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

184 statements  

« prev     ^ index     » next       coverage.py v6.5.0, created at 2022-10-12 09:01 +0000

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. 

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

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

692 the dataset within a collection. 

693 collections, optional. 

694 An expression that fully or partially identifies the collections to 

695 search for the dataset; see 

696 :ref:`daf_butler_collection_expressions` for more information. 

697 Defaults to ``self.defaults.collections``. 

698 timespan : `Timespan`, optional 

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

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

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

702 **kwargs 

703 Additional keyword arguments passed to 

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

705 `DataCoordinate` or augment an existing one. 

706 

707 Returns 

708 ------- 

709 ref : `DatasetRef` 

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

711 was found. 

712 

713 Raises 

714 ------ 

715 NoDefaultCollectionError 

716 Raised if ``collections`` is `None` and 

717 ``self.defaults.collections`` is `None`. 

718 LookupError 

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

720 MissingDatasetTypeError 

721 Raised if the dataset type does not exist. 

722 MissingCollectionError 

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

724 

725 Notes 

726 ----- 

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

728 when the set of collections searched is intrinsically incompatible with 

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

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

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

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

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

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

735 never changes the behavior. 

736 

737 This method handles component dataset types automatically, though most 

738 other registry operations do not. 

739 """ 

740 raise NotImplementedError() 

741 

742 @abstractmethod 

743 def insertDatasets( 

744 self, 

745 datasetType: Union[DatasetType, str], 

746 dataIds: Iterable[DataId], 

747 run: Optional[str] = None, 

748 expand: bool = True, 

749 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

750 ) -> List[DatasetRef]: 

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

752 

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

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

755 

756 Parameters 

757 ---------- 

758 datasetType : `DatasetType` or `str` 

759 A `DatasetType` or the name of one. 

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

761 Dimension-based identifiers for the new datasets. 

762 run : `str`, optional 

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

764 ``self.defaults.run``. 

765 expand : `bool`, optional 

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

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

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

769 unnecessary. 

770 idGenerationMode : `DatasetIdGenEnum`, optional 

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

772 are generated for each inserted dataset. 

773 

774 Returns 

775 ------- 

776 refs : `list` of `DatasetRef` 

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

778 order). 

779 

780 Raises 

781 ------ 

782 DatasetTypeError 

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

784 CollectionTypeError 

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

786 NoDefaultCollectionError 

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

788 ConflictingDefinitionError 

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

790 given already exists in ``run``. 

791 MissingCollectionError 

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

793 """ 

794 raise NotImplementedError() 

795 

796 @abstractmethod 

797 def _importDatasets( 

798 self, 

799 datasets: Iterable[DatasetRef], 

800 expand: bool = True, 

801 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

802 reuseIds: bool = False, 

803 ) -> List[DatasetRef]: 

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

805 

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

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

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

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

810 ignored if imported again. 

811 

812 Parameters 

813 ---------- 

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

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

816 identical ``datasetType`` and ``run`` attributes. ``run`` 

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

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

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

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

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

822 generated by backend. 

823 expand : `bool`, optional 

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

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

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

827 unnecessary. 

828 idGenerationMode : `DatasetIdGenEnum`, optional 

829 Specifies option for generating dataset IDs when IDs are not 

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

831 unique IDs are generated for each inserted dataset. 

832 reuseIds : `bool`, optional 

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

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

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

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

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

838 

839 Returns 

840 ------- 

841 refs : `list` of `DatasetRef` 

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

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

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

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

846 

847 Raises 

848 ------ 

849 NoDefaultCollectionError 

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

851 DatasetTypeError 

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

853 dataset type is not known to registry. 

854 ConflictingDefinitionError 

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

856 given already exists in ``run``. 

857 MissingCollectionError 

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

859 

860 Notes 

861 ----- 

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

863 implementation. Clients outside daf_butler package should not use this 

864 method. 

865 """ 

866 raise NotImplementedError() 

867 

868 @abstractmethod 

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

870 """Retrieve a Dataset entry. 

871 

872 Parameters 

873 ---------- 

874 id : `DatasetId` 

875 The unique identifier for the dataset. 

876 

877 Returns 

878 ------- 

879 ref : `DatasetRef` or `None` 

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

881 was found. 

882 """ 

883 raise NotImplementedError() 

884 

885 @abstractmethod 

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

887 """Remove datasets from the Registry. 

888 

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

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

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

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

893 removed from all Datastores. 

894 

895 Parameters 

896 ---------- 

897 refs : `Iterable` of `DatasetRef` 

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

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

900 

901 Raises 

902 ------ 

903 AmbiguousDatasetError 

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

905 OrphanedRecordError 

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

907 """ 

908 raise NotImplementedError() 

909 

910 @abstractmethod 

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

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

913 

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

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

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

917 `ConflictingDefinitionError` is raised. 

918 

919 Parameters 

920 ---------- 

921 collection : `str` 

922 Indicates the collection the datasets should be associated with. 

923 refs : `Iterable` [ `DatasetRef` ] 

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

925 in this `Registry`. 

926 

927 Raises 

928 ------ 

929 ConflictingDefinitionError 

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

931 given collection. 

932 AmbiguousDatasetError 

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

934 MissingCollectionError 

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

936 CollectionTypeError 

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

938 allowed. 

939 """ 

940 raise NotImplementedError() 

941 

942 @abstractmethod 

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

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

945 

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

947 associated are silently ignored. 

948 

949 Parameters 

950 ---------- 

951 collection : `str` 

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

953 refs : `Iterable` [ `DatasetRef` ] 

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

955 in this `Registry`. 

956 

957 Raises 

958 ------ 

959 AmbiguousDatasetError 

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

961 MissingCollectionError 

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

963 CollectionTypeError 

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

965 allowed. 

966 """ 

967 raise NotImplementedError() 

968 

969 @abstractmethod 

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

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

972 validity range within it. 

973 

974 Parameters 

975 ---------- 

976 collection : `str` 

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

978 collection. 

979 refs : `Iterable` [ `DatasetRef` ] 

980 Datasets to be associated. 

981 timespan : `Timespan` 

982 The validity range for these datasets within the collection. 

983 

984 Raises 

985 ------ 

986 AmbiguousDatasetError 

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

988 ConflictingDefinitionError 

989 Raised if the collection already contains a different dataset with 

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

991 range. 

992 CollectionTypeError 

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

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

995 which `DatasetType.isCalibration` returns `False`. 

996 """ 

997 raise NotImplementedError() 

998 

999 @abstractmethod 

1000 def decertify( 

1001 self, 

1002 collection: str, 

1003 datasetType: Union[str, DatasetType], 

1004 timespan: Timespan, 

1005 *, 

1006 dataIds: Optional[Iterable[DataId]] = None, 

1007 ) -> None: 

1008 """Remove or adjust datasets to clear a validity range within a 

1009 calibration collection. 

1010 

1011 Parameters 

1012 ---------- 

1013 collection : `str` 

1014 The name of an already-registered `~CollectionType.CALIBRATION` 

1015 collection. 

1016 datasetType : `str` or `DatasetType` 

1017 Name or `DatasetType` instance for the datasets to be decertified. 

1018 timespan : `Timespan`, optional 

1019 The validity range to remove datasets from within the collection. 

1020 Datasets that overlap this range but are not contained by it will 

1021 have their validity ranges adjusted to not overlap it, which may 

1022 split a single dataset validity range into two. 

1023 dataIds : `Iterable` [ `DataId` ], optional 

1024 Data IDs that should be decertified within the given validity range 

1025 If `None`, all data IDs for ``self.datasetType`` will be 

1026 decertified. 

1027 

1028 Raises 

1029 ------ 

1030 CollectionTypeError 

1031 Raised if ``collection`` is not a `~CollectionType.CALIBRATION` 

1032 collection or if ``datasetType.isCalibration() is False``. 

1033 """ 

1034 raise NotImplementedError() 

1035 

1036 @abstractmethod 

1037 def getDatastoreBridgeManager(self) -> DatastoreRegistryBridgeManager: 

1038 """Return an object that allows a new `Datastore` instance to 

1039 communicate with this `Registry`. 

1040 

1041 Returns 

1042 ------- 

1043 manager : `DatastoreRegistryBridgeManager` 

1044 Object that mediates communication between this `Registry` and its 

1045 associated datastores. 

1046 """ 

1047 raise NotImplementedError() 

1048 

1049 @abstractmethod 

1050 def getDatasetLocations(self, ref: DatasetRef) -> Iterable[str]: 

1051 """Retrieve datastore locations for a given dataset. 

1052 

1053 Parameters 

1054 ---------- 

1055 ref : `DatasetRef` 

1056 A reference to the dataset for which to retrieve storage 

1057 information. 

1058 

1059 Returns 

1060 ------- 

1061 datastores : `Iterable` [ `str` ] 

1062 All the matching datastores holding this dataset. 

1063 

1064 Raises 

1065 ------ 

1066 AmbiguousDatasetError 

1067 Raised if ``ref.id`` is `None`. 

1068 """ 

1069 raise NotImplementedError() 

1070 

1071 @abstractmethod 

1072 def expandDataId( 

1073 self, 

1074 dataId: Optional[DataId] = None, 

1075 *, 

1076 graph: Optional[DimensionGraph] = None, 

1077 records: Optional[NameLookupMapping[DimensionElement, Optional[DimensionRecord]]] = None, 

1078 withDefaults: bool = True, 

1079 **kwargs: Any, 

1080 ) -> DataCoordinate: 

1081 """Expand a dimension-based data ID to include additional information. 

1082 

1083 Parameters 

1084 ---------- 

1085 dataId : `DataCoordinate` or `dict`, optional 

1086 Data ID to be expanded; augmented and overridden by ``kwargs``. 

1087 graph : `DimensionGraph`, optional 

1088 Set of dimensions for the expanded ID. If `None`, the dimensions 

1089 will be inferred from the keys of ``dataId`` and ``kwargs``. 

1090 Dimensions that are in ``dataId`` or ``kwargs`` but not in 

1091 ``graph`` are silently ignored, providing a way to extract and 

1092 ``graph`` expand a subset of a data ID. 

1093 records : `Mapping` [`str`, `DimensionRecord`], optional 

1094 Dimension record data to use before querying the database for that 

1095 data, keyed by element name. 

1096 withDefaults : `bool`, optional 

1097 Utilize ``self.defaults.dataId`` to fill in missing governor 

1098 dimension key-value pairs. Defaults to `True` (i.e. defaults are 

1099 used). 

1100 **kwargs 

1101 Additional keywords are treated like additional key-value pairs for 

1102 ``dataId``, extending and overriding 

1103 

1104 Returns 

1105 ------- 

1106 expanded : `DataCoordinate` 

1107 A data ID that includes full metadata for all of the dimensions it 

1108 identifies, i.e. guarantees that ``expanded.hasRecords()`` and 

1109 ``expanded.hasFull()`` both return `True`. 

1110 

1111 Raises 

1112 ------ 

1113 DataIdError 

1114 Raised when ``dataId`` or keyword arguments specify unknown 

1115 dimensions or values, or when a resulting data ID contains 

1116 contradictory key-value pairs, according to dimension 

1117 relationships. 

1118 """ 

1119 raise NotImplementedError() 

1120 

1121 @abstractmethod 

1122 def insertDimensionData( 

1123 self, 

1124 element: Union[DimensionElement, str], 

1125 *data: Union[Mapping[str, Any], DimensionRecord], 

1126 conform: bool = True, 

1127 replace: bool = False, 

1128 skip_existing: bool = False, 

1129 ) -> None: 

1130 """Insert one or more dimension records into the database. 

1131 

1132 Parameters 

1133 ---------- 

1134 element : `DimensionElement` or `str` 

1135 The `DimensionElement` or name thereof that identifies the table 

1136 records will be inserted into. 

1137 data : `dict` or `DimensionRecord` (variadic) 

1138 One or more records to insert. 

1139 conform : `bool`, optional 

1140 If `False` (`True` is default) perform no checking or conversions, 

1141 and assume that ``element`` is a `DimensionElement` instance and 

1142 ``data`` is a one or more `DimensionRecord` instances of the 

1143 appropriate subclass. 

1144 replace : `bool`, optional 

1145 If `True` (`False` is default), replace existing records in the 

1146 database if there is a conflict. 

1147 skip_existing : `bool`, optional 

1148 If `True` (`False` is default), skip insertion if a record with 

1149 the same primary key values already exists. Unlike 

1150 `syncDimensionData`, this will not detect when the given record 

1151 differs from what is in the database, and should not be used when 

1152 this is a concern. 

1153 """ 

1154 raise NotImplementedError() 

1155 

1156 @abstractmethod 

1157 def syncDimensionData( 

1158 self, 

1159 element: Union[DimensionElement, str], 

1160 row: Union[Mapping[str, Any], DimensionRecord], 

1161 conform: bool = True, 

1162 update: bool = False, 

1163 ) -> Union[bool, Dict[str, Any]]: 

1164 """Synchronize the given dimension record with the database, inserting 

1165 if it does not already exist and comparing values if it does. 

1166 

1167 Parameters 

1168 ---------- 

1169 element : `DimensionElement` or `str` 

1170 The `DimensionElement` or name thereof that identifies the table 

1171 records will be inserted into. 

1172 row : `dict` or `DimensionRecord` 

1173 The record to insert. 

1174 conform : `bool`, optional 

1175 If `False` (`True` is default) perform no checking or conversions, 

1176 and assume that ``element`` is a `DimensionElement` instance and 

1177 ``data`` is a one or more `DimensionRecord` instances of the 

1178 appropriate subclass. 

1179 update: `bool`, optional 

1180 If `True` (`False` is default), update the existing record in the 

1181 database if there is a conflict. 

1182 

1183 Returns 

1184 ------- 

1185 inserted_or_updated : `bool` or `dict` 

1186 `True` if a new row was inserted, `False` if no changes were 

1187 needed, or a `dict` mapping updated column names to their old 

1188 values if an update was performed (only possible if 

1189 ``update=True``). 

1190 

1191 Raises 

1192 ------ 

1193 ConflictingDefinitionError 

1194 Raised if the record exists in the database (according to primary 

1195 key lookup) but is inconsistent with the given one. 

1196 """ 

1197 raise NotImplementedError() 

1198 

1199 @abstractmethod 

1200 def queryDatasetTypes( 

1201 self, 

1202 expression: Any = ..., 

1203 *, 

1204 components: Optional[bool] = None, 

1205 missing: Optional[List[str]] = None, 

1206 ) -> Iterable[DatasetType]: 

1207 """Iterate over the dataset types whose names match an expression. 

1208 

1209 Parameters 

1210 ---------- 

1211 expression : `Any`, optional 

1212 An expression that fully or partially identifies the dataset types 

1213 to return, such as a `str`, `re.Pattern`, or iterable thereof. 

1214 ``...`` can be used to return all dataset types, and is the 

1215 default. See :ref:`daf_butler_dataset_type_expressions` for more 

1216 information. 

1217 components : `bool`, optional 

1218 If `True`, apply all expression patterns to component dataset type 

1219 names as well. If `False`, never apply patterns to components. 

1220 If `None` (default), apply patterns to components only if their 

1221 parent datasets were not matched by the expression. 

1222 Fully-specified component datasets (`str` or `DatasetType` 

1223 instances) are always included. 

1224 

1225 Values other than `False` are deprecated, and only `False` will be 

1226 supported after v26. After v27 this argument will be removed 

1227 entirely. 

1228 missing : `list` of `str`, optional 

1229 String dataset type names that were explicitly given (i.e. not 

1230 regular expression patterns) but not found will be appended to this 

1231 list, if it is provided. 

1232 

1233 Returns 

1234 ------- 

1235 dataset_types : `Iterable` [ `DatasetType`] 

1236 An `Iterable` of `DatasetType` instances whose names match 

1237 ``expression``. 

1238 

1239 Raises 

1240 ------ 

1241 DatasetTypeExpressionError 

1242 Raised when ``expression`` is invalid. 

1243 """ 

1244 raise NotImplementedError() 

1245 

1246 @abstractmethod 

1247 def queryCollections( 

1248 self, 

1249 expression: Any = ..., 

1250 datasetType: Optional[DatasetType] = None, 

1251 collectionTypes: Union[Iterable[CollectionType], CollectionType] = CollectionType.all(), 

1252 flattenChains: bool = False, 

1253 includeChains: Optional[bool] = None, 

1254 ) -> Sequence[str]: 

1255 """Iterate over the collections whose names match an expression. 

1256 

1257 Parameters 

1258 ---------- 

1259 expression : `Any`, optional 

1260 An expression that identifies the collections to return, such as 

1261 a `str` (for full matches or partial matches via globs), 

1262 `re.Pattern` (for partial matches), or iterable thereof. ``...`` 

1263 can be used to return all collections, and is the default. 

1264 See :ref:`daf_butler_collection_expressions` for more information. 

1265 datasetType : `DatasetType`, optional 

1266 If provided, only yield collections that may contain datasets of 

1267 this type. This is a conservative approximation in general; it may 

1268 yield collections that do not have any such datasets. 

1269 collectionTypes : `AbstractSet` [ `CollectionType` ] or \ 

1270 `CollectionType`, optional 

1271 If provided, only yield collections of these types. 

1272 flattenChains : `bool`, optional 

1273 If `True` (`False` is default), recursively yield the child 

1274 collections of matching `~CollectionType.CHAINED` collections. 

1275 includeChains : `bool`, optional 

1276 If `True`, yield records for matching `~CollectionType.CHAINED` 

1277 collections. Default is the opposite of ``flattenChains``: include 

1278 either CHAINED collections or their children, but not both. 

1279 

1280 Returns 

1281 ------- 

1282 collections : `Sequence` [ `str` ] 

1283 The names of collections that match ``expression``. 

1284 

1285 Raises 

1286 ------ 

1287 CollectionExpressionError 

1288 Raised when ``expression`` is invalid. 

1289 

1290 Notes 

1291 ----- 

1292 The order in which collections are returned is unspecified, except that 

1293 the children of a `~CollectionType.CHAINED` collection are guaranteed 

1294 to be in the order in which they are searched. When multiple parent 

1295 `~CollectionType.CHAINED` collections match the same criteria, the 

1296 order in which the two lists appear is unspecified, and the lists of 

1297 children may be incomplete if a child has multiple parents. 

1298 """ 

1299 raise NotImplementedError() 

1300 

1301 @abstractmethod 

1302 def queryDatasets( 

1303 self, 

1304 datasetType: Any, 

1305 *, 

1306 collections: Any = None, 

1307 dimensions: Optional[Iterable[Union[Dimension, str]]] = None, 

1308 dataId: Optional[DataId] = None, 

1309 where: Optional[str] = None, 

1310 findFirst: bool = False, 

1311 components: Optional[bool] = None, 

1312 bind: Optional[Mapping[str, Any]] = None, 

1313 check: bool = True, 

1314 **kwargs: Any, 

1315 ) -> DatasetQueryResults: 

1316 """Query for and iterate over dataset references matching user-provided 

1317 criteria. 

1318 

1319 Parameters 

1320 ---------- 

1321 datasetType 

1322 An expression that fully or partially identifies the dataset types 

1323 to be queried. Allowed types include `DatasetType`, `str`, 

1324 `re.Pattern`, and iterables thereof. The special value ``...`` can 

1325 be used to query all dataset types. See 

1326 :ref:`daf_butler_dataset_type_expressions` for more information. 

1327 collections: optional 

1328 An expression that identifies the collections to search, such as a 

1329 `str` (for full matches or partial matches via globs), `re.Pattern` 

1330 (for partial matches), or iterable thereof. ``...`` can be used to 

1331 search all collections (actually just all `~CollectionType.RUN` 

1332 collections, because this will still find all datasets). 

1333 If not provided, ``self.default.collections`` is used. See 

1334 :ref:`daf_butler_collection_expressions` for more information. 

1335 dimensions : `~collections.abc.Iterable` of `Dimension` or `str` 

1336 Dimensions to include in the query (in addition to those used 

1337 to identify the queried dataset type(s)), either to constrain 

1338 the resulting datasets to those for which a matching dimension 

1339 exists, or to relate the dataset type's dimensions to dimensions 

1340 referenced by the ``dataId`` or ``where`` arguments. 

1341 dataId : `dict` or `DataCoordinate`, optional 

1342 A data ID whose key-value pairs are used as equality constraints 

1343 in the query. 

1344 where : `str`, optional 

1345 A string expression similar to a SQL WHERE clause. May involve 

1346 any column of a dimension table or (as a shortcut for the primary 

1347 key column of a dimension table) dimension name. See 

1348 :ref:`daf_butler_dimension_expressions` for more information. 

1349 findFirst : `bool`, optional 

1350 If `True` (`False` is default), for each result data ID, only 

1351 yield one `DatasetRef` of each `DatasetType`, from the first 

1352 collection in which a dataset of that dataset type appears 

1353 (according to the order of ``collections`` passed in). If `True`, 

1354 ``collections`` must not contain regular expressions and may not 

1355 be ``...``. 

1356 components : `bool`, optional 

1357 If `True`, apply all dataset expression patterns to component 

1358 dataset type names as well. If `False`, never apply patterns to 

1359 components. If `None` (default), apply patterns to components only 

1360 if their parent datasets were not matched by the expression. 

1361 Fully-specified component datasets (`str` or `DatasetType` 

1362 instances) are always included. 

1363 

1364 Values other than `False` are deprecated, and only `False` will be 

1365 supported after v26. After v27 this argument will be removed 

1366 entirely. 

1367 bind : `Mapping`, optional 

1368 Mapping containing literal values that should be injected into the 

1369 ``where`` expression, keyed by the identifiers they replace. 

1370 check : `bool`, optional 

1371 If `True` (default) check the query for consistency before 

1372 executing it. This may reject some valid queries that resemble 

1373 common mistakes (e.g. queries for visits without specifying an 

1374 instrument). 

1375 **kwargs 

1376 Additional keyword arguments are forwarded to 

1377 `DataCoordinate.standardize` when processing the ``dataId`` 

1378 argument (and may be used to provide a constraining data ID even 

1379 when the ``dataId`` argument is `None`). 

1380 

1381 Returns 

1382 ------- 

1383 refs : `queries.DatasetQueryResults` 

1384 Dataset references matching the given query criteria. Nested data 

1385 IDs are guaranteed to include values for all implied dimensions 

1386 (i.e. `DataCoordinate.hasFull` will return `True`), but will not 

1387 include dimension records (`DataCoordinate.hasRecords` will be 

1388 `False`) unless `~queries.DatasetQueryResults.expanded` is called 

1389 on the result object (which returns a new one). 

1390 

1391 Raises 

1392 ------ 

1393 DatasetTypeExpressionError 

1394 Raised when ``datasetType`` expression is invalid. 

1395 TypeError 

1396 Raised when the arguments are incompatible, such as when a 

1397 collection wildcard is passed when ``findFirst`` is `True`, or 

1398 when ``collections`` is `None` and``self.defaults.collections`` is 

1399 also `None`. 

1400 DataIdError 

1401 Raised when ``dataId`` or keyword arguments specify unknown 

1402 dimensions or values, or when they contain inconsistent values. 

1403 UserExpressionError 

1404 Raised when ``where`` expression is invalid. 

1405 

1406 Notes 

1407 ----- 

1408 When multiple dataset types are queried in a single call, the 

1409 results of this operation are equivalent to querying for each dataset 

1410 type separately in turn, and no information about the relationships 

1411 between datasets of different types is included. In contexts where 

1412 that kind of information is important, the recommended pattern is to 

1413 use `queryDataIds` to first obtain data IDs (possibly with the 

1414 desired dataset types and collections passed as constraints to the 

1415 query), and then use multiple (generally much simpler) calls to 

1416 `queryDatasets` with the returned data IDs passed as constraints. 

1417 """ 

1418 raise NotImplementedError() 

1419 

1420 @abstractmethod 

1421 def queryDataIds( 

1422 self, 

1423 dimensions: Union[Iterable[Union[Dimension, str]], Dimension, str], 

1424 *, 

1425 dataId: Optional[DataId] = None, 

1426 datasets: Any = None, 

1427 collections: Any = None, 

1428 where: Optional[str] = None, 

1429 components: Optional[bool] = None, 

1430 bind: Optional[Mapping[str, Any]] = None, 

1431 check: bool = True, 

1432 **kwargs: Any, 

1433 ) -> DataCoordinateQueryResults: 

1434 """Query for data IDs matching user-provided criteria. 

1435 

1436 Parameters 

1437 ---------- 

1438 dimensions : `Dimension` or `str`, or iterable thereof 

1439 The dimensions of the data IDs to yield, as either `Dimension` 

1440 instances or `str`. Will be automatically expanded to a complete 

1441 `DimensionGraph`. 

1442 dataId : `dict` or `DataCoordinate`, optional 

1443 A data ID whose key-value pairs are used as equality constraints 

1444 in the query. 

1445 datasets : `Any`, optional 

1446 An expression that fully or partially identifies dataset types 

1447 that should constrain the yielded data IDs. For example, including 

1448 "raw" here would constrain the yielded ``instrument``, 

1449 ``exposure``, ``detector``, and ``physical_filter`` values to only 

1450 those for which at least one "raw" dataset exists in 

1451 ``collections``. Allowed types include `DatasetType`, `str`, 

1452 and iterables thereof. Regular expression objects (i.e. 

1453 `re.Pattern`) are deprecated and will be removed after the v26 

1454 release. See :ref:`daf_butler_dataset_type_expressions` for more 

1455 information. 

1456 collections: `Any`, optional 

1457 An expression that identifies the collections to search for 

1458 datasets, such as a `str` (for full matches or partial matches 

1459 via globs), `re.Pattern` (for partial matches), or iterable 

1460 thereof. ``...`` can be used to search all collections (actually 

1461 just all `~CollectionType.RUN` collections, because this will 

1462 still find all datasets). If not provided, 

1463 ``self.default.collections`` is used. Ignored unless ``datasets`` 

1464 is also passed. See :ref:`daf_butler_collection_expressions` for 

1465 more information. 

1466 where : `str`, optional 

1467 A string expression similar to a SQL WHERE clause. May involve 

1468 any column of a dimension table or (as a shortcut for the primary 

1469 key column of a dimension table) dimension name. See 

1470 :ref:`daf_butler_dimension_expressions` for more information. 

1471 components : `bool`, optional 

1472 If `True`, apply all dataset expression patterns to component 

1473 dataset type names as well. If `False`, never apply patterns to 

1474 components. If `None` (default), apply patterns to components only 

1475 if their parent datasets were not matched by the expression. 

1476 Fully-specified component datasets (`str` or `DatasetType` 

1477 instances) are always included. 

1478 

1479 Values other than `False` are deprecated, and only `False` will be 

1480 supported after v26. After v27 this argument will be removed 

1481 entirely. 

1482 bind : `Mapping`, optional 

1483 Mapping containing literal values that should be injected into the 

1484 ``where`` expression, keyed by the identifiers they replace. 

1485 check : `bool`, optional 

1486 If `True` (default) check the query for consistency before 

1487 executing it. This may reject some valid queries that resemble 

1488 common mistakes (e.g. queries for visits without specifying an 

1489 instrument). 

1490 **kwargs 

1491 Additional keyword arguments are forwarded to 

1492 `DataCoordinate.standardize` when processing the ``dataId`` 

1493 argument (and may be used to provide a constraining data ID even 

1494 when the ``dataId`` argument is `None`). 

1495 

1496 Returns 

1497 ------- 

1498 dataIds : `queries.DataCoordinateQueryResults` 

1499 Data IDs matching the given query parameters. These are guaranteed 

1500 to identify all dimensions (`DataCoordinate.hasFull` returns 

1501 `True`), but will not contain `DimensionRecord` objects 

1502 (`DataCoordinate.hasRecords` returns `False`). Call 

1503 `DataCoordinateQueryResults.expanded` on the returned object to 

1504 fetch those (and consider using 

1505 `DataCoordinateQueryResults.materialize` on the returned object 

1506 first if the expected number of rows is very large). See 

1507 documentation for those methods for additional information. 

1508 

1509 Raises 

1510 ------ 

1511 NoDefaultCollectionError 

1512 Raised if ``collections`` is `None` and 

1513 ``self.defaults.collections`` is `None`. 

1514 CollectionExpressionError 

1515 Raised when ``collections`` expression is invalid. 

1516 DataIdError 

1517 Raised when ``dataId`` or keyword arguments specify unknown 

1518 dimensions or values, or when they contain inconsistent values. 

1519 DatasetTypeExpressionError 

1520 Raised when ``datasetType`` expression is invalid. 

1521 UserExpressionError 

1522 Raised when ``where`` expression is invalid. 

1523 """ 

1524 raise NotImplementedError() 

1525 

1526 @abstractmethod 

1527 def queryDimensionRecords( 

1528 self, 

1529 element: Union[DimensionElement, str], 

1530 *, 

1531 dataId: Optional[DataId] = None, 

1532 datasets: Any = None, 

1533 collections: Any = None, 

1534 where: Optional[str] = None, 

1535 components: Optional[bool] = None, 

1536 bind: Optional[Mapping[str, Any]] = None, 

1537 check: bool = True, 

1538 **kwargs: Any, 

1539 ) -> DimensionRecordQueryResults: 

1540 """Query for dimension information matching user-provided criteria. 

1541 

1542 Parameters 

1543 ---------- 

1544 element : `DimensionElement` or `str` 

1545 The dimension element to obtain records for. 

1546 dataId : `dict` or `DataCoordinate`, optional 

1547 A data ID whose key-value pairs are used as equality constraints 

1548 in the query. 

1549 datasets : `Any`, optional 

1550 An expression that fully or partially identifies dataset types 

1551 that should constrain the yielded records. See `queryDataIds` and 

1552 :ref:`daf_butler_dataset_type_expressions` for more information. 

1553 collections : `Any`, optional 

1554 An expression that identifies the collections to search for 

1555 datasets, such as a `str` (for full matches or partial matches 

1556 via globs), `re.Pattern` (for partial matches), or iterable 

1557 thereof. ``...`` can be used to search all collections (actually 

1558 just all `~CollectionType.RUN` collections, because this will 

1559 still find all datasets). If not provided, 

1560 ``self.default.collections`` is used. Ignored unless ``datasets`` 

1561 is also passed. See :ref:`daf_butler_collection_expressions` for 

1562 more information. 

1563 where : `str`, optional 

1564 A string expression similar to a SQL WHERE clause. See 

1565 `queryDataIds` and :ref:`daf_butler_dimension_expressions` for more 

1566 information. 

1567 components : `bool`, optional 

1568 Whether to apply dataset expressions to components as well. 

1569 See `queryDataIds` for more information. 

1570 

1571 Values other than `False` are deprecated, and only `False` will be 

1572 supported after v26. After v27 this argument will be removed 

1573 entirely. 

1574 bind : `Mapping`, optional 

1575 Mapping containing literal values that should be injected into the 

1576 ``where`` expression, keyed by the identifiers they replace. 

1577 check : `bool`, optional 

1578 If `True` (default) check the query for consistency before 

1579 executing it. This may reject some valid queries that resemble 

1580 common mistakes (e.g. queries for visits without specifying an 

1581 instrument). 

1582 **kwargs 

1583 Additional keyword arguments are forwarded to 

1584 `DataCoordinate.standardize` when processing the ``dataId`` 

1585 argument (and may be used to provide a constraining data ID even 

1586 when the ``dataId`` argument is `None`). 

1587 

1588 Returns 

1589 ------- 

1590 dataIds : `queries.DimensionRecordQueryResults` 

1591 Data IDs matching the given query parameters. 

1592 

1593 Raises 

1594 ------ 

1595 NoDefaultCollectionError 

1596 Raised if ``collections`` is `None` and 

1597 ``self.defaults.collections`` is `None`. 

1598 CollectionExpressionError 

1599 Raised when ``collections`` expression is invalid. 

1600 DataIdError 

1601 Raised when ``dataId`` or keyword arguments specify unknown 

1602 dimensions or values, or when they contain inconsistent values. 

1603 DatasetTypeExpressionError 

1604 Raised when ``datasetType`` expression is invalid. 

1605 UserExpressionError 

1606 Raised when ``where`` expression is invalid. 

1607 """ 

1608 raise NotImplementedError() 

1609 

1610 @abstractmethod 

1611 def queryDatasetAssociations( 

1612 self, 

1613 datasetType: Union[str, DatasetType], 

1614 collections: Any = ..., 

1615 *, 

1616 collectionTypes: Iterable[CollectionType] = CollectionType.all(), 

1617 flattenChains: bool = False, 

1618 ) -> Iterator[DatasetAssociation]: 

1619 """Iterate over dataset-collection combinations where the dataset is in 

1620 the collection. 

1621 

1622 This method is a temporary placeholder for better support for 

1623 association results in `queryDatasets`. It will probably be 

1624 removed in the future, and should be avoided in production code 

1625 whenever possible. 

1626 

1627 Parameters 

1628 ---------- 

1629 datasetType : `DatasetType` or `str` 

1630 A dataset type object or the name of one. 

1631 collections: `Any`, optional 

1632 An expression that identifies the collections to search for 

1633 datasets, such as a `str` (for full matches or partial matches 

1634 via globs), `re.Pattern` (for partial matches), or iterable 

1635 thereof. ``...`` can be used to search all collections (actually 

1636 just all `~CollectionType.RUN` collections, because this will still 

1637 find all datasets). If not provided, ``self.default.collections`` 

1638 is used. See :ref:`daf_butler_collection_expressions` for more 

1639 information. 

1640 collectionTypes : `AbstractSet` [ `CollectionType` ], optional 

1641 If provided, only yield associations from collections of these 

1642 types. 

1643 flattenChains : `bool`, optional 

1644 If `True` (default) search in the children of 

1645 `~CollectionType.CHAINED` collections. If `False`, ``CHAINED`` 

1646 collections are ignored. 

1647 

1648 Yields 

1649 ------ 

1650 association : `.DatasetAssociation` 

1651 Object representing the relationship between a single dataset and 

1652 a single collection. 

1653 

1654 Raises 

1655 ------ 

1656 NoDefaultCollectionError 

1657 Raised if ``collections`` is `None` and 

1658 ``self.defaults.collections`` is `None`. 

1659 CollectionExpressionError 

1660 Raised when ``collections`` expression is invalid. 

1661 """ 

1662 raise NotImplementedError() 

1663 

1664 storageClasses: StorageClassFactory 

1665 """All storage classes known to the registry (`StorageClassFactory`). 

1666 """ 

1667 

1668 datasetIdFactory: DatasetIdFactory 

1669 """Factory for dataset IDs."""