Package fr.eolya.crawler.queue

Examples of fr.eolya.crawler.queue.ISourceItemsQueue

845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
    SourceWeb src = (SourceWeb) source;

    StartingUrls startingUrls = src.getStartingUrls();
    if (startingUrls==null) return null;

    ISourceItemsQueue sourceQueue = QueueFactory.getSourceItemsQueueInstance(type, src.getId(), con, dbName, dbCollName);
    if (sourceQueue==null) return null;
   
    if (src.isReset() || src.isClear())
      sourceQueue.reset();

    if (src.isClear()) return sourceQueue;

    // TODO : v4 - Utile ???
    if (sourceQueue.getQueueSize()>0 && !src.isRescan() && !src.isRescanFromCache())
      sourceQueue.start();
    else {
      if (!src.isRescan() && !src.isRescanFromCache()) {
        int startDepth = 0;
        if (src.isDeeper()) {
          startDepth = sourceQueue.getCurrentMaxDepth();
        }
        sourceQueue.reStart(startDepth);
      } else {
        sourceQueue.reScan();
      }

    }
    // Utile ???

    if (!src.isRescan() && !src.isRescanFromCache() && !src.isDeeper()) {
      // Add or update starting urls
      boolean haveSiteMode = false;
      for (int i=0; i<startingUrls.size(); i++) {
        if ("s".equals(startingUrls.get(i).mode)) haveSiteMode = true;
      }
      for (int i=0; i<startingUrls.size(); i++) {
        if (!startingUrls.get(i).onlyFirstCrawl || !src.isFirstCrawlCompleted()) {
          if (haveSiteMode && !src.isFirstCrawlCompleted() && !"s".equals(startingUrls.get(i).mode)) continue;
          try {
            sourceQueue.push(startingUrls.get(i).getMap(src.getId()));
          } catch (Exception e) {
            e.printStackTrace();
          }
        }
      }
View Full Code Here
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
 
  @Test
  public void testMongoDBWebSiteUrlFifoQueue() {
    try {
      IDBConnection dbConnection = DBConnectionFactory.getDBConnectionInstance("mongodb", "localhost", 27017,  "", "");
      ISourceItemsQueue queue = QueueFactory.getSourceItemsQueueInstance("mongodb", 1, dbConnection, "testFifoQueue", "TestMongoDBWebSiteUrlFifoQueue");
     
      if (queue!=null) {
        queue.reset();
        queue.start();
        assertEquals(0, queue.size());
       
        assertTrue(queue.push(getItemWeb(1, "http://a.a.a/", 0).getMap()));
        assertTrue(queue.push(getItemWeb(1, "http://a.a.a/2.html", 1).getMap()));
        assertTrue(queue.push(getItemWeb(1, "http://a.a.a/3.html", 2).getMap()));

        assertTrue(queue.contains("http://a.a.a/"));
        assertEquals(3, queue.size());

        assertEquals(0, queue.getDoneQueueSize());
        assertFalse(queue.isDone("http://a.a.a/"));
       
        assertEquals("http://a.a.a/", queue.pop().get("url"));
        assertEquals("http://a.a.a/2.html", queue.pop().get("url"));
        assertEquals("http://a.a.a/3.html", queue.pop().get("url"));

        assertEquals(3, queue.getDoneQueueSize());
        assertTrue(queue.isDone("http://a.a.a/"));

        assertFalse(queue.contains("http://a.a.a/"));
        assertEquals(0, queue.size());

        assertFalse(queue.push(getItemWeb(1, "http://a.a.a/", 0).getMap()));
        assertTrue(queue.push(getItemWeb(1, "http://a.a.a/4.html", 2).getMap()));
        assertTrue(queue.push(getItemWeb(1, "http://a.a.a/3.html", 1).getMap()));
        assertFalse(queue.push(getItemWeb(1, "http://a.a.a/2.html", 2).getMap()));
       
        assertEquals(2, queue.size());

        assertEquals("http://a.a.a/3.html", queue.pop("depth").get("url"));
        assertEquals("http://a.a.a/4.html", queue.pop("depth").get("url"));

        assertEquals(0, queue.size());   
      }
     
      queue = QueueFactory.getSourceItemsQueueInstance("mongodb", 1, dbConnection, "testFifoQueue", "TestMongoDBWebSiteUrlFifoQueue"); 
      if (queue!=null) {
        queue.start();
        assertEquals(0, queue.size());
        assertFalse(queue.push(getItemWeb(1, "http://a.a.a/", 0).getMap()));
        queue.stop();
      }

      queue = QueueFactory.getSourceItemsQueueInstance("mongodb", 1, dbConnection, "testFifoQueue", "TestMongoDBWebSiteUrlFifoQueue"); 
      if (queue!=null) {
        queue.start();
        assertEquals(0, queue.size());
        assertTrue(queue.push(getItemWeb(1, "http://a.a.a/", 0).getMap()));
      }
    }
    catch (Exception e) {
      e.printStackTrace();
    }
View Full Code Here
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
  }
 
  public void testMongoDBWebSiteUrlFifoQueue2() {
    try {
      IDBConnection dbConnection = DBConnectionFactory.getDBConnectionInstance("mongodb", "localhost", 27017,  "", "");
      ISourceItemsQueue queue = QueueFactory.getSourceItemsQueueInstance("mongodb", 1, dbConnection, "testFifoQueue", "TestMongoDBWebSiteUrlFifoQueue");
     
      if (queue!=null) {
        queue.reset();
        queue.start();
        assertEquals(0, queue.size());
        assertTrue(queue.push(getItemWeb(1, "http://a.a.a/", 0).getMap()));
      }
    }
    catch (Exception e) {
      e.printStackTrace();
    }
View Full Code Here
TOP

Related Classes of fr.eolya.crawler.queue.ISourceItemsQueue

  • fr.eolya.crawler.connectors.web.WebConnector
  • fr.eolya.crawler.queue.mongodb.SourceItemsQueueTest
    • 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.