Package org.apache.nutch.parse.rss.structs

Examples of org.apache.nutch.parse.rss.structs.RSSChannel

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
140
141
142
143
144
145
146
147
148
149
150
151
152
153
        // also get the outlinks

        if (theRSSChannels != null) {
            for (int i = 0; i < theRSSChannels.size(); i++) {
                RSSChannel r = (RSSChannel) theRSSChannels.get(i);
                contentTitle.append(r.getTitle());
                contentTitle.append(" ");

                // concat the description to the index text
                indexText.append(r.getDescription());
                indexText.append(" ");

                if (r.getLink() != null) {
                    try {
                        // get the outlink
      if (r.getDescription()!= null ) {
          theOutlinks.add(new Outlink(r.getLink(), r.getDescription(), getConf()));
      } else {
          theOutlinks.add(new Outlink(r.getLink(), "", getConf()));
      }
                    } catch (MalformedURLException e) {
                        if (LOG.isWarnEnabled()) {
                          LOG.warn("MalformedURL: " + r.getLink());
                          LOG.warn("Attempting to continue processing outlinks");
                          e.printStackTrace(LogUtil.getWarnStream(LOG));
                        }
                        continue;
                    }
                }

                // now get the descriptions of all the underlying RSS Items and
                // then index them too
                for (int j = 0; j < r.getItems().size(); j++) {
                    RSSItem theRSSItem = (RSSItem) r.getItems().get(j);
                    indexText.append(theRSSItem.getDescription());
                    indexText.append(" ");

                    String whichLink = null;
View Full Code Here
92
93
94
95
96
97
98
99
            fRssChannels.add(fCurrentChannel);
        }

        //System.out.println("Found a new channel: " + title);

        fCurrentChannel = new RSSChannel(title, link, description);

    }
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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
        // also get the outlinks

        if (theRSSChannels != null) {
            for (int i = 0; i < theRSSChannels.size(); i++) {
                RSSChannel r = (RSSChannel) theRSSChannels.get(i);
                contentTitle.append(r.getTitle());
                contentTitle.append(" ");

                // concat the description to the index text
                indexText.append(r.getDescription());
                indexText.append(" ");

                if (r.getLink() != null) {
                    try {
                        // get the outlink
      if (r.getDescription()!= null ) {
          theOutlinks.add(new Outlink(r.getLink(), r.getDescription()));
      } else {
          theOutlinks.add(new Outlink(r.getLink(), ""));
      }
                    } catch (MalformedURLException e) {
                        if (LOG.isWarnEnabled()) {
                          LOG.warn("MalformedURL: " + r.getLink());
                          LOG.warn("Attempting to continue processing outlinks");
                          e.printStackTrace(LogUtil.getWarnStream(LOG));
                        }
                        continue;
                    }
                }

                // now get the descriptions of all the underlying RSS Items and
                // then index them too
                for (int j = 0; j < r.getItems().size(); j++) {
                    RSSItem theRSSItem = (RSSItem) r.getItems().get(j);
                    indexText.append(theRSSItem.getDescription());
                    indexText.append(" ");

                    String whichLink = null;
View Full Code Here
144
145
146
147
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
177
178
179
180
181
182
        // also get the outlinks

        if (theRSSChannels != null) {
            for (int i = 0; i < theRSSChannels.size(); i++) {
                RSSChannel r = (RSSChannel) theRSSChannels.get(i);
                contentTitle.append(r.getTitle());
                contentTitle.append(" ");

                // concat the description to the index text
                indexText.append(r.getDescription());
                indexText.append(" ");

                if (r.getLink() != null) {
                    try {
                        // get the outlink
      if (r.getDescription()!= null ) {
          theOutlinks.add(new Outlink(r.getLink(), r.getDescription()));
      } else {
          theOutlinks.add(new Outlink(r.getLink(), ""));
      }
                    } catch (MalformedURLException e) {
                        LOG.info("nutch:parse-rss:RSSParser Exception: MalformedURL: "
                                        + r.getLink()
                                        + ": Attempting to continue processing outlinks");
                        e.printStackTrace();
                        continue;
                    }
                }

                // now get the descriptions of all the underlying RSS Items and
                // then index them too
                for (int j = 0; j < r.getItems().size(); j++) {
                    RSSItem theRSSItem = (RSSItem) r.getItems().get(j);
                    indexText.append(theRSSItem.getDescription());
                    indexText.append(" ");

                    String whichLink = null;
View Full Code Here
92
93
94
95
96
97
98
99
            fRssChannels.add(fCurrentChannel);
        }

        //System.out.println("Found a new channel: " + title);

        fCurrentChannel = new RSSChannel(title, link, description);

    }
View Full Code Here
144
145
146
147
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
177
178
179
180
        // also get the outlinks

        if (theRSSChannels != null) {
            for (int i = 0; i < theRSSChannels.size(); i++) {
                RSSChannel r = (RSSChannel) theRSSChannels.get(i);
                contentTitle.append(r.getTitle());
                contentTitle.append(" ");

                // concat the description to the index text
                indexText.append(r.getDescription());
                indexText.append(" ");

                if (r.getLink() != null) {
                    try {
                        // get the outlink
                        theOutlinks.add(new Outlink(r.getLink(), r
                                .getDescription()));
                    } catch (MalformedURLException e) {
                        LOG
                                .info("nutch:parse-rss:RSSParser Exception: MalformedURL: "
                                        + r.getLink()
                                        + ": Attempting to continue processing outlinks");
                        e.printStackTrace();
                        continue;
                    }
                }

                // now get the descriptions of all the underlying RSS Items and
                // then index them too
                for (int j = 0; j < r.getItems().size(); j++) {
                    RSSItem theRSSItem = (RSSItem) r.getItems().get(j);
                    indexText.append(theRSSItem.getDescription());
                    indexText.append(" ");

                    String whichLink = null;
View Full Code Here
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
140
141
142
143
144
145
146
147
148
149
150
151
152
        // also get the outlinks

        if (theRSSChannels != null) {
            for (int i = 0; i < theRSSChannels.size(); i++) {
                RSSChannel r = (RSSChannel) theRSSChannels.get(i);
                contentTitle.append(r.getTitle());
                contentTitle.append(" ");

                // concat the description to the index text
                indexText.append(r.getDescription());
                indexText.append(" ");

                if (r.getLink() != null) {
                    try {
                        // get the outlink
      if (r.getDescription()!= null ) {
          theOutlinks.add(new Outlink(r.getLink(), r.getDescription(), getConf()));
      } else {
          theOutlinks.add(new Outlink(r.getLink(), "", getConf()));
      }
                    } catch (MalformedURLException e) {
                        if (LOG.isWarnEnabled()) {
                          LOG.warn("MalformedURL: " + r.getLink());
                          LOG.warn("Attempting to continue processing outlinks");
                          e.printStackTrace(LogUtil.getWarnStream(LOG));
                        }
                        continue;
                    }
                }

                // now get the descriptions of all the underlying RSS Items and
                // then index them too
                for (int j = 0; j < r.getItems().size(); j++) {
                    RSSItem theRSSItem = (RSSItem) r.getItems().get(j);
                    indexText.append(theRSSItem.getDescription());
                    indexText.append(" ");

                    String whichLink = null;
View Full Code Here
91
92
93
94
95
96
97
98
            fRssChannels.add(fCurrentChannel);
        }

        //System.out.println("Found a new channel: " + title);

        fCurrentChannel = new RSSChannel(title, link, description);

    }
View Full Code Here
TOP

Related Classes of org.apache.nutch.parse.rss.structs.RSSChannel

  • org.apache.nutch.parse.rss.FeedParserListenerImpl
  • org.apache.nutch.parse.rss.RSSParser
  • java.util.Vector
    • 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.