⭐ 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

📄 workqueuefrontier.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="421" href="#421">421</a> <em>     */</em><a name="422" href="#422">422</a>     <strong>public</strong> <strong>void</strong> receive(<a href="../../../../org/archive/crawler/datamodel/CandidateURI.html">CandidateURI</a> caUri) {<a name="423" href="#423">423</a>         <a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi = asCrawlUri(caUri);<a name="424" href="#424">424</a>         applySpecialHandling(curi);<a name="425" href="#425">425</a>         sendToQueue(curi);<a name="426" href="#426">426</a>         <em class="comment">// Update recovery log.</em><a name="427" href="#427">427</a>         doJournalAdded(curi);<a name="428" href="#428">428</a>     }<a name="429" href="#429">429</a> <a name="430" href="#430">430</a> 	<em class="comment">/*<em class="comment"> (non-Javadoc)</em></em><a name="431" href="#431">431</a> <em class="comment">	 * @see org.archive.crawler.frontier.AbstractFrontier#asCrawlUri(org.archive.crawler.datamodel.CandidateURI)</em><a name="432" href="#432">432</a> <em class="comment">	 */</em><a name="433" href="#433">433</a> 	<strong>protected</strong> <a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> asCrawlUri(<a href="../../../../org/archive/crawler/datamodel/CandidateURI.html">CandidateURI</a> caUri) {<a name="434" href="#434">434</a> 		CrawlURI curi = <strong>super</strong>.asCrawlUri(caUri);<a name="435" href="#435">435</a> 		<em class="comment">// force cost to be calculated, pre-insert</em><a name="436" href="#436">436</a> 		getCost(curi);<a name="437" href="#437">437</a> 		<strong>return</strong> curi;<a name="438" href="#438">438</a> 	}<a name="439" href="#439">439</a> 	<a name="440" href="#440">440</a>     <em>/**<em>*</em></em><a name="441" href="#441">441</a> <em>     * Send a CrawlURI to the appropriate subqueue.</em><a name="442" href="#442">442</a> <em>     * </em><a name="443" href="#443">443</a> <em>     * @param curi</em><a name="444" href="#444">444</a> <em>     */</em><a name="445" href="#445">445</a>     <strong>protected</strong> <strong>void</strong> sendToQueue(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi) {<a name="446" href="#446">446</a>         <a href="../../../../org/archive/crawler/frontier/WorkQueue.html">WorkQueue</a> wq = getQueueFor(curi);<a name="447" href="#447">447</a>         <strong>synchronized</strong> (wq) {<a name="448" href="#448">448</a>             wq.enqueue(<strong>this</strong>, curi);<a name="449" href="#449">449</a>             <strong>if</strong>(!wq.isRetired()) {<a name="450" href="#450">450</a>                 incrementQueuedUriCount();<a name="451" href="#451">451</a>             }<a name="452" href="#452">452</a>             <strong>if</strong>(!wq.isHeld()) {<a name="453" href="#453">453</a>                 wq.setHeld();<a name="454" href="#454">454</a>                 <strong>if</strong>(holdQueues() &amp;&amp; readyClassQueues.size()>=targetSizeForReadyQueues()) {<a name="455" href="#455">455</a>                     deactivateQueue(wq);<a name="456" href="#456">456</a>                 } <strong>else</strong> {<a name="457" href="#457">457</a>                     replenishSessionBalance(wq);<a name="458" href="#458">458</a>                     readyQueue(wq);<a name="459" href="#459">459</a>                 }<a name="460" href="#460">460</a>             }<a name="461" href="#461">461</a>             <a href="../../../../org/archive/crawler/frontier/WorkQueue.html">WorkQueue</a> laq = longestActiveQueue;<a name="462" href="#462">462</a>             <strong>if</strong>(!wq.isRetired()&amp;&amp;((laq==<strong>null</strong>) || wq.getCount() > laq.getCount())) {<a name="463" href="#463">463</a>                 longestActiveQueue = wq; <a name="464" href="#464">464</a>             }<a name="465" href="#465">465</a>         }<a name="466" href="#466">466</a>     }<a name="467" href="#467">467</a> <a name="468" href="#468">468</a>     <em>/**<em>*</em></em><a name="469" href="#469">469</a> <em>     * Whether queues should start inactive (only becoming active when needed</em><a name="470" href="#470">470</a> <em>     * to keep the crawler busy), or if queues should start out ready.</em><a name="471" href="#471">471</a> <em>     * </em><a name="472" href="#472">472</a> <em>     * @return true if new queues should held inactive</em><a name="473" href="#473">473</a> <em>     */</em><a name="474" href="#474">474</a>     <strong>private</strong> <strong>boolean</strong> holdQueues() {<a name="475" href="#475">475</a>         <strong>return</strong> ((Boolean) getUncheckedAttribute(<strong>null</strong>, ATTR_HOLD_QUEUES))<a name="476" href="#476">476</a>                 .booleanValue();<a name="477" href="#477">477</a>     }<a name="478" href="#478">478</a> <a name="479" href="#479">479</a>     <em>/**<em>*</em></em><a name="480" href="#480">480</a> <em>     * Put the given queue on the readyClassQueues queue</em><a name="481" href="#481">481</a> <em>     * @param wq</em><a name="482" href="#482">482</a> <em>     */</em><a name="483" href="#483">483</a>     <strong>private</strong> <strong>void</strong> readyQueue(<a href="../../../../org/archive/crawler/frontier/WorkQueue.html">WorkQueue</a> wq) {<a name="484" href="#484">484</a>         <strong>try</strong> {<a name="485" href="#485">485</a>             wq.setActive(<strong>this</strong>, <strong>true</strong>);<a name="486" href="#486">486</a>             readyClassQueues.put(wq.getClassKey());<a name="487" href="#487">487</a>         } <strong>catch</strong> (InterruptedException e) {<a name="488" href="#488">488</a>             e.printStackTrace();<a name="489" href="#489">489</a>             System.err.println(<span class="string">"unable to ready queue "</span>+wq);<a name="490" href="#490">490</a>             <em class="comment">// propagate interrupt up </em><a name="491" href="#491">491</a>             <strong>throw</strong> <strong>new</strong> RuntimeException(e);<a name="492" href="#492">492</a>         }<a name="493" href="#493">493</a>     }<a name="494" href="#494">494</a> <a name="495" href="#495">495</a>     <em>/**<em>*</em></em><a name="496" href="#496">496</a> <em>     * Put the given queue on the inactiveQueues queue</em><a name="497" href="#497">497</a> <em>     * @param wq</em><a name="498" href="#498">498</a> <em>     */</em><a name="499" href="#499">499</a>     <strong>private</strong> <strong>void</strong> deactivateQueue(<a href="../../../../org/archive/crawler/frontier/WorkQueue.html">WorkQueue</a> wq) {<a name="500" href="#500">500</a>         <strong>try</strong> {<a name="501" href="#501">501</a>             wq.setSessionBalance(0); <em class="comment">// zero out session balance</em><a name="502" href="#502">502</a>             inactiveQueues.put(wq.getClassKey());<a name="503" href="#503">503</a>             wq.setActive(<strong>this</strong>, false);<a name="504" href="#504">504</a>         } <strong>catch</strong> (InterruptedException e) {<a name="505" href="#505">505</a>             e.printStackTrace();<a name="506" href="#506">506</a>             System.err.println(<span class="string">"unable to deactivate queue "</span>+wq);<a name="507" href="#507">507</a>             <em class="comment">// propagate interrupt up </em><a name="508" href="#508">508</a>             <strong>throw</strong> <strong>new</strong> RuntimeException(e);<a name="509" href="#509">509</a>         }<a name="510" href="#510">510</a>     }<a name="511" href="#511">511</a>     <a name="512" href="#512">512</a>     <em>/**<em>*</em></em><a name="513" href="#513">513</a> <em>     * Put the given queue on the retiredQueues queue</em><a name="514" href="#514">514</a> <em>     * @param wq</em><a name="515" href="#515">515</a> <em>     */</em><a name="516" href="#516">516</a>     <strong>private</strong> <strong>void</strong> retireQueue(<a href="../../../../org/archive/crawler/frontier/WorkQueue.html">WorkQueue</a> wq) {<a name="517" href="#517">517</a>         <strong>try</strong> {<a name="518" href="#518">518</a>             retiredQueues.put(wq.getClassKey());<a name="519" href="#519">519</a>             decrementQueuedCount(wq.getCount());<a name="520" href="#520">520</a>             wq.setRetired(<strong>true</strong>);<a name="521" href="#521">521</a>             wq.setActive(<strong>this</strong>, false);<a name="522" href="#522">522</a>         } <strong>catch</strong> (InterruptedException e) {<a name="523" href="#523">523</a>             e.printStackTrace();<a name="524" href="#524">524</a>             System.err.println(<span class="string">"unable to retire queue "</span>+wq);<a name="525" href="#525">525</a>             <em class="comment">// propagate interrupt up </em><a name="526" href="#526">526</a>             <strong>throw</strong> <strong>new</strong> RuntimeException(e);<a name="527" href="#527">527</a>         }<a name="528" href="#528">528</a>     }<a name="529" href="#529">529</a>     <a name="530" href="#530">530</a>     <em>/**<em>* </em></em><a name="531" href="#531">531</a> <em>     * Accomodate any changes in settings.</em><a name="532" href="#532">532</a> <em>     * </em><a name="533" href="#533">533</a> <em>     * @see org.archive.crawler.framework.Frontier#kickUpdate()</em><a name="534" href="#534">534</a> <em>     */</em><a name="535" href="#535">535</a>     <strong>public</strong> <strong>void</strong> kickUpdate() {<a name="536" href="#536">536</a>         <strong>super</strong>.kickUpdate();<a name="537" href="#537">537</a>         <strong>int</strong> target = (Integer)getUncheckedAttribute(<strong>null</strong>,<a name="538" href="#538">538</a>                 ATTR_TARGET_READY_QUEUES_BACKLOG);<a name="539" href="#539">539</a>         <strong>if</strong> (target &lt; 1) {<a name="540" href="#540">540</a>             target = 1;<a name="541" href="#541">541</a>         }<a name="542" href="#542">542</a>         <strong>this</strong>.targetSizeForReadyQueues = target; <a name="543" href="#543">543</a>         <strong>try</strong> {<a name="544" href="#544">544</a>             initCostPolicy();<a name="545" href="#545">545</a>         } <strong>catch</strong> (FatalConfigurationException fce) {<a name="546" href="#546">546</a>             <strong>throw</strong> <strong>new</strong> RuntimeException(fce);<a name="547" href="#547">547</a>         }<a name="548" href="#548">548</a>         <em class="comment">// The rules for a 'retired' queue may have changed; so,</em><a name="549" href="#549">549</a>         <em class="comment">// unretire all queues to 'inactive'. If they still qualify</em><a name="550" href="#550">550</a>         <em class="comment">// as retired/overbudget next time they come up, they'll</em><a name="551" href="#551">551</a>         <em class="comment">// be re-retired; if not, they'll get a chance to become</em><a name="552" href="#552">552</a>         <em class="comment">// active under the new rules.</em><a name="553" href="#553">553</a>         Object key = <strong>this</strong>.retiredQueues.poll();<a name="554" href="#554">554</a>         <strong>while</strong> (key != <strong>null</strong>) {<a name="555" href="#555">555</a>             <a href="../../../../org/archive/crawler/frontier/WorkQueue.html">WorkQueue</a> q = (WorkQueue)<strong>this</strong>.allQueues.get(key);<a name="556" href="#556">556</a>             <strong>if</strong>(q != <strong>null</strong>) {<a name="557" href="#557">557</a>                 unretireQueue(q);<a name="558" href="#558">558</a>             }<a name="559" href="#559">559</a>             key = <strong>this</strong>.retiredQueues.poll();<a name="560" href="#560">560</a>         }<a name="561" href="#561">561</a>     }<a name="562" href="#562">562</a>     <em>/**<em>*</em></em><a name="563" href="#563">563</a> <em>     * Restore a retired queue to the 'inactive' state. </em>

⌨️ 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 + -