defaultblockfilesystem.html
来自「网络爬虫开源代码」· HTML 代码 · 共 301 行 · 第 1/2 页
HTML
301 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=UTF-8" /><title>DefaultBlockFileSystem xref</title><link type="text/css" rel="stylesheet" href="../../../../stylesheet.css" /></head><body><div id="overview"><a href="../../../../../apidocs/org/archive/util/ms/DefaultBlockFileSystem.html">View Javadoc</a></div><pre><a name="1" href="#1">1</a> <em class="comment">/*<em class="comment"> DefaultBlockFileSystem</em></em><a name="2" href="#2">2</a> <em class="comment">*</em><a name="3" href="#3">3</a> <em class="comment">* Created on September 12, 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/util/ms/package-summary.html">org.archive.util.ms</a>;<a name="24" href="#24">24</a> <a name="25" href="#25">25</a> <strong>import</strong> java.io.IOException;<a name="26" href="#26">26</a> <strong>import</strong> java.nio.ByteBuffer;<a name="27" href="#27">27</a> <strong>import</strong> java.nio.ByteOrder;<a name="28" href="#28">28</a> <strong>import</strong> java.util.Map;<a name="29" href="#29">29</a> <a name="30" href="#30">30</a> <strong>import</strong> org.archive.io.SeekInputStream;<a name="31" href="#31">31</a> <strong>import</strong> org.archive.util.IoUtils;<a name="32" href="#32">32</a> <strong>import</strong> org.archive.util.LRU;<a name="33" href="#33">33</a> <a name="34" href="#34">34</a> <a name="35" href="#35">35</a> <em>/**<em>*</em></em><a name="36" href="#36">36</a> <em> * Default implementation of the Block File System.</em><a name="37" href="#37">37</a> <em> * </em><a name="38" href="#38">38</a> <em> * <p>The overall structure of a BlockFileSystem file (such as a .doc file) is</em><a name="39" href="#39">39</a> <em> * as follows. The file is divided into blocks, which are of uniform length</em><a name="40" href="#40">40</a> <em> * (512 bytes). The first block (at file pointer 0) is called the header</em><a name="41" href="#41">41</a> <em> * block. It's used to look up other blocks in the file.</em><a name="42" href="#42">42</a> <em> * </em><a name="43" href="#43">43</a> <em> * <p>Subfiles contained within the .doc file are organized using a Block</em><a name="44" href="#44">44</a> <em> * Allocation Table, or BAT. The BAT is basically a linked list; given a </em><a name="45" href="#45">45</a> <em> * block number, the BAT will tell you the next block number. Note that</em><a name="46" href="#46">46</a> <em> * the header block has no number; block #0 is the first block after the</em><a name="47" href="#47">47</a> <em> * header. Thus, to convert a block number to a file pointer:</em><a name="48" href="#48">48</a> <em> * <code>int filePointer = (blockNumber + 1) * BLOCK_SIZE</code>.</em><a name="49" href="#49">49</a> <em> * </em><a name="50" href="#50">50</a> <em> * <p>The BAT itself is discontinuous, however. To find the blocks that </em><a name="51" href="#51">51</a> <em> * comprise the BAT, you have to look in the header block. The header block</em><a name="52" href="#52">52</a> <em> * contains an array of 109 pointers to the blocks that comprise the BAT.</em><a name="53" href="#53">53</a> <em> * If more than 109 BAT blocks are required (in other words, if the .doc</em><a name="54" href="#54">54</a> <em> * file is larger than ~6 megabytes), then something called the </em><a name="55" href="#55">55</a> <em> * XBAT comes into play.</em><a name="56" href="#56">56</a> <em> * </em><a name="57" href="#57">57</a> <em> * <p>XBAT blocks contain pointers to the 110th BAT block and beyond.</em><a name="58" href="#58">58</a> <em> * The first XBAT block is stored at a file pointer listed in the header.</em><a name="59" href="#59">59</a> <em> * The other XBAT blocks are always stored in order after the first; the </em><a name="60" href="#60">60</a> <em> * XBAT table is continuous. One is inclined to wonder why the BAT itself</em><a name="61" href="#61">61</a> <em> * is not so stored, but oh well.</em><a name="62" href="#62">62</a> <em> * </em><a name="63" href="#63">63</a> <em> * <p>The BAT only tells you the next block for a given block. To find the </em><a name="64" href="#64">64</a> <em> * first block for a subfile, you have to look up that subfile's directory</em><a name="65" href="#65">65</a> <em> * entry. Each directory entry is a 128 byte structure in the file, so four</em><a name="66" href="#66">66</a> <em> * of them fit in a block. The number of the first block of the entry list</em><a name="67" href="#67">67</a> <em> * is stored in the header. To find subsequent entry blocks, the BAT must</em><a name="68" href="#68">68</a> <em> * be used.</em><a name="69" href="#69">69</a> <em> * </em><a name="70" href="#70">70</a> <em> * <p>I'm telling you all this so that you understand the caching that this</em><a name="71" href="#71">71</a> <em> * class provides.</em><a name="72" href="#72">72</a> <em> * </em><a name="73" href="#73">73</a> <em> * <p>First, directory entries are not cached. It's assumed that they will</em><a name="74" href="#74">74</a> <em> * be looked up at the beginning of a lengthy operation, and then forgotten</em><a name="75" href="#75">75</a> <em> * about. This is certainly the case for {@link Doc#getText(BlockFileSystem)}. </em><a name="76" href="#76">76</a> <em> * If you need to remember directory entries, you can manually store the Entry </em><a name="77" href="#77">77</a> <em> * objects in a map or something, as they don't grow stale.</em><a name="78" href="#78">78</a> <em> * </em><a name="79" href="#79">79</a> <em> * <p>This class keeps all 512 bytes of the header block in memory at all </em><a name="80" href="#80">80</a> <em> * times. This prevents a potentially expensive file pointer repositioning</em><a name="81" href="#81">81</a> <em> * every time you're trying to figure out what comes next.</em><a name="82" href="#82">82</a> <em> * </em><a name="83" href="#83">83</a> <em> * <p>BAT and XBAT blocks are stored in a least-recently used cache. The </em><a name="84" href="#84">84</a> <em> * <i>n</i> most recent BAT and XBAT blocks are remembered, where <i>n</i></em><a name="85" href="#85">85</a> <em> * is set at construction time. The minimum value of <i>n</i> is 1. For small</em><a name="86" href="#86">86</a> <em> * files, this can prevent file pointer repositioning for BAT look ups.</em><a name="87" href="#87">87</a> <em> * </em><a name="88" href="#88">88</a> <em> * <p>The BAT/XBAT cache only takes up memory as needed. If the specified</em><a name="89" href="#89">89</a> <em> * cache size is 100 blocks, but the file only has 4 BAT blocks, then only </em><a name="90" href="#90">90</a> <em> * 2048 bytes will be used by the cache.</em><a name="91" href="#91">91</a> <em> * </em><a name="92" href="#92">92</a> <em> * <p>Note this class only caches BAT and XBAT blocks. It does not cache the</em><a name="93" href="#93">93</a> <em> * blocks that actually make up a subfile's contents. It is assumed that those</em><a name="94" href="#94">94</a> <em> * blocks will only be accessed once per operation (again, this is what</em><a name="95" href="#95">95</a> <em> * {Doc.getText(BlockFileSystem)} typically requires.)</em><a name="96" href="#96">96</a> <em> * </em><a name="97" href="#97">97</a> <em> * @author pjack</em><a name="98" href="#98">98</a> <em> * @see <a href="http://jakarta.apache.org/poi/poifs/fileformat.html" target="alexandria_uri">http://jakarta.apache.org/poi/poifs/fileformat.html</a></em><a name="99" href="#99">99</a> <em> */</em><a name="100" href="#100">100</a> <strong>public</strong> <strong>class</strong> <a href="../../../../org/archive/util/ms/DefaultBlockFileSystem.html">DefaultBlockFileSystem</a> implements <a href="../../../../org/archive/util/ms/BlockFileSystem.html">BlockFileSystem</a> {<a name="101" href="#101">101</a> <a name="102" href="#102">102</a> <a name="103" href="#103">103</a> <em>/**<em>*</em></em><a name="104" href="#104">104</a> <em> * Pointers per BAT block.</em><a name="105" href="#105">105</a> <em> */</em><a name="106" href="#106">106</a> <strong>final</strong> <strong>private</strong> <strong>static</strong> <strong>int</strong> POINTERS_PER_BAT = 128;<a name="107" href="#107">107</a> <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> * Size of a BAT pointer in bytes. (In other words, 4).</em><a name="111" href="#111">111</a> <em> */</em><a name="112" href="#112">112</a> <strong>final</strong> <strong>private</strong> <strong>static</strong> <strong>int</strong> BAT_POINTER_SIZE = BLOCK_SIZE / POINTERS_PER_BAT;<a name="113" href="#113">113</a> <a name="114" href="#114">114</a> <a name="115" href="#115">115</a> <em>/**<em>*</em></em><a name="116" href="#116">116</a> <em> * The number of BAT pointers in the header block. After this many </em><a name="117" href="#117">117</a> <em> * BAT blocks, the XBAT blocks must be consulted.</em><a name="118" href="#118">118</a> <em> */</em><a name="119" href="#119">119</a> <strong>final</strong> <strong>private</strong> <strong>static</strong> <strong>int</strong> HEADER_BAT_LIMIT = 109;<a name="120" href="#120">120</a> <a name="121" href="#121">121</a> <a name="122" href="#122">122</a> <em>/**<em>*</em></em><a name="123" href="#123">123</a> <em> * The size of an entry record in bytes.</em><a name="124" href="#124">124</a> <em> */</em><a name="125" href="#125">125</a> <strong>final</strong> <strong>private</strong> <strong>static</strong> <strong>int</strong> ENTRY_SIZE = 128;<a name="126" href="#126">126</a> <a name="127" href="#127">127</a> <a name="128" href="#128">128</a> <em>/**<em>*</em></em><a name="129" href="#129">129</a> <em> * The number of entries that can fit in a block.</em><a name="130" href="#130">130</a> <em> */</em><a name="131" href="#131">131</a> <strong>final</strong> <strong>private</strong> <strong>static</strong> <strong>int</strong> ENTRIES_PER_BLOCK = BLOCK_SIZE / ENTRY_SIZE;<a name="132" href="#132">132</a> <a name="133" href="#133">133</a> <a name="134" href="#134">134</a> <em>/**<em>*</em></em><a name="135" href="#135">135</a> <em> * The .doc file as a stream.</em><a name="136" href="#136">136</a> <em> */</em><a name="137" href="#137">137</a> <strong>private</strong> <a href="../../../../org/archive/io/SeekInputStream.html">SeekInputStream</a> input;<a name="138" href="#138">138</a> <a name="139" href="#139">139</a> <a name="140" href="#140">140</a> <em>/**<em>*</em></em><a name="141" href="#141">141</a> <em> * The header block.</em><a name="142" href="#142">142</a> <em> */</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?