[RESOLVED] XML restricting output

BarmyArmy

I have PHP code that collects data from different databases and generates an XML file of all the data without any problems except one - It's a little bit big.

I would like to limit its output to only generate the first 30 lines of data.

Is this possible?

If yes, then can somebody guide me to a tutorial, manual or example please.

Or even better, just post the answer or commands I'm looking for.

TIA

johanafm

Define 30 lines of data... 30 lines of XML? 30 database entries? 30 lines of data presented to the user after applying XSLT? ...

halojoy

This script will read max 30 lines from any file.
This means of course some XML tags will be missing.
To make it a valid XML would take a more complicated script.

Uses function [man]fgets[/man]

<?php

$num = 30; // Read max 30 lines
$fp = fopen("filename.xml", "r");

if($fp) {
	$cnt = 0;
	$text = '';
	while ($cnt<$num && !feof($fp)){//$num lines or end of file
		$text .= fgets($fp, 4096);
		$cnt++;
	}
	fclose($fp);
}

$text = nl2br($text);
echo $text;

?>

BarmyArmy

johanafm - Hello again and happy (belated) new year, sorry about the short question. It was supposed to be output 30 lines of XML.

halojoy - cheers for the code, I will give it a bash and get back to you.

BarmyArmy

halojoy;10941286 wrote:
This script will read max 30 lines from any file.
This means of course some XML tags will be missing.
To make it a valid XML would take a more complicated script.

Uses function [man]fgets[/man]
<?php

$num = 30; // Read max 30 lines
$fp = fopen("filename.xml", "r");

if($fp) {
	$cnt = 0;
	$text = '';
	while ($cnt<$num && !feof($fp)){//$num lines or end of file
		$text .= fgets($fp, 4096);
		$cnt++;
	}
	fclose($fp);
}

$text = nl2br($text);
echo $text;

?>

halojoy - I have tried your code and slighty amended it to write out to a file called data.XML. It works great for stripping the data down to 30 lines of XML but keeps placing statements at the start and end of every line and then removes some tags as stated. Any more input would be appreciated for removing the entries or is it a lot more complicated.

Looks like I will need to read up on the fgets function. 😃

johanafm

If I understood your initial inquiry correctly, you wish to build an xml document from a number of non-xml sources, rather than get part of an existing xml document. While it is of course entirely possible to build a huge xml document and then process it to get part of the tree, it seems unnecessary to go so.
Also, whatever way you choose to go, you must ensure that the end result is well formed. XML parsers are required to stop parsing when they encounter errors.

You will also need to decide what "30 lines of XML data" actually means. For example, would you consider either one or both of the following to be 12 lines of XML data? Or is one 13 and the other 1 line?

<?xml version="1.0"?><rnode><items><item><name>A</name><amount>12</amount></item><item><name>B</name><amount>1</amount></item></items></rnode>

<?xml version="1.0"?>
<rnode>
	<items>
		<item>
			<name>A</name>
			<amount>12</amount>
		</item>
		<item>
			<name>B</name>
			<amount>1</amount>
		</item>
	</items>
</rnode>

If you do choose to process a possibly too large XML document, I recommend using DOM to process the document by a pre-visit depth-first traversal appending and create a new xml document by appending the first 30 elements encountered (should this be what you mean by "30 lines of data"). Of course, what you actually need to do is dependent on what you wish to achieve. Your requirements might mean that you perform a pre-visit breadth first traversal.

If you generate the document from other sources, process that data in whatever way you see fit, but stop as soon as you have created 30 elements.

BarmyArmy

If you do choose to process a possibly too large XML document, I recommend using DOM to process the document by a pre-visit depth-first traversal appending and create a new xml document by appending the first 30 elements encountered (should this be what you mean by "30 lines of data"). Of course, what you actually need to do is dependent on what you wish to achieve. Your requirements might mean that you perform a pre-visit breadth first traversal.

If you generate the document from other sources, process that data in whatever way you see fit, but stop as soon as you have created 30 elements.

Johanafm - You have hit the nail on the head. I'm not to good with PHP and XML, but working on it. I really meant keeping exactly the same structure of the of the data recieved and creating a new file (data.XML) keeping the first 30 elements.

It is not a true XML file, but is in a structure that is easy for my flash website to read.
Here's and example of the file being recieved:

<?xml version="1.0" encoding="iso-8859-1"?>
<items>
  <item title="Title1" description="Description1" url="http://www.blahblahblah.com"/>
  <item title="Title2" description="Description2" url="http://www.blahblahblah.com"/>
  <item title="Title3" description="Description3" url="http://www.blahblahblah.com"/>
  <item title="Title4" description="Description4" url="http://www.blahblahblah.com"/>
  <item title="Title5" description="Description5" url="http://www.blahblahblah.com"/>
</items>
[B]And so on for a hundred times[/B]

But I only require the first 30 elements, but keep the same structure. It looks like its possible acording to you and halojoy. I just need some extra guidance on what the best way of acomplishing this.

So the final product looks like this:

<?xml version="1.0" encoding="iso-8859-1"?>
<items>
  <item title="Title1" description="Description1" url="http://www.blahblahblah.com"/>
  <item title="Title2" description="Description2" url="http://www.blahblahblah.com"/>
  <item title="Title3" description="Description3" url="http://www.blahblahblah.com"/>
  <item title="Title4" description="Description4" url="http://www.blahblahblah.com"/>
  <item title="Title5" description="Description5" url="http://www.blahblahblah.com"/>
</items>
[B]And so on for a maximum thirty times[/B]

And possibly a small example. :rolleyes:

So do you recommend reading up on DOM and find out about "pre-visit breadth first traversal"?

Sounds a mouthful and complicated or is there an easier or better way?

BarmyArmy

johanfm - I have been using this php:

$num = 32; // Read max 32 lines 
$fp = fopen("data1.xml", "r"); 

if($fp) { 
    $cnt = 0; 
    $text = ''; 
    while ($cnt<$num && !feof($fp)){//$num lines or end of file 
        $text .= fgets($fp, 4096); 
        $cnt++;
    }
        fclose($fp); 
} 
$fp2 = fopen('data2.xml', 'w');
fwrite($fp2, $text);
fclose($fp);

Untidy as it seems, it does get the job done except for one small fact. It stops exactly at 32 lines and I now have to find a way of putting the end Root element (</items>) back in for my website to read it.

Any ideas guys..?

Will I have to create a new DOMDocument, set the root element and parse the newly formed data2.xml file into it to get this result?

BarmyArmy

BarmyArmy;10941301 wrote:
halojoy - I have tried your code and slighty amended it to write out to a file called data.XML. It works great for stripping the data down to 30 lines of XML but keeps placing statements at the start and end of every line and then removes some tags as stated. Any more input would be appreciated for removing the entries or is it a lot more complicated.

Looks like I will need to read up on the fgets function. 😃

halojoy - Sorry dude, nl2br was a new command to me and I didn't realise what it did. But after a little reading and editing of your php I found out what it does with the tags.

johanafm

Using halojoys approach, and knowing that the two first lines are:
<?xml ?>
<items>

followed by one or more <item> and closed by </items>, it's very easy to make your final XML well formed.
You just need to keep track of wether you reached the end of file, or if you stop in the middle

while ($cnt<$num && !($reached_eof = feof($fp)) ){

And then

if (!$reached_eof)
	$text .= '</items>';
fwrite($fp2, $text);

BarmyArmy

Absolutely spot on and works like a charm. Cheers Guys for all the support it works.

Still cannot my head around how it works, but maybe some light reading about fgets might help. 😃