abstractfrontier.html

一个开源的网页爬虫一个开源的网页爬虫一个开源的网页爬虫一个开源的网页爬虫一个开源的网页爬虫一个开源的网页爬虫

<a name="415" href="#415">415</a> 
<a name="416" href="#416">416</a>     <strong>protected</strong> <strong>void</strong> doJournalEmitted(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> c) {
<a name="417" href="#417">417</a>         <strong>if</strong> (<strong>this</strong>.recover != <strong>null</strong>) {
<a name="418" href="#418">418</a>             <strong>this</strong>.recover.emitted(c);
<a name="419" href="#419">419</a>         }
<a name="420" href="#420">420</a>     }
<a name="421" href="#421">421</a> 
<a name="422" href="#422">422</a>     <em>/**<em>*</em></em>
<a name="423" href="#423">423</a> <em>     * Frontier is empty only if all queues are empty and no URIs are in-process</em>
<a name="424" href="#424">424</a> <em>     * </em>
<a name="425" href="#425">425</a> <em>     * @return True if queues are empty.</em>
<a name="426" href="#426">426</a> <em>     */</em>
<a name="427" href="#427">427</a>     <strong>public</strong> <strong>synchronized</strong> <strong>boolean</strong> isEmpty() {
<a name="428" href="#428">428</a>         <strong>return</strong> queuedUriCount == 0;
<a name="429" href="#429">429</a>     }
<a name="430" href="#430">430</a> 
<a name="431" href="#431">431</a>     <em>/**<em>*</em></em>
<a name="432" href="#432">432</a> <em>     * Increment the running count of queued URIs. Synchronized because</em>
<a name="433" href="#433">433</a> <em>     * operations on longs are not atomic.</em>
<a name="434" href="#434">434</a> <em>     */</em>
<a name="435" href="#435">435</a>     <strong>protected</strong> <strong>synchronized</strong> <strong>void</strong> incrementQueuedUriCount() {
<a name="436" href="#436">436</a>         queuedUriCount++;
<a name="437" href="#437">437</a>     }
<a name="438" href="#438">438</a> 
<a name="439" href="#439">439</a>     <em>/**<em>*</em></em>
<a name="440" href="#440">440</a> <em>     * Increment the running count of queued URIs. Synchronized because</em>
<a name="441" href="#441">441</a> <em>     * operations on longs are not atomic.</em>
<a name="442" href="#442">442</a> <em>     * </em>
<a name="443" href="#443">443</a> <em>     * @param increment</em>
<a name="444" href="#444">444</a> <em>     *            amount to increment the queued count</em>
<a name="445" href="#445">445</a> <em>     */</em>
<a name="446" href="#446">446</a>     <strong>protected</strong> <strong>synchronized</strong> <strong>void</strong> incrementQueuedUriCount(<strong>long</strong> increment) {
<a name="447" href="#447">447</a>         queuedUriCount += increment;
<a name="448" href="#448">448</a>     }
<a name="449" href="#449">449</a> 
<a name="450" href="#450">450</a>     <em>/**<em>*</em></em>
<a name="451" href="#451">451</a> <em>     * Note that a number of queued Uris have been deleted.</em>
<a name="452" href="#452">452</a> <em>     * </em>
<a name="453" href="#453">453</a> <em>     * @param numberOfDeletes</em>
<a name="454" href="#454">454</a> <em>     */</em>
<a name="455" href="#455">455</a>     <strong>protected</strong> <strong>synchronized</strong> <strong>void</strong> decrementQueuedCount(<strong>long</strong> numberOfDeletes) {
<a name="456" href="#456">456</a>         queuedUriCount -= numberOfDeletes;
<a name="457" href="#457">457</a>     }
<a name="458" href="#458">458</a> 
<a name="459" href="#459">459</a>     <em>/**<em>*</em></em>
<a name="460" href="#460">460</a> <em>     * (non-Javadoc)</em>
<a name="461" href="#461">461</a> <em>     * </em>
<a name="462" href="#462">462</a> <em>     * @see org.archive.crawler.framework.Frontier#queuedUriCount()</em>
<a name="463" href="#463">463</a> <em>     */</em>
<a name="464" href="#464">464</a>     <strong>public</strong> <strong>long</strong> queuedUriCount() {
<a name="465" href="#465">465</a>         <strong>return</strong> queuedUriCount;
<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>     * (non-Javadoc)</em>
<a name="470" href="#470">470</a> <em>     * </em>
<a name="471" href="#471">471</a> <em>     * @see org.archive.crawler.framework.Frontier#finishedUriCount()</em>
<a name="472" href="#472">472</a> <em>     */</em>
<a name="473" href="#473">473</a>     <strong>public</strong> <strong>long</strong> finishedUriCount() {
<a name="474" href="#474">474</a>         <strong>return</strong> succeededFetchCount + failedFetchCount + disregardedUriCount;
<a name="475" href="#475">475</a>     }
<a name="476" href="#476">476</a> 
<a name="477" href="#477">477</a>     <em>/**<em>*</em></em>
<a name="478" href="#478">478</a> <em>     * Increment the running count of successfully fetched URIs. Synchronized</em>
<a name="479" href="#479">479</a> <em>     * because operations on longs are not atomic.</em>
<a name="480" href="#480">480</a> <em>     */</em>
<a name="481" href="#481">481</a>     <strong>protected</strong> <strong>synchronized</strong> <strong>void</strong> incrementSucceededFetchCount() {
<a name="482" href="#482">482</a>         succeededFetchCount++;
<a name="483" href="#483">483</a>     }
<a name="484" href="#484">484</a> 
<a name="485" href="#485">485</a>     <em>/**<em>*</em></em>
<a name="486" href="#486">486</a> <em>     * (non-Javadoc)</em>
<a name="487" href="#487">487</a> <em>     * </em>
<a name="488" href="#488">488</a> <em>     * @see org.archive.crawler.framework.Frontier#succeededFetchCount()</em>
<a name="489" href="#489">489</a> <em>     */</em>
<a name="490" href="#490">490</a>     <strong>public</strong> <strong>long</strong> succeededFetchCount() {
<a name="491" href="#491">491</a>         <strong>return</strong> succeededFetchCount;
<a name="492" href="#492">492</a>     }
<a name="493" href="#493">493</a> 
<a name="494" href="#494">494</a>     <em>/**<em>*</em></em>
<a name="495" href="#495">495</a> <em>     * Increment the running count of failed URIs. Synchronized because</em>
<a name="496" href="#496">496</a> <em>     * operations on longs are not atomic.</em>
<a name="497" href="#497">497</a> <em>     */</em>
<a name="498" href="#498">498</a>     <strong>protected</strong> <strong>synchronized</strong> <strong>void</strong> incrementFailedFetchCount() {
<a name="499" href="#499">499</a>         failedFetchCount++;
<a name="500" href="#500">500</a>     }
<a name="501" href="#501">501</a> 
<a name="502" href="#502">502</a>     <em>/**<em>*</em></em>
<a name="503" href="#503">503</a> <em>     * (non-Javadoc)</em>
<a name="504" href="#504">504</a> <em>     * </em>
<a name="505" href="#505">505</a> <em>     * @see org.archive.crawler.framework.Frontier#failedFetchCount()</em>
<a name="506" href="#506">506</a> <em>     */</em>
<a name="507" href="#507">507</a>     <strong>public</strong> <strong>long</strong> failedFetchCount() {
<a name="508" href="#508">508</a>         <strong>return</strong> failedFetchCount;
<a name="509" href="#509">509</a>     }
<a name="510" href="#510">510</a> 
<a name="511" href="#511">511</a>     <em>/**<em>*</em></em>
<a name="512" href="#512">512</a> <em>     * Increment the running count of disregarded URIs. Synchronized because</em>
<a name="513" href="#513">513</a> <em>     * operations on longs are not atomic.</em>
<a name="514" href="#514">514</a> <em>     */</em>
<a name="515" href="#515">515</a>     <strong>protected</strong> <strong>synchronized</strong> <strong>void</strong> incrementDisregardedUriCount() {
<a name="516" href="#516">516</a>         disregardedUriCount++;
<a name="517" href="#517">517</a>     }
<a name="518" href="#518">518</a> 
<a name="519" href="#519">519</a>     <strong>public</strong> <strong>long</strong> disregardedUriCount() {
<a name="520" href="#520">520</a>         <strong>return</strong> disregardedUriCount;
<a name="521" href="#521">521</a>     }
<a name="522" href="#522">522</a> 
<a name="523" href="#523">523</a>     <strong>public</strong> <strong>long</strong> totalBytesWritten() {
<a name="524" href="#524">524</a>         <strong>return</strong> totalProcessedBytes;
<a name="525" href="#525">525</a>     }
<a name="526" href="#526">526</a> 
<a name="527" href="#527">527</a>     <em>/**<em>*</em></em>
<a name="528" href="#528">528</a> <em>     * Load up the seeds.</em>
<a name="529" href="#529">529</a> <em>     * </em>
<a name="530" href="#530">530</a> <em>     * This method is called on initialize and inside in the crawlcontroller</em>
<a name="531" href="#531">531</a> <em>     * when it wants to force reloading of configuration.</em>
<a name="532" href="#532">532</a> <em>     * </em>
<a name="533" href="#533">533</a> <em>     * @see org.archive.crawler.framework.CrawlController#kickUpdate()</em>
<a name="534" href="#534">534</a> <em>     */</em>
<a name="535" href="#535">535</a>     <strong>public</strong> <strong>void</strong> loadSeeds() {
<a name="536" href="#536">536</a>         Writer ignoredWriter = <strong>new</strong> StringWriter();
<a name="537" href="#537">537</a>         logger.info(<span class="string">"beginning"</span>);
<a name="538" href="#538">538</a>         <em class="comment">// Get the seeds to refresh.</em>
<a name="539" href="#539">539</a>         Iterator iter = <strong>this</strong>.controller.getScope().seedsIterator(ignoredWriter);
<a name="540" href="#540">540</a>         <strong>int</strong> count = 0; 
<a name="541" href="#541">541</a>         <strong>while</strong> (iter.hasNext()) {
<a name="542" href="#542">542</a>             <a href="../../../../org/archive/net/UURI.html">UURI</a> u = (UURI)iter.next();
<a name="543" href="#543">543</a>             <a href="../../../../org/archive/crawler/datamodel/CandidateURI.html">CandidateURI</a> caUri = CandidateURI.createSeedCandidateURI(u);
<a name="544" href="#544">544</a>             caUri.setSchedulingDirective(CandidateURI.MEDIUM);
<a name="545" href="#545">545</a>             <strong>if</strong> (((Boolean)getUncheckedAttribute(<strong>null</strong>, ATTR_SOURCE_TAG_SEEDS))
<a name="546" href="#546">546</a>                     .booleanValue()) {
<a name="547" href="#547">547</a>                 caUri.putString(CoreAttributeConstants.A_SOURCE_TAG,caUri.toString());
<a name="548" href="#548">548</a>                 caUri.makeHeritable(CoreAttributeConstants.A_SOURCE_TAG);
<a name="549" href="#549">549</a>             }
<a name="550" href="#550">550</a>             schedule(caUri);
<a name="551" href="#551">551</a>             count++;
<a name="552" href="#552">552</a>             <strong>if</strong>(count%1000==0) {
<a name="553" href="#553">553</a>                 logger.info(count+<span class="string">" seeds"</span>);
<a name="554" href="#554">554</a>             }
<a name="555" href="#555">555</a>         }