jerichoextractorhtml.html
来自「网络爬虫开源代码」· HTML 代码 · 共 477 行 · 第 1/3 页
HTML
477 行
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"><head><meta http-equiv="content-type" content="text/html; charset=ISO-8859-1" /><title>JerichoExtractorHTML xref</title><link type="text/css" rel="stylesheet" href="../../../../stylesheet.css" /></head><body><pre><a name="1" href="#1">1</a> <em class="comment">/*<em class="comment"> JerichoExtractorHTML</em></em><a name="2" href="#2">2</a> <em class="comment"> * </em><a name="3" href="#3">3</a> <em class="comment"> * Copyright (C) 2006 Olaf Freyer</em><a name="4" href="#4">4</a> <em class="comment"> *</em><a name="5" href="#5">5</a> <em class="comment"> * This file is part of the Heritrix web crawler (crawler.archive.org).</em><a name="6" href="#6">6</a> <em class="comment"> *</em><a name="7" href="#7">7</a> <em class="comment"> * Heritrix is free software; you can redistribute it and/or modify</em><a name="8" href="#8">8</a> <em class="comment"> * it under the terms of the GNU Lesser Public License as published by</em><a name="9" href="#9">9</a> <em class="comment"> * the Free Software Foundation; either version 2.1 of the License, or</em><a name="10" href="#10">10</a> <em class="comment"> * any later version.</em><a name="11" href="#11">11</a> <em class="comment"> *</em><a name="12" href="#12">12</a> <em class="comment"> * Heritrix is distributed in the hope that it will be useful,</em><a name="13" href="#13">13</a> <em class="comment"> * but WITHOUT ANY WARRANTY; without even the implied warranty of</em><a name="14" href="#14">14</a> <em class="comment"> * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the</em><a name="15" href="#15">15</a> <em class="comment"> * GNU Lesser Public License for more details.</em><a name="16" href="#16">16</a> <em class="comment"> *</em><a name="17" href="#17">17</a> <em class="comment"> * You should have received a copy of the GNU Lesser Public License</em><a name="18" href="#18">18</a> <em class="comment"> * along with Heritrix; if not, write to the Free Software</em><a name="19" href="#19">19</a> <em class="comment"> * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA</em><a name="20" href="#20">20</a> <em class="comment"> *</em><a name="21" href="#21">21</a> <em class="comment"> * $Id: JerichoExtractorHTML.java 4726 2006-11-15 17:57:11Z stack-sf $</em><a name="22" href="#22">22</a> <em class="comment"> */</em><a name="23" href="#23">23</a> <strong>package</strong> <a href="../../../../org/archive/crawler/extractor/package-summary.html">org.archive.crawler.extractor</a>;<a name="24" href="#24">24</a> <a name="25" href="#25">25</a> <strong>import</strong> java.util.ArrayList;<a name="26" href="#26">26</a> <strong>import</strong> java.util.Collection;<a name="27" href="#27">27</a> <strong>import</strong> java.util.Iterator;<a name="28" href="#28">28</a> <strong>import</strong> java.util.LinkedList;<a name="29" href="#29">29</a> <strong>import</strong> java.util.List;<a name="30" href="#30">30</a> <strong>import</strong> java.util.logging.Level;<a name="31" href="#31">31</a> <strong>import</strong> java.util.logging.Logger;<a name="32" href="#32">32</a> <a name="33" href="#33">33</a> <strong>import</strong> org.apache.commons.httpclient.URIException;<a name="34" href="#34">34</a> <strong>import</strong> org.apache.commons.lang.StringEscapeUtils;<a name="35" href="#35">35</a> <strong>import</strong> org.archive.crawler.datamodel.CoreAttributeConstants;<a name="36" href="#36">36</a> <strong>import</strong> org.archive.crawler.datamodel.CrawlURI;<a name="37" href="#37">37</a> <strong>import</strong> org.archive.crawler.datamodel.RobotsHonoringPolicy;<a name="38" href="#38">38</a> <strong>import</strong> org.archive.net.UURI;<a name="39" href="#39">39</a> <strong>import</strong> org.archive.net.UURIFactory;<a name="40" href="#40">40</a> <strong>import</strong> org.archive.util.DevUtils;<a name="41" href="#41">41</a> <strong>import</strong> org.archive.util.TextUtils;<a name="42" href="#42">42</a> <a name="43" href="#43">43</a> <strong>import</strong> au.id.jericho.lib.html.Attribute;<a name="44" href="#44">44</a> <strong>import</strong> au.id.jericho.lib.html.Attributes;<a name="45" href="#45">45</a> <strong>import</strong> au.id.jericho.lib.html.Element;<a name="46" href="#46">46</a> <strong>import</strong> au.id.jericho.lib.html.FormControl;<a name="47" href="#47">47</a> <strong>import</strong> au.id.jericho.lib.html.FormControlType;<a name="48" href="#48">48</a> <strong>import</strong> au.id.jericho.lib.html.FormField;<a name="49" href="#49">49</a> <strong>import</strong> au.id.jericho.lib.html.FormFields;<a name="50" href="#50">50</a> <strong>import</strong> au.id.jericho.lib.html.HTMLElementName;<a name="51" href="#51">51</a> <strong>import</strong> au.id.jericho.lib.html.Source;<a name="52" href="#52">52</a> <strong>import</strong> au.id.jericho.lib.html.StartTagType;<a name="53" href="#53">53</a> <a name="54" href="#54">54</a> <em>/**<em>*</em></em><a name="55" href="#55">55</a> <em> * Improved link-extraction from an HTML content-body using jericho-html parser.</em><a name="56" href="#56">56</a> <em> * This extractor extends ExtractorHTML and mimics its workflow - but has some</em><a name="57" href="#57">57</a> <em> * substantial differences when it comes to internal implementation. Instead</em><a name="58" href="#58">58</a> <em> * of heavily relying upon java regular expressions it uses a real html parser</em><a name="59" href="#59">59</a> <em> * library - namely Jericho HTML Parser (<a href="http://jerichohtml.sourceforge.net)." target="alexandria_uri">http://jerichohtml.sourceforge.net).</a></em><a name="60" href="#60">60</a> <em> * Using this parser it can better handle broken html (i.e. missing quotes)</em><a name="61" href="#61">61</a> <em> * and also offer improved extraction of HTML form URLs (not only extract</em><a name="62" href="#62">62</a> <em> * the action of a form, but also its default values).</em><a name="63" href="#63">63</a> <em> * Unfortunately this parser also has one major drawback - it has to read the</em><a name="64" href="#64">64</a> <em> * whole document into memory for parsing, thus has an inherent OOME risk.</em><a name="65" href="#65">65</a> <em> * This OOME risk can be reduced/eleminated by limiting the size of documents</em><a name="66" href="#66">66</a> <em> * to be parsed (i.e. using NotExceedsDocumentLengthTresholdDecideRule).</em><a name="67" href="#67">67</a> <em> * Also note that this extractor seems to have a lower overall memory </em><a name="68" href="#68">68</a> <em> * consumption compared to ExtractorHTML. (still to be confirmed on a larger </em><a name="69" href="#69">69</a> <em> * scale crawl) </em><a name="70" href="#70">70</a> <em> * </em><a name="71" href="#71">71</a> <em> * @author Olaf Freyer</em><a name="72" href="#72">72</a> <em> * @version $Date: 2006-11-15 17:57:11 +0000 (Wed, 15 Nov 2006) $ $Revision: 4726 $</em><a name="73" href="#73">73</a> <em> */</em><a name="74" href="#74">74</a> <strong>public</strong> <strong>class</strong> <a href="../../../../org/archive/crawler/extractor/JerichoExtractorHTML.html">JerichoExtractorHTML</a> <strong>extends</strong> <a href="../../../../org/archive/crawler/extractor/ExtractorHTML.html">ExtractorHTML</a> implements<a name="75" href="#75">75</a> <a href="../../../../org/archive/crawler/datamodel/CoreAttributeConstants.html">CoreAttributeConstants</a> {<a name="76" href="#76">76</a> <a name="77" href="#77">77</a> <strong>private</strong> <strong>static</strong> <strong>final</strong> <strong>long</strong> serialVersionUID = 1684681316546343615L;<a name="78" href="#78">78</a> <a name="79" href="#79">79</a> <strong>private</strong> Logger logger = Logger.getLogger(<strong>this</strong>.getClass().getName());<a name="80" href="#80">80</a> <a name="81" href="#81">81</a> <strong>protected</strong> <strong>long</strong> numberOfFormsProcessed = 0;<a name="82" href="#82">82</a> <a name="83" href="#83">83</a> <strong>public</strong> <a href="../../../../org/archive/crawler/extractor/JerichoExtractorHTML.html">JerichoExtractorHTML</a>(String name) {<a name="84" href="#84">84</a> <strong>this</strong>(name, <span class="string">"Jericho-HTML extractor. Extracts links from HTML "</span> +<a name="85" href="#85">85</a> <span class="string">"documents using Jericho HTML Parser. Offers same "</span> + <a name="86" href="#86">86</a> <span class="string">"basic functionality as ExtractorHTML but better "</span> +<a name="87" href="#87">87</a> <span class="string">"handles broken HTML and extraction of default "</span> +<a name="88" href="#88">88</a> <span class="string">"values from HTML forms. A word of warning: the used "</span> +<a name="89" href="#89">89</a> <span class="string">"parser, the Jericho HTML Parser, reads the whole "</span> +<a name="90" href="#90">90</a> <span class="string">"document into memory for "</span> +<a name="91" href="#91">91</a> <span class="string">"parsing - thus this extractor has an inherent OOME risk. "</span> +<a name="92" href="#92">92</a> <span class="string">"This OOME risk can be reduced/eleminated by limiting the "</span> +<a name="93" href="#93">93</a> <span class="string">"size of documents to be parsed (i.e. using "</span> +<a name="94" href="#94">94</a> <span class="string">"NotExceedsDocumentLengthTresholdDecideRule). "</span>);<a name="95" href="#95">95</a> }<a name="96" href="#96">96</a> <a name="97" href="#97">97</a> <strong>public</strong> <a href="../../../../org/archive/crawler/extractor/JerichoExtractorHTML.html">JerichoExtractorHTML</a>(String name, String description) {<a name="98" href="#98">98</a> <strong>super</strong>(name, description);<a name="99" href="#99">99</a> }<a name="100" href="#100">100</a> <a name="101" href="#101">101</a> <strong>private</strong> <strong>static</strong> List<Attribute> findOnAttributes(Attributes attributes) {<a name="102" href="#102">102</a> List<Attribute> result = <strong>new</strong> LinkedList<Attribute>();<a name="103" href="#103">103</a> <strong>for</strong> (Iterator attrIter = attributes.iterator(); attrIter.hasNext();) {<a name="104" href="#104">104</a> Attribute attr = (Attribute) attrIter.next();<a name="105" href="#105">105</a> <strong>if</strong> (attr.getKey().startsWith(<span class="string">"on"</span>))<a name="106" href="#106">106</a> result.add(attr);<a name="107" href="#107">107</a> }<a name="108" href="#108">108</a> <strong>return</strong> result;<a name="109" href="#109">109</a> }<a name="110" href="#110">110</a> <a name="111" href="#111">111</a> <strong>protected</strong> <strong>void</strong> processGeneralTag(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi, Element element,<a name="112" href="#112">112</a> Attributes attributes) {<a name="113" href="#113">113</a> Attribute attr;<a name="114" href="#114">114</a> String attrValue;<a name="115" href="#115">115</a> List attrList;<a name="116" href="#116">116</a> String elementName = element.getName();<a name="117" href="#117">117</a> <a name="118" href="#118">118</a> <em class="comment">// Just in case it's an OBJECT or APPLET tag</em><a name="119" href="#119">119</a> String codebase = <strong>null</strong>;<a name="120" href="#120">120</a> ArrayList<String> resources = <strong>null</strong>;<a name="121" href="#121">121</a> <a name="122" href="#122">122</a> <strong>final</strong> <strong>boolean</strong> framesAsEmbeds = ((Boolean) getUncheckedAttribute(curi,<a name="123" href="#123">123</a> ATTR_TREAT_FRAMES_AS_EMBED_LINKS)).booleanValue();<a name="124" href="#124">124</a> <a name="125" href="#125">125</a> <strong>final</strong> <strong>boolean</strong> ignoreFormActions = ((Boolean) getUncheckedAttribute(<a name="126" href="#126">126</a> curi, ATTR_IGNORE_FORM_ACTION_URLS)).booleanValue();<a name="127" href="#127">127</a> <a name="128" href="#128">128</a> <strong>final</strong> <strong>boolean</strong> overlyEagerLinkDetection =<a name="129" href="#129">129</a> ((Boolean)getUncheckedAttribute(<a name="130" href="#130">130</a> curi, ATTR_OVERLY_EAGER_LINK_DETECTION)).booleanValue();<a name="131" href="#131">131</a> <a name="132" href="#132">132</a> <em class="comment">// HREF</em><a name="133" href="#133">133</a> <strong>if</strong> (((attr = attributes.get(<span class="string">"href"</span>)) != <strong>null</strong>) &&<a name="134" href="#134">134</a> ((attrValue = attr.getValue()) != <strong>null</strong>)) {<a name="135" href="#135">135</a> CharSequence context = Link.elementContext(elementName, attr<a name="136" href="#136">136</a> .getKey());<a name="137" href="#137">137</a> <strong>if</strong> (<span class="string">"link"</span>.equals(elementName)) {<a name="138" href="#138">138</a> <em class="comment">// <LINK> elements treated as embeds (css, ico, etc)</em><a name="139" href="#139">139</a> processEmbed(curi, attrValue, context);<a name="140" href="#140">140</a> } <strong>else</strong> {<a name="141" href="#141">141</a> <em class="comment">// other HREFs treated as links</em><a name="142" href="#142">142</a> processLink(curi, attrValue, context);<a name="143" href="#143">143</a> }<a name="144" href="#144">144</a> <strong>if</strong> (<span class="string">"base"</span>.equals(elementName)) {<a name="145" href="#145">145</a> <strong>try</strong> {<a name="146" href="#146">146</a> curi.setBaseURI(attrValue);<a name="147" href="#147">147</a> } <strong>catch</strong> (URIException e) {<a name="148" href="#148">148</a> <strong>if</strong> (getController() != <strong>null</strong>) {<a name="149" href="#149">149</a> <em class="comment">// Controller can be null: e.g. when running</em><a name="150" href="#150">150</a> <em class="comment">// ExtractorTool.</em>⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?