Coverage for python/lsst/daf/butler/registry/dimensions/governor.py: 95%

82 statements  

« prev     ^ index     » next       coverage.py v6.5.0, created at 2022-11-19 01:58 -0800

1# This file is part of daf_butler. 

2# 

3# Developed for the LSST Data Management System. 

4# This product includes software developed by the LSST Project 

5# (http://www.lsst.org). 

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

7# for details of code ownership. 

8# 

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

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

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

12# (at your option) any later version. 

13# 

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

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

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

17# GNU General Public License for more details. 

18# 

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

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

21from __future__ import annotations 

22 

23__all__ = ["BasicGovernorDimensionRecordStorage"] 

24 

25from collections.abc import Callable, Iterable, Mapping, Set 

26from typing import Any 

27 

28import sqlalchemy 

29 

30from ...core import ( 

31 DataCoordinateIterable, 

32 DimensionElement, 

33 DimensionRecord, 

34 GovernorDimension, 

35 NamedKeyDict, 

36 TimespanDatabaseRepresentation, 

37) 

38from ..interfaces import Database, GovernorDimensionRecordStorage, StaticTablesContext 

39from ..queries import QueryBuilder 

40 

41 

42class BasicGovernorDimensionRecordStorage(GovernorDimensionRecordStorage): 

43 """A record storage implementation for `GovernorDimension` that 

44 aggressively fetches and caches all values from the database. 

45 

46 Parameters 

47 ---------- 

48 db : `Database` 

49 Interface to the database engine and namespace that will hold these 

50 dimension records. 

51 dimension : `GovernorDimension` 

52 The dimension whose records this storage will manage. 

53 table : `sqlalchemy.schema.Table` 

54 The logical table for the dimension. 

55 """ 

56 

57 def __init__(self, db: Database, dimension: GovernorDimension, table: sqlalchemy.schema.Table): 

58 self._db = db 

59 self._dimension = dimension 

60 self._table = table 

61 self._cache: dict[str, DimensionRecord] = {} 

62 self._callbacks: list[Callable[[DimensionRecord], None]] = [] 

63 

64 @classmethod 

65 def initialize( 

66 cls, 

67 db: Database, 

68 element: GovernorDimension, 

69 *, 

70 context: StaticTablesContext | None = None, 

71 config: Mapping[str, Any], 

72 ) -> GovernorDimensionRecordStorage: 

73 # Docstring inherited from GovernorDimensionRecordStorage. 

74 spec = element.RecordClass.fields.makeTableSpec( 

75 TimespanReprClass=db.getTimespanRepresentation(), 

76 ) 

77 if context is not None: 77 ↛ 80line 77 didn't jump to line 80, because the condition on line 77 was never false

78 table = context.addTable(element.name, spec) 

79 else: 

80 table = db.ensureTableExists(element.name, spec) 

81 return cls(db, element, table) 

82 

83 @property 

84 def element(self) -> GovernorDimension: 

85 # Docstring inherited from DimensionRecordStorage.element. 

86 return self._dimension 

87 

88 def refresh(self) -> None: 

89 # Docstring inherited from GovernorDimensionRecordStorage. 

90 RecordClass = self._dimension.RecordClass 

91 sql = sqlalchemy.sql.select( 

92 *[self._table.columns[name] for name in RecordClass.fields.standard.names] 

93 ).select_from(self._table) 

94 cache: dict[str, DimensionRecord] = {} 

95 for row in self._db.query(sql): 

96 record = RecordClass(**row._asdict()) 

97 cache[getattr(record, self._dimension.primaryKey.name)] = record 

98 self._cache = cache 

99 

100 @property 

101 def values(self) -> Set[str]: 

102 # Docstring inherited from GovernorDimensionRecordStorage. 

103 return self._cache.keys() 

104 

105 @property 

106 def table(self) -> sqlalchemy.schema.Table: 

107 return self._table 

108 

109 def registerInsertionListener(self, callback: Callable[[DimensionRecord], None]) -> None: 

110 # Docstring inherited from GovernorDimensionRecordStorage. 

111 self._callbacks.append(callback) 

112 

113 def clearCaches(self) -> None: 

114 # Docstring inherited from DimensionRecordStorage.clearCaches. 

115 self._cache.clear() 

116 

117 def join( 

118 self, 

119 builder: QueryBuilder, 

120 *, 

121 regions: NamedKeyDict[DimensionElement, sqlalchemy.sql.ColumnElement] | None = None, 

122 timespans: NamedKeyDict[DimensionElement, TimespanDatabaseRepresentation] | None = None, 

123 ) -> None: 

124 # Docstring inherited from DimensionRecordStorage. 

125 joinOn = builder.startJoin( 

126 self._table, self.element.dimensions, self.element.RecordClass.fields.dimensions.names 

127 ) 

128 builder.finishJoin(self._table, joinOn) 

129 return self._table 

130 

131 def insert(self, *records: DimensionRecord, replace: bool = False, skip_existing: bool = False) -> None: 

132 # Docstring inherited from DimensionRecordStorage.insert. 

133 elementRows = [record.toDict() for record in records] 

134 with self._db.transaction(): 

135 if replace: 135 ↛ 136line 135 didn't jump to line 136, because the condition on line 135 was never true

136 self._db.replace(self._table, *elementRows) 

137 elif skip_existing: 

138 self._db.ensure(self._table, *elementRows, primary_key_only=True) 

139 else: 

140 self._db.insert(self._table, *elementRows) 

141 for record in records: 

142 # We really shouldn't ever get into a situation where the record 

143 # here differs from the one in the DB, but the last thing we want 

144 # is to make it harder to debug by making the cache different from 

145 # the DB. 

146 if skip_existing: 

147 self._cache.setdefault(getattr(record, self.element.primaryKey.name), record) 

148 else: 

149 self._cache[getattr(record, self.element.primaryKey.name)] = record 

150 for callback in self._callbacks: 

151 callback(record) 

152 

153 def sync(self, record: DimensionRecord, update: bool = False) -> bool | dict[str, Any]: 

154 # Docstring inherited from DimensionRecordStorage.sync. 

155 compared = record.toDict() 

156 keys = {} 

157 for name in record.fields.required.names: 

158 keys[name] = compared.pop(name) 

159 with self._db.transaction(): 

160 _, inserted_or_updated = self._db.sync( 

161 self._table, 

162 keys=keys, 

163 compared=compared, 

164 update=update, 

165 ) 

166 if inserted_or_updated: 166 ↛ 170line 166 didn't jump to line 170, because the condition on line 166 was never false

167 self._cache[getattr(record, self.element.primaryKey.name)] = record 

168 for callback in self._callbacks: 

169 callback(record) 

170 return inserted_or_updated 

171 

172 def fetch(self, dataIds: DataCoordinateIterable) -> Iterable[DimensionRecord]: 

173 # Docstring inherited from DimensionRecordStorage.fetch. 

174 try: 

175 return [self._cache[dataId[self.element]] for dataId in dataIds] # type: ignore 

176 except KeyError: 

177 pass 

178 # If at first we don't succeed, refresh and try again. But this time 

179 # we use dict.get to return None if we don't find something. 

180 self.refresh() 

181 return [self._cache.get(dataId[self.element]) for dataId in dataIds] # type: ignore 

182 

183 def digestTables(self) -> Iterable[sqlalchemy.schema.Table]: 

184 # Docstring inherited from DimensionRecordStorage.digestTables. 

185 return [self._table]