statisticstracker.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="584" href="#584">584</a>     }
<a name="585" href="#585">585</a> 
<a name="586" href="#586">586</a>     <em>/**<em>*</em></em>
<a name="587" href="#587">587</a> <em>     * Number of URIs that have &lt;i>finished&lt;/i> processing.</em>
<a name="588" href="#588">588</a> <em>     *</em>
<a name="589" href="#589">589</a> <em>     * @return Number of URIs that have finished processing</em>
<a name="590" href="#590">590</a> <em>     *</em>
<a name="591" href="#591">591</a> <em>     * @see org.archive.crawler.framework.Frontier#finishedUriCount()</em>
<a name="592" href="#592">592</a> <em>     */</em>
<a name="593" href="#593">593</a>     <strong>public</strong> <strong>long</strong> finishedUriCount() {
<a name="594" href="#594">594</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="595" href="#595">595</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong> ?
<a name="596" href="#596">596</a>             controller.getFrontier().finishedUriCount() : finishedUriCount;
<a name="597" href="#597">597</a>     }
<a name="598" href="#598">598</a> 
<a name="599" href="#599">599</a>     <em>/**<em>*</em></em>
<a name="600" href="#600">600</a> <em>     * Get the total number of failed fetch attempts (connection failures -> give up, etc)</em>
<a name="601" href="#601">601</a> <em>     *</em>
<a name="602" href="#602">602</a> <em>     * @return The total number of failed fetch attempts</em>
<a name="603" href="#603">603</a> <em>     */</em>
<a name="604" href="#604">604</a>     <strong>public</strong> <strong>long</strong> failedFetchAttempts() {
<a name="605" href="#605">605</a>         <em class="comment">// While shouldrun is true we can use info direct from the crawler.</em>
<a name="606" href="#606">606</a>         <em class="comment">// After that our last snapshot will have to do.</em>
<a name="607" href="#607">607</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="608" href="#608">608</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong> ?
<a name="609" href="#609">609</a>             controller.getFrontier().failedFetchCount() : downloadFailures;
<a name="610" href="#610">610</a>     }
<a name="611" href="#611">611</a> 
<a name="612" href="#612">612</a>     <em>/**<em>*</em></em>
<a name="613" href="#613">613</a> <em>     * Get the total number of failed fetch attempts (connection failures -> give up, etc)</em>
<a name="614" href="#614">614</a> <em>     *</em>
<a name="615" href="#615">615</a> <em>     * @return The total number of failed fetch attempts</em>
<a name="616" href="#616">616</a> <em>     */</em>
<a name="617" href="#617">617</a>     <strong>public</strong> <strong>long</strong> disregardedFetchAttempts() {
<a name="618" href="#618">618</a>         <em class="comment">// While shouldrun is true we can use info direct from the crawler.</em>
<a name="619" href="#619">619</a>         <em class="comment">// After that our last snapshot will have to do.</em>
<a name="620" href="#620">620</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="621" href="#621">621</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong>?
<a name="622" href="#622">622</a>             controller.getFrontier().disregardedUriCount() : downloadDisregards;
<a name="623" href="#623">623</a>     }
<a name="624" href="#624">624</a> 
<a name="625" href="#625">625</a>     <strong>public</strong> <strong>long</strong> successfullyFetchedCount() {
<a name="626" href="#626">626</a>         <em class="comment">// While shouldrun is true we can use info direct from the crawler.</em>
<a name="627" href="#627">627</a>         <em class="comment">// After that our last snapshot will have to do.</em>
<a name="628" href="#628">628</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="629" href="#629">629</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong>?
<a name="630" href="#630">630</a>             controller.getFrontier().succeededFetchCount() : downloadedUriCount;
<a name="631" href="#631">631</a>     }
<a name="632" href="#632">632</a>     
<a name="633" href="#633">633</a>     <strong>public</strong> <strong>long</strong> totalCount() {
<a name="634" href="#634">634</a>         <strong>return</strong> queuedUriCount() + activeThreadCount() +
<a name="635" href="#635">635</a>             successfullyFetchedCount();
<a name="636" href="#636">636</a>     }
<a name="637" href="#637">637</a> 
<a name="638" href="#638">638</a>     <em>/**<em>*</em></em>
<a name="639" href="#639">639</a> <em>     * Ratio of number of threads that would theoretically allow</em>
<a name="640" href="#640">640</a> <em>     * maximum crawl progress (if each was as productive as current</em>
<a name="641" href="#641">641</a> <em>     * threads), to current number of threads.</em>
<a name="642" href="#642">642</a> <em>     * </em>
<a name="643" href="#643">643</a> <em>     * @return float congestion ratio </em>
<a name="644" href="#644">644</a> <em>     */</em>
<a name="645" href="#645">645</a>     <strong>public</strong> <strong>float</strong> congestionRatio() {
<a name="646" href="#646">646</a>         <em class="comment">// While shouldrun is true we can use info direct from the crawler.</em>
<a name="647" href="#647">647</a>         <em class="comment">// After that our last snapshot will have to do.</em>
<a name="648" href="#648">648</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="649" href="#649">649</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong> ?
<a name="650" href="#650">650</a>             controller.getFrontier().congestionRatio() : congestionRatio;
<a name="651" href="#651">651</a>     }
<a name="652" href="#652">652</a>     
<a name="653" href="#653">653</a>     <em>/**<em>*</em></em>
<a name="654" href="#654">654</a> <em>     * Ordinal position of the 'deepest' URI eligible </em>
<a name="655" href="#655">655</a> <em>     * for crawling. Essentially, the length of the longest</em>
<a name="656" href="#656">656</a> <em>     * frontier internal queue. </em>
<a name="657" href="#657">657</a> <em>     * </em>
<a name="658" href="#658">658</a> <em>     * @return long URI count to deepest URI</em>
<a name="659" href="#659">659</a> <em>     */</em>
<a name="660" href="#660">660</a>     <strong>public</strong> <strong>long</strong> deepestUri() {
<a name="661" href="#661">661</a>         <em class="comment">// While shouldrun is true we can use info direct from the crawler.</em>
<a name="662" href="#662">662</a>         <em class="comment">// After that our last snapshot will have to do.</em>
<a name="663" href="#663">663</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="664" href="#664">664</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong> ?
<a name="665" href="#665">665</a>             controller.getFrontier().deepestUri() : deepestUri;
<a name="666" href="#666">666</a>     }
<a name="667" href="#667">667</a>     
<a name="668" href="#668">668</a>     <em>/**<em>*</em></em>
<a name="669" href="#669">669</a> <em>     * Average depth of the last URI in all eligible queues.</em>
<a name="670" href="#670">670</a> <em>     * That is, the average length of all eligible queues.</em>
<a name="671" href="#671">671</a> <em>     * </em>
<a name="672" href="#672">672</a> <em>     * @return long average depth of last URIs in queues </em>
<a name="673" href="#673">673</a> <em>     */</em>
<a name="674" href="#674">674</a>     <strong>public</strong> <strong>long</strong> averageDepth() {
<a name="675" href="#675">675</a>         <em class="comment">// While shouldrun is true we can use info direct from the crawler.</em>
<a name="676" href="#676">676</a>         <em class="comment">// After that our last snapshot will have to do.</em>
<a name="677" href="#677">677</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="678" href="#678">678</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong> ?
<a name="679" href="#679">679</a>             controller.getFrontier().averageDepth() : averageDepth;
<a name="680" href="#680">680</a>     }
<a name="681" href="#681">681</a>     
<a name="682" href="#682">682</a>     <em>/**<em>*</em></em>
<a name="683" href="#683">683</a> <em>     * Number of URIs &lt;i>queued&lt;/i> up and waiting for processing.</em>
<a name="684" href="#684">684</a> <em>     *</em>
<a name="685" href="#685">685</a> <em>     * &lt;p>If crawl not running (paused or stopped) this will return the value</em>
<a name="686" href="#686">686</a> <em>     * of the last snapshot.</em>
<a name="687" href="#687">687</a> <em>     *</em>
<a name="688" href="#688">688</a> <em>     * @return Number of URIs queued up and waiting for processing.</em>
<a name="689" href="#689">689</a> <em>     *</em>
<a name="690" href="#690">690</a> <em>     * @see org.archive.crawler.framework.Frontier#queuedUriCount()</em>
<a name="691" href="#691">691</a> <em>     */</em>
<a name="692" href="#692">692</a>     <strong>public</strong> <strong>long</strong> queuedUriCount() {
<a name="693" href="#693">693</a>         <em class="comment">// While shouldrun is true we can use info direct from the crawler.</em>
<a name="694" href="#694">694</a>         <em class="comment">// After that our last snapshot will have to do.</em>
<a name="695" href="#695">695</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="696" href="#696">696</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong>?
<a name="697" href="#697">697</a>             controller.getFrontier().queuedUriCount() : queuedUriCount;
<a name="698" href="#698">698</a>     }
<a name="699" href="#699">699</a> 
<a name="700" href="#700">700</a>     <strong>public</strong> <strong>long</strong> totalBytesWritten() {
<a name="701" href="#701">701</a>         <strong>return</strong> shouldrun &amp;&amp; <strong>this</strong>.controller != <strong>null</strong> &amp;&amp;
<a name="702" href="#702">702</a>                 <strong>this</strong>.controller.getFrontier() != <strong>null</strong>?
<a name="703" href="#703">703</a>             controller.getFrontier().totalBytesWritten() : totalProcessedBytes;
<a name="704" href="#704">704</a>     }
<a name="705" href="#705">705</a> 
<a name="706" href="#706">706</a>     <em>/**<em>*</em></em>
<a name="707" href="#707">707</a> <em>     * If the curi is a seed, we update the processedSeeds table.</em>
<a name="708" href="#708">708</a> <em>     *</em>
<a name="709" href="#709">709</a> <em>     * @param curi The CrawlURI that may be a seed.</em>
<a name="710" href="#710">710</a> <em>     * @param disposition The dispositino of the CrawlURI.</em>
<a name="711" href="#711">711</a> <em>     */</em>
<a name="712" href="#712">712</a>     <strong>private</strong> <strong>void</strong> handleSeed(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi, String disposition) {
<a name="713" href="#713">713</a>         <strong>if</strong>(curi.isSeed()){
<a name="714" href="#714">714</a>             <a href="../../../../org/archive/crawler/admin/SeedRecord.html">SeedRecord</a> sr = <strong>new</strong> <a href="../../../../org/archive/crawler/admin/SeedRecord.html">SeedRecord</a>(curi, disposition);
<a name="715" href="#715">715</a>             processedSeedsRecords.put(sr.getUri(), sr);
<a name="716" href="#716">716</a>         }
<a name="717" href="#717">717</a>     }
<a name="718" href="#718">718</a> 
<a name="719" href="#719">719</a>     <strong>public</strong> <strong>void</strong> crawledURISuccessful(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi) {
<a name="720" href="#720">720</a>         handleSeed(curi,SEED_DISPOSITION_SUCCESS);
<a name="721" href="#721">721</a>         <em class="comment">// Save status codes</em>
<a name="722" href="#722">722</a>         incrementMapCount(statusCodeDistribution,
<a name="723" href="#723">723</a>             Integer.toString(curi.getFetchStatus()));
<a name="724" href="#724">724</a> 
<a name="725" href="#725">725</a>         <em class="comment">// Save mime types</em>
<a name="726" href="#726">726</a>         String mime = MimetypeUtils.truncate(curi.getContentType());
<a name="727" href="#727">727</a>         incrementMapCount(mimeTypeDistribution, mime);
<a name="728" href="#728">728</a>         incrementMapCount(mimeTypeBytes, mime, curi.getContentSize());
<a name="729" href="#729">729</a> 
<a name="730" href="#730">730</a>         <em class="comment">// Save hosts stats.</em>
<a name