httpmidfetchunchangedfilter.html
来自「网络爬虫开源代码」· HTML 代码 · 共 204 行 · 第 1/2 页
HTML
204 行
<a name="94" href="#94">94</a> <a name="95" href="#95">95</a> <strong>protected</strong> <strong>boolean</strong> innerAccepts(Object o) {<a name="96" href="#96">96</a> <em class="comment">// Return FALSE when the document has NOT changed!</em><a name="97" href="#97">97</a> <em class="comment">// Return TRUE if the document has changed or we can't tell</em><a name="98" href="#98">98</a> <strong>if</strong>(o instanceof <a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> == false){<a name="99" href="#99">99</a> <em class="comment">// Only handles CrawlURIs</em><a name="100" href="#100">100</a> <strong>if</strong> (logger.isLoggable(Level.INFO)) {<a name="101" href="#101">101</a> logger.info(<span class="string">"Error: Object passed for evaluation was not a "</span> +<a name="102" href="#102">102</a> <span class="string">"CrawlURI. "</span> + o.toString());<a name="103" href="#103">103</a> }<a name="104" href="#104">104</a> <strong>return</strong> <strong>true</strong>;<a name="105" href="#105">105</a> }<a name="106" href="#106">106</a> <a name="107" href="#107">107</a> <a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi = (CrawlURI)o;<a name="108" href="#108">108</a> <a name="109" href="#109">109</a> <strong>if</strong> (curi.isHttpTransaction() == false) {<a name="110" href="#110">110</a> <em class="comment">// Only handles HTTP</em><a name="111" href="#111">111</a> <strong>if</strong> (logger.isLoggable(Level.INFO)) {<a name="112" href="#112">112</a> logger.info(<span class="string">"Error: Non HTTP CrawlURI was passed for evalution. "</span><a name="113" href="#113">113</a> + curi.toString());<a name="114" href="#114">114</a> }<a name="115" href="#115">115</a> <strong>return</strong> <strong>true</strong>;<a name="116" href="#116">116</a> }<a name="117" href="#117">117</a> <a name="118" href="#118">118</a> <strong>if</strong>(curi.containsKey(A_HTTP_TRANSACTION) == false){<a name="119" href="#119">119</a> <em class="comment">// Missing header info, can't do anything.</em><a name="120" href="#120">120</a> <strong>if</strong> (logger.isLoggable(Level.INFO)) {<a name="121" href="#121">121</a> logger.info(<span class="string">"Error: Missing HttpMethod object in CrawlURI. "</span><a name="122" href="#122">122</a> + curi.toString());<a name="123" href="#123">123</a> }<a name="124" href="#124">124</a> <strong>return</strong> <strong>true</strong>;<a name="125" href="#125">125</a> }<a name="126" href="#126">126</a> <a name="127" href="#127">127</a> <em class="comment">// Intially assume header info is missing</em><a name="128" href="#128">128</a> <strong>int</strong> datestamp = HEADER_PREDICTS_MISSING;<a name="129" href="#129">129</a> <strong>int</strong> etag = HEADER_PREDICTS_MISSING;<a name="130" href="#130">130</a> HttpMethod method = (HttpMethod)curi.getObject(A_HTTP_TRANSACTION);<a name="131" href="#131">131</a> <a name="132" href="#132">132</a> <em class="comment">// Compare datestamps (last-modified)</em><a name="133" href="#133">133</a> String newDatestamp = <strong>null</strong>;<a name="134" href="#134">134</a> <strong>if</strong> (method.getResponseHeader(<span class="string">"last-modified"</span>) != <strong>null</strong>) {<a name="135" href="#135">135</a> newDatestamp = method.getResponseHeader(<span class="string">"last-modified"</span>).getValue();<a name="136" href="#136">136</a> }<a name="137" href="#137">137</a> <a name="138" href="#138">138</a> <strong>if</strong> (newDatestamp != <strong>null</strong> && newDatestamp.length() > 0) {<a name="139" href="#139">139</a> datestamp = HEADER_PREDICTS_CHANGED; <em class="comment">// Not missing, assume change</em><a name="140" href="#140">140</a> <strong>if</strong> (curi.containsKey(A_LAST_DATESTAMP)) {<a name="141" href="#141">141</a> <strong>if</strong> (newDatestamp.equals(curi.getString(A_LAST_DATESTAMP))) {<a name="142" href="#142">142</a> <em class="comment">// Both new and old are present and equal, datestamp</em><a name="143" href="#143">143</a> <em class="comment">// predicts no change</em><a name="144" href="#144">144</a> datestamp = HEADER_PREDICTS_UNCHANGED;<a name="145" href="#145">145</a> }<a name="146" href="#146">146</a> }<a name="147" href="#147">147</a> curi.putString(A_LAST_DATESTAMP, newDatestamp);<a name="148" href="#148">148</a> }<a name="149" href="#149">149</a> <a name="150" href="#150">150</a> <em class="comment">// Compare ETags</em><a name="151" href="#151">151</a> String newETag = <strong>null</strong>;<a name="152" href="#152">152</a> <strong>if</strong>(method.getResponseHeader(<span class="string">"last-etag"</span>) != <strong>null</strong>){<a name="153" href="#153">153</a> newETag = method.getResponseHeader(<span class="string">"last-etag"</span>).getValue();<a name="154" href="#154">154</a> }<a name="155" href="#155">155</a> <a name="156" href="#156">156</a> <strong>if</strong>(newETag != <strong>null</strong> && newETag.length() > 0){<a name="157" href="#157">157</a> etag = HEADER_PREDICTS_CHANGED; <em class="comment">// Not missing, assume change</em><a name="158" href="#158">158</a> <strong>if</strong>(curi.containsKey(A_LAST_ETAG)){<a name="159" href="#159">159</a> <strong>if</strong>(newETag.equals(curi.getString(A_LAST_ETAG))){<a name="160" href="#160">160</a> <em class="comment">// Both new and old are present and equal, etag </em><a name="161" href="#161">161</a> <em class="comment">// predicts no change</em><a name="162" href="#162">162</a> etag = HEADER_PREDICTS_UNCHANGED;<a name="163" href="#163">163</a> }<a name="164" href="#164">164</a> }<a name="165" href="#165">165</a> curi.putString(A_LAST_ETAG, newETag);<a name="166" href="#166">166</a> }<a name="167" href="#167">167</a> <a name="168" href="#168">168</a> <em class="comment">// If both are present, predict no change only if both agree</em><a name="169" href="#169">169</a> <strong>if</strong> (datestamp == HEADER_PREDICTS_UNCHANGED<a name="170" href="#170">170</a> && etag == HEADER_PREDICTS_UNCHANGED) {<a name="171" href="#171">171</a> <em class="comment">// Have both and they agree, no change</em><a name="172" href="#172">172</a> curi.putInt(A_CONTENT_STATE_KEY, CONTENT_UNCHANGED);<a name="173" href="#173">173</a> <strong>return</strong> false;<a name="174" href="#174">174</a> }<a name="175" href="#175">175</a> <em class="comment">// If one or the other is missing, trust the one that is present</em><a name="176" href="#176">176</a> <strong>if</strong> (datestamp == HEADER_PREDICTS_MISSING<a name="177" href="#177">177</a> && etag == HEADER_PREDICTS_UNCHANGED) {<a name="178" href="#178">178</a> <em class="comment">// Only have etag, and it predicts no change</em><a name="179" href="#179">179</a> curi.putInt(A_CONTENT_STATE_KEY, CONTENT_UNCHANGED);<a name="180" href="#180">180</a> <strong>return</strong> false;<a name="181" href="#181">181</a> }<a name="182" href="#182">182</a> <strong>if</strong> (datestamp == HEADER_PREDICTS_UNCHANGED<a name="183" href="#183">183</a> && etag == HEADER_PREDICTS_MISSING) {<a name="184" href="#184">184</a> <em class="comment">// Only have last-modified, and it predicts no change</em><a name="185" href="#185">185</a> curi.putInt(A_CONTENT_STATE_KEY, CONTENT_UNCHANGED);<a name="186" href="#186">186</a> <strong>return</strong> false;<a name="187" href="#187">187</a> }<a name="188" href="#188">188</a> <strong>return</strong> <strong>true</strong>; <em class="comment">// Default, assume change. </em><a name="189" href="#189">189</a> }<a name="190" href="#190">190</a> }</pre><hr/><div id="footer">This page was automatically generated by <a href="http://maven.apache.org/">Maven</a></div></body></html>
⌨️ 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?