Package com.iqbon.spider.domain

Examples of com.iqbon.spider.domain.Source

45
46
47
48
49
50
51
52
53
    sourceDao.insertOne(source);
  }

  @Test
  public void testQuerySourceByUrl() {
    Source s = sourceDao.querySourceByUrl(source.getUrl());
    logger.info(ToStringBuilder.reflectionToString(s));

  }
View Full Code Here
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
    }
  }

  @Test
  public void testUpdateSource() {
    Source s = sourceDao.querySourceByUrl(source.getUrl());
    SourceMatcher matcher = new SourceMatcher();
    matcher.setLinkParent("li[class=list-right]");
    matcher.setDescription("�Ҳ�����");
    Replace replace = new Replace();
    replace.setMatcher("<div class=\"gg200x300\">[\\s\\S]*?</div>");
    replace.setReplacement("");
    Replace replace2 = new Replace();
    replace2.setMatcher("<a [\\s\\S]*?>");
    replace2.setReplacement("");
    Replace replace3 = new Replace();
    replace3.setMatcher("</a>");
    replace3.setReplacement("");
    List<Replace> replaces = new ArrayList<Replace>();
    List<SourceMatcher> matchers = new ArrayList<SourceMatcher>();
    matchers.add(matcher);
    replaces.add(replace);
    replaces.add(replace2);
    replaces.add(replace3);
    s.setContentMatcher("#endText");
    s.setMatchers(matchers);
    s.setReplaces(replaces);
    s.setDescription("after test");
    sourceDao.updateSource(s);
  }
View Full Code Here
34
35
36
37
38
39
40
41
   * @param url
   * @param sourceUrl
   * @return
   */
  public Record getCrawlContentBySourceAndUrl(String url, String sourceUrl) {
    Source source = sourceDao.querySourceByUrl(sourceUrl);
    return crawlService.getContentFromLink(url, source);
  }
View Full Code Here
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
      return "输入的URL不合法";
    }
    if (StringUtils.isEmpty(contentMatcher)) {
      return "内容匹配规则不能为空";
    }
    Source duplicate = sourceDao.querySourceByUrl(url);
    if (duplicate != null) {
      return "已经存在重复的采集数据源" + url;
    }
    Source source = new Source();
    source.setUrl(url);
    source.setContentMatcher(contentMatcher);
    source.setDescription(description);
    source.setMatchers(matchers);
    sourceDao.insertOne(source);
    return null;
  }
View Full Code Here
23
24
25
26
27
28
29
30
   * 根据url获取该抓取数据源所能匹配的链接
   * @param sourceId
   * @return
   */
  public List<String> getCrawlRecordBySourceId(String url) {
    Source source = sourceDao.querySourceByUrl(url);
    return crawlService.getLinkFromSource(source);
  }
View Full Code Here
36
37
38
39
40
41
42
43
44
45
    fail("Not yet implemented");
  }

  @Test
  public void testGetContentFromLink() {
    Source source = sourceDao.querySourceByUrl(URL);
    Record record = crawlService.getContentFromLink(
        "http://home.163.com/13/0123/18/8LU3QD5700104JVC.html", source);
    logger.info(ToStringBuilder.reflectionToString(record));
  }
View Full Code Here
TOP

Related Classes of com.iqbon.spider.domain.Source

  • com.iqbon.spider.dao.SourceDaoTest
  • com.iqbon.spider.service.crawl.CrawlServiceTest
  • com.iqbon.spider.web.RecordController
  • com.iqbon.spider.web.SourceController
    • 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.