preconditionenforcer.html

用JAVA编写的,在做实验的时候留下来的,本来想删的,但是传上来,大家分享吧

<a name="162" href="#162">162</a>         <strong>if</strong> (isRobotsExpired(curi)) {
<a name="163" href="#163">163</a>         	<em class="comment">// Need to get robots</em>
<a name="164" href="#164">164</a>             <strong>if</strong> (logger.isLoggable(Level.FINE)) {
<a name="165" href="#165">165</a>                 logger.fine( <span class="string">"No valid robots for "</span> +
<a name="166" href="#166">166</a>                     getController().getServerCache().getServerFor(curi) +
<a name="167" href="#167">167</a>                     <span class="string">"; deferring "</span> + curi);
<a name="168" href="#168">168</a>             }
<a name="169" href="#169">169</a> 
<a name="170" href="#170">170</a>             <em class="comment">// Robots expired - should be refetched even though its already</em>
<a name="171" href="#171">171</a>             <em class="comment">// crawled.</em>
<a name="172" href="#172">172</a>             <strong>try</strong> {
<a name="173" href="#173">173</a>                 String prereq = curi.getUURI().resolve(<span class="string">"/robots.txt"</span>).toString();
<a name="174" href="#174">174</a>                 curi.markPrerequisite(prereq,
<a name="175" href="#175">175</a>                     getController().getPostprocessorChain());
<a name="176" href="#176">176</a>             }
<a name="177" href="#177">177</a>             <strong>catch</strong> (URIException e1) {
<a name="178" href="#178">178</a>                 logger.severe(<span class="string">"Failed resolve using "</span> + curi);
<a name="179" href="#179">179</a>                 <strong>throw</strong> <strong>new</strong> RuntimeException(e1); <em class="comment">// shouldn't ever happen</em>
<a name="180" href="#180">180</a>             }
<a name="181" href="#181">181</a>             <strong>return</strong> <strong>true</strong>;
<a name="182" href="#182">182</a>         }
<a name="183" href="#183">183</a>         <em class="comment">// test against robots.txt if available</em>
<a name="184" href="#184">184</a>         <a href="../../../../org/archive/crawler/datamodel/CrawlServer.html">CrawlServer</a> cs = getController().getServerCache().getServerFor(curi);
<a name="185" href="#185">185</a>         <strong>if</strong>(cs.isValidRobots()){
<a name="186" href="#186">186</a>             String ua = getController().getOrder().getUserAgent(curi);
<a name="187" href="#187">187</a>             <strong>if</strong>(cs.getRobots().disallows(curi, ua)) {
<a name="188" href="#188">188</a>                 <strong>if</strong>(((Boolean)getUncheckedAttribute(curi,ATTR_CALCULATE_ROBOTS_ONLY)).booleanValue() == <strong>true</strong>) {
<a name="189" href="#189">189</a>                     <em class="comment">// annotate URI as excluded, but continue to process normally</em>
<a name="190" href="#190">190</a>                     curi.addAnnotation(<span class="string">"robotExcluded"</span>);
<a name="191" href="#191">191</a>                     <strong>return</strong> false; 
<a name="192" href="#192">192</a>                 }
<a name="193" href="#193">193</a>                 <em class="comment">// mark as precluded; in FetchHTTP, this will</em>
<a name="194" href="#194">194</a>                 <em class="comment">// prevent fetching and cause a skip to the end</em>
<a name="195" href="#195">195</a>                 <em class="comment">// of processing (unless an intervening processor</em>
<a name="196" href="#196">196</a>                 <em class="comment">// overrules)</em>
<a name="197" href="#197">197</a>                 curi.setFetchStatus(S_ROBOTS_PRECLUDED);
<a name="198" href="#198">198</a>                 curi.putString(<span class="string">"error"</span>,<span class="string">"robots.txt exclusion"</span>);
<a name="199" href="#199">199</a>                 logger.fine(<span class="string">"robots.txt precluded "</span> + curi);
<a name="200" href="#200">200</a>                 <strong>return</strong> <strong>true</strong>;
<a name="201" href="#201">201</a>             }
<a name="202" href="#202">202</a>             <strong>return</strong> false;
<a name="203" href="#203">203</a>         }
<a name="204" href="#204">204</a>         <em class="comment">// No valid robots found => Attempt to get robots.txt failed</em>
<a name="205" href="#205">205</a>         curi.skipToProcessorChain(getController().getPostprocessorChain());
<a name="206" href="#206">206</a>         curi.setFetchStatus(S_ROBOTS_PREREQUISITE_FAILURE);
<a name="207" href="#207">207</a>         curi.putString(<span class="string">"error"</span>,<span class="string">"robots.txt prerequisite failed"</span>);
<a name="208" href="#208">208</a>         <strong>if</strong> (logger.isLoggable(Level.FINE)) {
<a name="209" href="#209">209</a>             logger.fine(<span class="string">"robots.txt prerequisite failed "</span> + curi);
<a name="210" href="#210">210</a>         }
<a name="211" href="#211">211</a>         <strong>return</strong> <strong>true</strong>;
<a name="212" href="#212">212</a>     }
<a name="213" href="#213">213</a> 
<a name="214" href="#214">214</a>     <em>/**<em>*</em></em>
<a name="215" href="#215">215</a> <em>     * @param curi CrawlURI whose dns prerequisite we're to check.</em>
<a name="216" href="#216">216</a> <em>     * @return true if no further processing in this module should occur</em>
<a name="217" href="#217">217</a> <em>     */</em>
<a name="218" href="#218">218</a>     <strong>private</strong> <strong>boolean</strong> considerDnsPreconditions(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi) {
<a name="219" href="#219">219</a>         <strong>if</strong>(curi.getUURI().getScheme().equals(<span class="string">"dns"</span>)){
<a name="220" href="#220">220</a>             <em class="comment">// DNS URIs never have a DNS precondition</em>
<a name="221" href="#221">221</a>             curi.setPrerequisite(<strong>true</strong>);
<a name="222" href="#222">222</a>             <strong>return</strong> false; 
<a name="223" href="#223">223</a>         }
<a name="224" href="#224">224</a>         
<a name="225" href="#225">225</a>         <a href="../../../../org/archive/crawler/datamodel/CrawlServer.html">CrawlServer</a> cs = getController().getServerCache().getServerFor(curi);
<a name="226" href="#226">226</a>         <strong>if</strong>(cs == <strong>null</strong>) {
<a name="227" href="#227">227</a>             curi.setFetchStatus(S_UNFETCHABLE_URI);
<a name="228" href="#228">228</a>             curi.skipToProcessorChain(getController().getPostprocessorChain());
<a name="229" href="#229">229</a>             <strong>return</strong> <strong>true</strong>;
<a name="230" href="#230">230</a>         }
<a name="231" href="#231">231</a> 
<a name="232" href="#232">232</a>         <em class="comment">// If we've done a dns lookup and it didn't resolve a host</em>
<a name="233" href="#233">233</a>         <em class="comment">// cancel further fetch-processing of this URI, because</em>
<a name="234" href="#234">234</a>         <em class="comment">// the domain is unresolvable</em>
<a name="235" href="#235">235</a>         <a href="../../../../org/archive/crawler/datamodel/CrawlHost.html">CrawlHost</a> ch = getController().getServerCache().getHostFor(curi);
<a name="236" href="#236">236</a>         <strong>if</strong> (ch == <strong>null</strong> || ch.hasBeenLookedUp() &amp;&amp; ch.getIP() == <strong>null</strong>) {
<a name="237" href="#237">237</a>             <strong>if</strong> (logger.isLoggable(Level.FINE)) {
<a name="238" href="#238">238</a>                 logger.fine( <span class="string">"no dns for "</span> + ch +
<a name="239" href="#239">239</a>                     <span class="string">" cancelling processing for CrawlURI "</span> + curi.toString());
<a name="240" href="#240">240</a>             }
<a name="241" href="#241">241</a>             curi.setFetchStatus(S_DOMAIN_PREREQUISITE_FAILURE);
<a name="242" href="#242">242</a>             curi.skipToProcessorChain(getController().getPostprocessorChain());
<a name="243" href="#243">243</a>             <strong>return</strong> <strong>true</strong>;
<a name="244" href="#244">244</a>         }
<a name="245" href="#245">245</a> 
<a name="246" href="#246">246</a>         <em class="comment">// If we haven't done a dns lookup  and this isn't a dns uri</em>
<a name="247" href="#247">247</a>         <em class="comment">// shoot that off and defer further processing</em>
<a name="248" href="#248">248</a>         <strong>if</strong> (isIpExpired(curi) &amp;&amp; !curi.getUURI().getScheme().equals(<span class="string">"dns"</span>)) {
<a name="249" href="#249">249</a>             logger.fine(<span class="string">"Deferring processing of CrawlURI "</span> + curi.toString()
<a name="250" href="#250">250</a>                 + <span class="string">" for dns lookup."</span>);
<a name="251" href="#251">251</a>             String preq = <span class="string">"dns:"</span> + ch.getHostName();
<a name="252" href="#252">252</a>             <strong>try</strong> {
<a name="253" href="#253">253</a>                 curi.markPrerequisite(preq,
<a name="254" href="#254">254</a>                     getController().getPostprocessorChain());
<a name="255" href="#255">255</a>             } <strong>catch</strong> (URIException e) {
<a name="256" href="#256">256</a>                 <strong>throw</strong> <strong>new</strong> RuntimeException(e); <em class="comment">// shouldn't ever happen</em>
<a name="257" href="#257">257</a>             }
<a name="258" href="#258">258</a>             <strong>return</strong> <strong>true</strong>;
<a name="259" href="#259">259</a>         }
<a name="260" href="#260">260</a>         
<a name="261" href="#261">261</a>         <em class="comment">// DNS preconditions OK</em>
<a name="262" href="#262">262</a>         <strong>return</strong> false;
<a name="263" href="#263">263</a>     }
<a name="264" href="#264">264</a> 
<a name="265" href="#265">265</a>     <em>/**<em>*</em></em>
<a name="266" href="#266">266</a> <em>     * Get the maximum time a dns-record is valid.</em>
<a name="267" href="#267">267</a> <em>     *</em>
<a name="268" href="#268">268</a> <em>     * @param curi the uri this time is valid for.</em>
<a name="269" href="#269">269</a> <em>     * @return the maximum time a dns-record is valid -- in seconds -- or</em>
<a name="270" href="#270">270</a> <em>     * negative if record's ttl should be used.</em>
<a name="271" href="#271">271</a> <em>     */</em>
<a name="272" href="#272">272</a>     <strong>public</strong> <strong>long</strong> getIPValidityDuration(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi) {
<a name="273" href="#273">273</a>         Integer d;
<a name="274" href="#274">274</a>         <strong>try</strong> {
<a name="275" href="#275">275</a>             d = (Integer)getAttribute(ATTR_IP_VALIDITY_DURATION, curi);
<a name="276" href="#276">276</a>         } <strong>catch</strong> (AttributeNotFoundException e) {
<a name="277" href="#277">277</a>             d = DEFAULT_IP_VALIDITY_DURATION;
<a name="278" href="#278">278</a>         }
<a name="279" href="#279">279</a> 
<a name="280" href="#280">280</a>         <strong>return</strong> d.longValue();
<a name="281" href="#281">281</a>     }
<a name="282" href="#282">282</a> 
<a name="283" href="#283">283</a>     <em>/**<em>* Return true if ip should be looked up.</em></em>
<a name="284" href="#284">284</a> <em>     *</em>
<a name="285" href="#285">285</a> <em>     * @param curi the URI to check.</em>
<a name="286" href="#286">286</a> <em>     * @return true if ip should be looked up.</em>
<a name="287" href="#287">287</a> <em>     */</em>
<a name="288" href="#288">288</a>     <strong>public</strong> <strong>boolean</strong> isIpExpired(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi) {
<a name="289" href="#289">289</a>         <a href="../../../../org/archive/crawler/datamodel/CrawlHost.html">CrawlHost</a> host = getController().getServerCache().getHostFor(curi);
<a name="290" href="#290">290</a>         <strong>if</strong> (!host.hasBeenLookedUp()) {
<a name="291" href="#291">291</a>             <em class="comment">// IP has not been looked up yet.</em>
<a name="292" href="#292">292</a>             <strong>return</strong> <strong>true</strong>;
<a name="293" href="#293">293</a>         }
<a name="294" href="#294">294</a> 
<a name="295" href="#295">295</a>         <strong>if</strong> (host.getIpTTL() == CrawlHost.IP_NEVER_EXPIRES) {
<a name="296" href="#296">296</a>             <em class="comment">// IP never expires (numeric IP)</em>
<a name="297" href="#297">297</a>             <strong>return</strong> false;
<a name="298" href="#298">298</a>         }
<a name="299" href="#299">299</a> 
<a name="300" href="#300">300</a>         <strong>long</strong> duration = getIPValidityDuration(curi);
<a name="301" href="#301">301</a>         <strong>if</strong> (duration == 0) {
<a name="302" href="#302">302</a>             <em class="comment">// Never expire ip if duration is null (set by user or more likely,</em>
<a name="303" href="#303">303</a>             <em class="comment">// set to zero in case where we tried in FetchDNS but failed).</em>
<a name="304" href="#304">304</a>             <strong>return</strong> false;
<a name="305" href="#305">305</a>         }
<a name="306" href="#306">306</a> 
<a name="307" href="#307">307</a>         <em class="comment">// catch old "default" -1 settings that are now problematic,</em>
<a name="308" href="#308">308</a>         <em class="comment">// convert to new minimum</em>
<a name="309" href="#309">309</a>         <strong>if</strong> (duration &lt;= 0) {
<a name="310" href="#310">310</a>             duration = DEFAULT_IP_VALIDITY_DURATION.intValue();
<a name="311" href="#311">311</a>         }
<a name="312" href="#312">312</a>         
<a name="313" href="#313">313</a>         <strong>long</strong> ttl = host.getIpTTL();
<a name="314" href="#314">314</a>         <strong>if</strong> (ttl > duration) {
<a name="315" href="#315">315</a>             <em class="comment">// Use the larger of the operator-set minimum duration </em>
<a name="316" href="#316">316</a>             <em class="comment">// or the DNS record TTL</em>
<a name="317" href="#317">317</a>             duration = ttl;
<a name="318" href="#318">318</a>         }
<a name="319" href="#319">319</a> 
<a name="320" href="#320">320</a>         <em class="comment">// Duration and ttl are in seconds.  Convert to millis.</em>
<a name="321" href="#321">321</a>         <strong>if</strong> (duration > 0) {
<a name="322" href="#322">322</a>             duration *= 1000;
<a name="323" href="#323">323</a>         }
<a name="324" href="#324">324</a> 
<a name="325" href="#325">325</a>         <strong>return</strong> (duration + host.getIpFetched()) &lt; System.currentTimeMillis();
<a name="326" href="#326">326</a>     }
<a name="327" href="#327">327</a> 
<a name="328" href="#328">328</a>     <em>/**<em>* Get the maximum time a robots.txt is valid.</em></em>
<a name="329" href="#329">329</a> <em>     *</em>
<a name="330" href="#330">330</a> <em>     * @param curi</em>
<a name="331" href="#331">331</a> <em>     * @return the time a robots.txt is valid in milliseconds.</em>