Exemplo n.º 1
0
 public CrawlRuleTester(int crawlDepth, long crawlDelay, String baseUrl, ArchivalUnit au) {
   super("crawlrule tester");
   m_crawlDepth = crawlDepth;
   long minFetchDelay =
       CurrentConfig.getLongParam(
           BaseArchivalUnit.PARAM_MIN_FETCH_DELAY, BaseArchivalUnit.DEFAULT_MIN_FETCH_DELAY);
   m_crawlDelay = Math.max(crawlDelay, minFetchDelay);
   m_baseUrl = baseUrl;
   m_au = au;
 }
  protected RateLimiter recomputeFetchRateLimiter(RateLimiter oldLimiter) {
    String rate = CurrentConfig.getParam(PARAM_REGISTRY_FETCH_RATE, DEFAULT_REGISTRY_FETCH_RATE);
    Object limiterKey = getFetchRateLimiterKey();

    if (limiterKey == null) {
      return RateLimiter.getRateLimiter(oldLimiter, rate, DEFAULT_REGISTRY_FETCH_RATE);
    } else {
      RateLimiter.Pool pool = RateLimiter.getPool();
      return pool.findNamedRateLimiter(limiterKey, rate, DEFAULT_REGISTRY_FETCH_RATE);
    }
  }
  public void loadAuConfigDescrs(Configuration config) throws ConfigurationException {
    super.loadAuConfigDescrs(config);
    this.m_registryUrl = config.get(ConfigParamDescr.BASE_URL.getKey());
    // Now we can construct a valid CC permission checker.
    m_permissionCheckers =
        //       ListUtil.list(new CreativeCommonsPermissionChecker(m_registryUrl));
        ListUtil.list(new CreativeCommonsPermissionChecker());

    paramMap.putLong(
        KEY_AU_NEW_CONTENT_CRAWL_INTERVAL,
        CurrentConfig.getTimeIntervalParam(
            PARAM_REGISTRY_CRAWL_INTERVAL, DEFAULT_REGISTRY_CRAWL_INTERVAL));
    if (log.isDebug2()) {
      log.debug2(
          "Setting Registry AU recrawl interval to "
              + StringUtil.timeIntervalToString(
                  paramMap.getLong(KEY_AU_NEW_CONTENT_CRAWL_INTERVAL)));
    }
  }
 protected String getFetchRateLimiterSource() {
   return CurrentConfig.getParam(
       PARAM_REGISTRY_FETCH_RATE_LIMITER_SOURCE, DEFAULT_REGISTRY_FETCH_RATE_LIMITER_SOURCE);
 }
Exemplo n.º 5
0
  /** Explode the archive into its constituent elements */
  public void explode() throws CacheException {
    CachedUrl cachedUrl = null;
    int goodEntries = 0;
    int badEntries = 0;
    int ignoredEntries = 0;
    int entriesBetweenSleep = 0;
    ArchiveReader arcReader = null;

    logger.info(
        (storeArchive ? "Storing" : "Fetching") + " WARC file: " + origUrl + " will explode");
    try {
      if (storeArchive) {
        UrlCacher uc = au.makeUrlCacher(new UrlData(arcStream, arcProps, fetchUrl));
        BitSet bs = new BitSet();
        bs.set(UrlCacher.DONT_CLOSE_INPUT_STREAM_FLAG);
        uc.setFetchFlags(bs);
        uc.storeContent();
        archiveData.resetInputStream();
        arcStream = archiveData.input;
      }
      // Wrap it in an ArchiveReader
      logger.debug3("About to wrap stream");
      arcReader = wrapStream(fetchUrl, arcStream);
      logger.debug3("wrapStream() returns " + (arcReader == null ? "null" : "non-null"));
      // Explode it
      if (arcReader == null) {
        throw new CacheException.ExploderException("no WarcReader for " + origUrl);
      }
      ArchivalUnit au = crawlFacade.getAu();
      Set stemSet = new HashSet();
      logger.debug("Exploding " + fetchUrl);
      // Iterate through the elements in the WARC file, except the first
      Iterator i = arcReader.iterator();
      // Skip first record
      for (i.next(); i.hasNext(); ) {
        // XXX probably not necessary
        helper.pokeWDog();
        if ((++entriesBetweenSleep % sleepAfter) == 0) {
          long pauseTime =
              CurrentConfig.getTimeIntervalParam(PARAM_RETRY_PAUSE, DEFAULT_RETRY_PAUSE);
          Deadline pause = Deadline.in(pauseTime);
          logger.debug3("Sleeping for " + StringUtil.timeIntervalToString(pauseTime));
          while (!pause.expired()) {
            try {
              pause.sleep();
            } catch (InterruptedException ie) {
              // no action
            }
          }
        }
        ArchiveRecord element = (ArchiveRecord) i.next();
        // Each element is a URL to be cached in a suitable AU
        ArchiveRecordHeader elementHeader = element.getHeader();
        String elementUrl = elementHeader.getUrl();
        String elementMimeType = elementHeader.getMimetype();
        long elementLength = elementHeader.getLength();
        logger.debug2("WARC url " + elementUrl + " mime " + elementMimeType);
        if (elementUrl.startsWith("http:")) {
          ArchiveEntry ae =
              new ArchiveEntry(
                  elementUrl,
                  elementLength,
                  0, // XXX need to convert getDate string to long
                  element, // ArchiveRecord extends InputStream
                  this,
                  fetchUrl);
          ae.setHeaderFields(makeCIProperties(elementHeader));
          long bytesStored = elementLength;
          logger.debug3("ArchiveEntry: " + ae.getName() + " bytes " + bytesStored);
          try {
            helper.process(ae);
          } catch (PluginException ex) {
            throw new CacheException.ExploderException("helper.process() threw", ex);
          }
          if (ae.getBaseUrl() != null) {
            if (ae.getRestOfUrl() != null && ae.getHeaderFields() != null) {
              storeEntry(ae);
              handleAddText(ae);
              goodEntries++;
              crawlFacade.getCrawlerStatus().addContentBytesFetched(bytesStored);
            } else {
              ignoredEntries++;
            }
          } else {
            badEntries++;
            logger.debug2("Can't map " + elementUrl + " from " + archiveUrl);
          }
        }
      }
    } catch (IOException ex) {
      throw new CacheException.ExploderException(ex);
    } finally {
      if (arcReader != null)
        try {
          arcReader.close();
          arcReader = null;
        } catch (IOException ex) {
          throw new CacheException.ExploderException(ex);
        }
      if (cachedUrl != null) {
        cachedUrl.release();
      }
      IOUtil.safeClose(arcStream);
    }
    if (badEntries == 0 && goodEntries > 0) {
      // Make it look like a new crawl finished on each AU to which
      // URLs were added.
      for (Iterator it = touchedAus.iterator(); it.hasNext(); ) {
        ArchivalUnit au = (ArchivalUnit) it.next();
        logger.debug3(archiveUrl + " touching " + au.toString());
        AuUtil.getDaemon(au).getNodeManager(au).newContentCrawlFinished();
      }
    } else {
      ArchivalUnit au = crawlFacade.getAu();
      String msg = archiveUrl + ": " + badEntries + "/" + goodEntries + " bad entries";
      throw new CacheException.UnretryableException(msg);
    }
  }