Package org.apache.nutch.util

Examples of org.apache.nutch.util.SuffixStringMatcher

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
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
    // handle missing config file
    if (reader == null) {
      if (LOG.isWarnEnabled()) {
        LOG.warn("Missing urlfilter.suffix.file, all URLs will be rejected!");
      }
      suffixes = new SuffixStringMatcher(new String[0]);
      modeAccept = false;
      ignoreCase = false;
      return;
    }
    BufferedReader in = new BufferedReader(reader);
    List aSuffixes = new ArrayList();
    boolean allow = false;
    boolean ignore = false;
    String line;

    while ((line = in.readLine()) != null) {
      if (line.length() == 0) continue;

      char first = line.charAt(0);
      switch (first) {
        case ' ':
        case '\n':
        case '#': // skip blank & comment lines
          break;
        case '-':
          allow = false;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        case '+':
          allow = true;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        default:
          aSuffixes.add(line);
      }
    }
    if (ignore) {
      for (int i = 0; i < aSuffixes.size(); i++) {
        aSuffixes.set(i, ((String) aSuffixes.get(i)).toLowerCase());
      }
    }
    suffixes = new SuffixStringMatcher(aSuffixes);
    modeAccept = allow;
    ignoreCase = ignore;
  }
View Full Code Here
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
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
    // handle missing config file
    if (reader == null) {
      if (LOG.isWarnEnabled()) {
        LOG.warn("Missing urlfilter.suffix.file, all URLs will be rejected!");
      }
      suffixes = new SuffixStringMatcher(new String[0]);
      modeAccept = false;
      ignoreCase = false;
      return;
    }
    BufferedReader in = new BufferedReader(reader);
    List<String> aSuffixes = new ArrayList<String>();
    boolean allow = false;
    boolean ignore = false;
    String line;

    while ((line = in.readLine()) != null) {
      if (line.length() == 0) continue;

      char first = line.charAt(0);
      switch (first) {
        case ' ':
        case '\n':
        case '#': // skip blank & comment lines
          break;
        case '-':
          allow = false;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        case '+':
          allow = true;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        default:
          aSuffixes.add(line);
      }
    }
    if (ignore) {
      for (int i = 0; i < aSuffixes.size(); i++) {
        aSuffixes.set(i, ((String) aSuffixes.get(i)).toLowerCase());
      }
    }
    suffixes = new SuffixStringMatcher(aSuffixes);
    modeAccept = allow;
    ignoreCase = ignore;
  }
View Full Code Here
161
162
163
164
165
166
167
168
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
201
202
203
204
205
206
207
208
    // handle missing config file
    if (reader == null) {
      if (LOG.isWarnEnabled()) {
        LOG.warn("Missing urlfilter.suffix.file, all URLs will be rejected!");
      }
      suffixes = new SuffixStringMatcher(new String[0]);
      modeAccept = false;
      ignoreCase = false;
      return;
    }
    BufferedReader in = new BufferedReader(reader);
    List aSuffixes = new ArrayList();
    boolean allow = false;
    boolean ignore = false;
    String line;

    while ((line = in.readLine()) != null) {
      if (line.length() == 0) continue;

      char first = line.charAt(0);
      switch (first) {
        case ' ':
        case '\n':
        case '#': // skip blank & comment lines
          break;
        case '-':
          allow = false;
          if (line.length() > 1 && line.charAt(1) == 'I')
            ignore = true;
          break;
        case '+':
          allow = true;
          if (line.length() > 1 && line.charAt(1) == 'I')
            ignore = true;
          break;
        default:
          aSuffixes.add(line);
      }
    }
    if (ignore) {
      for (int i = 0; i < aSuffixes.size(); i++) {
        aSuffixes.set(i, ((String) aSuffixes.get(i)).toLowerCase());
      }
    }
    suffixes = new SuffixStringMatcher(aSuffixes);
    modeAccept = allow;
    ignoreCase = ignore;
  }
View Full Code Here
160
161
162
163
164
165
166
167
168
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
201
202
203
204
205
206
207
    // handle missing config file
    if (reader == null) {
      if (LOG.isWarnEnabled()) {
        LOG.warn("Missing urlfilter.suffix.file, all URLs will be rejected!");
      }
      suffixes = new SuffixStringMatcher(new String[0]);
      modeAccept = false;
      ignoreCase = false;
      return;
    }
    BufferedReader in = new BufferedReader(reader);
    List aSuffixes = new ArrayList();
    boolean allow = false;
    boolean ignore = false;
    String line;

    while ((line = in.readLine()) != null) {
      if (line.length() == 0) continue;

      char first = line.charAt(0);
      switch (first) {
        case ' ':
        case '\n':
        case '#': // skip blank & comment lines
          break;
        case '-':
          allow = false;
          if (line.length() > 1 && line.charAt(1) == 'I')
            ignore = true;
          break;
        case '+':
          allow = true;
          if (line.length() > 1 && line.charAt(1) == 'I')
            ignore = true;
          break;
        default:
          aSuffixes.add(line);
      }
    }
    if (ignore) {
      for (int i = 0; i < aSuffixes.size(); i++) {
        aSuffixes.set(i, ((String) aSuffixes.get(i)).toLowerCase());
      }
    }
    suffixes = new SuffixStringMatcher(aSuffixes);
    modeAccept = allow;
    ignoreCase = ignore;
  }
View Full Code Here
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
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
    // handle missing config file
    if (reader == null) {
      if (LOG.isWarnEnabled()) {
        LOG.warn("Missing urlfilter.suffix.file, all URLs will be rejected!");
      }
      suffixes = new SuffixStringMatcher(new String[0]);
      modeAccept = false;
      ignoreCase = false;
      return;
    }
    BufferedReader in = new BufferedReader(reader);
    List aSuffixes = new ArrayList();
    boolean allow = false;
    boolean ignore = false;
    String line;

    while ((line = in.readLine()) != null) {
      if (line.length() == 0) continue;

      char first = line.charAt(0);
      switch (first) {
        case ' ':
        case '\n':
        case '#': // skip blank & comment lines
          break;
        case '-':
          allow = false;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        case '+':
          allow = true;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        default:
          aSuffixes.add(line);
      }
    }
    if (ignore) {
      for (int i = 0; i < aSuffixes.size(); i++) {
        aSuffixes.set(i, ((String) aSuffixes.get(i)).toLowerCase());
      }
    }
    suffixes = new SuffixStringMatcher(aSuffixes);
    modeAccept = allow;
    ignoreCase = ignore;
  }
View Full Code Here
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
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
    // handle missing config file
    if (reader == null) {
      if (LOG.isWarnEnabled()) {
        LOG.warn("Missing urlfilter.suffix.file, all URLs will be rejected!");
      }
      suffixes = new SuffixStringMatcher(new String[0]);
      modeAccept = false;
      ignoreCase = false;
      return;
    }
    BufferedReader in = new BufferedReader(reader);
    List aSuffixes = new ArrayList();
    boolean allow = false;
    boolean ignore = false;
    String line;

    while ((line = in.readLine()) != null) {
      if (line.length() == 0) continue;

      char first = line.charAt(0);
      switch (first) {
        case ' ':
        case '\n':
        case '#': // skip blank & comment lines
          break;
        case '-':
          allow = false;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        case '+':
          allow = true;
          if(line.contains("P"))
            filterFromPath = true;
          if(line.contains("I"))
            ignore = true;
          break;
        default:
          aSuffixes.add(line);
      }
    }
    if (ignore) {
      for (int i = 0; i < aSuffixes.size(); i++) {
        aSuffixes.set(i, ((String) aSuffixes.get(i)).toLowerCase());
      }
    }
    suffixes = new SuffixStringMatcher(aSuffixes);
    modeAccept = allow;
    ignoreCase = ignore;
  }
View Full Code Here
TOP

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

  • org.apache.nutch.urlfilter.suffix.SuffixURLFilter
  • java.util.Iterator
    • 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.