| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 1 | package de.ids_mannheim.korap.index; |
| 2 | |
| Eliza Margaretha | 51fd5c2 | 2014-10-14 13:12:33 +0000 | [diff] [blame] | 3 | import static org.junit.Assert.assertEquals; |
| 4 | |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 5 | import java.io.IOException; |
| 6 | |
| 7 | import org.apache.lucene.index.Term; |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 8 | import org.apache.lucene.sandbox.queries.regex.RegexQuery; |
| 9 | import org.apache.lucene.search.spans.SpanMultiTermQueryWrapper; |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 10 | import org.apache.lucene.search.spans.SpanQuery; |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 11 | import org.apache.lucene.search.spans.SpanTermQuery; |
| 12 | import org.junit.Test; |
| 13 | |
| Nils Diewald | a14ecd6 | 2015-02-26 21:00:20 +0000 | [diff] [blame] | 14 | import de.ids_mannheim.korap.KrillIndex; |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 15 | import de.ids_mannheim.korap.constants.RelationDirection; |
| Eliza Margaretha | 98c200e | 2014-10-15 13:59:58 +0000 | [diff] [blame] | 16 | import de.ids_mannheim.korap.query.SpanAttributeQuery; |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 17 | import de.ids_mannheim.korap.query.SpanClassQuery; |
| Eliza Margaretha | 3e50bc4 | 2014-10-22 15:29:15 +0000 | [diff] [blame] | 18 | import de.ids_mannheim.korap.query.SpanElementQuery; |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 19 | import de.ids_mannheim.korap.query.SpanFocusQuery; |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 20 | import de.ids_mannheim.korap.query.SpanRelationMatchQuery; |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 21 | import de.ids_mannheim.korap.query.SpanRelationQuery; |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 22 | import de.ids_mannheim.korap.query.SpanSegmentQuery; |
| Eliza Margaretha | d12cabb | 2014-10-27 17:45:34 +0000 | [diff] [blame] | 23 | import de.ids_mannheim.korap.query.SpanTermWithIdQuery; |
| Eliza Margaretha | 98c200e | 2014-10-15 13:59:58 +0000 | [diff] [blame] | 24 | import de.ids_mannheim.korap.query.SpanWithAttributeQuery; |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 25 | import de.ids_mannheim.korap.response.Result; |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 26 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 27 | /* |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 28 | * |
| 29 | * within(x,y) |
| 30 | * |
| 31 | * SpanRelationQuery-> |
| 32 | * rel("SUBJ", query1, query2) |
| 33 | * |
| 34 | * 1. return all words that are subjects of (that are linked by the |
| 35 | * “SUBJ” relation to) the string “beginnt” |
| 36 | * xip/syntax-dep_rel:beginnt >[func=”SUBJ”] xip/syntax-dep_rel:.* |
| 37 | * -> rel("SUBJ", highlight(query1), new TermQuery("s:beginnt")) |
| 38 | * |
| 39 | * |
| 40 | * SUBJ ist modelliert mit offset für den gesamten Bereich |
| 41 | * |
| 42 | * https://de.wikipedia.org/wiki/Dependenzgrammatik |
| 43 | * |
| 44 | * im regiert Wasser |
| 45 | * dass die Kinder im Wasser gespielt haben |
| 46 | * 3. im#16-18$ |
| 47 | * 3. >:COORD#16-25$3,4 |
| 48 | * 4. Wasser#19-25$ |
| 49 | * 4. <:COORD#16-25$3,4 |
| 50 | * |
| 51 | * # okay: return all main verbs that have no “SUBJ” relation |
| 52 | * specified |
| 53 | * |
| 54 | * |
| 55 | * # Not okay: 5. return all verbs with (at least?) 3 outgoing |
| 56 | * relations [think of ditransitives such as give] |
| 57 | * |
| 58 | * xip/morph_pos:VERB & xip/token:.* & xip/token:.* & xip/token:.* & |
| 59 | * xip/token:.* & #1 _=_#2 & #2 >[func=$x] #3 & #2 >[func=$x]#4 & #2 |
| 60 | * >[func=$x] #5 |
| 61 | * |
| 62 | * # Okay: return all verbs that have singular SUBJects and dative |
| 63 | * OBJects |
| 64 | * |
| 65 | * xip/morph_pos:VERB & mpt/morph_msd:Sg & mpt/morph_msd:Dat & #1 |
| 66 | * >[func=”SUBJ”] #2 & #1 >[func=”OBJ”] #3 |
| 67 | * |
| 68 | * -> [p:VVFIN](>SUBJ[nr:sg] & >OBJ[c:dat]) |
| 69 | * |
| 70 | * |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 71 | */ |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 72 | |
| 73 | public class TestRelationIndex { |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 74 | private KrillIndex ki; |
| 75 | private Result kr; |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 76 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 77 | |
| 78 | public TestRelationIndex () throws IOException { |
| 79 | ki = new KrillIndex(); |
| 80 | } |
| 81 | |
| 82 | |
| Akron | c12567c | 2016-06-03 00:40:52 +0200 | [diff] [blame] | 83 | public static FieldDocument createFieldDoc0 () { |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 84 | FieldDocument fd = new FieldDocument(); |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 85 | fd.addString("ID", "doc-0"); |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 86 | fd.addTV("base", "ceccecdeed", "[(0-1)s:c$<s>1|_0$<i>0<i>1" |
| 87 | + "|>:xip/syntax-dep_rel$<b>32<i>6<s>1<s>1<s>0]" |
| 88 | + "[(1-2)s:e$<s>1|_1$<i>1<i>2|" |
| 89 | + "<:xip/syntax-dep_rel$<b>32<i>9<s>1<s>1<s>0|" |
| 90 | + ">:xip/syntax-dep_rel$<b>32<i>4<s>1<s>1<s>0]" |
| 91 | + "[(2-3)s:c|_2$<i>2<i>3]" |
| 92 | + "[(3-4)s:c$<s>1|s:b$<s>2|_3$<i>3<i>4|<:xip/syntax-dep_rel$<b>32<i>9<s>1<s>1<s>0]" |
| 93 | + "[(4-5)s:e$<s>1|s:d$<s>2|_4$<i>4<i>5|<:xip/syntax-dep_rel$<b>32<i>1<s>1<s>1<s>0]" |
| 94 | + "[(5-6)s:c|_5$<i>5<i>6]" |
| 95 | + "[(6-7)s:d$<s>1|_6$<i>6<i>7|<:xip/syntax-dep_rel$<b>32<i>1<s>1<s>1<s>0]" |
| 96 | + "[(7-8)s:e|_7$<i>7<i>8]" + "[(8-9)s:e|s:b|_8$<i>8<i>9]" |
| 97 | + "[(9-10)s:d$<s>1|_9$<i>9<i>10|" |
| 98 | + ">:xip/syntax-dep_rel$<b>32<i>1<s>1<s>1<s>0|" |
| 99 | + ">:xip/syntax-dep_rel$<b>32<i>3<s>1<s>1<s>0]"); |
| Eliza Margaretha | f13b8ad | 2014-10-13 16:36:28 +0000 | [diff] [blame] | 100 | return fd; |
| 101 | } |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 102 | |
| 103 | |
| Akron | c12567c | 2016-06-03 00:40:52 +0200 | [diff] [blame] | 104 | public static FieldDocument createFieldDoc1 () { |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 105 | FieldDocument fd = new FieldDocument(); |
| Eliza Margaretha | 51fd5c2 | 2014-10-14 13:12:33 +0000 | [diff] [blame] | 106 | fd.addString("ID", "doc-1"); |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 107 | fd.addTV("base", "ceccecdeed", |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 108 | "[(0-1)s:c$<s>2|<>:p$<b>64<i>0<i>3<i>3<b>0<s>1|_0$<i>0<i>1|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 109 | + ">:xip/syntax-dep_rel$<b>35<i>0<i>1<i>1<i>2<i>3<i>6<i>9<s>1<s>1<s>2|" |
| 110 | + ">:xip/syntax-dep_rel$<b>33<i>1<i>2<i>6<i>9<s>2<s>1<s>0|" |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 111 | + "@:func=subj$<b>18<s>2]" |
| 112 | + "[(1-2)s:e|_1$<i>1<i>2|<>:p$<b>64<i>1<i>3<i>3<b>0<s>1]" |
| 113 | + "[(2-3)s:c|_2$<i>2<i>3]" |
| 114 | + "[(3-4)s:c|s:b|_3$<i>3<i>4]" |
| 115 | + "[(4-5)s:e|s:d|_4$<i>4<i>5]" |
| 116 | + "[(5-6)s:c|_5$<i>5<i>6]" |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 117 | + "[(6-7)s:d$<s>2|<>:p$<b>64<i>6<i>9<i>9<b>0<s>1|_6$<i>6<i>7|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 118 | + ">:xip/syntax-dep_rel$<b>34<i>3<i>4<i>9<i>9<s>1<s>1<s>0|" |
| 119 | + "<:xip/syntax-dep_rel$<b>35<i>3<i>4<i>4<i>5<i>9<i>1<i>3<s>1<s>1<s>2|" |
| 120 | + "<:xip/syntax-dep_rel$<b>34<i>5<i>6<i>9<i>1<s>1<s>2<s>0|" |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 121 | + "@:func=obj$<b>18<s>2]" + "[(7-8)s:e|_7$<i>7<i>8]" |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 122 | + "[(8-9)s:e|s:b|_8$<i>8<i>9]" |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 123 | + "[(9-10)s:d$<s>1|_9$<i>9<i>10|<" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 124 | + ":xip/syntax-dep_rel$<b>33<i>6<i>7<i>6<i>9<s>2<s>1<s>0]"); |
| Eliza Margaretha | 51fd5c2 | 2014-10-14 13:12:33 +0000 | [diff] [blame] | 125 | return fd; |
| 126 | } |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 127 | |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 128 | public static FieldDocument createFieldDoc3 () { |
| 129 | FieldDocument fd = new FieldDocument(); |
| 130 | fd.addString("ID", "doc-1"); |
| 131 | fd.addTV("base", "ceccecdeed", |
| 132 | "[(0-1)s:c$<s>2|<>:p$<b>64<i>0<i>3<i>3<b>0<s>1|_0$<i>0<i>1|" |
| 133 | + ">:xip/dep_rel$<b>35<i>0<i>1<i>1<i>2<i>3<i>6<i>9<s>1<s>1<s>2|" |
| 134 | + ">:xip/dep_rel$<b>33<i>1<i>2<i>6<i>9<s>2<s>1<s>0|" |
| 135 | + "@:func=subj$<b>18<s>2]" |
| 136 | + "[(1-2)s:e|_1$<i>1<i>2|<>:p$<b>64<i>1<i>3<i>3<b>0<s>1]" |
| 137 | + "[(2-3)s:c|_2$<i>2<i>3]" |
| 138 | + "[(3-4)s:c|s:b|_3$<i>3<i>4]" |
| 139 | + "[(4-5)s:e|s:d|_4$<i>4<i>5]" |
| 140 | + "[(5-6)s:c|_5$<i>5<i>6]" |
| 141 | + "[(6-7)s:d$<s>2|<>:p$<b>64<i>6<i>9<i>9<b>0<s>1|_6$<i>6<i>7|" |
| 142 | + ">:xip/dep_rel$<b>34<i>3<i>4<i>9<i>9<s>1<s>1<s>0|" |
| 143 | + "<:xip/dep_rel$<b>35<i>3<i>4<i>4<i>5<i>9<i>1<i>3<s>1<s>1<s>2|" |
| 144 | + "<:xip/dep_rel$<b>34<i>5<i>6<i>9<i>1<s>1<s>2<s>0|" |
| 145 | + "@:func=obj$<b>18<s>2]" + "[(7-8)s:e|_7$<i>7<i>8]" |
| 146 | + "[(8-9)s:e|s:b|_8$<i>8<i>9]" |
| 147 | + "[(9-10)s:d$<s>1|_9$<i>9<i>10|<" |
| 148 | + ":xip/dep_rel$<b>33<i>6<i>7<i>6<i>9<s>2<s>1<s>0]"); |
| 149 | return fd; |
| 150 | } |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 151 | |
| 152 | |
| margaretha | b097bac | 2015-04-15 11:37:02 +0200 | [diff] [blame] | 153 | public static FieldDocument createFieldDoc2 () { |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 154 | FieldDocument fd = new FieldDocument(); |
| Eliza Margaretha | 98c200e | 2014-10-15 13:59:58 +0000 | [diff] [blame] | 155 | fd.addString("ID", "doc-2"); |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 156 | fd.addTV("base", "Ich kaufe die Blümen für meine Mutter.", |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 157 | "[(0-3)s:Ich|_0$<i>0<i>3|pos:NN$<s>1|<>:s$<b>64<i>0<i>38<i>7<b>0<s>2|<>:np$<b>64<i>0<i>3<i>1<b>0<s>3|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 158 | + ">:child-of$<b>33<i>0<i>38<i>0<i>7<s>3<s>2<s>4|" |
| 159 | + ">:child-of$<b>33<i>0<i>3<i>0<i>1<s>1<s>3<s>0|" |
| 160 | + "<:child-of$<b>32<i>0<s>3<s>1<s>0|" |
| 161 | + "<:child-of$<b>35<i>0<i>38<i>0<i>3<i>7<i>0<i>1<s>2<s>3<s>0|" |
| 162 | + "<:child-of$<b>35<i>0<i>38<i>4<i>9<i>7<i>1<i>7<s>2<s>2<s>0|" |
| 163 | + "<:dep$<b>32<i>0<s>1<s>1<s>0|" |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 164 | + "@:func=sbj$<b>18<i>7<s>4|" // attribute belongs to a relation |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 165 | + "@:case=nominative$<b>18<i>1<s>3]" // attribute belongs to a relation node |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 166 | |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 167 | + "[(1-2)s:kaufe|_1$<i>4<i>9|pos:V$<s>1|<>:vp$<b>64<i>4<i>38<i>7<b>0<s>2|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 168 | + ">:child-of$<b>35<i>4<i>38<i>0<i>38<i>7<i>0<i>7<s>2<s>2<s>0|" |
| 169 | + ">:child-of$<b>33<i>4<i>9<i>1<i>7<s>2<s>7<s>0|" |
| 170 | + "<:child-of$<b>34<i>4<i>38<i>7<i>2<s>2<s>1<s>0|" |
| 171 | + "<:child-of$<b>35<i>4<i>38<i>10<i>38<i>7<i>2<i>7<s>2<s>4<s>0|" |
| 172 | + ">:dep$<b>32<i>0<s>1<s>1<s>0|" |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 173 | + ">:dep$<b>32<i>3<s>1<s>1<s>0]" |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 174 | |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 175 | + "[(2-3)s:die|_2$<i>10<i>13|pos:ART$<s>1|tt:DET$<s>2|" |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 176 | + "<>:np$<b>64<i>10<i>20<i>4<b>0<s>3|<>:np$<b>64<i>10<i>38<i>7<b>0<s>4|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 177 | + ">:child-of$<b>35<i>10<i>24<i>10<i>38<i>4<i>2<i>7<s>3<s>4<s>0|" |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 178 | |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 179 | + ">:child-of$<b>33<i>10<i>24<i>2<i>4<s>1<s>3<s>0|" |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 180 | |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 181 | + ">:child-of$<b>35<i>10<i>38<i>4<i>38<i>7<i>1<i>7<s>4<s>2<s>0|" |
| 182 | + "<:child-of$<b>34<i>10<i>24<i>4<i>2<s>3<s>1<s>0|" |
| 183 | + "<:child-of$<b>34<i>10<i>24<i>4<i>3<s>3<s>1<s>0|" |
| 184 | + "<:child-of$<b>35<i>10<i>38<i>10<i>24<i>7<i>2<i>4<s>4<s>3<s>0|" |
| 185 | + "<:child-of$<b>35<i>10<i>38<i>21<i>38<i>7<i>4<i>7<s>4<s>2<s>0|" |
| 186 | + ">:parent-of$<b>35<i>10<i>38<i>21<i>38<i>7<i>4<i>7<s>4<s>2<s>0|" |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 187 | + "<:dep$<b>32<i>3<s>1<s>1<s>3|" |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 188 | + "@:func=head$<b>18<i>4<s>3|" |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 189 | + "@:case=accusative$<b>18<i>4<s>3|" // attribute belongs to a relation node |
| 190 | + "@:case=accusative$<b>18<i>7<s>2]" // attribute belongs to a relation node |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 191 | |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 192 | + "[(3-4)s:Blümen|_3$<i>14<i>20|pos:NN$<s>1|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 193 | + ">:child-of$<b>33<i>10<i>24<i>2<i>4<s>1<s>3<s>0|" |
| 194 | + "<:dep$<b>32<i>1<s>1<s>1<s>0|" |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 195 | + ">:dep$<b>32<i>2<s>1<s>1<s>2|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 196 | + ">:dep$<b>32<i>4<s>1<s>1<s>0|" |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 197 | + "@:func=obj$<b>18<i>4<s>2]" |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 198 | |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 199 | + "[(4-5)s:für|_4$<i>21<i>24|pos:PREP$<s>1|<>:pp$<b>64<i>21<i>38<i>7<b>0<s>2|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 200 | + ">:child-of$<b>33<i>21<i>38<i>4<i>7<s>1<s>2<s>0|" |
| 201 | + ">:child-of$<b>35<i>21<i>38<i>10<i>38<i>7<i>2<i>7<s>2<s>4<s>0|" |
| 202 | + "<:child-of$<b>34<i>21<i>38<i>7<i>5<s>2<s>1<s>0|" |
| 203 | + "<:child-of$<b>35<i>21<i>38<i>25<i>38<i>7<i>5<i>7<s>2<s>2<s>0|" |
| 204 | + "<:dep$<b>32<i>3<s>1<s>1<s>0|" |
| 205 | + ">:dep$<b>32<i>6<s>1<s>1<s>0]" |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 206 | |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 207 | + "[(5-6)s:meine|_5$<i>25<i>30|pos:ART$<s>1|<>:np$<b>64<i>25<i>38<i>7<b>0<s>2|" |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 208 | |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 209 | + ">:child-of$<b>33<i>25<i>38<i>5<i>7<s>1<s>2<s>0|" |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 210 | |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 211 | + ">:child-of$<b>35<i>25<i>38<i>21<i>38<i>7<i>4<i>7<s>2<s>2<s>0|" |
| 212 | + "<:child-of$<b>34<i>25<i>38<i>7<i>5<s>2<s>1<s>0|" |
| 213 | + "<:child-of$<b>34<i>25<i>38<i>7<i>6<s>2<s>1<s>0|" |
| 214 | + "<:dep$<b>32<i>6<s>1<s>1<s>3<s>0|" |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 215 | + "@:func=head$<b>18<i>7<s>3|" |
| 216 | + "@:case=accusative$<b>18<i>7<s>2]" // attribute belongs to a relation node |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 217 | |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 218 | + "[(6-7)s:Mutter.|_6$<i>31<i>38|pos:NN$<s>1|" |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 219 | + ">:child-of$<b>33<i>25<i>38<i>5<i>7<s>1<s>2<s>0|" |
| 220 | + ">:dep$<b>32<i>5<s>1<s>1<s>0|" |
| 221 | + "<:dep$<b>32<i>4<s>1<s>1<s>0]"); |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 222 | |
| Eliza Margaretha | 98c200e | 2014-10-15 13:59:58 +0000 | [diff] [blame] | 223 | return fd; |
| 224 | } |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 225 | |
| 226 | |
| 227 | public static FieldDocument createFieldDoc4 () { |
| 228 | FieldDocument fd = new FieldDocument(); |
| 229 | fd.addString("ID", "doc-0"); |
| 230 | fd.addTV("base", "element term", |
| 231 | "[(0-1)s:element|_0$<i>0<i>7|<>:s$<b>64<i>0<i>7<i>0<b>0<s>1|" |
| 232 | // this relation is missing TUIs for source, target and relation ids |
| 233 | + ">:dep$<b>34<i>0<i>7<i>0<i>1|" |
| 234 | //+ ">:dep$<b>34<i>0<i>7<i>0<i>1<s>1<s>2<s>3|" |
| 235 | + "[(1-2)s:term$<b>128<s>2|_1$<i>8<i>12|]"); |
| 236 | return fd; |
| 237 | } |
| 238 | |
| 239 | @Test |
| 240 | public void testRelationMissingPayloads () throws IOException { |
| 241 | ki.addDoc(createFieldDoc4()); |
| 242 | ki.commit(); |
| 243 | |
| 244 | SpanRelationQuery sq = new SpanRelationQuery( |
| 245 | new SpanTermQuery(new Term("base", ">:dep")), true, |
| 246 | RelationDirection.RIGHT); |
| 247 | kr = ki.search(sq, (short) 10); |
| 248 | |
| 249 | assertEquals((long) 1, kr.getTotalResults()); |
| 250 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 251 | assertEquals(0, kr.getMatch(0).getEndPos()); |
| 252 | |
| 253 | SpanFocusQuery fq = new SpanFocusQuery(sq, sq.getTempClassNumbers()); |
| 254 | fq.setMatchTemporaryClass(true); |
| 255 | fq.setRemoveTemporaryClasses(true); |
| 256 | fq.setSorted(false); |
| 257 | |
| 258 | kr = ki.search(fq, (short) 20); |
| 259 | assertEquals((long) 1, kr.getTotalResults()); |
| 260 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 261 | assertEquals(2, kr.getMatch(0).getEndPos()); |
| 262 | } |
| 263 | |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 264 | @Test |
| 265 | public void testRelationWithRegex () throws IOException { |
| 266 | ki.addDoc(createFieldDoc0()); |
| 267 | ki.addDoc(createFieldDoc3()); |
| 268 | ki.commit(); |
| 269 | |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 270 | SpanQuery sq; |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 271 | sq = new SpanRelationQuery( |
| 272 | new SpanMultiTermQueryWrapper<RegexQuery>( |
| 273 | new RegexQuery(new Term("base", ">:xip/.*"))), |
| 274 | true, RelationDirection.RIGHT); |
| 275 | kr = ki.search(sq, (short) 10); |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 276 | |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 277 | assertEquals((long) 7, kr.getTotalResults()); |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 278 | |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 279 | sq = new SpanRelationQuery( |
| 280 | new SpanMultiTermQueryWrapper<RegexQuery>( |
| 281 | new RegexQuery(new Term("base", "<:xip/.*"))), |
| 282 | true, RelationDirection.LEFT); |
| 283 | kr = ki.search(sq, (short) 10); |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 284 | |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 285 | assertEquals((long) 7, kr.getTotalResults()); |
| 286 | } |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 287 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 288 | /** |
| 289 | * Relations: token to token, token to span, span to span |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 290 | */ |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 291 | @Test |
| 292 | public void testCase1 () throws IOException { |
| 293 | ki.addDoc(createFieldDoc0()); |
| 294 | ki.addDoc(createFieldDoc1()); |
| 295 | ki.commit(); |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 296 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 297 | SpanRelationQuery sq = new SpanRelationQuery( |
| 298 | new SpanTermQuery(new Term("base", ">:xip/syntax-dep_rel")), |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 299 | true, RelationDirection.RIGHT); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 300 | kr = ki.search(sq, (short) 10); |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 301 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 302 | assertEquals((long) 7, kr.getTotalResults()); |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 303 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 304 | // token to token |
| 305 | assertEquals(0, kr.getMatch(0).getLocalDocID()); |
| 306 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 307 | assertEquals(1, kr.getMatch(0).getEndPos()); |
| 308 | assertEquals(1, kr.getMatch(1).getStartPos()); |
| 309 | assertEquals(2, kr.getMatch(1).getEndPos()); |
| 310 | assertEquals(9, kr.getMatch(2).getStartPos()); |
| 311 | assertEquals(10, kr.getMatch(2).getEndPos()); |
| 312 | assertEquals(9, kr.getMatch(3).getStartPos()); |
| 313 | assertEquals(10, kr.getMatch(3).getEndPos()); |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 314 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 315 | // token to span |
| 316 | assertEquals(1, kr.getMatch(4).getLocalDocID()); |
| 317 | assertEquals(0, kr.getMatch(4).getStartPos()); |
| 318 | assertEquals(1, kr.getMatch(4).getEndPos()); |
| 319 | assertEquals(0, kr.getMatch(5).getStartPos()); |
| 320 | assertEquals(3, kr.getMatch(5).getEndPos()); |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 321 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 322 | // span to span |
| 323 | assertEquals(6, kr.getMatch(6).getStartPos()); |
| 324 | assertEquals(9, kr.getMatch(6).getEndPos()); |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 325 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 326 | // check target |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 327 | } |
| 328 | |
| 329 | |
| 330 | /** |
| 331 | * Relation span to token |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 332 | */ |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 333 | @Test |
| 334 | public void testCase2 () throws IOException { |
| 335 | ki.addDoc(createFieldDoc0()); |
| 336 | ki.addDoc(createFieldDoc1()); |
| 337 | ki.commit(); |
| 338 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 339 | SpanRelationQuery sq = new SpanRelationQuery( |
| 340 | new SpanTermQuery(new Term("base", "<:xip/syntax-dep_rel")), |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 341 | true, RelationDirection.LEFT); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 342 | kr = ki.search(sq, (short) 10); |
| 343 | |
| 344 | assertEquals((long) 7, kr.getTotalResults()); |
| 345 | // token to token |
| 346 | assertEquals(0, kr.getMatch(0).getLocalDocID()); |
| 347 | assertEquals(1, kr.getMatch(0).getStartPos()); |
| 348 | assertEquals(2, kr.getMatch(0).getEndPos()); |
| 349 | assertEquals(3, kr.getMatch(1).getStartPos()); |
| 350 | assertEquals(4, kr.getMatch(1).getEndPos()); |
| 351 | assertEquals(4, kr.getMatch(2).getStartPos()); |
| 352 | assertEquals(5, kr.getMatch(2).getEndPos()); |
| 353 | assertEquals(6, kr.getMatch(3).getStartPos()); |
| 354 | assertEquals(7, kr.getMatch(3).getEndPos()); |
| 355 | |
| 356 | assertEquals(1, kr.getMatch(4).getLocalDocID()); |
| 357 | // span to token |
| 358 | assertEquals(6, kr.getMatch(4).getStartPos()); |
| 359 | assertEquals(9, kr.getMatch(4).getEndPos()); |
| 360 | assertEquals(6, kr.getMatch(5).getStartPos()); |
| 361 | assertEquals(9, kr.getMatch(5).getEndPos()); |
| 362 | // span to span |
| 363 | assertEquals(9, kr.getMatch(6).getStartPos()); |
| 364 | assertEquals(10, kr.getMatch(6).getEndPos()); |
| 365 | } |
| 366 | |
| 367 | |
| 368 | /** |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 369 | * Relations only |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 370 | */ |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 371 | @Test |
| 372 | public void testCase3 () throws IOException { |
| 373 | ki.addDoc(createFieldDoc2()); |
| 374 | ki.commit(); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 375 | |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 376 | // child-of relations |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 377 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 378 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 379 | RelationDirection.RIGHT); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 380 | kr = ki.search(srq, (short) 20); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 381 | |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 382 | assertEquals((long) 13, kr.getTotalResults()); |
| 383 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 384 | assertEquals(1, kr.getMatch(0).getEndPos()); |
| 385 | assertEquals(0, kr.getMatch(1).getStartPos()); |
| 386 | assertEquals(1, kr.getMatch(1).getEndPos()); |
| 387 | assertEquals(1, kr.getMatch(2).getStartPos()); |
| 388 | assertEquals(2, kr.getMatch(2).getEndPos()); |
| 389 | assertEquals(1, kr.getMatch(3).getStartPos()); |
| 390 | assertEquals(7, kr.getMatch(3).getEndPos()); |
| 391 | assertEquals(2, kr.getMatch(4).getStartPos()); |
| 392 | assertEquals(3, kr.getMatch(4).getEndPos()); |
| 393 | assertEquals(2, kr.getMatch(5).getStartPos()); |
| 394 | assertEquals(4, kr.getMatch(5).getEndPos()); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 395 | } |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 396 | |
| Akron | bb5d173 | 2015-06-22 01:22:40 +0200 | [diff] [blame] | 397 | |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 398 | /** |
| 399 | * Relations only with/out attribute |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 400 | */ |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 401 | @Test |
| 402 | public void testCase4 () throws IOException { |
| 403 | ki.addDoc(createFieldDoc2()); |
| 404 | ki.commit(); |
| 405 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 406 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 407 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 408 | RelationDirection.RIGHT); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 409 | |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 410 | SpanFocusQuery fq = new SpanFocusQuery(srq, srq.getTempClassNumbers()); |
| 411 | fq.setMatchTemporaryClass(true); |
| 412 | fq.setRemoveTemporaryClasses(true); |
| 413 | fq.setSorted(false); |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 414 | // kr = ki.search(fq, (short) 20); |
| 415 | // for (Match m : kr.getMatches()) { |
| 416 | // System.out.println(m.getStartPos() + " " + m.getEndPos()); |
| 417 | // } |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 418 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 419 | SpanAttributeQuery saq = new SpanAttributeQuery( |
| 420 | new SpanTermQuery(new Term("base", "@:func=sbj")), true); |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 421 | // kr = ki.search(saq, (short) 20); |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 422 | |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 423 | // child-of with attr func=sbj |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 424 | SpanWithAttributeQuery wq; |
| margaretha | 29a3911 | 2016-04-04 12:10:11 +0200 | [diff] [blame] | 425 | wq = new SpanWithAttributeQuery(fq, saq, true); |
| Akron | 4299355 | 2016-02-04 13:24:24 +0100 | [diff] [blame] | 426 | // kr = ki.search(wq, (short) 20); |
| 427 | // assertEquals((long) 1, kr.getTotalResults()); |
| 428 | // assertEquals(0, kr.getMatch(0).getStartPos()); // token |
| 429 | // assertEquals(7, kr.getMatch(0).getEndPos()); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 430 | |
| 431 | // child-of without attr func=sbj |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 432 | wq = new SpanWithAttributeQuery(fq, |
| 433 | new SpanAttributeQuery( |
| 434 | new SpanTermQuery(new Term("base", "@:func=sbj")), true, |
| 435 | true), |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 436 | true); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 437 | kr = ki.search(wq, (short) 20); |
| 438 | assertEquals((long) 12, kr.getTotalResults()); |
| 439 | } |
| 440 | |
| 441 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 442 | /** |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 443 | * Relation directions <br/> |
| 444 | * Relation with specific sources, return the sources |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 445 | */ |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 446 | @Test |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 447 | public void testCase6 () throws IOException { |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 448 | ki.addDoc(createFieldDoc2()); |
| 449 | ki.commit(); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 450 | |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 451 | // return all children that are NP |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 452 | SpanElementQuery seq1 = new SpanElementQuery("base", "np"); |
| 453 | SpanClassQuery scq1 = new SpanClassQuery(seq1, (byte) 1); |
| 454 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 455 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 456 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 457 | RelationDirection.RIGHT); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 458 | |
| 459 | SpanRelationMatchQuery rm = new SpanRelationMatchQuery(srq, scq1, true); |
| 460 | SpanFocusQuery rv = new SpanFocusQuery(rm, (byte) 1); |
| 461 | rv.setSorted(false); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 462 | |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 463 | assertEquals( |
| Akron | a26184e | 2018-12-05 15:37:34 +0100 | [diff] [blame] | 464 | "focus(1: focus(#[1,2]spanSegment(spanRelation(base:>:child-of), {1: <base:np />}),sorting),sorting)", |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 465 | rv.toString()); |
| 466 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 467 | kr = ki.search(rv, (short) 10); |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 468 | |
| 469 | assertEquals(4, kr.getTotalResults()); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 470 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 471 | assertEquals(1, kr.getMatch(0).getEndPos()); |
| 472 | assertEquals(2, kr.getMatch(1).getStartPos()); |
| 473 | assertEquals(4, kr.getMatch(1).getEndPos()); |
| 474 | assertEquals(2, kr.getMatch(2).getStartPos()); |
| 475 | assertEquals(7, kr.getMatch(2).getEndPos()); |
| 476 | assertEquals(5, kr.getMatch(3).getStartPos()); |
| 477 | assertEquals(7, kr.getMatch(3).getEndPos()); |
| 478 | |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 479 | // return all parents that are NP |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 480 | srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 481 | new SpanTermQuery(new Term("base", "<:child-of")), true, |
| 482 | RelationDirection.LEFT); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 483 | rm = new SpanRelationMatchQuery(srq, scq1, true); |
| 484 | rv = new SpanFocusQuery(rm, (byte) 1); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 485 | kr = ki.search(rv, (short) 10); |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 486 | |
| 487 | assertEquals(7, kr.getTotalResults()); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 488 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 489 | assertEquals(1, kr.getMatch(0).getEndPos()); |
| 490 | assertEquals(2, kr.getMatch(1).getStartPos()); |
| 491 | assertEquals(4, kr.getMatch(1).getEndPos()); |
| 492 | assertEquals(2, kr.getMatch(2).getStartPos()); |
| 493 | assertEquals(4, kr.getMatch(2).getEndPos()); |
| 494 | assertEquals(2, kr.getMatch(3).getStartPos()); |
| 495 | assertEquals(7, kr.getMatch(3).getEndPos()); |
| 496 | assertEquals(2, kr.getMatch(4).getStartPos()); |
| 497 | assertEquals(7, kr.getMatch(4).getEndPos()); |
| 498 | assertEquals(5, kr.getMatch(5).getStartPos()); |
| 499 | assertEquals(7, kr.getMatch(5).getEndPos()); |
| 500 | assertEquals(5, kr.getMatch(6).getStartPos()); |
| 501 | assertEquals(7, kr.getMatch(6).getEndPos()); |
| 502 | } |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 503 | |
| Akron | bb5d173 | 2015-06-22 01:22:40 +0200 | [diff] [blame] | 504 | |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 505 | /** |
| 506 | * Dependency relations with attribute |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 507 | */ |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 508 | @Test |
| Akron | bb5d173 | 2015-06-22 01:22:40 +0200 | [diff] [blame] | 509 | public void testCase5 () throws IOException { |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 510 | ki.addDoc(createFieldDoc2()); |
| 511 | ki.commit(); |
| 512 | |
| 513 | // target of a dependency relation |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 514 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 515 | new SpanTermQuery(new Term("base", "<:dep")), true, |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 516 | RelationDirection.LEFT); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 517 | kr = ki.search(srq, (short) 10); |
| 518 | assertEquals((long) 6, kr.getTotalResults()); |
| 519 | |
| 520 | SpanFocusQuery fq = new SpanFocusQuery(srq, srq.getTempClassNumbers()); |
| 521 | fq.setMatchTemporaryClass(true); |
| 522 | fq.setRemoveTemporaryClasses(true); |
| 523 | // fq.setSorted(false); |
| 524 | |
| 525 | kr = ki.search(fq, (short) 10); |
| 526 | assertEquals((long) 6, kr.getTotalResults()); |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 527 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 528 | SpanAttributeQuery aq = new SpanAttributeQuery( |
| 529 | new SpanTermQuery(new Term("base", "@:func=head")), true); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 530 | kr = ki.search(aq, (short) 10); |
| 531 | |
| 532 | // dependency relation, which is also a head |
| margaretha | 7ee6595 | 2015-12-14 15:39:12 +0100 | [diff] [blame] | 533 | SpanWithAttributeQuery wq = new SpanWithAttributeQuery(fq, aq, true); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 534 | |
| 535 | kr = ki.search(wq, (short) 20); |
| 536 | |
| 537 | assertEquals((long) 2, kr.getTotalResults()); |
| 538 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 539 | assertEquals(4, kr.getMatch(0).getEndPos()); |
| 540 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 541 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 542 | |
| 543 | } |
| Eliza Margaretha | 2db5e23 | 2015-03-04 10:20:01 +0000 | [diff] [blame] | 544 | |
| Akron | bb5d173 | 2015-06-22 01:22:40 +0200 | [diff] [blame] | 545 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 546 | /** |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 547 | * Relation with specific sources, return any targets |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 548 | */ |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 549 | @Test |
| 550 | public void testCase7 () throws IOException { |
| 551 | ki.addDoc(createFieldDoc2()); |
| 552 | ki.commit(); |
| Eliza Margaretha | d12cabb | 2014-10-27 17:45:34 +0000 | [diff] [blame] | 553 | |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 554 | // match all children that are NP |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 555 | SpanElementQuery seq1 = new SpanElementQuery("base", "np"); |
| 556 | SpanClassQuery scq1 = new SpanClassQuery(seq1, (byte) 1); |
| 557 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 558 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 559 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 560 | RelationDirection.RIGHT); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 561 | srq.setTargetClass((byte) 2); |
| 562 | |
| 563 | SpanRelationMatchQuery rm = new SpanRelationMatchQuery(srq, scq1, true); |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 564 | // SpanQuery rv = new SpanFocusQuery(rm, (byte) 1); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 565 | |
| 566 | //return all parents of np |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 567 | SpanFocusQuery rv2 = new SpanFocusQuery(rm, (byte) 2); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 568 | rv2.setSorted(false); |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 569 | |
| 570 | assertEquals( |
| Akron | a26184e | 2018-12-05 15:37:34 +0100 | [diff] [blame] | 571 | "focus(2: focus(#[1,2]spanSegment({2: target:spanRelation(base:>:child-of)}, {1: <base:np />}),sorting),sorting)", |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 572 | rv2.toString()); |
| 573 | |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 574 | kr = ki.search(rv2, (short) 10); |
| 575 | |
| 576 | assertEquals((long) 4, kr.getTotalResults()); |
| 577 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 578 | assertEquals(7, kr.getMatch(0).getEndPos()); |
| 579 | assertEquals(1, kr.getMatch(1).getStartPos()); |
| 580 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 581 | assertEquals(2, kr.getMatch(2).getStartPos()); |
| 582 | assertEquals(7, kr.getMatch(2).getEndPos()); |
| 583 | assertEquals(4, kr.getMatch(3).getStartPos()); |
| 584 | assertEquals(7, kr.getMatch(3).getEndPos()); |
| 585 | // id problem (solved) |
| 586 | |
| 587 | // return all parents of np that are PP |
| 588 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 589 | } |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 590 | |
| 591 | |
| 592 | /** |
| 593 | * Relations whose source/target do not embed |
| 594 | * its counterparts. |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 595 | */ |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 596 | @Test |
| 597 | public void testCase8 () throws IOException { |
| 598 | ki.addDoc(createFieldDoc2()); |
| 599 | ki.commit(); |
| 600 | |
| 601 | //return source of dep relations to pos:NN |
| Akron | bb5d173 | 2015-06-22 01:22:40 +0200 | [diff] [blame] | 602 | |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 603 | SpanTermWithIdQuery tq = |
| 604 | new SpanTermWithIdQuery(new Term("base", "pos:NN"), true); |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 605 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 606 | new SpanTermQuery(new Term("base", "<:dep")), true, |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 607 | RelationDirection.LEFT); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 608 | srq.setSourceClass((byte) 1); |
| 609 | SpanRelationMatchQuery rm = new SpanRelationMatchQuery(srq, tq, true); |
| 610 | SpanQuery rv = new SpanFocusQuery(rm, (byte) 1); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 611 | |
| 612 | kr = ki.search(rv, (short) 10); |
| 613 | assertEquals((long) 3, kr.getTotalResults()); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 614 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 615 | assertEquals(1, kr.getMatch(0).getEndPos()); |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 616 | assertEquals(1, kr.getMatch(1).getStartPos()); |
| 617 | assertEquals(2, kr.getMatch(1).getEndPos()); |
| 618 | assertEquals(4, kr.getMatch(2).getStartPos()); |
| 619 | assertEquals(5, kr.getMatch(2).getEndPos()); |
| 620 | |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 621 | // return target of dep relations from pos:NN |
| 622 | srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 623 | new SpanTermQuery(new Term("base", ">:dep")), true, |
| margaretha | f09fdd4 | 2017-12-19 17:47:50 +0100 | [diff] [blame] | 624 | RelationDirection.RIGHT); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 625 | srq.setTargetClass((byte) 1); |
| 626 | rm = new SpanRelationMatchQuery(srq, tq, true); |
| 627 | rv = new SpanFocusQuery(rm, (byte) 1); |
| 628 | |
| Nils Diewald | bb33da2 | 2015-03-04 16:24:25 +0000 | [diff] [blame] | 629 | kr = ki.search(rv, (short) 10); |
| 630 | assertEquals((long) 3, kr.getTotalResults()); |
| 631 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 632 | assertEquals(3, kr.getMatch(0).getEndPos()); |
| 633 | assertEquals(4, kr.getMatch(1).getStartPos()); |
| 634 | assertEquals(5, kr.getMatch(1).getEndPos()); |
| 635 | assertEquals(5, kr.getMatch(2).getStartPos()); |
| 636 | assertEquals(6, kr.getMatch(2).getEndPos()); |
| 637 | |
| 638 | } |
| 639 | |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 640 | |
| 641 | /** |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 642 | * Relation with specific sources and return any targets <br/> |
| Akron | bb5d173 | 2015-06-22 01:22:40 +0200 | [diff] [blame] | 643 | * Relation with specific sources and targets, return the targets |
| 644 | * <br/> |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 645 | * Relation with specific sources and targets, return the sources |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 646 | * |
| 647 | * @throws IOException |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 648 | */ |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 649 | @Test |
| 650 | public void testCase9 () throws IOException { |
| 651 | ki.addDoc(createFieldDoc2()); |
| 652 | ki.commit(); |
| 653 | |
| 654 | // return all children of np |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 655 | SpanElementQuery seq1 = new SpanElementQuery("base", "np"); |
| 656 | SpanClassQuery scq1 = new SpanClassQuery(seq1, (byte) 1); |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 657 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 658 | new SpanTermQuery(new Term("base", "<:child-of")), true, |
| 659 | RelationDirection.LEFT); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 660 | srq.setSourceClass((byte) 2); |
| 661 | SpanRelationMatchQuery rm = new SpanRelationMatchQuery(srq, scq1, true); |
| 662 | SpanFocusQuery rv = new SpanFocusQuery(rm, (byte) 2); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 663 | rv.setSorted(false); |
| 664 | |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 665 | assertEquals( |
| Akron | a26184e | 2018-12-05 15:37:34 +0100 | [diff] [blame] | 666 | "focus(2: focus(#[1,2]spanSegment({2: source:spanRelation(base:<:child-of)}, {1: <base:np />}),sorting),sorting)", |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 667 | rv.toString()); |
| 668 | |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 669 | kr = ki.search(rv, (short) 10); |
| 670 | |
| 671 | assertEquals((long) 7, kr.getTotalResults()); |
| 672 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 673 | assertEquals(1, kr.getMatch(0).getEndPos()); |
| 674 | assertEquals(2, kr.getMatch(1).getStartPos()); |
| 675 | assertEquals(3, kr.getMatch(1).getEndPos()); |
| 676 | assertEquals(2, kr.getMatch(2).getStartPos()); |
| 677 | assertEquals(4, kr.getMatch(2).getEndPos()); |
| 678 | assertEquals(3, kr.getMatch(3).getStartPos()); |
| 679 | assertEquals(4, kr.getMatch(3).getEndPos()); |
| 680 | assertEquals(4, kr.getMatch(4).getStartPos()); |
| 681 | assertEquals(7, kr.getMatch(4).getEndPos()); |
| 682 | assertEquals(5, kr.getMatch(5).getStartPos()); |
| 683 | assertEquals(6, kr.getMatch(5).getEndPos()); |
| 684 | assertEquals(6, kr.getMatch(6).getStartPos()); |
| 685 | assertEquals(7, kr.getMatch(6).getEndPos()); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 686 | } |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 687 | |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 688 | |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 689 | @Test |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 690 | public void testCase9b () throws IOException { |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 691 | ki.addDoc(createFieldDoc2()); |
| 692 | ki.commit(); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 693 | |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 694 | SpanTermWithIdQuery tiq = |
| 695 | new SpanTermWithIdQuery(new Term("base", "pos:ART"), true); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 696 | SpanClassQuery scq1 = new SpanClassQuery(tiq, (byte) 1); |
| 697 | SpanElementQuery seq = new SpanElementQuery("base", "np"); |
| 698 | SpanClassQuery scq2 = new SpanClassQuery(seq, (byte) 2); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 699 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 700 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 701 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 702 | RelationDirection.RIGHT); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 703 | srq.setSourceClass((byte) 1); |
| 704 | srq.setTargetClass((byte) 2); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 705 | |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 706 | SpanRelationMatchQuery rm = |
| 707 | new SpanRelationMatchQuery(srq, scq1, scq2, true); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 708 | |
| 709 | // return all nps whose children are articles |
| 710 | SpanFocusQuery rv = new SpanFocusQuery(rm, (byte) 2); |
| 711 | rv.setSorted(false); |
| margaretha | b9ee3e5 | 2015-05-04 12:42:23 +0200 | [diff] [blame] | 712 | |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 713 | kr = ki.search(rv, (short) 10); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 714 | assertEquals((long) 2, kr.getTotalResults()); |
| 715 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 716 | assertEquals(4, kr.getMatch(0).getEndPos()); |
| 717 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 718 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 719 | |
| 720 | // for (Match m : kr.getMatches()) { |
| 721 | // System.out.println(m.getStartPos() + " " + m.getEndPos()); |
| 722 | // } |
| 723 | } |
| 724 | |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 725 | |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 726 | @Test |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 727 | public void testCase9c () throws IOException { |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 728 | ki.addDoc(createFieldDoc2()); |
| 729 | ki.commit(); |
| 730 | |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 731 | SpanTermWithIdQuery tiq = |
| 732 | new SpanTermWithIdQuery(new Term("base", "pos:ART"), true); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 733 | SpanClassQuery scq1 = new SpanClassQuery(tiq, (byte) 1); |
| 734 | |
| 735 | kr = ki.search(scq1, (short) 10); |
| 736 | assertEquals((long) 2, kr.getTotalResults()); |
| 737 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 738 | assertEquals(3, kr.getMatch(0).getEndPos()); |
| 739 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 740 | assertEquals(6, kr.getMatch(1).getEndPos()); |
| 741 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 742 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 743 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 744 | RelationDirection.RIGHT); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 745 | srq.setSourceClass((byte) 1); |
| 746 | srq.setTargetClass((byte) 2); |
| 747 | |
| 748 | // match articles as relation sources (left side) |
| 749 | SpanSegmentQuery ssq1 = new SpanSegmentQuery(srq, scq1); |
| 750 | // NOTE: SegmentSpans can only always adopt the relation left |
| 751 | // side id. |
| 752 | |
| 753 | kr = ki.search(ssq1, (short) 10); |
| 754 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 755 | assertEquals(3, kr.getMatch(0).getEndPos()); |
| 756 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 757 | assertEquals(6, kr.getMatch(1).getEndPos()); |
| 758 | |
| 759 | // return all parents of articles |
| 760 | SpanFocusQuery sfq = new SpanFocusQuery(ssq1, (byte) 2); |
| 761 | sfq.setSorted(false); |
| 762 | |
| 763 | kr = ki.search(sfq, (short) 10); |
| 764 | assertEquals((long) 2, kr.getTotalResults()); |
| 765 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 766 | assertEquals(4, kr.getMatch(0).getEndPos()); |
| 767 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 768 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 769 | |
| 770 | SpanElementQuery seq = new SpanElementQuery("base", "np"); |
| 771 | SpanClassQuery scq2 = new SpanClassQuery(seq, (byte) 2); |
| 772 | |
| 773 | kr = ki.search(scq2, (short) 10); |
| 774 | assertEquals((long) 4, kr.getTotalResults()); |
| 775 | assertEquals(0, kr.getMatch(0).getStartPos()); |
| 776 | assertEquals(1, kr.getMatch(0).getEndPos()); |
| 777 | assertEquals(2, kr.getMatch(1).getStartPos()); |
| 778 | assertEquals(4, kr.getMatch(1).getEndPos()); |
| 779 | assertEquals(2, kr.getMatch(2).getStartPos()); |
| 780 | assertEquals(7, kr.getMatch(2).getEndPos()); |
| 781 | assertEquals(5, kr.getMatch(3).getStartPos()); |
| 782 | assertEquals(7, kr.getMatch(3).getEndPos()); |
| 783 | |
| 784 | // match nps as relation targets (right side) |
| 785 | // return all nps whose children are articles |
| 786 | SpanSegmentQuery ssq2 = new SpanSegmentQuery(sfq, scq2); |
| 787 | kr = ki.search(ssq2, (short) 10); |
| 788 | assertEquals((long) 2, kr.getTotalResults()); |
| 789 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 790 | assertEquals(4, kr.getMatch(0).getEndPos()); |
| 791 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 792 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 793 | |
| 794 | // NOTE: cannot match the span id of the embedded relation |
| 795 | // target and the np. So the segment match here is only by |
| 796 | // positions. All nps in the same position will be matches. |
| 797 | |
| 798 | // return the articles whos parents are nps |
| 799 | SpanFocusQuery sfq2 = new SpanFocusQuery(ssq2, (byte) 1); |
| 800 | sfq2.setSorted(false); |
| 801 | assertEquals("focus(1: spanSegment(focus(2: spanSegment({1: source:" |
| 802 | + "{2: target:spanRelation(base:>:child-of)}}, " |
| Akron | a26184e | 2018-12-05 15:37:34 +0100 | [diff] [blame] | 803 | + "{1: spanTermWithId(base:pos:ART)}),sorting), " |
| 804 | + "{2: <base:np />}),sorting)", sfq2.toString()); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 805 | |
| 806 | kr = ki.search(sfq2, (short) 10); |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 807 | |
| 808 | assertEquals((long) 2, kr.getTotalResults()); |
| 809 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 810 | assertEquals(3, kr.getMatch(0).getEndPos()); |
| 811 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 812 | assertEquals(6, kr.getMatch(1).getEndPos()); |
| 813 | |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 814 | // for (Match m : kr.getMatches()) { |
| 815 | // System.out.println(m.getStartPos() + " " + m.getEndPos()); |
| 816 | // } |
| margaretha | 50c7633 | 2015-03-19 10:10:39 +0100 | [diff] [blame] | 817 | } |
| 818 | |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 819 | |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 820 | /** |
| 821 | * Relation whose nodes have a specific attribute. |
| 822 | * |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 823 | */ |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 824 | @Test |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 825 | public void testCase10a () throws IOException { |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 826 | ki.addDoc(createFieldDoc2()); |
| 827 | ki.commit(); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 828 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 829 | SpanAttributeQuery aq = new SpanAttributeQuery( |
| 830 | new SpanTermQuery(new Term("base", "@:case=accusative")), true); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 831 | kr = ki.search(aq, (short) 10); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 832 | assertEquals((long) 3, kr.getTotalResults()); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 833 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 834 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 835 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 836 | RelationDirection.RIGHT); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 837 | srq.setSourceClass((byte) 1); |
| 838 | srq.setTargetClass((byte) 2); |
| 839 | kr = ki.search(srq, (short) 20); |
| 840 | assertEquals((long) 13, kr.getTotalResults()); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 841 | |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 842 | // Matching relation source node with an attribute |
| 843 | SpanFocusQuery sfq1 = new SpanFocusQuery(srq, (byte) 1); |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 844 | SpanWithAttributeQuery swaq = |
| 845 | new SpanWithAttributeQuery(sfq1, aq, true); |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 846 | |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 847 | kr = ki.search(swaq, (short) 10); |
| 848 | assertEquals((long) 2, kr.getTotalResults()); |
| 849 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 850 | assertEquals(4, kr.getMatch(0).getEndPos()); |
| 851 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 852 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 853 | |
| 854 | // Returning relations whose source has a specific attribute |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 855 | SpanFocusQuery fqr = |
| 856 | new SpanFocusQuery(swaq, srq.getTempClassNumbers()); |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 857 | fqr.setMatchTemporaryClass(true); |
| 858 | fqr.setRemoveTemporaryClasses(true); |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 859 | assertEquals( |
| 860 | "focus(#[1,2]spanRelationWithAttribute(focus(1: " |
| 861 | + "{1: source:{2: target:spanRelation(base:>:child-of)}}), " |
| 862 | + "spanAttribute(base:@:case=accusative)))", |
| 863 | fqr.toString()); |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 864 | |
| 865 | kr = ki.search(fqr, (short) 10); |
| 866 | assertEquals((long) 2, kr.getTotalResults()); |
| 867 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 868 | assertEquals(7, kr.getMatch(0).getEndPos()); |
| 869 | assertEquals(4, kr.getMatch(1).getStartPos()); |
| 870 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 871 | |
| 872 | // Matching relation target nodes with an attribute |
| 873 | SpanFocusQuery sfq2 = new SpanFocusQuery(srq, (byte) 2); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 874 | sfq2.setSorted(false); |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 875 | SpanWithAttributeQuery swaq2 = |
| 876 | new SpanWithAttributeQuery(sfq2, aq, true); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 877 | |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 878 | kr = ki.search(aq, (short) 20); |
| 879 | |
| 880 | // for (Match m : kr.getMatches()) { |
| 881 | // System.out.println(m.getStartPos() + " " + m.getEndPos()); |
| 882 | // } |
| 883 | kr = ki.search(swaq2, (short) 10); |
| 884 | assertEquals((long) 2, kr.getTotalResults()); |
| 885 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 886 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 887 | |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 888 | // Returning relations whose target has a specific attribute |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 889 | SpanFocusQuery fqr2 = |
| 890 | new SpanFocusQuery(swaq2, srq.getTempClassNumbers()); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 891 | fqr2.setMatchTemporaryClass(true); |
| 892 | fqr2.setRemoveTemporaryClasses(true); |
| 893 | |
| 894 | kr = ki.search(fqr2, (short) 10); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 895 | assertEquals((long) 2, kr.getTotalResults()); |
| 896 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 897 | assertEquals(7, kr.getMatch(0).getEndPos()); |
| 898 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 899 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 900 | } |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 901 | |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 902 | |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 903 | /** |
| 904 | * Relation whose nodes have a specific attribute. Alternative |
| 905 | * query (actually used in serialization) |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 906 | */ |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 907 | @Test |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 908 | public void testCase10b () throws IOException { |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 909 | ki.addDoc(createFieldDoc2()); |
| 910 | ki.commit(); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 911 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 912 | SpanAttributeQuery aq = new SpanAttributeQuery( |
| 913 | new SpanTermQuery(new Term("base", "@:case=accusative")), true); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 914 | kr = ki.search(aq, (short) 10); |
| 915 | assertEquals((long) 3, kr.getTotalResults()); |
| 916 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 917 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 918 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 919 | RelationDirection.RIGHT); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 920 | srq.setSourceClass((byte) 1); |
| 921 | srq.setTargetClass((byte) 2); |
| 922 | kr = ki.search(srq, (short) 20); |
| 923 | assertEquals((long) 13, kr.getTotalResults()); |
| 924 | |
| 925 | // Matching relation source node with an attribute |
| 926 | SpanWithAttributeQuery swaq = new SpanWithAttributeQuery(aq, true); |
| 927 | SpanSegmentQuery ssq = new SpanSegmentQuery(srq, swaq); |
| 928 | |
| 929 | assertEquals( |
| 930 | "spanSegment({1: source:{2: target:spanRelation(base:>:child-of)}}, " |
| 931 | + "spanWithAttribute(spanAttribute(base:@:case=accusative)))", |
| 932 | ssq.toString()); |
| 933 | |
| 934 | kr = ki.search(ssq, (short) 10); |
| margaretha | 24d2d01 | 2016-04-13 18:43:30 +0200 | [diff] [blame] | 935 | assertEquals((long) 2, kr.getTotalResults()); |
| 936 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 937 | assertEquals(4, kr.getMatch(0).getEndPos()); |
| 938 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 939 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 940 | |
| 941 | // Matching relation target nodes with an attribute |
| 942 | // NOTE: swaq must be the first parameter |
| 943 | SpanFocusQuery sfq2 = new SpanFocusQuery(srq, (byte) 2); |
| 944 | sfq2.setSorted(false); |
| 945 | SpanSegmentQuery ssq2 = new SpanSegmentQuery(swaq, sfq2); |
| 946 | |
| 947 | kr = ki.search(ssq2, (short) 10); |
| 948 | assertEquals((long) 2, kr.getTotalResults()); |
| 949 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 950 | assertEquals(7, kr.getMatch(0).getEndPos()); |
| 951 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 952 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 953 | |
| 954 | // Matching specific relation source node with an attribute |
| 955 | SpanElementQuery seq = new SpanElementQuery("base", "np"); |
| 956 | swaq = new SpanWithAttributeQuery(seq, aq, true); |
| 957 | ssq = new SpanSegmentQuery(srq, swaq); |
| 958 | |
| 959 | assertEquals( |
| 960 | "spanSegment({1: source:{2: target:spanRelation(base:>:child-of)}}, " |
| 961 | + "spanElementWithAttribute(<base:np />, " |
| 962 | + "spanAttribute(base:@:case=accusative)))", |
| 963 | ssq.toString()); |
| 964 | |
| 965 | kr = ki.search(ssq, (short) 10); |
| 966 | assertEquals((long) 2, kr.getTotalResults()); |
| 967 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 968 | assertEquals(4, kr.getMatch(0).getEndPos()); |
| 969 | assertEquals(5, kr.getMatch(1).getStartPos()); |
| 970 | assertEquals(7, kr.getMatch(1).getEndPos()); |
| 971 | } |
| 972 | |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 973 | |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 974 | /** |
| 975 | * Matching both relation nodes whose a specific attribute |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 976 | */ |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 977 | @Test |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 978 | public void testCase11 () throws IOException { |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 979 | ki.addDoc(createFieldDoc2()); |
| 980 | ki.commit(); |
| 981 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 982 | SpanAttributeQuery aq = new SpanAttributeQuery( |
| 983 | new SpanTermQuery(new Term("base", "@:case=accusative")), true); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 984 | kr = ki.search(aq, (short) 10); |
| 985 | assertEquals((long) 3, kr.getTotalResults()); |
| 986 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 987 | SpanAttributeQuery aq2 = new SpanAttributeQuery( |
| 988 | new SpanTermQuery(new Term("base", "@:case=accusative")), true); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 989 | kr = ki.search(aq2, (short) 10); |
| 990 | assertEquals((long) 3, kr.getTotalResults()); |
| 991 | |
| Eliza Margaretha | 6f98920 | 2016-10-14 21:48:29 +0200 | [diff] [blame] | 992 | SpanRelationQuery srq = new SpanRelationQuery( |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 993 | new SpanTermQuery(new Term("base", ">:child-of")), true, |
| 994 | RelationDirection.RIGHT); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 995 | srq.setSourceClass((byte) 1); |
| 996 | srq.setTargetClass((byte) 2); |
| 997 | kr = ki.search(srq, (short) 20); |
| 998 | assertEquals((long) 13, kr.getTotalResults()); |
| Akron | 6759b04 | 2016-04-28 01:25:00 +0200 | [diff] [blame] | 999 | |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 1000 | SpanWithAttributeQuery swaq1 = new SpanWithAttributeQuery(aq, true); |
| 1001 | SpanWithAttributeQuery swaq2 = new SpanWithAttributeQuery(aq2, true); |
| 1002 | |
| 1003 | kr = ki.search(swaq1, (short) 10); |
| 1004 | assertEquals((long) 3, kr.getTotalResults()); |
| 1005 | |
| margaretha | 362c18b | 2018-01-17 17:30:56 +0100 | [diff] [blame] | 1006 | SpanRelationMatchQuery srmq = |
| 1007 | new SpanRelationMatchQuery(srq, swaq1, swaq2, true); |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 1008 | assertEquals( |
| 1009 | "focus(#[1,2]spanSegment(spanWithAttribute(spanAttribute(base:@:case=accusative)), " |
| 1010 | + "focus(#2: spanSegment({1: source:{2: target:spanRelation(base:>:child-of)}}, " |
| Akron | a26184e | 2018-12-05 15:37:34 +0100 | [diff] [blame] | 1011 | + "spanWithAttribute(spanAttribute(base:@:case=accusative))),sorting)),sorting)", |
| margaretha | 32d2bee | 2016-04-22 16:06:53 +0200 | [diff] [blame] | 1012 | srmq.toString()); |
| 1013 | kr = ki.search(srmq, (short) 10); |
| 1014 | assertEquals((long) 1, kr.getTotalResults()); |
| 1015 | assertEquals(2, kr.getMatch(0).getStartPos()); |
| 1016 | assertEquals(7, kr.getMatch(0).getEndPos()); |
| margaretha | f70addb | 2015-04-27 13:17:18 +0200 | [diff] [blame] | 1017 | } |
| margaretha | c44b228 | 2016-04-26 16:40:01 +0200 | [diff] [blame] | 1018 | |
| Akron | 700c1eb | 2015-09-25 16:57:30 +0200 | [diff] [blame] | 1019 | } |