⭐ 欢迎来到虫虫下载站! | 📦 资源下载 📁 资源专辑 ℹ️ 关于我们
⭐ 虫虫下载站

📄 adaptiverevisitfrontier.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
💻 HTML
📖 第 1 页 / 共 5 页
字号:
<a name="262" href="#262">262</a>         loadSeeds();<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>     * Create a UriUniqFilter that will serve as record </em><a name="267" href="#267">267</a> <em>     * of already seen URIs.</em><a name="268" href="#268">268</a> <em>     *</em><a name="269" href="#269">269</a> <em>     * @return A UURISet that will serve as a record of already seen URIs</em><a name="270" href="#270">270</a> <em>     * @throws IOException</em><a name="271" href="#271">271</a> <em>     */</em><a name="272" href="#272">272</a>     <strong>protected</strong> <a href="../../../../org/archive/crawler/datamodel/UriUniqFilter.html">UriUniqFilter</a> createAlreadyIncluded() throws IOException {<a name="273" href="#273">273</a>         <a href="../../../../org/archive/crawler/datamodel/UriUniqFilter.html">UriUniqFilter</a> uuf = <strong>new</strong> <a href="../../../../org/archive/crawler/util/BdbUriUniqFilter.html">BdbUriUniqFilter</a>(<a name="274" href="#274">274</a>                 <strong>this</strong>.controller.getBdbEnvironment());<a name="275" href="#275">275</a>         uuf.setDestination(<strong>this</strong>);<a name="276" href="#276">276</a>         <strong>return</strong> uuf;<a name="277" href="#277">277</a>     }<a name="278" href="#278">278</a>     <a name="279" href="#279">279</a>     <em>/**<em>*</em></em><a name="280" href="#280">280</a> <em>     * Loads the seeds</em><a name="281" href="#281">281</a> <em>     * &lt;p></em><a name="282" href="#282">282</a> <em>     * This method is called by initialize() and kickUpdate()</em><a name="283" href="#283">283</a> <em>     */</em><a name="284" href="#284">284</a>     <strong>public</strong> <strong>void</strong> loadSeeds() {<a name="285" href="#285">285</a>         Writer ignoredWriter = <strong>new</strong> StringWriter();<a name="286" href="#286">286</a>         <em class="comment">// Get the seeds to refresh.</em><a name="287" href="#287">287</a>         Iterator iter = <strong>this</strong>.controller.getScope().seedsIterator(ignoredWriter);<a name="288" href="#288">288</a>         <strong>while</strong> (iter.hasNext()) {<a name="289" href="#289">289</a>             <a href="../../../../org/archive/crawler/datamodel/CandidateURI.html">CandidateURI</a> caUri =<a name="290" href="#290">290</a>                 CandidateURI.createSeedCandidateURI((UURI)iter.next());<a name="291" href="#291">291</a>             caUri.setSchedulingDirective(CandidateURI.MEDIUM);<a name="292" href="#292">292</a>             schedule(caUri);<a name="293" href="#293">293</a>         }<a name="294" href="#294">294</a>         batchFlush();<a name="295" href="#295">295</a>         <em class="comment">// save ignored items (if any) where they can be consulted later</em><a name="296" href="#296">296</a>         AbstractFrontier.saveIgnoredItems(<a name="297" href="#297">297</a>                 ignoredWriter.toString(), <a name="298" href="#298">298</a>                 controller.getDisk());<a name="299" href="#299">299</a>     }<a name="300" href="#300">300</a>     <a name="301" href="#301">301</a>     <strong>public</strong> String getClassKey(<a href="../../../../org/archive/crawler/datamodel/CandidateURI.html">CandidateURI</a> cauri) {<a name="302" href="#302">302</a>         String queueKey = (String)getUncheckedAttribute(cauri,<a name="303" href="#303">303</a>                 ATTR_FORCE_QUEUE);<a name="304" href="#304">304</a>             <strong>if</strong> (<span class="string">""</span>.equals(queueKey)) {<a name="305" href="#305">305</a>                 <em class="comment">// Typical case, barring overrides</em><a name="306" href="#306">306</a>                 queueKey =<a name="307" href="#307">307</a>                     queueAssignmentPolicy.getClassKey(controller, cauri);<a name="308" href="#308">308</a>                 <em class="comment">// The queueAssignmentPolicy is always based on Hostnames</em><a name="309" href="#309">309</a>                 <em class="comment">// We may need to remove any www[0-9]{0,}\. prefixes from the</em><a name="310" href="#310">310</a>                 <em class="comment">// hostnames</em><a name="311" href="#311">311</a>                 <strong>if</strong>(((Boolean)getUncheckedAttribute(<a name="312" href="#312">312</a>                         cauri,ATTR_QUEUE_IGNORE_WWW)).booleanValue()){<a name="313" href="#313">313</a>                     queueKey = queueKey.replaceAll(<span class="string">"^www[0-9]{0,}&#47;&#47;."</span>,<span class="string">""</span>);<a name="314" href="#314">314</a>                 }<a name="315" href="#315">315</a>             }<a name="316" href="#316">316</a>             <strong>return</strong> queueKey;<a name="317" href="#317">317</a>     }<a name="318" href="#318">318</a> <a name="319" href="#319">319</a>     <em>/**<em>*</em></em><a name="320" href="#320">320</a> <em>     * Canonicalize passed uuri. Its would be sweeter if this canonicalize</em><a name="321" href="#321">321</a> <em>     * function was encapsulated by that which it canonicalizes but because</em><a name="322" href="#322">322</a> <em>     * settings change with context -- i.e. there may be overrides in operation</em><a name="323" href="#323">323</a> <em>     * for a particular URI -- its not so easy; Each CandidateURI would need a</em><a name="324" href="#324">324</a> <em>     * reference to the settings system. That's awkward to pass in.</em><a name="325" href="#325">325</a> <em>     * </em><a name="326" href="#326">326</a> <em>     * @param uuri Candidate URI to canonicalize.</em><a name="327" href="#327">327</a> <em>     * @return Canonicalized version of passed &lt;code>uuri&lt;/code>.</em><a name="328" href="#328">328</a> <em>     */</em><a name="329" href="#329">329</a>     <strong>protected</strong> String canonicalize(<a href="../../../../org/archive/net/UURI.html">UURI</a> uuri) {<a name="330" href="#330">330</a>         <strong>return</strong> Canonicalizer.canonicalize(uuri, <strong>this</strong>.controller.getOrder());<a name="331" href="#331">331</a>     }<a name="332" href="#332">332</a> <a name="333" href="#333">333</a>     <em>/**<em>*</em></em><a name="334" href="#334">334</a> <em>     * Canonicalize passed CandidateURI. This method differs from</em><a name="335" href="#335">335</a> <em>     * {@link #canonicalize(UURI)} in that it takes a look at</em><a name="336" href="#336">336</a> <em>     * the CandidateURI context possibly overriding any canonicalization effect if</em><a name="337" href="#337">337</a> <em>     * it could make us miss content. If canonicalization produces an URL that</em><a name="338" href="#338">338</a> <em>     * was 'alreadyseen', but the entry in the 'alreadyseen' database did</em><a name="339" href="#339">339</a> <em>     * nothing but redirect to the current URL, we won't get the current URL;</em><a name="340" href="#340">340</a> <em>     * we'll think we've already see it. Examples would be archive.org</em><a name="341" href="#341">341</a> <em>     * redirecting to www.archive.org or the inverse, www.netarkivet.net</em><a name="342" href="#342">342</a> <em>     * redirecting to netarkivet.net (assuming stripWWW rule enabled).</em><a name="343" href="#343">343</a> <em>     * &lt;p>Note, this method under circumstance sets the forceFetch flag.</em><a name="344" href="#344">344</a> <em>     * </em><a name="345" href="#345">345</a> <em>     * @param cauri CandidateURI to examine.</em><a name="346" href="#346">346</a> <em>     * @return Canonicalized &lt;code>cacuri&lt;/code>.</em><a name="347" href="#347">347</a> <em>     */</em><a name="348" href="#348">348</a>     <strong>protected</strong> String canonicalize(<a href="../../../../org/archive/crawler/datamodel/CandidateURI.html">CandidateURI</a> cauri) {<a name="349" href="#349">349</a>         String canon = canonicalize(cauri.getUURI());<a name="350" href="#350">350</a>         <strong>if</strong> (cauri.isLocation()) {<a name="351" href="#351">351</a>             <em class="comment">// If the via is not the same as where we're being redirected (i.e.</em><a name="352" href="#352">352</a>             <em class="comment">// we're not being redirected back to the same page, AND the</em><a name="353" href="#353">353</a>             <em class="comment">// canonicalization of the via is equal to the the current cauri, </em><a name="354" href="#354">354</a>             <em class="comment">// THEN forcefetch (Forcefetch so no chance of our not crawling</em><a name="355" href="#355">355</a>             <em class="comment">// content because alreadyseen check things its seen the url before.</em><a name="356" href="#356">356</a>             <em class="comment">// An example of an URL that redirects to itself is:</em><a name="357" href="#357">357</a>             <em class="comment">// http://bridalelegance.com/images/buttons3/tuxedos-off.gif.</em><a name="358" href="#358">358</a>             <em class="comment">// An example of an URL whose canonicalization equals its via's</em><a name="359" href="#359">359</a>             <em class="comment">// canonicalization, and we want to fetch content at the</em><a name="360" href="#360">360</a>             <em class="comment">// redirection (i.e. need to set forcefetch), is netarkivet.dk.</em><a name="361" href="#361">361</a>             <strong>if</strong> (!cauri.toString().equals(cauri.getVia().toString()) &amp;&amp;<a name="362" href="#362">362</a>                     canonicalize(cauri.getVia()).equals(canon)) {<a name="363" href="#363">363</a>                 cauri.setForceFetch(<strong>true</strong>);<a name="364" href="#364">364</a>             }<a name="365" href="#365">365</a>         }<a name="366" href="#366">366</a>         <strong>return</strong> canon;<a name="367" href="#367">367</a>     }<a name="368" href="#368">368</a> <a name="369" href="#369">369</a>     <em>/**<em>*</em></em><a name="370" href="#370">370</a> <em>     * </em><a name="371" href="#371">371</a> <em>     * @param caUri The URI to schedule.</em><a name="372" href="#372">372</a> <em>     */</em><a name="373" href="#373">373</a>     <strong>protected</strong> <strong>void</strong> innerSchedule(<a href="../../../../org/archive/crawler/datamodel/CandidateURI.html">CandidateURI</a> caUri) {<a name="374" href="#374">374</a>         <a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi;<a name="375" href="#375">375</a>         <strong>if</strong>(caUri instanceof <a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a>) {<a name="376" href="#376">376</a>             curi = (CrawlURI) caUri;<a name="377" href="#377">377</a>         } <strong>else</strong> {<a name="378" href="#378">378</a>             curi = CrawlURI.from(caUri,System.currentTimeMillis());<a name="379" href="#379">379</a>             curi.putLong(A_TIME_OF_NEXT_PROCESSING,<a name="380" href="#380">380</a>                 System.currentTimeMillis());<a name="381" href="#381">381</a>             <em class="comment">// New CrawlURIs get 'current time' as the time of next processing.</em><a name="382" href="#382">382</a>         }<a name="383" href="#383">383</a>         <a name="384" href="#384">384</a>         <strong>if</strong>(curi.getClassKey() == <strong>null</strong>){<a name="385" href="#385">385</a>             curi.setClassKey(getClassKey(curi));<a name="386" href="#386">386</a>         }<a name="387" href="#387">387</a> <a name="388" href="#388">388</a>         <strong>if</strong>(curi.isSeed() &amp;&amp; curi.getVia() != <strong>null</strong><a name="389" href="#389">389</a>                 &amp;&amp; curi.flattenVia().length() > 0) {<a name="390" href="#390">390</a>             <em class="comment">// The only way a seed can have a non-empty via is if it is the</em><a name="391" href="#391">391</a>             <em class="comment">// result of a seed redirect.  Add it to the seeds list.</em><a name="392" href="#392">392</a>             <em class="comment">//</em><a name="393" href="#393">393</a>             <em class="comment">// This is a feature.  This is handling for case where a seed</em><a name="394" href="#394">394</a>             <em class="comment">// gets immediately redirected to another page.  What we're doing</em><a name="395" href="#395">395</a>             <em class="comment">// is treating the immediate redirect target as a seed.</em><a name="396" href="#396">396</a>             <strong>this</strong>.controller.getScope().addSeed(curi);

⌨️ 快捷键说明

复制代码 Ctrl + C
搜索代码 Ctrl + F
全屏模式 F11
切换主题 Ctrl + Shift + D
显示快捷键 ?
增大字号 Ctrl + =
减小字号 Ctrl + -