|
| 1 | +/* |
| 2 | + * Licensed to the Apache Software Foundation (ASF) under one |
| 3 | + * or more contributor license agreements. See the NOTICE file |
| 4 | + * distributed with this work for additional information |
| 5 | + * regarding copyright ownership. The ASF licenses this file |
| 6 | + * to you under the Apache License, Version 2.0 (the |
| 7 | + * "License"); you may not use this file except in compliance |
| 8 | + * with the License. You may obtain a copy of the License at |
| 9 | + * |
| 10 | + * http://www.apache.org/licenses/LICENSE-2.0 |
| 11 | + * |
| 12 | + * Unless required by applicable law or agreed to in writing, software |
| 13 | + * distributed under the License is distributed on an "AS IS" BASIS, |
| 14 | + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 15 | + * See the License for the specific language governing permissions and |
| 16 | + * limitations under the License. |
| 17 | + */ |
| 18 | + |
| 19 | +package org.apache.paimon.utils; |
| 20 | + |
| 21 | +import org.apache.paimon.fs.Path; |
| 22 | + |
| 23 | +import org.junit.jupiter.api.Test; |
| 24 | + |
| 25 | +import static org.assertj.core.api.Assertions.assertThat; |
| 26 | + |
| 27 | +/** Tests for {@link FileType}. */ |
| 28 | +public class FileTypeTest { |
| 29 | + |
| 30 | + private static final String TABLE_ROOT = "hdfs://cluster/warehouse/db.db/table"; |
| 31 | + |
| 32 | + // ===== META files ===== |
| 33 | + |
| 34 | + @Test |
| 35 | + public void testMetaFiles() { |
| 36 | + // snapshot |
| 37 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/snapshot/snapshot-1"))) |
| 38 | + .isEqualTo(FileType.META); |
| 39 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/snapshot/snapshot-100"))) |
| 40 | + .isEqualTo(FileType.META); |
| 41 | + // schema |
| 42 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/schema/schema-0"))) |
| 43 | + .isEqualTo(FileType.META); |
| 44 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/schema/schema-5"))) |
| 45 | + .isEqualTo(FileType.META); |
| 46 | + // manifest |
| 47 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/manifest/manifest-a1b2c3d4-0"))) |
| 48 | + .isEqualTo(FileType.META); |
| 49 | + // manifest-list |
| 50 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/manifest/manifest-list-a1b2c3d4-0"))) |
| 51 | + .isEqualTo(FileType.META); |
| 52 | + // index-manifest |
| 53 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/manifest/index-manifest-a1b2c3d4-0"))) |
| 54 | + .isEqualTo(FileType.META); |
| 55 | + // statistics |
| 56 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/statistics/stat-a1b2c3d4-0"))) |
| 57 | + .isEqualTo(FileType.META); |
| 58 | + // tag |
| 59 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/tag/tag-2024-01-01"))) |
| 60 | + .isEqualTo(FileType.META); |
| 61 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/tag/tag-myTag"))) |
| 62 | + .isEqualTo(FileType.META); |
| 63 | + // changelog metadata |
| 64 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/changelog/changelog-1"))) |
| 65 | + .isEqualTo(FileType.META); |
| 66 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/changelog/changelog-100"))) |
| 67 | + .isEqualTo(FileType.META); |
| 68 | + // hint files |
| 69 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/snapshot/EARLIEST"))) |
| 70 | + .isEqualTo(FileType.META); |
| 71 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/snapshot/LATEST"))) |
| 72 | + .isEqualTo(FileType.META); |
| 73 | + // success files |
| 74 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/dt=2024-01-01/bucket-0/_SUCCESS"))) |
| 75 | + .isEqualTo(FileType.META); |
| 76 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/tag/tag-success-file/myTag_SUCCESS"))) |
| 77 | + .isEqualTo(FileType.META); |
| 78 | + // consumer |
| 79 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/consumer/consumer-myGroup"))) |
| 80 | + .isEqualTo(FileType.META); |
| 81 | + // service |
| 82 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/service/service-primary-key-lookup"))) |
| 83 | + .isEqualTo(FileType.META); |
| 84 | + } |
| 85 | + |
| 86 | + // ===== BUCKET_INDEX files ===== |
| 87 | + |
| 88 | + @Test |
| 89 | + public void testBucketIndexFiles() { |
| 90 | + // under /index/ dir |
| 91 | + assertThat(FileType.classify(new Path(TABLE_ROOT + "/index/index-a1b2c3d4-0"))) |
| 92 | + .isEqualTo(FileType.BUCKET_INDEX); |
| 93 | + // under bucket dir |
| 94 | + assertThat( |
| 95 | + FileType.classify( |
| 96 | + new Path(TABLE_ROOT + "/dt=2024-01-01/bucket-0/index-a1b2c3d4-0"))) |
| 97 | + .isEqualTo(FileType.BUCKET_INDEX); |
| 98 | + } |
| 99 | + |
| 100 | + // ===== GLOBAL_INDEX files ===== |
| 101 | + |
| 102 | + @Test |
| 103 | + public void testGlobalIndexFiles() { |
| 104 | + // btree global index |
| 105 | + assertThat( |
| 106 | + FileType.classify( |
| 107 | + new Path( |
| 108 | + TABLE_ROOT |
| 109 | + + "/index/btree-global-index-a1b2c3d4-e5f6.index"))) |
| 110 | + .isEqualTo(FileType.GLOBAL_INDEX); |
| 111 | + // bitmap global index |
| 112 | + assertThat( |
| 113 | + FileType.classify( |
| 114 | + new Path( |
| 115 | + TABLE_ROOT |
| 116 | + + "/index/bitmap-global-index-a1b2c3d4-e5f6.index"))) |
| 117 | + .isEqualTo(FileType.GLOBAL_INDEX); |
| 118 | + // lumina vector global index |
| 119 | + assertThat( |
| 120 | + FileType.classify( |
| 121 | + new Path( |
| 122 | + TABLE_ROOT |
| 123 | + + "/index/lumina-vector-ann-global-index-a1b2c3d4.index"))) |
| 124 | + .isEqualTo(FileType.GLOBAL_INDEX); |
| 125 | + // tantivy fulltext global index |
| 126 | + assertThat( |
| 127 | + FileType.classify( |
| 128 | + new Path( |
| 129 | + TABLE_ROOT |
| 130 | + + "/index/tantivy-fulltext-global-index-a1b2c3d4.index"))) |
| 131 | + .isEqualTo(FileType.GLOBAL_INDEX); |
| 132 | + } |
| 133 | + |
| 134 | + // ===== FILE_INDEX files ===== |
| 135 | + |
| 136 | + @Test |
| 137 | + public void testFileIndexFiles() { |
| 138 | + assertThat( |
| 139 | + FileType.classify( |
| 140 | + new Path( |
| 141 | + TABLE_ROOT |
| 142 | + + "/dt=2024-01-01/bucket-0/data-a1b2c3d4-0.orc.index"))) |
| 143 | + .isEqualTo(FileType.FILE_INDEX); |
| 144 | + assertThat( |
| 145 | + FileType.classify( |
| 146 | + new Path( |
| 147 | + TABLE_ROOT |
| 148 | + + "/dt=2024-01-01/bucket-0/data-a1b2c3d4-0.parquet.index"))) |
| 149 | + .isEqualTo(FileType.FILE_INDEX); |
| 150 | + } |
| 151 | + |
| 152 | + // ===== isIndex() ===== |
| 153 | + |
| 154 | + @Test |
| 155 | + public void testIsIndex() { |
| 156 | + assertThat(FileType.BUCKET_INDEX.isIndex()).isTrue(); |
| 157 | + assertThat(FileType.GLOBAL_INDEX.isIndex()).isTrue(); |
| 158 | + assertThat(FileType.FILE_INDEX.isIndex()).isTrue(); |
| 159 | + assertThat(FileType.META.isIndex()).isFalse(); |
| 160 | + assertThat(FileType.DATA.isIndex()).isFalse(); |
| 161 | + } |
| 162 | + |
| 163 | + // ===== DATA files ===== |
| 164 | + |
| 165 | + @Test |
| 166 | + public void testDataFiles() { |
| 167 | + // orc data file |
| 168 | + assertThat( |
| 169 | + FileType.classify( |
| 170 | + new Path( |
| 171 | + TABLE_ROOT |
| 172 | + + "/dt=2024-01-01/bucket-0/data-a1b2c3d4-0.orc"))) |
| 173 | + .isEqualTo(FileType.DATA); |
| 174 | + // parquet data file |
| 175 | + assertThat( |
| 176 | + FileType.classify( |
| 177 | + new Path( |
| 178 | + TABLE_ROOT |
| 179 | + + "/dt=2024-01-01/bucket-0/data-a1b2c3d4-0.parquet"))) |
| 180 | + .isEqualTo(FileType.DATA); |
| 181 | + // changelog data file |
| 182 | + assertThat( |
| 183 | + FileType.classify( |
| 184 | + new Path( |
| 185 | + TABLE_ROOT |
| 186 | + + "/dt=2024-01-01/bucket-0/changelog-a1b2c3d4-0.orc"))) |
| 187 | + .isEqualTo(FileType.DATA); |
| 188 | + // blob file |
| 189 | + assertThat( |
| 190 | + FileType.classify( |
| 191 | + new Path( |
| 192 | + TABLE_ROOT |
| 193 | + + "/dt=2024-01-01/bucket-0/data-a1b2c3d4-0.blob"))) |
| 194 | + .isEqualTo(FileType.DATA); |
| 195 | + // vector file |
| 196 | + assertThat( |
| 197 | + FileType.classify( |
| 198 | + new Path( |
| 199 | + TABLE_ROOT |
| 200 | + + "/dt=2024-01-01/bucket-0/data-a1b2c3d4-0.vector.lance"))) |
| 201 | + .isEqualTo(FileType.DATA); |
| 202 | + // unknown file defaults to DATA |
| 203 | + assertThat( |
| 204 | + FileType.classify( |
| 205 | + new Path(TABLE_ROOT + "/dt=2024-01-01/bucket-0/unknown-file.bin"))) |
| 206 | + .isEqualTo(FileType.DATA); |
| 207 | + } |
| 208 | + |
| 209 | + // ===== Edge cases ===== |
| 210 | + |
| 211 | + @Test |
| 212 | + public void testChangelogDirInParentPathNotMisjudged() { |
| 213 | + // table root path itself contains "changelog", should not be misjudged as META |
| 214 | + String tricky = "hdfs://cluster/changelog/warehouse/db.db/table"; |
| 215 | + assertThat( |
| 216 | + FileType.classify( |
| 217 | + new Path(tricky + "/dt=2024-01-01/bucket-0/data-a1b2c3d4-0.orc"))) |
| 218 | + .isEqualTo(FileType.DATA); |
| 219 | + assertThat( |
| 220 | + FileType.classify( |
| 221 | + new Path( |
| 222 | + tricky |
| 223 | + + "/dt=2024-01-01/bucket-0/changelog-a1b2c3d4-0.orc"))) |
| 224 | + .isEqualTo(FileType.DATA); |
| 225 | + } |
| 226 | + |
| 227 | + @Test |
| 228 | + public void testBranchPaths() { |
| 229 | + String branchRoot = TABLE_ROOT + "/branch/branch-dev"; |
| 230 | + assertThat(FileType.classify(new Path(branchRoot + "/snapshot/snapshot-1"))) |
| 231 | + .isEqualTo(FileType.META); |
| 232 | + assertThat(FileType.classify(new Path(branchRoot + "/schema/schema-0"))) |
| 233 | + .isEqualTo(FileType.META); |
| 234 | + assertThat(FileType.classify(new Path(branchRoot + "/changelog/changelog-1"))) |
| 235 | + .isEqualTo(FileType.META); |
| 236 | + assertThat(FileType.classify(new Path(branchRoot + "/index/index-a1b2c3d4-0"))) |
| 237 | + .isEqualTo(FileType.BUCKET_INDEX); |
| 238 | + assertThat( |
| 239 | + FileType.classify( |
| 240 | + new Path(branchRoot + "/index/btree-global-index-a1b2c3d4.index"))) |
| 241 | + .isEqualTo(FileType.GLOBAL_INDEX); |
| 242 | + } |
| 243 | +} |
0 commit comments