extractoruniversal.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="452" href="#452">452</a>     <em>/**<em>*</em></em>
<a name="453" href="#453">453</a> <em>     * This method takes a look at a string and determines if it could be a URL.</em>
<a name="454" href="#454">454</a> <em>     * To qualify the string must either begin with "<a href="http://" target="alexandria_uri">http://</a>" (https would also</em>
<a name="455" href="#455">455</a> <em>     * work) followed by something that looks like an IP address or contain</em>
<a name="456" href="#456">456</a> <em>     * within the string (possible at the end but not at the beginning) a TLD</em>
<a name="457" href="#457">457</a> <em>     * (Top Level Domain) preceded by a dot.</em>
<a name="458" href="#458">458</a> <em>     *</em>
<a name="459" href="#459">459</a> <em>     * @param lookat The string to examine in an effort to determine if it</em>
<a name="460" href="#460">460</a> <em>     * could be a URL</em>
<a name="461" href="#461">461</a> <em>     * @return True if the string matches the above criteria for a URL.</em>
<a name="462" href="#462">462</a> <em>     */</em>
<a name="463" href="#463">463</a>     <strong>private</strong> <strong>boolean</strong> looksLikeAnURL(String lookat) {
<a name="464" href="#464">464</a>         <strong>if</strong>(lookat.indexOf(<span class="string">"http://"</span>)==0 || lookat.indexOf(<span class="string">"https://"</span>)==0){
<a name="465" href="#465">465</a>             <em class="comment">//Check if the rest of the string looks like an IP address.</em>
<a name="466" href="#466">466</a>             <em class="comment">//if so return true. Otherwise continue on.</em>
<a name="467" href="#467">467</a>             Matcher ip = TextUtils.getMatcher(IP_ADDRESS, lookat);
<a name="468" href="#468">468</a>             <strong>boolean</strong> testVal = ip.matches();
<a name="469" href="#469">469</a>             TextUtils.recycleMatcher(ip);
<a name="470" href="#470">470</a>             <strong>if</strong>(testVal){
<a name="471" href="#471">471</a>                 <strong>return</strong> <strong>true</strong>;
<a name="472" href="#472">472</a>             }
<a name="473" href="#473">473</a>         }
<a name="474" href="#474">474</a> 
<a name="475" href="#475">475</a>         <strong>int</strong> dot = lookat.indexOf(<span class="string">"."</span>);
<a name="476" href="#476">476</a>         <strong>if</strong>(dot!=0){<em class="comment">//An URL can't start with a .tld.</em>
<a name="477" href="#477">477</a>             <strong>while</strong>(dot != -1 &amp;&amp; dot &lt; lookat.length()){
<a name="478" href="#478">478</a>                 lookat = lookat.substring(dot+1);
<a name="479" href="#479">479</a>                 <strong>if</strong> (isTLD(lookat.substring(0, lookat.length() &lt;= 6?
<a name="480" href="#480">480</a>                     lookat.length(): 6)))
<a name="481" href="#481">481</a>                 {
<a name="482" href="#482">482</a>                     <strong>return</strong> <strong>true</strong>;
<a name="483" href="#483">483</a>                 }
<a name="484" href="#484">484</a>                 dot = lookat.indexOf(<span class="string">"."</span>);
<a name="485" href="#485">485</a>             }
<a name="486" href="#486">486</a>         }
<a name="487" href="#487">487</a> 
<a name="488" href="#488">488</a>         <strong>return</strong> false;
<a name="489" href="#489">489</a>     }
<a name="490" href="#490">490</a> 
<a name="491" href="#491">491</a>     <em>/**<em>*</em></em>
<a name="492" href="#492">492</a> <em>     * Checks if a string is equal to known Top Level Domain. The string may</em>
<a name="493" href="#493">493</a> <em>     * contain additional characters &lt;i>after&lt;/i> the TLD but not before.</em>
<a name="494" href="#494">494</a> <em>     * @param potentialTLD The string (usually 2-6 chars) to check if it starts</em>
<a name="495" href="#495">495</a> <em>     * with a TLD.</em>
<a name="496" href="#496">496</a> <em>     * @return True if the given string starts with the name of a known TLD</em>
<a name="497" href="#497">497</a> <em>     *</em>
<a name="498" href="#498">498</a> <em>     * @see #TLDs</em>
<a name="499" href="#499">499</a> <em>     */</em>
<a name="500" href="#500">500</a>     <strong>private</strong> <strong>boolean</strong> isTLD(String potentialTLD) {
<a name="501" href="#501">501</a>         <strong>if</strong>(potentialTLD.length()&lt;2){
<a name="502" href="#502">502</a>             <strong>return</strong> false;
<a name="503" href="#503">503</a>         }
<a name="504" href="#504">504</a> 
<a name="505" href="#505">505</a>         potentialTLD.toLowerCase();
<a name="506" href="#506">506</a>         Matcher uri = TextUtils.getMatcher(TLDs, potentialTLD);
<a name="507" href="#507">507</a>         <strong>boolean</strong> ret = uri.matches();
<a name="508" href="#508">508</a>         TextUtils.recycleMatcher(uri);
<a name="509" href="#509">509</a>         <strong>return</strong> ret;
<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>     * Determines if a char (as represented by an int in the range of 0-255) is</em>
<a name="514" href="#514">514</a> <em>     * a character (in the Ansi character set) that can be present in a URL.</em>
<a name="515" href="#515">515</a> <em>     * This method takes a &lt;b>strict&lt;/b> approach to what characters can be in</em>
<a name="516" href="#516">516</a> <em>     * a URL.</em>
<a name="517" href="#517">517</a> <em>     * &lt;p></em>
<a name="518" href="#518">518</a> <em>     * The following are considered to be 'URLable'&lt;br></em>
<a name="519" href="#519">519</a> <em>     * &lt;ul></em>
<a name="520" href="#520">520</a> <em>     *  &lt;li> &lt;code># $ % &amp; + , - . /&lt;/code> values 35-38,43-47</em>
<a name="521" href="#521">521</a> <em>     *  &lt;li> &lt;code>[0-9]&lt;/code> values 48-57</em>
<a name="522" href="#522">522</a> <em>     *  &lt;li> &lt;code>: ; = ? @&lt;/code> value 58-59,61,63-64</em>
<a name="523" href="#523">523</a> <em>     *  &lt;li> &lt;code>[A-Z]&lt;/code> values 65-90</em>
<a name="524" href="#524">524</a> <em>     *  &lt;li> &lt;code>_&lt;/code> value 95</em>
<a name="525" href="#525">525</a> <em>     *  &lt;li> &lt;code>[a-z]&lt;/code> values 97-122</em>
<a name="526" href="#526">526</a> <em>     *  &lt;li> &lt;code>~&lt;/code> value 126</em>
<a name="527" href="#527">527</a> <em>     * &lt;/ul></em>
<a name="528" href="#528">528</a> <em>     * &lt;p></em>
<a name="529" href="#529">529</a> <em>     * To summerize, the following ranges are considered URLable:&lt;br></em>
<a name="530" href="#530">530</a> <em>     * 35-38,43-59,61,63-90,95,97-122,126</em>
<a name="531" href="#531">531</a> <em>     *</em>
<a name="532" href="#532">532</a> <em>     * @param ch The character (represented by an int) to test.</em>
<a name="533" href="#533">533</a> <em>     * @return True if it is a URLable character, false otherwise.</em>
<a name="534" href="#534">534</a> <em>     */</em>
<a name="535" href="#535">535</a>     <strong>private</strong> <strong>boolean</strong> isURLableChar(<strong>int</strong> ch) {
<a name="536" href="#536">536</a>         <strong>return</strong> (ch>=35 &amp;&amp; ch&lt;=38)
<a name="537" href="#537">537</a>             || (ch>=43 &amp;&amp; ch&lt;=59)
<a name="538" href="#538">538</a>             || (ch==61)
<a name="539" href="#539">539</a>             || (ch>=63 &amp;&amp; ch&lt;=90)
<a name="540" href="#540">540</a>             || (ch==95)
<a name="541" href="#541">541</a>             || (ch>=97 &amp;&amp; ch&lt;=122)
<a name="542" href="#542">542</a>             || (ch==126);
<a name="543" href="#543">543</a>     }
<a name="544" href="#544">544</a> 
<a name="545" href="#545">545</a>     <em class="comment">/*<em class="comment"> (non-Javadoc)</em></em>
<a name="546" href="#546">546</a> <em class="comment">     * @see org.archive.crawler.framework.Processor#report()</em>
<a name="547" href="#547">547</a> <em class="comment">     */</em>
<a name="548" href="#548">548</a>     <strong>public</strong> String report() {
<a name="549" href="#549">549</a>         StringBuffer ret = <strong>new</strong> StringBuffer();
<a name="550" href="#550">550</a>         ret.append(<span class="string">"Processor: org.archive.crawler.extractor."</span> +
<a name="551" href="#551">551</a>             <span class="string">"ExtractorUniversal\n"</span>);
<a name="552" href="#552">552</a>         ret.append(<span class="string">"  Function:          Link extraction on unknown file"</span> +
<a name="553" href="#553">553</a>             <span class="string">" types.\n"</span>);
<a name="554" href="#554">554</a>         ret.append(<span class="string">"  CrawlURIs handled: "</span> + numberOfCURIsHandled + <span class="string">"\n"</span>);
<a name="555" href="#555">555</a>         ret.append(<span class="string">"  Links extracted:   "</span> + numberOfLinksExtracted + <span class="string">"\n\n"</span>);
<a name="556" href="#556">556</a> 
<a name="557" href="#557">557</a>         <strong>return</strong> ret.toString();
<a name="558" href="#558">558</a>     }
<a name="559" href="#559">559</a> }
</pre>
<hr/><div id="footer">This page was automatically generated by <a href="http://maven.apache.org/">Maven</a></div></body>
</html>