Coverage for python/lsst/daf/butler/core/dimensions/_records.py: 24%

161 statements  

« prev     ^ index     » next       coverage.py v7.2.7, created at 2023-08-12 09:20 +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__ = ("DimensionRecord", "SerializedDimensionRecord") 

25 

26from collections.abc import Hashable 

27from typing import TYPE_CHECKING, Any, ClassVar, Optional, Tuple 

28 

29import lsst.sphgeom 

30from lsst.daf.butler._compat import PYDANTIC_V2, _BaseModelCompat 

31from lsst.utils.classes import immutable 

32from pydantic import Field, StrictBool, StrictFloat, StrictInt, StrictStr, create_model 

33 

34from ..json import from_json_pydantic, to_json_pydantic 

35from ..persistenceContext import PersistenceContextVars 

36from ..timespan import Timespan, TimespanDatabaseRepresentation 

37from ._elements import Dimension, DimensionElement 

38 

39if TYPE_CHECKING: # Imports needed only for type annotations; may be circular. 

40 from ...registry import Registry 

41 from ._coordinate import DataCoordinate 

42 from ._graph import DimensionUniverse 

43 from ._schema import DimensionElementFields 

44 

45 

46def _reconstructDimensionRecord(definition: DimensionElement, mapping: dict[str, Any]) -> DimensionRecord: 

47 """Unpickle implementation for `DimensionRecord` subclasses. 

48 

49 For internal use by `DimensionRecord`. 

50 """ 

51 return definition.RecordClass(**mapping) 

52 

53 

54def _subclassDimensionRecord(definition: DimensionElement) -> type[DimensionRecord]: 

55 """Create a dynamic subclass of `DimensionRecord` for the given element. 

56 

57 For internal use by `DimensionRecord`. 

58 """ 

59 from ._schema import DimensionElementFields 

60 

61 fields = DimensionElementFields(definition) 

62 slots = list(fields.standard.names) 

63 if definition.spatial: 

64 slots.append("region") 

65 if definition.temporal: 

66 slots.append(TimespanDatabaseRepresentation.NAME) 

67 d = {"definition": definition, "__slots__": tuple(slots), "fields": fields} 

68 return type(definition.name + ".RecordClass", (DimensionRecord,), d) 

69 

70 

71class SpecificSerializedDimensionRecord(_BaseModelCompat, extra="forbid"): 

72 """Base model for a specific serialized record content.""" 

73 

74 

75_SIMPLE_RECORD_CLASS_CACHE: dict[ 

76 tuple[DimensionElement, DimensionUniverse], type[SpecificSerializedDimensionRecord] 

77] = {} 

78 

79 

80def _createSimpleRecordSubclass(definition: DimensionElement) -> type[SpecificSerializedDimensionRecord]: 

81 from ._schema import DimensionElementFields 

82 

83 # Cache on the definition (which hashes as the name) and the 

84 # associated universe. 

85 cache_key = (definition, definition.universe) 

86 if cache_key in _SIMPLE_RECORD_CLASS_CACHE: 

87 return _SIMPLE_RECORD_CLASS_CACHE[cache_key] 

88 

89 fields = DimensionElementFields(definition) 

90 members = {} 

91 # Prefer strict typing for external data 

92 type_map = { 

93 str: StrictStr, 

94 float: StrictFloat, 

95 bool: StrictBool, 

96 int: StrictInt, 

97 } 

98 

99 for field in fields.standard: 

100 field_type = field.getPythonType() 

101 field_type = type_map.get(field_type, field_type) 

102 if field.nullable: 

103 field_type = Optional[field_type] # type: ignore 

104 members[field.name] = (field_type, ...) 

105 if definition.temporal: 

106 members["timespan"] = (Tuple[int, int], ...) # type: ignore 

107 if definition.spatial: 

108 members["region"] = (str, ...) 

109 

110 # mypy does not seem to like create_model 

111 model = create_model( 

112 f"SpecificSerializedDimensionRecord{definition.name.capitalize()}", 

113 __base__=SpecificSerializedDimensionRecord, 

114 **members, # type: ignore 

115 ) 

116 

117 _SIMPLE_RECORD_CLASS_CACHE[cache_key] = model 

118 return model 

119 

120 

121# While supporting pydantic v1 and v2 keep this outside the model. 

122_serialized_dimension_record_schema_extra = { 

123 "examples": [ 

124 { 

125 "definition": "detector", 

126 "record": { 

127 "instrument": "HSC", 

128 "id": 72, 

129 "full_name": "0_01", 

130 "name_in_raft": "01", 

131 "raft": "0", 

132 "purpose": "SCIENCE", 

133 }, 

134 } 

135 ] 

136} 

137 

138 

139class SerializedDimensionRecord(_BaseModelCompat): 

140 """Simplified model for serializing a `DimensionRecord`.""" 

141 

142 definition: str = Field( 

143 ..., 

144 title="Name of dimension associated with this record.", 

145 examples=["exposure"], 

146 ) 

147 

148 # Use strict types to prevent casting 

149 record: dict[str, None | StrictFloat | StrictStr | StrictBool | StrictInt | tuple[int, int]] = Field( 

150 ..., 

151 title="Dimension record keys and values.", 

152 examples=[ 

153 { 

154 "definition": "exposure", 

155 "record": { 

156 "instrument": "LATISS", 

157 "exposure": 2021050300044, 

158 "obs_id": "AT_O_20210503_00044", 

159 }, 

160 } 

161 ], 

162 ) 

163 

164 if PYDANTIC_V2: 164 ↛ 165line 164 didn't jump to line 165

165 model_config = { 

166 "json_schema_extra": _serialized_dimension_record_schema_extra, # type: ignore[typeddict-item] 

167 } 

168 else: 

169 

170 class Config: 

171 """Local configuration overrides for model.""" 

172 

173 schema_extra = _serialized_dimension_record_schema_extra 

174 

175 @classmethod 

176 def direct( 

177 cls, 

178 *, 

179 definition: str, 

180 record: dict[str, None | StrictFloat | StrictStr | StrictBool | StrictInt | tuple[int, int]], 

181 ) -> SerializedDimensionRecord: 

182 """Construct a `SerializedDimensionRecord` directly without validators. 

183 

184 This differs from the pydantic "construct" method in that the arguments 

185 are explicitly what the model requires, and it will recurse through 

186 members, constructing them from their corresponding `direct` methods. 

187 

188 This method should only be called when the inputs are trusted. 

189 """ 

190 # This method requires tuples as values of the mapping, but JSON 

191 # readers will read things in as lists. Be kind and transparently 

192 # transform to tuples 

193 _recItems = {k: v if type(v) != list else tuple(v) for k, v in record.items()} # type: ignore 

194 

195 # Type ignore because the ternary statement seems to confuse mypy 

196 # based on conflicting inferred types of v. 

197 key = ( 

198 definition, 

199 frozenset(_recItems.items()), 

200 ) 

201 cache = PersistenceContextVars.serializedDimensionRecordMapping.get() 

202 if cache is not None and (result := cache.get(key)) is not None: 

203 return result 

204 

205 node = cls.model_construct(definition=definition, record=_recItems) # type: ignore 

206 

207 if cache is not None: 

208 cache[key] = node 

209 return node 

210 

211 

212@immutable 

213class DimensionRecord: 

214 """Base class for the Python representation of database records. 

215 

216 Parameters 

217 ---------- 

218 **kwargs 

219 Field values for this record. Unrecognized keys are ignored. If this 

220 is the record for a `Dimension`, its primary key value may be provided 

221 with the actual name of the field (e.g. "id" or "name"), the name of 

222 the `Dimension`, or both. If this record class has a "timespan" 

223 attribute, "datetime_begin" and "datetime_end" keyword arguments may 

224 be provided instead of a single "timespan" keyword argument (but are 

225 ignored if a "timespan" argument is provided). 

226 

227 Notes 

228 ----- 

229 `DimensionRecord` subclasses are created dynamically for each 

230 `DimensionElement` in a `DimensionUniverse`, and are accessible via the 

231 `DimensionElement.RecordClass` attribute. The `DimensionRecord` base class 

232 itself is pure abstract, but does not use the `abc` module to indicate this 

233 because it does not have overridable methods. 

234 

235 Record classes have attributes that correspond exactly to the 

236 `~DimensionElementFields.standard` fields in the related database table, 

237 plus "region" and "timespan" attributes for spatial and/or temporal 

238 elements (respectively). 

239 

240 Instances are usually obtained from a `Registry`, but can be constructed 

241 directly from Python as well. 

242 

243 `DimensionRecord` instances are immutable. 

244 """ 

245 

246 # Derived classes are required to define __slots__ as well, and it's those 

247 # derived-class slots that other methods on the base class expect to see 

248 # when they access self.__slots__. 

249 __slots__ = ("dataId",) 

250 

251 _serializedType = SerializedDimensionRecord 

252 

253 def __init__(self, **kwargs: Any): 

254 # Accept either the dimension name or the actual name of its primary 

255 # key field; ensure both are present in the dict for convenience below. 

256 if isinstance(self.definition, Dimension): 

257 v = kwargs.get(self.definition.primaryKey.name) 

258 if v is None: 

259 v = kwargs.get(self.definition.name) 

260 if v is None: 

261 raise ValueError( 

262 f"No value provided for {self.definition.name}.{self.definition.primaryKey.name}." 

263 ) 

264 kwargs[self.definition.primaryKey.name] = v 

265 else: 

266 v2 = kwargs.setdefault(self.definition.name, v) 

267 if v != v2: 

268 raise ValueError( 

269 "Multiple inconsistent values for " 

270 f"{self.definition.name}.{self.definition.primaryKey.name}: {v!r} != {v2!r}." 

271 ) 

272 for name in self.__slots__: 

273 object.__setattr__(self, name, kwargs.get(name)) 

274 if self.definition.temporal is not None and self.timespan is None: 

275 object.__setattr__( 

276 self, 

277 "timespan", 

278 Timespan( 

279 kwargs.get("datetime_begin"), 

280 kwargs.get("datetime_end"), 

281 ), 

282 ) 

283 

284 from ._coordinate import DataCoordinate 

285 

286 object.__setattr__( 

287 self, 

288 "dataId", 

289 DataCoordinate.fromRequiredValues( 

290 self.definition.graph, 

291 tuple(kwargs[dimension] for dimension in self.definition.required.names), 

292 ), 

293 ) 

294 

295 def __eq__(self, other: Any) -> bool: 

296 if type(other) != type(self): 

297 return False 

298 return self.dataId == other.dataId 

299 

300 def __hash__(self) -> int: 

301 return hash(self.dataId) 

302 

303 def __str__(self) -> str: 

304 lines = [f"{self.definition.name}:"] 

305 lines.extend(f" {name}: {getattr(self, name)!r}" for name in self.__slots__) 

306 return "\n".join(lines) 

307 

308 def __repr__(self) -> str: 

309 return "{}.RecordClass({})".format( 

310 self.definition.name, ", ".join(f"{name}={getattr(self, name)!r}" for name in self.__slots__) 

311 ) 

312 

313 def __reduce__(self) -> tuple: 

314 mapping = {name: getattr(self, name) for name in self.__slots__} 

315 return (_reconstructDimensionRecord, (self.definition, mapping)) 

316 

317 def _repr_html_(self) -> str: 

318 """Override the default representation in IPython/Jupyter notebooks. 

319 

320 This gives a more readable output that understands embedded newlines. 

321 """ 

322 return f"<pre>{self}<pre>" 

323 

324 def to_simple(self, minimal: bool = False) -> SerializedDimensionRecord: 

325 """Convert this class to a simple python type. 

326 

327 This makes it suitable for serialization. 

328 

329 Parameters 

330 ---------- 

331 minimal : `bool`, optional 

332 Use minimal serialization. Has no effect on for this class. 

333 

334 Returns 

335 ------- 

336 names : `list` 

337 The names of the dimensions. 

338 """ 

339 # The DataId is sufficient if you are willing to do a deferred 

340 # query. This may not be overly useful since to reconstruct 

341 # a collection of records will require repeated registry queries. 

342 # For now do not implement minimal form. 

343 

344 mapping = {name: getattr(self, name) for name in self.__slots__} 

345 # If the item in mapping supports simplification update it 

346 for k, v in mapping.items(): 

347 try: 

348 mapping[k] = v.to_simple(minimal=minimal) 

349 except AttributeError: 

350 if isinstance(v, lsst.sphgeom.Region): 

351 # YAML serialization specifies the class when it 

352 # doesn't have to. This is partly for explicitness 

353 # and also history. Here use a different approach. 

354 # This code needs to be migrated to sphgeom 

355 mapping[k] = v.encode().hex() 

356 if isinstance(v, bytes): 

357 # We actually can't handle serializing out to bytes for 

358 # hash objects, encode it here to a hex string 

359 mapping[k] = v.hex() 

360 definition = self.definition.to_simple(minimal=minimal) 

361 return SerializedDimensionRecord(definition=definition, record=mapping) 

362 

363 @classmethod 

364 def from_simple( 

365 cls, 

366 simple: SerializedDimensionRecord, 

367 universe: DimensionUniverse | None = None, 

368 registry: Registry | None = None, 

369 cacheKey: Hashable | None = None, 

370 ) -> DimensionRecord: 

371 """Construct a new object from the simplified form. 

372 

373 This is generally data returned from the `to_simple` 

374 method. 

375 

376 Parameters 

377 ---------- 

378 simple : `SerializedDimensionRecord` 

379 Value return from `to_simple`. 

380 universe : `DimensionUniverse` 

381 The special graph of all known dimensions of which this graph will 

382 be a subset. Can be `None` if `Registry` is provided. 

383 registry : `lsst.daf.butler.Registry`, optional 

384 Registry from which a universe can be extracted. Can be `None` 

385 if universe is provided explicitly. 

386 cacheKey : `Hashable` or `None` 

387 If this is not None, it will be used as a key for any cached 

388 reconstruction instead of calculating a value from the serialized 

389 format. 

390 

391 Returns 

392 ------- 

393 record : `DimensionRecord` 

394 Newly-constructed object. 

395 """ 

396 if universe is None and registry is None: 

397 raise ValueError("One of universe or registry is required to convert names to a DimensionGraph") 

398 if universe is None and registry is not None: 

399 universe = registry.dimensions 

400 if universe is None: 

401 # this is for mypy 

402 raise ValueError("Unable to determine a usable universe") 

403 # Type ignore because the ternary statement seems to confuse mypy 

404 # based on conflicting inferred types of v. 

405 key = cacheKey or ( 

406 simple.definition, 

407 frozenset(simple.record.items()), # type: ignore 

408 ) 

409 cache = PersistenceContextVars.dimensionRecords.get() 

410 if cache is not None and (result := cache.get(key)) is not None: 

411 return result 

412 

413 definition = DimensionElement.from_simple(simple.definition, universe=universe) 

414 

415 # Create a specialist subclass model with type validation. 

416 # This allows us to do simple checks of external data (possibly 

417 # sent as JSON) since for now _reconstructDimensionRecord does not 

418 # do any validation. 

419 record_model_cls = _createSimpleRecordSubclass(definition) 

420 record_model = record_model_cls(**simple.record) 

421 

422 # Timespan and region have to be converted to native form 

423 # for now assume that those keys are special 

424 rec = record_model.model_dump() 

425 

426 if (ts := "timespan") in rec: 

427 rec[ts] = Timespan.from_simple(rec[ts], universe=universe, registry=registry) 

428 if (reg := "region") in rec: 

429 encoded = bytes.fromhex(rec[reg]) 

430 rec[reg] = lsst.sphgeom.Region.decode(encoded) 

431 if (hsh := "hash") in rec: 

432 rec[hsh] = bytes.fromhex(rec[hsh].decode()) 

433 

434 dimRec = _reconstructDimensionRecord(definition, rec) 

435 if cache is not None: 

436 cache[key] = dimRec 

437 return dimRec 

438 

439 to_json = to_json_pydantic 

440 from_json: ClassVar = classmethod(from_json_pydantic) 

441 

442 def toDict(self, splitTimespan: bool = False) -> dict[str, Any]: 

443 """Return a vanilla `dict` representation of this record. 

444 

445 Parameters 

446 ---------- 

447 splitTimespan : `bool`, optional 

448 If `True` (`False` is default) transform any "timespan" key value 

449 from a `Timespan` instance into a pair of regular 

450 ("datetime_begin", "datetime_end") fields. 

451 """ 

452 results = {name: getattr(self, name) for name in self.__slots__} 

453 if splitTimespan: 

454 timespan = results.pop("timespan", None) 

455 if timespan is not None: 

456 results["datetime_begin"] = timespan.begin 

457 results["datetime_end"] = timespan.end 

458 return results 

459 

460 # DimensionRecord subclasses are dynamically created, so static type 

461 # checkers can't know about them or their attributes. To avoid having to 

462 # put "type: ignore", everywhere, add a dummy __getattr__ that tells type 

463 # checkers not to worry about missing attributes. 

464 def __getattr__(self, name: str) -> Any: 

465 raise AttributeError(name) 

466 

467 # Class attributes below are shadowed by instance attributes, and are 

468 # present just to hold the docstrings for those instance attributes. 

469 

470 dataId: DataCoordinate 

471 """A dict-like identifier for this record's primary keys 

472 (`DataCoordinate`). 

473 """ 

474 

475 definition: ClassVar[DimensionElement] 

476 """The `DimensionElement` whose records this class represents 

477 (`DimensionElement`). 

478 """ 

479 

480 fields: ClassVar[DimensionElementFields] 

481 """A categorized view of the fields in this class 

482 (`DimensionElementFields`). 

483 """