arcreader.html
来自「网络爬虫开源代码」· HTML 代码 · 共 753 行 · 第 1/5 页
HTML
753 行
<a name="294" href="#294">294</a> }<a name="295" href="#295">295</a> <em class="comment">// LOOP TERMINATION.</em><a name="296" href="#296">296</a> <strong>break</strong>;<a name="297" href="#297">297</a> } <strong>else</strong> <strong>if</strong> (c == HEADER_FIELD_SEPARATOR) {<a name="298" href="#298">298</a> <strong>if</strong> (!isStrict() && previous == HEADER_FIELD_SEPARATOR) {<a name="299" href="#299">299</a> <em class="comment">// Early ARCs sometimes had multiple spaces between fields.</em><a name="300" href="#300">300</a> <strong>continue</strong>;<a name="301" href="#301">301</a> }<a name="302" href="#302">302</a> <strong>if</strong> (list != <strong>null</strong>) {<a name="303" href="#303">303</a> list.add(buffer.toString());<a name="304" href="#304">304</a> }<a name="305" href="#305">305</a> <em class="comment">// reset to empty</em><a name="306" href="#306">306</a> buffer.setLength(0);<a name="307" href="#307">307</a> } <strong>else</strong> {<a name="308" href="#308">308</a> buffer.append((<strong>char</strong>)c);<a name="309" href="#309">309</a> }<a name="310" href="#310">310</a> }<a name="311" href="#311">311</a> <a name="312" href="#312">312</a> <em class="comment">// List must have at least 3 elements in it and no more than 10. If</em><a name="313" href="#313">313</a> <em class="comment">// it has other than this, then bogus parse.</em><a name="314" href="#314">314</a> <strong>if</strong> (list != <strong>null</strong> && (list.size() < 3 || list.size() > 100)) {<a name="315" href="#315">315</a> <strong>throw</strong> <strong>new</strong> IOException(<span class="string">"Unparseable header line: "</span> + list);<a name="316" href="#316">316</a> }<a name="317" href="#317">317</a> <a name="318" href="#318">318</a> <strong>return</strong> read;<a name="319" href="#319">319</a> }<a name="320" href="#320">320</a> <a name="321" href="#321">321</a> <em>/**<em>*</em></em><a name="322" href="#322">322</a> <em> * Compute metadata fields.</em><a name="323" href="#323">323</a> <em> *</em><a name="324" href="#324">324</a> <em> * Here we check the meta field has right number of items in it.</em><a name="325" href="#325">325</a> <em> *</em><a name="326" href="#326">326</a> <em> * @param keys Keys to use composing headerFields map.</em><a name="327" href="#327">327</a> <em> * @param values Values to set into the headerFields map.</em><a name="328" href="#328">328</a> <em> * @param v The version of this ARC file.</em><a name="329" href="#329">329</a> <em> * @param offset Offset into arc file.</em><a name="330" href="#330">330</a> <em> *</em><a name="331" href="#331">331</a> <em> * @return Metadata structure for this record.</em><a name="332" href="#332">332</a> <em> *</em><a name="333" href="#333">333</a> <em> * @exception IOException If no. of keys doesn't match no. of values.</em><a name="334" href="#334">334</a> <em> */</em><a name="335" href="#335">335</a> <strong>private</strong> <a href="../../../../org/archive/io/arc/ARCRecordMetaData.html">ARCRecordMetaData</a> computeMetaData(List<String> keys,<a name="336" href="#336">336</a> List<String> values, String v, <strong>long</strong> offset)<a name="337" href="#337">337</a> throws IOException {<a name="338" href="#338">338</a> <strong>if</strong> (keys.size() != values.size()) {<a name="339" href="#339">339</a> List<String> originalValues = values;<a name="340" href="#340">340</a> <strong>if</strong> (!isStrict()) {<a name="341" href="#341">341</a> values = fixSpaceInURL(values, keys.size());<a name="342" href="#342">342</a> <em class="comment">// If values still doesn't match key size, try and do</em><a name="343" href="#343">343</a> <em class="comment">// further repair.</em><a name="344" href="#344">344</a> <strong>if</strong> (keys.size() != values.size()) {<a name="345" href="#345">345</a> <em class="comment">// Early ARCs had a space in mimetype.</em><a name="346" href="#346">346</a> <strong>if</strong> (values.size() == (keys.size() + 1) &&<a name="347" href="#347">347</a> values.get(4).toLowerCase().startsWith(<span class="string">"charset="</span>)) {<a name="348" href="#348">348</a> List<String> nuvalues =<a name="349" href="#349">349</a> <strong>new</strong> ArrayList<String>(keys.size());<a name="350" href="#350">350</a> nuvalues.add(0, values.get(0));<a name="351" href="#351">351</a> nuvalues.add(1, values.get(1));<a name="352" href="#352">352</a> nuvalues.add(2, values.get(2));<a name="353" href="#353">353</a> nuvalues.add(3, values.get(3) + values.get(4));<a name="354" href="#354">354</a> nuvalues.add(4, values.get(5));<a name="355" href="#355">355</a> values = nuvalues;<a name="356" href="#356">356</a> } <strong>else</strong> <strong>if</strong>((values.size() + 1) == keys.size() &&<a name="357" href="#357">357</a> isLegitimateIPValue(values.get(1)) &&<a name="358" href="#358">358</a> isDate(values.get(2)) && isNumber(values.get(3))) {<a name="359" href="#359">359</a> <em class="comment">// Mimetype is empty.</em><a name="360" href="#360">360</a> List<String> nuvalues =<a name="361" href="#361">361</a> <strong>new</strong> ArrayList<String>(keys.size());<a name="362" href="#362">362</a> nuvalues.add(0, values.get(0));<a name="363" href="#363">363</a> nuvalues.add(1, values.get(1));<a name="364" href="#364">364</a> nuvalues.add(2, values.get(2));<a name="365" href="#365">365</a> nuvalues.add(3, <span class="string">"-"</span>);<a name="366" href="#366">366</a> nuvalues.add(4, values.get(3));<a name="367" href="#367">367</a> values = nuvalues;<a name="368" href="#368">368</a> }<a name="369" href="#369">369</a> }<a name="370" href="#370">370</a> }<a name="371" href="#371">371</a> <strong>if</strong> (keys.size() != values.size()) {<a name="372" href="#372">372</a> <strong>throw</strong> <strong>new</strong> IOException(<span class="string">"Size of field name keys does"</span> +<a name="373" href="#373">373</a> <span class="string">" not match count of field values: "</span> + values);<a name="374" href="#374">374</a> }<a name="375" href="#375">375</a> <em class="comment">// Note that field was fixed on stderr.</em><a name="376" href="#376">376</a> logStdErr(Level.WARNING, <span class="string">"Fixed spaces in metadata line at "</span> +<a name="377" href="#377">377</a> <span class="string">"offset "</span> + offset +<a name="378" href="#378">378</a> <span class="string">" Original: "</span> + originalValues + <span class="string">", New: "</span> + values);<a name="379" href="#379">379</a> }<a name="380" href="#380">380</a> <a name="381" href="#381">381</a> Map<Object, Object> headerFields =<a name="382" href="#382">382</a> <strong>new</strong> HashMap<Object, Object>(keys.size() + 2);<a name="383" href="#383">383</a> <strong>for</strong> (<strong>int</strong> i = 0; i < keys.size(); i++) {<a name="384" href="#384">384</a> headerFields.put(keys.get(i), values.get(i));<a name="385" href="#385">385</a> }<a name="386" href="#386">386</a> <a name="387" href="#387">387</a> <em class="comment">// Add a check for tabs in URLs. If any, replace with '%09'.</em><a name="388" href="#388">388</a> <em class="comment">// See https://sourceforge.net/tracker/?group_id=73833&atid=539099&func=detail&aid=1010966,</em><a name="389" href="#389">389</a> <em class="comment">// [ 1010966 ] crawl.log has URIs with spaces in them.</em><a name="390" href="#390">390</a> String url = (String)headerFields.get(URL_FIELD_KEY);<a name="391" href="#391">391</a> <strong>if</strong> (url != <strong>null</strong> && url.indexOf('\t') >= 0) {<a name="392" href="#392">392</a> headerFields.put(URL_FIELD_KEY,<a name="393" href="#393">393</a> TextUtils.replaceAll(<span class="string">"\t"</span>, url, <span class="string">"%09"</span>));<a name="394" href="#394">394</a> }<a name="395" href="#395">395</a> <a name="396" href="#396">396</a> headerFields.put(VERSION_FIELD_KEY, v);<a name="397" href="#397">397</a> headerFields.put(ABSOLUTE_OFFSET_KEY, <strong>new</strong> Long(offset));<a name="398" href="#398">398</a> <a name="399" href="#399">399</a> <strong>return</strong> <strong>new</strong> <a href="../../../../org/archive/io/arc/ARCRecordMetaData.html">ARCRecordMetaData</a>(getReaderIdentifier(), headerFields);<a name="400" href="#400">400</a> }<a name="401" href="#401">401</a> <a name="402" href="#402">402</a> <strong>protected</strong> <strong>boolean</strong> isDate(<strong>final</strong> String date) {<a name="403" href="#403">403</a> <strong>if</strong> (date.length() != 14) {<a name="404" href="#404">404</a> <strong>return</strong> false;<a name="405" href="#405">405</a> }<a name="406" href="#406">406</a> <strong>return</strong> isNumber(date);<a name="407" href="#407">407</a> }<a name="408" href="#408">408</a> <a name="409" href="#409">409</a> <strong>protected</strong> <strong>boolean</strong> isNumber(<strong>final</strong> String n) {<a name="410" href="#410">410</a> <strong>for</strong> (<strong>int</strong> i = 0; i < n.length(); i++) {<a name="411" href="#411">411</a> <strong>if</strong> (!Character.isDigit(n.charAt(i))) {<a name="412" href="#412">412</a> <strong>return</strong> false;<a name="413" href="#413">413</a> }<a name="414" href="#414">414</a> }<a name="415" href="#415">415</a> <strong>return</strong> <strong>true</strong>;<a name="416" href="#416">416</a> }<a name="417" href="#417">417</a> <a name="418" href="#418">418</a> <strong>protected</strong> <strong>boolean</strong> isLegitimateIPValue(<strong>final</strong> String ip) {<a name="419" href="#419">419</a> <strong>if</strong> (<span class="string">"-"</span>.equals(ip)) {<a name="420" href="#420">420</a> <strong>return</strong> <strong>true</strong>;<a name="421" href="#421">421</a> }<a name="422" href="#422">422</a> Matcher m = InetAddressUtil.IPV4_QUADS.matcher(ip);<a name="423" href="#423">423</a> <strong>return</strong> m != <strong>null</strong> && m.matches();<a name="424" href="#424">424</a> }<a name="425" href="#425">425</a> <a name="426" href="#426">426</a> <em>/**<em>*</em></em><a name="427" href="#427">427</a> <em> * Fix space in URLs.</em><a name="428" href="#428">428</a> <em> * The ARCWriter used to write into the ARC URLs with spaces in them.</em><a name="429" href="#429">429</a> <em> * See <a</em><a name="430" href="#430">430</a> <em> * href="https://sourceforge.net/tracker/?group_id=73833&atid=539099&func=detail&aid=1010966">[ 1010966 ]</em><a name="431" href="#431">431</a> <em> * crawl.log has URIs with spaces in them</a>.</em><a name="432" href="#432">432</a> <em> * This method does fix up on such headers converting all spaces found</em><a name="433" href="#433">433</a> <em> * to '%20'.</em><a name="434" href="#434">434</a> <em> * @param values List of metadata values.</em><a name="435" href="#435">435</a> <em> * @param requiredSize Expected size of resultant values list.</em><a name="436" href="#436">436</a> <em> * @return New list if we successfully fixed up values or original if</em><a name="437" href="#437">437</a> <em> * fixup failed.</em><a name="438" href="#438">438</a> <em> */</em><a name="439" href="#439">439</a> <strong>protected</strong> List<String> fixSpaceInURL(List<String> values, <strong>int</strong> requiredSize) {<a name="440" href="#440">440</a> <em class="comment">// Do validity check. 3rd from last is a date of 14 numeric</em><a name="441" href="#441">441</a> <em class="comment">// characters. The 4th from last is IP, all before the IP</em><a name="442" href="#442">442</a> <em class="comment">// should be concatenated together with a '%20' joiner.</em><a name="443" href="#443">443</a> <em class="comment">// In the below, '4' is 4th field from end which has the IP.</em><a name="444" href="#444">444</a> <strong>if</strong> (!(values.size() > requiredSize) || values.size() < 4) {
⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?