function cData($parser, $data) {
	if ($GLOBALS["handlerMode"] == OPTION_GET_ARTICLE) {
		//Set variable values
		switch ($GLOBALS["currentNode"]) {
			case "article":
				break;
			case "title":
				$GLOBALS["articleTitle"] = $data;
				break;
			case "date":
				$GLOBALS["articleDate"] = $data;
				break;
			case "name":
				$GLOBALS["articleAuthorName"] = $data;
				break;
			case "rank":
				$GLOBALS["articleAuthorRank"] = $data;
				break;
			case "pict":
				$GLOBALS["articleAuthorPict"] = $data;
				break;
			case "content":
				$GLOBALS["articleContent"] .= $data;
				break;
			case "link":
				$GLOBALS["articleContent"] .= $data;
				break;
		}
	}
}

?>

To be continued...

[h3x3n]

Continued from previous post...

news.xml (containing sample news)

<?xml version="1.0"?>
<news>
	<article id="2">
		<title>H3X3N's Lair is looking for staff!</title>
		<date>11:32AM Tuesday November 24 2005</date>
		<author>
			<name>ANGELUS</name>
			<rank>Adminstrator</rank>
			<pict>img/admin.jpg</pict>
		</author>
		<content>
		Currently, H3X3N's Lair is being developed at a sluggish pace, and that has to change. With that being said, we are looking for talented volunteers to help lead and maintain the website. The following posts are currently available:
				<br />
				<br />- Content Manager(s)
				<br />- Graphics Artist(s)
				<br />
				<br />
			A page with more details will be posted soon. Please direct all queries and submit your applications to <link id="newsLink1" style="color:#FFFFFF" onmouseover="changeStyle(1,'newsLink1')" onmouseout="changeStyle(0,'newsLink1')" onclick="linkTo('mailto:[e-mail]')">[e-mail]</link>.
		</content>
	</article>
	<article id="1">
		<title>H3X3N's Lair is XHTML and CSS compliant!</title>
		<date>12:49AM Tuesday November 22 2005</date>
		<author>
			<name>ANGELUS</name>
			<rank>Adminstrator</rank>
			<pict>img/admin.jpg</pict>
		</author>
		<content>
		After hours spent on restructuring and rewriting mountains of code, we are proud to announce that H3X3N's Lair has passed both the W3C XHTML 1.0 Strict and CSS standards. This means that H3X3N's Lair should work and display well regardless of whatever browsers and platforms are used. Also, passing W3C's XHTML and CSS validators entitles us to include the following logos in our website:
			<br />
			<br />
			<a href="http://validator.w3.org/check?uri=referer">
			<img src="img/xhtml.jpg" style="border:0;width:88px;height:31px;" alt="Valid XHTML 1.0 Strict" />
			</a>
			<a href="http://jigsaw.w3.org/css-validator/">
			<img src="img/css.jpg" style="border:0;width:88px;height:31px;" alt="Valid CSS!" />
			</a>
		</content>
	</article>
</news>

In order to see the results, create two files newsHandler.php and news.xml in your HTDOCS folder and copy and paste the above code into the respective files. Afterwards, create another PHP file, include newsHandler.php and call either showArticle($articleId) or showRecentArticles().

CONSIDERATIONS

Bear in mind that the articles in the XML file may not be sorted according to IDs. Also, while optimizing code performance is important, it is even more important that the code should be easy to customize (by changing values under the CONFIGURATION SECTION).

Thanks a lot.

[h3x3n]

Merry Christmas to all of you! 🙂

This morning, I've rewrote the newsHandler.php in an OOP fashion, and've also managed to fix a few errors. However, there're still problems:

• ($this->foundArticle == 1) never returns true and hence nothing besides the template is displayed. Testing revealed that $this->foundArticle always held a value of 0.
• When I changed the code to test for $this->foundArticle == 0 or left the whole testing out completely, the articles still repeat. Checking showArticle() revealed that only Article ID 1 shows regardless of the articleID argument specificed.

Here is the rewritten newsHandler.php:

<?php

class NewsProcessor {
	//CLASS MEMBERS
	private $xmlSource;
	private $noOfRecentArticles;
	private $articleFormat;

//PARSER INFORMATION
private $xmlParser;
private $requestedArticle = 0;
private $foundArticle = 0;
private $resultString = '';
private	$currentNode = '';
private $handlerMode = 0;
private $articleId = 0;
private $articleTitle = '';
private $articleDate = '';
private $articleAuthorName = '';
private $articleAuthorRank = '';
private $articleAuthorPict = '';
private $articleContent = '';

//CLASS CONSTANTS
const PARSER_OPTION_UNDEFINED = 0;
const PARSER_OPTION_GET_ARTICLE = 1;
const PARSER_OPTION_GET_LATEST_ARTICLE_ID = 2;

//CONSTRUCTOR
public function __construct($xmlSource, $noOfRecentArticles, $articleFormat) {
	$this->xmlSource = $xmlSource;
	$this->noOfRecentArticles = $noOfRecentArticles;
	$this->articleFormat = $articleFormat;
}

//ACCESSOR METHODS
public function getXMLSource() {
	return $this->xmlSource;
}
public function getNoOfRecentArticles() {
	return $this->noOfRecentArticles;
}
public function getArticleFormat() {
	return $this->articleFormat;
}

//SETTER METHODS
public function setXMLSource($xmlSource) {
	$this->xmlSource = $xmlSource;
}
public function setNoOfRecentArticles($noOfRecentArticles) {
	$this->xmlSource = $noOfRecentArticles;
}
public function setArticleFormat($articleFormat) {
	$this->xmlSource = $articleFormat;
}

//FRONT-END METHODS
public function showArticle() {
	$this->getArticle($articleId);
	$this->transformArticle();
    print $this->resultString;
	$this->resetProcessor();
}

public function showRecentArticles() {
	$temp = $this->getLatestArticleId();
	for ($i = $temp; $i > ($temp - $this->noOfRecentArticles); $i--) {
		//Get and display that specific article
		$this->getArticle($i);
		$this->transformArticle();
		print $this->resultString;
		$this->resetProcessor();
	}
}

//BACK-END METHODS
private function resetProcessor() {
	$this->requestedArticle = 0;
	$this->foundArticle = 0;
	$this->resultString = '';
	$this->currentNode = '';
	$this->handlerMode = 0;
	$this->articleId = 0;
	$this->articleTitle = '';
	$this->articleDate = '';
	$this->articleAuthorName = '';
	$this->articleAuthorRank = '';
	$this->articleAuthorPict = '';
	$this->articleContent = '';
}

private function getArticle($articleId) {
	//Set parser variables
	$this->requestedArticle = $articleId;
	$this->handlerMode = PARSER_OPTION_GET_ARTICLE;

	//Parse XML
	$file = fopen($this->xmlSource, 'r') or die ('Unable to open XML file');
	$this->xmlParser = xml_parser_create();
	xml_set_object($this->xmlParser, $this);
	xml_parser_set_option($this->xmlParser, XML_OPTION_CASE_FOLDING, false);
	xml_parser_set_option($this->xmlParser, XML_OPTION_SKIP_WHITE, true);
	xml_set_element_handler($this->xmlParser, "startElement", "endElement");
	xml_set_character_data_handler($this->xmlParser, "cData");
	while ($s = fread($file, 4096)) {
		xml_parse($this->xmlParser, $s, feof($file));
	}
	xml_parser_free($this->xmlParser);
}

private function getLatestArticleId() {
	//Set parser variables
	$this->requestedArticle = 0;
	$this->handlerMode = PARSER_OPTION_GET_LATEST_ARTICLE_ID;

	//Parse XML
	$file = fopen($this->xmlSource, 'r') or die ('Unable to open XML file');
	$this->xmlParser = xml_parser_create();
	xml_set_object($this->xmlParser, $this);
	xml_parser_set_option($this->xmlParser, XML_OPTION_CASE_FOLDING, false);
	xml_parser_set_option($this->xmlParser, XML_OPTION_SKIP_WHITE, true);
	xml_set_element_handler($this->xmlParser, "startElement", "endElement");
	xml_set_character_data_handler($this->xmlParser, "cData");
	while ($s = fread($file, 4096)) {
		xml_parse($this->xmlParser, $s, feof($file));
	}
	xml_parser_free($this->xmlParser);
	$result = $this->requestedArticle;
	$this->resetProcessor();					
	return $result;
}

private function transformArticle() {
	$result = $this->articleFormat;

	//Replace placeholders with actual values
	$result = str_replace('[ID]', $this->articleId, $result);
	$result = str_replace('[TITLE]', $this->articleTitle, $result);
	$result = str_replace('[DATE]', $this->articleDate, $result);
	$result = str_replace('[AUTHORNAME]', $this->articleAuthorName, $result);
	$result = str_replace('[AUTHORRANK]', $this->articleAuthorRank, $result);
	$result = str_replace('[AUTHORPICT]', $this->articleAuthorPict, $result);
	$result = str_replace('[CONTENT]', $this->articleContent, $result);

	$this->resultString = $result;
}

function startElement($parser, $name, $attribs) {
	$this->currentNode = $name;

	if ($this->handlerMode == PARSER_OPTION_GET_ARTICLE) {
		if ($this->currentNode == 'article') {
			$this->articleId = $attribs['id'];
			if ($this->articleId == $this->requestedArticle) {
				$this->foundArticle = 1;
			}
			else {
				$this->foundArticle = 0;
			}
		}
		elseif ($this->currentsNode == 'a' && $this->foundArticle == 1) {
			$this->articleContent .= "<a class=\"newsLink\" href=\"" . $attribs['href'] . "\">";
		}
		elseif ($this->currentNode == 'link' && $this->foundArticle == 1) {
			$this->articleContent .= "<ins id=\"" . $attribs['id'] . "\" style=\"" . $attribs['style'] . "\" onmouseover=\"" . $attribs['onmouseover']  . "\" onmouseout=\"" . $attribs['onmouseout'] . "\" onclick=\"" . $attribs['onclick'] . "\">";
		}
		elseif ($this->currentNode == 'br' && $this->foundArticle == 1) {
			$this->articleContent .= '<ins><br /></ins>';
		}
		elseif ($this->currentNode == 'img' && $this->foundArticle == 1) {
			$this->articleContent .= "<img src=\"" . $attribs['src'] . "\" alt=\"" . $attribs['alt'] . "\" style=\"" . $attribs['style'] . "\" />";
		}
	}
	elseif ($this->handlerMode == PARSER_OPTION_GET_LATEST_ARTICLE_ID) {
		if ($this->currentNode == 'article') {
			$this->articleId = (int) $attribs['id'];
			if ($this->articleId > $this->requestedArticle) {
				$this->requestedArticle = $this->articleId;
			}
		}
	}
	else {
		die ('Undefined option for function startElement()');
	}
}

function endElement($parser, $name) {
	if ($this->currentNode == 'article') {
		//Reset article information
		$this->resetProcessor();
		$this->currentNode = 'news';
	}
	elseif ($this->currentNode == 'title') {
		$this->currentNode = 'article';
	}
	elseif ($this->currentNode == 'date') {
		$this->currentNode = 'article';
	}
	elseif ($this->currentNode == 'name') {
		$this->currentNode = 'author';
	}
	elseif ($this->currentNode == 'rank') {
		$this->currentNode = 'author';
	}
	elseif ($this->currentNode == 'pict') {
		$this->currentNode = 'author';
	}
	elseif ($this->currentNode == 'author') {
		$this->currentNode = 'article';
	}
	elseif ($this->currentNode == 'a') {
		$this->articleContent .= '</a>';
		$this->currentNode = 'content';
	}
	elseif ($this->currentNode == 'br') {
		$this->currentNode = "content";
	}
	elseif ($this->currentNode == 'link') {
		$this->articleContent .= '</ins>';
		$this->currentNode = 'content';
	}
	elseif ($this->currentNode == 'content') {
		$this->currentNode = 'article';
	}
	else {
		//Reset parser variables
		$this->foundArticle = 0;
		$this->currentNode = "";
	}
}

function cData($parser, $data) {
	if ($this->handlerMode == PARSER_OPTION_GET_ARTICLE && $this->foundArticle == 1) {
		//Set variable values
		switch ($this->currentNode) {
			case 'article':
				break;
			case 'title':
				$this->articleTitle = $data;
				break;
			case 'date':
				$this->articleDate = $data;
				break;
			case 'name':
				$this->articleAuthorName = $data;
				break;
			case 'rank':
				$this->articleAuthorRank = $data;
				break;
			case 'pict':
				$this->articleAuthorPict = $data;
				break;
			case 'content':
				$this->articleContent .= $data;
				break;
			case 'link':
				$this->articleContent .= $data;
				break;
		}
	}
}
}

?>

[h3x3n]

Continued...

To test, simply use the following code in a new PHP file -

<?php
                 require_once("newsHandler.php");
	/* CONFIGURATION SECTION
	 *
	 * $xmlSource
	 * The location of the XML file containing news
	 *
	 * $noOfRecentArticles
	 * The number of recent news articles to be displayed
	 *
	 * $articleFormat
	 * The HTML used to layout the article. Use the placeholders
	 * e.g. [TITLE] instead of the actual values. The names of
	 * the placeholders should be self-explanatory.
	 */
	$xmlSource = 'news.xml';
	$noOfRecentArticles = 2;
	$articleFormat = <<<ENDOFSTRING
	<table class="newsholder">
		<tr>
			<td style="width:100px" align="center" colspan="1" rowspan="3">
				<div class="bold">[AUTHORNAME]</div>
				<div>[AUTHORRANK]</div>
				<img src="[AUTHORPICT]" alt='' />
			</td>
			<td style="width:450px;" colspan="1" rowspan="1">
				<div class="newstitle">[TITLE]</div>
				<ins><hr class="newsDivider" /></ins>
			</td>
		</tr>
		<tr>
			<td style="width:450px;" colspan="1" rowspan="1">
				<div class="newscontent">
				[CONTENT]
				</div>
			</td>
		</tr>
  		<tr>
    		<td style="width:450px;">
			<div class="date">[DATE]</div>
			</td>
		</tr>
	</table>
	<ins><img src="img/area.jpg" width="550" height="25" alt='' /></ins>
ENDOFSTRING;
	$news = new NewsProcessor($xmlSource, $noOfRecentArticles, $articleFormat);
	$news->showRecentArticles();
	?>