Coverage for python/lsst/daf/butler/registry/datasets/byDimensions/tables.py: 97%
72 statements
« prev ^ index » next coverage.py v7.5.0, created at 2024-04-27 02:59 -0700
« prev ^ index » next coverage.py v7.5.0, created at 2024-04-27 02:59 -0700
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 software is dual licensed under the GNU General Public License and also
10# under a 3-clause BSD license. Recipients may choose which of these licenses
11# to use; please see the files gpl-3.0.txt and/or bsd_license.txt,
12# respectively. If you choose the GPL option then the following text applies
13# (but note that there is still no warranty even if you opt for BSD instead):
14#
15# This program is free software: you can redistribute it and/or modify
16# it under the terms of the GNU General Public License as published by
17# the Free Software Foundation, either version 3 of the License, or
18# (at your option) any later version.
19#
20# This program is distributed in the hope that it will be useful,
21# but WITHOUT ANY WARRANTY; without even the implied warranty of
22# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23# GNU General Public License for more details.
24#
25# You should have received a copy of the GNU General Public License
26# along with this program. If not, see <http://www.gnu.org/licenses/>.
28from __future__ import annotations
30from .... import ddl
32__all__ = (
33 "addDatasetForeignKey",
34 "makeCalibTableName",
35 "makeCalibTableSpec",
36 "makeStaticTableSpecs",
37 "makeTagTableName",
38 "makeTagTableSpec",
39 "StaticDatasetTablesTuple",
40)
42from collections import namedtuple
43from typing import Any
45import sqlalchemy
47from ...._dataset_type import DatasetType
48from ....dimensions import DimensionUniverse, GovernorDimension, addDimensionForeignKey
49from ....timespan_database_representation import TimespanDatabaseRepresentation
50from ...interfaces import CollectionManager, VersionTuple
52DATASET_TYPE_NAME_LENGTH = 128
55StaticDatasetTablesTuple = namedtuple(
56 "StaticDatasetTablesTuple",
57 [
58 "dataset_type",
59 "dataset",
60 ],
61)
64def addDatasetForeignKey(
65 tableSpec: ddl.TableSpec,
66 dtype: type,
67 *,
68 name: str = "dataset",
69 onDelete: str | None = None,
70 constraint: bool = True,
71 **kwargs: Any,
72) -> ddl.FieldSpec:
73 """Add a foreign key column for datasets and (optionally) a constraint to
74 a table.
76 This is an internal interface for the ``byDimensions`` package; external
77 code should use `DatasetRecordStorageManager.addDatasetForeignKey` instead.
79 Parameters
80 ----------
81 tableSpec : `ddl.TableSpec`
82 Specification for the table that should reference the dataset
83 table. Will be modified in place.
84 dtype : `type`
85 Type of the column, same as the column type of the PK column of
86 a referenced table (``dataset.id``).
87 name : `str`, optional
88 A name to use for the prefix of the new field; the full name is
89 ``{name}_id``.
90 onDelete : `str`, optional
91 One of "CASCADE" or "SET NULL", indicating what should happen to
92 the referencing row if the collection row is deleted. `None`
93 indicates that this should be an integrity error.
94 constraint : `bool`, optional
95 If `False` (`True` is default), add a field that can be joined to
96 the dataset primary key, but do not add a foreign key constraint.
97 **kwargs
98 Additional keyword arguments are forwarded to the `ddl.FieldSpec`
99 constructor (only the ``name`` and ``dtype`` arguments are
100 otherwise provided).
102 Returns
103 -------
104 idSpec : `ddl.FieldSpec`
105 Specification for the ID field.
106 """
107 idFieldSpec = ddl.FieldSpec(f"{name}_id", dtype=dtype, **kwargs)
108 tableSpec.fields.add(idFieldSpec)
109 if constraint:
110 tableSpec.foreignKeys.append(
111 ddl.ForeignKeySpec("dataset", source=(idFieldSpec.name,), target=("id",), onDelete=onDelete)
112 )
113 return idFieldSpec
116def makeStaticTableSpecs(
117 collections: type[CollectionManager],
118 universe: DimensionUniverse,
119 dtype: type,
120 autoincrement: bool,
121 schema_version: VersionTuple,
122) -> StaticDatasetTablesTuple:
123 """Construct all static tables used by the classes in this package.
125 Static tables are those that are present in all Registries and do not
126 depend on what DatasetTypes have been registered.
128 Parameters
129 ----------
130 collections : `CollectionManager`
131 Manager object for the collections in this `Registry`.
132 universe : `DimensionUniverse`
133 Universe graph containing all dimensions known to this `Registry`.
134 dtype : `type`
135 Type of the dataset ID (primary key) column.
136 autoincrement : `bool`
137 If `True` then dataset ID column will be auto-incrementing.
138 schema_version : `VersionTuple`
139 The version of this schema.
141 Returns
142 -------
143 specs : `StaticDatasetTablesTuple`
144 A named tuple containing `ddl.TableSpec` instances.
145 """
146 ingest_date_type: type
147 ingest_date_default: Any = None
148 if schema_version.major > 1:
149 ingest_date_type = ddl.AstropyTimeNsecTai
150 else:
151 ingest_date_type = sqlalchemy.TIMESTAMP
152 # New code provides explicit values for ingest_data, but we keep
153 # default just to be consistent with the existing schema.
154 ingest_date_default = sqlalchemy.sql.func.now()
156 specs = StaticDatasetTablesTuple(
157 dataset_type=ddl.TableSpec(
158 fields=[
159 ddl.FieldSpec(
160 name="id",
161 dtype=sqlalchemy.BigInteger,
162 autoincrement=True,
163 primaryKey=True,
164 doc=(
165 "Autoincrement ID that uniquely identifies a dataset "
166 "type in other tables. Python code outside the "
167 "`Registry` class should never interact with this; "
168 "its existence is considered an implementation detail."
169 ),
170 ),
171 ddl.FieldSpec(
172 name="name",
173 dtype=sqlalchemy.String,
174 length=DATASET_TYPE_NAME_LENGTH,
175 nullable=False,
176 doc="String name that uniquely identifies a dataset type.",
177 ),
178 ddl.FieldSpec(
179 name="storage_class",
180 dtype=sqlalchemy.String,
181 length=64,
182 nullable=False,
183 doc=(
184 "Name of the storage class associated with all "
185 "datasets of this type. Storage classes are "
186 "generally associated with a Python class, and are "
187 "enumerated in butler configuration."
188 ),
189 ),
190 ddl.FieldSpec(
191 name="dimensions_key",
192 dtype=sqlalchemy.BigInteger,
193 nullable=False,
194 doc="Unique key for the set of dimensions that identifies datasets of this type.",
195 ),
196 ddl.FieldSpec(
197 name="tag_association_table",
198 dtype=sqlalchemy.String,
199 length=128,
200 nullable=False,
201 doc=(
202 "Name of the table that holds associations between "
203 "datasets of this type and most types of collections."
204 ),
205 ),
206 ddl.FieldSpec(
207 name="calibration_association_table",
208 dtype=sqlalchemy.String,
209 length=128,
210 nullable=True,
211 doc=(
212 "Name of the table that holds associations between "
213 "datasets of this type and CALIBRATION collections. "
214 "NULL values indicate dataset types with "
215 "isCalibration=False."
216 ),
217 ),
218 ],
219 unique=[("name",)],
220 ),
221 dataset=ddl.TableSpec(
222 fields=[
223 ddl.FieldSpec(
224 name="id",
225 dtype=dtype,
226 autoincrement=autoincrement,
227 primaryKey=True,
228 doc="A unique field used as the primary key for dataset.",
229 ),
230 ddl.FieldSpec(
231 name="dataset_type_id",
232 dtype=sqlalchemy.BigInteger,
233 nullable=False,
234 doc="Reference to the associated entry in the dataset_type table.",
235 ),
236 ddl.FieldSpec(
237 name="ingest_date",
238 dtype=ingest_date_type,
239 default=ingest_date_default,
240 nullable=False,
241 doc="Time of dataset ingestion.",
242 ),
243 # Foreign key field/constraint to run added below.
244 ],
245 foreignKeys=[
246 ddl.ForeignKeySpec("dataset_type", source=("dataset_type_id",), target=("id",)),
247 ],
248 ),
249 )
250 # Add foreign key fields programmatically.
251 collections.addRunForeignKey(specs.dataset, onDelete="CASCADE", nullable=False)
252 return specs
255def makeTagTableName(datasetType: DatasetType, dimensionsKey: int) -> str:
256 """Construct the name for a dynamic (DatasetType-dependent) tag table used
257 by the classes in this package.
259 Parameters
260 ----------
261 datasetType : `DatasetType`
262 Dataset type to construct a name for. Multiple dataset types may
263 share the same table.
264 dimensionsKey : `int`
265 Integer key used to save ``datasetType.dimensions`` to the database.
267 Returns
268 -------
269 name : `str`
270 Name for the table.
271 """
272 return f"dataset_tags_{dimensionsKey:08d}"
275def makeCalibTableName(datasetType: DatasetType, dimensionsKey: int) -> str:
276 """Construct the name for a dynamic (DatasetType-dependent) tag + validity
277 range table used by the classes in this package.
279 Parameters
280 ----------
281 datasetType : `DatasetType`
282 Dataset type to construct a name for. Multiple dataset types may
283 share the same table.
284 dimensionsKey : `int`
285 Integer key used to save ``datasetType.dimensions`` to the database.
287 Returns
288 -------
289 name : `str`
290 Name for the table.
291 """
292 assert datasetType.isCalibration()
293 return f"dataset_calibs_{dimensionsKey:08d}"
296def makeTagTableSpec(
297 datasetType: DatasetType, collections: type[CollectionManager], dtype: type, *, constraints: bool = True
298) -> ddl.TableSpec:
299 """Construct the specification for a dynamic (DatasetType-dependent) tag
300 table used by the classes in this package.
302 Parameters
303 ----------
304 datasetType : `DatasetType`
305 Dataset type to construct a spec for. Multiple dataset types may
306 share the same table.
307 collections : `type` [ `CollectionManager` ]
308 `CollectionManager` subclass that can be used to construct foreign keys
309 to the run and/or collection tables.
310 dtype : `type`
311 Type of the FK column, same as the column type of the PK column of
312 a referenced table (``dataset.id``).
313 constraints : `bool`, optional
314 If `False` (`True` is default), do not define foreign key constraints.
316 Returns
317 -------
318 spec : `ddl.TableSpec`
319 Specification for the table.
320 """
321 tableSpec = ddl.TableSpec(
322 fields=[
323 # Foreign key fields to dataset, collection, and usually dimension
324 # tables added below.
325 # The dataset_type_id field here would be redundant with the one
326 # in the main monolithic dataset table, but we need it here for an
327 # important unique constraint.
328 ddl.FieldSpec("dataset_type_id", dtype=sqlalchemy.BigInteger, nullable=False),
329 ]
330 )
331 if constraints:
332 tableSpec.foreignKeys.append(
333 ddl.ForeignKeySpec("dataset_type", source=("dataset_type_id",), target=("id",))
334 )
335 # We'll also have a unique constraint on dataset type, collection, and data
336 # ID. We only include the required part of the data ID, as that's
337 # sufficient and saves us from worrying about nulls in the constraint.
338 constraint = ["dataset_type_id"]
339 # Add foreign key fields to dataset table (part of the primary key)
340 addDatasetForeignKey(tableSpec, dtype, primaryKey=True, onDelete="CASCADE", constraint=constraints)
341 # Add foreign key fields to collection table (part of the primary key and
342 # the data ID unique constraint).
343 collectionFieldSpec = collections.addCollectionForeignKey(
344 tableSpec, primaryKey=True, onDelete="CASCADE", constraint=constraints
345 )
346 constraint.append(collectionFieldSpec.name)
347 # Add foreign key constraint to the collection_summary_dataset_type table.
348 if constraints:
349 tableSpec.foreignKeys.append(
350 ddl.ForeignKeySpec(
351 "collection_summary_dataset_type",
352 source=(collectionFieldSpec.name, "dataset_type_id"),
353 target=(collectionFieldSpec.name, "dataset_type_id"),
354 )
355 )
356 for dimension_name in datasetType.dimensions.required.names:
357 dimension = datasetType.dimensions.universe.dimensions[dimension_name]
358 fieldSpec = addDimensionForeignKey(
359 tableSpec, dimension=dimension, nullable=False, primaryKey=False, constraint=constraints
360 )
361 constraint.append(fieldSpec.name)
362 # If this is a governor dimension, add a foreign key constraint to the
363 # collection_summary_<dimension> table.
364 if isinstance(dimension, GovernorDimension) and constraints:
365 tableSpec.foreignKeys.append(
366 ddl.ForeignKeySpec(
367 f"collection_summary_{dimension.name}",
368 source=(collectionFieldSpec.name, fieldSpec.name),
369 target=(collectionFieldSpec.name, fieldSpec.name),
370 )
371 )
372 # Actually add the unique constraint.
373 tableSpec.unique.add(tuple(constraint))
374 return tableSpec
377def makeCalibTableSpec(
378 datasetType: DatasetType,
379 collections: type[CollectionManager],
380 TimespanReprClass: type[TimespanDatabaseRepresentation],
381 dtype: type,
382) -> ddl.TableSpec:
383 """Construct the specification for a dynamic (DatasetType-dependent) tag +
384 validity range table used by the classes in this package.
386 Parameters
387 ----------
388 datasetType : `DatasetType`
389 Dataset type to construct a spec for. Multiple dataset types may
390 share the same table.
391 collections : `type` [ `CollectionManager` ]
392 `CollectionManager` subclass that can be used to construct foreign keys
393 to the run and/or collection tables.
394 TimespanReprClass : `type` of `TimespanDatabaseRepresentation`
395 The Python type to use to represent a timespan.
396 dtype : `type`
397 Type of the FK column, same as the column type of the PK column of
398 a referenced table (``dataset.id``).
400 Returns
401 -------
402 spec : `ddl.TableSpec`
403 Specification for the table.
404 """
405 tableSpec = ddl.TableSpec(
406 fields=[
407 # This table has no natural primary key, compound or otherwise, so
408 # we add an autoincrement key. We may use this field a bit
409 # internally, but its presence is an implementation detail and it
410 # shouldn't appear as a foreign key in any other tables.
411 ddl.FieldSpec("id", dtype=sqlalchemy.BigInteger, autoincrement=True, primaryKey=True),
412 # Foreign key fields to dataset, collection, and usually dimension
413 # tables added below. The dataset_type_id field here is redundant
414 # with the one in the main monolithic dataset table, but this bit
415 # of denormalization lets us define what should be a much more
416 # useful index.
417 ddl.FieldSpec("dataset_type_id", dtype=sqlalchemy.BigInteger, nullable=False),
418 ],
419 foreignKeys=[
420 ddl.ForeignKeySpec("dataset_type", source=("dataset_type_id",), target=("id",)),
421 ],
422 )
423 # Record fields that should go in the temporal lookup index/constraint,
424 # starting with the dataset type.
425 index: list[str | type[TimespanDatabaseRepresentation]] = ["dataset_type_id"]
426 # Add foreign key fields to dataset table (not part of the temporal
427 # lookup/constraint).
428 addDatasetForeignKey(tableSpec, dtype, nullable=False, onDelete="CASCADE")
429 # Add foreign key fields to collection table (part of the temporal lookup
430 # index/constraint).
431 collectionFieldSpec = collections.addCollectionForeignKey(tableSpec, nullable=False, onDelete="CASCADE")
432 index.append(collectionFieldSpec.name)
433 # Add foreign key constraint to the collection_summary_dataset_type table.
434 tableSpec.foreignKeys.append(
435 ddl.ForeignKeySpec(
436 "collection_summary_dataset_type",
437 source=(collectionFieldSpec.name, "dataset_type_id"),
438 target=(collectionFieldSpec.name, "dataset_type_id"),
439 )
440 )
441 # Add dimension fields (part of the temporal lookup index.constraint).
442 for dimension_name in datasetType.dimensions.required.names:
443 dimension = datasetType.dimensions.universe.dimensions[dimension_name]
444 fieldSpec = addDimensionForeignKey(tableSpec, dimension=dimension, nullable=False, primaryKey=False)
445 index.append(fieldSpec.name)
446 # If this is a governor dimension, add a foreign key constraint to the
447 # collection_summary_<dimension> table.
448 if isinstance(dimension, GovernorDimension):
449 tableSpec.foreignKeys.append(
450 ddl.ForeignKeySpec(
451 f"collection_summary_{dimension.name}",
452 source=(collectionFieldSpec.name, fieldSpec.name),
453 target=(collectionFieldSpec.name, fieldSpec.name),
454 )
455 )
456 # Add validity-range field(s) (part of the temporal lookup
457 # index/constraint).
458 tsFieldSpecs = TimespanReprClass.makeFieldSpecs(nullable=False)
459 for fieldSpec in tsFieldSpecs:
460 tableSpec.fields.add(fieldSpec)
461 if TimespanReprClass.hasExclusionConstraint(): 461 ↛ 466line 461 didn't jump to line 466, because the condition on line 461 was never true
462 # This database's timespan representation can define a database-level
463 # constraint that prevents overlapping validity ranges for entries with
464 # the same DatasetType, collection, and data ID.
465 # This also creates an index.
466 index.append(TimespanReprClass)
467 tableSpec.exclusion.add(tuple(index))
468 else:
469 # No database-level constraint possible. We'll have to simulate that
470 # in our DatasetRecordStorage.certify() implementation, and just create
471 # a regular index here in the hope that helps with lookups.
472 index.extend(fieldSpec.name for fieldSpec in tsFieldSpecs)
473 tableSpec.indexes.add(ddl.IndexSpec(*index)) # type: ignore
474 return tableSpec