Hide keyboard shortcuts

Hot-keys on this page

r m x p   toggle line displays

j k   next/prev highlighted chunk

0   (zero) top of page

1   (one) first highlighted chunk

1# This file is part of daf_butler. 

2# 

3# Developed for the LSST Data Management System. 

4# This product includes software developed by the LSST Project 

5# (http://www.lsst.org). 

6# See the COPYRIGHT file at the top-level directory of this distribution 

7# for details of code ownership. 

8# 

9# This program is free software: you can redistribute it and/or modify 

10# it under the terms of the GNU General Public License as published by 

11# the Free Software Foundation, either version 3 of the License, or 

12# (at your option) any later version. 

13# 

14# This program is distributed in the hope that it will be useful, 

15# but WITHOUT ANY WARRANTY; without even the implied warranty of 

16# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 

17# GNU General Public License for more details. 

18# 

19# You should have received a copy of the GNU General Public License 

20# along with this program. If not, see <http://www.gnu.org/licenses/>. 

21from __future__ import annotations 

22 

23__all__ = ["QuerySummary", "RegistryManagers"] # other classes here are local to subpackage 

24 

25from dataclasses import dataclass 

26from typing import Iterator, List, Optional, Union 

27 

28from sqlalchemy.sql import ColumnElement 

29 

30from lsst.sphgeom import Region 

31from ...core import ( 

32 DataCoordinate, 

33 DatasetType, 

34 Dimension, 

35 DimensionElement, 

36 DimensionGraph, 

37 DimensionUniverse, 

38 NamedKeyDict, 

39 NamedValueSet, 

40 SkyPixDimension, 

41 Timespan, 

42) 

43from ..interfaces import ( 

44 CollectionManager, 

45 DatasetRecordStorageManager, 

46 DimensionRecordStorageManager, 

47) 

48# We're not trying to add typing to the lex/yacc parser code, so MyPy 

49# doesn't know about some of these imports. 

50from .exprParser import Node, ParserYacc # type: ignore 

51 

52 

53@dataclass 

54class QueryWhereExpression: 

55 """A struct representing a parsed user-provided WHERE expression. 

56 

57 Parameters 

58 ---------- 

59 universe : `DimensionUniverse` 

60 All known dimensions. 

61 expression : `str`, optional 

62 The string expression to parse. 

63 """ 

64 def __init__(self, universe: DimensionUniverse, expression: Optional[str] = None): 

65 if expression: 

66 from .expressions import InspectionVisitor 

67 try: 

68 parser = ParserYacc() 

69 self.tree = parser.parse(expression) 

70 except Exception as exc: 

71 raise RuntimeError(f"Failed to parse user expression `{expression}'.") from exc 

72 visitor = InspectionVisitor(universe) 

73 assert self.tree is not None 

74 self.tree.visit(visitor) 

75 self.keys = visitor.keys 

76 self.metadata = visitor.metadata 

77 else: 

78 self.tree = None 

79 self.keys = NamedValueSet() 

80 self.metadata = NamedKeyDict() 

81 

82 tree: Optional[Node] 

83 """The parsed user expression tree, if present (`Node` or `None`). 

84 """ 

85 

86 keys: NamedValueSet[Dimension] 

87 """All dimensions whose keys are referenced by the expression 

88 (`NamedValueSet` of `Dimension`). 

89 """ 

90 

91 metadata: NamedKeyDict[DimensionElement, List[str]] 

92 """All dimension elements metadata fields referenced by the expression 

93 (`NamedKeyDict` mapping `DimensionElement` to a `set` of field names). 

94 """ 

95 

96 

97@dataclass 

98class QuerySummary: 

99 """A struct that holds and categorizes the dimensions involved in a query. 

100 

101 A `QuerySummary` instance is necessary to construct a `QueryBuilder`, and 

102 it needs to include all of the dimensions that will be included in the 

103 query (including any needed for querying datasets). 

104 

105 Parameters 

106 ---------- 

107 requested : `DimensionGraph` 

108 The dimensions whose primary keys should be included in the result rows 

109 of the query. 

110 dataId : `DataCoordinate`, optional 

111 A fully-expanded data ID identifying dimensions known in advance. If 

112 not provided, will be set to an empty data ID. ``dataId.hasRecords()`` 

113 must return `True`. 

114 expression : `str` or `QueryWhereExpression`, optional 

115 A user-provided string WHERE expression. 

116 whereRegion : `lsst.sphgeom.Region`, optional 

117 A spatial region that all rows must overlap. If `None` and ``dataId`` 

118 is not `None`, ``dataId.region`` will be used. 

119 """ 

120 def __init__(self, requested: DimensionGraph, *, 

121 dataId: Optional[DataCoordinate] = None, 

122 expression: Optional[Union[str, QueryWhereExpression]] = None, 

123 whereRegion: Optional[Region] = None): 

124 self.requested = requested 

125 self.dataId = dataId if dataId is not None else DataCoordinate.makeEmpty(requested.universe) 

126 self.expression = (expression if isinstance(expression, QueryWhereExpression) 

127 else QueryWhereExpression(requested.universe, expression)) 

128 if whereRegion is None and self.dataId is not None: 

129 whereRegion = self.dataId.region 

130 self.whereRegion = whereRegion 

131 

132 requested: DimensionGraph 

133 """Dimensions whose primary keys should be included in the result rows of 

134 the query (`DimensionGraph`). 

135 """ 

136 

137 dataId: DataCoordinate 

138 """A data ID identifying dimensions known before query construction 

139 (`DataCoordinate`). 

140 

141 ``dataId.hasRecords()`` is guaranteed to return `True`. 

142 """ 

143 

144 whereRegion: Optional[Region] 

145 """A spatial region that all result rows must overlap 

146 (`lsst.sphgeom.Region` or `None`). 

147 """ 

148 

149 expression: QueryWhereExpression 

150 """Information about any parsed user WHERE expression 

151 (`QueryWhereExpression`). 

152 """ 

153 

154 @property 

155 def universe(self) -> DimensionUniverse: 

156 """All known dimensions (`DimensionUniverse`). 

157 """ 

158 return self.requested.universe 

159 

160 @property 

161 def spatial(self) -> NamedValueSet[DimensionElement]: 

162 """Dimension elements whose regions and skypix IDs should be included 

163 in the query (`NamedValueSet` of `DimensionElement`). 

164 """ 

165 # An element may participate spatially in the query if: 

166 # - it's the most precise spatial element for its system in the 

167 # requested dimensions (i.e. in `self.requested.spatial`); 

168 # - it isn't also given at query construction time. 

169 result = NamedValueSet(self.mustHaveKeysJoined.spatial - self.dataId.graph.elements) 

170 if len(result) == 1: 

171 # There's no spatial join, but there might be a WHERE filter based 

172 # on a given region. 

173 if self.dataId.graph.spatial: 

174 # We can only perform those filters against SkyPix dimensions, 

175 # so if what we have isn't one, add the common SkyPix dimension 

176 # to the query; the element we have will be joined to that. 

177 element, = result 

178 if not isinstance(element, SkyPixDimension): 

179 result.add(self.universe.commonSkyPix) 

180 else: 

181 # There is no spatial join or filter in this query. Even 

182 # if this element might be associated with spatial 

183 # information, we don't need it for this query. 

184 return NamedValueSet() 

185 elif len(result) > 1: 

186 # There's a spatial join. Those require the common SkyPix 

187 # system to be included in the query in order to connect them. 

188 result.add(self.universe.commonSkyPix) 

189 return result 

190 

191 @property 

192 def temporal(self) -> NamedValueSet[DimensionElement]: 

193 """Dimension elements whose timespans should be included in the 

194 query (`NamedValueSet` of `DimensionElement`). 

195 """ 

196 # An element may participate temporally in the query if: 

197 # - it's the most precise temporal element for its system in the 

198 # requested dimensions (i.e. in `self.requested.temporal`); 

199 # - it isn't also given at query construction time. 

200 result = NamedValueSet(self.mustHaveKeysJoined.temporal - self.dataId.graph.elements) 

201 if len(result) == 1 and not self.dataId.graph.temporal: 

202 # No temporal join or filter. Even if this element might be 

203 # associated with temporal information, we don't need it for this 

204 # query. 

205 return NamedValueSet() 

206 return result 

207 

208 @property 

209 def mustHaveKeysJoined(self) -> DimensionGraph: 

210 """Dimensions whose primary keys must be used in the JOIN ON clauses 

211 of the query, even if their tables do not appear (`DimensionGraph`). 

212 

213 A `Dimension` primary key can appear in a join clause without its table 

214 via a foreign key column in table of a dependent dimension element or 

215 dataset. 

216 """ 

217 names = set(self.requested.names | self.expression.keys.names) 

218 return DimensionGraph(self.universe, names=names) 

219 

220 @property 

221 def mustHaveTableJoined(self) -> NamedValueSet[DimensionElement]: 

222 """Dimension elements whose associated tables must appear in the 

223 query's FROM clause (`NamedValueSet` of `DimensionElement`). 

224 """ 

225 result = NamedValueSet(self.spatial | self.temporal | self.expression.metadata.keys()) 

226 for dimension in self.mustHaveKeysJoined: 

227 if dimension.implied: 

228 result.add(dimension) 

229 for element in self.mustHaveKeysJoined.union(self.dataId.graph).elements: 

230 if element.alwaysJoin: 

231 result.add(element) 

232 return result 

233 

234 

235@dataclass 

236class DatasetQueryColumns: 

237 """A struct containing the columns used to reconstruct `DatasetRef` 

238 instances from query results. 

239 """ 

240 

241 datasetType: DatasetType 

242 """The dataset type being queried (`DatasetType`). 

243 """ 

244 

245 id: ColumnElement 

246 """Column containing the unique integer ID for this dataset. 

247 """ 

248 

249 runKey: ColumnElement 

250 """Foreign key column to the `~CollectionType.RUN` collection that holds 

251 this dataset. 

252 """ 

253 

254 def __iter__(self) -> Iterator[ColumnElement]: 

255 yield self.id 

256 yield self.runKey 

257 

258 

259@dataclass 

260class QueryColumns: 

261 """A struct organizing the columns in an under-construction or currently- 

262 executing query. 

263 

264 Takes no parameters at construction, as expected usage is to add elements 

265 to its container attributes incrementally. 

266 """ 

267 def __init__(self) -> None: 

268 self.keys = NamedKeyDict() 

269 self.timespans = NamedKeyDict() 

270 self.regions = NamedKeyDict() 

271 self.datasets = None 

272 

273 keys: NamedKeyDict[Dimension, List[ColumnElement]] 

274 """Columns that correspond to the primary key values of dimensions 

275 (`NamedKeyDict` mapping `Dimension` to a `list` of `ColumnElement`). 

276 

277 Each value list contains columns from multiple tables corresponding to the 

278 same dimension, and the query should constrain the values of those columns 

279 to be the same. 

280 

281 In a `Query`, the keys of this dictionary must include at least the 

282 dimensions in `QuerySummary.requested` and `QuerySummary.dataId.graph`. 

283 """ 

284 

285 timespans: NamedKeyDict[DimensionElement, Timespan[ColumnElement]] 

286 """Columns that correspond to timespans for elements that participate in a 

287 temporal join or filter in the query (`NamedKeyDict` mapping 

288 `DimensionElement` to `Timespan` of `ColumnElement`). 

289 

290 In a `Query`, the keys of this dictionary must be exactly the elements 

291 in `QuerySummary.temporal`. 

292 """ 

293 

294 regions: NamedKeyDict[DimensionElement, ColumnElement] 

295 """Columns that correspond to regions for elements that participate in a 

296 spatial join or filter in the query (`NamedKeyDict` mapping 

297 `DimensionElement` to `ColumnElement`). 

298 

299 In a `Query`, the keys of this dictionary must be exactly the elements 

300 in `QuerySummary.spatial`. 

301 """ 

302 

303 datasets: Optional[DatasetQueryColumns] 

304 """Columns that can be used to construct `DatasetRef` instances from query 

305 results. 

306 (`DatasetQueryColumns` or `None`). 

307 """ 

308 

309 def isEmpty(self) -> bool: 

310 """Return `True` if this query has no columns at all. 

311 """ 

312 return not (self.keys or self.timespans or self.regions or self.datasets is not None) 

313 

314 def getKeyColumn(self, dimension: Union[Dimension, str]) -> ColumnElement: 

315 """ Return one of the columns in self.keys for the given dimension. 

316 

317 The column selected is an implentation detail but is guaranteed to 

318 be deterministic and consistent across multiple calls. 

319 

320 Parameters 

321 ---------- 

322 dimension : `Dimension` or `str` 

323 Dimension for which to obtain a key column. 

324 

325 Returns 

326 ------- 

327 column : `sqlalchemy.sql.ColumnElement` 

328 SQLAlchemy column object. 

329 """ 

330 # Choosing the last element here is entirely for human readers of the 

331 # query (e.g. developers debugging things); it makes it more likely a 

332 # dimension key will be provided by the dimension's own table, or 

333 # failing that, some closely related dimension, which might be less 

334 # surprising to see than e.g. some dataset subquery. From the 

335 # database's perspective this is entirely arbitrary, because the query 

336 # guarantees they all have equal values. 

337 return self.keys[dimension][-1] 

338 

339 

340@dataclass 

341class RegistryManagers: 

342 """Struct used to pass around the manager objects that back a `Registry` 

343 and are used internally by the query system. 

344 """ 

345 

346 collections: CollectionManager 

347 """Manager for collections (`CollectionManager`). 

348 """ 

349 

350 datasets: DatasetRecordStorageManager 

351 """Manager for datasets and dataset types (`DatasetRecordStorageManager`). 

352 """ 

353 

354 dimensions: DimensionRecordStorageManager 

355 """Manager for dimensions (`DimensionRecordStorageManager`). 

356 """