[php] wyświetlanie RSS z polskimi znakami

[kruzyk]

Witam.

No i mam problem.

Mam skrypt do wyświetlania RSS na stronie, jednak wyświetla tekst bez polskich znaków. Zamiast nich są krzaki.

Co zmienić i jak naprawić?

# NewsParser2 - Etomite 0.6.1 - 2005-01-29
# By: ColdRain - coldrain@workingonit.org
#
# RSS parser shamelessly adapted from the RSS_feed class
# see https://www.phpclasses.org/browse/package/1299.html
#
# This parser allows for more sorts of feeds 
# and by using a simple URL translation allows for parsing of URLs
# which use variables within the URL
# The class can also limit the number of items shown and show the image
# For interoperatebility, the same CSS classes are used
# as in rad's NewsParser snippet;-)
# 
# usage:
# [[NewsParser2?url=https://kanaly.rss.interia.pl/finanse.xml]]
#
# original credits and comments kept:
/********************************************************************
Class: RSS_feed
Author: Dr. Timothy Sakach
Version 2.0
This script will parse a RSS/XML file that comes from a URL feed.
It will return an HTML unordered list.

Getting a feed from another web site, should be quick and simple. Any
unneeded complications should be avoided. One feed processor from a
large Content Management System written in PHP nearly got me kicked 
off my server. It ran wild with overly complex coding and included 
opening hundreds of objects in a loop and used the XML-DOM (ugh!).

This is far simpler, faster and more reliable.

There are several versions of RSS/RDF. However, in practical use
most all feed implementations follow the same pattern: A publication link is
contained within <item> tags and consists of a <title>, <link>, and
a <description>. We tested feeds from many different sources using
all flavors and versions and this class was able to parse all of them.

Granted there are many options with version 1.0 (RDF). However,  
the full RDF syntax and vocabulary are not needed in a feed. 
All of the implementations we found that included RDF tags 
really added very little to the feed and  were safely ignored. 

Version 2.0 extends versions 0.9x by adding new tags. This presented
no problems and this class can be extended to process those tags. But 
our goal was to provide a simple solution that can easily add syndicated
publications to any web site.

 Changes: by Dr. Timothy Sakach
 1. 8/31/03 Corrected bug because the description tag (and all other tags, for that
 matter} can occur in any sequence within the container tags. 
 2. 8/31/03 Corrected bug created by the way the XML parser returns parsed cdata.
 3. 8/31/03 Added Set_Limit property to control the number of links to show.
 4. 8/31/03 Added image control and Show_Image property.
 5. 8/31/03 Channel, Image, and Items now use arrays as buffers.
********************************************************************/

// defaults and variable translation

$url = str_replace ( "|xq|", "?", $url );
$url = str_replace ( "|xe|", "=", $url );
$url = str_replace ( "|xa|", "&", $url );
$url = htmlentities($url);

if ( $url == "" ) $url = "https://www.php.net/news.rss";
if ( $limit == "" ) $limit = "4";
if ( $image == "" )
{ $bimage = false; }
else
{ $bimage = true; }

if ( $descr == "" )
{ $bdescr = true; }
else
{ $bdescr = true; }

// end of defaults and variable translation
//
// see the end of this snippet where the below class is called

#
# +++++++++++++++++ beginning of class ++++++++++++
#

class RSS_feed 
{
 // The object of the parser is to determine the "State" of the RSS/XML
 // and set up the class to respond accordingly. The critical information
 // is in the _handle_character_data function as that is where the 
 // feed information can be found.

 var $flag;	  // To control the state of the unordered list
 var $state;	 // To determine which element tag is being worked
 var $level;	 // Simple XML level control
 var $output;	// Where the results will be stored
 var $showdesc;  // A flag to indicate whether or not to show the description.
 var $showimage; // A flag to indicate whether or not to show the image.
 var $URL;	   // The location of the external feed.
 var $psr;	   // Our parser object.
 var $contents;  // The RSS/XML from the feed
 var $rss_version; // Stores version "number"
 var $limit;	 // The maximum number of links we want
 var $channel;   // Array of channel cdata
 var $image;	 // Array of image cdata
 var $item;	  // Array of item cdata
 var $channelclosed; // To indicate state if channel closes before image tags.

// Define the functions required for handling the different pieces.

 function RSS_feed()
 {
// Constructor
$this->output = "";
$this->channel = array();
$this->image = array();
$this->item = array();
$this->flag = false;
$this->state = 0;
$this->level = 0;
$this->showdesc = false;
$this->showimage = false;
$this->channelclosed = false;
$this->version = 9;   // set 0.9x as default;
$this->limit = 0;   // use 0 as default == all

 }

// METHODS AND PROPERTIES *****************************************

 function Show_Description($tf)
 {
// By default the description is not included in the results
// This public function allows for the description to be
// included, if desired.
if (!$tf === false) 
{
  $this->showdesc = true;
} else {
  $this->showdesc = false;
}
 }

 function Show_Image($tf)
 {
// By default the image is not included in the results
// This allows for the image to be included, if desired.
if (!$tf === false)
{
  $this->showimage = true;
} else {
  $this->showimage = false;
}
 }

 function Set_URL($url)
 {
// This is the URL to the feed. The class expects that RSS/XML will
// be returned.
$this->URL = $url;
// Knowing this, we can get the feed contents now.

// Get the RSS/XML from the feed URL
$this->_load_file();

// Check the version of the XML and set the version state
$this->_get_rss_version();
 }

 function Set_Limit($cnt)
 {
// This property sets the limit of links to return
// if $cnt is not numeric, 0 is returned! You get the entire list!
$i = intval($cnt);
if ($i > 0) $this->limit = $i;
 }

 function Get_Results()
 {
// When the properties have been set, then this function should
// be called. It will return the HTML unordered list.

$c = $this->contents;

// Create the parser and set handlers.
$this->psr = xml_parser_create();

xml_set_object($this->psr, $this);
xml_parser_set_option($this->psr,XML_OPTION_CASE_FOLDING,1);

// Set the parser element handlers based upon the version.
switch ($this->version)
{
  case 9:
  case 2;
	xml_set_element_handler($this->psr, '_handle_open_element', '_handle_close_element');
	break;
  case 1:
	xml_set_element_handler($this->psr, '_rdf_handle_open_element', '_handle_close_element');
	break;
}
// Set the handler for the cdata
xml_set_character_data_handler($this->psr, "_handle_character_data");

// Parse it.
if (!xml_parse ($this->psr, $c))
{
  // This returns an error message if the RSS/XML cannot be parsed. 
  // Too bad.
  // This indicates a bad or malformed feed!
  $ln =  xml_get_current_line_number($this->psr);
  $msg =  xml_error_string(xml_get_error_code($this->psr));
  return "An XML error occurred on line $ln: $msg";
}

// Free up the parser and clear memory
xml_parser_free($this->psr);
$this->contents = "";

// return the results
return $this->output;
 }

//**************************************************************
// HANDLER FUNCTIONS

 function _handle_open_element (&$p, &$element, &$attributes)
 {
// parser for rss version 0.9x and 2.0
// Set the state of the class for the benefit of the cdata handler.
$element = strtolower($element);

switch($element) 
{
  case 'rss':
	// data at this level may not be needed
	$this->level = 0;
	$this->state = 0;
	break;
  case 'channel':
	// The channel may have a title and a link and a description
	// This data will not be part of the list, but will precede it.
	$this->flag = true;
	$this->level = 1;
	$this->state = 1;
	break;
  case 'item':
	// We have an item to process
	$this->level = 3;
	$this->state = 3;
	break;
  // some tags that will appear under 'channel'
  // for now, ignore these tags. 
  case 'pubdate':
  case 'managingeditor':
  case 'webmaster':
  case 'width':
  case 'height':
  case 'language':
	$this->state = 99;
	break;
  case 'image':
	// This assumes the image is a container element, as this is most common
	$this->level = 2;
	break;
  case 'title':
	$this->state = 4;
	break;
  case 'link':
	$this->state = 5;
	break;
  case 'description':
	$this->state = 6;
	break;
  case 'url':
	$this->state = 7;
	break;
  default:
	// ignore any undefined tags
	$this->state = 99;
	break;
}
 }

 function _rdf_handle_open_element(&$p, &$element, &$attributes)
 {
// RDF mixes things up a bit so we need to pay attention
// However, when you get right down to it. There may be
// no difference in the RDF feed other than more stuff to
// ignore.

$element = strtolower($element);

// Include Dublin Core tags and full RDF specs? 
// Nah!
switch($element)
{
  case "rdf:rdf":
	// The parser takes care of the Namespace, so we can ignore this.
	$this->level = 0;
	$this->state = 0;
	break;
  case "channel":
	// Same as above.
	$this->flag = true;
	$this->level = 1;
	$this->state = 1;
	break;
  case "image":
	// Assumes <image> is a container element. This is most common.
	$this->level = 2;
	break;
  case "items":
	// channel parameters stop and items begin. We can ignore this.
	$this->state = 99;
	break;
  case "item":
	// We have an item to process.
	$this->level = 3;
	$this->state = 3;
	break;	
  case "title":
	$this->state = 4;
	break;
  case "link":
	$this->state = 5;
	break;
  case "description":
	$this->state = 6;
	break;
  case "url":
	$this->state = 7;
	break;
  // These next two are somewhat redundant, unless strict RDF format is followed.
  // Ho hum ...
  // We will ignore them.
  case "rdf:seq":
  case "rdf:li":
  default:
	// ignore tags
	$this->state = 99;
	break;
}
 }

 function _handle_character_data(&$p, &$cdata)
 {
/* 
 This function is trivialized in many examples. However, this is
 where the real action lies. We have set the state of the class in order 
 to determine what we should do with cdata. 

 Changes had to be made here because the PHP parser, under special conditions
 would parse the data in the elements and return it in pieces. This showed
 up when the <title> contained ' or " 

 This function only accumulates $cdata text in arrays.
*/

// Ignore $cdata filled with blanks or nothing
$s = trim($cdata);
if (strlen($s)) {
  // We really only need these things now: Title, Links and, if desired, Description and Image URL
  switch ($this->state) 
  {
	case 4:  // title
	  switch ($this->level) 
	  {
	  // This are the only levels that are important here.
		case 1:  // Channel
		  $this->channel["title"] .= $cdata;
		  break;
		case 2; // Image
		  $this->image["title"] .= $cdata;
		  break;
		case 3:  // Item
		  $this->item["title"] .= $cdata;
		  break;
	  }
	  break;
	case 5:  // link
	  switch ($this->level) 
	  {
		case 1: // Channel
		// Make the link for the channel and change to item level. We're done here.
		  $this->channel["link"] .= $cdata;
		  break;
		case 2: // Image
		  $this->image["link"] .= $cdata;
		  break;
		case 3: // Item
		// Make the link for the item. Reset the flag and initialize the unordered list.
		// Add the link for the item

		  $this->item["link"] .= $cdata;
		  break;
	  }
	  break;
	case 6:  // description
	  // If the description is desired, add it now.
	  if ($this->showdesc)
	  {
		switch ($this->level) 
		{
		  case 1: // Channel
			$this->channel["desc"] .= $cdata;
			break;
		  case 3: // Item
			$this->item["desc"] .= $cdata;
			break;
		}
	  }
	  break;
	case 7: // Image url
	  if ($this->showimage)
	  {
		switch ($this->level)
		{
		  case 2: // Image
			$this->image["url"] .= $cdata;
			break;
		}
	  }
	  break;
  }
}
 }

 function _handle_close_element(&$p, &$element)
 {
// Closing elements for all versions.
// Because the elements can appear in orders differing from each other
// the output is now created at the close of each of the critical elements.
$element = strtolower($element);
static $cnt;
switch ($element)
{
  case 'channel':  // major elements -- define closing event
	// put channel information on the top. This should work even if the channel close element
	// occurs before the item tags
	if ($this->showimage && $this->image["link"])
	{
	  $outbuffer .= '';
	}
	$outbuffer .= '' 
	  . $this->channel["title"] = "";
	if ($this->showdesc)
	  $outbuffer .= $this->channel["desc"] = "";
	$outbuffer .= $this->output;
	$this->output = $outbuffer;
	$this->channel["link"] = "";
	$this->channel["title"] = "";
	$this->channel["desc"] = "";
	$this->channelclosed = true;
	break;
  case 'image': // Image tags closed
	// A little tricky here as image is often part of the channel container tags
	// Some feeds close the channel element and then provide the image.
	// Image has its own close element, unless the values are provided by Attributes!
	if ($this->channelclosed)
	{
	  //Put the image on the top
	  if ($this->showimage && $this->image["link"])
	  {
		$outbuffer .= '';
		$this->output = $outbuffer . $this->output;
	  }
	}
	break;
  case 'item':
	// Each item has its own close element
	if ($this->flag) 
	{
	  // Initialize list of newsitems
	  $this->flag = false;
	  $cnt = 0;
	}
	if ($this->limit > $cnt || !$this->limit)
	{
	  $this->output .= "<fieldset class='rss_fieldset'>";
	  $this->output .= "<legend class='rss_legend'><b>";
	  $this->output .= $this->item["title"];
	  $this->output .= "</b></legend>";

	  if ($this->showdesc)
		$this->output .= $this->item["desc"];

	  $this->output .= "<br /><br /></fieldset>";

	  if ($this->limit > 0) $cnt++;
	}
	$this->item["link"] = "";
	$this->item["title"] = "";
	$this->item["desc"] = "";
	break;
  default: // ignore all other close elements
	break;
}
 }

 function _load_file()
 {
// Get the raw feed from the URL. Because this uses a URL as the feed source
// it can be used to process an RSS/XML feed from any web site, including local.
$data = @file($this->URL) or die ("Could not open a feed called: " . $this->URL);
$this->contents = implode('',$data);  // Put data into an array
 }

 function _get_rss_version()
 {
// Set the version state
if (strpos($this->contents,'version="0.9'))  $this->rss_version = 9;
elseif (strpos($this->contents,"rdf:")) $this->rss_version = 1;
elseif (strpos($this->contents,'version="2.0"')) $this->rss_version = 2;
 }

}  // end of class

#
# +++++++++++++++++ end of class ++++++++++++
#

$px = new RSS_feed();

$px->Set_URL($url);
$px->Set_Limit($limit);		// 0 = Default = All
$px->Show_Image($bimage);		// Default = false
$px->Show_Description($bdescr);	// Default = false

// $output .= "<br>" . $url . "<br>"; // for debugging
$output .= $px->Get_Results();

return $output;

[MMP]

Strzelam że Twoja strona ma inne kodowanie niż RSS?

[kruzyk]

No właśnie strona ma kodowanie iso-8859-2.

A RSS ma kodowanie utf-8.

Jak to naprawić ? (bez zmiany kodowania strony )

[MMP]

Użyj biblioteki iconv, lub mbstring. Zależy którą posiada Twój serwer.

[kruzyk]

Ale o so chozii?

To nie wystarczy czegoś dopisać do tego skryptu?

ps. gdybym się znał nie używałbym gotowych skryptów tylko pisałbym własne

[kruzyk]

Z tego co wyczytałem muszę dodać

echo iconv("UTF-8","ISO-8859-2","TREŚĆ RSS'a");

Ten fragment odpowiada za wyświetlanie tekstu:

{
	  $this->output .= "<fieldset class='rss_fieldset'>";
	  $this->output .= "<legend class='rss_legend'><b>";
	  $this->output .= $this->item["title"];
	  $this->output .= "</b></legend>";

	  if ($this->showdesc)
		$this->output .= $this->item["desc"];

	  $this->output .= "<br /><br /></fieldset>";

Jak mam tego użyć?

[MMP]

Zamiast "TREŚĆ RSS'a" daj $this->output

[kruzyk]

Kurde nie mogę sobie poradzić...

Kombinowałem tak:

{
	  $this->output .= "<fieldset class='rss_fieldset'>";
	  $this->output .= "<legend class='rss_legend'><b>";

echo iconv("UTF-8","ISO-8859-2",$this->output);

	  $this->output .= $this->item["title"];
	  $this->output .= "</b></legend>";

	  if ($this->showdesc)

echo iconv("UTF-8","ISO-8859-2",$this->output);

		$this->output .= $this->item["desc"];

	  $this->output .= "<br /><br /></fieldset>";

{
	  $this->output .= "<fieldset class='rss_fieldset'>";
	  $this->output .= "<legend class='rss_legend'><b>";

echo iconv("UTF-8","ISO-8859-2",$this->output);

	  $this->output .= $this->item["title"];
	  $this->output .= "</b></legend>";

	  if ($this->showdesc)
		$this->output .= $this->item["desc"];

	  $this->output .= "<br /><br /></fieldset>";

{
	  $this->output .= "<fieldset class='rss_fieldset'>";
	  $this->output .= "<legend class='rss_legend'><b>";

echo iconv("UTF-8","ISO-8859-2",$this->output .= $this->item["title"]);


	  $this->output .= "</b></legend>";

	  if ($this->showdesc)

echo iconv("UTF-8","ISO-8859-2",$this->output .= $this->item["desc"]);

	  $this->output .= "<br /><br /></fieldset>";

{
	  $this->output .= "<fieldset class='rss_fieldset'>";
	  $this->output .= "<legend class='rss_legend'><b>";

echo iconv("UTF-8","ISO-8859-2",$this->item["title"]);

	  $this->output .= $this->item["title"];
	  $this->output .= "</b></legend>";

	  if ($this->showdesc)

echo iconv("UTF-8","ISO-8859-2",$this->item["desc"]);

		$this->output .= $this->item["desc"];

	  $this->output .= "<br /><br /></fieldset>";

Niestety za każdym razem osiągałęm niezamierzony wynik. Najlepsze jest to, że w częsci przypadków było poprawna translacja z utf na iso ale tekst nie podporządkowywał się atrybutom i pojawiał się np. na samej górze strony, nad samym logo.

/edit:

Adres wysłałem na PW.

[MMP]

Pod

$this->output .= "<br /><br /></fieldset>";

daj

$this->output = iconv( 'UTF-8', 'ISO-8859-2', $this->output ); //Konwersja z utf-8 do iso-8859-2

[kruzyk]

MMP, to co podałeś prawie nic nie wyświetlało więc dodałem kropkę

$this->output .= iconv( 'UTF-8', 'ISO-8859-2', $this->output );

Jednak coś z tym $this->output jest nie tak, bo powtarza mi te same newsy. Na dodatek raz je wyświetla poprawnie (bez krzaków), poniżej ten sam news ale już z krzakami. Jeszcze niżej między innymi ten sam news ale obcięty dziwnie...

Nic nie zmieniam na stronie żebyś mógł to zaobserwować.

Dziękuję Ci za pomoc i mam nadzieję, że mi pomożesz.

PS. to powiadamianie o odpowiedziach chyba już przestało działać na PiO... !!!

[MMP]

Ahh, nie przejrzałem całego kodu. Zamiast tego co wcześniej wkleiłeś daj:

$output = "<fieldset class='rss_fieldset'>";
	  $output .= "<legend class='rss_legend'><b>";

	  $output .= $this->item["title"];
	  $output .= "</b></legend>";

	  if ($this->showdesc)


		$output .= $this->item["desc"];

	  $output .= "<br /><br /></fieldset>";
$this->output .= iconv( 'UTF-8', 'ISO-8859-2', $output );

[kruzyk]

MMP, jesteś wielki Chociażby za cierpliwość.

Dziękuję i pozdrawiam.