Add XML News Feeds to Your Site

>>You can also display, for example only 3 news items, if you change the foreach with a different loop construct.

How exactly would you write that? Please help.

- Matt

there are loadsa ways of only showing a limited number of items..:

i.e.

$id=0;

foreach ($items as $item){

if($id=="5") {

break;

}

$id++;

$title = untag($item, 'title');

$link = untag($item, 'link');

rest of code....

Hi there,

Thanks for the use of this script. The problem I have is that the xml feed I've got has links that look like this:

http%3A%2F%2Fwww%2Enzherald%2Eco%2Enz%2Fstorydisplay%2Ecfm%3FstoryID%3D3523776%26thesection %3Dtechnology%26thesubsection%3Dgeneral

Can the parser be modified to deal with this?

works great exept it whipes out the code that comes after it iused an inclde function to iclude thx php script into my html when i veiw the page everything below the php is not seen by th browser huh if you can help thanks

another quick question how can i make the head lines open a new browser for the hole story when clicked

I am wanting to do more than just show the links and title....

I want to display the title as a link and then the description underneath it. This is completely allowed by the site feeding the rss.

My problem lies in the fact that some descriptions will have <p> tag in them....this causes that description to not be displayed....how do i correct?

sam: set up your link tags as:

a target=_blank href=[html link here]

the target=_blank causes the link to open in a new window when clicked.

john: unwanted tags. just process the descriptions prior to displaying them. copy all the xml into a char string until you encounter the tag, skip past it, continue copying.

It don't work :-)

I get the error ...

Warning: fopen(http://slashdot.org/slashdot.rdf): failed to open stream: Bad file descriptor in C:\htdocs\test.php on line 6

Moreover has a big list of news headline feeds:

http://w.moreover.com/categories/category_list_xml.html

Unfortunately Moreover is atrocious at weeding out identical stories. Using the feeds they provide, how might I go about discarding any <article> where the <headline_text> is not unique?

Here's an example of one article:

- <article id="_113573634">

<url>http://c.moreover.com/click/here.pl?x113573634</url>

<headline_text>British Airways to Resume Flight to D.C</headline_text>

<source>AP via New York Post</source>

<media_type>text</media_type>

<cluster>UK business news</cluster>

<tagline />

<document_url>http://breakingnews.nypost.com</document_url>

<harvest_time>Jan 3 2004 2:34PM</harvest_time>

<access_registration />

<access_status />

</article>

What a great article, it really helped me get an XML document into a CSV file.

Hey thanks for the awesome code

if anyone is interested, here's mine with some small changes (messy but works)

grabs 3 random headlines, puts the full headline in the title attribute of the a tag and shortens the headline if necessary (ie length is > 20) with '...' on the end

$feed = 'http://www.some.com.au/FEED/xml.html';

ini_set('allow_url_fopen', true);

$fp = fopen($feed, 'r');

$xml = '';

while (!feof($fp)) {

$xml .= fread($fp, 128);

}

fclose($fp);

function untag($string, $tag) {

$tmpval = array();

$preg = "|<$tag>(.*?)</$tag>|s";

preg_match_all($preg, $string, $tags);

foreach ($tags[1] as $tmpcont){

$tmpval[] = $tmpcont;

}

return $tmpval;

}

$items = untag($xml, 'resource');

$max=count($items)-3;

$start=rand(0,$max);

$html = '';

for($i=$start;$i<$start+3;$i++){

$title = untag($items[$i], 'headline');

$link = untag($items[$i], 'linkto');

$html .= '<a href="' . $link[0] . '" title="' . $title[0] . '">';

$len=strlen($title[0]);

if($len>20){

$words=explode(' ',$title[0]);

$title[0]='';

for($w=0;$w<count($words);$w++){

$title[0] .= $words[$w] . ' ';

if(strlen($title[0])>20){

$title[0] .= '...';

break;

}

}

}

$html .= $title[0] . "</a><br />\n";

}

echo $html;

cheers

Is there any way to get a particular area like all the articles on the job market or the housing market?

i'm fiddling with it, but i'm having a hard time trying to get the date inserted in the html string as well...any ideas?

is there a way to limit the amount of feeds that come through, e.g I only want the first 5 news headlines

foreach ($items as $item) {

$description = untag ($item, 'description');

$title = untag($item, 'title');

$link = untag($item, 'link');

$html .= '<a href="' . $link[0] . '">' . $title[0] . "</a><br />\n";

$html2 .= strip_tags($description[0]);

$descricao=explode('(&lt;i&gt;',$description[0]);

//limited number of items

$a++;

if($a==4) {

exit; }

//end limited number of items

print $descricao[0];

print '<br><a href="' . $link[0] . '">' . $title[0] . "</a><br />\n";

}

I want to know what the HTML code would be for the XML code at http://rss.groups.yahoo.com/group/nascar_pitstop/rss. How would you make the code?

Very nice work. I've been toying with this for hours here at work in textpad (it's so difficult to pull these things together when you have no place to debug it either!)

This article has really helped me out. Thanks:)

I was grabbing the top five stories from Yahoo and wanted to show the first 100 characters of the description. I wrote a quick function that starts at the 100th character and loops until it finds a space and a suitable break point.

function strclnup($description) {

for($i = 100; $i <= 150; $i++):

if (substr($description,$i,1) == " ") {

return substr($description,0,$i);

break;

}

endfor;

}

$html = '<p>';

$id=0;

foreach ($items as $item){

if($id=="5") {

break;

}

$id++;

$title = untag($item, 'title');

$link = untag($item, 'link');

$descr = untag($item, 'description');

$html .= "<a href='" . $link[0] . "' target='_new'>" . $title[0] . "</a><br>" . strclnup($descr[0]) . "...<br>\n";

}

$html .= "</p>";

Hi all, my name is Daniele Leone, i read someone ask for reading news from Yahoo!. I just built a simple but powerfull script that make you able to search in yahoo news feeds. Try the demo here http://www.danieleleone.com and download it for free if you like ! ;-) Bye, Daniele

What a great XML feed routine, thanks! I use it succesfully taking feeds from Silicon.com for my site www.FirstAidforComputers.com

By the way SteveT I use a routine to shorten descriptions as follows:

$sh = array_slice(explode(" ",$description),0,100);

$shortened = implode(" ",$sh);

Hope it comes in useful.

Instead of exploding the string in to an array just to convert it back, a similar thing can be done using a regular expression all in one line:

$title = preg_replace("/(.{20}.)\s.$/","\${1}...",$title);

Here the code trims the title to 20 characters but keeps the last word as a complete word and places an elipse ('...') at the end if the length is over this.

Um... seems as though the server decided to chew my code up! Let's try that again:

$title = preg_replace("/(.{20}._ASTERIX_)\s._ASTERIX_$/","\${1}...",$title);

where _ASTERIX_ is an asterix symbol (as in one of those star characters - shift-8 on a Mac).

Hope that helps!

how do I grab a password protected feed if I know the password?

If it seems to good to be true, it probably isn't. If this script worked, it would have been great...

I just get a blank page. I'm using PHP5. Does anybody knows why it's not working?

I used your second method and it gives the links to the newsfeeds now.

Thanks a million. I stand corrected. This is the real thing.

I used thiscode to display the pubDate localtime, but there must be a better way

$pubDate = untag($item, 'pubDate');

//Tue, 02-11-2004 02:10:49 GMT

//Thu, 14 Oct 2004 08:21:00 GMT

$Section = substr($pubDate[0], 5);

$day = substr($Section, 0, 2);

$mth = substr($Section, 3, 2);

$match = preg_replace("/.[0-9]/", "match", $mth);

if( $match == "match"){

$yrs = substr($Section, 6, 4);

$hrs = substr($Section, 11, 2);

$min = substr($Section, 14, 2);

$sec = substr($Section, 17, 2);

$newpubDate = mktime( $hrs, $min, $sec, $mth, $day, $yrs, 1);

$newpubDate = date("D, j M \a\\t g:i a", $newpubDate);

}

else

{

$mth = substr($Section, 3, 3);

$mth = strtotime("10 $mth 2000");

$mth = date("m",$mth);

$yrs = substr($Section, 7, 4);

$hrs = substr($Section, 12, 2);

$min = substr($Section, 15, 2);

$sec = substr($Section, 18, 2);

$newpubDate = mktime( $hrs, $min, $sec, $mth, $day, $yrs, 1);

$newpubDate = date("D, j M \a\\t g:i a", $newpubDate);

}

BTW excellent script, easy to impliment, and no php_extensions required!

http://xoap.weather.com/weather/local/ASXX0083?cc=*&unit=m&par=1004027791&key=b046c68db097b8d1

How do I read tags with ID numbers

As Wasabi mentions in the post above, tags with content inside of them do in fact exist now. I've encountered one at this feed (http://www.core77.com/corehome/index.xml). Specifically they embed some RDF data in the XML tag for Item. The item tag reads like this:

<item rdf:about="http://www.example.com/stories/001.html">

The easiest answer would seem that I should write a custom function JUST to handle the parsing of the $xml variable to extract the information into the $items array. The problem is, each of the articles listed has a different URL embedded inside its <item> tag... however I don't know how to write wildcards in PHP!!

Golly gee whillikers it would sure be neat to know how to handle this....

well I "kind of" solved this problem. First off, I made a duplicate of the untag function and gave it a new name (lets call it 'workaround_link'). Turns out that you really don't NEED to include the ENTIRE opening tag for the variable $preg, just the closing carat of the tag. So it could look like this:

$preg = "|>(.*?)</$tag>|s"

All I know is that it WORKS!!!! (and that I can now rest)