<?php
/*
    instantRSS.php
    version   : 1.3 (2004.04.29)
    copyright : ben wilson 2003-2004
    email     : ben[at]thelocust[dot]org
    site      : http://thelocust.org/projects/instantrss

    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with this program; if not, write to the Free Software
    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

    (or you can peep it here: http://www.gnu.org/licenses/gpl.txt)
*/

/* CHANGELOG                                                                  */
/* v1.3 - (2004.04.29)                                                        */
/* updated RSS output to RSS 2.0, and none of this RDF funny business         */
/* output now validates with archive.org and scripting.org's rss validators   */
/*                                                                            */
/* improved HTML entity handling.  by default, we now use php's htmlentities  */
/* to turn all html entities into their & codes, and after that step we turn  */
/* the resulting &'s into &amp;. I swear it works.                            */
/*                                                                            */
/* v1.2 - (2003.10.15)                                                        */
/* minor fix to ensure that the arrValidURL array is initialized              */
/* instantRSS tags.  thanks to nicolas kicillof (nicok[at]dc[dot]uba[dot]ar   */
/*                                                                            */
/* v1.1 - (2003.07.14)                                                        */
/* minor fix for correctly grabbing HTML-commented values from between        */
/* instantRSS tags.  thanks to ronald heft                                    */
/*                                                                            */
/* v1.0 - initial release                                                     */

/* DESCRIPTION                                                                */
/* instantRSS is a simple RSS creator for folks who don't have RSS capability */
/* for their site/blog/whatever.  As long as a page is marked up appropriately*/
/* with the <!--instantRSS:xxxxx--> tags, then instantRSS can spit out an RSS */
/* file!                                                                      */

/* instantRSS can be used two ways: remotely or locally.  both methods of     */
/* of usage will return the same output.                                      */

/* Remote usage implies that the instantRSS script resides on a server that   */
/* does not host your site/blog. This is handy for folks that don't have any  */
/* type of scripting on their hosts, or just don't want to fool with it.  If  */
/* you'd like to use thelocust.org's instantRSS script to generate your RSS   */
/* then check out instantRSS's site listed above for the url to use           */

/* Local usage implies that the instantRSS script resides on the same server  */
/* that hosts your site, and that instantRSS script is used *only* for your   */
/* instantRSS marked pages.                                                   */  

/* In either case, to give instantRSS a URL to parse:                         */
/* http://yourblog.com/instantRSS.php?url=http://yourblog.com/index.html      */
/* just specify the page to get the RSS from in the url= variable in the URL  */
/* (replacing yourblog.com with the name of the server with instantRSS on it) */

/* instantRSS markup tags                                                     */
/* i'm not going to explain RSS here, because i assume you already know what  */
/* all that is about. http://voidstar.com/rssfaq should provide more info.    */

/* all instantRSS tags are in this format: <!--instantRSS:xxxxx--> where      */
/* xxxxx is the name of the tag.                                              */

/* example instantRSS markup */
/*
<!--instantRSS:beginchantitle-->
  Ben's Blog
<!--instantRSS:endchantitle-->

<!--instantRSS:beginchanlink-->
  http://thelocust.org
<!--instantRSS:endchanlink-->

<!--instantRSS:beginchandesc-->
  the blog of ben wilson
<!--instantRSS:endchandesc-->

<!--instantRSS:beginitem-->
      <!--instantRSS:begintitle-->
        February 11, 2003
      <!--instantRSS:endtitle-->

      <!--instantRSS:beginlink-->
      <!--http://thelocust.org?newsid=364-->
      <!--instantRSS:endlink-->

      <!--instantRSS:begindesc-->
        Man, I've got SO MUCH news today!  The sun is shining!
      <!--instantRSS:enddesc-->
<!--instantRSS:enditem-->

*/

/* instantRSS will use whatever is in between the two corresponding tags      */
/* as the value for that tag.  for instance, the channel title in the above   */
/* example is "Ben's Blog".  instantRSS will also take values in HTML comments*/
/* as seen in the item link tag in the example above.  This allows you to hide*/
/* values so that they are not see on your actual site.  This may be done with*/
/* ANY valid instantRSS tag.  When outputting that hidden value, the <!-- and */
/* --> header and footer will be removed automagically.                       */

/* instantRSS TAGS */

/* CHANNEL TAGS                                                               */
/* title - (beginchantitle / endchantitle).  if instantRSS fails to find this */
/*   tag, it will attempt to find and use the <title></title> tag which is    */
/*   found in the <header></header> of most pages.                            */
/* link  - (beginchanlink / endchanlink). if instantRSS fails to find this    */
/*   tag, it will use the link passed to it via the URL variable.             */
/* desc  - (beginchandesc / endchandesc). if instantRSS fails to find this tag*/
/*   it will attempt to use the META description tag found on many pages in   */
/*   in the <header>. e.g.:                                                   */
/*   <meta name="description" content="AE4RV.com is....">                     */

/* ITEM TAGS                                                                  */
/* item - (beginitem / enditem). Note that each item must begin and end with  */
/*   the corresponding tag.  all item sub-tags must be within the beginning   */
/*   and ending item tags                                                     */
/* item title - (begintitle / endtitle)                                       */
/* item link  - (beginlink / endlink)                                         */
/* item desc  - (begindesc / enddesc)                                         */


/* NOTES */

/* HTML will be stripped out of ALL tag values, unless specified through the  */
/* HTML variable as such:                                                     */
/*   http://yoursite.com/instantRSS.php?url=http://thelocust.org&html=1       */
/* There are 3 possible values for HTML=:                                     */
/*   0 - no HTML in RSS output                                                */
/*   1 - any and all HTML in RSS                                              */ 
/*   2 - escaped HTML in RSS (spaces become %20, etc)                         */ 

/* HOW TO LINK TO IT */
/* http://yourserver.com/instantRSS.php?url=http://yourserver.com/index.html  */


/* HACKING NOTES */
/* You want to hack on this and send me a patch?  AWESOME.  Send your changes */
/* (in a patch or as the whole file) to ben[at]thelocust[dot]org              */
/* Please use spaces instead of tabs when indenting.  Tabs are the work of    */
/* the devil.                                                                 */


/* CONFIGURATION                                                              */

/* URL RESTRICTION                                                            */

/* set bRestrictURL to 1 to restrict instantRSS to ONLY parsing URLs from the */
/* $sValidURL list. If $bRestrictURL is 0, then whatever is passed to instant */
/* RSS in the URL variable will be parsed.  use at your own risk!!            */
$bRestrictURL = 0;


/* set arrValidURL to the URLs that instantRSS is allowed to accept. use commas */
/* to delimit. i.e. array("thelocust.org","ae4rv.com");                       */
/* please note that this should ALWAYS be at least defined like:              */
/* $arrValidURL = array();                                                    */
/* even if it is empty.                                                       */
//$arrValidURL = array();
$arrValidURL = array("thelocust.org","ae4rv.com");
//$arrValidURL = array();


/* set iHTMLLevel to set the level of HTML compatibility.                     */
/*   0 - no HTML in RSS output                                                */
/*   1 - any and all HTML in RSS                                              */ 
/*   2 - escaped HTML in RSS (spaces become %20, etc)                         */ 
/* NOTE: setting this variable will override an value passed in the URL.      */
/*   leave it commented out (with "//") to not set it.                        */ 
//$iHTMLLevel = 0;


if (!isset($iHTMLLevel) && isset($_GET['html']))
  $iHTMLLevel = $_GET['html'];
else
  $iHTMLLevel = 0;

if (isset($_GET['url'])) {
  $URL = $_GET['url'];
  if (substr($URL,0,7) != 'http://') {
    $bFailure = TRUE;
    echo "<BR>The URL you have provided '$URL' does not start with http://, and it should!";
  }
  else {
    $tmp = explode("/",substr($URL,7,strlen($URL)-7));
    if ( is_array($arrValidURL) ) {
      if ( !in_array($tmp[0],$arrValidURL) && $bRestrictURL == 1  ) {
        $bFailure = TRUE;
        echo "<BR>The URL specified (<a href=\"$URL\">$URL</a>) is not in the list of valid server URLs!";
      } 
    }
    else {
      echo "<BR>You MUST initialize the $arrValidURL array, even if it is to be empty.";
    }
  }
}
else
  $URL = "";


if ($bFailure)
     echo "<br><br>FAILURE!<br>Some sort of horribly catastrophic failure has occured.  See the above error messages for details.";
else {
  //read into an array lines from the URL
  $file = file($URL);

  //make sure the file read was successfull
  if (!$file)
    echo("The URL you have given me either contains no data or could not be read.<br>".
	 "<a href=\"$URL\">Click here</a> to go to the URL you specified.");
  else {

    $file = implode("",$file);
    $itemtags = array('title','link','desc');
    $items = array();

    //first up, split the page using the begin item tag as a delimiter
    //thus, we will have an array with the beginning of the page as index 0,
    //the first item at 1...
    $arrfile = explode('<!--instantRSS:beginitem-->',$file);

    //find CHANNEL TITLE
    $tmp = split("<!--instantRSS:beginchantitle-->",$arrfile[0]);
    $tmp = split("<!--instantRSS:endchantitle-->",$tmp[1]);
    if (strlen($tmp[0]))
      $channeltitle = stripHTML($tmp[0]);
    else {
      $tmp = split("<title>",$arrfile[0]);
      $tmp = split("</title>",$tmp[1]);
      $channeltitle = stripHTML($tmp[0]);
    }

    //find CHANNEL DESC
    $tmp = split("<!--instantRSS:beginchandesc-->",$arrfile[0]);
    $tmp = split("<!--instantRSS:endchandesc-->",$tmp[1]);
    if (strlen($tmp[0]))
      $channeldesc = stripHTML($tmp[0]);
    else {
      $tmp = split("<meta name=\"description\" content=\"",$arrfile[0]);
      $tmp = split("\">",$tmp[1]);
      $channeldesc = stripHTML($tmp[0]);
    }

    //find CHANNEL LINK
    $tmp = split("<!--instantRSS:beginchanlink-->",$arrfile[0]);
    $tmp = split("<!--instantRSS:endchanlink-->",$tmp[1]);
    if (strlen($tmp[0]))
      $channellink = stripHTML($tmp[0]);
    else
      $channellink = stripHTML($URL);


    //for each item in the array, starting with index of 1
    for ($i=1;$i<=sizeof($arrfile);$i++) {

      $items[$i] = array();

      //take the item, and split it using the enditem tag as delimiter
      //this will give us an array with the actual item in index 0, and the leftover 
      //crap in index 1
      $item = explode('<!--instantRSS:enditem-->',$arrfile[$i]);

      //set the item to the item text in index 0
      $item = $item[0];

      //for each tag in "itemtags"
      for ($x=0;$x<sizeof($itemtags);$x++) {
        //take this cleaned up item, and split it using the begin tag as delimiter
        //leaving us with an array with the title+restofitem in index 1
        $arritem = explode("<!--instantRSS:begin$itemtags[$x]-->",$item);
        
        //if we didn't find a begin title tag, then set the title to item #X
        //else splie the title+restofitem using the endtitle tag!
        if ( sizeof($arritem) <= 1 ) {
          if ($itemtags[$x] == 'link')
	    $items[$i][$x] = $URL;
          elseif ($itemtags[$x] == 'title')
            $items[$i][$x] = "item #$i";
          elseif ($itemtags[$x] == 'desc')
            $items[$i][$x] = "no description";
        } //if sizeof
        else {
          $tmp = explode("<!--instantRSS:end$itemtags[$x]-->",$arritem[1]);
          $items[$i][$x] = stripHTML($tmp[0]);
        } //if sizeof else
      }  //for sizeof itemtags

    } //for each newsitem


/* START OUTPUT */

if (!headers_sent()) {

//print the XML header so the browser or whatever is picking this up
//will know wtf it is!
if (!isset($_GET['debug']))
  header ("content-type: text/xml");
print '<?xml version="1.0"?>';

?>

<!--
RSS generated by instantRSS (http://thelocust.org/projects/instantrss) 
for <?=$channeltitle?> (<?=$channellink?>)
at <?=date('r')?>

-->
<rss version="2.0" xmlns:blogChannel="http://backend.userland.com/blogChannelModule">

<channel>
<title><?=$channeltitle?></title>
<link><?=$channellink?></link>
<description><?=$channeldesc?></description>


<?php
  //start individual ITEMs

  //for each item, again
  for($j=1;$j<sizeof($items);$j++) {
?>

<item>
  <title><?=$items[$j][0]?></title>        
  <link><?=$items[$j][1]?></link>        
  <description><?=$items[$j][2]?></description>        
</item>

<?php
    } //end for items

?>
</channel>
</rss>

<?php
	} //if headers sent

  } //if file was good


} // if a URL has been specified


function stripHTML($string) {
  global $iHTMLLevel;

  $string = trim(str_replace("-->","",str_replace("<!--","",$string)));

  
  //escape HTML
  if ($iHTMLLevel == 2)                            //escapeHTML
    $string = urlencode(htmlentities($string));
  elseif ($iHTMLLevel == 1)                        //leave HTML alone
    $string = $string;
  elseif ($iHTMLLevel == 0 || !isset($iHTMLLevel)) //strip all HTML
    $string = str_replace('&','&amp;',rssencode(htmlentities(strip_tags($string))));
  
  //return a trimmed, un-HTML-commented string
  return $string;
}


//rss encoding function
//from http://www.pmichaud.com/examples/pmwiki/rss.php
//Copyright 2002-2004 Patrick R. Michaud (pmichaud@pobox.com)  
function rssencode($s) 
{ return preg_replace('/([\\x80-\\xff])/e',"'&#'.ord('\$1').';'",$s); }

?>