📄 regexphtmllinkextractor.html
字号:
<!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=UTF-8" /><title>RegexpHTMLLinkExtractor xref</title><link type="text/css" rel="stylesheet" href="../../../stylesheet.css" /></head><body><div id="overview"><a href="../../../../apidocs/org/archive/extractor/RegexpHTMLLinkExtractor.html">View Javadoc</a></div><pre><a name="1" href="#1">1</a> <em class="comment">/*<em class="comment"> Copyright (C) 2003 Internet Archive.</em></em><a name="2" href="#2">2</a> <em class="comment"> *</em><a name="3" href="#3">3</a> <em class="comment"> * This file is part of the Heritrix web crawler (crawler.archive.org).</em><a name="4" href="#4">4</a> <em class="comment"> *</em><a name="5" href="#5">5</a> <em class="comment"> * Heritrix is free software; you can redistribute it and/or modify</em><a name="6" href="#6">6</a> <em class="comment"> * it under the terms of the GNU Lesser Public License as published by</em><a name="7" href="#7">7</a> <em class="comment"> * the Free Software Foundation; either version 2.1 of the License, or</em><a name="8" href="#8">8</a> <em class="comment"> * any later version.</em><a name="9" href="#9">9</a> <em class="comment"> *</em><a name="10" href="#10">10</a> <em class="comment"> * Heritrix is distributed in the hope that it will be useful,</em><a name="11" href="#11">11</a> <em class="comment"> * but WITHOUT ANY WARRANTY; without even the implied warranty of</em><a name="12" href="#12">12</a> <em class="comment"> * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the</em><a name="13" href="#13">13</a> <em class="comment"> * GNU Lesser Public License for more details.</em><a name="14" href="#14">14</a> <em class="comment"> *</em><a name="15" href="#15">15</a> <em class="comment"> * You should have received a copy of the GNU Lesser Public License</em><a name="16" href="#16">16</a> <em class="comment"> * along with Heritrix; if not, write to the Free Software</em><a name="17" href="#17">17</a> <em class="comment"> * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA</em><a name="18" href="#18">18</a> <em class="comment"> *</em><a name="19" href="#19">19</a> <em class="comment"> * SimpleHTMLExtractor.java</em><a name="20" href="#20">20</a> <em class="comment"> * Created on Jun 5, 2003</em><a name="21" href="#21">21</a> <em class="comment"> *</em><a name="22" href="#22">22</a> <em class="comment"> * $Header: /cvsroot/archive-crawler/ArchiveOpenCrawler/src/java/org/archive/extractor/RegexpHTMLLinkExtractor.java,v 1.7 2005/11/11 22:29:53 gojomo Exp $</em><a name="23" href="#23">23</a> <em class="comment"> */</em><a name="24" href="#24">24</a> <strong>package</strong> <a href="../../../org/archive/extractor/package-summary.html">org.archive.extractor</a>;<a name="25" href="#25">25</a> <a name="26" href="#26">26</a> <strong>import</strong> java.util.ArrayList;<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.logging.Level;<a name="30" href="#30">30</a> <strong>import</strong> java.util.logging.Logger;<a name="31" href="#31">31</a> <strong>import</strong> java.util.regex.Matcher;<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.archive.crawler.extractor.Link;<a name="35" href="#35">35</a> <strong>import</strong> org.archive.net.UURI;<a name="36" href="#36">36</a> <strong>import</strong> org.archive.net.UURIFactory;<a name="37" href="#37">37</a> <strong>import</strong> org.archive.util.DevUtils;<a name="38" href="#38">38</a> <strong>import</strong> org.archive.util.TextUtils;<a name="39" href="#39">39</a> <a name="40" href="#40">40</a> <a name="41" href="#41">41</a> <em>/**<em>*</em></em><a name="42" href="#42">42</a> <em> * Basic link-extraction, from an HTML content-body,</em><a name="43" href="#43">43</a> <em> * using regular expressions.</em><a name="44" href="#44">44</a> <em> *</em><a name="45" href="#45">45</a> <em> * ROUGH DRAFT IN PROGRESS / incomplete... untested...</em><a name="46" href="#46">46</a> <em> *</em><a name="47" href="#47">47</a> <em> * @author gojomo</em><a name="48" href="#48">48</a> <em> */</em><a name="49" href="#49">49</a> <strong>public</strong> <strong>class</strong> <a href="../../../org/archive/extractor/RegexpHTMLLinkExtractor.html">RegexpHTMLLinkExtractor</a> <strong>extends</strong> <a href="../../../org/archive/extractor/CharSequenceLinkExtractor.html">CharSequenceLinkExtractor</a> {<a name="50" href="#50">50</a> <strong>private</strong> <strong>static</strong> Logger logger =<a name="51" href="#51">51</a> Logger.getLogger(RegexpHTMLLinkExtractor.<strong>class</strong>.getName());<a name="52" href="#52">52</a> <a name="53" href="#53">53</a> <strong>boolean</strong> honorRobots = <strong>true</strong>;<a name="54" href="#54">54</a> <strong>boolean</strong> extractInlineCss = <strong>true</strong>;<a name="55" href="#55">55</a> <strong>boolean</strong> extractInlineJs = <strong>true</strong>;<a name="56" href="#56">56</a> <a name="57" href="#57">57</a> <strong>protected</strong> LinkedList next = <strong>new</strong> LinkedList();<a name="58" href="#58">58</a> <strong>protected</strong> Matcher tags;<a name="59" href="#59">59</a> <a name="60" href="#60">60</a> <em class="comment">/*<em class="comment"> (non-Javadoc)</em></em><a name="61" href="#61">61</a> <em class="comment"> * @see org.archive.extractor.CharSequenceLinkExtractor#findNextLink()</em><a name="62" href="#62">62</a> <em class="comment"> */</em><a name="63" href="#63">63</a> <strong>protected</strong> <strong>boolean</strong> findNextLink() {<a name="64" href="#64">64</a> <strong>if</strong> (tags == <strong>null</strong>) {<a name="65" href="#65">65</a> tags = TextUtils.getMatcher(RELEVANT_TAG_EXTRACTOR, sourceContent);<a name="66" href="#66">66</a> }<a name="67" href="#67">67</a> <strong>while</strong>(tags.find()) {<a name="68" href="#68">68</a> <strong>if</strong>(Thread.interrupted()){<a name="69" href="#69">69</a> <em class="comment">// TODO: throw an exception, perhaps, rather than just clear & break?</em><a name="70" href="#70">70</a> <strong>break</strong>;<a name="71" href="#71">71</a> }<a name="72" href="#72">72</a> <strong>if</strong> (tags.start(8) > 0) {<a name="73" href="#73">73</a> <em class="comment">// comment match</em><a name="74" href="#74">74</a> <em class="comment">// for now do nothing</em><a name="75" href="#75">75</a> } <strong>else</strong> <strong>if</strong> (tags.start(7) > 0) {<a name="76" href="#76">76</a> <em class="comment">// <meta> match</em><a name="77" href="#77">77</a> <strong>int</strong> start = tags.start(5);<a name="78" href="#78">78</a> <strong>int</strong> end = tags.end(5);<a name="79" href="#79">79</a> processMeta(sourceContent.subSequence(start, end));<a name="80" href="#80">80</a> } <strong>else</strong> <strong>if</strong> (tags.start(5) > 0) {<a name="81" href="#81">81</a> <em class="comment">// generic <whatever> match</em><a name="82" href="#82">82</a> <strong>int</strong> start5 = tags.start(5);<a name="83" href="#83">83</a> <strong>int</strong> end5 = tags.end(5);<a name="84" href="#84">84</a> <strong>int</strong> start6 = tags.start(6);<a name="85" href="#85">85</a> <strong>int</strong> end6 = tags.end(6);<a name="86" href="#86">86</a> processGeneralTag(sourceContent.subSequence(start6, end6),<a name="87" href="#87">87</a> sourceContent.subSequence(start5, end5));<a name="88" href="#88">88</a> } <strong>else</strong> <strong>if</strong> (tags.start(1) > 0) {<a name="89" href="#89">89</a> <em class="comment">// <script> match</em><a name="90" href="#90">90</a> <strong>int</strong> start = tags.start(1);<a name="91" href="#91">91</a> <strong>int</strong> end = tags.end(1);<a name="92" href="#92">92</a> processScript(sourceContent.subSequence(start, end),<a name="93" href="#93">93</a> tags.end(2) - start);<a name="94" href="#94">94</a> } <strong>else</strong> <strong>if</strong> (tags.start(3) > 0){<a name="95" href="#95">95</a> <em class="comment">// <style... match</em><a name="96" href="#96">96</a> <strong>int</strong> start = tags.start(3);<a name="97" href="#97">97</a> <strong>int</strong> end = tags.end(3);<a name="98" href="#98">98</a> processStyle(sourceContent.subSequence(start, end),<a name="99" href="#99">99</a> tags.end(4) - start);<a name="100" href="#100">100</a> }<a name="101" href="#101">101</a> <strong>if</strong>(!next.isEmpty()) {<a name="102" href="#102">102</a> <em class="comment">// at least one link found</em><a name="103" href="#103">103</a> <strong>return</strong> <strong>true</strong>;<a name="104" href="#104">104</a> }<a name="105" href="#105">105</a> }<a name="106" href="#106">106</a> <em class="comment">// no relevant tags found</em><a name="107" href="#107">107</a> <strong>return</strong> false;<a name="108" href="#108">108</a> }<a name="109" href="#109">109</a> <a name="110" href="#110">110</a> <em>/**<em>*</em></em><a name="111" href="#111">111</a> <em> * Compiled relevant tag extractor.</em><a name="112" href="#112">112</a> <em> *</em><a name="113" href="#113">113</a> <em> * <p></em><a name="114" href="#114">114</a> <em> * This pattern extracts either:</em><a name="115" href="#115">115</a> <em> * <li> (1) whole &lt;script&gt;...&lt;/script&gt; or</em><a name="116" href="#116">116</a> <em> * <li> (2) &lt;style&gt;...&lt;/style&gt; or</em><a name="117" href="#117">117</a> <em> * <li> (3) &lt;meta ...&gt; or</em><a name="118" href="#118">118</a> <em> * <li> (4) any other open-tag with at least one attribute</em><a name="119" href="#119">119</a> <em> * (eg matches "&lt;a href='boo'&gt;" but not "&lt;/a&gt;" or "&lt;br&gt;")</em><a name="120" href="#120">120</a> <em> * <p></em><a name="121" href="#121">121</a> <em> * groups:</em><a name="122" href="#122">122</a> <em> * <li> 1: SCRIPT SRC=foo&gt;boo&lt;/SCRIPT</em><a name="123" href="#123">123</a> <em> * <li> 2: just script open tag</em><a name="124" href="#124">124</a> <em> * <li> 3: STYLE TYPE=moo&gt;zoo&lt;/STYLE</em><a name="125" href="#125">125</a> <em> * <li> 4: just style open tag</em><a name="126" href="#126">126</a> <em> * <li> 5: entire other tag, without '<' '>'</em><a name="127" href="#127">127</a> <em> * <li> 6: element</em><a name="128" href="#128">128</a> <em> * <li> 7: META</em><a name="129" href="#129">129</a> <em> * <li> 8: !-- comment --</em><a name="130" href="#130">130</a> <em> */</em><a name="131" href="#131">131</a> <strong>static</strong> <strong>final</strong> String RELEVANT_TAG_EXTRACTOR =<a name="132" href="#132">132</a> <span class="string">"(?is)<(?:((script[^>]*+)>.*?</script)|((style[^>]*+)>[^<]*+</style)|(((meta)|(?://w+))//s+[^>]*+)|(!--.*?--))>"</span>;<a name="133" href="#133">133</a> <a name="134" href="#134">134</a> <em class="comment">// this pattern extracts attributes from any open-tag innards</em><a name="135" href="#135">135</a> <em class="comment">// matched by the above. attributes known to be URIs of various</em><a name="136" href="#136">136</a> <em class="comment">// sorts are matched specially</em><a name="137" href="#137">137</a> <strong>static</strong> <strong>final</strong> String EACH_ATTRIBUTE_EXTRACTOR =<a name="138" href="#138">138</a> <span class="string">"(?is)//s((href)|(action)|(on//w*)"</span><a name="139" href="#139">139</a> +<span class="string">"|((?:src)|(?:lowsrc)|(?:background)|(?:cite)|(?:longdesc)"</span><a name="140" href="#140">140</a> +<span class="string">"|(?:usemap)|(?:profile)|(?:datasrc)|(?:for))"</span><a name="141" href="#141">141</a> +<span class="string">"|(codebase)|((?:classid)|(?:data))|(archive)|(code)"</span><a name="142" href="#142">142</a> +<span class="string">"|(value)|([-//w]+))"</span><a name="143" href="#143">143</a> +<span class="string">"//s*=//s*"</span><a name="144" href="#144">144</a> +<span class="string">"(?:(?:\"(.*?)(?:\"|$))"</span><a name="145" href="#145">145</a> +<span class="string">"|(?:'(.*?)(?:'|$))"</span><a name="146" href="#146">146</a> +<span class="string">"|(//S+))"</span>;<a name="147" href="#147">147</a> <em class="comment">// groups:</em><a name="148" href="#148">148</a> <em class="comment">// 1: attribute name</em><a name="149" href="#149">149</a> <em class="comment">// 2: HREF - single URI relative to doc base, or occasionally javascript:</em>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -