Package org.apache.nutch.util.domain

Examples of org.apache.nutch.util.domain.DomainSuffixes

82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
   *  getDomainName(conf, new URL(http://lucene.apache.org/))
   *  </code><br>
   *  will return <br><code> apache.org</code>
   *   */
  public static String getDomainName(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    //it seems that java returns hostnames ending with .
    if(host.endsWith("."))
      host = host.substring(0, host.length() - 1);
    if(IP_PATTERN.matcher(host).matches())
      return host;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      if(tlds.isDomainSuffix(subCandidate)) {
        return candidate;
      }
      candidate = subCandidate;
    }
    return candidate;
View Full Code Here
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
  /** Returns the {@link DomainSuffix} corresponding to the
   * last public part of the hostname
   */
  public static DomainSuffix getDomainSuffix(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    if(IP_PATTERN.matcher(host).matches())
      return null;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      DomainSuffix d = tlds.get(subCandidate);
      if(d != null) {
        return d;
      }
      candidate = subCandidate;
    }
View Full Code Here
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
   *  getDomainName(conf, new URL(http://lucene.apache.org/))
   *  </code><br>
   *  will return <br><code> apache.org</code>
   *   */
  public static String getDomainName(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    //it seems that java returns hostnames ending with .
    if(host.endsWith("."))
      host = host.substring(0, host.length() - 1);
    if(IP_PATTERN.matcher(host).matches())
      return host;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      if(tlds.isDomainSuffix(subCandidate)) {
        return candidate;
      }
      candidate = subCandidate;
    }
    return candidate;
View Full Code Here
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
  /** Returns the {@link DomainSuffix} corresponding to the
   * last public part of the hostname
   */
  public static DomainSuffix getDomainSuffix(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    if(IP_PATTERN.matcher(host).matches())
      return null;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      DomainSuffix d = tlds.get(subCandidate);
      if(d != null) {
        return d;
      }
      candidate = subCandidate;
    }
View Full Code Here
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
   *  getDomainName(conf, new URL(http://lucene.apache.org/))
   *  </code><br>
   *  will return <br><code> apache.org</code>
   *   */
  public static String getDomainName(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    //it seems that java returns hostnames ending with .
    if(host.endsWith("."))
      host = host.substring(0, host.length() - 1);
    if(IP_PATTERN.matcher(host).matches())
      return host;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      if(tlds.isDomainSuffix(subCandidate)) {
        return candidate;
      }
      candidate = subCandidate;
    }
    return candidate;
View Full Code Here
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
  /** Returns the {@link DomainSuffix} corresponding to the
   * last public part of the hostname
   */
  public static DomainSuffix getDomainSuffix(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    if(IP_PATTERN.matcher(host).matches())
      return null;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      DomainSuffix d = tlds.get(subCandidate);
      if(d != null) {
        return d;
      }
      candidate = subCandidate;
    }
View Full Code Here
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
   *  getDomainName(conf, new URL(http://lucene.apache.org/))
   *  </code><br>
   *  will return <br><code> apache.org</code>
   *   */
  public static String getDomainName(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    //it seems that java returns hostnames ending with .
    if(host.endsWith("."))
      host = host.substring(0, host.length() - 1);
    if(IP_PATTERN.matcher(host).matches())
      return host;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      if(tlds.isDomainSuffix(subCandidate)) {
        return candidate;
      }
      candidate = subCandidate;
    }
    return candidate;
View Full Code Here
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
  /** Returns the {@link DomainSuffix} corresponding to the
   * last public part of the hostname
   */
  public static DomainSuffix getDomainSuffix(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    if(IP_PATTERN.matcher(host).matches())
      return null;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      DomainSuffix d = tlds.get(subCandidate);
      if(d != null) {
        return d;
      }
      candidate = subCandidate;
    }
View Full Code Here
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
   *  getDomainName(conf, new URL(http://lucene.apache.org/))
   *  </code><br>
   *  will return <br><code> apache.org</code>
   *   */
  public static String getDomainName(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    //it seems that java returns hostnames ending with .
    if(host.endsWith("."))
      host = host.substring(0, host.length() - 1);
    if(IP_PATTERN.matcher(host).matches())
      return host;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      if(tlds.isDomainSuffix(subCandidate)) {
        return candidate;
      }
      candidate = subCandidate;
    }
    return candidate;
View Full Code Here
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
  /** Returns the {@link DomainSuffix} corresponding to the
   * last public part of the hostname
   */
  public static DomainSuffix getDomainSuffix(URL url) {
    DomainSuffixes tlds = DomainSuffixes.getInstance();
    String host = url.getHost();
    if(IP_PATTERN.matcher(host).matches())
      return null;
   
    int index = 0;
    String candidate = host;
    for(;index >= 0;) {
      index = candidate.indexOf('.');
      String subCandidate = candidate.substring(index+1);
      DomainSuffix d = tlds.get(subCandidate);
      if(d != null) {
        return d;
      }
      candidate = subCandidate;
    }
View Full Code Here
TOP

Related Classes of org.apache.nutch.util.domain.DomainSuffixes

  • org.apache.nutch.util.URLUtil
  • java.io.InputStream
    • 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.