Beispiel #1
0
 private void checkSubstance(ArchivalUnit au) {
   SubstanceChecker subChecker = new SubstanceChecker(au);
   if (!subChecker.isEnabled()) {
     errMsg = "No substance patterns defined for plugin.";
     return;
   }
   AuState auState = AuUtil.getAuState(au);
   SubstanceChecker.State oldState = auState.getSubstanceState();
   SubstanceChecker.State newState = subChecker.findSubstance();
   String chtxt = (newState == oldState ? "(unchanged)" : "(was " + oldState.toString() + ")");
   switch (newState) {
     case Unknown:
       log.error("Shouldn't happen: SubstanceChecker returned Unknown");
       errMsg = "Error in SubstanceChecker; see log.";
       break;
     case Yes:
       statusMsg = "AU has substance " + chtxt + ": " + au.getName();
       auState.setSubstanceState(SubstanceChecker.State.Yes);
       break;
     case No:
       statusMsg = "AU has no substance " + chtxt + ": " + au.getName();
       auState.setSubstanceState(SubstanceChecker.State.No);
       break;
   }
 }
Beispiel #2
0
  private boolean startReindexingMetadata(ArchivalUnit au, boolean force) {
    if (metadataMgr == null) {
      errMsg = "Metadata processing is not enabled.";
      return false;
    }

    if (!force) {
      if (!AuUtil.hasCrawled(au)) {
        errMsg = "Au has never crawled. Click again to reindex metadata";
        showForceReindexMetadata = true;
        return false;
      }

      AuState auState = AuUtil.getAuState(au);
      switch (auState.getSubstanceState()) {
        case No:
          errMsg = "Au has no substance. Click again to reindex metadata";
          showForceReindexMetadata = true;
          return false;
        case Unknown:
          errMsg = "Unknown substance for Au. Click again to reindex metadata.";
          showForceReindexMetadata = true;
          return false;
        case Yes:
          // fall through
      }
    }

    // Fully reindex metadata with the highest priority.
    Connection conn = null;
    PreparedStatement insertPendingAuBatchStatement = null;

    try {
      conn = dbMgr.getConnection();
      insertPendingAuBatchStatement = metadataMgr.getPrioritizedInsertPendingAuBatchStatement(conn);

      if (metadataMgr.enableAndAddAuToReindex(
          au, conn, insertPendingAuBatchStatement, false, true)) {
        statusMsg = "Reindexing metadata for " + au.getName();
        return true;
      }
    } catch (DbException dbe) {
      log.error("Cannot reindex metadata for " + au.getName(), dbe);
    } finally {
      DbManager.safeCloseStatement(insertPendingAuBatchStatement);
      DbManager.safeRollbackAndClose(conn);
    }

    if (force) {
      errMsg = "Still cannot reindex metadata for " + au.getName();
    } else {
      errMsg = "Cannot reindex metadata for " + au.getName();
    }
    return false;
  }
Beispiel #3
0
  private boolean disableMetadataIndexing(ArchivalUnit au, boolean force) {
    if (metadataMgr == null) {
      errMsg = "Metadata processing is not enabled.";
      return false;
    }

    try {
      metadataMgr.disableAuIndexing(au);
      statusMsg = "Disabled metadata indexing for " + au.getName();
      return true;
    } catch (Exception e) {
      errMsg = "Cannot reindex metadata for " + au.getName() + ": " + e.getMessage();
      return false;
    }
  }
Beispiel #4
0
 private void crawlPluginRegistries() {
   StringBuilder sb = new StringBuilder();
   for (ArchivalUnit au : pluginMgr.getAllRegistryAus()) {
     sb.append(au.getName());
     sb.append(": ");
     try {
       startCrawl(au, true, false);
       sb.append("Queued.");
     } catch (CrawlManagerImpl.NotEligibleException e) {
       sb.append("Failed: ");
       sb.append(e.getMessage());
     }
     sb.append("\n");
   }
   statusMsg = sb.toString();
 }
Beispiel #5
0
  private boolean startCrawl(ArchivalUnit au, boolean force, boolean deep)
      throws CrawlManagerImpl.NotEligibleException {
    CrawlManagerImpl cmi = (CrawlManagerImpl) crawlMgr;
    if (force) {
      RateLimiter limit = cmi.getNewContentRateLimiter(au);
      if (!limit.isEventOk()) {
        limit.unevent();
      }
    }
    cmi.checkEligibleToQueueNewContentCrawl(au);
    String delayMsg = "";
    String deepMsg = "";
    try {
      cmi.checkEligibleForNewContentCrawl(au);
    } catch (CrawlManagerImpl.NotEligibleException e) {
      delayMsg = ", Start delayed due to: " + e.getMessage();
    }
    Configuration config = ConfigManager.getCurrentConfig();
    int pri = config.getInt(PARAM_CRAWL_PRIORITY, DEFAULT_CRAWL_PRIORITY);

    CrawlReq req;
    try {
      req = new CrawlReq(au);
      req.setPriority(pri);
      if (deep) {
        int d = Integer.parseInt(formDepth);
        if (d < 0) {
          errMsg = "Illegal refetch depth: " + d;
          return false;
        }
        req.setRefetchDepth(d);
        deepMsg = "Deep (" + req.getRefetchDepth() + ") ";
      }
    } catch (NumberFormatException e) {
      errMsg = "Illegal refetch depth: " + formDepth;
      return false;
    } catch (RuntimeException e) {
      log.error("Couldn't create CrawlReq: " + au, e);
      errMsg = "Couldn't create CrawlReq: " + e.toString();
      return false;
    }
    cmi.startNewContentCrawl(req, null);
    statusMsg = deepMsg + "Crawl requested for " + au.getName() + delayMsg;
    return true;
  }
Beispiel #6
0
 private void callV3ContentPoll(ArchivalUnit au) throws PollManager.NotEligibleException {
   log.debug("Enqueuing a V3 Content Poll on " + au.getName());
   PollSpec spec = new PollSpec(au.getAuCachedUrlSet(), Poll.V3_POLL);
   pollManager.enqueueHighPriorityPoll(au, spec);
   statusMsg = "Enqueued V3 poll for " + au.getName();
 }
Beispiel #7
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);
    }
  }