?? recoverylogmapper.html
字號:
<!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>RecoveryLogMapper xref</title><link type="text/css" rel="stylesheet" href="../../../../stylesheet.css" /></head><body><div id="overview"><a href="../../../../../apidocs/org/archive/crawler/util/RecoveryLogMapper.html">View Javadoc</a></div><pre><a name="1" href="#1">1</a> <em class="comment">/*<em class="comment"> RecoveryLogMapper.java</em></em><a name="2" href="#2">2</a> <em class="comment">*</em><a name="3" href="#3">3</a> <em class="comment">* $Id: RecoveryLogMapper.java 4647 2006-09-22 18:39:39Z paul_jack $</em><a name="4" href="#4">4</a> <em class="comment">*</em><a name="5" href="#5">5</a> <em class="comment">* Created on Mar 7, 2005</em><a name="6" href="#6">6</a> <em class="comment">*</em><a name="7" href="#7">7</a> <em class="comment">* Copyright (C) 2005 Mike Schwartz.</em><a name="8" href="#8">8</a> <em class="comment">*</em><a name="9" href="#9">9</a> <em class="comment">* This file is part of the Heritrix web crawler (crawler.archive.org).</em><a name="10" href="#10">10</a> <em class="comment">*</em><a name="11" href="#11">11</a> <em class="comment">* Heritrix is free software; you can redistribute it and/or modify</em><a name="12" href="#12">12</a> <em class="comment">* it under the terms of the GNU Lesser Public License as published by</em><a name="13" href="#13">13</a> <em class="comment">* the Free Software Foundation; either version 2.1 of the License, or</em><a name="14" href="#14">14</a> <em class="comment">* any later version.</em><a name="15" href="#15">15</a> <em class="comment">*</em><a name="16" href="#16">16</a> <em class="comment">* Heritrix is distributed in the hope that it will be useful,</em><a name="17" href="#17">17</a> <em class="comment">* but WITHOUT ANY WARRANTY; without even the implied warranty of</em><a name="18" href="#18">18</a> <em class="comment">* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the</em><a name="19" href="#19">19</a> <em class="comment">* GNU Lesser Public License for more details.</em><a name="20" href="#20">20</a> <em class="comment">*</em><a name="21" href="#21">21</a> <em class="comment">* You should have received a copy of the GNU Lesser Public License</em><a name="22" href="#22">22</a> <em class="comment">* along with Heritrix; if not, write to the Free Software</em><a name="23" href="#23">23</a> <em class="comment">* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA</em><a name="24" href="#24">24</a> <em class="comment">*/</em><a name="25" href="#25">25</a> <a name="26" href="#26">26</a> <em>/**<em>*</em></em><a name="27" href="#27">27</a> <em> * Parses a Heritrix recovery log file (recover.gz), and builds maps</em><a name="28" href="#28">28</a> <em> * that allow a caller to look up any seed URL and get back an Iterator of all</em><a name="29" href="#29">29</a> <em> * URLs successfully crawled from given seed.</em><a name="30" href="#30">30</a> <em> *</em><a name="31" href="#31">31</a> <em> * Also allows lookup on any crawled</em><a name="32" href="#32">32</a> <em> * URL to find the seed URL from which the crawler reached that URL (through 1</em><a name="33" href="#33">33</a> <em> * or more discovered URL hops, which are collapsed in this lookup).</em><a name="34" href="#34">34</a> <em> * </em><a name="35" href="#35">35</a> <em> * <p>This code creates some fairly large collections (proprotionate in size to</em><a name="36" href="#36">36</a> <em> * # discovered URLs) so make sure you allocate</em><a name="37" href="#37">37</a> <em> * it a large heap to work in. It also takes a while to process a recover log.</em><a name="38" href="#38">38</a> <em> * <p>See {@link #main()} method at end for test/demo code.</em><a name="39" href="#39">39</a> <em> * @author Mike Schwartz, schwartz at CodeOnTheRoad dot com</em><a name="40" href="#40">40</a> <em> */</em><a name="41" href="#41">41</a> <strong>package</strong> <a href="../../../../org/archive/crawler/util/package-summary.html">org.archive.crawler.util</a>;<a name="42" href="#42">42</a> <a name="43" href="#43">43</a> <strong>import</strong> org.archive.crawler.frontier.RecoveryJournal;<a name="44" href="#44">44</a> <a name="45" href="#45">45</a> <strong>import</strong> java.io.File;<a name="46" href="#46">46</a> <strong>import</strong> java.io.LineNumberReader;<a name="47" href="#47">47</a> <strong>import</strong> java.io.PrintWriter;<a name="48" href="#48">48</a> <strong>import</strong> java.io.FileOutputStream;<a name="49" href="#49">49</a> <strong>import</strong> java.util.Collection;<a name="50" href="#50">50</a> <strong>import</strong> java.util.HashMap;<a name="51" href="#51">51</a> <strong>import</strong> java.util.HashSet;<a name="52" href="#52">52</a> <strong>import</strong> java.util.Iterator;<a name="53" href="#53">53</a> <strong>import</strong> java.util.Map;<a name="54" href="#54">54</a> <strong>import</strong> java.util.Set;<a name="55" href="#55">55</a> <strong>import</strong> java.util.logging.Level;<a name="56" href="#56">56</a> <strong>import</strong> java.util.logging.Logger;<a name="57" href="#57">57</a> <a name="58" href="#58">58</a> <strong>public</strong> <strong>class</strong> <a href="../../../../org/archive/crawler/util/RecoveryLogMapper.html">RecoveryLogMapper</a> {<a name="59" href="#59">59</a> <strong>private</strong> <strong>static</strong> <strong>final</strong> <strong>char</strong> LOG_LINE_START_CHAR =<a name="60" href="#60">60</a> RecoveryJournal.F_ADD.charAt(0);<a name="61" href="#61">61</a> <strong>private</strong> <strong>static</strong> <strong>final</strong> Logger logger =<a name="62" href="#62">62</a> Logger.getLogger(RecoveryLogMapper.<strong>class</strong>.getName());<a name="63" href="#63">63</a> <strong>private</strong> PrintWriter seedNotFoundPrintWriter = <strong>null</strong>;<a name="64" href="#64">64</a> <a name="65" href="#65">65</a> <em>/**<em>*</em></em><a name="66" href="#66">66</a> <em> * Tracks seed for each crawled URL</em><a name="67" href="#67">67</a> <em> */</em><a name="68" href="#68">68</a> <strong>private</strong> Map<String,String> crawledUrlToSeedMap<a name="69" href="#69">69</a> = <strong>new</strong> HashMap<String,String>();<a name="70" href="#70">70</a> <a name="71" href="#71">71</a> <em>/**<em>*</em></em><a name="72" href="#72">72</a> <em> * Maps seed URLs to Set of discovered URLs</em><a name="73" href="#73">73</a> <em> */</em><a name="74" href="#74">74</a> <strong>private</strong> Map<String,Set<String>> seedUrlToDiscoveredUrlsMap<a name="75" href="#75">75</a> = <strong>new</strong> HashMap<String,Set<String>>();<a name="76" href="#76">76</a> <a name="77" href="#77">77</a> <em>/**<em>*</em></em><a name="78" href="#78">78</a> <em> * Tracks which URLs were successfully crawled</em><a name="79" href="#79">79</a> <em> */</em><a name="80" href="#80">80</a> <strong>private</strong> Set<String> successfullyCrawledUrls = <strong>new</strong> HashSet<String>();<a name="81" href="#81">81</a> <a name="82" href="#82">82</a> <em>/**<em>*</em></em><a name="83" href="#83">83</a> <em> * Normal constructor - if encounter not-found seeds while loading</em><a name="84" href="#84">84</a> <em> * recoverLogFileName, will throw throw SeedUrlNotFoundException.</em><a name="85" href="#85">85</a> <em> * Use {@link #RecoveryLogMapper(String)} if you want to just log</em><a name="86" href="#86">86</a> <em> * such cases and keep going. (Those should not happen if the</em><a name="87" href="#87">87</a> <em> * recover log is written correctly, but we see them in pratice.)</em><a name="88" href="#88">88</a> <em> * @param recoverLogFileName</em><a name="89" href="#89">89</a> <em> * @throws java.io.FileNotFoundException </em><a name="90" href="#90">90</a> <em> * @throws java.io.IOException </em><a name="91" href="#91">91</a> <em> * @throws SeedUrlNotFoundException </em><a name="92" href="#92">92</a> <em> */</em><a name="93" href="#93">93</a> <strong>public</strong> <a href="../../../../org/archive/crawler/util/RecoveryLogMapper.html">RecoveryLogMapper</a>(String recoverLogFileName)<a name="94" href="#94">94</a> throws java.io.FileNotFoundException, java.io.IOException,<a name="95" href="#95">95</a> <a href="../../../../org/archive/crawler/util/SeedUrlNotFoundException.html">SeedUrlNotFoundException</a> {<a name="96" href="#96">96</a> load(recoverLogFileName);<a name="97" href="#97">97</a> }<a name="98" href="#98">98</a> <a name="99" href="#99">99</a> <em>/**<em>*</em></em><a name="100" href="#100">100</a> <em> * Constructor to use if you want to allow not-found seeds, logging</em><a name="101" href="#101">101</a> <em> * them to seedNotFoundLogFileName. In contrast, {@link</em><a name="102" href="#102">102</a> <em> * #RecoveryLogMapper(String)} will throw SeedUrlNotFoundException</em><a name="103" href="#103">103</a> <em> * when a seed isn't found.</em><a name="104" href="#104">104</a> <em> * @param recoverLogFileName</em><a name="105" href="#105">105</a> <em> * @param seedNotFoundLogFileName</em><a name="106" href="#106">106</a> <em> */</em><a name="107" href="#107">107</a> <strong>public</strong> <a href="../../../../org/archive/crawler/util/RecoveryLogMapper.html">RecoveryLogMapper</a>(String recoverLogFileName,<a name="108" href="#108">108</a> String seedNotFoundLogFileName)<a name="109" href="#109">109</a> throws java.io.FileNotFoundException, java.io.IOException,<a name="110" href="#110">110</a> <a href="../../../../org/archive/crawler/util/SeedUrlNotFoundException.html">SeedUrlNotFoundException</a> {<a name="111" href="#111">111</a> seedNotFoundPrintWriter = <strong>new</strong> PrintWriter(<strong>new</strong> FileOutputStream(<a name="112" href="#112">112</a> seedNotFoundLogFileName));<a name="113" href="#113">113</a> load(recoverLogFileName);<a name="114" href="#114">114</a> }<a name="115" href="#115">115</a> <a name="116" href="#116">116</a> <strong>protected</strong> <strong>void</strong> load(String recoverLogFileName)<a name="117" href="#117">117</a> throws java.io.FileNotFoundException, java.io.IOException,<a name="118" href="#118">118</a> <a href="../../../../org/archive/crawler/util/SeedUrlNotFoundException.html">SeedUrlNotFoundException</a> {<a name="119" href="#119">119</a> LineNumberReader reader = <strong>new</strong> LineNumberReader(RecoveryJournal.<a name="120" href="#120">120</a> getBufferedReader(<strong>new</strong> File(recoverLogFileName)));<a name="121" href="#121">121</a> String curLine = <strong>null</strong>;<a name="122" href="#122">122</a> <strong>while</strong> ((curLine = reader.readLine()) != <strong>null</strong>) {<a name="123" href="#123">123</a> <strong>if</strong> (curLine.length() == 0<a name="124" href="#124">124</a> || curLine.charAt(0) != LOG_LINE_START_CHAR) {<a name="125" href="#125">125</a> <strong>continue</strong>;<a name="126" href="#126">126</a> }<a name="127" href="#127">127</a> String args[] = curLine.split(<span class="string">"//s+"</span>);<a name="128" href="#128">128</a> <strong>int</strong> curLineNumWords = args.length;<a name="129" href="#129">129</a> String firstUrl = args[1];<a name="130" href="#130">130</a> <em class="comment">// Ignore DNS log entries</em><a name="131" href="#131">131</a> <strong>if</strong> (firstUrl.startsWith(<span class="string">"dns:"</span>)) {<a name="132" href="#132">132</a> <strong>continue</strong>;<a name="133" href="#133">133</a> }<a name="134" href="#134">134</a> <strong>if</strong> (curLine.startsWith(RecoveryJournal.F_ADD)) {<a name="135" href="#135">135</a> <em class="comment">// Seed URL</em><a name="136" href="#136">136</a> <strong>if</strong> (curLineNumWords == 2) {<a name="137" href="#137">137</a> <strong>if</strong> (logger.isLoggable(Level.FINE)) {<a name="138" href="#138">138</a> logger.fine(<span class="string">"F_ADD with 2 words --> seed URL ("</span> +<a name="139" href="#139">139</a> firstUrl + <span class="string">")"</span>);<a name="140" href="#140">140</a> }<a name="141" href="#141">141</a> <em class="comment">// Add seed the first time we find it</em><a name="142" href="#142">142</a> <strong>if</strong> (seedUrlToDiscoveredUrlsMap.get(firstUrl) == <strong>null</strong>) {<a name="143" href="#143">143</a> seedUrlToDiscoveredUrlsMap.put(firstUrl,<a name="144" href="#144">144</a> <strong>new</strong> HashSet<String>());<a name="145" href="#145">145</a> }<a name="146" href="#146">146</a> } <strong>else</strong> {<a name="147" href="#147">147</a> <em class="comment">// URL found via an earlier seeded / discovered URL</em><a name="148" href="#148">148</a> <em class="comment">// Look for the seed from which firstUrlString came, so</em><a name="149" href="#149">149</a> <em class="comment">// we can collapse new URLString back to it</em><a name="150" href="#150">150</a> String viaUrl = args[curLineNumWords - 1];<a name="151" href="#151">151</a> <strong>if</strong> (logger.isLoggable(Level.FINE)) {<a name="152" href="#152">152</a> logger.fine(<span class="string">"F_ADD with 3+ words --> new URL "</span><a name="153" href="#153">153</a> + firstUrl + <span class="string">" via URL "</span> + viaUrl);<a name="154" href="#154">154</a> }<a name="155" href="#155">155</a> String seedForFirstUrl =<a name="156" href="#156">156</a> (String) crawledUrlToSeedMap.get(viaUrl);<a name="157" href="#157">157</a> <em class="comment">// viaUrlString is a seed URL</em><a name="158" href="#158">158</a> <strong>if</strong> (seedForFirstUrl == <strong>null</strong>) {<a name="159" href="#159">159</a> <strong>if</strong> (logger.isLoggable(Level.FINE)) {<a name="160" href="#160">160</a> logger.fine(<span class="string">"\tvia URL is a seed"</span>);<a name="161" href="#161">161</a> }?? 快捷鍵說明
復制代碼
Ctrl + C
搜索代碼
Ctrl + F
全屏模式
F11
切換主題
Ctrl + Shift + D
顯示快捷鍵
?
增大字號
Ctrl + =
減小字號
Ctrl + -