regexphtmllinkextractor.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="308" href="#308">308</a> <em>     * @param cs</em>
<a name="309" href="#309">309</a> <em>     */</em>
<a name="310" href="#310">310</a>     <strong>protected</strong> <strong>void</strong> processScriptCode(CharSequence cs) {
<a name="311" href="#311">311</a>         RegexpJSLinkExtractor.extract(cs, source, base, next,
<a name="312" href="#312">312</a>                 extractErrorListener);
<a name="313" href="#313">313</a>     }
<a name="314" href="#314">314</a> 
<a name="315" href="#315">315</a>     <strong>static</strong> <strong>final</strong> String JAVASCRIPT = <span class="string">"(?i)^javascript:.*"</span>;
<a name="316" href="#316">316</a> 
<a name="317" href="#317">317</a>     <em>/**<em>*</em></em>
<a name="318" href="#318">318</a> <em>     * @param value</em>
<a name="319" href="#319">319</a> <em>     * @param context</em>
<a name="320" href="#320">320</a> <em>     */</em>
<a name="321" href="#321">321</a>     <strong>protected</strong> <strong>void</strong> processLink(CharSequence value, CharSequence context) {
<a name="322" href="#322">322</a>         String link = TextUtils.replaceAll(ESCAPED_AMP, value, <span class="string">"&amp;"</span>);
<a name="323" href="#323">323</a> 
<a name="324" href="#324">324</a>         <strong>if</strong>(TextUtils.matches(JAVASCRIPT, link)) {
<a name="325" href="#325">325</a>             processScriptCode(value.subSequence(11, value.length()));
<a name="326" href="#326">326</a>         } <strong>else</strong> {
<a name="327" href="#327">327</a>             addLinkFromString(link, context,Link.NAVLINK_HOP);
<a name="328" href="#328">328</a>         }
<a name="329" href="#329">329</a>     }
<a name="330" href="#330">330</a> 
<a name="331" href="#331">331</a>     <em>/**<em>*</em></em>
<a name="332" href="#332">332</a> <em>     * @param uri</em>
<a name="333" href="#333">333</a> <em>     * @param context</em>
<a name="334" href="#334">334</a> <em>     */</em>
<a name="335" href="#335">335</a>     <strong>private</strong> <strong>void</strong> addLinkFromString(String uri, CharSequence context, <strong>char</strong> hopType) {
<a name="336" href="#336">336</a>         <strong>try</strong> {
<a name="337" href="#337">337</a>             <a href="../../../org/archive/crawler/extractor/Link.html">Link</a> link = <strong>new</strong> <a href="../../../org/archive/crawler/extractor/Link.html">Link</a>(source, UURIFactory.getInstance(
<a name="338" href="#338">338</a>                     base, uri), context, hopType);
<a name="339" href="#339">339</a>             next.addLast(link);
<a name="340" href="#340">340</a>         } <strong>catch</strong> (URIException e) {
<a name="341" href="#341">341</a>            extractErrorListener.noteExtractError(e,source,uri);
<a name="342" href="#342">342</a>         }
<a name="343" href="#343">343</a>     }
<a name="344" href="#344">344</a> 
<a name="345" href="#345">345</a>     <strong>protected</strong> <strong>long</strong> processEmbed(CharSequence value, CharSequence context) {
<a name="346" href="#346">346</a>         String embed = TextUtils.replaceAll(ESCAPED_AMP, value, <span class="string">"&amp;"</span>);
<a name="347" href="#347">347</a>         addLinkFromString(embed, context,Link.EMBED_HOP);
<a name="348" href="#348">348</a>         <strong>return</strong> 1;
<a name="349" href="#349">349</a>     }
<a name="350" href="#350">350</a> 
<a name="351" href="#351">351</a>     <strong>static</strong> <strong>final</strong> String NON_HTML_PATH_EXTENSION =
<a name="352" href="#352">352</a>         <span class="string">"(?i)(gif)|(jp(e)?g)|(png)|(tif(f)?)|(bmp)|(avi)|(mov)|(mp(e)?g)"</span>+
<a name="353" href="#353">353</a>         <span class="string">"|(mp3)|(mp4)|(swf)|(wav)|(au)|(aiff)|(mid)"</span>;
<a name="354" href="#354">354</a> 
<a name="355" href="#355">355</a>     <strong>protected</strong> <strong>void</strong> processScript(CharSequence sequence, <strong>int</strong> endOfOpenTag) {
<a name="356" href="#356">356</a>         <strong>long</strong> tally = 0;
<a name="357" href="#357">357</a>         <em class="comment">// first, get attributes of script-open tag</em>
<a name="358" href="#358">358</a>         <em class="comment">// as per any other tag</em>
<a name="359" href="#359">359</a>         processGeneralTag(sequence.subSequence(0,6),
<a name="360" href="#360">360</a>             sequence.subSequence(0,endOfOpenTag));
<a name="361" href="#361">361</a> 
<a name="362" href="#362">362</a>         <em class="comment">// then, apply best-effort string-analysis heuristics</em>
<a name="363" href="#363">363</a>         <em class="comment">// against any code present (false positives are OK)</em>
<a name="364" href="#364">364</a>         processScriptCode(sequence.subSequence(endOfOpenTag, sequence.length()));
<a name="365" href="#365">365</a>     }
<a name="366" href="#366">366</a> 
<a name="367" href="#367">367</a>     <strong>protected</strong> <strong>void</strong> processMeta(CharSequence cs) {
<a name="368" href="#368">368</a>         Matcher attr = TextUtils.getMatcher(EACH_ATTRIBUTE_EXTRACTOR, cs);
<a name="369" href="#369">369</a> 
<a name="370" href="#370">370</a>         String name = <strong>null</strong>;
<a name="371" href="#371">371</a>         String httpEquiv = <strong>null</strong>;
<a name="372" href="#372">372</a>         String content = <strong>null</strong>;
<a name="373" href="#373">373</a> 
<a name="374" href="#374">374</a>         <strong>while</strong> (attr.find()) {
<a name="375" href="#375">375</a>             <strong>int</strong> valueGroup =
<a name="376" href="#376">376</a>                 (attr.start(12) > -1) ? 12 : (attr.start(13) > -1) ? 13 : 14;
<a name="377" href="#377">377</a>             CharSequence value =
<a name="378" href="#378">378</a>                 cs.subSequence(attr.start(valueGroup), attr.end(valueGroup));
<a name="379" href="#379">379</a>             <strong>if</strong> (attr.group(1).equalsIgnoreCase(<span class="string">"name"</span>)) {
<a name="380" href="#380">380</a>                 name = value.toString();
<a name="381" href="#381">381</a>             } <strong>else</strong> <strong>if</strong> (attr.group(1).equalsIgnoreCase(<span class="string">"http-equiv"</span>)) {
<a name="382" href="#382">382</a>                 httpEquiv = value.toString();
<a name="383" href="#383">383</a>             } <strong>else</strong> <strong>if</strong> (attr.group(1).equalsIgnoreCase(<span class="string">"content"</span>)) {
<a name="384" href="#384">384</a>                 content = value.toString();
<a name="385" href="#385">385</a>             }
<a name="386" href="#386">386</a>             <em class="comment">// TODO: handle other stuff</em>
<a name="387" href="#387">387</a>         }
<a name="388" href="#388">388</a>         TextUtils.recycleMatcher(attr);
<a name="389" href="#389">389</a>         
<a name="390" href="#390">390</a>         <em class="comment">// Look for the 'robots' meta-tag</em>
<a name="391" href="#391">391</a>         <strong>if</strong>(<span class="string">"robots"</span>.equalsIgnoreCase(name) &amp;&amp; content != <strong>null</strong> ) {
<a name="392" href="#392">392</a>             <strong>if</strong> (getHonorRobots())  {
<a name="393" href="#393">393</a>             String contentLower = content.toLowerCase();
<a name="394" href="#394">394</a>                 <strong>if</strong> ((contentLower.indexOf(<span class="string">"nofollow"</span>) >= 0
<a name="395" href="#395">395</a>                         || contentLower.indexOf(<span class="string">"none"</span>) >= 0)) {
<a name="396" href="#396">396</a>                     <em class="comment">// if 'nofollow' or 'none' is specified and we</em>
<a name="397" href="#397">397</a>                     <em class="comment">// are honoring robots, end html extraction</em>
<a name="398" href="#398">398</a>                     logger.fine(<span class="string">"HTML extraction skipped due to robots meta-tag for: "</span>
<a name="399" href="#399">399</a>                                     + source);
<a name="400" href="#400">400</a>                     cancelFurtherExtraction();
<a name="401" href="#401">401</a>                     <strong>return</strong>;
<a name="402" href="#402">402</a>                 }
<a name="403" href="#403">403</a>             }
<a name="404" href="#404">404</a>         } <strong>else</strong> <strong>if</strong> (<span class="string">"refresh"</span>.equalsIgnoreCase(httpEquiv) &amp;&amp; content != <strong>null</strong>) {
<a name="405" href="#405">405</a>             String refreshUri = content.substring(content.indexOf(<span class="string">"="</span>) + 1);
<a name="406" href="#406">406</a>             <strong>try</strong> {
<a name="407" href="#407">407</a>                 <a href="../../../org/archive/crawler/extractor/Link.html">Link</a> refreshLink = <strong>new</strong> <a href="../../../org/archive/crawler/extractor/Link.html">Link</a>(source, UURIFactory.getInstance(base,refreshUri), Link.elementContext(<span class="string">"meta"</span>,httpEquiv),Link.REFER_HOP);
<a name="408" href="#408">408</a>                 next.addLast(refreshLink);
<a name="409" href="#409">409</a>             } <strong>catch</strong> (URIException e) {
<a name="410" href="#410">410</a>                 extractErrorListener.noteExtractError(e,source,refreshUri);
<a name="411" href="#411">411</a>             }
<a name="412" href="#412">412</a>         }
<a name="413" href="#413">413</a>     }
<a name="414" href="#414">414</a> 
<a name="415" href="#415">415</a>     <em>/**<em>*</em></em>
<a name="416" href="#416">416</a> <em>     * @return whether to honor internal robots directives (eg meta robots)</em>
<a name="417" href="#417">417</a> <em>     */</em>
<a name="418" href="#418">418</a>     <strong>private</strong> <strong>boolean</strong> getHonorRobots() {
<a name="419" href="#419">419</a>         <strong>return</strong> honorRobots;
<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>     * Ensure no further Links are extracted (by setting matcher up to fail)</em>
<a name="424" href="#424">424</a> <em>     */</em>
<a name="425" href="#425">425</a>     <strong>private</strong> <strong>void</strong> cancelFurtherExtraction() {
<a name="426" href="#426">426</a>         <em class="comment">// java 1.5 only:</em>
<a name="427" href="#427">427</a>         <em class="comment">// tags.region(tags.regionEnd(),tags.regionEnd());</em>
<a name="428" href="#428">428</a>         tags.reset(<span class="string">""</span>); 
<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>     * @param sequence</em>
<a name="433" href="#433">433</a> <em>     * @param endOfOpenTag</em>
<a name="434" href="#434">434</a> <em>     */</em>
<a name="435" href="#435">435</a>     <strong>protected</strong> <strong>void</strong> processStyle(CharSequence sequence,
<a name="436" href="#436">436</a>             <strong>int</strong> endOfOpenTag)
<a name="437" href="#437">437</a>     {
<a name="438" href="#438">438</a>         <em class="comment">// First, get attributes of script-open tag as per any other tag.</em>
<a name="439" href="#439">439</a>         processGeneralTag(sequence.subSequence(0,6),
<a name="440" href="#440">440</a>             sequence.subSequence(0,endOfOpenTag));
<a name="441" href="#441">441</a> 
<a name="442" href="#442">442</a>         <em class="comment">// then, parse for URIs</em>
<a name="443" href="#443">443</a>         RegexpCSSLinkExtractor.extract(sequence.subSequence(endOfOpenTag,
<a name="444" href="#444">444</a>                 sequence.length()), source, base, next, extractErrorListener);
<a name="445" href="#445">445</a>     }
<a name="446" href="#446">446</a> 
<a name="447" href="#447">447</a>     <em>/**<em>*</em></em>
<a name="448" href="#448">448</a> <em>     * Discard all state. Another setup() is required to use again.</em>
<a name="449" href="#449">449</a> <em>     */</em>
<a name="450" href="#450">450</a>     <strong>public</strong> <strong>void</strong> reset() {
<a name="451" href="#451">451</a>         <strong>super</strong>.reset();
<a name="452" href="#452">452</a>         TextUtils.recycleMatcher(tags);
<a name="453" href="#453">453</a>         tags = <strong>null</strong>;
<a name="454" href="#454">454</a>     }
<a name="455" href="#455">455</a> 
<a name="456" href="#456">456</a>     <strong>protected</strong> <strong>static</strong> <a href="../../../org/archive/extractor/CharSequenceLinkExtractor.html">CharSequenceLinkExtractor</a> newDefaultInstance() {
<a name="457" href="#457">457</a>         <strong>return</strong> <strong>new</strong> <a href="../../../org/archive/extractor/RegexpHTMLLinkExtractor.html">RegexpHTMLLinkExtractor</a>();
<a name="458" href="#458">458</a>     }
<a name="459" href="#459">459</a> }
<a name="460" href="#460">460</a> 
</pre>
<hr/><div id="footer">This page was automatically generated by <a href="http://maven.apache.org/">Maven</a></div></body>
</html>