Google Git
Sign in
apache / hbase-site / fbcf6c89b23841dd0de8ef3466b759a547994d66 / . / devapidocs / src-html / org / apache / hadoop / hbase / io / hfile / HFileScanner.html
blob: 87894a37ad5cddddfd5104f7506183b2f0dff78d [file] [log] [blame]
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html lang="en">
<head>
<title>Source code</title>
<link rel="stylesheet" type="text/css" href="../../../../../../../stylesheet.css" title="Style">
</head>
<body>
<div class="sourceContainer">
<pre><span class="sourceLineNo">001</span>/*<a name="line.1"></a>
<span class="sourceLineNo">002</span> * Licensed to the Apache Software Foundation (ASF) under one<a name="line.2"></a>
<span class="sourceLineNo">003</span> * or more contributor license agreements. See the NOTICE file<a name="line.3"></a>
<span class="sourceLineNo">004</span> * distributed with this work for additional information<a name="line.4"></a>
<span class="sourceLineNo">005</span> * regarding copyright ownership. The ASF licenses this file<a name="line.5"></a>
<span class="sourceLineNo">006</span> * to you under the Apache License, Version 2.0 (the<a name="line.6"></a>
<span class="sourceLineNo">007</span> * "License"); you may not use this file except in compliance<a name="line.7"></a>
<span class="sourceLineNo">008</span> * with the License. You may obtain a copy of the License at<a name="line.8"></a>
<span class="sourceLineNo">009</span> *<a name="line.9"></a>
<span class="sourceLineNo">010</span> * http://www.apache.org/licenses/LICENSE-2.0<a name="line.10"></a>
<span class="sourceLineNo">011</span> *<a name="line.11"></a>
<span class="sourceLineNo">012</span> * Unless required by applicable law or agreed to in writing, software<a name="line.12"></a>
<span class="sourceLineNo">013</span> * distributed under the License is distributed on an "AS IS" BASIS,<a name="line.13"></a>
<span class="sourceLineNo">014</span> * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.<a name="line.14"></a>
<span class="sourceLineNo">015</span> * See the License for the specific language governing permissions and<a name="line.15"></a>
<span class="sourceLineNo">016</span> * limitations under the License.<a name="line.16"></a>
<span class="sourceLineNo">017</span> */<a name="line.17"></a>
<span class="sourceLineNo">018</span>package org.apache.hadoop.hbase.io.hfile;<a name="line.18"></a>
<span class="sourceLineNo">019</span><a name="line.19"></a>
<span class="sourceLineNo">020</span>import java.io.Closeable;<a name="line.20"></a>
<span class="sourceLineNo">021</span>import java.io.IOException;<a name="line.21"></a>
<span class="sourceLineNo">022</span>import java.nio.ByteBuffer;<a name="line.22"></a>
<span class="sourceLineNo">023</span>import java.util.function.IntConsumer;<a name="line.23"></a>
<span class="sourceLineNo">024</span>import org.apache.hadoop.hbase.Cell;<a name="line.24"></a>
<span class="sourceLineNo">025</span>import org.apache.hadoop.hbase.regionserver.Shipper;<a name="line.25"></a>
<span class="sourceLineNo">026</span>import org.apache.yetus.audience.InterfaceAudience;<a name="line.26"></a>
<span class="sourceLineNo">027</span><a name="line.27"></a>
<span class="sourceLineNo">028</span>/**<a name="line.28"></a>
<span class="sourceLineNo">029</span> * A scanner allows you to position yourself within a HFile and scan through it. It allows you to<a name="line.29"></a>
<span class="sourceLineNo">030</span> * reposition yourself as well.<a name="line.30"></a>
<span class="sourceLineNo">031</span> * &lt;p&gt;<a name="line.31"></a>
<span class="sourceLineNo">032</span> * A scanner doesn't always have a key/value that it is pointing to when it is first created and<a name="line.32"></a>
<span class="sourceLineNo">033</span> * before {@link #seekTo()}/{@link #seekTo(Cell)} are called. In this case,<a name="line.33"></a>
<span class="sourceLineNo">034</span> * {@link #getKey()}/{@link #getValue()} returns null. At most other times, a key and value will be<a name="line.34"></a>
<span class="sourceLineNo">035</span> * available. The general pattern is that you position the Scanner using the seekTo variants and<a name="line.35"></a>
<span class="sourceLineNo">036</span> * then getKey and getValue.<a name="line.36"></a>
<span class="sourceLineNo">037</span> */<a name="line.37"></a>
<span class="sourceLineNo">038</span>@InterfaceAudience.Private<a name="line.38"></a>
<span class="sourceLineNo">039</span>public interface HFileScanner extends Shipper, Closeable {<a name="line.39"></a>
<span class="sourceLineNo">040</span> /**<a name="line.40"></a>
<span class="sourceLineNo">041</span> * SeekTo or just before the passed &lt;code&gt;cell&lt;/code&gt;. Examine the return code to figure whether<a name="line.41"></a>
<span class="sourceLineNo">042</span> * we found the cell or not. Consider the cell stream of all the cells in the file,<a name="line.42"></a>
<span class="sourceLineNo">043</span> * &lt;code&gt;c[0] .. c[n]&lt;/code&gt;, where there are n cells in the file.<a name="line.43"></a>
<span class="sourceLineNo">044</span> * @return -1, if cell &amp;lt; c[0], no position; 0, such that c[i] = cell and scanner is left in<a name="line.44"></a>
<span class="sourceLineNo">045</span> * position i; and 1, such that c[i] &amp;lt; cell, and scanner is left in position i. The<a name="line.45"></a>
<span class="sourceLineNo">046</span> * scanner will position itself between c[i] and c[i+1] where c[i] &amp;lt; cell &amp;lt;= c[i+1].<a name="line.46"></a>
<span class="sourceLineNo">047</span> * If there is no cell c[i+1] greater than or equal to the input cell, then the scanner<a name="line.47"></a>
<span class="sourceLineNo">048</span> * will position itself at the end of the file and next() will return false when it is<a name="line.48"></a>
<span class="sourceLineNo">049</span> * called.<a name="line.49"></a>
<span class="sourceLineNo">050</span> */<a name="line.50"></a>
<span class="sourceLineNo">051</span> int seekTo(Cell cell) throws IOException;<a name="line.51"></a>
<span class="sourceLineNo">052</span><a name="line.52"></a>
<span class="sourceLineNo">053</span> /**<a name="line.53"></a>
<span class="sourceLineNo">054</span> * Reseek to or just before the passed &lt;code&gt;cell&lt;/code&gt;. Similar to seekTo except that this can<a name="line.54"></a>
<span class="sourceLineNo">055</span> * be called even if the scanner is not at the beginning of a file. This can be used to seek only<a name="line.55"></a>
<span class="sourceLineNo">056</span> * to cells which come after the current position of the scanner. Consider the cell stream of all<a name="line.56"></a>
<span class="sourceLineNo">057</span> * the cells in the file, &lt;code&gt;c[0] .. c[n]&lt;/code&gt;, where there are n cellc in the file after<a name="line.57"></a>
<span class="sourceLineNo">058</span> * current position of HFileScanner. The scanner will position itself between c[i] and c[i+1]<a name="line.58"></a>
<span class="sourceLineNo">059</span> * where c[i] &amp;lt; cell &amp;lt;= c[i+1]. If there is no cell c[i+1] greater than or equal to the<a name="line.59"></a>
<span class="sourceLineNo">060</span> * input cell, then the scanner will position itself at the end of the file and next() will return<a name="line.60"></a>
<span class="sourceLineNo">061</span> * false when it is called.<a name="line.61"></a>
<span class="sourceLineNo">062</span> * @param cell Cell to find (should be non-null)<a name="line.62"></a>
<span class="sourceLineNo">063</span> * @return -1, if cell &amp;lt; c[0], no position; 0, such that c[i] = cell and scanner is left in<a name="line.63"></a>
<span class="sourceLineNo">064</span> * position i; and 1, such that c[i] &amp;lt; cell, and scanner is left in position i.<a name="line.64"></a>
<span class="sourceLineNo">065</span> */<a name="line.65"></a>
<span class="sourceLineNo">066</span> int reseekTo(Cell cell) throws IOException;<a name="line.66"></a>
<span class="sourceLineNo">067</span><a name="line.67"></a>
<span class="sourceLineNo">068</span> /**<a name="line.68"></a>
<span class="sourceLineNo">069</span> * Consider the cell stream of all the cells in the file, &lt;code&gt;c[0] .. c[n]&lt;/code&gt;, where there<a name="line.69"></a>
<span class="sourceLineNo">070</span> * are n cells in the file.<a name="line.70"></a>
<span class="sourceLineNo">071</span> * @param cell Cell to find<a name="line.71"></a>
<span class="sourceLineNo">072</span> * @return false if cell &amp;lt;= c[0] or true with scanner in position 'i' such that: c[i] &amp;lt;<a name="line.72"></a>
<span class="sourceLineNo">073</span> * cell. Furthermore: there may be a c[i+1], such that c[i] &amp;lt; cell &amp;lt;= c[i+1] but<a name="line.73"></a>
<span class="sourceLineNo">074</span> * there may also NOT be a c[i+1], and next() will return false (EOF).<a name="line.74"></a>
<span class="sourceLineNo">075</span> */<a name="line.75"></a>
<span class="sourceLineNo">076</span> boolean seekBefore(Cell cell) throws IOException;<a name="line.76"></a>
<span class="sourceLineNo">077</span><a name="line.77"></a>
<span class="sourceLineNo">078</span> /**<a name="line.78"></a>
<span class="sourceLineNo">079</span> * Positions this scanner at the start of the file.<a name="line.79"></a>
<span class="sourceLineNo">080</span> * @return False if empty file; i.e. a call to next would return false and the current key and<a name="line.80"></a>
<span class="sourceLineNo">081</span> * value are undefined.<a name="line.81"></a>
<span class="sourceLineNo">082</span> */<a name="line.82"></a>
<span class="sourceLineNo">083</span> boolean seekTo() throws IOException;<a name="line.83"></a>
<span class="sourceLineNo">084</span><a name="line.84"></a>
<span class="sourceLineNo">085</span> /**<a name="line.85"></a>
<span class="sourceLineNo">086</span> * Scans to the next entry in the file.<a name="line.86"></a>
<span class="sourceLineNo">087</span> * @return Returns false if you are at the end otherwise true if more in file.<a name="line.87"></a>
<span class="sourceLineNo">088</span> */<a name="line.88"></a>
<span class="sourceLineNo">089</span> boolean next() throws IOException;<a name="line.89"></a>
<span class="sourceLineNo">090</span><a name="line.90"></a>
<span class="sourceLineNo">091</span> /**<a name="line.91"></a>
<span class="sourceLineNo">092</span> * Gets the current key in the form of a cell. You must call {@link #seekTo(Cell)} before this<a name="line.92"></a>
<span class="sourceLineNo">093</span> * method.<a name="line.93"></a>
<span class="sourceLineNo">094</span> * @return gets the current key as a Cell.<a name="line.94"></a>
<span class="sourceLineNo">095</span> */<a name="line.95"></a>
<span class="sourceLineNo">096</span> Cell getKey();<a name="line.96"></a>
<span class="sourceLineNo">097</span><a name="line.97"></a>
<span class="sourceLineNo">098</span> /**<a name="line.98"></a>
<span class="sourceLineNo">099</span> * Gets a buffer view to the current value. You must call {@link #seekTo(Cell)} before this<a name="line.99"></a>
<span class="sourceLineNo">100</span> * method.<a name="line.100"></a>
<span class="sourceLineNo">101</span> * @return byte buffer for the value. The limit is set to the value size, and the position is 0,<a name="line.101"></a>
<span class="sourceLineNo">102</span> * the start of the buffer view.<a name="line.102"></a>
<span class="sourceLineNo">103</span> */<a name="line.103"></a>
<span class="sourceLineNo">104</span> ByteBuffer getValue();<a name="line.104"></a>
<span class="sourceLineNo">105</span><a name="line.105"></a>
<span class="sourceLineNo">106</span> /** Returns Instance of {@link org.apache.hadoop.hbase.Cell}. */<a name="line.106"></a>
<span class="sourceLineNo">107</span> Cell getCell();<a name="line.107"></a>
<span class="sourceLineNo">108</span><a name="line.108"></a>
<span class="sourceLineNo">109</span> /** Returns Reader that underlies this Scanner instance. */<a name="line.109"></a>
<span class="sourceLineNo">110</span> HFile.Reader getReader();<a name="line.110"></a>
<span class="sourceLineNo">111</span><a name="line.111"></a>
<span class="sourceLineNo">112</span> /**<a name="line.112"></a>
<span class="sourceLineNo">113</span> * @return True is scanner has had one of the seek calls invoked; i.e. {@link #seekBefore(Cell)}<a name="line.113"></a>
<span class="sourceLineNo">114</span> * or {@link #seekTo()} or {@link #seekTo(Cell)}. Otherwise returns false.<a name="line.114"></a>
<span class="sourceLineNo">115</span> */<a name="line.115"></a>
<span class="sourceLineNo">116</span> boolean isSeeked();<a name="line.116"></a>
<span class="sourceLineNo">117</span><a name="line.117"></a>
<span class="sourceLineNo">118</span> /** Returns the next key in the index (the key to seek to the next block) */<a name="line.118"></a>
<span class="sourceLineNo">119</span> Cell getNextIndexedKey();<a name="line.119"></a>
<span class="sourceLineNo">120</span><a name="line.120"></a>
<span class="sourceLineNo">121</span> /**<a name="line.121"></a>
<span class="sourceLineNo">122</span> * Close this HFile scanner and do necessary cleanup.<a name="line.122"></a>
<span class="sourceLineNo">123</span> */<a name="line.123"></a>
<span class="sourceLineNo">124</span> @Override<a name="line.124"></a>
<span class="sourceLineNo">125</span> void close();<a name="line.125"></a>
<span class="sourceLineNo">126</span><a name="line.126"></a>
<span class="sourceLineNo">127</span> /**<a name="line.127"></a>
<span class="sourceLineNo">128</span> * Record the size of the current block in bytes, passing as an argument to the blockSizeConsumer.<a name="line.128"></a>
<span class="sourceLineNo">129</span> * Implementations should ensure that blockSizeConsumer is only called once per block.<a name="line.129"></a>
<span class="sourceLineNo">130</span> * @param blockSizeConsumer to be called with block size in bytes, once per block.<a name="line.130"></a>
<span class="sourceLineNo">131</span> */<a name="line.131"></a>
<span class="sourceLineNo">132</span> void recordBlockSize(IntConsumer blockSizeConsumer);<a name="line.132"></a>
<span class="sourceLineNo">133</span>}<a name="line.133"></a>
</pre>
</div>
</body>
</html>