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 

22""" 

23Butler top level classes. 

24""" 

25from __future__ import annotations 

26 

27__all__ = ( 

28 "Butler", 

29 "ButlerValidationError", 

30 "PruneCollectionsArgsError", 

31 "PurgeWithoutUnstorePruneCollectionsError", 

32 "RunWithoutPurgePruneCollectionsError", 

33 "PurgeUnsupportedPruneCollectionsError", 

34) 

35 

36import collections.abc 

37from collections import defaultdict 

38import contextlib 

39import logging 

40import numbers 

41import os 

42from typing import ( 

43 Any, 

44 ClassVar, 

45 Counter, 

46 Dict, 

47 Iterable, 

48 Iterator, 

49 List, 

50 MutableMapping, 

51 Optional, 

52 Set, 

53 TextIO, 

54 Tuple, 

55 Type, 

56 Union, 

57) 

58 

59try: 

60 import boto3 

61except ImportError: 

62 boto3 = None 

63 

64from lsst.utils import doImport 

65from .core import ( 

66 AmbiguousDatasetError, 

67 ButlerURI, 

68 Config, 

69 ConfigSubset, 

70 DataCoordinate, 

71 DataId, 

72 DataIdValue, 

73 DatasetRef, 

74 DatasetType, 

75 Datastore, 

76 Dimension, 

77 DimensionConfig, 

78 FileDataset, 

79 Progress, 

80 StorageClassFactory, 

81 Timespan, 

82 ValidationError, 

83 VERBOSE, 

84) 

85from .core.repoRelocation import BUTLER_ROOT_TAG 

86from .core.utils import transactional, getClassOf 

87from ._deferredDatasetHandle import DeferredDatasetHandle 

88from ._butlerConfig import ButlerConfig 

89from .registry import ( 

90 Registry, 

91 RegistryConfig, 

92 RegistryDefaults, 

93 CollectionSearch, 

94 CollectionType, 

95 ConflictingDefinitionError, 

96 DatasetIdGenEnum, 

97) 

98from .transfers import RepoExportContext 

99 

100log = logging.getLogger(__name__) 

101 

102 

103class ButlerValidationError(ValidationError): 

104 """There is a problem with the Butler configuration.""" 

105 pass 

106 

107 

108class PruneCollectionsArgsError(TypeError): 

109 """Base class for errors relating to Butler.pruneCollections input 

110 arguments. 

111 """ 

112 pass 

113 

114 

115class PurgeWithoutUnstorePruneCollectionsError(PruneCollectionsArgsError): 

116 """Raised when purge and unstore are both required to be True, and 

117 purge is True but unstore is False. 

118 """ 

119 

120 def __init__(self) -> None: 

121 super().__init__("Cannot pass purge=True without unstore=True.") 

122 

123 

124class RunWithoutPurgePruneCollectionsError(PruneCollectionsArgsError): 

125 """Raised when pruning a RUN collection but purge is False.""" 

126 

127 def __init__(self, collectionType: CollectionType): 

128 self.collectionType = collectionType 

129 super().__init__(f"Cannot prune RUN collection {self.collectionType.name} without purge=True.") 

130 

131 

132class PurgeUnsupportedPruneCollectionsError(PruneCollectionsArgsError): 

133 """Raised when purge is True but is not supported for the given 

134 collection.""" 

135 

136 def __init__(self, collectionType: CollectionType): 

137 self.collectionType = collectionType 

138 super().__init__( 

139 f"Cannot prune {self.collectionType} collection {self.collectionType.name} with purge=True.") 

140 

141 

142class Butler: 

143 """Main entry point for the data access system. 

144 

145 Parameters 

146 ---------- 

147 config : `ButlerConfig`, `Config` or `str`, optional. 

148 Configuration. Anything acceptable to the 

149 `ButlerConfig` constructor. If a directory path 

150 is given the configuration will be read from a ``butler.yaml`` file in 

151 that location. If `None` is given default values will be used. 

152 butler : `Butler`, optional. 

153 If provided, construct a new Butler that uses the same registry and 

154 datastore as the given one, but with the given collection and run. 

155 Incompatible with the ``config``, ``searchPaths``, and ``writeable`` 

156 arguments. 

157 collections : `str` or `Iterable` [ `str` ], optional 

158 An expression specifying the collections to be searched (in order) when 

159 reading datasets. 

160 This may be a `str` collection name or an iterable thereof. 

161 See :ref:`daf_butler_collection_expressions` for more information. 

162 These collections are not registered automatically and must be 

163 manually registered before they are used by any method, but they may be 

164 manually registered after the `Butler` is initialized. 

165 run : `str`, optional 

166 Name of the `~CollectionType.RUN` collection new datasets should be 

167 inserted into. If ``collections`` is `None` and ``run`` is not `None`, 

168 ``collections`` will be set to ``[run]``. If not `None`, this 

169 collection will automatically be registered. If this is not set (and 

170 ``writeable`` is not set either), a read-only butler will be created. 

171 searchPaths : `list` of `str`, optional 

172 Directory paths to search when calculating the full Butler 

173 configuration. Not used if the supplied config is already a 

174 `ButlerConfig`. 

175 writeable : `bool`, optional 

176 Explicitly sets whether the butler supports write operations. If not 

177 provided, a read-write butler is created if any of ``run``, ``tags``, 

178 or ``chains`` is non-empty. 

179 inferDefaults : `bool`, optional 

180 If `True` (default) infer default data ID values from the values 

181 present in the datasets in ``collections``: if all collections have the 

182 same value (or no value) for a governor dimension, that value will be 

183 the default for that dimension. Nonexistent collections are ignored. 

184 If a default value is provided explicitly for a governor dimension via 

185 ``**kwargs``, no default will be inferred for that dimension. 

186 **kwargs : `str` 

187 Default data ID key-value pairs. These may only identify "governor" 

188 dimensions like ``instrument`` and ``skymap``. 

189 

190 Examples 

191 -------- 

192 While there are many ways to control exactly how a `Butler` interacts with 

193 the collections in its `Registry`, the most common cases are still simple. 

194 

195 For a read-only `Butler` that searches one collection, do:: 

196 

197 butler = Butler("/path/to/repo", collections=["u/alice/DM-50000"]) 

198 

199 For a read-write `Butler` that writes to and reads from a 

200 `~CollectionType.RUN` collection:: 

201 

202 butler = Butler("/path/to/repo", run="u/alice/DM-50000/a") 

203 

204 The `Butler` passed to a ``PipelineTask`` is often much more complex, 

205 because we want to write to one `~CollectionType.RUN` collection but read 

206 from several others (as well):: 

207 

208 butler = Butler("/path/to/repo", run="u/alice/DM-50000/a", 

209 collections=["u/alice/DM-50000/a", 

210 "u/bob/DM-49998", 

211 "HSC/defaults"]) 

212 

213 This butler will `put` new datasets to the run ``u/alice/DM-50000/a``. 

214 Datasets will be read first from that run (since it appears first in the 

215 chain), and then from ``u/bob/DM-49998`` and finally ``HSC/defaults``. 

216 

217 Finally, one can always create a `Butler` with no collections:: 

218 

219 butler = Butler("/path/to/repo", writeable=True) 

220 

221 This can be extremely useful when you just want to use ``butler.registry``, 

222 e.g. for inserting dimension data or managing collections, or when the 

223 collections you want to use with the butler are not consistent. 

224 Passing ``writeable`` explicitly here is only necessary if you want to be 

225 able to make changes to the repo - usually the value for ``writeable`` can 

226 be guessed from the collection arguments provided, but it defaults to 

227 `False` when there are not collection arguments. 

228 """ 

229 def __init__(self, config: Union[Config, str, None] = None, *, 

230 butler: Optional[Butler] = None, 

231 collections: Any = None, 

232 run: Optional[str] = None, 

233 searchPaths: Optional[List[str]] = None, 

234 writeable: Optional[bool] = None, 

235 inferDefaults: bool = True, 

236 **kwargs: str, 

237 ): 

238 defaults = RegistryDefaults(collections=collections, run=run, infer=inferDefaults, **kwargs) 

239 # Load registry, datastore, etc. from config or existing butler. 

240 if butler is not None: 

241 if config is not None or searchPaths is not None or writeable is not None: 

242 raise TypeError("Cannot pass 'config', 'searchPaths', or 'writeable' " 

243 "arguments with 'butler' argument.") 

244 self.registry = butler.registry.copy(defaults) 

245 self.datastore = butler.datastore 

246 self.storageClasses = butler.storageClasses 

247 self._config: ButlerConfig = butler._config 

248 self._allow_put_of_predefined_dataset = butler._allow_put_of_predefined_dataset 

249 else: 

250 self._config = ButlerConfig(config, searchPaths=searchPaths) 

251 if "root" in self._config: 

252 butlerRoot = self._config["root"] 

253 else: 

254 butlerRoot = self._config.configDir 

255 if writeable is None: 

256 writeable = run is not None 

257 self.registry = Registry.fromConfig(self._config, butlerRoot=butlerRoot, writeable=writeable, 

258 defaults=defaults) 

259 self.datastore = Datastore.fromConfig(self._config, self.registry.getDatastoreBridgeManager(), 

260 butlerRoot=butlerRoot) 

261 self.storageClasses = StorageClassFactory() 

262 self.storageClasses.addFromConfig(self._config) 

263 self._allow_put_of_predefined_dataset = self._config.get("allow_put_of_predefined_dataset", False) 

264 if "run" in self._config or "collection" in self._config: 

265 raise ValueError("Passing a run or collection via configuration is no longer supported.") 

266 

267 GENERATION: ClassVar[int] = 3 

268 """This is a Generation 3 Butler. 

269 

270 This attribute may be removed in the future, once the Generation 2 Butler 

271 interface has been fully retired; it should only be used in transitional 

272 code. 

273 """ 

274 

275 @staticmethod 

276 def makeRepo(root: str, config: Union[Config, str, None] = None, 

277 dimensionConfig: Union[Config, str, None] = None, standalone: bool = False, 

278 searchPaths: Optional[List[str]] = None, forceConfigRoot: bool = True, 

279 outfile: Optional[str] = None, overwrite: bool = False) -> Config: 

280 """Create an empty data repository by adding a butler.yaml config 

281 to a repository root directory. 

282 

283 Parameters 

284 ---------- 

285 root : `str` or `ButlerURI` 

286 Path or URI to the root location of the new repository. Will be 

287 created if it does not exist. 

288 config : `Config` or `str`, optional 

289 Configuration to write to the repository, after setting any 

290 root-dependent Registry or Datastore config options. Can not 

291 be a `ButlerConfig` or a `ConfigSubset`. If `None`, default 

292 configuration will be used. Root-dependent config options 

293 specified in this config are overwritten if ``forceConfigRoot`` 

294 is `True`. 

295 dimensionConfig : `Config` or `str`, optional 

296 Configuration for dimensions, will be used to initialize registry 

297 database. 

298 standalone : `bool` 

299 If True, write all expanded defaults, not just customized or 

300 repository-specific settings. 

301 This (mostly) decouples the repository from the default 

302 configuration, insulating it from changes to the defaults (which 

303 may be good or bad, depending on the nature of the changes). 

304 Future *additions* to the defaults will still be picked up when 

305 initializing `Butlers` to repos created with ``standalone=True``. 

306 searchPaths : `list` of `str`, optional 

307 Directory paths to search when calculating the full butler 

308 configuration. 

309 forceConfigRoot : `bool`, optional 

310 If `False`, any values present in the supplied ``config`` that 

311 would normally be reset are not overridden and will appear 

312 directly in the output config. This allows non-standard overrides 

313 of the root directory for a datastore or registry to be given. 

314 If this parameter is `True` the values for ``root`` will be 

315 forced into the resulting config if appropriate. 

316 outfile : `str`, optional 

317 If not-`None`, the output configuration will be written to this 

318 location rather than into the repository itself. Can be a URI 

319 string. Can refer to a directory that will be used to write 

320 ``butler.yaml``. 

321 overwrite : `bool`, optional 

322 Create a new configuration file even if one already exists 

323 in the specified output location. Default is to raise 

324 an exception. 

325 

326 Returns 

327 ------- 

328 config : `Config` 

329 The updated `Config` instance written to the repo. 

330 

331 Raises 

332 ------ 

333 ValueError 

334 Raised if a ButlerConfig or ConfigSubset is passed instead of a 

335 regular Config (as these subclasses would make it impossible to 

336 support ``standalone=False``). 

337 FileExistsError 

338 Raised if the output config file already exists. 

339 os.error 

340 Raised if the directory does not exist, exists but is not a 

341 directory, or cannot be created. 

342 

343 Notes 

344 ----- 

345 Note that when ``standalone=False`` (the default), the configuration 

346 search path (see `ConfigSubset.defaultSearchPaths`) that was used to 

347 construct the repository should also be used to construct any Butlers 

348 to avoid configuration inconsistencies. 

349 """ 

350 if isinstance(config, (ButlerConfig, ConfigSubset)): 

351 raise ValueError("makeRepo must be passed a regular Config without defaults applied.") 

352 

353 # Ensure that the root of the repository exists or can be made 

354 uri = ButlerURI(root, forceDirectory=True) 

355 uri.mkdir() 

356 

357 config = Config(config) 

358 

359 # If we are creating a new repo from scratch with relative roots, 

360 # do not propagate an explicit root from the config file 

361 if "root" in config: 

362 del config["root"] 

363 

364 full = ButlerConfig(config, searchPaths=searchPaths) # this applies defaults 

365 datastoreClass: Type[Datastore] = doImport(full["datastore", "cls"]) 

366 datastoreClass.setConfigRoot(BUTLER_ROOT_TAG, config, full, overwrite=forceConfigRoot) 

367 

368 # if key exists in given config, parse it, otherwise parse the defaults 

369 # in the expanded config 

370 if config.get(("registry", "db")): 

371 registryConfig = RegistryConfig(config) 

372 else: 

373 registryConfig = RegistryConfig(full) 

374 defaultDatabaseUri = registryConfig.makeDefaultDatabaseUri(BUTLER_ROOT_TAG) 

375 if defaultDatabaseUri is not None: 

376 Config.updateParameters(RegistryConfig, config, full, 

377 toUpdate={"db": defaultDatabaseUri}, 

378 overwrite=forceConfigRoot) 

379 else: 

380 Config.updateParameters(RegistryConfig, config, full, toCopy=("db",), 

381 overwrite=forceConfigRoot) 

382 

383 if standalone: 

384 config.merge(full) 

385 else: 

386 # Always expand the registry.managers section into the per-repo 

387 # config, because after the database schema is created, it's not 

388 # allowed to change anymore. Note that in the standalone=True 

389 # branch, _everything_ in the config is expanded, so there's no 

390 # need to special case this. 

391 Config.updateParameters(RegistryConfig, config, full, toMerge=("managers",), overwrite=False) 

392 configURI: Union[str, ButlerURI] 

393 if outfile is not None: 

394 # When writing to a separate location we must include 

395 # the root of the butler repo in the config else it won't know 

396 # where to look. 

397 config["root"] = uri.geturl() 

398 configURI = outfile 

399 else: 

400 configURI = uri 

401 config.dumpToUri(configURI, overwrite=overwrite) 

402 

403 # Create Registry and populate tables 

404 registryConfig = RegistryConfig(config.get("registry")) 

405 dimensionConfig = DimensionConfig(dimensionConfig) 

406 Registry.createFromConfig(registryConfig, dimensionConfig=dimensionConfig, butlerRoot=root) 

407 

408 log.log(VERBOSE, "Wrote new Butler configuration file to %s", configURI) 

409 

410 return config 

411 

412 @classmethod 

413 def _unpickle(cls, config: ButlerConfig, collections: Optional[CollectionSearch], run: Optional[str], 

414 defaultDataId: Dict[str, str], writeable: bool) -> Butler: 

415 """Callable used to unpickle a Butler. 

416 

417 We prefer not to use ``Butler.__init__`` directly so we can force some 

418 of its many arguments to be keyword-only (note that ``__reduce__`` 

419 can only invoke callables with positional arguments). 

420 

421 Parameters 

422 ---------- 

423 config : `ButlerConfig` 

424 Butler configuration, already coerced into a true `ButlerConfig` 

425 instance (and hence after any search paths for overrides have been 

426 utilized). 

427 collections : `CollectionSearch` 

428 Names of the default collections to read from. 

429 run : `str`, optional 

430 Name of the default `~CollectionType.RUN` collection to write to. 

431 defaultDataId : `dict` [ `str`, `str` ] 

432 Default data ID values. 

433 writeable : `bool` 

434 Whether the Butler should support write operations. 

435 

436 Returns 

437 ------- 

438 butler : `Butler` 

439 A new `Butler` instance. 

440 """ 

441 # MyPy doesn't recognize that the kwargs below are totally valid; it 

442 # seems to think '**defaultDataId* is a _positional_ argument! 

443 return cls(config=config, collections=collections, run=run, writeable=writeable, 

444 **defaultDataId) # type: ignore 

445 

446 def __reduce__(self) -> tuple: 

447 """Support pickling. 

448 """ 

449 return (Butler._unpickle, (self._config, self.collections, self.run, 

450 self.registry.defaults.dataId.byName(), 

451 self.registry.isWriteable())) 

452 

453 def __str__(self) -> str: 

454 return "Butler(collections={}, run={}, datastore='{}', registry='{}')".format( 

455 self.collections, self.run, self.datastore, self.registry) 

456 

457 def isWriteable(self) -> bool: 

458 """Return `True` if this `Butler` supports write operations. 

459 """ 

460 return self.registry.isWriteable() 

461 

462 @contextlib.contextmanager 

463 def transaction(self) -> Iterator[None]: 

464 """Context manager supporting `Butler` transactions. 

465 

466 Transactions can be nested. 

467 """ 

468 with self.registry.transaction(): 

469 with self.datastore.transaction(): 

470 yield 

471 

472 def _standardizeArgs(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

473 dataId: Optional[DataId] = None, **kwargs: Any 

474 ) -> Tuple[DatasetType, Optional[DataId]]: 

475 """Standardize the arguments passed to several Butler APIs. 

476 

477 Parameters 

478 ---------- 

479 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

480 When `DatasetRef` the `dataId` should be `None`. 

481 Otherwise the `DatasetType` or name thereof. 

482 dataId : `dict` or `DataCoordinate` 

483 A `dict` of `Dimension` link name, value pairs that label the 

484 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

485 should be provided as the second argument. 

486 **kwargs 

487 Additional keyword arguments used to augment or construct a 

488 `DataCoordinate`. See `DataCoordinate.standardize` 

489 parameters. 

490 

491 Returns 

492 ------- 

493 datasetType : `DatasetType` 

494 A `DatasetType` instance extracted from ``datasetRefOrType``. 

495 dataId : `dict` or `DataId`, optional 

496 Argument that can be used (along with ``kwargs``) to construct a 

497 `DataId`. 

498 

499 Notes 

500 ----- 

501 Butler APIs that conceptually need a DatasetRef also allow passing a 

502 `DatasetType` (or the name of one) and a `DataId` (or a dict and 

503 keyword arguments that can be used to construct one) separately. This 

504 method accepts those arguments and always returns a true `DatasetType` 

505 and a `DataId` or `dict`. 

506 

507 Standardization of `dict` vs `DataId` is best handled by passing the 

508 returned ``dataId`` (and ``kwargs``) to `Registry` APIs, which are 

509 generally similarly flexible. 

510 """ 

511 externalDatasetType: Optional[DatasetType] = None 

512 internalDatasetType: Optional[DatasetType] = None 

513 if isinstance(datasetRefOrType, DatasetRef): 

514 if dataId is not None or kwargs: 

515 raise ValueError("DatasetRef given, cannot use dataId as well") 

516 externalDatasetType = datasetRefOrType.datasetType 

517 dataId = datasetRefOrType.dataId 

518 else: 

519 # Don't check whether DataId is provided, because Registry APIs 

520 # can usually construct a better error message when it wasn't. 

521 if isinstance(datasetRefOrType, DatasetType): 

522 externalDatasetType = datasetRefOrType 

523 else: 

524 internalDatasetType = self.registry.getDatasetType(datasetRefOrType) 

525 

526 # Check that they are self-consistent 

527 if externalDatasetType is not None: 

528 internalDatasetType = self.registry.getDatasetType(externalDatasetType.name) 

529 if externalDatasetType != internalDatasetType: 

530 raise ValueError(f"Supplied dataset type ({externalDatasetType}) inconsistent with " 

531 f"registry definition ({internalDatasetType})") 

532 

533 assert internalDatasetType is not None 

534 return internalDatasetType, dataId 

535 

536 def _rewrite_data_id(self, dataId: Optional[DataId], datasetType: DatasetType, 

537 **kwargs: Any) -> Tuple[Optional[DataId], Dict[str, Any]]: 

538 """Rewrite a data ID taking into account dimension records. 

539 

540 Take a Data ID and keyword args and rewrite it if necessary to 

541 allow the user to specify dimension records rather than dimension 

542 primary values. 

543 

544 This allows a user to include a dataId dict with keys of 

545 ``exposure.day_obs`` and ``exposure.seq_num`` instead of giving 

546 the integer exposure ID. It also allows a string to be given 

547 for a dimension value rather than the integer ID if that is more 

548 convenient. For example, rather than having to specifyin the 

549 detector with ``detector.full_name``, a string given for ``detector`` 

550 will be interpreted as the full name and converted to the integer 

551 value. 

552 

553 Keyword arguments can also use strings for dimensions like detector 

554 and exposure but python does not allow them to include ``.`` and 

555 so the ``exposure.day_obs`` syntax can not be used in a keyword 

556 argument. 

557 

558 Parameters 

559 ---------- 

560 dataId : `dict` or `DataCoordinate` 

561 A `dict` of `Dimension` link name, value pairs that will label the 

562 `DatasetRef` within a Collection. 

563 datasetType : `DatasetType` 

564 The dataset type associated with this dataId. Required to 

565 determine the relevant dimensions. 

566 **kwargs 

567 Additional keyword arguments used to augment or construct a 

568 `DataId`. See `DataId` parameters. 

569 

570 Returns 

571 ------- 

572 dataId : `dict` or `DataCoordinate` 

573 The, possibly rewritten, dataId. If given a `DataCoordinate` and 

574 no keyword arguments, the orginal dataId will be returned 

575 unchanged. 

576 **kwargs : `dict` 

577 Any unused keyword arguments. 

578 """ 

579 # Do nothing if we have a standalone DataCoordinate. 

580 if isinstance(dataId, DataCoordinate) and not kwargs: 

581 return dataId, kwargs 

582 

583 # Process dimension records that are using record information 

584 # rather than ids 

585 newDataId: Dict[str, DataIdValue] = {} 

586 byRecord: Dict[str, Dict[str, Any]] = defaultdict(dict) 

587 

588 # if all the dataId comes from keyword parameters we do not need 

589 # to do anything here because they can't be of the form 

590 # exposure.obs_id because a "." is not allowed in a keyword parameter. 

591 if dataId: 

592 for k, v in dataId.items(): 

593 # If we have a Dimension we do not need to do anything 

594 # because it cannot be a compound key. 

595 if isinstance(k, str) and "." in k: 

596 # Someone is using a more human-readable dataId 

597 dimensionName, record = k.split(".", 1) 

598 byRecord[dimensionName][record] = v 

599 elif isinstance(k, Dimension): 

600 newDataId[k.name] = v 

601 else: 

602 newDataId[k] = v 

603 

604 # Go through the updated dataId and check the type in case someone is 

605 # using an alternate key. We have already filtered out the compound 

606 # keys dimensions.record format. 

607 not_dimensions = {} 

608 

609 # Will need to look in the dataId and the keyword arguments 

610 # and will remove them if they need to be fixed or are unrecognized. 

611 for dataIdDict in (newDataId, kwargs): 

612 # Use a list so we can adjust the dict safely in the loop 

613 for dimensionName in list(dataIdDict): 

614 value = dataIdDict[dimensionName] 

615 try: 

616 dimension = self.registry.dimensions.getStaticDimensions()[dimensionName] 

617 except KeyError: 

618 # This is not a real dimension 

619 not_dimensions[dimensionName] = value 

620 del dataIdDict[dimensionName] 

621 continue 

622 

623 # Convert an integral type to an explicit int to simplify 

624 # comparisons here 

625 if isinstance(value, numbers.Integral): 

626 value = int(value) 

627 

628 if not isinstance(value, dimension.primaryKey.getPythonType()): 

629 for alternate in dimension.alternateKeys: 

630 if isinstance(value, alternate.getPythonType()): 

631 byRecord[dimensionName][alternate.name] = value 

632 del dataIdDict[dimensionName] 

633 log.debug("Converting dimension %s to %s.%s=%s", 

634 dimensionName, dimensionName, alternate.name, value) 

635 break 

636 else: 

637 log.warning("Type mismatch found for value '%r' provided for dimension %s. " 

638 "Could not find matching alternative (primary key has type %s) " 

639 "so attempting to use as-is.", 

640 value, dimensionName, dimension.primaryKey.getPythonType()) 

641 

642 # If we have some unrecognized dimensions we have to try to connect 

643 # them to records in other dimensions. This is made more complicated 

644 # by some dimensions having records with clashing names. A mitigation 

645 # is that we can tell by this point which dimensions are missing 

646 # for the DatasetType but this does not work for calibrations 

647 # where additional dimensions can be used to constrain the temporal 

648 # axis. 

649 if not_dimensions: 

650 # Calculate missing dimensions 

651 provided = set(newDataId) | set(kwargs) | set(byRecord) 

652 missingDimensions = datasetType.dimensions.names - provided 

653 

654 # For calibrations we may well be needing temporal dimensions 

655 # so rather than always including all dimensions in the scan 

656 # restrict things a little. It is still possible for there 

657 # to be confusion over day_obs in visit vs exposure for example. 

658 # If we are not searching calibration collections things may 

659 # fail but they are going to fail anyway because of the 

660 # ambiguousness of the dataId... 

661 candidateDimensions: Set[str] = set() 

662 candidateDimensions.update(missingDimensions) 

663 if datasetType.isCalibration(): 

664 for dim in self.registry.dimensions.getStaticDimensions(): 

665 if dim.temporal: 

666 candidateDimensions.add(str(dim)) 

667 

668 # Look up table for the first association with a dimension 

669 guessedAssociation: Dict[str, Dict[str, Any]] = defaultdict(dict) 

670 

671 # Keep track of whether an item is associated with multiple 

672 # dimensions. 

673 counter: Counter[str] = Counter() 

674 assigned: Dict[str, Set[str]] = defaultdict(set) 

675 

676 # Go through the missing dimensions and associate the 

677 # given names with records within those dimensions 

678 for dimensionName in candidateDimensions: 

679 dimension = self.registry.dimensions.getStaticDimensions()[dimensionName] 

680 fields = dimension.metadata.names | dimension.uniqueKeys.names 

681 for field in not_dimensions: 

682 if field in fields: 

683 guessedAssociation[dimensionName][field] = not_dimensions[field] 

684 counter[dimensionName] += 1 

685 assigned[field].add(dimensionName) 

686 

687 # There is a chance we have allocated a single dataId item 

688 # to multiple dimensions. Need to decide which should be retained. 

689 # For now assume that the most popular alternative wins. 

690 # This means that day_obs with seq_num will result in 

691 # exposure.day_obs and not visit.day_obs 

692 # Also prefer an explicitly missing dimension over an inferred 

693 # temporal dimension. 

694 for fieldName, assignedDimensions in assigned.items(): 

695 if len(assignedDimensions) > 1: 

696 # Pick the most popular (preferring mandatory dimensions) 

697 requiredButMissing = assignedDimensions.intersection(missingDimensions) 

698 if requiredButMissing: 

699 candidateDimensions = requiredButMissing 

700 else: 

701 candidateDimensions = assignedDimensions 

702 

703 # Select the relevant items and get a new restricted 

704 # counter. 

705 theseCounts = {k: v for k, v in counter.items() if k in candidateDimensions} 

706 duplicatesCounter: Counter[str] = Counter() 

707 duplicatesCounter.update(theseCounts) 

708 

709 # Choose the most common. If they are equally common 

710 # we will pick the one that was found first. 

711 # Returns a list of tuples 

712 selected = duplicatesCounter.most_common(1)[0][0] 

713 

714 log.debug("Ambiguous dataId entry '%s' associated with multiple dimensions: %s." 

715 " Removed ambiguity by choosing dimension %s.", 

716 fieldName, ", ".join(assignedDimensions), selected) 

717 

718 for candidateDimension in assignedDimensions: 

719 if candidateDimension != selected: 

720 del guessedAssociation[candidateDimension][fieldName] 

721 

722 # Update the record look up dict with the new associations 

723 for dimensionName, values in guessedAssociation.items(): 

724 if values: # A dict might now be empty 

725 log.debug("Assigned non-dimension dataId keys to dimension %s: %s", 

726 dimensionName, values) 

727 byRecord[dimensionName].update(values) 

728 

729 if byRecord: 

730 # Some record specifiers were found so we need to convert 

731 # them to the Id form 

732 for dimensionName, values in byRecord.items(): 

733 if dimensionName in newDataId: 

734 log.warning("DataId specified explicit %s dimension value of %s in addition to" 

735 " general record specifiers for it of %s. Ignoring record information.", 

736 dimensionName, newDataId[dimensionName], str(values)) 

737 continue 

738 

739 # Build up a WHERE expression 

740 bind = {k: v for k, v in values.items()} 

741 where = " AND ".join(f"{dimensionName}.{k} = {k}" 

742 for k in bind) 

743 

744 # Hopefully we get a single record that matches 

745 records = set(self.registry.queryDimensionRecords(dimensionName, dataId=newDataId, 

746 where=where, bind=bind, **kwargs)) 

747 

748 if len(records) != 1: 

749 if len(records) > 1: 

750 log.debug("Received %d records from constraints of %s", len(records), str(values)) 

751 for r in records: 

752 log.debug("- %s", str(r)) 

753 raise RuntimeError(f"DataId specification for dimension {dimensionName} is not" 

754 f" uniquely constrained to a single dataset by {values}." 

755 f" Got {len(records)} results.") 

756 raise RuntimeError(f"DataId specification for dimension {dimensionName} matched no" 

757 f" records when constrained by {values}") 

758 

759 # Get the primary key from the real dimension object 

760 dimension = self.registry.dimensions.getStaticDimensions()[dimensionName] 

761 if not isinstance(dimension, Dimension): 

762 raise RuntimeError( 

763 f"{dimension.name} is not a true dimension, and cannot be used in data IDs." 

764 ) 

765 newDataId[dimensionName] = getattr(records.pop(), dimension.primaryKey.name) 

766 

767 # We have modified the dataId so need to switch to it 

768 dataId = newDataId 

769 

770 return dataId, kwargs 

771 

772 def _findDatasetRef(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

773 dataId: Optional[DataId] = None, *, 

774 collections: Any = None, 

775 allowUnresolved: bool = False, 

776 **kwargs: Any) -> DatasetRef: 

777 """Shared logic for methods that start with a search for a dataset in 

778 the registry. 

779 

780 Parameters 

781 ---------- 

782 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

783 When `DatasetRef` the `dataId` should be `None`. 

784 Otherwise the `DatasetType` or name thereof. 

785 dataId : `dict` or `DataCoordinate`, optional 

786 A `dict` of `Dimension` link name, value pairs that label the 

787 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

788 should be provided as the first argument. 

789 collections : Any, optional 

790 Collections to be searched, overriding ``self.collections``. 

791 Can be any of the types supported by the ``collections`` argument 

792 to butler construction. 

793 allowUnresolved : `bool`, optional 

794 If `True`, return an unresolved `DatasetRef` if finding a resolved 

795 one in the `Registry` fails. Defaults to `False`. 

796 **kwargs 

797 Additional keyword arguments used to augment or construct a 

798 `DataId`. See `DataId` parameters. 

799 

800 Returns 

801 ------- 

802 ref : `DatasetRef` 

803 A reference to the dataset identified by the given arguments. 

804 

805 Raises 

806 ------ 

807 LookupError 

808 Raised if no matching dataset exists in the `Registry` (and 

809 ``allowUnresolved is False``). 

810 ValueError 

811 Raised if a resolved `DatasetRef` was passed as an input, but it 

812 differs from the one found in the registry. 

813 TypeError 

814 Raised if no collections were provided. 

815 """ 

816 datasetType, dataId = self._standardizeArgs(datasetRefOrType, dataId, **kwargs) 

817 if isinstance(datasetRefOrType, DatasetRef): 

818 idNumber = datasetRefOrType.id 

819 else: 

820 idNumber = None 

821 timespan: Optional[Timespan] = None 

822 

823 dataId, kwargs = self._rewrite_data_id(dataId, datasetType, **kwargs) 

824 

825 if datasetType.isCalibration(): 

826 # Because this is a calibration dataset, first try to make a 

827 # standardize the data ID without restricting the dimensions to 

828 # those of the dataset type requested, because there may be extra 

829 # dimensions that provide temporal information for a validity-range 

830 # lookup. 

831 dataId = DataCoordinate.standardize(dataId, universe=self.registry.dimensions, 

832 defaults=self.registry.defaults.dataId, **kwargs) 

833 if dataId.graph.temporal: 

834 dataId = self.registry.expandDataId(dataId) 

835 timespan = dataId.timespan 

836 else: 

837 # Standardize the data ID to just the dimensions of the dataset 

838 # type instead of letting registry.findDataset do it, so we get the 

839 # result even if no dataset is found. 

840 dataId = DataCoordinate.standardize(dataId, graph=datasetType.dimensions, 

841 defaults=self.registry.defaults.dataId, **kwargs) 

842 # Always lookup the DatasetRef, even if one is given, to ensure it is 

843 # present in the current collection. 

844 ref = self.registry.findDataset(datasetType, dataId, collections=collections, timespan=timespan) 

845 if ref is None: 

846 if allowUnresolved: 

847 return DatasetRef(datasetType, dataId) 

848 else: 

849 if collections is None: 

850 collections = self.registry.defaults.collections 

851 raise LookupError(f"Dataset {datasetType.name} with data ID {dataId} " 

852 f"could not be found in collections {collections}.") 

853 if idNumber is not None and idNumber != ref.id: 

854 if collections is None: 

855 collections = self.registry.defaults.collections 

856 raise ValueError(f"DatasetRef.id provided ({idNumber}) does not match " 

857 f"id ({ref.id}) in registry in collections {collections}.") 

858 return ref 

859 

860 @transactional 

861 def put(self, obj: Any, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

862 dataId: Optional[DataId] = None, *, 

863 run: Optional[str] = None, 

864 **kwargs: Any) -> DatasetRef: 

865 """Store and register a dataset. 

866 

867 Parameters 

868 ---------- 

869 obj : `object` 

870 The dataset. 

871 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

872 When `DatasetRef` is provided, ``dataId`` should be `None`. 

873 Otherwise the `DatasetType` or name thereof. 

874 dataId : `dict` or `DataCoordinate` 

875 A `dict` of `Dimension` link name, value pairs that label the 

876 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

877 should be provided as the second argument. 

878 run : `str`, optional 

879 The name of the run the dataset should be added to, overriding 

880 ``self.run``. 

881 **kwargs 

882 Additional keyword arguments used to augment or construct a 

883 `DataCoordinate`. See `DataCoordinate.standardize` 

884 parameters. 

885 

886 Returns 

887 ------- 

888 ref : `DatasetRef` 

889 A reference to the stored dataset, updated with the correct id if 

890 given. 

891 

892 Raises 

893 ------ 

894 TypeError 

895 Raised if the butler is read-only or if no run has been provided. 

896 """ 

897 log.debug("Butler put: %s, dataId=%s, run=%s", datasetRefOrType, dataId, run) 

898 if not self.isWriteable(): 

899 raise TypeError("Butler is read-only.") 

900 datasetType, dataId = self._standardizeArgs(datasetRefOrType, dataId, **kwargs) 

901 if isinstance(datasetRefOrType, DatasetRef) and datasetRefOrType.id is not None: 

902 raise ValueError("DatasetRef must not be in registry, must have None id") 

903 

904 # Add Registry Dataset entry. 

905 dataId = self.registry.expandDataId(dataId, graph=datasetType.dimensions, **kwargs) 

906 

907 # For an execution butler the datasets will be pre-defined. 

908 # If the butler is configured that way datasets should only be inserted 

909 # if they do not already exist in registry. Trying and catching 

910 # ConflictingDefinitionError will not work because the transaction 

911 # will be corrupted. Instead, in this mode always check first. 

912 ref = None 

913 ref_is_predefined = False 

914 if self._allow_put_of_predefined_dataset: 

915 # Get the matching ref for this run. 

916 ref = self.registry.findDataset(datasetType, collections=run, 

917 dataId=dataId) 

918 

919 if ref: 

920 # Must be expanded form for datastore templating 

921 dataId = self.registry.expandDataId(dataId, graph=datasetType.dimensions) 

922 ref = ref.expanded(dataId) 

923 ref_is_predefined = True 

924 

925 if not ref: 

926 ref, = self.registry.insertDatasets(datasetType, run=run, dataIds=[dataId]) 

927 

928 # If the ref is predefined it is possible that the datastore also 

929 # has the record. Asking datastore to put it again will result in 

930 # the artifact being recreated, overwriting previous, then will cause 

931 # a failure in writing the record which will cause the artifact 

932 # to be removed. Much safer to ask first before attempting to 

933 # overwrite. Race conditions should not be an issue for the 

934 # execution butler environment. 

935 if ref_is_predefined: 

936 if self.datastore.knows(ref): 

937 raise ConflictingDefinitionError(f"Dataset associated {ref} already exists.") 

938 

939 self.datastore.put(obj, ref) 

940 

941 return ref 

942 

943 def getDirect(self, ref: DatasetRef, *, parameters: Optional[Dict[str, Any]] = None) -> Any: 

944 """Retrieve a stored dataset. 

945 

946 Unlike `Butler.get`, this method allows datasets outside the Butler's 

947 collection to be read as long as the `DatasetRef` that identifies them 

948 can be obtained separately. 

949 

950 Parameters 

951 ---------- 

952 ref : `DatasetRef` 

953 Resolved reference to an already stored dataset. 

954 parameters : `dict` 

955 Additional StorageClass-defined options to control reading, 

956 typically used to efficiently read only a subset of the dataset. 

957 

958 Returns 

959 ------- 

960 obj : `object` 

961 The dataset. 

962 """ 

963 return self.datastore.get(ref, parameters=parameters) 

964 

965 def getDirectDeferred(self, ref: DatasetRef, *, 

966 parameters: Union[dict, None] = None) -> DeferredDatasetHandle: 

967 """Create a `DeferredDatasetHandle` which can later retrieve a dataset, 

968 from a resolved `DatasetRef`. 

969 

970 Parameters 

971 ---------- 

972 ref : `DatasetRef` 

973 Resolved reference to an already stored dataset. 

974 parameters : `dict` 

975 Additional StorageClass-defined options to control reading, 

976 typically used to efficiently read only a subset of the dataset. 

977 

978 Returns 

979 ------- 

980 obj : `DeferredDatasetHandle` 

981 A handle which can be used to retrieve a dataset at a later time. 

982 

983 Raises 

984 ------ 

985 AmbiguousDatasetError 

986 Raised if ``ref.id is None``, i.e. the reference is unresolved. 

987 """ 

988 if ref.id is None: 

989 raise AmbiguousDatasetError( 

990 f"Dataset of type {ref.datasetType.name} with data ID {ref.dataId} is not resolved." 

991 ) 

992 return DeferredDatasetHandle(butler=self, ref=ref, parameters=parameters) 

993 

994 def getDeferred(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

995 dataId: Optional[DataId] = None, *, 

996 parameters: Union[dict, None] = None, 

997 collections: Any = None, 

998 **kwargs: Any) -> DeferredDatasetHandle: 

999 """Create a `DeferredDatasetHandle` which can later retrieve a dataset, 

1000 after an immediate registry lookup. 

1001 

1002 Parameters 

1003 ---------- 

1004 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1005 When `DatasetRef` the `dataId` should be `None`. 

1006 Otherwise the `DatasetType` or name thereof. 

1007 dataId : `dict` or `DataCoordinate`, optional 

1008 A `dict` of `Dimension` link name, value pairs that label the 

1009 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1010 should be provided as the first argument. 

1011 parameters : `dict` 

1012 Additional StorageClass-defined options to control reading, 

1013 typically used to efficiently read only a subset of the dataset. 

1014 collections : Any, optional 

1015 Collections to be searched, overriding ``self.collections``. 

1016 Can be any of the types supported by the ``collections`` argument 

1017 to butler construction. 

1018 **kwargs 

1019 Additional keyword arguments used to augment or construct a 

1020 `DataId`. See `DataId` parameters. 

1021 

1022 Returns 

1023 ------- 

1024 obj : `DeferredDatasetHandle` 

1025 A handle which can be used to retrieve a dataset at a later time. 

1026 

1027 Raises 

1028 ------ 

1029 LookupError 

1030 Raised if no matching dataset exists in the `Registry` (and 

1031 ``allowUnresolved is False``). 

1032 ValueError 

1033 Raised if a resolved `DatasetRef` was passed as an input, but it 

1034 differs from the one found in the registry. 

1035 TypeError 

1036 Raised if no collections were provided. 

1037 """ 

1038 ref = self._findDatasetRef(datasetRefOrType, dataId, collections=collections, **kwargs) 

1039 return DeferredDatasetHandle(butler=self, ref=ref, parameters=parameters) 

1040 

1041 def get(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

1042 dataId: Optional[DataId] = None, *, 

1043 parameters: Optional[Dict[str, Any]] = None, 

1044 collections: Any = None, 

1045 **kwargs: Any) -> Any: 

1046 """Retrieve a stored dataset. 

1047 

1048 Parameters 

1049 ---------- 

1050 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1051 When `DatasetRef` the `dataId` should be `None`. 

1052 Otherwise the `DatasetType` or name thereof. 

1053 dataId : `dict` or `DataCoordinate` 

1054 A `dict` of `Dimension` link name, value pairs that label the 

1055 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1056 should be provided as the first argument. 

1057 parameters : `dict` 

1058 Additional StorageClass-defined options to control reading, 

1059 typically used to efficiently read only a subset of the dataset. 

1060 collections : Any, optional 

1061 Collections to be searched, overriding ``self.collections``. 

1062 Can be any of the types supported by the ``collections`` argument 

1063 to butler construction. 

1064 **kwargs 

1065 Additional keyword arguments used to augment or construct a 

1066 `DataCoordinate`. See `DataCoordinate.standardize` 

1067 parameters. 

1068 

1069 Returns 

1070 ------- 

1071 obj : `object` 

1072 The dataset. 

1073 

1074 Raises 

1075 ------ 

1076 ValueError 

1077 Raised if a resolved `DatasetRef` was passed as an input, but it 

1078 differs from the one found in the registry. 

1079 LookupError 

1080 Raised if no matching dataset exists in the `Registry`. 

1081 TypeError 

1082 Raised if no collections were provided. 

1083 

1084 Notes 

1085 ----- 

1086 When looking up datasets in a `~CollectionType.CALIBRATION` collection, 

1087 this method requires that the given data ID include temporal dimensions 

1088 beyond the dimensions of the dataset type itself, in order to find the 

1089 dataset with the appropriate validity range. For example, a "bias" 

1090 dataset with native dimensions ``{instrument, detector}`` could be 

1091 fetched with a ``{instrument, detector, exposure}`` data ID, because 

1092 ``exposure`` is a temporal dimension. 

1093 """ 

1094 log.debug("Butler get: %s, dataId=%s, parameters=%s", datasetRefOrType, dataId, parameters) 

1095 ref = self._findDatasetRef(datasetRefOrType, dataId, collections=collections, **kwargs) 

1096 return self.getDirect(ref, parameters=parameters) 

1097 

1098 def getURIs(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

1099 dataId: Optional[DataId] = None, *, 

1100 predict: bool = False, 

1101 collections: Any = None, 

1102 run: Optional[str] = None, 

1103 **kwargs: Any) -> Tuple[Optional[ButlerURI], Dict[str, ButlerURI]]: 

1104 """Returns the URIs associated with the dataset. 

1105 

1106 Parameters 

1107 ---------- 

1108 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1109 When `DatasetRef` the `dataId` should be `None`. 

1110 Otherwise the `DatasetType` or name thereof. 

1111 dataId : `dict` or `DataCoordinate` 

1112 A `dict` of `Dimension` link name, value pairs that label the 

1113 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1114 should be provided as the first argument. 

1115 predict : `bool` 

1116 If `True`, allow URIs to be returned of datasets that have not 

1117 been written. 

1118 collections : Any, optional 

1119 Collections to be searched, overriding ``self.collections``. 

1120 Can be any of the types supported by the ``collections`` argument 

1121 to butler construction. 

1122 run : `str`, optional 

1123 Run to use for predictions, overriding ``self.run``. 

1124 **kwargs 

1125 Additional keyword arguments used to augment or construct a 

1126 `DataCoordinate`. See `DataCoordinate.standardize` 

1127 parameters. 

1128 

1129 Returns 

1130 ------- 

1131 primary : `ButlerURI` 

1132 The URI to the primary artifact associated with this dataset. 

1133 If the dataset was disassembled within the datastore this 

1134 may be `None`. 

1135 components : `dict` 

1136 URIs to any components associated with the dataset artifact. 

1137 Can be empty if there are no components. 

1138 """ 

1139 ref = self._findDatasetRef(datasetRefOrType, dataId, allowUnresolved=predict, 

1140 collections=collections, **kwargs) 

1141 if ref.id is None: # only possible if predict is True 

1142 if run is None: 

1143 run = self.run 

1144 if run is None: 

1145 raise TypeError("Cannot predict location with run=None.") 

1146 # Lie about ID, because we can't guess it, and only 

1147 # Datastore.getURIs() will ever see it (and it doesn't use it). 

1148 ref = ref.resolved(id=0, run=run) 

1149 return self.datastore.getURIs(ref, predict) 

1150 

1151 def getURI(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

1152 dataId: Optional[DataId] = None, *, 

1153 predict: bool = False, 

1154 collections: Any = None, 

1155 run: Optional[str] = None, 

1156 **kwargs: Any) -> ButlerURI: 

1157 """Return the URI to the Dataset. 

1158 

1159 Parameters 

1160 ---------- 

1161 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1162 When `DatasetRef` the `dataId` should be `None`. 

1163 Otherwise the `DatasetType` or name thereof. 

1164 dataId : `dict` or `DataCoordinate` 

1165 A `dict` of `Dimension` link name, value pairs that label the 

1166 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1167 should be provided as the first argument. 

1168 predict : `bool` 

1169 If `True`, allow URIs to be returned of datasets that have not 

1170 been written. 

1171 collections : Any, optional 

1172 Collections to be searched, overriding ``self.collections``. 

1173 Can be any of the types supported by the ``collections`` argument 

1174 to butler construction. 

1175 run : `str`, optional 

1176 Run to use for predictions, overriding ``self.run``. 

1177 **kwargs 

1178 Additional keyword arguments used to augment or construct a 

1179 `DataCoordinate`. See `DataCoordinate.standardize` 

1180 parameters. 

1181 

1182 Returns 

1183 ------- 

1184 uri : `ButlerURI` 

1185 URI pointing to the Dataset within the datastore. If the 

1186 Dataset does not exist in the datastore, and if ``predict`` is 

1187 `True`, the URI will be a prediction and will include a URI 

1188 fragment "#predicted". 

1189 If the datastore does not have entities that relate well 

1190 to the concept of a URI the returned URI string will be 

1191 descriptive. The returned URI is not guaranteed to be obtainable. 

1192 

1193 Raises 

1194 ------ 

1195 LookupError 

1196 A URI has been requested for a dataset that does not exist and 

1197 guessing is not allowed. 

1198 ValueError 

1199 Raised if a resolved `DatasetRef` was passed as an input, but it 

1200 differs from the one found in the registry. 

1201 TypeError 

1202 Raised if no collections were provided. 

1203 RuntimeError 

1204 Raised if a URI is requested for a dataset that consists of 

1205 multiple artifacts. 

1206 """ 

1207 primary, components = self.getURIs(datasetRefOrType, dataId=dataId, predict=predict, 

1208 collections=collections, run=run, **kwargs) 

1209 

1210 if primary is None or components: 

1211 raise RuntimeError(f"Dataset ({datasetRefOrType}) includes distinct URIs for components. " 

1212 "Use Butler.getURIs() instead.") 

1213 return primary 

1214 

1215 def retrieveArtifacts(self, refs: Iterable[DatasetRef], 

1216 destination: Union[str, ButlerURI], transfer: str = "auto", 

1217 preserve_path: bool = True, 

1218 overwrite: bool = False) -> List[ButlerURI]: 

1219 """Retrieve the artifacts associated with the supplied refs. 

1220 

1221 Parameters 

1222 ---------- 

1223 refs : iterable of `DatasetRef` 

1224 The datasets for which artifacts are to be retrieved. 

1225 A single ref can result in multiple artifacts. The refs must 

1226 be resolved. 

1227 destination : `ButlerURI` or `str` 

1228 Location to write the artifacts. 

1229 transfer : `str`, optional 

1230 Method to use to transfer the artifacts. Must be one of the options 

1231 supported by `ButlerURI.transfer_from()`. "move" is not allowed. 

1232 preserve_path : `bool`, optional 

1233 If `True` the full path of the artifact within the datastore 

1234 is preserved. If `False` the final file component of the path 

1235 is used. 

1236 overwrite : `bool`, optional 

1237 If `True` allow transfers to overwrite existing files at the 

1238 destination. 

1239 

1240 Returns 

1241 ------- 

1242 targets : `list` of `ButlerURI` 

1243 URIs of file artifacts in destination location. Order is not 

1244 preserved. 

1245 

1246 Notes 

1247 ----- 

1248 For non-file datastores the artifacts written to the destination 

1249 may not match the representation inside the datastore. For example 

1250 a hierarchical data structure in a NoSQL database may well be stored 

1251 as a JSON file. 

1252 """ 

1253 return self.datastore.retrieveArtifacts(refs, ButlerURI(destination), transfer=transfer, 

1254 preserve_path=preserve_path, overwrite=overwrite) 

1255 

1256 def datasetExists(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

1257 dataId: Optional[DataId] = None, *, 

1258 collections: Any = None, 

1259 **kwargs: Any) -> bool: 

1260 """Return True if the Dataset is actually present in the Datastore. 

1261 

1262 Parameters 

1263 ---------- 

1264 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1265 When `DatasetRef` the `dataId` should be `None`. 

1266 Otherwise the `DatasetType` or name thereof. 

1267 dataId : `dict` or `DataCoordinate` 

1268 A `dict` of `Dimension` link name, value pairs that label the 

1269 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1270 should be provided as the first argument. 

1271 collections : Any, optional 

1272 Collections to be searched, overriding ``self.collections``. 

1273 Can be any of the types supported by the ``collections`` argument 

1274 to butler construction. 

1275 **kwargs 

1276 Additional keyword arguments used to augment or construct a 

1277 `DataCoordinate`. See `DataCoordinate.standardize` 

1278 parameters. 

1279 

1280 Raises 

1281 ------ 

1282 LookupError 

1283 Raised if the dataset is not even present in the Registry. 

1284 ValueError 

1285 Raised if a resolved `DatasetRef` was passed as an input, but it 

1286 differs from the one found in the registry. 

1287 TypeError 

1288 Raised if no collections were provided. 

1289 """ 

1290 ref = self._findDatasetRef(datasetRefOrType, dataId, collections=collections, **kwargs) 

1291 return self.datastore.exists(ref) 

1292 

1293 def removeRuns(self, names: Iterable[str], unstore: bool = True) -> None: 

1294 """Remove one or more `~CollectionType.RUN` collections and the 

1295 datasets within them. 

1296 

1297 Parameters 

1298 ---------- 

1299 names : `Iterable` [ `str` ] 

1300 The names of the collections to remove. 

1301 unstore : `bool`, optional 

1302 If `True` (default), delete datasets from all datastores in which 

1303 they are present, and attempt to rollback the registry deletions if 

1304 datastore deletions fail (which may not always be possible). If 

1305 `False`, datastore records for these datasets are still removed, 

1306 but any artifacts (e.g. files) will not be. 

1307 

1308 Raises 

1309 ------ 

1310 TypeError 

1311 Raised if one or more collections are not of type 

1312 `~CollectionType.RUN`. 

1313 """ 

1314 if not self.isWriteable(): 

1315 raise TypeError("Butler is read-only.") 

1316 names = list(names) 

1317 refs: List[DatasetRef] = [] 

1318 for name in names: 

1319 collectionType = self.registry.getCollectionType(name) 

1320 if collectionType is not CollectionType.RUN: 

1321 raise TypeError(f"The collection type of '{name}' is {collectionType.name}, not RUN.") 

1322 refs.extend(self.registry.queryDatasets(..., collections=name, findFirst=True)) 

1323 with self.registry.transaction(): 

1324 if unstore: 

1325 self.datastore.trash(refs) 

1326 else: 

1327 self.datastore.forget(refs) 

1328 for name in names: 

1329 self.registry.removeCollection(name) 

1330 if unstore: 

1331 # Point of no return for removing artifacts 

1332 self.datastore.emptyTrash() 

1333 

1334 def pruneCollection(self, name: str, purge: bool = False, unstore: bool = False, 

1335 unlink: Optional[List[str]] = None) -> None: 

1336 """Remove a collection and possibly prune datasets within it. 

1337 

1338 Parameters 

1339 ---------- 

1340 name : `str` 

1341 Name of the collection to remove. If this is a 

1342 `~CollectionType.TAGGED` or `~CollectionType.CHAINED` collection, 

1343 datasets within the collection are not modified unless ``unstore`` 

1344 is `True`. If this is a `~CollectionType.RUN` collection, 

1345 ``purge`` and ``unstore`` must be `True`, and all datasets in it 

1346 are fully removed from the data repository. 

1347 purge : `bool`, optional 

1348 If `True`, permit `~CollectionType.RUN` collections to be removed, 

1349 fully removing datasets within them. Requires ``unstore=True`` as 

1350 well as an added precaution against accidental deletion. Must be 

1351 `False` (default) if the collection is not a ``RUN``. 

1352 unstore: `bool`, optional 

1353 If `True`, remove all datasets in the collection from all 

1354 datastores in which they appear. 

1355 unlink: `list` [`str`], optional 

1356 Before removing the given `collection` unlink it from from these 

1357 parent collections. 

1358 

1359 Raises 

1360 ------ 

1361 TypeError 

1362 Raised if the butler is read-only or arguments are mutually 

1363 inconsistent. 

1364 """ 

1365 # See pruneDatasets comments for more information about the logic here; 

1366 # the cases are almost the same, but here we can rely on Registry to 

1367 # take care everything but Datastore deletion when we remove the 

1368 # collection. 

1369 if not self.isWriteable(): 

1370 raise TypeError("Butler is read-only.") 

1371 collectionType = self.registry.getCollectionType(name) 

1372 if purge and not unstore: 

1373 raise PurgeWithoutUnstorePruneCollectionsError() 

1374 if collectionType is CollectionType.RUN and not purge: 

1375 raise RunWithoutPurgePruneCollectionsError(collectionType) 

1376 if collectionType is not CollectionType.RUN and purge: 

1377 raise PurgeUnsupportedPruneCollectionsError(collectionType) 

1378 

1379 def remove(child: str, parent: str) -> None: 

1380 """Remove a child collection from a parent collection.""" 

1381 # Remove child from parent. 

1382 chain = list(self.registry.getCollectionChain(parent)) 

1383 try: 

1384 chain.remove(name) 

1385 except ValueError as e: 

1386 raise RuntimeError(f"{name} is not a child of {parent}") from e 

1387 self.registry.setCollectionChain(parent, chain) 

1388 

1389 with self.registry.transaction(): 

1390 if (unlink): 

1391 for parent in unlink: 

1392 remove(name, parent) 

1393 if unstore: 

1394 refs = self.registry.queryDatasets(..., collections=name, findFirst=True) 

1395 self.datastore.trash(refs) 

1396 self.registry.removeCollection(name) 

1397 

1398 if unstore: 

1399 # Point of no return for removing artifacts 

1400 self.datastore.emptyTrash() 

1401 

1402 def pruneDatasets(self, refs: Iterable[DatasetRef], *, 

1403 disassociate: bool = True, 

1404 unstore: bool = False, 

1405 tags: Iterable[str] = (), 

1406 purge: bool = False, 

1407 run: Optional[str] = None) -> None: 

1408 """Remove one or more datasets from a collection and/or storage. 

1409 

1410 Parameters 

1411 ---------- 

1412 refs : `~collections.abc.Iterable` of `DatasetRef` 

1413 Datasets to prune. These must be "resolved" references (not just 

1414 a `DatasetType` and data ID). 

1415 disassociate : `bool`, optional 

1416 Disassociate pruned datasets from ``tags``, or from all collections 

1417 if ``purge=True``. 

1418 unstore : `bool`, optional 

1419 If `True` (`False` is default) remove these datasets from all 

1420 datastores known to this butler. Note that this will make it 

1421 impossible to retrieve these datasets even via other collections. 

1422 Datasets that are already not stored are ignored by this option. 

1423 tags : `Iterable` [ `str` ], optional 

1424 `~CollectionType.TAGGED` collections to disassociate the datasets 

1425 from. Ignored if ``disassociate`` is `False` or ``purge`` is 

1426 `True`. 

1427 purge : `bool`, optional 

1428 If `True` (`False` is default), completely remove the dataset from 

1429 the `Registry`. To prevent accidental deletions, ``purge`` may 

1430 only be `True` if all of the following conditions are met: 

1431 

1432 - All given datasets are in the given run. 

1433 - ``disassociate`` is `True`; 

1434 - ``unstore`` is `True`. 

1435 

1436 This mode may remove provenance information from datasets other 

1437 than those provided, and should be used with extreme care. 

1438 

1439 Raises 

1440 ------ 

1441 TypeError 

1442 Raised if the butler is read-only, if no collection was provided, 

1443 or the conditions for ``purge=True`` were not met. 

1444 """ 

1445 if not self.isWriteable(): 

1446 raise TypeError("Butler is read-only.") 

1447 if purge: 

1448 if not disassociate: 

1449 raise TypeError("Cannot pass purge=True without disassociate=True.") 

1450 if not unstore: 

1451 raise TypeError("Cannot pass purge=True without unstore=True.") 

1452 elif disassociate: 

1453 tags = tuple(tags) 

1454 if not tags: 

1455 raise TypeError("No tags provided but disassociate=True.") 

1456 for tag in tags: 

1457 collectionType = self.registry.getCollectionType(tag) 

1458 if collectionType is not CollectionType.TAGGED: 

1459 raise TypeError(f"Cannot disassociate from collection '{tag}' " 

1460 f"of non-TAGGED type {collectionType.name}.") 

1461 # Transform possibly-single-pass iterable into something we can iterate 

1462 # over multiple times. 

1463 refs = list(refs) 

1464 # Pruning a component of a DatasetRef makes no sense since registry 

1465 # doesn't know about components and datastore might not store 

1466 # components in a separate file 

1467 for ref in refs: 

1468 if ref.datasetType.component(): 

1469 raise ValueError(f"Can not prune a component of a dataset (ref={ref})") 

1470 # We don't need an unreliable Datastore transaction for this, because 

1471 # we've been extra careful to ensure that Datastore.trash only involves 

1472 # mutating the Registry (it can _look_ at Datastore-specific things, 

1473 # but shouldn't change them), and hence all operations here are 

1474 # Registry operations. 

1475 with self.registry.transaction(): 

1476 if unstore: 

1477 self.datastore.trash(refs) 

1478 if purge: 

1479 self.registry.removeDatasets(refs) 

1480 elif disassociate: 

1481 assert tags, "Guaranteed by earlier logic in this function." 

1482 for tag in tags: 

1483 self.registry.disassociate(tag, refs) 

1484 # We've exited the Registry transaction, and apparently committed. 

1485 # (if there was an exception, everything rolled back, and it's as if 

1486 # nothing happened - and we never get here). 

1487 # Datastore artifacts are not yet gone, but they're clearly marked 

1488 # as trash, so if we fail to delete now because of (e.g.) filesystem 

1489 # problems we can try again later, and if manual administrative 

1490 # intervention is required, it's pretty clear what that should entail: 

1491 # deleting everything on disk and in private Datastore tables that is 

1492 # in the dataset_location_trash table. 

1493 if unstore: 

1494 # Point of no return for removing artifacts 

1495 self.datastore.emptyTrash() 

1496 

1497 @transactional 

1498 def ingest(self, *datasets: FileDataset, transfer: Optional[str] = "auto", run: Optional[str] = None, 

1499 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

1500 ) -> None: 

1501 """Store and register one or more datasets that already exist on disk. 

1502 

1503 Parameters 

1504 ---------- 

1505 datasets : `FileDataset` 

1506 Each positional argument is a struct containing information about 

1507 a file to be ingested, including its URI (either absolute or 

1508 relative to the datastore root, if applicable), a `DatasetRef`, 

1509 and optionally a formatter class or its fully-qualified string 

1510 name. If a formatter is not provided, the formatter that would be 

1511 used for `put` is assumed. On successful return, all 

1512 `FileDataset.ref` attributes will have their `DatasetRef.id` 

1513 attribute populated and all `FileDataset.formatter` attributes will 

1514 be set to the formatter class used. `FileDataset.path` attributes 

1515 may be modified to put paths in whatever the datastore considers a 

1516 standardized form. 

1517 transfer : `str`, optional 

1518 If not `None`, must be one of 'auto', 'move', 'copy', 'direct', 

1519 'split', 'hardlink', 'relsymlink' or 'symlink', indicating how to 

1520 transfer the file. 

1521 run : `str`, optional 

1522 The name of the run ingested datasets should be added to, 

1523 overriding ``self.run``. 

1524 idGenerationMode : `DatasetIdGenEnum`, optional 

1525 Specifies option for generating dataset IDs. By default unique IDs 

1526 are generated for each inserted dataset. 

1527 

1528 Raises 

1529 ------ 

1530 TypeError 

1531 Raised if the butler is read-only or if no run was provided. 

1532 NotImplementedError 

1533 Raised if the `Datastore` does not support the given transfer mode. 

1534 DatasetTypeNotSupportedError 

1535 Raised if one or more files to be ingested have a dataset type that 

1536 is not supported by the `Datastore`.. 

1537 FileNotFoundError 

1538 Raised if one of the given files does not exist. 

1539 FileExistsError 

1540 Raised if transfer is not `None` but the (internal) location the 

1541 file would be moved to is already occupied. 

1542 

1543 Notes 

1544 ----- 

1545 This operation is not fully exception safe: if a database operation 

1546 fails, the given `FileDataset` instances may be only partially updated. 

1547 

1548 It is atomic in terms of database operations (they will either all 

1549 succeed or all fail) providing the database engine implements 

1550 transactions correctly. It will attempt to be atomic in terms of 

1551 filesystem operations as well, but this cannot be implemented 

1552 rigorously for most datastores. 

1553 """ 

1554 if not self.isWriteable(): 

1555 raise TypeError("Butler is read-only.") 

1556 progress = Progress("lsst.daf.butler.Butler.ingest", level=logging.DEBUG) 

1557 # Reorganize the inputs so they're grouped by DatasetType and then 

1558 # data ID. We also include a list of DatasetRefs for each FileDataset 

1559 # to hold the resolved DatasetRefs returned by the Registry, before 

1560 # it's safe to swap them into FileDataset.refs. 

1561 # Some type annotation aliases to make that clearer: 

1562 GroupForType = Dict[DataCoordinate, Tuple[FileDataset, List[DatasetRef]]] 

1563 GroupedData = MutableMapping[DatasetType, GroupForType] 

1564 # The actual data structure: 

1565 groupedData: GroupedData = defaultdict(dict) 

1566 # And the nested loop that populates it: 

1567 for dataset in progress.wrap(datasets, desc="Grouping by dataset type"): 

1568 # This list intentionally shared across the inner loop, since it's 

1569 # associated with `dataset`. 

1570 resolvedRefs: List[DatasetRef] = [] 

1571 

1572 # Somewhere to store pre-existing refs if we have an 

1573 # execution butler. 

1574 existingRefs: List[DatasetRef] = [] 

1575 

1576 for ref in dataset.refs: 

1577 if ref.dataId in groupedData[ref.datasetType]: 

1578 raise ConflictingDefinitionError(f"Ingest conflict. Dataset {dataset.path} has same" 

1579 " DataId as other ingest dataset" 

1580 f" {groupedData[ref.datasetType][ref.dataId][0].path} " 

1581 f" ({ref.dataId})") 

1582 if self._allow_put_of_predefined_dataset: 

1583 existing_ref = self.registry.findDataset(ref.datasetType, 

1584 dataId=ref.dataId, 

1585 collections=run) 

1586 if existing_ref: 

1587 if self.datastore.knows(existing_ref): 

1588 raise ConflictingDefinitionError(f"Dataset associated with path {dataset.path}" 

1589 f" already exists as {existing_ref}.") 

1590 # Store this ref elsewhere since it already exists 

1591 # and we do not want to remake it but we do want 

1592 # to store it in the datastore. 

1593 existingRefs.append(existing_ref) 

1594 

1595 # Nothing else to do until we have finished 

1596 # iterating. 

1597 continue 

1598 

1599 groupedData[ref.datasetType][ref.dataId] = (dataset, resolvedRefs) 

1600 

1601 if existingRefs: 

1602 

1603 if len(dataset.refs) != len(existingRefs): 

1604 # Keeping track of partially pre-existing datasets is hard 

1605 # and should generally never happen. For now don't allow 

1606 # it. 

1607 raise ConflictingDefinitionError(f"For dataset {dataset.path} some dataIds already exist" 

1608 " in registry but others do not. This is not supported.") 

1609 

1610 # Attach the resolved refs if we found them. 

1611 dataset.refs = existingRefs 

1612 

1613 # Now we can bulk-insert into Registry for each DatasetType. 

1614 for datasetType, groupForType in progress.iter_item_chunks(groupedData.items(), 

1615 desc="Bulk-inserting datasets by type"): 

1616 refs = self.registry.insertDatasets( 

1617 datasetType, 

1618 dataIds=groupForType.keys(), 

1619 run=run, 

1620 expand=self.datastore.needs_expanded_data_ids(transfer, datasetType), 

1621 idGenerationMode=idGenerationMode, 

1622 ) 

1623 # Append those resolved DatasetRefs to the new lists we set up for 

1624 # them. 

1625 for ref, (_, resolvedRefs) in zip(refs, groupForType.values()): 

1626 resolvedRefs.append(ref) 

1627 

1628 # Go back to the original FileDatasets to replace their refs with the 

1629 # new resolved ones. 

1630 for groupForType in progress.iter_chunks(groupedData.values(), 

1631 desc="Reassociating resolved dataset refs with files"): 

1632 for dataset, resolvedRefs in groupForType.values(): 

1633 dataset.refs = resolvedRefs 

1634 

1635 # Bulk-insert everything into Datastore. 

1636 self.datastore.ingest(*datasets, transfer=transfer) 

1637 

1638 @contextlib.contextmanager 

1639 def export(self, *, directory: Optional[str] = None, 

1640 filename: Optional[str] = None, 

1641 format: Optional[str] = None, 

1642 transfer: Optional[str] = None) -> Iterator[RepoExportContext]: 

1643 """Export datasets from the repository represented by this `Butler`. 

1644 

1645 This method is a context manager that returns a helper object 

1646 (`RepoExportContext`) that is used to indicate what information from 

1647 the repository should be exported. 

1648 

1649 Parameters 

1650 ---------- 

1651 directory : `str`, optional 

1652 Directory dataset files should be written to if ``transfer`` is not 

1653 `None`. 

1654 filename : `str`, optional 

1655 Name for the file that will include database information associated 

1656 with the exported datasets. If this is not an absolute path and 

1657 ``directory`` is not `None`, it will be written to ``directory`` 

1658 instead of the current working directory. Defaults to 

1659 "export.{format}". 

1660 format : `str`, optional 

1661 File format for the database information file. If `None`, the 

1662 extension of ``filename`` will be used. 

1663 transfer : `str`, optional 

1664 Transfer mode passed to `Datastore.export`. 

1665 

1666 Raises 

1667 ------ 

1668 TypeError 

1669 Raised if the set of arguments passed is inconsistent. 

1670 

1671 Examples 

1672 -------- 

1673 Typically the `Registry.queryDataIds` and `Registry.queryDatasets` 

1674 methods are used to provide the iterables over data IDs and/or datasets 

1675 to be exported:: 

1676 

1677 with butler.export("exports.yaml") as export: 

1678 # Export all flats, but none of the dimension element rows 

1679 # (i.e. data ID information) associated with them. 

1680 export.saveDatasets(butler.registry.queryDatasets("flat"), 

1681 elements=()) 

1682 # Export all datasets that start with "deepCoadd_" and all of 

1683 # their associated data ID information. 

1684 export.saveDatasets(butler.registry.queryDatasets("deepCoadd_*")) 

1685 """ 

1686 if directory is None and transfer is not None: 

1687 raise TypeError("Cannot transfer without providing a directory.") 

1688 if transfer == "move": 

1689 raise TypeError("Transfer may not be 'move': export is read-only") 

1690 if format is None: 

1691 if filename is None: 

1692 raise TypeError("At least one of 'filename' or 'format' must be provided.") 

1693 else: 

1694 _, format = os.path.splitext(filename) 

1695 elif filename is None: 

1696 filename = f"export.{format}" 

1697 if directory is not None: 

1698 filename = os.path.join(directory, filename) 

1699 BackendClass = getClassOf(self._config["repo_transfer_formats"][format]["export"]) 

1700 with open(filename, 'w') as stream: 

1701 backend = BackendClass(stream) 

1702 try: 

1703 helper = RepoExportContext(self.registry, self.datastore, backend=backend, 

1704 directory=directory, transfer=transfer) 

1705 yield helper 

1706 except BaseException: 

1707 raise 

1708 else: 

1709 helper._finish() 

1710 

1711 def import_(self, *, directory: Optional[str] = None, 

1712 filename: Union[str, TextIO, None] = None, 

1713 format: Optional[str] = None, 

1714 transfer: Optional[str] = None, 

1715 skip_dimensions: Optional[Set] = None, 

1716 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

1717 reuseIds: bool = False) -> None: 

1718 """Import datasets into this repository that were exported from a 

1719 different butler repository via `~lsst.daf.butler.Butler.export`. 

1720 

1721 Parameters 

1722 ---------- 

1723 directory : `str`, optional 

1724 Directory containing dataset files to import from. If `None`, 

1725 ``filename`` and all dataset file paths specified therein must 

1726 be absolute. 

1727 filename : `str` or `TextIO`, optional 

1728 A stream or name of file that contains database information 

1729 associated with the exported datasets, typically generated by 

1730 `~lsst.daf.butler.Butler.export`. If this a string (name) and 

1731 is not an absolute path, does not exist in the current working 

1732 directory, and ``directory`` is not `None`, it is assumed to be in 

1733 ``directory``. Defaults to "export.{format}". 

1734 format : `str`, optional 

1735 File format for ``filename``. If `None`, the extension of 

1736 ``filename`` will be used. 

1737 transfer : `str`, optional 

1738 Transfer mode passed to `~lsst.daf.butler.Datastore.ingest`. 

1739 skip_dimensions : `set`, optional 

1740 Names of dimensions that should be skipped and not imported. 

1741 idGenerationMode : `DatasetIdGenEnum`, optional 

1742 Specifies option for generating dataset IDs when IDs are not 

1743 provided or their type does not match backend type. By default 

1744 unique IDs are generated for each inserted dataset. 

1745 reuseIds : `bool`, optional 

1746 If `True` then forces re-use of imported dataset IDs for integer 

1747 IDs which are normally generated as auto-incremented; exception 

1748 will be raised if imported IDs clash with existing ones. This 

1749 option has no effect on the use of globally-unique IDs which are 

1750 always re-used (or generated if integer IDs are being imported). 

1751 

1752 Raises 

1753 ------ 

1754 TypeError 

1755 Raised if the set of arguments passed is inconsistent, or if the 

1756 butler is read-only. 

1757 """ 

1758 if not self.isWriteable(): 

1759 raise TypeError("Butler is read-only.") 

1760 if format is None: 

1761 if filename is None: 

1762 raise TypeError("At least one of 'filename' or 'format' must be provided.") 

1763 else: 

1764 _, format = os.path.splitext(filename) # type: ignore 

1765 elif filename is None: 

1766 filename = f"export.{format}" 

1767 if isinstance(filename, str) and directory is not None and not os.path.exists(filename): 

1768 filename = os.path.join(directory, filename) 

1769 BackendClass = getClassOf(self._config["repo_transfer_formats"][format]["import"]) 

1770 

1771 def doImport(importStream: TextIO) -> None: 

1772 backend = BackendClass(importStream, self.registry) 

1773 backend.register() 

1774 with self.transaction(): 

1775 backend.load(self.datastore, directory=directory, transfer=transfer, 

1776 skip_dimensions=skip_dimensions, idGenerationMode=idGenerationMode, 

1777 reuseIds=reuseIds) 

1778 

1779 if isinstance(filename, str): 

1780 with open(filename, "r") as stream: 

1781 doImport(stream) 

1782 else: 

1783 doImport(filename) 

1784 

1785 def transfer_from(self, source_butler: Butler, source_refs: Iterable[DatasetRef], 

1786 transfer: str = "auto", 

1787 id_gen_map: Dict[str, DatasetIdGenEnum] = None, 

1788 skip_missing: bool = True) -> List[DatasetRef]: 

1789 """Transfer datasets to this Butler from a run in another Butler. 

1790 

1791 Parameters 

1792 ---------- 

1793 source_butler : `Butler` 

1794 Butler from which the datasets are to be transferred. 

1795 source_refs : iterable of `DatasetRef` 

1796 Datasets defined in the source butler that should be transferred to 

1797 this butler. 

1798 transfer : `str`, optional 

1799 Transfer mode passed to `~lsst.daf.butler.Datastore.transfer_from`. 

1800 id_gen_map : `dict` of [`str`, `DatasetIdGenEnum`], optional 

1801 A mapping of dataset type to ID generation mode. Only used if 

1802 the source butler is using integer IDs. Should not be used 

1803 if this receiving butler uses integer IDs. Without this dataset 

1804 import always uses unique. 

1805 skip_missing : `bool` 

1806 If `True`, datasets with no datastore artifact associated with 

1807 them are not transferred. 

1808 

1809 Returns 

1810 ------- 

1811 refs : `list` of `DatasetRef` 

1812 The refs added to this Butler. 

1813 

1814 Notes 

1815 ----- 

1816 Requires that any dimension definitions are already present in the 

1817 receiving Butler. The datastore artifact has to exist for a transfer 

1818 to be made but non-existence is not an error. 

1819 

1820 Datasets that already exist in this run will be skipped. 

1821 

1822 The datasets are imported as part of a transaction, although 

1823 dataset types are registered before the transaction is started. 

1824 This means that it is possible for a dataset type to be registered 

1825 even though transfer has failed. 

1826 """ 

1827 if not self.isWriteable(): 

1828 raise TypeError("Butler is read-only.") 

1829 progress = Progress("lsst.daf.butler.Butler.transfer_from", level=VERBOSE) 

1830 

1831 # Will iterate through the refs multiple times so need to convert 

1832 # to a list if this isn't a collection. 

1833 if not isinstance(source_refs, collections.abc.Collection): 

1834 source_refs = list(source_refs) 

1835 

1836 log.info("Transferring %d datasets into %s", len(source_refs), str(self)) 

1837 

1838 if id_gen_map is None: 

1839 id_gen_map = {} 

1840 

1841 # In some situations the datastore artifact may be missing 

1842 # and we do not want that registry entry to be imported. 

1843 # Asking datastore is not sufficient, the records may have been 

1844 # purged, we have to ask for the (predicted) URI and check 

1845 # existence explicitly. Execution butler is set up exactly like 

1846 # this with no datastore records. 

1847 if skip_missing: 

1848 source_refs = [ref for ref in source_refs if source_butler.datastore.exists(ref)] 

1849 

1850 # Importing requires that we group the refs by dataset type and run 

1851 # before doing the import. 

1852 grouped_refs = defaultdict(list) 

1853 grouped_indices = defaultdict(list) 

1854 for i, ref in enumerate(source_refs): 

1855 grouped_refs[ref.datasetType, ref.run].append(ref) 

1856 grouped_indices[ref.datasetType, ref.run].append(i) 

1857 

1858 # Register any dataset types we need. This has to be done outside 

1859 # of a transaction and so will not be rolled back on failure. 

1860 for datasetType, _ in grouped_refs: 

1861 self.registry.registerDatasetType(datasetType) 

1862 

1863 # The returned refs should be identical for UUIDs. 

1864 # For now must also support integers and so need to retain the 

1865 # newly-created refs from this registry. 

1866 # Pre-size it so we can assign refs into the correct slots 

1867 transferred_refs_tmp: List[Optional[DatasetRef]] = [None] * len(source_refs) 

1868 default_id_gen = DatasetIdGenEnum.UNIQUE 

1869 

1870 # Do all the importing in a single transaction. 

1871 with self.transaction(): 

1872 for (datasetType, run), refs_to_import in progress.iter_item_chunks(grouped_refs.items(), 

1873 desc="Importing to registry" 

1874 " by run and dataset type"): 

1875 run_doc = source_butler.registry.getCollectionDocumentation(run) 

1876 self.registry.registerCollection(run, CollectionType.RUN, doc=run_doc) 

1877 

1878 id_generation_mode = default_id_gen 

1879 if isinstance(refs_to_import[0].id, int): 

1880 # ID generation mode might need to be overridden when 

1881 # targetting UUID 

1882 id_generation_mode = id_gen_map.get(datasetType.name, default_id_gen) 

1883 

1884 n_refs = len(refs_to_import) 

1885 log.log(VERBOSE, "Importing %d ref%s of dataset type %s into run %s", 

1886 n_refs, "" if n_refs == 1 else "s", datasetType.name, run) 

1887 

1888 # No way to know if this butler's registry uses UUID. 

1889 # We have to trust the caller on this. If it fails they will 

1890 # have to change their approach. We can't catch the exception 

1891 # and retry with unique because that will mess up the 

1892 # transaction handling. We aren't allowed to ask the registry 

1893 # manager what type of ID it is using. 

1894 imported_refs = self.registry._importDatasets(refs_to_import, 

1895 idGenerationMode=id_generation_mode, 

1896 expand=False) 

1897 

1898 # Map them into the correct slots to match the initial order 

1899 for i, ref in zip(grouped_indices[datasetType, run], imported_refs): 

1900 transferred_refs_tmp[i] = ref 

1901 

1902 # Mypy insists that we might have None in here so we have to make 

1903 # that explicit by assigning to a new variable and filtering out 

1904 # something that won't be there. 

1905 transferred_refs = [ref for ref in transferred_refs_tmp if ref is not None] 

1906 

1907 # Check consistency 

1908 assert len(source_refs) == len(transferred_refs), "Different number of refs imported than given" 

1909 

1910 log.log(VERBOSE, "Imported %d datasets into destination butler", len(transferred_refs)) 

1911 

1912 # The transferred refs need to be reordered to match the original 

1913 # ordering given by the caller. Without this the datastore transfer 

1914 # will be broken. 

1915 

1916 # Ask the datastore to transfer. The datastore has to check that 

1917 # the source datastore is compatible with the target datastore. 

1918 self.datastore.transfer_from(source_butler.datastore, source_refs, 

1919 local_refs=transferred_refs, transfer=transfer) 

1920 

1921 return transferred_refs 

1922 

1923 def validateConfiguration(self, logFailures: bool = False, 

1924 datasetTypeNames: Optional[Iterable[str]] = None, 

1925 ignore: Iterable[str] = None) -> None: 

1926 """Validate butler configuration. 

1927 

1928 Checks that each `DatasetType` can be stored in the `Datastore`. 

1929 

1930 Parameters 

1931 ---------- 

1932 logFailures : `bool`, optional 

1933 If `True`, output a log message for every validation error 

1934 detected. 

1935 datasetTypeNames : iterable of `str`, optional 

1936 The `DatasetType` names that should be checked. This allows 

1937 only a subset to be selected. 

1938 ignore : iterable of `str`, optional 

1939 Names of DatasetTypes to skip over. This can be used to skip 

1940 known problems. If a named `DatasetType` corresponds to a 

1941 composite, all components of that `DatasetType` will also be 

1942 ignored. 

1943 

1944 Raises 

1945 ------ 

1946 ButlerValidationError 

1947 Raised if there is some inconsistency with how this Butler 

1948 is configured. 

1949 """ 

1950 if datasetTypeNames: 

1951 datasetTypes = [self.registry.getDatasetType(name) for name in datasetTypeNames] 

1952 else: 

1953 datasetTypes = list(self.registry.queryDatasetTypes()) 

1954 

1955 # filter out anything from the ignore list 

1956 if ignore: 

1957 ignore = set(ignore) 

1958 datasetTypes = [e for e in datasetTypes 

1959 if e.name not in ignore and e.nameAndComponent()[0] not in ignore] 

1960 else: 

1961 ignore = set() 

1962 

1963 # Find all the registered instruments 

1964 instruments = set( 

1965 record.name for record in self.registry.queryDimensionRecords("instrument") 

1966 ) 

1967 

1968 # For each datasetType that has an instrument dimension, create 

1969 # a DatasetRef for each defined instrument 

1970 datasetRefs = [] 

1971 

1972 for datasetType in datasetTypes: 

1973 if "instrument" in datasetType.dimensions: 

1974 for instrument in instruments: 

1975 datasetRef = DatasetRef(datasetType, {"instrument": instrument}, # type: ignore 

1976 conform=False) 

1977 datasetRefs.append(datasetRef) 

1978 

1979 entities: List[Union[DatasetType, DatasetRef]] = [] 

1980 entities.extend(datasetTypes) 

1981 entities.extend(datasetRefs) 

1982 

1983 datastoreErrorStr = None 

1984 try: 

1985 self.datastore.validateConfiguration(entities, logFailures=logFailures) 

1986 except ValidationError as e: 

1987 datastoreErrorStr = str(e) 

1988 

1989 # Also check that the LookupKeys used by the datastores match 

1990 # registry and storage class definitions 

1991 keys = self.datastore.getLookupKeys() 

1992 

1993 failedNames = set() 

1994 failedDataId = set() 

1995 for key in keys: 

1996 if key.name is not None: 

1997 if key.name in ignore: 

1998 continue 

1999 

2000 # skip if specific datasetType names were requested and this 

2001 # name does not match 

2002 if datasetTypeNames and key.name not in datasetTypeNames: 

2003 continue 

2004 

2005 # See if it is a StorageClass or a DatasetType 

2006 if key.name in self.storageClasses: 

2007 pass 

2008 else: 

2009 try: 

2010 self.registry.getDatasetType(key.name) 

2011 except KeyError: 

2012 if logFailures: 

2013 log.critical("Key '%s' does not correspond to a DatasetType or StorageClass", key) 

2014 failedNames.add(key) 

2015 else: 

2016 # Dimensions are checked for consistency when the Butler 

2017 # is created and rendezvoused with a universe. 

2018 pass 

2019 

2020 # Check that the instrument is a valid instrument 

2021 # Currently only support instrument so check for that 

2022 if key.dataId: 

2023 dataIdKeys = set(key.dataId) 

2024 if set(["instrument"]) != dataIdKeys: 

2025 if logFailures: 

2026 log.critical("Key '%s' has unsupported DataId override", key) 

2027 failedDataId.add(key) 

2028 elif key.dataId["instrument"] not in instruments: 

2029 if logFailures: 

2030 log.critical("Key '%s' has unknown instrument", key) 

2031 failedDataId.add(key) 

2032 

2033 messages = [] 

2034 

2035 if datastoreErrorStr: 

2036 messages.append(datastoreErrorStr) 

2037 

2038 for failed, msg in ((failedNames, "Keys without corresponding DatasetType or StorageClass entry: "), 

2039 (failedDataId, "Keys with bad DataId entries: ")): 

2040 if failed: 

2041 msg += ", ".join(str(k) for k in failed) 

2042 messages.append(msg) 

2043 

2044 if messages: 

2045 raise ValidationError(";\n".join(messages)) 

2046 

2047 @property 

2048 def collections(self) -> CollectionSearch: 

2049 """The collections to search by default, in order (`CollectionSearch`). 

2050 

2051 This is an alias for ``self.registry.defaults.collections``. It cannot 

2052 be set directly in isolation, but all defaults may be changed together 

2053 by assigning a new `RegistryDefaults` instance to 

2054 ``self.registry.defaults``. 

2055 """ 

2056 return self.registry.defaults.collections 

2057 

2058 @property 

2059 def run(self) -> Optional[str]: 

2060 """Name of the run this butler writes outputs to by default (`str` or 

2061 `None`). 

2062 

2063 This is an alias for ``self.registry.defaults.run``. It cannot be set 

2064 directly in isolation, but all defaults may be changed together by 

2065 assigning a new `RegistryDefaults` instance to 

2066 ``self.registry.defaults``. 

2067 """ 

2068 return self.registry.defaults.run 

2069 

2070 registry: Registry 

2071 """The object that manages dataset metadata and relationships (`Registry`). 

2072 

2073 Most operations that don't involve reading or writing butler datasets are 

2074 accessible only via `Registry` methods. 

2075 """ 

2076 

2077 datastore: Datastore 

2078 """The object that manages actual dataset storage (`Datastore`). 

2079 

2080 Direct user access to the datastore should rarely be necessary; the primary 

2081 exception is the case where a `Datastore` implementation provides extra 

2082 functionality beyond what the base class defines. 

2083 """ 

2084 

2085 storageClasses: StorageClassFactory 

2086 """An object that maps known storage class names to objects that fully 

2087 describe them (`StorageClassFactory`). 

2088 """ 

2089 

2090 _allow_put_of_predefined_dataset: bool 

2091 """Allow a put to succeed even if there is already a registry entry for it 

2092 but not a datastore record. (`bool`)."""