multibytereplaycharsequence.html
来自「网络爬虫开源代码」· HTML 代码 · 共 356 行 · 第 1/2 页
HTML
356 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>MultiByteReplayCharSequence 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"> MultiByteReplayCharSequenceFactory</em></em><a name="2" href="#2">2</a> <em class="comment"> *</em><a name="3" href="#3">3</a> <em class="comment"> * (Re)Created on Dec 21, 2006</em><a name="4" href="#4">4</a> <em class="comment"> *</em><a name="5" href="#5">5</a> <em class="comment"> * Copyright (C) 2006 Internet Archive.</em><a name="6" href="#6">6</a> <em class="comment"> *</em><a name="7" href="#7">7</a> <em class="comment"> * This file is part of the Heritrix web crawler (crawler.archive.org).</em><a name="8" href="#8">8</a> <em class="comment"> *</em><a name="9" href="#9">9</a> <em class="comment"> * Heritrix is free software; you can redistribute it and/or modify</em><a name="10" href="#10">10</a> <em class="comment"> * it under the terms of the GNU Lesser Public License as published by</em><a name="11" href="#11">11</a> <em class="comment"> * the Free Software Foundation; either version 2.1 of the License, or</em><a name="12" href="#12">12</a> <em class="comment"> * any later version.</em><a name="13" href="#13">13</a> <em class="comment"> *</em><a name="14" href="#14">14</a> <em class="comment"> * Heritrix is distributed in the hope that it will be useful,</em><a name="15" href="#15">15</a> <em class="comment"> * but WITHOUT ANY WARRANTY; without even the implied warranty of</em><a name="16" href="#16">16</a> <em class="comment"> * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the</em><a name="17" href="#17">17</a> <em class="comment"> * GNU Lesser Public License for more details.</em><a name="18" href="#18">18</a> <em class="comment"> *</em><a name="19" href="#19">19</a> <em class="comment"> * You should have received a copy of the GNU Lesser Public License</em><a name="20" href="#20">20</a> <em class="comment"> * along with Heritrix; if not, write to the Free Software</em><a name="21" href="#21">21</a> <em class="comment"> * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA</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/io/package-summary.html">org.archive.io</a>;<a name="24" href="#24">24</a> <a name="25" href="#25">25</a> <strong>import</strong> java.io.BufferedReader;<a name="26" href="#26">26</a> <strong>import</strong> java.io.BufferedWriter;<a name="27" href="#27">27</a> <strong>import</strong> java.io.File;<a name="28" href="#28">28</a> <strong>import</strong> java.io.FileInputStream;<a name="29" href="#29">29</a> <strong>import</strong> java.io.FileNotFoundException;<a name="30" href="#30">30</a> <strong>import</strong> java.io.FileOutputStream;<a name="31" href="#31">31</a> <strong>import</strong> java.io.IOException;<a name="32" href="#32">32</a> <strong>import</strong> java.io.InputStreamReader;<a name="33" href="#33">33</a> <strong>import</strong> java.io.OutputStreamWriter;<a name="34" href="#34">34</a> <strong>import</strong> java.io.Writer;<a name="35" href="#35">35</a> <strong>import</strong> java.nio.ByteBuffer;<a name="36" href="#36">36</a> <strong>import</strong> java.nio.CharBuffer;<a name="37" href="#37">37</a> <strong>import</strong> java.nio.channels.FileChannel;<a name="38" href="#38">38</a> <strong>import</strong> java.nio.charset.Charset;<a name="39" href="#39">39</a> <strong>import</strong> java.nio.charset.CharsetDecoder;<a name="40" href="#40">40</a> <strong>import</strong> java.nio.charset.CoderResult;<a name="41" href="#41">41</a> <strong>import</strong> java.nio.charset.CodingErrorAction;<a name="42" href="#42">42</a> <strong>import</strong> java.util.logging.Level;<a name="43" href="#43">43</a> <strong>import</strong> java.util.logging.Logger;<a name="44" href="#44">44</a> <a name="45" href="#45">45</a> <em>/**<em>*</em></em><a name="46" href="#46">46</a> <em> * Provides a (Replay)CharSequence view on recorded streams (a prefix</em><a name="47" href="#47">47</a> <em> * buffer and overflow backing file) that can handle streams of multibyte</em><a name="48" href="#48">48</a> <em> * characters.</em><a name="49" href="#49">49</a> <em> *</em><a name="50" href="#50">50</a> <em> * If possible, use {@link ByteReplayCharSequence}. It performs better even</em><a name="51" href="#51">51</a> <em> * for the single byte case (Decoding is an expensive process).</em><a name="52" href="#52">52</a> <em> *</em><a name="53" href="#53">53</a> <em> * <p>Call close on this class when done so can clean up resources.</em><a name="54" href="#54">54</a> <em> *</em><a name="55" href="#55">55</a> <em> * <p>Implementation currently works by checking to see if content to read</em><a name="56" href="#56">56</a> <em> * all fits the in-memory buffer. If so, we decode into a CharBuffer and</em><a name="57" href="#57">57</a> <em> * keep this around for CharSequence operations. This CharBuffer is</em><a name="58" href="#58">58</a> <em> * discarded on close.</em><a name="59" href="#59">59</a> <em> *</em><a name="60" href="#60">60</a> <em> * <p>If content length is greater than in-memory buffer, we decode the</em><a name="61" href="#61">61</a> <em> * buffer plus backing file into a new file named for the backing file w/</em><a name="62" href="#62">62</a> <em> * a suffix of the encoding we write the file as. We then run w/ a</em><a name="63" href="#63">63</a> <em> * memory-mapped CharBuffer against this file to implement CharSequence.</em><a name="64" href="#64">64</a> <em> * Reasons for this implemenation are that CharSequence wants to return the</em><a name="65" href="#65">65</a> <em> * length of the CharSequence.</em><a name="66" href="#66">66</a> <em> *</em><a name="67" href="#67">67</a> <em> * <p>Obvious optimizations would keep around decodings whether the</em><a name="68" href="#68">68</a> <em> * in-memory decoded buffer or the file of decodings written to disk but the</em><a name="69" href="#69">69</a> <em> * general usage pattern processing URIs is that the decoding is used by one</em><a name="70" href="#70">70</a> <em> * processor only. Also of note, files usually fit into the in-memory</em><a name="71" href="#71">71</a> <em> * buffer.</em><a name="72" href="#72">72</a> <em> *</em><a name="73" href="#73">73</a> <em> * <p>We might also be able to keep up 3 windows that moved across the file</em><a name="74" href="#74">74</a> <em> * decoding a window at a time trying to keep one of the buffers just in</em><a name="75" href="#75">75</a> <em> * front of the regex processing returning it a length that would be only</em><a name="76" href="#76">76</a> <em> * the length of current position to end of current block or else the length</em><a name="77" href="#77">77</a> <em> * could be got by multipling the backing files length by the decoders'</em><a name="78" href="#78">78</a> <em> * estimate of average character size. This would save us writing out the</em><a name="79" href="#79">79</a> <em> * decoded file. We'd have to do the latter for files that are</em><a name="80" href="#80">80</a> <em> * > Integer.MAX_VALUE.</em><a name="81" href="#81">81</a> <em> *</em><a name="82" href="#82">82</a> <em> * @author stack</em><a name="83" href="#83">83</a> <em> * @version $Revision: 4844 $, $Date: 2007-01-10 17:18:34 +0000 (Wed, 10 Jan 2007) $</em><a name="84" href="#84">84</a> <em> */</em><a name="85" href="#85">85</a> <strong>public</strong> <strong>class</strong> <a href="../../../org/archive/io/MultiByteReplayCharSequence.html">MultiByteReplayCharSequence</a> implements <a href="../../../org/archive/io/ReplayCharSequence.html">ReplayCharSequence</a> {<a name="86" href="#86">86</a> <a name="87" href="#87">87</a> <strong>protected</strong> <strong>static</strong> Logger logger =<a name="88" href="#88">88</a> Logger.getLogger(MultiByteReplayCharSequence.<strong>class</strong>.getName());<a name="89" href="#89">89</a> <a name="90" href="#90">90</a> <em>/**<em>*</em></em><a name="91" href="#91">91</a> <em> * Name of the encoding we use writing out concatenated decoded prefix</em><a name="92" href="#92">92</a> <em> * buffer and decoded backing file.</em><a name="93" href="#93">93</a> <em> *</em><a name="94" href="#94">94</a> <em> * <p>This define is also used as suffix for the file that holds the</em><a name="95" href="#95">95</a> <em> * decodings. The name of the file that holds the decoding is the name</em><a name="96" href="#96">96</a> <em> * of the backing file w/ this encoding for a suffix.</em><a name="97" href="#97">97</a> <em> *</em><a name="98" href="#98">98</a> <em> * <p>See <a ref="<a href="http://java.sun.com/j2se/1.4.2/docs/guide/intl/encoding.doc.html" target="alexandria_uri">http://java.sun.com/j2se/1.4.2/docs/guide/intl/encoding.doc.html</a>">Encoding</a>.</em><a name="99" href="#99">99</a> <em> */</em><a name="100" href="#100">100</a> <strong>private</strong> <strong>static</strong> <strong>final</strong> String WRITE_ENCODING = <span class="string">"UTF-16BE"</span>;<a name="101" href="#101">101</a> <a name="102" href="#102">102</a> <em>/**<em>*</em></em><a name="103" href="#103">103</a> <em> * CharBuffer of decoded content.</em><a name="104" href="#104">104</a> <em> *</em><a name="105" href="#105">105</a> <em> * Content of this buffer is unicode.</em><a name="106" href="#106">106</a> <em> */</em><a name="107" href="#107">107</a> <strong>private</strong> CharBuffer content = <strong>null</strong>;<a name="108" href="#108">108</a> <a name="109" href="#109">109</a> <em>/**<em>*</em></em><a name="110" href="#110">110</a> <em> * File that has decoded content.</em><a name="111" href="#111">111</a> <em> *</em><a name="112" href="#112">112</a> <em> * Keep it around so we can remove on close.</em><a name="113" href="#113">113</a> <em> */</em><a name="114" href="#114">114</a> <strong>private</strong> File decodedFile = <strong>null</strong>;<a name="115" href="#115">115</a> <a name="116" href="#116">116</a> <a name="117" href="#117">117</a> <em>/**<em>*</em></em><a name="118" href="#118">118</a> <em> * Constructor for all in-memory operation.</em><a name="119" href="#119">119</a> <em> *</em><a name="120" href="#120">120</a> <em> * @param buffer In-memory buffer of recordings prefix. We read from</em><a name="121" href="#121">121</a> <em> * here first and will only go to the backing file if <code>size</code></em><a name="122" href="#122">122</a> <em> * requested is greater than <code>buffer.length</code>.</em><a name="123" href="#123">123</a> <em> * @param size Total size of stream to replay in bytes. Used to find</em><a name="124" href="#124">124</a> <em> * EOS. This is total length of content including HTTP headers if</em><a name="125" href="#125">125</a> <em> * present.</em><a name="126" href="#126">126</a> <em> * @param responseBodyStart Where the response body starts in bytes.</em><a name="127" href="#127">127</a> <em> * Used to skip over the HTTP headers if present.</em><a name="128" href="#128">128</a> <em> * @param backingFilename Path to backing file with content in excess of</em><a name="129" href="#129">129</a> <em> * whats in <code>buffer</code>.</em><a name="130" href="#130">130</a> <em> * @param encoding Encoding to use reading the passed prefix buffer and</em><a name="131" href="#131">131</a> <em> * backing file. For now, should be java canonical name for the</em><a name="132" href="#132">132</a> <em> * encoding. (If null is passed, we will default to</em><a name="133" href="#133">133</a> <em> * ByteReplayCharSequence).</em><a name="134" href="#134">134</a> <em> *</em><a name="135" href="#135">135</a> <em> * @throws IOException</em><a name="136" href="#136">136</a> <em> */</em><a name="137" href="#137">137</a> <strong>public</strong> <a href="../../../org/archive/io/MultiByteReplayCharSequence.html">MultiByteReplayCharSequence</a>(byte[] buffer, <strong>long</strong> size,<a name="138" href="#138">138</a> <strong>long</strong> responseBodyStart, String encoding)<a name="139" href="#139">139</a> throws IOException {<a name="140" href="#140">140</a> <strong>super</strong>();<a name="141" href="#141">141</a> <strong>this</strong>.content = decodeInMemory(buffer, size, responseBodyStart, <a name="142" href="#142">142</a> encoding);<a name="143" href="#143">143</a> }<a name="144" href="#144">144</a> <a name="145" href="#145">145</a> <em>/**<em>*</em></em><a name="146" href="#146">146</a> <em> * Constructor for overflow-to-disk-file operation.</em><a name="147" href="#147">147</a> <em> *</em><a name="148" href="#148">148</a> <em> * @param contentReplayInputStream inputStream of content</em><a name="149" href="#149">149</a> <em> * @param backingFilename hint for name of temp file</em><a name="150" href="#150">150</a> <em> * @param characterEncoding Encoding to use reading the stream.</em><a name="151" href="#151">151</a> <em> * For now, should be java canonical name for the</em><a name="152" href="#152">152</a> <em> * encoding. </em><a name="153" href="#153">153</a> <em> *</em><a name="154" href="#154">154</a> <em> * @throws IOException</em><a name="155" href="#155">155</a> <em> */</em><a name="156" href="#156">156</a> <strong>public</strong> <a href="../../../org/archive/io/MultiByteReplayCharSequence.html">MultiByteReplayCharSequence</a>(<a name="157" href="#157">157</a> <a href="../../../org/archive/io/ReplayInputStream.html">ReplayInputStream</a> contentReplayInputStream,<a name="158" href="#158">158</a> String backingFilename,<a name="159" href="#159">159</a> String characterEncoding)<a name="160" href="#160">160</a> throws IOException {<a name="161" href="#161">161</a> <strong>super</strong>();<a name="162" href="#162">162</a> <strong>this</strong>.content = decodeToFile(contentReplayInputStream, <a name="163" href="#163">163</a> backingFilename, characterEncoding);<a name="164" href="#164">164</a> }<a name="165" href="#165">165</a> <a name="166" href="#166">166</a> <em>/**<em>*</em></em><a name="167" href="#167">167</a> <em> * Decode passed buffer and backing file into a CharBuffer.</em><a name="168" href="#168">168</a> <em> *</em><a name="169" href="#169">169</a> <em> * This method writes a new file made of the decoded concatenation of</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?