
平台论坛博客文库

› 论坛 › 程序设计 › PHP › [RSS解析原创]RSS文档解析类

[RSS解析原创]RSS文档解析类 [复制链接]

lovered

白手起家

论坛徽章:: 0

电梯直达

1楼 [收藏(0)] [报告]

发表于 2005-09-27 15:35 |只看该作者 |倒序浏览

<?PHP
/*****************************
* RSS文档解析类-(lovered.GV)
******************************/
class RssParse
{
var $url; //rss文件的地址
var $data; //rss文件的内容
var $version; //rss文件的版本号
var $channel; //rss文件中的频道信息
var $items;
//与XML解析有关的属性####################
var $xml_parser; //xml解析器句柄
var $depth; //XML当前解析深度
var $tag; //当前正在解析的XML元素
var $prev_tag; //当前正在解析的上一个元素
var $marker; //用来标记制定的深度
var $event; //实践名称:CHANNEL and ITEM
var $item_index; //item元素索引
//初始化#################################
function RssParse($rss_url)
{
$h=fopen($rss_url,"r");
if($h)
{
$this->;url=$rss_url;
while(!feof($h))
$this->;data.=fgets($h,4096);
fclose($h);
//初始化xml解析器
$this->;xml_parser = xml_parser_create("UTF-8");
xml_set_object($this->;xml_parser, &$this);
xml_parser_set_option($this->;xml_parser, XML_OPTION_CASE_FOLDING, 1);
xml_set_element_handler($this->;xml_parser, "startElement", "endElement");
xml_set_character_data_handler($this->;xml_parser, "characterData");
//开始解析数据
if (!xml_parse($this->;xml_parser, $this->;data))
trigger_error("XML error: ".xml_error_string(xml_get_error_code($this->;xml_parser))." at line ".xml_get_current_line_number($this->;xml_parser),E_USER_ERROR);
}
else
trigger_error("无法初始化RSS对象,{$rss_url}无法访问或者不存在",E_USER_ERROR);
}
//返回所有ITEM信息#######################
function GetItems()
{
return $this->;items;
}
//返回频道信息###########################
function GetChannel()
{
return $this->;channel;
}
//返回RSS文件的版本信息##################
function GetVersion()
{
return $this->;version;
}
//开始解析XML元素########################
function startElement($parser, $name, $attribs)
{
$this->;depth++;
$this->;tag=$name;
switch($name)
{
case "RSS":
$this->;event=$name;
$this->;version=$attribs["VERSION"];
break;
case "CHANNEL":
$this->;event=$name;
$this->;marker=$this->;depth+1;
break;
case "ITEM":
$this->;item_index++;
$this->;event=$name;
$this->;marker=$this->;depth+1;
break;
default:
return NULL;
}
}
//结束某个元素解析时#####################
function endElement($parser, $name)
{
$this->;depth--;
return;
}
//处理数据###############################
function characterData($parser, $data)
{
$data=iconv("utf-8","gb2312",trim($data));
//当数据为chanel下的数据时执行
if( $this->;event=="CHANNEL" && $this->;marker==$this->;depth )
{
$this->;channel[$this->;tag]=$data;
}
//当数据为item下的数据时执行
if( $this->;event=="ITEM" && $this->;marker==$this->;depth )
{
if($this->;prev_tag==$this->;tag)
$this->;items[$this->;item_index][$this->;tag].=$data;
else
$this->;items[$this->;item_index][$this->;tag]=$data;
}
$this->;prev_tag=$this->;tag;
}
//是否是一个有效的RSS地址################
function IsRss($rss_url)
{
$rss_url=trim($rss_url);
if($rss_url=="")
return false;
if($h=@fopen($rss_url,"r"))
{
$text=@fread($h,512);
@fclose($h);
if(eregi("<RSS",$text))
return true;
else
return false;
}
else
return false;
}
}
?>;

复制代码

文库|博客

lovered

白手起家

论坛徽章:: 0

2楼 [报告]

发表于 2005-09-27 15:45 |只看该作者

[RSS解析原创]RSS文档解析类

调用方法
if(RssParse::IsRss($rss_url))
$rss=new RssParse($rss_url);

实战分享：从技术角度谈机器学习入门| 【大话IT】RadonDB低门槛向MySQL集群下战书 | ChinaUnix打赏功能已上线！ | 新一代分布式关系型数据库RadonDB知多少？

kapil

白手起家

论坛徽章:: 0

3楼 [报告]

发表于 2005-12-24 08:39 |只看该作者

请教楼主一个问题，如果有新数据的话，是把整个RSS文件重新生成一下吗？还是只更新某个节点？
如果方便的话能否给出一个简单的小程序为例？呵呵，谢了~~

ydlhero

丰衣足食

论坛徽章:: 0

4楼 [报告]

发表于 2005-12-24 09:32 |只看该作者

请教lz $this->;url =类似这样的为什么要这样写有什么好处?

lovered

白手起家

论坛徽章:: 0

5楼 [报告]

发表于 2005-12-24 14:42 |只看该作者

原帖由 kapil 于 2005-12-24 08:39 发表
请教楼主一个问题，如果有新数据的话，是把整个RSS文件重新生成一下吗？还是只更新某个节点？
如果方便的话能否给出一个简单的小程序为例？呵呵，谢了~~

关于更新的问题我也很苦脑.需要重新读取文件.不生成新文件.目前我解决更新的办法是通过报存到数据库中来判断.效率不高,也在找解决办法.
例子创建实例后 print_r看一下酒知道了.

lovered

白手起家

论坛徽章:: 0

6楼 [报告]

发表于 2005-12-24 14:44 |只看该作者

原帖由 ydlhero 于 2005-12-24 09:32 发表
请教lz $this->;url =类似这样的为什么要这样写有什么好处?

答,粘贴上来就变成这样了,我没有注意.

gaoyikun

家境小康

论坛徽章:: 0

7楼 [报告]

发表于 2005-12-24 15:40 |只看该作者

贡献一个目前自己正在用的RSS Reader类，下载地址http://lastrss.webdot.cz/，感觉还可以，更新上似乎用的是在磁盘上建Cache文件的办法。判断和转换字符编码上好像也更灵活些。

<?php
class lastRSS {
// -------------------------------------------------------------------
// Public properties
// -------------------------------------------------------------------
var $default_cp = 'UTF-8';
var $CDATA = 'nochange';
var $cp = 'gb2312';//默认值为本人所加
var $items_limit = 0;
var $stripHTML = False;
var $date_format = '';
// -------------------------------------------------------------------
// Private variables
// -------------------------------------------------------------------
var $channeltags = array ('title', 'link', 'description', 'language', 'copyright', 'managingEditor', 'webMaster', 'lastBuildDate', 'rating', 'docs');
var $itemtags = array('title', 'link', 'description', 'author', 'category', 'comments', 'enclosure', 'guid', 'pubDate', 'source');
var $imagetags = array('title', 'url', 'link', 'width', 'height');
var $textinputtags = array('title', 'description', 'name', 'link');
// -------------------------------------------------------------------
// Parse RSS file and returns associative array.
// -------------------------------------------------------------------
function Get ($rss_url) {
// If CACHE ENABLED
if ($this->cache_dir != '') {
$cache_file = $this->cache_dir . '/rsscache_' . md5($rss_url);
$timedif = @(time() - filemtime($cache_file));
if ($timedif < $this->cache_time) {
// cached file is fresh enough, return cached array
$result = unserialize(join('', file($cache_file)));
// set 'cached' to 1 only if cached file is correct
if ($result) $result['cached'] = 1;
} else {
// cached file is too old, create new
$result = $this->Parse($rss_url);
$serialized = serialize($result);
if ($f = @fopen($cache_file, 'w')) {
fwrite ($f, $serialized, strlen($serialized));
fclose($f);
}
if ($result) $result['cached'] = 0;
}
}
// If CACHE DISABLED >> load and parse the file directly
else {
$result = $this->Parse($rss_url);
if ($result) $result['cached'] = 0;
}
// return result
return $result;
}
// -------------------------------------------------------------------
// Modification of preg_match(); return trimed field with index 1
// from 'classic' preg_match() array output
// -------------------------------------------------------------------
function my_preg_match ($pattern, $subject) {
// start regullar expression
preg_match($pattern, $subject, $out);
// if there is some result... process it and return it
if(isset($out[1])) {
// Process CDATA (if present)
if ($this->CDATA == 'content') { // Get CDATA content (without CDATA tag)
$out[1] = strtr($out[1], array('<![CDATA['=>'', ']]>'=>''));
} elseif ($this->CDATA == 'strip') { // Strip CDATA
$out[1] = strtr($out[1], array('<![CDATA['=>'', ']]>'=>''));
}
// If code page is set convert character encoding to required
if ($this->cp != '')
//$out[1] = $this->MyConvertEncoding($this->rsscp, $this->cp, $out[1]);
$out[1] = iconv($this->rsscp, $this->cp.'//TRANSLIT', $out[1]);
// Return result
return trim($out[1]);
} else {
// if there is NO result, return empty string
return '';
}
}
// -------------------------------------------------------------------
// Replace HTML entities &something; by real characters
// -------------------------------------------------------------------
function unhtmlentities ($string) {
// Get HTML entities table
$trans_tbl = get_html_translation_table (HTML_ENTITIES, ENT_QUOTES);
// Flip keys<==>values
$trans_tbl = array_flip ($trans_tbl);
// Add support for ' entity (missing in HTML_ENTITIES)
$trans_tbl += array(''' => "'");
// Replace entities by values
return strtr ($string, $trans_tbl);
}
// -------------------------------------------------------------------
// Parse() is private method used by Get() to load and parse RSS file.
// Don't use Parse() in your scripts - use Get($rss_file) instead.
// -------------------------------------------------------------------
function Parse ($rss_url) {
// Open and load RSS file
if ($f = @fopen($rss_url, 'r')) {
$rss_content = '';
while (!feof($f)) {
$rss_content .= fgets($f, 4096);
}
fclose($f);
// Parse document encoding
$result['encoding'] = $this->my_preg_match("'encoding=[\'\"](.*?)[\'\"]'si", $rss_content);
// if document codepage is specified, use it
if ($result['encoding'] != '')
{ $this->rsscp = $result['encoding']; } // This is used in my_preg_match()
// otherwise use the default codepage
else
{ $this->rsscp = $this->default_cp; } // This is used in my_preg_match()
// Parse CHANNEL info
preg_match("'<channel.*?>(.*?)</channel>'si", $rss_content, $out_channel);
foreach($this->channeltags as $channeltag)
{
$temp = $this->my_preg_match("'<$channeltag.*?>(.*?)</$channeltag>'si", $out_channel[1]);
if ($temp != '') $result[$channeltag] = $temp; // Set only if not empty
}
// If date_format is specified and lastBuildDate is valid
if ($this->date_format != '' && ($timestamp = strtotime($result['lastBuildDate'])) !==-1) {
// convert lastBuildDate to specified date format
$result['lastBuildDate'] = date($this->date_format, $timestamp);
}
// Parse TEXTINPUT info
preg_match("'<textinput(|[^>]*[^/])>(.*?)</textinput>'si", $rss_content, $out_textinfo);
// This a little strange regexp means:
// Look for tag <textinput> with or without any attributes, but skip truncated version <textinput /> (it's not beggining tag)
if (isset($out_textinfo[2])) {
foreach($this->textinputtags as $textinputtag) {
$temp = $this->my_preg_match("'<$textinputtag.*?>(.*?)</$textinputtag>'si", $out_textinfo[2]);
if ($temp != '') $result['textinput_'.$textinputtag] = $temp; // Set only if not empty
}
}
// Parse IMAGE info
preg_match("'<image.*?>(.*?)</image>'si", $rss_content, $out_imageinfo);
if (isset($out_imageinfo[1])) {
foreach($this->imagetags as $imagetag) {
$temp = $this->my_preg_match("'<$imagetag.*?>(.*?)</$imagetag>'si", $out_imageinfo[1]);
if ($temp != '') $result['image_'.$imagetag] = $temp; // Set only if not empty
}
}
// Parse ITEMS
preg_match_all("'<item(| .*?)>(.*?)</item>'si", $rss_content, $items);
$rss_items = $items[2];
$i = 0;
$result['items'] = array(); // create array even if there are no items
foreach($rss_items as $rss_item) {
// If number of items is lower then limit: Parse one item
if ($i < $this->items_limit || $this->items_limit == 0) {
foreach($this->itemtags as $itemtag) {
$temp = $this->my_preg_match("'<$itemtag.*?>(.*?)</$itemtag>'si", $rss_item);
if ($temp != '') $result['items'][$i][$itemtag] = $temp; // Set only if not empty
}
// Strip HTML tags and other bullshit from DESCRIPTION
if ($this->stripHTML && $result['items'][$i]['description'])
$result['items'][$i]['description'] = strip_tags($this->unhtmlentities(strip_tags($result['items'][$i]['description'])));
// Strip HTML tags and other bullshit from TITLE
if ($this->stripHTML && $result['items'][$i]['title'])
$result['items'][$i]['title'] = strip_tags($this->unhtmlentities(strip_tags($result['items'][$i]['title'])));
// If date_format is specified and pubDate is valid
if ($this->date_format != '' && ($timestamp = strtotime($result['items'][$i]['pubDate'])) !==-1) {
// convert pubDate to specified date format
$result['items'][$i]['pubDate'] = date($this->date_format, $timestamp);
}
// Item counter
$i++;
}
}
$result['items_count'] = $i;
return $result;
}
else // Error in opening return False
{
return False;
}
}
}
?>

复制代码

[ 本帖最后由 gaoyikun 于 2005-12-24 15:42 编辑 ]

返回列表

Chinaunix › 论坛 › 程序设计 › PHP › [RSS解析原创]RSS文档解析类

[RSS解析 原创]RSS文档解析类 [复制链接]

[RSS解析 原创]RSS文档解析类

浏览过的版块

[RSS解析原创]RSS文档解析类 [复制链接]

[RSS解析原创]RSS文档解析类