1 package us.codecraft.webmagic.model.annotation;
2
3 import java.lang.annotation.ElementType;
4 import java.lang.annotation.Retention;
5 import java.lang.annotation.Target;
6
7 /**
8 * Combo 'ExtractBy' extractor with and/or operator.
9 *
10 * @author code4crafter@gmail.com <br>
11 * @since 0.2.1
12 */
13 @Retention(java.lang.annotation.RetentionPolicy.RUNTIME)
14 @Target({ElementType.FIELD, ElementType.TYPE})
15 public @interface ComboExtract {
16
17 /**
18 * The extractors to be combined.
19 *
20 * @return the extractors to be combined
21 */
22 ExtractBy[] value();
23
24 public static enum Op {
25 /**
26 * All extractors will be arranged as a pipeline. <br>
27 * The next extractor uses the result of the previous as source.
28 */
29 And,
30 /**
31 * All extractors will do extracting separately, <br>
32 * and the results of extractors will combined as the final result.
33 */
34 Or;
35 }
36
37 /**
38 * Combining operation of extractors.<br>
39 *
40 * @return combining operation of extractors
41 */
42 Op op() default Op.And;
43
44 /**
45 * Define whether the field can be null.<br>
46 * If set to 'true' and the extractor get no result, the entire class will be discarded. <br>
47 *
48 * @return whether the field can be null
49 */
50 boolean notNull() default false;
51
52 /**
53 * types of source for extracting.
54 */
55 public static enum Source {
56 /**
57 * extract from the content extracted by class extractor
58 */
59 SelectedHtml,
60 /**
61 * extract from the raw html
62 */
63 RawHtml
64 }
65
66 /**
67 * The source for extracting. <br>
68 * It works only if you already added 'ExtractBy' to Class. <br>
69 *
70 * @return the source for extracting
71 */
72 Source source() default Source.SelectedHtml;
73
74 /**
75 * Define whether the extractor return more than one result.
76 * When set to 'true', the extractor return a list of string (so you should define the field as List). <br>
77 *
78 * Deprecated since 0.4.2. This option is determined automatically by the class of field.
79 * @deprecated since 0.4.2
80 * @return whether the extractor return more than one result
81 */
82 boolean multi() default false;
83
84 }