Hide keyboard shortcuts

Hot-keys on this page

r m x p   toggle line displays

j k   next/prev highlighted chunk

0   (zero) top of page

1   (one) first highlighted chunk

1# This file is part of daf_butler. 

2# 

3# Developed for the LSST Data Management System. 

4# This product includes software developed by the LSST Project 

5# (http://www.lsst.org). 

6# See the COPYRIGHT file at the top-level directory of this distribution 

7# for details of code ownership. 

8# 

9# This program is free software: you can redistribute it and/or modify 

10# it under the terms of the GNU General Public License as published by 

11# the Free Software Foundation, either version 3 of the License, or 

12# (at your option) any later version. 

13# 

14# This program is distributed in the hope that it will be useful, 

15# but WITHOUT ANY WARRANTY; without even the implied warranty of 

16# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 

17# GNU General Public License for more details. 

18# 

19# You should have received a copy of the GNU General Public License 

20# along with this program. If not, see <http://www.gnu.org/licenses/>. 

21 

22from __future__ import annotations 

23 

24__all__ = ["YamlRepoExportBackend", "YamlRepoImportBackend"] 

25 

26from datetime import datetime 

27from typing import ( 

28 Any, 

29 Dict, 

30 IO, 

31 Iterable, 

32 List, 

33 Mapping, 

34 Optional, 

35 Set, 

36 Tuple, 

37 Type, 

38) 

39import uuid 

40import warnings 

41from collections import defaultdict 

42 

43import yaml 

44import astropy.time 

45 

46from lsst.utils import doImport 

47from ..core import ( 

48 DatasetAssociation, 

49 DatasetId, 

50 DatasetRef, 

51 DatasetType, 

52 Datastore, 

53 DimensionElement, 

54 DimensionRecord, 

55 FileDataset, 

56 Timespan, 

57) 

58from ..core._butlerUri import ButlerURI 

59from ..core.utils import iterable 

60from ..core.named import NamedValueSet 

61from ..registry import CollectionType, Registry 

62from ..registry.interfaces import ( 

63 ChainedCollectionRecord, 

64 CollectionRecord, 

65 DatasetIdGenEnum, 

66 RunRecord, 

67 VersionTuple, 

68) 

69from ..registry.versions import IncompatibleVersionError 

70from ._interfaces import RepoExportBackend, RepoImportBackend 

71 

72 

73EXPORT_FORMAT_VERSION = VersionTuple(1, 0, 1) 

74"""Export format version. 

75 

76Files with a different major version or a newer minor version cannot be read by 

77this version of the code. 

78""" 

79 

80 

81def _uuid_representer(dumper: yaml.Dumper, data: uuid.UUID) -> yaml.Node: 

82 """Generate YAML representation for UUID. 

83 

84 This produces a scalar node with a tag "!uuid" and value being a regular 

85 string representation of UUID. 

86 """ 

87 return dumper.represent_scalar("!uuid", str(data)) 

88 

89 

90def _uuid_constructor(loader: yaml.Loader, node: yaml.Node) -> Optional[uuid.UUID]: 

91 if node.value is not None: 

92 return uuid.UUID(hex=node.value) 

93 return None 

94 

95 

96yaml.Dumper.add_representer(uuid.UUID, _uuid_representer) 

97yaml.SafeLoader.add_constructor("!uuid", _uuid_constructor) 

98 

99 

100class YamlRepoExportBackend(RepoExportBackend): 

101 """A repository export implementation that saves to a YAML file. 

102 

103 Parameters 

104 ---------- 

105 stream 

106 A writeable file-like object. 

107 """ 

108 

109 def __init__(self, stream: IO): 

110 self.stream = stream 

111 self.data: List[Dict[str, Any]] = [] 

112 

113 def saveDimensionData(self, element: DimensionElement, *data: DimensionRecord) -> None: 

114 # Docstring inherited from RepoExportBackend.saveDimensionData. 

115 data_dicts = [record.toDict(splitTimespan=True) for record in data] 

116 self.data.append({ 

117 "type": "dimension", 

118 "element": element.name, 

119 "records": data_dicts, 

120 }) 

121 

122 def saveCollection(self, record: CollectionRecord, doc: Optional[str]) -> None: 

123 # Docstring inherited from RepoExportBackend.saveCollections. 

124 data: Dict[str, Any] = { 

125 "type": "collection", 

126 "collection_type": record.type.name, 

127 "name": record.name, 

128 } 

129 if doc is not None: 

130 data["doc"] = doc 

131 if isinstance(record, RunRecord): 

132 data["host"] = record.host 

133 data["timespan_begin"] = record.timespan.begin 

134 data["timespan_end"] = record.timespan.end 

135 elif isinstance(record, ChainedCollectionRecord): 

136 data["children"] = list(record.children) 

137 self.data.append(data) 

138 

139 def saveDatasets(self, datasetType: DatasetType, run: str, *datasets: FileDataset) -> None: 

140 # Docstring inherited from RepoExportBackend.saveDatasets. 

141 self.data.append({ 

142 "type": "dataset_type", 

143 "name": datasetType.name, 

144 "dimensions": [d.name for d in datasetType.dimensions], 

145 "storage_class": datasetType.storageClass.name, 

146 "is_calibration": datasetType.isCalibration(), 

147 }) 

148 self.data.append({ 

149 "type": "dataset", 

150 "dataset_type": datasetType.name, 

151 "run": run, 

152 "records": [ 

153 { 

154 "dataset_id": [ref.id for ref in sorted(dataset.refs)], 

155 "data_id": [ref.dataId.byName() for ref in sorted(dataset.refs)], 

156 "path": dataset.path, 

157 "formatter": dataset.formatter, 

158 # TODO: look up and save other collections 

159 } 

160 for dataset in datasets 

161 ] 

162 }) 

163 

164 def saveDatasetAssociations(self, collection: str, collectionType: CollectionType, 

165 associations: Iterable[DatasetAssociation]) -> None: 

166 # Docstring inherited from RepoExportBackend.saveDatasetAssociations. 

167 if collectionType is CollectionType.TAGGED: 

168 self.data.append({ 

169 "type": "associations", 

170 "collection": collection, 

171 "collection_type": collectionType.name, 

172 "dataset_ids": [assoc.ref.id for assoc in associations], 

173 }) 

174 elif collectionType is CollectionType.CALIBRATION: 

175 idsByTimespan: Dict[Timespan, List[DatasetId]] = defaultdict(list) 

176 for association in associations: 

177 assert association.timespan is not None 

178 assert association.ref.id is not None 

179 idsByTimespan[association.timespan].append(association.ref.id) 

180 self.data.append({ 

181 "type": "associations", 

182 "collection": collection, 

183 "collection_type": collectionType.name, 

184 "validity_ranges": [ 

185 { 

186 "begin": timespan.begin, 

187 "end": timespan.end, 

188 "dataset_ids": dataset_ids, 

189 } 

190 for timespan, dataset_ids in idsByTimespan.items() 

191 ] 

192 }) 

193 

194 def finish(self) -> None: 

195 # Docstring inherited from RepoExportBackend. 

196 yaml.dump( 

197 { 

198 "description": "Butler Data Repository Export", 

199 "version": str(EXPORT_FORMAT_VERSION), 

200 "data": self.data, 

201 }, 

202 stream=self.stream, 

203 sort_keys=False, 

204 ) 

205 

206 

207class YamlRepoImportBackend(RepoImportBackend): 

208 """A repository import implementation that reads from a YAML file. 

209 

210 Parameters 

211 ---------- 

212 stream 

213 A readable file-like object. 

214 registry : `Registry` 

215 The registry datasets will be imported into. Only used to retreive 

216 dataset types during construction; all write happen in `register` 

217 and `load`. 

218 """ 

219 

220 def __init__(self, stream: IO, registry: Registry): 

221 # We read the file fully and convert its contents to Python objects 

222 # instead of loading incrementally so we can spot some problems early; 

223 # because `register` can't be put inside a transaction, we'd rather not 

224 # run that at all if there's going to be problem later in `load`. 

225 wrapper = yaml.safe_load(stream) 

226 if wrapper["version"] == 0: 

227 # Grandfather-in 'version: 0' -> 1.0.0, which is what we wrote 

228 # before we really tried to do versioning here. 

229 fileVersion = VersionTuple(1, 0, 0) 

230 else: 

231 fileVersion = VersionTuple.fromString(wrapper["version"]) 

232 if fileVersion.major != EXPORT_FORMAT_VERSION.major: 

233 raise IncompatibleVersionError( 

234 f"Cannot read repository export file with version={fileVersion} " 

235 f"({EXPORT_FORMAT_VERSION.major}.x.x required)." 

236 ) 

237 if fileVersion.minor > EXPORT_FORMAT_VERSION.minor: 

238 raise IncompatibleVersionError( 

239 f"Cannot read repository export file with version={fileVersion} " 

240 f"< {EXPORT_FORMAT_VERSION.major}.{EXPORT_FORMAT_VERSION.minor}.x required." 

241 ) 

242 self.runs: Dict[str, Tuple[Optional[str], Timespan]] = {} 

243 self.chains: Dict[str, List[str]] = {} 

244 self.collections: Dict[str, CollectionType] = {} 

245 self.collectionDocs: Dict[str, str] = {} 

246 self.datasetTypes: NamedValueSet[DatasetType] = NamedValueSet() 

247 self.dimensions: Mapping[DimensionElement, List[DimensionRecord]] = defaultdict(list) 

248 self.tagAssociations: Dict[str, List[DatasetId]] = defaultdict(list) 

249 self.calibAssociations: Dict[str, Dict[Timespan, List[DatasetId]]] = defaultdict(dict) 

250 self.refsByFileId: Dict[DatasetId, DatasetRef] = {} 

251 self.registry: Registry = registry 

252 datasetData = [] 

253 for data in wrapper["data"]: 

254 if data["type"] == "dimension": 

255 # convert all datetime values to astropy 

256 for record in data["records"]: 

257 for key in record: 

258 # Some older YAML files were produced with native 

259 # YAML support for datetime, we support reading that 

260 # data back. Newer conversion uses _AstropyTimeToYAML 

261 # class with special YAML tag. 

262 if isinstance(record[key], datetime): 

263 record[key] = astropy.time.Time(record[key], scale="utc") 

264 element = self.registry.dimensions[data["element"]] 

265 RecordClass: Type[DimensionRecord] = element.RecordClass 

266 self.dimensions[element].extend( 

267 RecordClass(**r) for r in data["records"] 

268 ) 

269 elif data["type"] == "collection": 

270 collectionType = CollectionType.from_name(data["collection_type"]) 

271 if collectionType is CollectionType.RUN: 

272 self.runs[data["name"]] = ( 

273 data["host"], 

274 Timespan(begin=data["timespan_begin"], end=data["timespan_end"]) 

275 ) 

276 elif collectionType is CollectionType.CHAINED: 

277 children = [] 

278 for child in data["children"]: 

279 if not isinstance(child, str): 

280 warnings.warn( 

281 f"CHAINED collection {data['name']} includes restrictions on child " 

282 "collection searches, which are no longer suppored and will be ignored." 

283 ) 

284 # Old form with dataset type restrictions only, 

285 # supported for backwards compatibility. 

286 child, _ = child 

287 children.append(child) 

288 self.chains[data["name"]] = children 

289 else: 

290 self.collections[data["name"]] = collectionType 

291 doc = data.get("doc") 

292 if doc is not None: 

293 self.collectionDocs[data["name"]] = doc 

294 elif data["type"] == "run": 

295 # Also support old form of saving a run with no extra info. 

296 self.runs[data["name"]] = (None, Timespan(None, None)) 

297 elif data["type"] == "dataset_type": 

298 self.datasetTypes.add( 

299 DatasetType(data["name"], dimensions=data["dimensions"], 

300 storageClass=data["storage_class"], universe=self.registry.dimensions, 

301 isCalibration=data.get("is_calibration", False)) 

302 ) 

303 elif data["type"] == "dataset": 

304 # Save raw dataset data for a second loop, so we can ensure we 

305 # know about all dataset types first. 

306 datasetData.append(data) 

307 elif data["type"] == "associations": 

308 collectionType = CollectionType.from_name(data["collection_type"]) 

309 if collectionType is CollectionType.TAGGED: 

310 self.tagAssociations[data["collection"]].extend(data["dataset_ids"]) 

311 elif collectionType is CollectionType.CALIBRATION: 

312 assocsByTimespan = self.calibAssociations[data["collection"]] 

313 for d in data["validity_ranges"]: 

314 assocsByTimespan[Timespan(begin=d["begin"], end=d["end"])] = d["dataset_ids"] 

315 else: 

316 raise ValueError(f"Unexpected calibration type for association: {collectionType.name}.") 

317 else: 

318 raise ValueError(f"Unexpected dictionary type: {data['type']}.") 

319 # key is (dataset type name, run) 

320 self.datasets: Mapping[Tuple[str, str], List[FileDataset]] = defaultdict(list) 

321 for data in datasetData: 

322 datasetType = self.datasetTypes.get(data["dataset_type"]) 

323 if datasetType is None: 

324 datasetType = self.registry.getDatasetType(data["dataset_type"]) 

325 self.datasets[data["dataset_type"], data["run"]].extend( 

326 FileDataset( 

327 d.get("path"), 

328 [DatasetRef(datasetType, dataId, run=data["run"], id=refid) 

329 for dataId, refid in zip(iterable(d["data_id"]), iterable(d["dataset_id"]))], 

330 formatter=doImport(d.get("formatter")) if "formatter" in d else None 

331 ) 

332 for d in data["records"] 

333 ) 

334 

335 def register(self) -> None: 

336 # Docstring inherited from RepoImportBackend.register. 

337 for datasetType in self.datasetTypes: 

338 self.registry.registerDatasetType(datasetType) 

339 for run in self.runs: 

340 self.registry.registerRun(run, doc=self.collectionDocs.get(run)) 

341 # No way to add extra run info to registry yet. 

342 for collection, collection_type in self.collections.items(): 

343 self.registry.registerCollection(collection, collection_type, 

344 doc=self.collectionDocs.get(collection)) 

345 for chain, children in self.chains.items(): 

346 self.registry.registerCollection(chain, CollectionType.CHAINED, 

347 doc=self.collectionDocs.get(chain)) 

348 self.registry.setCollectionChain(chain, children) 

349 

350 def load(self, datastore: Optional[Datastore], *, 

351 directory: Optional[str] = None, transfer: Optional[str] = None, 

352 skip_dimensions: Optional[Set] = None, 

353 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

354 reuseIds: bool = False) -> None: 

355 # Docstring inherited from RepoImportBackend.load. 

356 for element, dimensionRecords in self.dimensions.items(): 

357 if skip_dimensions and element in skip_dimensions: 

358 continue 

359 self.registry.insertDimensionData(element, *dimensionRecords) 

360 # FileDatasets to ingest into the datastore (in bulk): 

361 fileDatasets = [] 

362 for (datasetTypeName, run), records in self.datasets.items(): 

363 # Make a big flattened list of all data IDs and dataset_ids, while 

364 # remembering slices that associate them with the FileDataset 

365 # instances they came from. 

366 datasets: List[DatasetRef] = [] 

367 dataset_ids: List[DatasetId] = [] 

368 slices = [] 

369 for fileDataset in records: 

370 start = len(datasets) 

371 datasets.extend(fileDataset.refs) 

372 dataset_ids.extend(ref.id for ref in fileDataset.refs) # type: ignore 

373 stop = len(datasets) 

374 slices.append(slice(start, stop)) 

375 # Insert all of those DatasetRefs at once. 

376 # For now, we ignore the dataset_id we pulled from the file 

377 # and just insert without one to get a new autoincrement value. 

378 # Eventually (once we have origin in IDs) we'll preserve them. 

379 resolvedRefs = self.registry._importDatasets(datasets, idGenerationMode=idGenerationMode, 

380 reuseIds=reuseIds) 

381 # Populate our dictionary that maps int dataset_id values from the 

382 # export file to the new DatasetRefs 

383 for fileId, ref in zip(dataset_ids, resolvedRefs): 

384 self.refsByFileId[fileId] = ref 

385 # Now iterate over the original records, and install the new 

386 # resolved DatasetRefs to replace the unresolved ones as we 

387 # reorganize the collection information. 

388 for sliceForFileDataset, fileDataset in zip(slices, records): 

389 fileDataset.refs = resolvedRefs[sliceForFileDataset] 

390 if directory is not None: 

391 fileDataset.path = ButlerURI(directory, forceDirectory=True).join(fileDataset.path) 

392 fileDatasets.append(fileDataset) 

393 # Ingest everything into the datastore at once. 

394 if datastore is not None and fileDatasets: 

395 datastore.ingest(*fileDatasets, transfer=transfer) 

396 # Associate datasets with tagged collections. 

397 for collection, dataset_ids in self.tagAssociations.items(): 

398 self.registry.associate(collection, [self.refsByFileId[i] for i in dataset_ids]) 

399 # Associate datasets with calibration collections. 

400 for collection, idsByTimespan in self.calibAssociations.items(): 

401 for timespan, dataset_ids in idsByTimespan.items(): 

402 self.registry.certify(collection, [self.refsByFileId[i] for i in dataset_ids], timespan)