bytereplaycharsequence.html
来自「网络爬虫开源代码」· HTML 代码 · 共 377 行 · 第 1/3 页
HTML
377 行
<a name="118" href="#118">118</a> <em> * Random access to the backing file.</em><a name="119" href="#119">119</a> <em> */</em><a name="120" href="#120">120</a> <strong>private</strong> RandomAccessFile raFile;<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> * Offset into prefix buffer at which content beings.</em><a name="124" href="#124">124</a> <em> */</em><a name="125" href="#125">125</a> <strong>private</strong> <strong>int</strong> contentOffset;<a name="126" href="#126">126</a> <a name="127" href="#127">127</a> <em>/**<em>*</em></em><a name="128" href="#128">128</a> <em> * 8-bit encoding used reading single bytes from buffer and</em><a name="129" href="#129">129</a> <em> * stream.</em><a name="130" href="#130">130</a> <em> */</em><a name="131" href="#131">131</a> <strong>private</strong> <strong>static</strong> <strong>final</strong> String DEFAULT_SINGLE_BYTE_ENCODING =<a name="132" href="#132">132</a> <span class="string">"ISO-8859-1"</span>;<a name="133" href="#133">133</a> <a name="134" href="#134">134</a> <a name="135" href="#135">135</a> <em>/**<em>*</em></em><a name="136" href="#136">136</a> <em> * Constructor.</em><a name="137" href="#137">137</a> <em> *</em><a name="138" href="#138">138</a> <em> * @param buffer In-memory buffer of recordings prefix. We read from</em><a name="139" href="#139">139</a> <em> * here first and will only go to the backing file if <code>size</code></em><a name="140" href="#140">140</a> <em> * requested is greater than <code>buffer.length</code>.</em><a name="141" href="#141">141</a> <em> * @param size Total size of stream to replay in bytes. Used to find</em><a name="142" href="#142">142</a> <em> * EOS. This is total length of content including HTTP headers if</em><a name="143" href="#143">143</a> <em> * present.</em><a name="144" href="#144">144</a> <em> * @param responseBodyStart Where the response body starts in bytes.</em><a name="145" href="#145">145</a> <em> * Used to skip over the HTTP headers if present.</em><a name="146" href="#146">146</a> <em> * @param backingFilename Path to backing file with content in excess of</em><a name="147" href="#147">147</a> <em> * whats in <code>buffer</code>.</em><a name="148" href="#148">148</a> <em> *</em><a name="149" href="#149">149</a> <em> * @throws IOException</em><a name="150" href="#150">150</a> <em> */</em><a name="151" href="#151">151</a> <strong>public</strong> <a href="../../../org/archive/io/ByteReplayCharSequence.html">ByteReplayCharSequence</a>(byte[] buffer, <strong>long</strong> size,<a name="152" href="#152">152</a> <strong>long</strong> responseBodyStart, String backingFilename)<a name="153" href="#153">153</a> throws IOException {<a name="154" href="#154">154</a> <a name="155" href="#155">155</a> <strong>this</strong>.length = (<strong>int</strong>)(size - responseBodyStart);<a name="156" href="#156">156</a> <strong>this</strong>.absoluteLength = (<strong>int</strong>)size;<a name="157" href="#157">157</a> <strong>this</strong>.prefixBuffer = buffer;<a name="158" href="#158">158</a> <strong>this</strong>.contentOffset = (<strong>int</strong>)responseBodyStart;<a name="159" href="#159">159</a> <a name="160" href="#160">160</a> <em class="comment">// If amount to read is > than what is in our prefix buffer, then</em><a name="161" href="#161">161</a> <em class="comment">// open the backing file.</em><a name="162" href="#162">162</a> <strong>if</strong> (size > buffer.length) {<a name="163" href="#163">163</a> <strong>this</strong>.backingFilename = backingFilename;<a name="164" href="#164">164</a> <strong>this</strong>.raFile = <strong>new</strong> RandomAccessFile(backingFilename, <span class="string">"r"</span>);<a name="165" href="#165">165</a> <strong>this</strong>.wraparoundBuffer = <strong>new</strong> byte[<strong>this</strong>.prefixBuffer.length];<a name="166" href="#166">166</a> <strong>this</strong>.wrapOrigin = <strong>this</strong>.prefixBuffer.length;<a name="167" href="#167">167</a> <strong>this</strong>.wrapOffset = 0;<a name="168" href="#168">168</a> loadBuffer();<a name="169" href="#169">169</a> }<a name="170" href="#170">170</a> }<a name="171" href="#171">171</a> <a name="172" href="#172">172</a> <em>/**<em>*</em></em><a name="173" href="#173">173</a> <em> * @return Length of characters in stream to replay. Starts counting</em><a name="174" href="#174">174</a> <em> * at the HTTP header/body boundary.</em><a name="175" href="#175">175</a> <em> */</em><a name="176" href="#176">176</a> <strong>public</strong> <strong>int</strong> length() {<a name="177" href="#177">177</a> <strong>return</strong> <strong>this</strong>.length;<a name="178" href="#178">178</a> }<a name="179" href="#179">179</a> <a name="180" href="#180">180</a> <em>/**<em>*</em></em><a name="181" href="#181">181</a> <em> * Get character at passed absolute position.</em><a name="182" href="#182">182</a> <em> *</em><a name="183" href="#183">183</a> <em> * Called by {@link #charAt(int)} which has a relative index into the</em><a name="184" href="#184">184</a> <em> * content, one that doesn't account for HTTP header if present.</em><a name="185" href="#185">185</a> <em> *</em><a name="186" href="#186">186</a> <em> * @param index Index into content adjusted to accomodate initial offset</em><a name="187" href="#187">187</a> <em> * to get us past the HTTP header if present (i.e.</em><a name="188" href="#188">188</a> <em> * {@link #contentOffset}).</em><a name="189" href="#189">189</a> <em> *</em><a name="190" href="#190">190</a> <em> * @return Characater at offset <code>index</code>.</em><a name="191" href="#191">191</a> <em> */</em><a name="192" href="#192">192</a> <strong>public</strong> <strong>char</strong> <strong>char</strong>At(<strong>int</strong> index) {<a name="193" href="#193">193</a> <strong>int</strong> c = -1;<a name="194" href="#194">194</a> <em class="comment">// Add to index start-of-content offset to get us over HTTP header</em><a name="195" href="#195">195</a> <em class="comment">// if present.</em><a name="196" href="#196">196</a> index += <strong>this</strong>.contentOffset;<a name="197" href="#197">197</a> <strong>if</strong> (index < <strong>this</strong>.prefixBuffer.length) {<a name="198" href="#198">198</a> <em class="comment">// If index is into our prefix buffer.</em><a name="199" href="#199">199</a> c = <strong>this</strong>.prefixBuffer[index];<a name="200" href="#200">200</a> } <strong>else</strong> <strong>if</strong> (index >= <strong>this</strong>.wrapOrigin &&<a name="201" href="#201">201</a> (index - <strong>this</strong>.wrapOrigin) < <strong>this</strong>.wraparoundBuffer.length) {<a name="202" href="#202">202</a> <em class="comment">// If index is into our buffer window on underlying backing file.</em><a name="203" href="#203">203</a> c = <strong>this</strong>.wraparoundBuffer[<a name="204" href="#204">204</a> ((index - <strong>this</strong>.wrapOrigin) + <strong>this</strong>.wrapOffset) %<a name="205" href="#205">205</a> <strong>this</strong>.wraparoundBuffer.length];<a name="206" href="#206">206</a> } <strong>else</strong> {<a name="207" href="#207">207</a> <em class="comment">// Index is outside of both prefix buffer and our buffer window</em><a name="208" href="#208">208</a> <em class="comment">// onto the underlying backing file. Fix the buffer window</em><a name="209" href="#209">209</a> <em class="comment">// location.</em><a name="210" href="#210">210</a> c = faultCharAt(index);<a name="211" href="#211">211</a> }<a name="212" href="#212">212</a> <em class="comment">// Stream is treated as single byte. Make sure characters returned</em><a name="213" href="#213">213</a> <em class="comment">// are not negative.</em><a name="214" href="#214">214</a> <strong>return</strong> (<strong>char</strong>)(c & 0xff);<a name="215" href="#215">215</a> }<a name="216" href="#216">216</a> <a name="217" href="#217">217</a> <em>/**<em>*</em></em><a name="218" href="#218">218</a> <em> * Get a character that's outside the current buffers.</em><a name="219" href="#219">219</a> <em> *</em><a name="220" href="#220">220</a> <em> * will cause the wraparoundBuffer to be changed to</em><a name="221" href="#221">221</a> <em> * cover a region including the index</em><a name="222" href="#222">222</a> <em> *</em><a name="223" href="#223">223</a> <em> * if index is higher than the highest index in the</em><a name="224" href="#224">224</a> <em> * wraparound buffer, buffer is moved forward such</em><a name="225" href="#225">225</a> <em> * that requested char is last item in buffer</em><a name="226" href="#226">226</a> <em> *</em><a name="227" href="#227">227</a> <em> * if index is lower than lowest index in the</em><a name="228" href="#228">228</a> <em> * wraparound buffer, buffet is reset centered around</em><a name="229" href="#229">229</a> <em> * index</em><a name="230" href="#230">230</a> <em> *</em><a name="231" href="#231">231</a> <em> * @param index Index of character to fetch.</em><a name="232" href="#232">232</a> <em> * @return A character that's outside the current buffers</em><a name="233" href="#233">233</a> <em> */</em><a name="234" href="#234">234</a> <strong>private</strong> <strong>int</strong> faultCharAt(<strong>int</strong> index) {<a name="235" href="#235">235</a> <strong>if</strong>(Thread.interrupted()) {<a name="236" href="#236">236</a> <strong>throw</strong> <strong>new</strong> RuntimeException(<span class="string">"thread interrupted"</span>);<a name="237" href="#237">237</a> }<a name="238" href="#238">238</a> <strong>if</strong>(index >= <strong>this</strong>.wrapOrigin + <strong>this</strong>.wraparoundBuffer.length) {<a name="239" href="#239">239</a> <em class="comment">// Moving forward</em><a name="240" href="#240">240</a> <strong>while</strong> (index >= <strong>this</strong>.wrapOrigin + <strong>this</strong>.wraparoundBuffer.length)<a name="241" href="#241">241</a> {<a name="242" href="#242">242</a> <em class="comment">// TODO optimize this</em><a name="243" href="#243">243</a> advanceBuffer();
⌨️ 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?