Package org.apache.droids.exception

Examples of org.apache.droids.exception.DroidsException

116
117
118
119
120
121
122
123
  public void handle(URI uri, ContentEntity entity) throws IOException, DroidsException {
    SolrInputDocument doc = createSolrInputDocument(uri, entity);
    try {
      server.add(doc);
    } catch (SolrServerException e) {
      throw new DroidsException(e);
    }
  }
View Full Code Here
49
50
51
52
53
54
55
56
57
58
59
60
      ParseImpl parse = new ParseImpl(handler.toString(),null);
     
      return parse;

    } catch (SAXException ex) {
      throw new DroidsException("Failure parsing document " + link.getId(), ex);
    } catch (TikaException ex) {
      throw new DroidsException("Failure parsing document " + link.getId(), ex);
    } finally {
      instream.close();
    }
  }
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
79
    try {
      parser.parse(instream, parallelHandler, metadata);
     
      return new ParseImpl(data.toString(), extractor.getLinks());
    } catch (SAXException ex) {
      throw new DroidsException("Failure parsing document " + link.getId(), ex);
    } catch (TikaException ex) {
      throw new DroidsException("Failure parsing document " + link.getId(), ex);
    } finally {
      instream.close();
    }
  }
View Full Code Here
38
39
40
41
42
43
44
45
    SolrInputDocument doc = createSolrInputDocument(uri, entity);
    try {
      solr.add( doc );
    }
    catch (SolrServerException e) {
      throw new DroidsException( e );
    }
  }
View Full Code Here
117
118
119
120
121
122
123
124
  public void handle(URI uri, ContentEntity entity) throws IOException, DroidsException {
    SolrInputDocument doc = createSolrInputDocument(uri, entity);
    try {
      server.add(doc);
    } catch (SolrServerException e) {
      throw new DroidsException(e);
    }
  }
View Full Code Here
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
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
    SAXTransformerFactory factory = (SAXTransformerFactory) SAXTransformerFactory.newInstance();
    TransformerHandler xmlHandler;
    try {
      xmlHandler = factory.newTransformerHandler();
    } catch (TransformerConfigurationException e) {
      throw new DroidsException(e);
    }
    xmlHandler.getTransformer().setOutputProperty(OutputKeys.METHOD, "xml");
    xmlHandler.setResult(new StreamResult(dataBuffer));
   
    BoilerpipeContentHandler mainContentHandler = new BoilerpipeContentHandler(mainContentBuffer);
    BodyContentHandler bodyHandler = new BodyContentHandler(bodyBuffer);
    LinkContentHandler linkHandler = new LinkContentHandler();
   
    TeeContentHandler parallelHandler = new TeeContentHandler(xmlHandler, mainContentHandler, bodyHandler, linkHandler );

    InputStream instream = entity.obtainContent();
    try {
      parser.parse(instream, parallelHandler, metadata, new ParseContext());
     
      ArrayList<Link> extractedTasks = new ArrayList<Link>();
      int depth = task.getDepth() + 1;
      if (task instanceof LinkTask) {
        for(org.apache.tika.sax.Link tikaLink : linkHandler.getLinks()) {
          try {
            URI uri = new URI(tikaLink.getUri());
            // Test to see if the scheme is empty
            // This would indicate a relative URL, so resolve it against the task URI
            if(uri.getScheme() == null) {
              uri = ((Link) task).getURI().resolve(uri);
            }
            extractedTasks.add(new LinkTask((Link)task, uri, depth, tikaLink.getText()));
          } catch (URISyntaxException e) {
            if(LOG.isWarnEnabled()) {
              LOG.warn("URI not valid: "+ tikaLink.getUri());
            }
          }
        }
      }
      return new TikaParseImpl(dataBuffer.toString(), extractedTasks, bodyBuffer.toString(), mainContentBuffer.toString(), metadata);
    } catch (SAXException ex) {
      throw new DroidsException("Failure parsing document " + task.getId(), ex);
    } catch (TikaException ex) {
      throw new DroidsException("Failure parsing document " + task.getId(), ex);
    } finally {
      instream.close();
    }
  }
View Full Code Here
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
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
    SAXTransformerFactory factory = (SAXTransformerFactory) SAXTransformerFactory.newInstance();
    TransformerHandler xmlHandler;
    try {
      xmlHandler = factory.newTransformerHandler();
    } catch (TransformerConfigurationException e) {
      throw new DroidsException(e);
    }
    xmlHandler.getTransformer().setOutputProperty(OutputKeys.METHOD, "xml");
    xmlHandler.setResult(new StreamResult(dataBuffer));
   
    BoilerpipeContentHandler mainContentHandler = new BoilerpipeContentHandler(mainContentBuffer);
    BodyContentHandler bodyHandler = new BodyContentHandler(bodyBuffer);
    LinkContentHandler linkHandler = new LinkContentHandler();
   
    TeeContentHandler parallelHandler = new TeeContentHandler(xmlHandler, mainContentHandler, bodyHandler, linkHandler );

    InputStream instream = entity.obtainContent();
    try {
      parser.parse(instream, parallelHandler, metadata, new ParseContext());
     
      ArrayList<Link> extractedTasks = new ArrayList<Link>();
      if (task instanceof Link) {
        int depth = task.getDepth() + 1;
        for(org.apache.tika.sax.Link tikaLink : linkHandler.getLinks()) {
          try {
            URI uri = new URI(tikaLink.getUri());
            // Test to see if the scheme is empty
            // This would indicate a relative URL, so resolve it against the task URI
            if(uri.getScheme() == null) {
              uri = ((Link) task).getURI().resolve(uri);
            }
            extractedTasks.add(new LinkTask((Link)task, uri, depth, tikaLink.getText()));
          } catch (URISyntaxException e) {
            if(log.isWarnEnabled()) {
              log.warn("URI not valid: "+ tikaLink.getUri());
            }
          }
        }
      }
      return new TikaParseImpl(dataBuffer.toString(), extractedTasks, bodyBuffer.toString(), mainContentBuffer.toString(), metadata);
    } catch (SAXException ex) {
      throw new DroidsException("Failure parsing document " + task.getId(), ex);
    } catch (TikaException ex) {
      throw new DroidsException("Failure parsing document " + task.getId(), ex);
    } finally {
      instream.close();
    }
  }
View Full Code Here
73
74
75
76
77
78
79
80
81
82
83
      ManagedContentEntity entity = null;
      try {
        entity = protocol.load(uri);
      } catch(OutOfMemoryError e) {
        LOG.error("Out of memory processing: " + uri + " skipping", e);
        throw new DroidsException(e);
      }
      try {
        String contentType = entity.getMimeType();
        if (LOG.isDebugEnabled()) {
          LOG.debug("Content type " + contentType);
View Full Code Here
TOP

Related Classes of org.apache.droids.exception.DroidsException

  • org.apache.droids.robot.crawler.CrawlingWorker
  • org.apache.droids.solr.AdvancedSolrHandler
  • org.apache.droids.solr.SolrHandler
  • org.apache.droids.tika.TikaDocumentParser
  • org.apache.droids.tika.TikaHtmlParser
    • 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.