Examples of org.apache.nutch.net.URLNormalizers

org.apache.nutch.net.URLNormalizers
This class uses a "chained filter" pattern to run defined normalizers. Different lists of normalizers may be defined for different "scopes", or contexts where they are used (note however that they need to be activated first through plugin.include property).
There is one global scope defined by default, which consists of all active normalizers. The order in which these normalizers are executed may be defined in "urlnormalizer.order" property, which lists space-separated implementation classes (if this property is missing normalizers will be run in random order). If there are more normalizers activated than explicitly named on this list, the remaining ones will be run in random order after the ones specified on the list are executed.

You can define a set of contexts (or scopes) in which normalizers may be called. Each scope can have its own list of normalizers (defined in "urlnormalizer.scope." property) and its own order (defined in "urlnormalizer.order." property). If any of these properties are missing, default settings are used for the global scope.

In case no normalizers are required for any given scope, a org.apache.nutch.net.urlnormalizer.pass.PassURLNormalizer should be used.

Each normalizer may further select among many configurations, depending on the scope in which it is called, because the scope name is passed as a parameter to each normalizer. You can also use the same normalizer for many scopes.

Several scopes have been defined, and various Nutch tools will attempt using scope-specific normalizers first (and fall back to default config if scope-specific configuration is missing).

Normalizers may be run several times, to ensure that modifications introduced by normalizers at the end of the list can be further reduced by normalizers executed at the beginning. By default this loop is executed just once - if you want to ensure that all possible combinations have been applied you may want to run this loop up to the number of activated normalizers. This loop count can be configured through urlnormalizer.loop.count property. As soon as the url is unchanged the loop will stop and return the result.
@author Andrzej Bialecki

    if (!mode.equals(PARTITION_MODE_IP) && !mode.equals(PARTITION_MODE_DOMAIN)
        && !mode.equals(PARTITION_MODE_HOST)) {
      LOG.error("Unknown partition mode : " + mode + " - forcing to byHost");
      mode = PARTITION_MODE_HOST;
    }
    normalizers = new URLNormalizers(conf, URLNormalizers.SCOPE_PARTITION);
  }

View Full Code Here

    normalise = conf.getBoolean(GeneratorJob.GENERATOR_NORMALISE, true);
    if (filter) {
      filters = new URLFilters(conf);
    }
    if (normalise) {
      normalizers = new URLNormalizers(conf, URLNormalizers.SCOPE_GENERATE_HOST_COUNT);
    }
    maxDistance=conf.getInt("generate.max.distance", -1);
    curTime = conf.getLong(GeneratorJob.GENERATOR_CUR_TIME, System.currentTimeMillis());
    schedule = FetchScheduleFactory.getFetchSchedule(conf);
    scoringFilters = new ScoringFilters(conf);

View Full Code Here

    if (conf.getBoolean("segment.merger.filter", false)) {
      filters = new URLFilters(conf);
      mergeFilters = new SegmentMergeFilters(conf);
    }
    if (conf.getBoolean("segment.merger.normalizer", false))
      normalizers = new URLNormalizers(conf, URLNormalizers.SCOPE_DEFAULT);
    sliceSize = conf.getLong("segment.merger.slice", -1);
    if ((sliceSize > 0) && (LOG.isInfoEnabled())) {
      LOG.info("Slice size: " + sliceSize + " URLs.");
    }
  }

View Full Code Here


    normalize = job.getBoolean(URL_NORMALIZING, false);
    filter = job.getBoolean(URL_FILTERING, false);


    if (normalize) {
      urlNormalizers = new URLNormalizers(getConf(), URLNormalizers.SCOPE_DEFAULT);
    }


    if (filter) {
      urlFilters = new URLFilters(getConf());
    }

View Full Code Here

        byDomain = false;
      }
      if (GENERATOR_COUNT_VALUE_DOMAIN.equals(job.get(GENERATOR_COUNT_MODE))) byDomain = true;
      filters = new URLFilters(job);
      normalise = job.getBoolean(GENERATOR_NORMALISE, true);
      if (normalise) normalizers = new URLNormalizers(job,
          URLNormalizers.SCOPE_GENERATE_HOST_COUNT);
      scfilters = new ScoringFilters(job);
      partitioner.configure(job);
      filter = job.getBoolean(GENERATOR_FILTER, true);
      genDelay = job.getLong(GENERATOR_DELAY, 7L) * 3600L * 24L * 1000L;

View Full Code Here

    if (conf.getBoolean("segment.merger.filter", false)) {
      filters = new URLFilters(conf);
      mergeFilters = new SegmentMergeFilters(conf);
    }
    if (conf.getBoolean("segment.merger.normalizer", false))
      normalizers = new URLNormalizers(conf, URLNormalizers.SCOPE_DEFAULT);
    sliceSize = conf.getLong("segment.merger.slice", -1);
    if ((sliceSize > 0) && (LOG.isInfoEnabled())) {
      LOG.info("Slice size: " + sliceSize + " URLs.");
    }
  }

View Full Code Here

   * 
   */
  public void setConf(Configuration conf) {
    this.conf = conf;
    this.parserFactory = new ParserFactory(conf);
    this.normalizers = new URLNormalizers(conf, URLNormalizers.SCOPE_OUTLINK);
    this.filters = new URLFilters(conf);
    this.defaultEncoding =
      conf.get("parser.character.encoding.default", "windows-1252");
  }

View Full Code Here

    if (filter) {
      filters = new URLFilters(job);
    }
    if (normalize) {
      scope = job.get(URL_NORMALIZING_SCOPE, URLNormalizers.SCOPE_LINKDB);
      normalizers = new URLNormalizers(job, scope);
    }
  }

View Full Code Here

    if (urlFiltering) {
      filters = new URLFilters(job);
    }
    if (urlNormalizers) {
      scope = job.get(URL_NORMALIZING_SCOPE, URLNormalizers.SCOPE_CRAWLDB);
      normalizers = new URLNormalizers(job, scope);
    }
  }

View Full Code Here

    ignoreInternalLinks = job.getBoolean("db.ignore.internal.links", true);
    if (job.getBoolean(LinkDbFilter.URL_FILTERING, false)) {
      urlFilters = new URLFilters(job);
    }
    if (job.getBoolean(LinkDbFilter.URL_NORMALIZING, false)) {
      urlNormalizers = new URLNormalizers(job, URLNormalizers.SCOPE_LINKDB);
    }
  }

View Full Code Here

0 1 2 3 4 5 6 7

TOP

Related Classes of org.apache.nutch.net.URLNormalizers

org.apache.access.nutch.utils.UrlNormalizer

org.apache.nutch.crawl.CrawlDbFilter

org.apache.nutch.crawl.Generator$Selector

org.apache.nutch.crawl.GeneratorMapper

org.apache.nutch.crawl.InjectorJob$UrlMapper

org.apache.nutch.crawl.LinkDb

org.apache.nutch.crawl.LinkDbFilter

org.apache.nutch.crawl.PartitionUrlByHost

org.apache.nutch.crawl.URLPartitioner

org.apache.nutch.fetcher.FetcherReducer$FetcherThread

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.