📄 extractordoc.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>ExtractorDOC xref</title><link type="text/css" rel="stylesheet" href="../../../../stylesheet.css" /></head><body><div id="overview"><a href="../../../../../apidocs/org/archive/crawler/extractor/ExtractorDOC.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"> * Created on Jul 7, 2003</em><a name="20" href="#20">20</a> <em class="comment"> *</em><a name="21" href="#21">21</a> <em class="comment"> */</em><a name="22" href="#22">22</a> <strong>package</strong> <a href="../../../../org/archive/crawler/extractor/package-summary.html">org.archive.crawler.extractor</a>;<a name="23" href="#23">23</a> <a name="24" href="#24">24</a> <strong>import</strong> java.io.IOException;<a name="25" href="#25">25</a> <strong>import</strong> java.io.InputStream;<a name="26" href="#26">26</a> <strong>import</strong> java.io.StringWriter;<a name="27" href="#27">27</a> <strong>import</strong> java.io.Writer;<a name="28" href="#28">28</a> <strong>import</strong> java.util.ArrayList;<a name="29" href="#29">29</a> <strong>import</strong> java.util.logging.Logger;<a name="30" href="#30">30</a> <a name="31" href="#31">31</a> <strong>import</strong> org.apache.commons.httpclient.URIException;<a name="32" href="#32">32</a> <strong>import</strong> org.apache.poi.hdf.extractor.WordDocument;<a name="33" href="#33">33</a> <strong>import</strong> org.archive.crawler.datamodel.CoreAttributeConstants;<a name="34" href="#34">34</a> <strong>import</strong> org.archive.crawler.datamodel.CrawlURI;<a name="35" href="#35">35</a> <a name="36" href="#36">36</a> <em>/**<em>*</em></em><a name="37" href="#37">37</a> <em> * This class allows the caller to extract href style links from word97-format word documents.</em><a name="38" href="#38">38</a> <em> *</em><a name="39" href="#39">39</a> <em> * @author Parker Thompson</em><a name="40" href="#40">40</a> <em> *</em><a name="41" href="#41">41</a> <em> */</em><a name="42" href="#42">42</a> <strong>public</strong> <strong>class</strong> <a href="../../../../org/archive/crawler/extractor/ExtractorDOC.html">ExtractorDOC</a> <strong>extends</strong> <a href="../../../../org/archive/crawler/extractor/Extractor.html">Extractor</a> implements <a href="../../../../org/archive/crawler/datamodel/CoreAttributeConstants.html">CoreAttributeConstants</a> {<a name="43" href="#43">43</a> <a name="44" href="#44">44</a> <strong>private</strong> <strong>static</strong> Logger logger =<a name="45" href="#45">45</a> Logger.getLogger(<span class="string">"org.archive.crawler.extractor.ExtractorDOC"</span>);<a name="46" href="#46">46</a> <strong>private</strong> <strong>long</strong> numberOfCURIsHandled = 0;<a name="47" href="#47">47</a> <strong>private</strong> <strong>long</strong> numberOfLinksExtracted = 0;<a name="48" href="#48">48</a> <a name="49" href="#49">49</a> <em>/**<em>*</em></em><a name="50" href="#50">50</a> <em> * @param name</em><a name="51" href="#51">51</a> <em> */</em><a name="52" href="#52">52</a> <strong>public</strong> <a href="../../../../org/archive/crawler/extractor/ExtractorDOC.html">ExtractorDOC</a>(String name) {<a name="53" href="#53">53</a> <strong>super</strong>(name, <span class="string">"MS-Word document Extractor. Extracts links from MS-Word"</span> +<a name="54" href="#54">54</a> <span class="string">" '.doc' documents."</span>);<a name="55" href="#55">55</a> }<a name="56" href="#56">56</a> <a name="57" href="#57">57</a> <em>/**<em>*</em></em><a name="58" href="#58">58</a> <em> * Processes a word document and extracts any hyperlinks from it.</em><a name="59" href="#59">59</a> <em> * This only extracts href style links, and does not examine the actual</em><a name="60" href="#60">60</a> <em> * text for valid URIs.</em><a name="61" href="#61">61</a> <em> * @param curi CrawlURI to process.</em><a name="62" href="#62">62</a> <em> */</em><a name="63" href="#63">63</a> <strong>protected</strong> <strong>void</strong> extract(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi){<a name="64" href="#64">64</a> <em class="comment">// Assumes docs will be coming in through http.</em><a name="65" href="#65">65</a> <em class="comment">// TODO make this more general (currently we're only fetching via http</em><a name="66" href="#66">66</a> <em class="comment">// so it doesn't matter)</em><a name="67" href="#67">67</a> <strong>if</strong> (!isHttpTransactionContentToProcess(curi) ||<a name="68" href="#68">68</a> !isExpectedMimeType(curi.getContentType(),<a name="69" href="#69">69</a> <span class="string">"application/msword"</span>)) {<a name="70" href="#70">70</a> <strong>return</strong>;<a name="71" href="#71">71</a> }<a name="72" href="#72">72</a> <a name="73" href="#73">73</a> ArrayList links = <strong>new</strong> ArrayList();<a name="74" href="#74">74</a> InputStream documentStream = <strong>null</strong>;<a name="75" href="#75">75</a> Writer out = <strong>null</strong>;<a name="76" href="#76">76</a> <a name="77" href="#77">77</a> numberOfCURIsHandled++;<a name="78" href="#78">78</a> <a name="79" href="#79">79</a> <em class="comment">// Get the doc as a File</em><a name="80" href="#80">80</a> <strong>try</strong><a name="81" href="#81">81</a> {<a name="82" href="#82">82</a> documentStream = curi.getHttpRecorder().getRecordedInput().<a name="83" href="#83">83</a> getContentReplayInputStream();<a name="84" href="#84">84</a> <a name="85" href="#85">85</a> <strong>if</strong> (documentStream==<strong>null</strong>) {<a name="86" href="#86">86</a> <em class="comment">// TODO: note problem</em><a name="87" href="#87">87</a> <strong>return</strong>;<a name="88" href="#88">88</a> }<a name="89" href="#89">89</a> <em class="comment">// extract the text from the doc and write it to a stream we</em><a name="90" href="#90">90</a> <em class="comment">// can then process</em><a name="91" href="#91">91</a> out = <strong>new</strong> StringWriter();<a name="92" href="#92">92</a> WordDocument w = <strong>null</strong>;<a name="93" href="#93">93</a> <a name="94" href="#94">94</a> w = <strong>new</strong> WordDocument( documentStream );<a name="95" href="#95">95</a> w.writeAllText(out);<a name="96" href="#96">96</a> }<strong>catch</strong>(Exception e){<a name="97" href="#97">97</a> curi.addLocalizedError(getName(),e,<span class="string">"ExtractorDOC Exception"</span>);<a name="98" href="#98">98</a> <strong>return</strong>;<a name="99" href="#99">99</a> } <strong>finally</strong> {<a name="100" href="#100">100</a> <strong>try</strong> {<a name="101" href="#101">101</a> documentStream.close();<a name="102" href="#102">102</a> } <strong>catch</strong> (IOException ignored) {<a name="103" href="#103">103</a> <a name="104" href="#104">104</a> }<a name="105" href="#105">105</a> }<a name="106" href="#106">106</a> <a name="107" href="#107">107</a> <em class="comment">// get doc text out of stream</em><a name="108" href="#108">108</a> String page = out.toString();<a name="109" href="#109">109</a> <a name="110" href="#110">110</a> <em class="comment">// find HYPERLINKs</em><a name="111" href="#111">111</a> <strong>int</strong> currentPos = -1;<a name="112" href="#112">112</a> <strong>int</strong> linkStart = -1;<a name="113" href="#113">113</a> <strong>int</strong> linkEnd = -1;<a name="114" href="#114">114</a> <strong>char</strong> quote = '\"';<a name="115" href="#115">115</a> <a name="116" href="#116">116</a> currentPos = page.indexOf(<span class="string">"HYPERLINK"</span>);<a name="117" href="#117">117</a> <strong>while</strong>(currentPos >= 0){<a name="118" href="#118">118</a> <a name="119" href="#119">119</a> linkStart = page.indexOf(quote, currentPos) + 1;<a name="120" href="#120">120</a> linkEnd = page.indexOf(quote, linkStart);<a name="121" href="#121">121</a> <a name="122" href="#122">122</a> String hyperlink = page.substring(linkStart, linkEnd);<a name="123" href="#123">123</a> <a name="124" href="#124">124</a> <strong>try</strong> {<a name="125" href="#125">125</a> curi.createAndAddLink(hyperlink,Link.NAVLINK_MISC,Link.NAVLINK_HOP);<a name="126" href="#126">126</a> } <strong>catch</strong> (URIException e1) {<a name="127" href="#127">127</a> getController().logUriError(e1, curi.getUURI(), hyperlink);<a name="128" href="#128">128</a> <strong>if</strong> (getController() != <strong>null</strong>) {<a name="129" href="#129">129</a> <em class="comment">// Controller can be null: e.g. when running</em><a name="130" href="#130">130</a> <em class="comment">// ExtractorTool.</em><a name="131" href="#131">131</a> getController().logUriError(e1, curi.getUURI(), hyperlink);<a name="132" href="#132">132</a> } <strong>else</strong> {<a name="133" href="#133">133</a> logger.info(curi + <span class="string">", "</span> + hyperlink + <span class="string">": "</span><a name="134" href="#134">134</a> + e1.getMessage());<a name="135" href="#135">135</a> }<a name="136" href="#136">136</a> }<a name="137" href="#137">137</a> numberOfLinksExtracted++;<a name="138" href="#138">138</a> currentPos = page.indexOf(<span class="string">"HYPERLINK"</span>, linkEnd + 1);<a name="139" href="#139">139</a> }<a name="140" href="#140">140</a> <a name="141" href="#141">141</a> curi.linkExtractorFinished(); <em class="comment">// Set flag to indicate that link extraction is completed.</em><a name="142" href="#142">142</a> logger.fine(curi + <span class="string">" has "</span> + links.size() + <span class="string">" links."</span>);<a name="143" href="#143">143</a> }<a name="144" href="#144">144</a> <a name="145" href="#145">145</a> <em class="comment">/*<em class="comment"> (non-Javadoc)</em></em><a name="146" href="#146">146</a> <em class="comment"> * @see org.archive.crawler.framework.Processor#report()</em><a name="147" href="#147">147</a> <em class="comment"> */</em><a name="148" href="#148">148</a> <strong>public</strong> String report() {<a name="149" href="#149">149</a> StringBuffer ret = <strong>new</strong> StringBuffer();<a name="150" href="#150">150</a> ret.append(<span class="string">"Processor: org.archive.crawler.extractor.ExtractorDOC\n"</span>);<a name="151" href="#151">151</a> ret.append(<span class="string">" Function: Link extraction on MS Word documents (.doc)\n"</span>);<a name="152" href="#152">152</a> ret.append(<span class="string">" CrawlURIs handled: "</span> + numberOfCURIsHandled + <span class="string">"\n"</span>);<a name="153" href="#153">153</a> ret.append(<span class="string">" Links extracted: "</span> + numberOfLinksExtracted + <span class="string">"\n\n"</span>);<a name="154" href="#154">154</a> <a name="155" href="#155">155</a> <strong>return</strong> ret.toString();<a name="156" href="#156">156</a> }<a name="157" href="#157">157</a> }</pre><hr/><div id="footer">This page was automatically generated by <a href="http://maven.apache.org/">Maven</a></div></body></html>⌨️ 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 + -