Package org.apache.nutch.util

Examples of org.apache.nutch.util.ObjectCache

212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
   * @return a list of extensions to be used for this contentType.
   *         If none, returns <code>null</code>.
   */
  protected List<Extension> getExtensions(String contentType) {
   
    ObjectCache objectCache = ObjectCache.get(conf);
    // First of all, tries to clean the content-type
    String type = null;
    type = MimeUtil.cleanMimeType(contentType);


    List<Extension> extensions = (List<Extension>) objectCache.getObject(type);

    // Just compare the reference:
    // if this is the empty list, we know we will find no extension.
    if (extensions == EMPTY_EXTENSION_LIST) {
      return null;
    }
   
    if (extensions == null) {
      extensions = findExtensions(type);
      if (extensions != null) {
        objectCache.setObject(type, extensions);
      } else {
        // Put the empty extension list into cache
        // to remember we don't know any related extension.
        objectCache.setObject(type, EMPTY_EXTENSION_LIST);
      }
    }
    return extensions;
  }
View Full Code Here
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
    }
  }
 
  public static CollectionManager getCollectionManager(Configuration conf) {
    String key = "collectionmanager";
    ObjectCache objectCache = ObjectCache.get(conf);
    CollectionManager impl = (CollectionManager)objectCache.getObject(key);
    if (impl == null) {
      try {
        if (LOG.isInfoEnabled()) {
          LOG.info("Instantiating CollectionManager");
        }
        impl=new CollectionManager(conf);
        objectCache.setObject(key,impl);
      } catch (Exception e) {
        throw new RuntimeException("Couldn't create CollectionManager", e);
      }
    }
    return impl;
View Full Code Here
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
  private ScoringFilter[] filters;

  public ScoringFilters(Configuration conf) {
    super(conf);
    ObjectCache objectCache = ObjectCache.get(conf);
    String order = conf.get("scoring.filter.order");
    this.filters = (ScoringFilter[]) objectCache.getObject(ScoringFilter.class.getName());

    if (this.filters == null) {
      String[] orderedFilters = null;
      if (order != null && !order.trim().equals("")) {
        orderedFilters = order.trim().split("\\s+");
      }

      try {
        ExtensionPoint point = PluginRepository.get(conf).getExtensionPoint(ScoringFilter.X_POINT_ID);
        if (point == null) throw new RuntimeException(ScoringFilter.X_POINT_ID + " not found.");
        Extension[] extensions = point.getExtensions();
        HashMap<String, ScoringFilter> filterMap =
          new HashMap<String, ScoringFilter>();
        for (int i = 0; i < extensions.length; i++) {
          Extension extension = extensions[i];
          ScoringFilter filter = (ScoringFilter) extension.getExtensionInstance();
          if (!filterMap.containsKey(filter.getClass().getName())) {
            filterMap.put(filter.getClass().getName(), filter);
          }
        }
        if (orderedFilters == null) {
          objectCache.setObject(ScoringFilter.class.getName(), filterMap.values().toArray(new ScoringFilter[0]));
        } else {
          ScoringFilter[] filter = new ScoringFilter[orderedFilters.length];
          for (int i = 0; i < orderedFilters.length; i++) {
            filter[i] = filterMap.get(orderedFilters[i]);
          }
          objectCache.setObject(ScoringFilter.class.getName(), filter);
        }
      } catch (PluginRuntimeException e) {
        throw new RuntimeException(e);
      }
      this.filters = (ScoringFilter[]) objectCache.getObject(ScoringFilter.class.getName());
    }
  }
View Full Code Here
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
  public static final String URLFILTER_ORDER = "urlfilter.order";
  private URLFilter[] filters;

  public URLFilters(Configuration conf) {
    String order = conf.get(URLFILTER_ORDER);
    ObjectCache objectCache = ObjectCache.get(conf);
    this.filters = (URLFilter[]) objectCache.getObject(URLFilter.class.getName());

    if (this.filters == null) {
      String[] orderedFilters = null;
      if (order != null && !order.trim().equals("")) {
        orderedFilters = order.trim().split("\\s+");
      }

      try {
        ExtensionPoint point = PluginRepository.get(conf).getExtensionPoint(
            URLFilter.X_POINT_ID);
        if (point == null)
          throw new RuntimeException(URLFilter.X_POINT_ID + " not found.");
        Extension[] extensions = point.getExtensions();
        Map<String, URLFilter> filterMap = new HashMap<String, URLFilter>();
        for (int i = 0; i < extensions.length; i++) {
          Extension extension = extensions[i];
          URLFilter filter = (URLFilter) extension.getExtensionInstance();
          if (!filterMap.containsKey(filter.getClass().getName())) {
            filterMap.put(filter.getClass().getName(), filter);
          }
        }
        if (orderedFilters == null) {
          objectCache.setObject(URLFilter.class.getName(), filterMap.values().toArray(
              new URLFilter[0]));
        } else {
          ArrayList<URLFilter> filters = new ArrayList<URLFilter>();
          for (int i = 0; i < orderedFilters.length; i++) {
            URLFilter filter = filterMap.get(orderedFilters[i]);
            if (filter != null) {
              filters.add(filter);
            }
          }
          objectCache.setObject(URLFilter.class.getName(), filters
              .toArray(new URLFilter[filters.size()]));
        }
      } catch (PluginRuntimeException e) {
        throw new RuntimeException(e);
      }
      this.filters = (URLFilter[]) objectCache.getObject(URLFilter.class.getName());
    }
  }
View Full Code Here
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
  public URLNormalizers(Configuration conf, String scope) {
    this.conf = conf;
    this.extensionPoint = PluginRepository.get(conf).getExtensionPoint(
            URLNormalizer.X_POINT_ID);
    ObjectCache objectCache = ObjectCache.get(conf);
   
    if (this.extensionPoint == null) {
      throw new RuntimeException("x point " + URLNormalizer.X_POINT_ID
              + " not found.");
    }

    normalizers = (URLNormalizer[])objectCache.getObject(URLNormalizer.X_POINT_ID + "_" + scope);
    if (normalizers == null) {
      normalizers = getURLNormalizers(scope);
    }
    if (normalizers == EMPTY_NORMALIZERS) {
      normalizers = (URLNormalizer[])objectCache.getObject(URLNormalizer.X_POINT_ID + "_" + SCOPE_DEFAULT);
      if (normalizers == null) {
        normalizers = getURLNormalizers(SCOPE_DEFAULT);
      }
    }
   
View Full Code Here
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
   *         scope.
   * @throws PluginRuntimeException
   */
  URLNormalizer[] getURLNormalizers(String scope) {
    List<Extension> extensions = getExtensions(scope);
    ObjectCache objectCache = ObjectCache.get(conf);
   
    if (extensions == EMPTY_EXTENSION_LIST) {
      return EMPTY_NORMALIZERS;
    }
   
    List<URLNormalizer> normalizers = new Vector<URLNormalizer>(extensions.size());

    Iterator<Extension> it = extensions.iterator();
    while (it.hasNext()) {
      Extension ext = it.next();
      URLNormalizer normalizer = null;
      try {
        // check to see if we've cached this URLNormalizer instance yet
        normalizer = (URLNormalizer) objectCache.getObject(ext.getId());
        if (normalizer == null) {
          // go ahead and instantiate it and then cache it
          normalizer = (URLNormalizer) ext.getExtensionInstance();
          objectCache.setObject(ext.getId(), normalizer);
        }
        normalizers.add(normalizer);
      } catch (PluginRuntimeException e) {
        e.printStackTrace();
        LOG.warn("URLNormalizers:PluginRuntimeException when "
View Full Code Here
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
   * @return a list of extensions to be used for this scope. If none, returns
   *         empty list.
   * @throws PluginRuntimeException
   */
  private List<Extension> getExtensions(String scope) {
    ObjectCache objectCache = ObjectCache.get(conf);
    List<Extension> extensions =
      (List<Extension>) objectCache.getObject(URLNormalizer.X_POINT_ID + "_x_"
                                                + scope);

    // Just compare the reference:
    // if this is the empty list, we know we will find no extension.
    if (extensions == EMPTY_EXTENSION_LIST) {
      return EMPTY_EXTENSION_LIST;
    }

    if (extensions == null) {
      extensions = findExtensions(scope);
      if (extensions != null) {
        objectCache.setObject(URLNormalizer.X_POINT_ID + "_x_" + scope, extensions);
      } else {
        // Put the empty extension list into cache
        // to remember we don't know any related extension.
        objectCache.setObject(URLNormalizer.X_POINT_ID + "_x_" + scope, EMPTY_EXTENSION_LIST);
        extensions = EMPTY_EXTENSION_LIST;
      }
    }
    return extensions;
  }
View Full Code Here
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
  private IndexingFilter[] indexingFilters;

  public IndexingFilters(Configuration conf) {
    /* Get indexingfilter.order property */
    String order = conf.get(INDEXINGFILTER_ORDER);
    ObjectCache objectCache = ObjectCache.get(conf);
    this.indexingFilters = (IndexingFilter[]) objectCache
        .getObject(IndexingFilter.class.getName());
    if (this.indexingFilters == null) {
      /*
       * If ordered filters are required, prepare array of filters based on
       * property
       */
      String[] orderedFilters = null;
      if (order != null && !order.trim().equals("")) {
        orderedFilters = order.trim().split("\\s+");
      }
      try {
        ExtensionPoint point = PluginRepository.get(conf).getExtensionPoint(
            IndexingFilter.X_POINT_ID);
        if (point == null)
          throw new RuntimeException(IndexingFilter.X_POINT_ID + " not found.");
        Extension[] extensions = point.getExtensions();
        HashMap<String, IndexingFilter> filterMap =
          new HashMap<String, IndexingFilter>();
        for (int i = 0; i < extensions.length; i++) {
          Extension extension = extensions[i];
          IndexingFilter filter = (IndexingFilter) extension
              .getExtensionInstance();
          LOG.info("Adding " + filter.getClass().getName());
          if (!filterMap.containsKey(filter.getClass().getName())) {
            filterMap.put(filter.getClass().getName(), filter);
          }
        }
        /*
         * If no ordered filters required, just get the filters in an
         * indeterminate order
         */
        if (orderedFilters == null) {
          objectCache.setObject(IndexingFilter.class.getName(),
              filterMap.values().toArray(
                  new IndexingFilter[0]));
          /* Otherwise run the filters in the required order */
        } else {
          ArrayList<IndexingFilter> filters = new ArrayList<IndexingFilter>();
          for (int i = 0; i < orderedFilters.length; i++) {
            IndexingFilter filter = filterMap
                .get(orderedFilters[i]);
            if (filter != null) {
              filters.add(filter);
            }
          }
          objectCache.setObject(IndexingFilter.class.getName(), filters
              .toArray(new IndexingFilter[filters.size()]));
        }
      } catch (PluginRuntimeException e) {
        throw new RuntimeException(e);
      }
      this.indexingFilters = (IndexingFilter[]) objectCache
          .getObject(IndexingFilter.class.getName());
    }
  }                 
View Full Code Here
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
 
  public static final String HTMLPARSEFILTER_ORDER = "htmlparsefilter.order";

  public HtmlParseFilters(Configuration conf) {
        String order = conf.get(HTMLPARSEFILTER_ORDER);
        ObjectCache objectCache = ObjectCache.get(conf);
        this.htmlParseFilters = (HtmlParseFilter[]) objectCache.getObject(HtmlParseFilter.class.getName());
        if (htmlParseFilters == null) {
          /*
           * If ordered filters are required, prepare array of filters based on
           * property
           */
          String[] orderedFilters = null;
          if (order != null && !order.trim().equals("")) {
            orderedFilters = order.trim().split("\\s+");
          }
            HashMap<String, HtmlParseFilter> filterMap =
              new HashMap<String, HtmlParseFilter>();
            try {
                ExtensionPoint point = PluginRepository.get(conf).getExtensionPoint(HtmlParseFilter.X_POINT_ID);
                if (point == null)
                    throw new RuntimeException(HtmlParseFilter.X_POINT_ID + " not found.");
                Extension[] extensions = point.getExtensions();
                for (int i = 0; i < extensions.length; i++) {
                    Extension extension = extensions[i];
                    HtmlParseFilter parseFilter = (HtmlParseFilter) extension.getExtensionInstance();
                    if (!filterMap.containsKey(parseFilter.getClass().getName())) {
                        filterMap.put(parseFilter.getClass().getName(), parseFilter);
                    }
                }
                HtmlParseFilter[] htmlParseFilters = filterMap.values().toArray(new HtmlParseFilter[filterMap.size()]);
                /*
                 * If no ordered filters required, just get the filters in an
                 * indeterminate order
                 */
                if (orderedFilters == null) {
                  objectCache.setObject(HtmlParseFilter.class.getName(), htmlParseFilters);
                }
                /* Otherwise run the filters in the required order */
                else {
                  ArrayList<HtmlParseFilter> filters = new ArrayList<HtmlParseFilter>();
                  for (int i = 0; i < orderedFilters.length; i++) {
                    HtmlParseFilter filter = filterMap
                        .get(orderedFilters[i]);
                    if (filter != null) {
                      filters.add(filter);
                    }
                  }
                  objectCache.setObject(HtmlParseFilter.class.getName(), filters
                      .toArray(new HtmlParseFilter[filters.size()]));
                }
            } catch (PluginRuntimeException e) {
                throw new RuntimeException(e);
            }
            this.htmlParseFilters = (HtmlParseFilter[]) objectCache.getObject(HtmlParseFilter.class.getName());
        }
    }                 
View Full Code Here
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
  private ScoringFilter[] filters;

  public ScoringFilters(Configuration conf) {
    super(conf);
    ObjectCache objectCache = ObjectCache.get(conf);
    String order = conf.get("scoring.filter.order");
    this.filters = (ScoringFilter[]) objectCache.getObject(ScoringFilter.class.getName());

    if (this.filters == null) {
      String[] orderedFilters = null;
      if (order != null && !order.trim().equals("")) {
        orderedFilters = order.split("\\s+");
      }

      try {
        ExtensionPoint point = PluginRepository.get(conf).getExtensionPoint(ScoringFilter.X_POINT_ID);
        if (point == null) throw new RuntimeException(ScoringFilter.X_POINT_ID + " not found.");
        Extension[] extensions = point.getExtensions();
        HashMap<String, ScoringFilter> filterMap =
          new HashMap<String, ScoringFilter>();
        for (int i = 0; i < extensions.length; i++) {
          Extension extension = extensions[i];
          ScoringFilter filter = (ScoringFilter) extension.getExtensionInstance();
          if (!filterMap.containsKey(filter.getClass().getName())) {
            filterMap.put(filter.getClass().getName(), filter);
          }
        }
        if (orderedFilters == null) {
          objectCache.setObject(ScoringFilter.class.getName(), filterMap.values().toArray(new ScoringFilter[0]));
        } else {
          ScoringFilter[] filter = new ScoringFilter[orderedFilters.length];
          for (int i = 0; i < orderedFilters.length; i++) {
            filter[i] = filterMap.get(orderedFilters[i]);
          }
          objectCache.setObject(ScoringFilter.class.getName(), filter);
        }
      } catch (PluginRuntimeException e) {
        throw new RuntimeException(e);
      }
      this.filters = (ScoringFilter[]) objectCache.getObject(ScoringFilter.class.getName());
    }
  }
View Full Code Here
TOP

Related Classes of org.apache.nutch.util.ObjectCache

  • org.apache.nutch.analysis.AnalyzerFactory
  • org.apache.nutch.analysis.CommonGrams
  • org.apache.nutch.collection.CollectionManager
  • org.apache.nutch.crawl.FetchScheduleFactory
  • org.apache.nutch.crawl.SignatureFactory
  • org.apache.nutch.indexer.field.FieldFilters
  • org.apache.nutch.indexer.IndexingFilters
  • org.apache.nutch.indexer.IndexWriters
  • org.apache.nutch.indexer.solr.SolrMappingReader
  • org.apache.nutch.indexwriter.solr.SolrMappingReader
    • Copyright © 2018 www.massapicom. All rights reserved.
    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.