|
| 1 | +setup: |
| 2 | + - do: |
| 3 | + indices.create: |
| 4 | + index: test |
| 5 | + body: |
| 6 | + settings: |
| 7 | + number_of_shards: 1 |
| 8 | + mappings: |
| 9 | + properties: |
| 10 | + tags: |
| 11 | + type: text |
| 12 | + author: |
| 13 | + type: keyword |
| 14 | + number: |
| 15 | + type: integer |
| 16 | + class: |
| 17 | + type: integer |
| 18 | + |
| 19 | + - do: |
| 20 | + bulk: |
| 21 | + index: test |
| 22 | + refresh: true |
| 23 | + body: |
| 24 | + - '{"index": {}}' |
| 25 | + - '{"tags": "kibana", "author": "alice", "number": 1, "class": 1}' |
| 26 | + - '{"index": {}}' |
| 27 | + - '{"tags": "kibana", "author": "bob", "number": 2, "class": 1}' |
| 28 | + - '{"index": {}}' |
| 29 | + - '{"tags": "kibana", "author": "alice", "number": 3, "class": 2}' |
| 30 | + - '{"index": {}}' |
| 31 | + - '{"tags": "javascript", "author": "alice", "number": 4, "class": 1}' |
| 32 | + |
| 33 | +--- |
| 34 | +small shard_size: |
| 35 | + - do: |
| 36 | + search: |
| 37 | + rest_total_hits_as_int: true |
| 38 | + body: |
| 39 | + size: 0 |
| 40 | + query: |
| 41 | + query_string: |
| 42 | + query: 'tags:kibana OR tags:javascript' |
| 43 | + aggs: |
| 44 | + diversified: |
| 45 | + diversified_sampler: |
| 46 | + field: author |
| 47 | + shard_size: 1 |
| 48 | + aggs: |
| 49 | + min_number: |
| 50 | + min: |
| 51 | + field: number |
| 52 | + max_number: |
| 53 | + max: |
| 54 | + field: number |
| 55 | + |
| 56 | + |
| 57 | + - match: { hits.total: 4 } |
| 58 | + - match: { aggregations.diversified.doc_count: 1 } |
| 59 | + # The most relevant document has a value of 4 so we only aggregate that. |
| 60 | + - match: { aggregations.diversified.min_number.value: 4.0 } |
| 61 | + - match: { aggregations.diversified.max_number.value: 4.0 } |
| 62 | + |
| 63 | +--- |
| 64 | +defaults: |
| 65 | + - do: |
| 66 | + search: |
| 67 | + rest_total_hits_as_int: true |
| 68 | + body: |
| 69 | + size: 0 |
| 70 | + query: |
| 71 | + query_string: |
| 72 | + query: 'tags:kibana OR tags:javascript' |
| 73 | + aggs: |
| 74 | + diversified: |
| 75 | + diversified_sampler: |
| 76 | + field: author |
| 77 | + aggs: |
| 78 | + min_number: |
| 79 | + min: |
| 80 | + field: number |
| 81 | + max_number: |
| 82 | + max: |
| 83 | + field: number |
| 84 | + |
| 85 | + - match: { hits.total: 4 } |
| 86 | + # The max_docs_per_value defaults to 1 so we just get one doc for each of the two authors |
| 87 | + - match: { aggregations.diversified.doc_count: 2 } |
| 88 | + # Bob's only document is 2 so we get that as the min. |
| 89 | + - match: { aggregations.diversified.min_number.value: 2.0 } |
| 90 | + # Alice's most relevant document is 2 so we get that as the max. |
| 91 | + - match: { aggregations.diversified.max_number.value: 4.0 } |
| 92 | + |
| 93 | +--- |
| 94 | +override max_docs_per_value: |
| 95 | + - do: |
| 96 | + search: |
| 97 | + rest_total_hits_as_int: true |
| 98 | + body: |
| 99 | + size: 0 |
| 100 | + query: |
| 101 | + query_string: |
| 102 | + query: 'tags:kibana OR tags:javascript' |
| 103 | + aggs: |
| 104 | + diversified: |
| 105 | + diversified_sampler: |
| 106 | + field: author |
| 107 | + max_docs_per_value: 3 |
| 108 | + aggs: |
| 109 | + min_number: |
| 110 | + min: |
| 111 | + field: number |
| 112 | + max_number: |
| 113 | + max: |
| 114 | + field: number |
| 115 | + |
| 116 | + - match: { hits.total: 4 } |
| 117 | + # We've bumped the max_docs_per_value high enough to get all docs |
| 118 | + - match: { aggregations.diversified.doc_count: 4 } |
| 119 | + - match: { aggregations.diversified.min_number.value: 1.0 } |
| 120 | + - match: { aggregations.diversified.max_number.value: 4.0 } |
| 121 | + |
| 122 | +--- |
| 123 | +run on number: |
| 124 | + - do: |
| 125 | + search: |
| 126 | + rest_total_hits_as_int: true |
| 127 | + body: |
| 128 | + size: 0 |
| 129 | + query: |
| 130 | + query_string: |
| 131 | + query: 'tags:kibana OR tags:javascript' |
| 132 | + aggs: |
| 133 | + diversified: |
| 134 | + diversified_sampler: |
| 135 | + field: class |
| 136 | + aggs: |
| 137 | + min_number: |
| 138 | + min: |
| 139 | + field: number |
| 140 | + max_number: |
| 141 | + max: |
| 142 | + field: number |
| 143 | + |
| 144 | + - match: { hits.total: 4 } |
| 145 | + # The max_docs_per_value defaults to 1 so we just get one doc for each of the two classes |
| 146 | + - match: { aggregations.diversified.doc_count: 2 } |
| 147 | + - match: { aggregations.diversified.min_number.value: 3.0 } |
| 148 | + - match: { aggregations.diversified.max_number.value: 4.0 } |
| 149 | + |
| 150 | +--- |
| 151 | +force map mode: |
| 152 | + - do: |
| 153 | + search: |
| 154 | + rest_total_hits_as_int: true |
| 155 | + body: |
| 156 | + size: 0 |
| 157 | + query: |
| 158 | + query_string: |
| 159 | + query: 'tags:kibana OR tags:javascript' |
| 160 | + aggs: |
| 161 | + diversified: |
| 162 | + diversified_sampler: |
| 163 | + field: author |
| 164 | + execution_hint: map |
| 165 | + aggs: |
| 166 | + min_number: |
| 167 | + min: |
| 168 | + field: number |
| 169 | + max_number: |
| 170 | + max: |
| 171 | + field: number |
| 172 | + |
| 173 | + - match: { hits.total: 4 } |
| 174 | + - match: { aggregations.diversified.doc_count: 2 } |
| 175 | + - match: { aggregations.diversified.min_number.value: 2.0 } |
| 176 | + - match: { aggregations.diversified.max_number.value: 4.0 } |
| 177 | + |
| 178 | +--- |
| 179 | +force global ordinals mode: |
| 180 | + - do: |
| 181 | + search: |
| 182 | + rest_total_hits_as_int: true |
| 183 | + body: |
| 184 | + size: 0 |
| 185 | + query: |
| 186 | + query_string: |
| 187 | + query: 'tags:kibana OR tags:javascript' |
| 188 | + aggs: |
| 189 | + diversified: |
| 190 | + diversified_sampler: |
| 191 | + field: author |
| 192 | + execution_hint: global_ordinals |
| 193 | + aggs: |
| 194 | + min_number: |
| 195 | + min: |
| 196 | + field: number |
| 197 | + max_number: |
| 198 | + max: |
| 199 | + field: number |
| 200 | + |
| 201 | + - match: { hits.total: 4 } |
| 202 | + - match: { aggregations.diversified.doc_count: 2 } |
| 203 | + - match: { aggregations.diversified.min_number.value: 2.0 } |
| 204 | + - match: { aggregations.diversified.max_number.value: 4.0 } |
| 205 | + |
| 206 | +--- |
| 207 | +enable hash mode mode: |
| 208 | + - do: |
| 209 | + search: |
| 210 | + rest_total_hits_as_int: true |
| 211 | + body: |
| 212 | + size: 0 |
| 213 | + query: |
| 214 | + query_string: |
| 215 | + query: 'tags:kibana OR tags:javascript' |
| 216 | + aggs: |
| 217 | + diversified: |
| 218 | + diversified_sampler: |
| 219 | + field: author |
| 220 | + execution_hint: bytes_hash |
| 221 | + aggs: |
| 222 | + min_number: |
| 223 | + min: |
| 224 | + field: number |
| 225 | + max_number: |
| 226 | + max: |
| 227 | + field: number |
| 228 | + |
| 229 | + # This mode can have hash collisions. The hash is seeded with tests.seed |
| 230 | + # so we have to have weaker constraints on these hits |
| 231 | + - match: { hits.total: 4 } |
| 232 | + - lte: { aggregations.diversified.doc_count: 2 } |
| 233 | + - gte: { aggregations.diversified.doc_count: 1 } |
| 234 | + - gte: { aggregations.diversified.min_number.value: 2.0 } |
| 235 | + - lte: { aggregations.diversified.min_number.value: 4.0 } |
| 236 | + - match: { aggregations.diversified.max_number.value: 4.0 } |
0 commit comments