Coverage for python/lsst/daf/butler/registry/_registry.py: 61%

184 statements  

« prev     ^ index     » next       coverage.py v6.5.0, created at 2023-01-26 02:04 -0800

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__ = ("Registry",) 

25 

26import contextlib 

27import logging 

28from abc import ABC, abstractmethod 

29from typing import ( 

30 TYPE_CHECKING, 

31 Any, 

32 Dict, 

33 Iterable, 

34 Iterator, 

35 List, 

36 Mapping, 

37 Optional, 

38 Sequence, 

39 Set, 

40 Tuple, 

41 Type, 

42 Union, 

43) 

44 

45from lsst.resources import ResourcePathExpression 

46from lsst.utils import doImportType 

47 

48from ..core import ( 

49 Config, 

50 DataCoordinate, 

51 DataId, 

52 DatasetAssociation, 

53 DatasetId, 

54 DatasetRef, 

55 DatasetType, 

56 Dimension, 

57 DimensionConfig, 

58 DimensionElement, 

59 DimensionGraph, 

60 DimensionRecord, 

61 DimensionUniverse, 

62 NameLookupMapping, 

63 StorageClassFactory, 

64 Timespan, 

65) 

66from ._collection_summary import CollectionSummary 

67from ._collectionType import CollectionType 

68from ._config import RegistryConfig 

69from ._defaults import RegistryDefaults 

70from .interfaces import DatasetIdFactory, DatasetIdGenEnum 

71from .queries import DataCoordinateQueryResults, DatasetQueryResults, DimensionRecordQueryResults 

72 

73if TYPE_CHECKING: 73 ↛ 74line 73 didn't jump to line 74, because the condition on line 73 was never true

74 from .._butlerConfig import ButlerConfig 

75 from .interfaces import CollectionRecord, DatastoreRegistryBridgeManager 

76 

77_LOG = logging.getLogger(__name__) 

78 

79 

80class Registry(ABC): 

81 """Abstract Registry interface. 

82 

83 Each registry implementation can have its own constructor parameters. 

84 The assumption is that an instance of a specific subclass will be 

85 constructed from configuration using `Registry.fromConfig()`. 

86 The base class will look for a ``cls`` entry and call that specific 

87 `fromConfig()` method. 

88 

89 All subclasses should store `RegistryDefaults` in a ``_defaults`` 

90 property. No other properties are assumed shared between implementations. 

91 """ 

92 

93 defaultConfigFile: Optional[str] = None 

94 """Path to configuration defaults. Accessed within the ``configs`` resource 

95 or relative to a search path. Can be None if no defaults specified. 

96 """ 

97 

98 @classmethod 

99 def forceRegistryConfig( 

100 cls, config: Optional[Union[ButlerConfig, RegistryConfig, Config, str]] 

101 ) -> RegistryConfig: 

102 """Force the supplied config to a `RegistryConfig`. 

103 

104 Parameters 

105 ---------- 

106 config : `RegistryConfig`, `Config` or `str` or `None` 

107 Registry configuration, if missing then default configuration will 

108 be loaded from registry.yaml. 

109 

110 Returns 

111 ------- 

112 registry_config : `RegistryConfig` 

113 A registry config. 

114 """ 

115 if not isinstance(config, RegistryConfig): 

116 if isinstance(config, (str, Config)) or config is None: 

117 config = RegistryConfig(config) 

118 else: 

119 raise ValueError(f"Incompatible Registry configuration: {config}") 

120 return config 

121 

122 @classmethod 

123 def determineTrampoline( 

124 cls, config: Optional[Union[ButlerConfig, RegistryConfig, Config, str]] 

125 ) -> Tuple[Type[Registry], RegistryConfig]: 

126 """Return class to use to instantiate real registry. 

127 

128 Parameters 

129 ---------- 

130 config : `RegistryConfig` or `str`, optional 

131 Registry configuration, if missing then default configuration will 

132 be loaded from registry.yaml. 

133 

134 Returns 

135 ------- 

136 requested_cls : `type` of `Registry` 

137 The real registry class to use. 

138 registry_config : `RegistryConfig` 

139 The `RegistryConfig` to use. 

140 """ 

141 config = cls.forceRegistryConfig(config) 

142 

143 # Default to the standard registry 

144 registry_cls_name = config.get("cls", "lsst.daf.butler.registries.sql.SqlRegistry") 

145 registry_cls = doImportType(registry_cls_name) 

146 if registry_cls is cls: 

147 raise ValueError("Can not instantiate the abstract base Registry from config") 

148 if not issubclass(registry_cls, Registry): 

149 raise TypeError( 

150 f"Registry class obtained from config {registry_cls_name} is not a Registry class." 

151 ) 

152 return registry_cls, config 

153 

154 @classmethod 

155 def createFromConfig( 

156 cls, 

157 config: Optional[Union[RegistryConfig, str]] = None, 

158 dimensionConfig: Optional[Union[DimensionConfig, str]] = None, 

159 butlerRoot: Optional[ResourcePathExpression] = None, 

160 ) -> Registry: 

161 """Create registry database and return `Registry` instance. 

162 

163 This method initializes database contents, database must be empty 

164 prior to calling this method. 

165 

166 Parameters 

167 ---------- 

168 config : `RegistryConfig` or `str`, optional 

169 Registry configuration, if missing then default configuration will 

170 be loaded from registry.yaml. 

171 dimensionConfig : `DimensionConfig` or `str`, optional 

172 Dimensions configuration, if missing then default configuration 

173 will be loaded from dimensions.yaml. 

174 butlerRoot : convertible to `lsst.resources.ResourcePath`, optional 

175 Path to the repository root this `Registry` will manage. 

176 

177 Returns 

178 ------- 

179 registry : `Registry` 

180 A new `Registry` instance. 

181 

182 Notes 

183 ----- 

184 This class will determine the concrete `Registry` subclass to 

185 use from configuration. Each subclass should implement this method 

186 even if it can not create a registry. 

187 """ 

188 registry_cls, registry_config = cls.determineTrampoline(config) 

189 return registry_cls.createFromConfig(registry_config, dimensionConfig, butlerRoot) 

190 

191 @classmethod 

192 def fromConfig( 

193 cls, 

194 config: Union[ButlerConfig, RegistryConfig, Config, str], 

195 butlerRoot: Optional[ResourcePathExpression] = None, 

196 writeable: bool = True, 

197 defaults: Optional[RegistryDefaults] = None, 

198 ) -> Registry: 

199 """Create `Registry` subclass instance from `config`. 

200 

201 Registry database must be initialized prior to calling this method. 

202 

203 Parameters 

204 ---------- 

205 config : `ButlerConfig`, `RegistryConfig`, `Config` or `str` 

206 Registry configuration 

207 butlerRoot : `lsst.resources.ResourcePathExpression`, optional 

208 Path to the repository root this `Registry` will manage. 

209 writeable : `bool`, optional 

210 If `True` (default) create a read-write connection to the database. 

211 defaults : `RegistryDefaults`, optional 

212 Default collection search path and/or output `~CollectionType.RUN` 

213 collection. 

214 

215 Returns 

216 ------- 

217 registry : `Registry` (subclass) 

218 A new `Registry` subclass instance. 

219 

220 Notes 

221 ----- 

222 This class will determine the concrete `Registry` subclass to 

223 use from configuration. Each subclass should implement this method. 

224 """ 

225 # The base class implementation should trampoline to the correct 

226 # subclass. No implementation should ever use this implementation 

227 # directly. If no class is specified, default to the standard 

228 # registry. 

229 registry_cls, registry_config = cls.determineTrampoline(config) 

230 return registry_cls.fromConfig(config, butlerRoot, writeable, defaults) 

231 

232 @abstractmethod 

233 def isWriteable(self) -> bool: 

234 """Return `True` if this registry allows write operations, and `False` 

235 otherwise. 

236 """ 

237 raise NotImplementedError() 

238 

239 @abstractmethod 

240 def copy(self, defaults: Optional[RegistryDefaults] = None) -> Registry: 

241 """Create a new `Registry` backed by the same data repository and 

242 connection as this one, but independent defaults. 

243 

244 Parameters 

245 ---------- 

246 defaults : `RegistryDefaults`, optional 

247 Default collections and data ID values for the new registry. If 

248 not provided, ``self.defaults`` will be used (but future changes 

249 to either registry's defaults will not affect the other). 

250 

251 Returns 

252 ------- 

253 copy : `Registry` 

254 A new `Registry` instance with its own defaults. 

255 

256 Notes 

257 ----- 

258 Because the new registry shares a connection with the original, they 

259 also share transaction state (despite the fact that their `transaction` 

260 context manager methods do not reflect this), and must be used with 

261 care. 

262 """ 

263 raise NotImplementedError() 

264 

265 @property 

266 @abstractmethod 

267 def dimensions(self) -> DimensionUniverse: 

268 """Definitions of all dimensions recognized by this `Registry` 

269 (`DimensionUniverse`). 

270 """ 

271 raise NotImplementedError() 

272 

273 @property 

274 def defaults(self) -> RegistryDefaults: 

275 """Default collection search path and/or output `~CollectionType.RUN` 

276 collection (`RegistryDefaults`). 

277 

278 This is an immutable struct whose components may not be set 

279 individually, but the entire struct can be set by assigning to this 

280 property. 

281 """ 

282 return self._defaults 

283 

284 @defaults.setter 

285 def defaults(self, value: RegistryDefaults) -> None: 

286 if value.run is not None: 

287 self.registerRun(value.run) 

288 value.finish(self) 

289 self._defaults = value 

290 

291 @abstractmethod 

292 def refresh(self) -> None: 

293 """Refresh all in-memory state by querying the database. 

294 

295 This may be necessary to enable querying for entities added by other 

296 registry instances after this one was constructed. 

297 """ 

298 raise NotImplementedError() 

299 

300 @contextlib.contextmanager 

301 @abstractmethod 

302 def transaction(self, *, savepoint: bool = False) -> Iterator[None]: 

303 """Return a context manager that represents a transaction.""" 

304 raise NotImplementedError() 

305 

306 def resetConnectionPool(self) -> None: 

307 """Reset connection pool for registry if relevant. 

308 

309 This operation can be used reset connections to servers when 

310 using registry with fork-based multiprocessing. This method should 

311 usually be called by the child process immediately 

312 after the fork. 

313 

314 The base class implementation is a no-op. 

315 """ 

316 pass 

317 

318 @abstractmethod 

319 def registerCollection( 

320 self, name: str, type: CollectionType = CollectionType.TAGGED, doc: Optional[str] = None 

321 ) -> bool: 

322 """Add a new collection if one with the given name does not exist. 

323 

324 Parameters 

325 ---------- 

326 name : `str` 

327 The name of the collection to create. 

328 type : `CollectionType` 

329 Enum value indicating the type of collection to create. 

330 doc : `str`, optional 

331 Documentation string for the collection. 

332 

333 Returns 

334 ------- 

335 registered : `bool` 

336 Boolean indicating whether the collection was already registered 

337 or was created by this call. 

338 

339 Notes 

340 ----- 

341 This method cannot be called within transactions, as it needs to be 

342 able to perform its own transaction to be concurrent. 

343 """ 

344 raise NotImplementedError() 

345 

346 @abstractmethod 

347 def getCollectionType(self, name: str) -> CollectionType: 

348 """Return an enumeration value indicating the type of the given 

349 collection. 

350 

351 Parameters 

352 ---------- 

353 name : `str` 

354 The name of the collection. 

355 

356 Returns 

357 ------- 

358 type : `CollectionType` 

359 Enum value indicating the type of this collection. 

360 

361 Raises 

362 ------ 

363 MissingCollectionError 

364 Raised if no collection with the given name exists. 

365 """ 

366 raise NotImplementedError() 

367 

368 @abstractmethod 

369 def _get_collection_record(self, name: str) -> CollectionRecord: 

370 """Return the record for this collection. 

371 

372 Parameters 

373 ---------- 

374 name : `str` 

375 Name of the collection for which the record is to be retrieved. 

376 

377 Returns 

378 ------- 

379 record : `CollectionRecord` 

380 The record for this collection. 

381 """ 

382 raise NotImplementedError() 

383 

384 @abstractmethod 

385 def registerRun(self, name: str, doc: Optional[str] = None) -> bool: 

386 """Add a new run if one with the given name does not exist. 

387 

388 Parameters 

389 ---------- 

390 name : `str` 

391 The name of the run to create. 

392 doc : `str`, optional 

393 Documentation string for the collection. 

394 

395 Returns 

396 ------- 

397 registered : `bool` 

398 Boolean indicating whether a new run was registered. `False` 

399 if it already existed. 

400 

401 Notes 

402 ----- 

403 This method cannot be called within transactions, as it needs to be 

404 able to perform its own transaction to be concurrent. 

405 """ 

406 raise NotImplementedError() 

407 

408 @abstractmethod 

409 def removeCollection(self, name: str) -> None: 

410 """Remove the given collection from the registry. 

411 

412 Parameters 

413 ---------- 

414 name : `str` 

415 The name of the collection to remove. 

416 

417 Raises 

418 ------ 

419 MissingCollectionError 

420 Raised if no collection with the given name exists. 

421 sqlalchemy.IntegrityError 

422 Raised if the database rows associated with the collection are 

423 still referenced by some other table, such as a dataset in a 

424 datastore (for `~CollectionType.RUN` collections only) or a 

425 `~CollectionType.CHAINED` collection of which this collection is 

426 a child. 

427 

428 Notes 

429 ----- 

430 If this is a `~CollectionType.RUN` collection, all datasets and quanta 

431 in it will removed from the `Registry` database. This requires that 

432 those datasets be removed (or at least trashed) from any datastores 

433 that hold them first. 

434 

435 A collection may not be deleted as long as it is referenced by a 

436 `~CollectionType.CHAINED` collection; the ``CHAINED`` collection must 

437 be deleted or redefined first. 

438 """ 

439 raise NotImplementedError() 

440 

441 @abstractmethod 

442 def getCollectionChain(self, parent: str) -> Sequence[str]: 

443 """Return the child collections in a `~CollectionType.CHAINED` 

444 collection. 

445 

446 Parameters 

447 ---------- 

448 parent : `str` 

449 Name of the chained collection. Must have already been added via 

450 a call to `Registry.registerCollection`. 

451 

452 Returns 

453 ------- 

454 children : `Sequence` [ `str` ] 

455 An ordered sequence of collection names that are searched when the 

456 given chained collection is searched. 

457 

458 Raises 

459 ------ 

460 MissingCollectionError 

461 Raised if ``parent`` does not exist in the `Registry`. 

462 CollectionTypeError 

463 Raised if ``parent`` does not correspond to a 

464 `~CollectionType.CHAINED` collection. 

465 """ 

466 raise NotImplementedError() 

467 

468 @abstractmethod 

469 def setCollectionChain(self, parent: str, children: Any, *, flatten: bool = False) -> None: 

470 """Define or redefine a `~CollectionType.CHAINED` collection. 

471 

472 Parameters 

473 ---------- 

474 parent : `str` 

475 Name of the chained collection. Must have already been added via 

476 a call to `Registry.registerCollection`. 

477 children : `Any` 

478 An expression defining an ordered search of child collections, 

479 generally an iterable of `str`; see 

480 :ref:`daf_butler_collection_expressions` for more information. 

481 flatten : `bool`, optional 

482 If `True` (`False` is default), recursively flatten out any nested 

483 `~CollectionType.CHAINED` collections in ``children`` first. 

484 

485 Raises 

486 ------ 

487 MissingCollectionError 

488 Raised when any of the given collections do not exist in the 

489 `Registry`. 

490 CollectionTypeError 

491 Raised if ``parent`` does not correspond to a 

492 `~CollectionType.CHAINED` collection. 

493 ValueError 

494 Raised if the given collections contains a cycle. 

495 """ 

496 raise NotImplementedError() 

497 

498 @abstractmethod 

499 def getCollectionParentChains(self, collection: str) -> Set[str]: 

500 """Return the CHAINED collections that directly contain the given one. 

501 

502 Parameters 

503 ---------- 

504 name : `str` 

505 Name of the collection. 

506 

507 Returns 

508 ------- 

509 chains : `set` of `str` 

510 Set of `~CollectionType.CHAINED` collection names. 

511 """ 

512 raise NotImplementedError() 

513 

514 @abstractmethod 

515 def getCollectionDocumentation(self, collection: str) -> Optional[str]: 

516 """Retrieve the documentation string for a collection. 

517 

518 Parameters 

519 ---------- 

520 name : `str` 

521 Name of the collection. 

522 

523 Returns 

524 ------- 

525 docs : `str` or `None` 

526 Docstring for the collection with the given name. 

527 """ 

528 raise NotImplementedError() 

529 

530 @abstractmethod 

531 def setCollectionDocumentation(self, collection: str, doc: Optional[str]) -> None: 

532 """Set the documentation string for a collection. 

533 

534 Parameters 

535 ---------- 

536 name : `str` 

537 Name of the collection. 

538 docs : `str` or `None` 

539 Docstring for the collection with the given name; will replace any 

540 existing docstring. Passing `None` will remove any existing 

541 docstring. 

542 """ 

543 raise NotImplementedError() 

544 

545 @abstractmethod 

546 def getCollectionSummary(self, collection: str) -> CollectionSummary: 

547 """Return a summary for the given collection. 

548 

549 Parameters 

550 ---------- 

551 collection : `str` 

552 Name of the collection for which a summary is to be retrieved. 

553 

554 Returns 

555 ------- 

556 summary : `CollectionSummary` 

557 Summary of the dataset types and governor dimension values in 

558 this collection. 

559 """ 

560 raise NotImplementedError() 

561 

562 @abstractmethod 

563 def registerDatasetType(self, datasetType: DatasetType) -> bool: 

564 """ 

565 Add a new `DatasetType` to the Registry. 

566 

567 It is not an error to register the same `DatasetType` twice. 

568 

569 Parameters 

570 ---------- 

571 datasetType : `DatasetType` 

572 The `DatasetType` to be added. 

573 

574 Returns 

575 ------- 

576 inserted : `bool` 

577 `True` if ``datasetType`` was inserted, `False` if an identical 

578 existing `DatsetType` was found. Note that in either case the 

579 DatasetType is guaranteed to be defined in the Registry 

580 consistently with the given definition. 

581 

582 Raises 

583 ------ 

584 ValueError 

585 Raised if the dimensions or storage class are invalid. 

586 ConflictingDefinitionError 

587 Raised if this DatasetType is already registered with a different 

588 definition. 

589 

590 Notes 

591 ----- 

592 This method cannot be called within transactions, as it needs to be 

593 able to perform its own transaction to be concurrent. 

594 """ 

595 raise NotImplementedError() 

596 

597 @abstractmethod 

598 def removeDatasetType(self, name: str) -> None: 

599 """Remove the named `DatasetType` from the registry. 

600 

601 .. warning:: 

602 

603 Registry implementations can cache the dataset type definitions. 

604 This means that deleting the dataset type definition may result in 

605 unexpected behavior from other butler processes that are active 

606 that have not seen the deletion. 

607 

608 Parameters 

609 ---------- 

610 name : `str` 

611 Name of the type to be removed. 

612 

613 Raises 

614 ------ 

615 lsst.daf.butler.registry.OrphanedRecordError 

616 Raised if an attempt is made to remove the dataset type definition 

617 when there are already datasets associated with it. 

618 

619 Notes 

620 ----- 

621 If the dataset type is not registered the method will return without 

622 action. 

623 """ 

624 raise NotImplementedError() 

625 

626 @abstractmethod 

627 def getDatasetType(self, name: str) -> DatasetType: 

628 """Get the `DatasetType`. 

629 

630 Parameters 

631 ---------- 

632 name : `str` 

633 Name of the type. 

634 

635 Returns 

636 ------- 

637 type : `DatasetType` 

638 The `DatasetType` associated with the given name. 

639 

640 Raises 

641 ------ 

642 MissingDatasetTypeError 

643 Raised if the requested dataset type has not been registered. 

644 

645 Notes 

646 ----- 

647 This method handles component dataset types automatically, though most 

648 other registry operations do not. 

649 """ 

650 raise NotImplementedError() 

651 

652 @abstractmethod 

653 def supportsIdGenerationMode(self, mode: DatasetIdGenEnum) -> bool: 

654 """Test whether the given dataset ID generation mode is supported by 

655 `insertDatasets`. 

656 

657 Parameters 

658 ---------- 

659 mode : `DatasetIdGenEnum` 

660 Enum value for the mode to test. 

661 

662 Returns 

663 ------- 

664 supported : `bool` 

665 Whether the given mode is supported. 

666 """ 

667 raise NotImplementedError() 

668 

669 @abstractmethod 

670 def findDataset( 

671 self, 

672 datasetType: Union[DatasetType, str], 

673 dataId: Optional[DataId] = None, 

674 *, 

675 collections: Any = None, 

676 timespan: Optional[Timespan] = None, 

677 **kwargs: Any, 

678 ) -> Optional[DatasetRef]: 

679 """Find a dataset given its `DatasetType` and data ID. 

680 

681 This can be used to obtain a `DatasetRef` that permits the dataset to 

682 be read from a `Datastore`. If the dataset is a component and can not 

683 be found using the provided dataset type, a dataset ref for the parent 

684 will be returned instead but with the correct dataset type. 

685 

686 Parameters 

687 ---------- 

688 datasetType : `DatasetType` or `str` 

689 A `DatasetType` or the name of one. If this is a `DatasetType` 

690 instance, its storage class will be respected and propagated to 

691 the output, even if it differs from the dataset type definition 

692 in the registry, as long as the storage classes are convertible. 

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

694 A `dict`-like object containing the `Dimension` links that identify 

695 the dataset within a collection. 

696 collections, optional. 

697 An expression that fully or partially identifies the collections to 

698 search for the dataset; see 

699 :ref:`daf_butler_collection_expressions` for more information. 

700 Defaults to ``self.defaults.collections``. 

701 timespan : `Timespan`, optional 

702 A timespan that the validity range of the dataset must overlap. 

703 If not provided, any `~CollectionType.CALIBRATION` collections 

704 matched by the ``collections`` argument will not be searched. 

705 **kwargs 

706 Additional keyword arguments passed to 

707 `DataCoordinate.standardize` to convert ``dataId`` to a true 

708 `DataCoordinate` or augment an existing one. 

709 

710 Returns 

711 ------- 

712 ref : `DatasetRef` 

713 A reference to the dataset, or `None` if no matching Dataset 

714 was found. 

715 

716 Raises 

717 ------ 

718 NoDefaultCollectionError 

719 Raised if ``collections`` is `None` and 

720 ``self.defaults.collections`` is `None`. 

721 LookupError 

722 Raised if one or more data ID keys are missing. 

723 MissingDatasetTypeError 

724 Raised if the dataset type does not exist. 

725 MissingCollectionError 

726 Raised if any of ``collections`` does not exist in the registry. 

727 

728 Notes 

729 ----- 

730 This method simply returns `None` and does not raise an exception even 

731 when the set of collections searched is intrinsically incompatible with 

732 the dataset type, e.g. if ``datasetType.isCalibration() is False``, but 

733 only `~CollectionType.CALIBRATION` collections are being searched. 

734 This may make it harder to debug some lookup failures, but the behavior 

735 is intentional; we consider it more important that failed searches are 

736 reported consistently, regardless of the reason, and that adding 

737 additional collections that do not contain a match to the search path 

738 never changes the behavior. 

739 

740 This method handles component dataset types automatically, though most 

741 other registry operations do not. 

742 """ 

743 raise NotImplementedError() 

744 

745 @abstractmethod 

746 def insertDatasets( 

747 self, 

748 datasetType: Union[DatasetType, str], 

749 dataIds: Iterable[DataId], 

750 run: Optional[str] = None, 

751 expand: bool = True, 

752 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

753 ) -> List[DatasetRef]: 

754 """Insert one or more datasets into the `Registry` 

755 

756 This always adds new datasets; to associate existing datasets with 

757 a new collection, use ``associate``. 

758 

759 Parameters 

760 ---------- 

761 datasetType : `DatasetType` or `str` 

762 A `DatasetType` or the name of one. 

763 dataIds : `~collections.abc.Iterable` of `dict` or `DataCoordinate` 

764 Dimension-based identifiers for the new datasets. 

765 run : `str`, optional 

766 The name of the run that produced the datasets. Defaults to 

767 ``self.defaults.run``. 

768 expand : `bool`, optional 

769 If `True` (default), expand data IDs as they are inserted. This is 

770 necessary in general to allow datastore to generate file templates, 

771 but it may be disabled if the caller can guarantee this is 

772 unnecessary. 

773 idGenerationMode : `DatasetIdGenEnum`, optional 

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

775 are generated for each inserted dataset. 

776 

777 Returns 

778 ------- 

779 refs : `list` of `DatasetRef` 

780 Resolved `DatasetRef` instances for all given data IDs (in the same 

781 order). 

782 

783 Raises 

784 ------ 

785 DatasetTypeError 

786 Raised if ``datasetType`` is not known to registry. 

787 CollectionTypeError 

788 Raised if ``run`` collection type is not `~CollectionType.RUN`. 

789 NoDefaultCollectionError 

790 Raised if ``run`` is `None` and ``self.defaults.run`` is `None`. 

791 ConflictingDefinitionError 

792 If a dataset with the same dataset type and data ID as one of those 

793 given already exists in ``run``. 

794 MissingCollectionError 

795 Raised if ``run`` does not exist in the registry. 

796 """ 

797 raise NotImplementedError() 

798 

799 @abstractmethod 

800 def _importDatasets( 

801 self, 

802 datasets: Iterable[DatasetRef], 

803 expand: bool = True, 

804 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

805 reuseIds: bool = False, 

806 ) -> List[DatasetRef]: 

807 """Import one or more datasets into the `Registry`. 

808 

809 Difference from `insertDatasets` method is that this method accepts 

810 `DatasetRef` instances which should already be resolved and have a 

811 dataset ID. If registry supports globally-unique dataset IDs (e.g. 

812 `uuid.UUID`) then datasets which already exist in the registry will be 

813 ignored if imported again. 

814 

815 Parameters 

816 ---------- 

817 datasets : `~collections.abc.Iterable` of `DatasetRef` 

818 Datasets to be inserted. All `DatasetRef` instances must have 

819 identical ``datasetType`` and ``run`` attributes. ``run`` 

820 attribute can be `None` and defaults to ``self.defaults.run``. 

821 Datasets can specify ``id`` attribute which will be used for 

822 inserted datasets. All dataset IDs must have the same type 

823 (`int` or `uuid.UUID`), if type of dataset IDs does not match 

824 configured backend then IDs will be ignored and new IDs will be 

825 generated by backend. 

826 expand : `bool`, optional 

827 If `True` (default), expand data IDs as they are inserted. This is 

828 necessary in general to allow datastore to generate file templates, 

829 but it may be disabled if the caller can guarantee this is 

830 unnecessary. 

831 idGenerationMode : `DatasetIdGenEnum`, optional 

832 Specifies option for generating dataset IDs when IDs are not 

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

834 unique IDs are generated for each inserted dataset. 

835 reuseIds : `bool`, optional 

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

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

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

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

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

841 

842 Returns 

843 ------- 

844 refs : `list` of `DatasetRef` 

845 Resolved `DatasetRef` instances for all given data IDs (in the same 

846 order). If any of ``datasets`` has an ID which already exists in 

847 the database then it will not be inserted or updated, but a 

848 resolved `DatasetRef` will be returned for it in any case. 

849 

850 Raises 

851 ------ 

852 NoDefaultCollectionError 

853 Raised if ``run`` is `None` and ``self.defaults.run`` is `None`. 

854 DatasetTypeError 

855 Raised if datasets correspond to more than one dataset type or 

856 dataset type is not known to registry. 

857 ConflictingDefinitionError 

858 If a dataset with the same dataset type and data ID as one of those 

859 given already exists in ``run``. 

860 MissingCollectionError 

861 Raised if ``run`` does not exist in the registry. 

862 

863 Notes 

864 ----- 

865 This method is considered package-private and internal to Butler 

866 implementation. Clients outside daf_butler package should not use this 

867 method. 

868 """ 

869 raise NotImplementedError() 

870 

871 @abstractmethod 

872 def getDataset(self, id: DatasetId) -> Optional[DatasetRef]: 

873 """Retrieve a Dataset entry. 

874 

875 Parameters 

876 ---------- 

877 id : `DatasetId` 

878 The unique identifier for the dataset. 

879 

880 Returns 

881 ------- 

882 ref : `DatasetRef` or `None` 

883 A ref to the Dataset, or `None` if no matching Dataset 

884 was found. 

885 """ 

886 raise NotImplementedError() 

887 

888 @abstractmethod 

889 def removeDatasets(self, refs: Iterable[DatasetRef]) -> None: 

890 """Remove datasets from the Registry. 

891 

892 The datasets will be removed unconditionally from all collections, and 

893 any `Quantum` that consumed this dataset will instead be marked with 

894 having a NULL input. `Datastore` records will *not* be deleted; the 

895 caller is responsible for ensuring that the dataset has already been 

896 removed from all Datastores. 

897 

898 Parameters 

899 ---------- 

900 refs : `Iterable` of `DatasetRef` 

901 References to the datasets to be removed. Must include a valid 

902 ``id`` attribute, and should be considered invalidated upon return. 

903 

904 Raises 

905 ------ 

906 AmbiguousDatasetError 

907 Raised if any ``ref.id`` is `None`. 

908 OrphanedRecordError 

909 Raised if any dataset is still present in any `Datastore`. 

910 """ 

911 raise NotImplementedError() 

912 

913 @abstractmethod 

914 def associate(self, collection: str, refs: Iterable[DatasetRef]) -> None: 

915 """Add existing datasets to a `~CollectionType.TAGGED` collection. 

916 

917 If a DatasetRef with the same exact ID is already in a collection 

918 nothing is changed. If a `DatasetRef` with the same `DatasetType` and 

919 data ID but with different ID exists in the collection, 

920 `ConflictingDefinitionError` is raised. 

921 

922 Parameters 

923 ---------- 

924 collection : `str` 

925 Indicates the collection the datasets should be associated with. 

926 refs : `Iterable` [ `DatasetRef` ] 

927 An iterable of resolved `DatasetRef` instances that already exist 

928 in this `Registry`. 

929 

930 Raises 

931 ------ 

932 ConflictingDefinitionError 

933 If a Dataset with the given `DatasetRef` already exists in the 

934 given collection. 

935 AmbiguousDatasetError 

936 Raised if ``any(ref.id is None for ref in refs)``. 

937 MissingCollectionError 

938 Raised if ``collection`` does not exist in the registry. 

939 CollectionTypeError 

940 Raise adding new datasets to the given ``collection`` is not 

941 allowed. 

942 """ 

943 raise NotImplementedError() 

944 

945 @abstractmethod 

946 def disassociate(self, collection: str, refs: Iterable[DatasetRef]) -> None: 

947 """Remove existing datasets from a `~CollectionType.TAGGED` collection. 

948 

949 ``collection`` and ``ref`` combinations that are not currently 

950 associated are silently ignored. 

951 

952 Parameters 

953 ---------- 

954 collection : `str` 

955 The collection the datasets should no longer be associated with. 

956 refs : `Iterable` [ `DatasetRef` ] 

957 An iterable of resolved `DatasetRef` instances that already exist 

958 in this `Registry`. 

959 

960 Raises 

961 ------ 

962 AmbiguousDatasetError 

963 Raised if any of the given dataset references is unresolved. 

964 MissingCollectionError 

965 Raised if ``collection`` does not exist in the registry. 

966 CollectionTypeError 

967 Raise adding new datasets to the given ``collection`` is not 

968 allowed. 

969 """ 

970 raise NotImplementedError() 

971 

972 @abstractmethod 

973 def certify(self, collection: str, refs: Iterable[DatasetRef], timespan: Timespan) -> None: 

974 """Associate one or more datasets with a calibration collection and a 

975 validity range within it. 

976 

977 Parameters 

978 ---------- 

979 collection : `str` 

980 The name of an already-registered `~CollectionType.CALIBRATION` 

981 collection. 

982 refs : `Iterable` [ `DatasetRef` ] 

983 Datasets to be associated. 

984 timespan : `Timespan` 

985 The validity range for these datasets within the collection. 

986 

987 Raises 

988 ------ 

989 AmbiguousDatasetError 

990 Raised if any of the given `DatasetRef` instances is unresolved. 

991 ConflictingDefinitionError 

992 Raised if the collection already contains a different dataset with 

993 the same `DatasetType` and data ID and an overlapping validity 

994 range. 

995 CollectionTypeError 

996 Raised if ``collection`` is not a `~CollectionType.CALIBRATION` 

997 collection or if one or more datasets are of a dataset type for 

998 which `DatasetType.isCalibration` returns `False`. 

999 """ 

1000 raise NotImplementedError() 

1001 

1002 @abstractmethod 

1003 def decertify( 

1004 self, 

1005 collection: str, 

1006 datasetType: Union[str, DatasetType], 

1007 timespan: Timespan, 

1008 *, 

1009 dataIds: Optional[Iterable[DataId]] = None, 

1010 ) -> None: 

1011 """Remove or adjust datasets to clear a validity range within a 

1012 calibration collection. 

1013 

1014 Parameters 

1015 ---------- 

1016 collection : `str` 

1017 The name of an already-registered `~CollectionType.CALIBRATION` 

1018 collection. 

1019 datasetType : `str` or `DatasetType` 

1020 Name or `DatasetType` instance for the datasets to be decertified. 

1021 timespan : `Timespan`, optional 

1022 The validity range to remove datasets from within the collection. 

1023 Datasets that overlap this range but are not contained by it will 

1024 have their validity ranges adjusted to not overlap it, which may 

1025 split a single dataset validity range into two. 

1026 dataIds : `Iterable` [ `DataId` ], optional 

1027 Data IDs that should be decertified within the given validity range 

1028 If `None`, all data IDs for ``self.datasetType`` will be 

1029 decertified. 

1030 

1031 Raises 

1032 ------ 

1033 CollectionTypeError 

1034 Raised if ``collection`` is not a `~CollectionType.CALIBRATION` 

1035 collection or if ``datasetType.isCalibration() is False``. 

1036 """ 

1037 raise NotImplementedError() 

1038 

1039 @abstractmethod 

1040 def getDatastoreBridgeManager(self) -> DatastoreRegistryBridgeManager: 

1041 """Return an object that allows a new `Datastore` instance to 

1042 communicate with this `Registry`. 

1043 

1044 Returns 

1045 ------- 

1046 manager : `DatastoreRegistryBridgeManager` 

1047 Object that mediates communication between this `Registry` and its 

1048 associated datastores. 

1049 """ 

1050 raise NotImplementedError() 

1051 

1052 @abstractmethod 

1053 def getDatasetLocations(self, ref: DatasetRef) -> Iterable[str]: 

1054 """Retrieve datastore locations for a given dataset. 

1055 

1056 Parameters 

1057 ---------- 

1058 ref : `DatasetRef` 

1059 A reference to the dataset for which to retrieve storage 

1060 information. 

1061 

1062 Returns 

1063 ------- 

1064 datastores : `Iterable` [ `str` ] 

1065 All the matching datastores holding this dataset. 

1066 

1067 Raises 

1068 ------ 

1069 AmbiguousDatasetError 

1070 Raised if ``ref.id`` is `None`. 

1071 """ 

1072 raise NotImplementedError() 

1073 

1074 @abstractmethod 

1075 def expandDataId( 

1076 self, 

1077 dataId: Optional[DataId] = None, 

1078 *, 

1079 graph: Optional[DimensionGraph] = None, 

1080 records: Optional[NameLookupMapping[DimensionElement, Optional[DimensionRecord]]] = None, 

1081 withDefaults: bool = True, 

1082 **kwargs: Any, 

1083 ) -> DataCoordinate: 

1084 """Expand a dimension-based data ID to include additional information. 

1085 

1086 Parameters 

1087 ---------- 

1088 dataId : `DataCoordinate` or `dict`, optional 

1089 Data ID to be expanded; augmented and overridden by ``kwargs``. 

1090 graph : `DimensionGraph`, optional 

1091 Set of dimensions for the expanded ID. If `None`, the dimensions 

1092 will be inferred from the keys of ``dataId`` and ``kwargs``. 

1093 Dimensions that are in ``dataId`` or ``kwargs`` but not in 

1094 ``graph`` are silently ignored, providing a way to extract and 

1095 ``graph`` expand a subset of a data ID. 

1096 records : `Mapping` [`str`, `DimensionRecord`], optional 

1097 Dimension record data to use before querying the database for that 

1098 data, keyed by element name. 

1099 withDefaults : `bool`, optional 

1100 Utilize ``self.defaults.dataId`` to fill in missing governor 

1101 dimension key-value pairs. Defaults to `True` (i.e. defaults are 

1102 used). 

1103 **kwargs 

1104 Additional keywords are treated like additional key-value pairs for 

1105 ``dataId``, extending and overriding 

1106 

1107 Returns 

1108 ------- 

1109 expanded : `DataCoordinate` 

1110 A data ID that includes full metadata for all of the dimensions it 

1111 identifies, i.e. guarantees that ``expanded.hasRecords()`` and 

1112 ``expanded.hasFull()`` both return `True`. 

1113 

1114 Raises 

1115 ------ 

1116 DataIdError 

1117 Raised when ``dataId`` or keyword arguments specify unknown 

1118 dimensions or values, or when a resulting data ID contains 

1119 contradictory key-value pairs, according to dimension 

1120 relationships. 

1121 

1122 Notes 

1123 ----- 

1124 This method cannot be relied upon to reject invalid data ID values 

1125 for dimensions that do actually not have any record columns. For 

1126 efficiency reasons the records for these dimensions (which have only 

1127 dimension key values that are given by the caller) may be constructed 

1128 directly rather than obtained from the registry database. 

1129 """ 

1130 raise NotImplementedError() 

1131 

1132 @abstractmethod 

1133 def insertDimensionData( 

1134 self, 

1135 element: Union[DimensionElement, str], 

1136 *data: Union[Mapping[str, Any], DimensionRecord], 

1137 conform: bool = True, 

1138 replace: bool = False, 

1139 skip_existing: bool = False, 

1140 ) -> None: 

1141 """Insert one or more dimension records into the database. 

1142 

1143 Parameters 

1144 ---------- 

1145 element : `DimensionElement` or `str` 

1146 The `DimensionElement` or name thereof that identifies the table 

1147 records will be inserted into. 

1148 data : `dict` or `DimensionRecord` (variadic) 

1149 One or more records to insert. 

1150 conform : `bool`, optional 

1151 If `False` (`True` is default) perform no checking or conversions, 

1152 and assume that ``element`` is a `DimensionElement` instance and 

1153 ``data`` is a one or more `DimensionRecord` instances of the 

1154 appropriate subclass. 

1155 replace : `bool`, optional 

1156 If `True` (`False` is default), replace existing records in the 

1157 database if there is a conflict. 

1158 skip_existing : `bool`, optional 

1159 If `True` (`False` is default), skip insertion if a record with 

1160 the same primary key values already exists. Unlike 

1161 `syncDimensionData`, this will not detect when the given record 

1162 differs from what is in the database, and should not be used when 

1163 this is a concern. 

1164 """ 

1165 raise NotImplementedError() 

1166 

1167 @abstractmethod 

1168 def syncDimensionData( 

1169 self, 

1170 element: Union[DimensionElement, str], 

1171 row: Union[Mapping[str, Any], DimensionRecord], 

1172 conform: bool = True, 

1173 update: bool = False, 

1174 ) -> Union[bool, Dict[str, Any]]: 

1175 """Synchronize the given dimension record with the database, inserting 

1176 if it does not already exist and comparing values if it does. 

1177 

1178 Parameters 

1179 ---------- 

1180 element : `DimensionElement` or `str` 

1181 The `DimensionElement` or name thereof that identifies the table 

1182 records will be inserted into. 

1183 row : `dict` or `DimensionRecord` 

1184 The record to insert. 

1185 conform : `bool`, optional 

1186 If `False` (`True` is default) perform no checking or conversions, 

1187 and assume that ``element`` is a `DimensionElement` instance and 

1188 ``data`` is a one or more `DimensionRecord` instances of the 

1189 appropriate subclass. 

1190 update: `bool`, optional 

1191 If `True` (`False` is default), update the existing record in the 

1192 database if there is a conflict. 

1193 

1194 Returns 

1195 ------- 

1196 inserted_or_updated : `bool` or `dict` 

1197 `True` if a new row was inserted, `False` if no changes were 

1198 needed, or a `dict` mapping updated column names to their old 

1199 values if an update was performed (only possible if 

1200 ``update=True``). 

1201 

1202 Raises 

1203 ------ 

1204 ConflictingDefinitionError 

1205 Raised if the record exists in the database (according to primary 

1206 key lookup) but is inconsistent with the given one. 

1207 """ 

1208 raise NotImplementedError() 

1209 

1210 @abstractmethod 

1211 def queryDatasetTypes( 

1212 self, 

1213 expression: Any = ..., 

1214 *, 

1215 components: Optional[bool] = None, 

1216 missing: Optional[List[str]] = None, 

1217 ) -> Iterable[DatasetType]: 

1218 """Iterate over the dataset types whose names match an expression. 

1219 

1220 Parameters 

1221 ---------- 

1222 expression : `Any`, optional 

1223 An expression that fully or partially identifies the dataset types 

1224 to return, such as a `str`, `re.Pattern`, or iterable thereof. 

1225 ``...`` can be used to return all dataset types, and is the 

1226 default. See :ref:`daf_butler_dataset_type_expressions` for more 

1227 information. 

1228 components : `bool`, optional 

1229 If `True`, apply all expression patterns to component dataset type 

1230 names as well. If `False`, never apply patterns to components. 

1231 If `None` (default), apply patterns to components only if their 

1232 parent datasets were not matched by the expression. 

1233 Fully-specified component datasets (`str` or `DatasetType` 

1234 instances) are always included. 

1235 

1236 Values other than `False` are deprecated, and only `False` will be 

1237 supported after v26. After v27 this argument will be removed 

1238 entirely. 

1239 missing : `list` of `str`, optional 

1240 String dataset type names that were explicitly given (i.e. not 

1241 regular expression patterns) but not found will be appended to this 

1242 list, if it is provided. 

1243 

1244 Returns 

1245 ------- 

1246 dataset_types : `Iterable` [ `DatasetType`] 

1247 An `Iterable` of `DatasetType` instances whose names match 

1248 ``expression``. 

1249 

1250 Raises 

1251 ------ 

1252 DatasetTypeExpressionError 

1253 Raised when ``expression`` is invalid. 

1254 """ 

1255 raise NotImplementedError() 

1256 

1257 @abstractmethod 

1258 def queryCollections( 

1259 self, 

1260 expression: Any = ..., 

1261 datasetType: Optional[DatasetType] = None, 

1262 collectionTypes: Union[Iterable[CollectionType], CollectionType] = CollectionType.all(), 

1263 flattenChains: bool = False, 

1264 includeChains: Optional[bool] = None, 

1265 ) -> Sequence[str]: 

1266 """Iterate over the collections whose names match an expression. 

1267 

1268 Parameters 

1269 ---------- 

1270 expression : `Any`, optional 

1271 An expression that identifies the collections to return, such as 

1272 a `str` (for full matches or partial matches via globs), 

1273 `re.Pattern` (for partial matches), or iterable thereof. ``...`` 

1274 can be used to return all collections, and is the default. 

1275 See :ref:`daf_butler_collection_expressions` for more information. 

1276 datasetType : `DatasetType`, optional 

1277 If provided, only yield collections that may contain datasets of 

1278 this type. This is a conservative approximation in general; it may 

1279 yield collections that do not have any such datasets. 

1280 collectionTypes : `AbstractSet` [ `CollectionType` ] or \ 

1281 `CollectionType`, optional 

1282 If provided, only yield collections of these types. 

1283 flattenChains : `bool`, optional 

1284 If `True` (`False` is default), recursively yield the child 

1285 collections of matching `~CollectionType.CHAINED` collections. 

1286 includeChains : `bool`, optional 

1287 If `True`, yield records for matching `~CollectionType.CHAINED` 

1288 collections. Default is the opposite of ``flattenChains``: include 

1289 either CHAINED collections or their children, but not both. 

1290 

1291 Returns 

1292 ------- 

1293 collections : `Sequence` [ `str` ] 

1294 The names of collections that match ``expression``. 

1295 

1296 Raises 

1297 ------ 

1298 CollectionExpressionError 

1299 Raised when ``expression`` is invalid. 

1300 

1301 Notes 

1302 ----- 

1303 The order in which collections are returned is unspecified, except that 

1304 the children of a `~CollectionType.CHAINED` collection are guaranteed 

1305 to be in the order in which they are searched. When multiple parent 

1306 `~CollectionType.CHAINED` collections match the same criteria, the 

1307 order in which the two lists appear is unspecified, and the lists of 

1308 children may be incomplete if a child has multiple parents. 

1309 """ 

1310 raise NotImplementedError() 

1311 

1312 @abstractmethod 

1313 def queryDatasets( 

1314 self, 

1315 datasetType: Any, 

1316 *, 

1317 collections: Any = None, 

1318 dimensions: Optional[Iterable[Union[Dimension, str]]] = None, 

1319 dataId: Optional[DataId] = None, 

1320 where: str = "", 

1321 findFirst: bool = False, 

1322 components: Optional[bool] = None, 

1323 bind: Optional[Mapping[str, Any]] = None, 

1324 check: bool = True, 

1325 **kwargs: Any, 

1326 ) -> DatasetQueryResults: 

1327 """Query for and iterate over dataset references matching user-provided 

1328 criteria. 

1329 

1330 Parameters 

1331 ---------- 

1332 datasetType 

1333 An expression that fully or partially identifies the dataset types 

1334 to be queried. Allowed types include `DatasetType`, `str`, 

1335 `re.Pattern`, and iterables thereof. The special value ``...`` can 

1336 be used to query all dataset types. See 

1337 :ref:`daf_butler_dataset_type_expressions` for more information. 

1338 collections: optional 

1339 An expression that identifies the collections to search, such as a 

1340 `str` (for full matches or partial matches via globs), `re.Pattern` 

1341 (for partial matches), or iterable thereof. ``...`` can be used to 

1342 search all collections (actually just all `~CollectionType.RUN` 

1343 collections, because this will still find all datasets). 

1344 If not provided, ``self.default.collections`` is used. See 

1345 :ref:`daf_butler_collection_expressions` for more information. 

1346 dimensions : `~collections.abc.Iterable` of `Dimension` or `str` 

1347 Dimensions to include in the query (in addition to those used 

1348 to identify the queried dataset type(s)), either to constrain 

1349 the resulting datasets to those for which a matching dimension 

1350 exists, or to relate the dataset type's dimensions to dimensions 

1351 referenced by the ``dataId`` or ``where`` arguments. 

1352 dataId : `dict` or `DataCoordinate`, optional 

1353 A data ID whose key-value pairs are used as equality constraints 

1354 in the query. 

1355 where : `str`, optional 

1356 A string expression similar to a SQL WHERE clause. May involve 

1357 any column of a dimension table or (as a shortcut for the primary 

1358 key column of a dimension table) dimension name. See 

1359 :ref:`daf_butler_dimension_expressions` for more information. 

1360 findFirst : `bool`, optional 

1361 If `True` (`False` is default), for each result data ID, only 

1362 yield one `DatasetRef` of each `DatasetType`, from the first 

1363 collection in which a dataset of that dataset type appears 

1364 (according to the order of ``collections`` passed in). If `True`, 

1365 ``collections`` must not contain regular expressions and may not 

1366 be ``...``. 

1367 components : `bool`, optional 

1368 If `True`, apply all dataset expression patterns to component 

1369 dataset type names as well. If `False`, never apply patterns to 

1370 components. If `None` (default), apply patterns to components only 

1371 if their parent datasets were not matched by the expression. 

1372 Fully-specified component datasets (`str` or `DatasetType` 

1373 instances) are always included. 

1374 

1375 Values other than `False` are deprecated, and only `False` will be 

1376 supported after v26. After v27 this argument will be removed 

1377 entirely. 

1378 bind : `Mapping`, optional 

1379 Mapping containing literal values that should be injected into the 

1380 ``where`` expression, keyed by the identifiers they replace. 

1381 check : `bool`, optional 

1382 If `True` (default) check the query for consistency before 

1383 executing it. This may reject some valid queries that resemble 

1384 common mistakes (e.g. queries for visits without specifying an 

1385 instrument). 

1386 **kwargs 

1387 Additional keyword arguments are forwarded to 

1388 `DataCoordinate.standardize` when processing the ``dataId`` 

1389 argument (and may be used to provide a constraining data ID even 

1390 when the ``dataId`` argument is `None`). 

1391 

1392 Returns 

1393 ------- 

1394 refs : `queries.DatasetQueryResults` 

1395 Dataset references matching the given query criteria. Nested data 

1396 IDs are guaranteed to include values for all implied dimensions 

1397 (i.e. `DataCoordinate.hasFull` will return `True`), but will not 

1398 include dimension records (`DataCoordinate.hasRecords` will be 

1399 `False`) unless `~queries.DatasetQueryResults.expanded` is called 

1400 on the result object (which returns a new one). 

1401 

1402 Raises 

1403 ------ 

1404 DatasetTypeExpressionError 

1405 Raised when ``datasetType`` expression is invalid. 

1406 TypeError 

1407 Raised when the arguments are incompatible, such as when a 

1408 collection wildcard is passed when ``findFirst`` is `True`, or 

1409 when ``collections`` is `None` and``self.defaults.collections`` is 

1410 also `None`. 

1411 DataIdError 

1412 Raised when ``dataId`` or keyword arguments specify unknown 

1413 dimensions or values, or when they contain inconsistent values. 

1414 UserExpressionError 

1415 Raised when ``where`` expression is invalid. 

1416 

1417 Notes 

1418 ----- 

1419 When multiple dataset types are queried in a single call, the 

1420 results of this operation are equivalent to querying for each dataset 

1421 type separately in turn, and no information about the relationships 

1422 between datasets of different types is included. In contexts where 

1423 that kind of information is important, the recommended pattern is to 

1424 use `queryDataIds` to first obtain data IDs (possibly with the 

1425 desired dataset types and collections passed as constraints to the 

1426 query), and then use multiple (generally much simpler) calls to 

1427 `queryDatasets` with the returned data IDs passed as constraints. 

1428 """ 

1429 raise NotImplementedError() 

1430 

1431 @abstractmethod 

1432 def queryDataIds( 

1433 self, 

1434 dimensions: Union[Iterable[Union[Dimension, str]], Dimension, str], 

1435 *, 

1436 dataId: Optional[DataId] = None, 

1437 datasets: Any = None, 

1438 collections: Any = None, 

1439 where: str = "", 

1440 components: Optional[bool] = None, 

1441 bind: Optional[Mapping[str, Any]] = None, 

1442 check: bool = True, 

1443 **kwargs: Any, 

1444 ) -> DataCoordinateQueryResults: 

1445 """Query for data IDs matching user-provided criteria. 

1446 

1447 Parameters 

1448 ---------- 

1449 dimensions : `Dimension` or `str`, or iterable thereof 

1450 The dimensions of the data IDs to yield, as either `Dimension` 

1451 instances or `str`. Will be automatically expanded to a complete 

1452 `DimensionGraph`. 

1453 dataId : `dict` or `DataCoordinate`, optional 

1454 A data ID whose key-value pairs are used as equality constraints 

1455 in the query. 

1456 datasets : `Any`, optional 

1457 An expression that fully or partially identifies dataset types 

1458 that should constrain the yielded data IDs. For example, including 

1459 "raw" here would constrain the yielded ``instrument``, 

1460 ``exposure``, ``detector``, and ``physical_filter`` values to only 

1461 those for which at least one "raw" dataset exists in 

1462 ``collections``. Allowed types include `DatasetType`, `str`, 

1463 and iterables thereof. Regular expression objects (i.e. 

1464 `re.Pattern`) are deprecated and will be removed after the v26 

1465 release. See :ref:`daf_butler_dataset_type_expressions` for more 

1466 information. 

1467 collections: `Any`, optional 

1468 An expression that identifies the collections to search for 

1469 datasets, such as a `str` (for full matches or partial matches 

1470 via globs), `re.Pattern` (for partial matches), or iterable 

1471 thereof. ``...`` can be used to search all collections (actually 

1472 just all `~CollectionType.RUN` collections, because this will 

1473 still find all datasets). If not provided, 

1474 ``self.default.collections`` is used. Ignored unless ``datasets`` 

1475 is also passed. See :ref:`daf_butler_collection_expressions` for 

1476 more information. 

1477 where : `str`, optional 

1478 A string expression similar to a SQL WHERE clause. May involve 

1479 any column of a dimension table or (as a shortcut for the primary 

1480 key column of a dimension table) dimension name. See 

1481 :ref:`daf_butler_dimension_expressions` for more information. 

1482 components : `bool`, optional 

1483 If `True`, apply all dataset expression patterns to component 

1484 dataset type names as well. If `False`, never apply patterns to 

1485 components. If `None` (default), apply patterns to components only 

1486 if their parent datasets were not matched by the expression. 

1487 Fully-specified component datasets (`str` or `DatasetType` 

1488 instances) are always included. 

1489 

1490 Values other than `False` are deprecated, and only `False` will be 

1491 supported after v26. After v27 this argument will be removed 

1492 entirely. 

1493 bind : `Mapping`, optional 

1494 Mapping containing literal values that should be injected into the 

1495 ``where`` expression, keyed by the identifiers they replace. 

1496 check : `bool`, optional 

1497 If `True` (default) check the query for consistency before 

1498 executing it. This may reject some valid queries that resemble 

1499 common mistakes (e.g. queries for visits without specifying an 

1500 instrument). 

1501 **kwargs 

1502 Additional keyword arguments are forwarded to 

1503 `DataCoordinate.standardize` when processing the ``dataId`` 

1504 argument (and may be used to provide a constraining data ID even 

1505 when the ``dataId`` argument is `None`). 

1506 

1507 Returns 

1508 ------- 

1509 dataIds : `queries.DataCoordinateQueryResults` 

1510 Data IDs matching the given query parameters. These are guaranteed 

1511 to identify all dimensions (`DataCoordinate.hasFull` returns 

1512 `True`), but will not contain `DimensionRecord` objects 

1513 (`DataCoordinate.hasRecords` returns `False`). Call 

1514 `DataCoordinateQueryResults.expanded` on the returned object to 

1515 fetch those (and consider using 

1516 `DataCoordinateQueryResults.materialize` on the returned object 

1517 first if the expected number of rows is very large). See 

1518 documentation for those methods for additional information. 

1519 

1520 Raises 

1521 ------ 

1522 NoDefaultCollectionError 

1523 Raised if ``collections`` is `None` and 

1524 ``self.defaults.collections`` is `None`. 

1525 CollectionExpressionError 

1526 Raised when ``collections`` expression is invalid. 

1527 DataIdError 

1528 Raised when ``dataId`` or keyword arguments specify unknown 

1529 dimensions or values, or when they contain inconsistent values. 

1530 DatasetTypeExpressionError 

1531 Raised when ``datasetType`` expression is invalid. 

1532 UserExpressionError 

1533 Raised when ``where`` expression is invalid. 

1534 """ 

1535 raise NotImplementedError() 

1536 

1537 @abstractmethod 

1538 def queryDimensionRecords( 

1539 self, 

1540 element: Union[DimensionElement, str], 

1541 *, 

1542 dataId: Optional[DataId] = None, 

1543 datasets: Any = None, 

1544 collections: Any = None, 

1545 where: str = "", 

1546 components: Optional[bool] = None, 

1547 bind: Optional[Mapping[str, Any]] = None, 

1548 check: bool = True, 

1549 **kwargs: Any, 

1550 ) -> DimensionRecordQueryResults: 

1551 """Query for dimension information matching user-provided criteria. 

1552 

1553 Parameters 

1554 ---------- 

1555 element : `DimensionElement` or `str` 

1556 The dimension element to obtain records for. 

1557 dataId : `dict` or `DataCoordinate`, optional 

1558 A data ID whose key-value pairs are used as equality constraints 

1559 in the query. 

1560 datasets : `Any`, optional 

1561 An expression that fully or partially identifies dataset types 

1562 that should constrain the yielded records. See `queryDataIds` and 

1563 :ref:`daf_butler_dataset_type_expressions` for more information. 

1564 collections : `Any`, optional 

1565 An expression that identifies the collections to search for 

1566 datasets, such as a `str` (for full matches or partial matches 

1567 via globs), `re.Pattern` (for partial matches), or iterable 

1568 thereof. ``...`` can be used to search all collections (actually 

1569 just all `~CollectionType.RUN` collections, because this will 

1570 still find all datasets). If not provided, 

1571 ``self.default.collections`` is used. Ignored unless ``datasets`` 

1572 is also passed. See :ref:`daf_butler_collection_expressions` for 

1573 more information. 

1574 where : `str`, optional 

1575 A string expression similar to a SQL WHERE clause. See 

1576 `queryDataIds` and :ref:`daf_butler_dimension_expressions` for more 

1577 information. 

1578 components : `bool`, optional 

1579 Whether to apply dataset expressions to components as well. 

1580 See `queryDataIds` for more information. 

1581 

1582 Values other than `False` are deprecated, and only `False` will be 

1583 supported after v26. After v27 this argument will be removed 

1584 entirely. 

1585 bind : `Mapping`, optional 

1586 Mapping containing literal values that should be injected into the 

1587 ``where`` expression, keyed by the identifiers they replace. 

1588 check : `bool`, optional 

1589 If `True` (default) check the query for consistency before 

1590 executing it. This may reject some valid queries that resemble 

1591 common mistakes (e.g. queries for visits without specifying an 

1592 instrument). 

1593 **kwargs 

1594 Additional keyword arguments are forwarded to 

1595 `DataCoordinate.standardize` when processing the ``dataId`` 

1596 argument (and may be used to provide a constraining data ID even 

1597 when the ``dataId`` argument is `None`). 

1598 

1599 Returns 

1600 ------- 

1601 dataIds : `queries.DimensionRecordQueryResults` 

1602 Data IDs matching the given query parameters. 

1603 

1604 Raises 

1605 ------ 

1606 NoDefaultCollectionError 

1607 Raised if ``collections`` is `None` and 

1608 ``self.defaults.collections`` is `None`. 

1609 CollectionExpressionError 

1610 Raised when ``collections`` expression is invalid. 

1611 DataIdError 

1612 Raised when ``dataId`` or keyword arguments specify unknown 

1613 dimensions or values, or when they contain inconsistent values. 

1614 DatasetTypeExpressionError 

1615 Raised when ``datasetType`` expression is invalid. 

1616 UserExpressionError 

1617 Raised when ``where`` expression is invalid. 

1618 """ 

1619 raise NotImplementedError() 

1620 

1621 @abstractmethod 

1622 def queryDatasetAssociations( 

1623 self, 

1624 datasetType: Union[str, DatasetType], 

1625 collections: Any = ..., 

1626 *, 

1627 collectionTypes: Iterable[CollectionType] = CollectionType.all(), 

1628 flattenChains: bool = False, 

1629 ) -> Iterator[DatasetAssociation]: 

1630 """Iterate over dataset-collection combinations where the dataset is in 

1631 the collection. 

1632 

1633 This method is a temporary placeholder for better support for 

1634 association results in `queryDatasets`. It will probably be 

1635 removed in the future, and should be avoided in production code 

1636 whenever possible. 

1637 

1638 Parameters 

1639 ---------- 

1640 datasetType : `DatasetType` or `str` 

1641 A dataset type object or the name of one. 

1642 collections: `Any`, optional 

1643 An expression that identifies the collections to search for 

1644 datasets, such as a `str` (for full matches or partial matches 

1645 via globs), `re.Pattern` (for partial matches), or iterable 

1646 thereof. ``...`` can be used to search all collections (actually 

1647 just all `~CollectionType.RUN` collections, because this will still 

1648 find all datasets). If not provided, ``self.default.collections`` 

1649 is used. See :ref:`daf_butler_collection_expressions` for more 

1650 information. 

1651 collectionTypes : `AbstractSet` [ `CollectionType` ], optional 

1652 If provided, only yield associations from collections of these 

1653 types. 

1654 flattenChains : `bool`, optional 

1655 If `True` (default) search in the children of 

1656 `~CollectionType.CHAINED` collections. If `False`, ``CHAINED`` 

1657 collections are ignored. 

1658 

1659 Yields 

1660 ------ 

1661 association : `.DatasetAssociation` 

1662 Object representing the relationship between a single dataset and 

1663 a single collection. 

1664 

1665 Raises 

1666 ------ 

1667 NoDefaultCollectionError 

1668 Raised if ``collections`` is `None` and 

1669 ``self.defaults.collections`` is `None`. 

1670 CollectionExpressionError 

1671 Raised when ``collections`` expression is invalid. 

1672 """ 

1673 raise NotImplementedError() 

1674 

1675 storageClasses: StorageClassFactory 

1676 """All storage classes known to the registry (`StorageClassFactory`). 

1677 """ 

1678 

1679 datasetIdFactory: DatasetIdFactory 

1680 """Factory for dataset IDs."""