Package org.apache.nutch.util

Examples of org.apache.nutch.util.ObjectCache

114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
  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
148
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
   *         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
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
  private ExtensionPoint extensionPoint;
  private ParsePluginList parsePluginList;

  public ParserFactory(Configuration conf) {
    this.conf = conf;
    ObjectCache objectCache = ObjectCache.get(conf);
    this.extensionPoint = PluginRepository.get(conf).getExtensionPoint(
        Parser.X_POINT_ID);
    this.parsePluginList = (ParsePluginList)objectCache.getObject(ParsePluginList.class.getName());
    if (this.parsePluginList == null) {
      this.parsePluginList = new ParsePluginsReader().parse(conf);
      objectCache.setObject(ParsePluginList.class.getName(), this.parsePluginList);
    }

    if (this.extensionPoint == null) {
      throw new RuntimeException("x point " + Parser.X_POINT_ID + " not found.");
    }
View Full Code Here
192
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
   * @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
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
  throws ParserNotFound {
   
    List<Parser> parsers = null;
    List<Extension> parserExts = null;
   
    ObjectCache objectCache = ObjectCache.get(conf);
   
    // TODO once the MimeTypes is available
    // parsers = getExtensions(MimeUtils.map(contentType));
    // if (parsers != null) {
    //   return parsers;
    // }
    // Last Chance: Guess content-type from file url...
    // parsers = getExtensions(MimeUtils.getMimeType(url));

    parserExts = getExtensions(contentType);
    if (parserExts == null) {
      throw new ParserNotFound(url, contentType);
    }

    parsers = new Vector<Parser>(parserExts.size());
    for (Iterator i=parserExts.iterator(); i.hasNext(); ){
      Extension ext = (Extension) i.next();
      Parser p = null;
      try {
        //check to see if we've cached this parser instance yet
        p = (Parser) objectCache.getObject(ext.getId());
        if (p == null) {
          // go ahead and instantiate it and then cache it
          p = (Parser) ext.getExtensionInstance();
          objectCache.setObject(ext.getId(),p);
        }
        parsers.add(p);
      } catch (PluginRuntimeException e) {
        if (LOG.isWarnEnabled()) {
          e.printStackTrace(LogUtil.getWarnStream(LOG));
View Full Code Here
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
  public Parser getParserById(String id) throws ParserNotFound {

    Extension[] extensions = this.extensionPoint.getExtensions();
    Extension parserExt = null;

    ObjectCache objectCache = ObjectCache.get(conf);
   
    if (id != null) {
      parserExt = getExtension(extensions, id);
    }
    if (parserExt == null) {
      parserExt = getExtensionFromAlias(extensions, id);
    }

    if (parserExt == null) {
      throw new ParserNotFound("No Parser Found for id [" + id + "]");
    }
   
    // first check the cache          
    if (objectCache.getObject(parserExt.getId()) != null) {
      return (Parser) objectCache.getObject(parserExt.getId());

    // if not found in cache, instantiate the Parser   
    } else {
      try {
        Parser p = (Parser) parserExt.getExtensionInstance();
        objectCache.setObject(parserExt.getId(), p);
        return p;
      } catch (PluginRuntimeException e) {
        if (LOG.isWarnEnabled()) {
          LOG.warn("Canno initialize parser " +
                   parserExt.getDescriptor().getPluginId() +
View Full Code Here
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
243
244
   * @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
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
  private ExtensionPoint extensionPoint;
  private ParsePluginList parsePluginList;

  public ParserFactory(Configuration conf) {
    this.conf = conf;
    ObjectCache objectCache = ObjectCache.get(conf);
    this.extensionPoint = PluginRepository.get(conf).getExtensionPoint(
        Parser.X_POINT_ID);
    this.parsePluginList = (ParsePluginList)objectCache.getObject(ParsePluginList.class.getName());
    if (this.parsePluginList == null) {
      this.parsePluginList = new ParsePluginsReader().parse(conf);
      objectCache.setObject(ParsePluginList.class.getName(), this.parsePluginList);
    }

    if (this.extensionPoint == null) {
      throw new RuntimeException("x point " + Parser.X_POINT_ID + " not found.");
    }
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
  throws ParserNotFound {
   
    List<Parser> parsers = null;
    List<Extension> parserExts = null;
   
    ObjectCache objectCache = ObjectCache.get(conf);
   
    // TODO once the MimeTypes is available
    // parsers = getExtensions(MimeUtils.map(contentType));
    // if (parsers != null) {
    //   return parsers;
    // }
    // Last Chance: Guess content-type from file url...
    // parsers = getExtensions(MimeUtils.getMimeType(url));

    parserExts = getExtensions(contentType);
    if (parserExts == null) {
      throw new ParserNotFound(url, contentType);
    }

    parsers = new Vector<Parser>(parserExts.size());
    for (Iterator i=parserExts.iterator(); i.hasNext(); ){
      Extension ext = (Extension) i.next();
      Parser p = null;
      try {
        //check to see if we've cached this parser instance yet
        p = (Parser) objectCache.getObject(ext.getId());
        if (p == null) {
          // go ahead and instantiate it and then cache it
          p = (Parser) ext.getExtensionInstance();
          objectCache.setObject(ext.getId(),p);
        }
        parsers.add(p);
      } catch (PluginRuntimeException e) {
        if (LOG.isWarnEnabled()) {
          LOG.warn("ParserFactory:PluginRuntimeException when "
View Full Code Here
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
  public Parser getParserById(String id) throws ParserNotFound {

    Extension[] extensions = this.extensionPoint.getExtensions();
    Extension parserExt = null;

    ObjectCache objectCache = ObjectCache.get(conf);
   
    if (id != null) {
      parserExt = getExtension(extensions, id);
    }
    if (parserExt == null) {
      parserExt = getExtensionFromAlias(extensions, id);
    }

    if (parserExt == null) {
      throw new ParserNotFound("No Parser Found for id [" + id + "]");
    }
   
    // first check the cache          
    if (objectCache.getObject(parserExt.getId()) != null) {
      return (Parser) objectCache.getObject(parserExt.getId());

    // if not found in cache, instantiate the Parser   
    } else {
      try {
        Parser p = (Parser) parserExt.getExtensionInstance();
        objectCache.setObject(parserExt.getId(), p);
        return p;
      } catch (PluginRuntimeException e) {
        if (LOG.isWarnEnabled()) {
          LOG.warn("Canno initialize parser " +
                   parserExt.getDescriptor().getPluginId() +
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.