Coverage for python/lsst/pipe/base/caching_limited_butler.py: 22%

1# This file is part of pipe_base. 

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/>. 

27 

28from __future__ import annotations 

29 

30__all__ = ["CachingLimitedButler"] 

31 

32import logging 

33from collections.abc import Set 

34from typing import Any, Iterable 

35 

36from lsst.daf.butler import ( 

37 DatasetId, 

38 DatasetRef, 

39 DeferredDatasetHandle, 

40 DimensionUniverse, 

41 LimitedButler, 

42 StorageClass, 

43) 

44 

45from ._dataset_handle import InMemoryDatasetHandle 

46 

47_LOG = logging.getLogger(__name__) 

48 

49 

50class CachingLimitedButler(LimitedButler): 

51 """A `LimitedButler` that caches datasets. 

52 

53 A `CachingLimitedButler` caches on both `.put()` and `.get()`, and holds a 

54 single instance of the most recently used dataset type for that put/get. 

55 

56 The dataset types which will be cached on put/get are controlled via the 

57 `cache_on_put` and `cache_on_get` attributes, respectively. 

58 

59 By default, copies of the cached items are returned on `get`, so that code 

60 is free to operate on data in-place. A `no_copy_on_cache` attribute also 

61 exists to tell the `CachingLimitedButler` not to return copies when it is 

62 known that the calling code can be trusted not to change values, e.g. when 

63 passing calibs to `isrTask`. 

64 

65 Parameters 

66 ---------- 

67 wrapped : `LimitedButler` 

68 The butler to wrap. 

69 cache_on_put : `set` [`str`], optional 

70 The dataset types to cache on put. 

71 cache_on_get : `set` [`str`], optional 

72 The dataset types to cache on get. 

73 no_copy_on_cache : `set` [`str`], optional 

74 The dataset types for which to not return copies when cached. 

75 """ 

76 

77 def __init__( 

78 self, 

79 wrapped: LimitedButler, 

80 cache_on_put: Set[str] = frozenset(), 

81 cache_on_get: Set[str] = frozenset(), 

82 no_copy_on_cache: Set[str] = frozenset(), 

83 ): 

84 self._wrapped = wrapped 

85 self._datastore = self._wrapped._datastore 

86 self.storageClasses = self._wrapped.storageClasses 

87 self._cache_on_put = cache_on_put 

88 self._cache_on_get = cache_on_get 

89 self._cache: dict[str, tuple[DatasetId, InMemoryDatasetHandle]] = {} 

90 self._no_copy_on_cache = no_copy_on_cache 

91 

92 def get( 

93 self, 

94 ref: DatasetRef, 

95 /, 

96 *, 

97 parameters: dict[str, Any] | None = None, 

98 storageClass: StorageClass | str | None = None, 

99 ) -> Any: 

100 if storageClass is None: 

101 storageClass = ref.datasetType.storageClass 

102 elif isinstance(storageClass, str): 

103 storageClass = self.storageClasses.getStorageClass(storageClass) 

104 

105 # check if we have this dataset type in the cache 

106 if cached := self._cache.get(ref.datasetType.name): 

107 dataset_id, handle = cached 

108 if dataset_id == ref.id: # if we do, check it's the right object 

109 _LOG.debug("Returning cached dataset %s", ref) 

110 return handle.get(parameters=parameters, storageClass=storageClass) 

111 

112 obj = self._wrapped.get(ref, parameters=parameters, storageClass=storageClass) 

113 if ref.datasetType.name in self._cache_on_get and not parameters: 

114 handle = InMemoryDatasetHandle( 

115 obj, 

116 storageClass=storageClass, 

117 dataId=ref.dataId, 

118 copy=ref.datasetType.name not in self._no_copy_on_cache, 

119 ) 

120 # and not parameters is to make sure we don't cache sub-images etc 

121 self._cache[ref.datasetType.name] = (ref.id, handle) 

122 _LOG.debug("Cached dataset %s", ref) 

123 # make sure copy fires if needed 

124 return handle.get() 

125 return obj 

126 

127 def getDeferred( 

128 self, 

129 ref: DatasetRef, 

130 /, 

131 *, 

132 parameters: dict[str, Any] | None = None, 

133 storageClass: str | StorageClass | None = None, 

134 ) -> DeferredDatasetHandle: 

135 # note that this does not use the cache at all 

136 return self._wrapped.getDeferred(ref, parameters=parameters, storageClass=storageClass) 

137 

138 def stored(self, ref: DatasetRef) -> bool: 

139 return self.stored_many([ref])[ref] # TODO: remove this once DM-43086 is done. 

140 

141 def stored_many(self, refs: Iterable[DatasetRef]) -> dict[DatasetRef, bool]: 

142 result = {} 

143 unknown_refs = [] 

144 for ref in refs: 

145 if cached := self._cache.get(ref.datasetType.name): 

146 dataset_id, _ = cached 

147 if dataset_id == ref.id: 

148 result[ref] = True 

149 continue 

150 unknown_refs.append(ref) 

151 

152 result.update(self._wrapped.stored_many(unknown_refs)) 

153 return result 

154 

155 def isWriteable(self) -> bool: 

156 return self._wrapped.isWriteable() 

157 

158 def put(self, obj: Any, ref: DatasetRef) -> DatasetRef: 

159 if ref.datasetType.name in self._cache_on_put: 

160 self._cache[ref.datasetType.name] = ( 

161 ref.id, 

162 InMemoryDatasetHandle( 

163 obj, 

164 storageClass=ref.datasetType.storageClass, 

165 dataId=ref.dataId, 

166 copy=ref.datasetType.name not in self._no_copy_on_cache, 

167 ), 

168 ) 

169 _LOG.debug("Cached dataset %s on put", ref) 

170 return self._wrapped.put(obj, ref) 

171 

172 def pruneDatasets( 

173 self, 

174 refs: Iterable[DatasetRef], 

175 *, 

176 disassociate: bool = True, 

177 unstore: bool = False, 

178 tags: Iterable[str] = (), 

179 purge: bool = False, 

180 ) -> None: 

181 refs = list(refs) 

182 for ref in refs: 

183 if cached := self._cache.get(ref.datasetType.name): 

184 dataset_id, _ = cached 

185 if dataset_id == ref.id: 

186 del self._cache[ref.datasetType.name] 

187 

188 return self._wrapped.pruneDatasets( 

189 refs, disassociate=disassociate, unstore=unstore, tags=tags, purge=purge 

190 ) 

191 

192 @property 

193 def dimensions(self) -> DimensionUniverse: 

194 return self._wrapped.dimensions