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 obs_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 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"""Concrete implementations of `PathElementHandler`. 

22 

23The `PathElementHandler` ABC is defined in ``scanner.py`` instead of here to 

24avoid a circular dependency between modules. 

25""" 

26from __future__ import annotations 

27 

28__all__ = ["IgnoreHandler", "SkipHandler", "SubdirectoryHandler", "TargetFileHandler"] 

29 

30from abc import abstractmethod 

31import re 

32from typing import ( 

33 Callable, 

34 List, 

35 Mapping, 

36 Optional, 

37) 

38 

39from lsst.log import Log 

40from lsst.daf.butler import ( 

41 DataCoordinate, 

42 DatasetRef, 

43 DatasetType, 

44 FileDataset, 

45) 

46from ..translators import Translator 

47from .parser import PathElementParser 

48from .scanner import PathElementHandler, DirectoryScanner 

49 

50 

51class IgnoreHandler(PathElementHandler): 

52 """A `PathElementHandler` that matches via a regular expression, and does 

53 nothing. 

54 

55 An `IgnoreHandler` is used to ignore file or directory patterns that can 

56 occur at any level in the directory tree, and have no relation to any 

57 Gen2 filename template. 

58 

59 Parameters 

60 ---------- 

61 pattern : `re.Pattern` 

62 A regular expression pattern. 

63 isForFiles : `bool` 

64 Whether this handler should be applied to files (`True`) or 

65 directories (`False`). 

66 """ 

67 def __init__(self, pattern: re.Pattern, isForFiles: bool): 

68 super().__init__() 

69 self._pattern = pattern 

70 self._isForFiles = isForFiles 

71 

72 __slots__ = ("_pattern", "_isForFiles") 

73 

74 def isForFiles(self) -> bool: 

75 # Docstring inherited from PathElementHandler. 

76 return self._isForFiles 

77 

78 @property 

79 def rank(self) -> int: 

80 # Docstring inherited from PathElementHandler. 

81 return 0 

82 

83 def __call__(self, path: str, name: str, datasets: Mapping[DatasetType, List[FileDataset]], *, 

84 log: Log, predicate: Callable[[DataCoordinate], bool]) -> bool: 

85 # Docstring inherited from PathElementHandler. 

86 if self._pattern.fullmatch(name): 

87 return True 

88 else: 

89 return False 

90 

91 

92class ParsedPathElementHandler(PathElementHandler): 

93 """An intermediate base class for `PathElementHandler` classes that utilize 

94 a `PathElementParser` to match a Gen2 filename template. 

95 

96 Parameters 

97 ---------- 

98 parser : `PathElementParser` 

99 An object that matches the path element this handler is responsible for 

100 and extracts a (partial) Gen2 data ID from it. 

101 """ 

102 def __init__(self, parser: PathElementParser): 

103 super().__init__() 

104 self._parser = parser 

105 

106 __slots__ = ("_parser",) 

107 

108 def __call__(self, path: str, name: str, datasets: Mapping[DatasetType, List[FileDataset]], *, 

109 log: Log, predicate: Callable[[DataCoordinate], bool]) -> bool: 

110 # Docstring inherited from PathElementParser. 

111 nextDataId2 = self._parser.parse(name, self.lastDataId2, log=log) 

112 if nextDataId2 is None: 

113 return False 

114 self.handle(path, nextDataId2, datasets, log=log, predicate=predicate) 

115 return True 

116 

117 @property 

118 def rank(self) -> int: 

119 # Docstring inherited from PathElementParser. 

120 return len(self._parser.keys) 

121 

122 @abstractmethod 

123 def handle(self, path: str, nextDataId2: dict, datasets: Mapping[DatasetType, List[FileDataset]], *, 

124 log: Log, predicate: Callable[[DataCoordinate], bool]): 

125 """Customization hook for ``__call__``. 

126 

127 Subclasses must override this method, while external callers (i.e. 

128 `DirectoryScanner` should instead invoke `__call__`. 

129 

130 Parameters 

131 ---------- 

132 path : `str` 

133 Full path of the file or directory. 

134 nextDataId2 : `dict` 

135 Gen2 data ID (usually partial) extracted from the path so far. 

136 datasets : `dict` [`DatasetType`, `list` [`FileDataset`] ] 

137 Dictionary that found datasets should be added to. 

138 log : `Log`, optional 

139 Log to use to report warnings and debug information. 

140 predicate : `~collections.abc.Callable` 

141 A callable taking a single `DataCoordinate` argument and returning 

142 `bool`, indicating whether that (Gen3) data ID represents one 

143 that should be included in the scan. 

144 """ 

145 raise NotImplementedError() 

146 

147 

148class SkipHandler(ParsedPathElementHandler): 

149 """A `ParsedPathElementHandler` that does nothing with an entry other 

150 optionally logging a warning message. 

151 

152 A `SkipHandler` is used for Gen2 datasets that we can recognize but do not 

153 want to (or cannot) extract Gen3 datasets from, or other files/directories 

154 that alway appears at a fixed level in the diectory tree. 

155 

156 Parameters 

157 ---------- 

158 parser : `PathElementParser` 

159 An object that matches the path element this handler is responsible for 

160 and extracts a (partial) Gen2 data ID from it. 

161 isForFiles : `bool` 

162 Whether this handler should be applied to files (`True`) or 

163 directories (`False`). 

164 message : `str`, optional 

165 A message to log at warning level when this handler matches a path 

166 entry. If `None`, matched entrie will be silently skipped. 

167 """ 

168 def __init__(self, parser: PathElementParser, isForFiles: bool, message: Optional[str]): 

169 super().__init__(parser=parser) 

170 self._isForFiles = isForFiles 

171 self._message = message 

172 

173 __slots__ = ("_message", "_isForFiles") 

174 

175 def isForFiles(self) -> bool: 

176 # Docstring inherited from PathElementHandler. 

177 return self._isForFiles 

178 

179 def handle(self, path: str, nextDataId2: dict, datasets: Mapping[DatasetType, List[FileDataset]], *, 

180 log: Log, predicate: Callable[[DataCoordinate], bool]): 

181 # Docstring inherited from ParsedPathElementHandler. 

182 if self._message is not None: 

183 log.warn("Skipping %s: %s", path, self._message) 

184 

185 

186class SubdirectoryHandler(ParsedPathElementHandler): 

187 """A `PathElementHandler` that uses a `DirectoryScanner` to recurse. 

188 

189 Parameters 

190 ---------- 

191 parser : `PathElementParser` 

192 An object that matches the path element this handler is responsible for 

193 and extracts a (partial) Gen2 data ID from it. 

194 

195 Notes 

196 ----- 

197 The nested `DirectoryScanner` is default-constructed and should be 

198 populated with child handlers after the `SubdirectoryHandler` is created. 

199 """ 

200 

201 def __init__(self, parser: PathElementParser): 

202 super().__init__(parser=parser) 

203 self.scanner = DirectoryScanner() 

204 

205 __slots__ = ("scanner",) 

206 

207 def isForFiles(self) -> bool: 

208 # Docstring inherited from PathElementHandler. 

209 return False 

210 

211 def handle(self, path: str, nextDataId2, datasets: Mapping[DatasetType, List[FileDataset]], *, 

212 log: Log, predicate: Callable[[DataCoordinate], bool]): 

213 # Docstring inherited from ParsedPathElementHandler. 

214 if not nextDataId2: 

215 # We matched, and there's no data ID at all yet. That means the 

216 # full path so far is just a fixed string so we should descend 

217 # and the match is exclusive. 

218 scan = True 

219 else: 

220 dataId3 = self.translate(nextDataId2, partial=True, log=log) 

221 if dataId3 is not None: 

222 scan = predicate(dataId3) 

223 else: 

224 scan = True 

225 if scan: 

226 for handler in self.scanner: 

227 handler.lastDataId2 = nextDataId2 

228 self.scanner.scan(path, datasets, log=log, predicate=predicate) 

229 

230 def translate(self, dataId2: dict, *, partial: bool = False, log: Log) -> Optional[DataCoordinate]: 

231 # Docstring inherited from PathElementHandler. 

232 for handler in self.scanner: 

233 # Since we're recursing, we're always asking for a partial match, 

234 # because the data ID we have corresponds to different level than 

235 # the one child handlers operate at. 

236 result = handler.translate(dataId2, partial=True, log=log) 

237 if result is not None: 

238 return result 

239 return None 

240 

241 scanner: DirectoryScanner 

242 """Scanner object that holds handlers for the entries of the subdirectory 

243 matched by this handler (`DirectoryScanner`). 

244 """ 

245 

246 

247class TargetFileHandler(ParsedPathElementHandler): 

248 """A `PathElementHandler` that matches files that correspond to target 

249 datasets and outputs `FileDataset` instances for them. 

250 

251 Parameters 

252 ---------- 

253 parser : `PathElementParser` 

254 An object that matches the path element this handler is responsible for 

255 and extracts a (partial) Gen2 data ID from it. 

256 translator : `Translator` 

257 Object that translates data IDs from Gen2 to Gen3. 

258 datasetType : `lsst.daf.butler.DatasetType` 

259 Gen3 dataset type for the datasets this handler matches. 

260 """ 

261 def __init__(self, parser: PathElementParser, translator: Translator, datasetType: DatasetType): 

262 super().__init__(parser=parser) 

263 self._translator = translator 

264 self._datasetType = datasetType 

265 

266 __slots__ = ("_translator", "_datasetType") 

267 

268 def isForFiles(self) -> bool: 

269 # Docstring inherited from PathElementHandler. 

270 return True 

271 

272 def handle(self, path: str, nextDataId2, datasets: Mapping[DatasetType, List[FileDataset]], *, 

273 log: Log, predicate: Callable[[DataCoordinate], bool]): 

274 # Docstring inherited from ParsedPathElementHandler. 

275 dataId3 = self.translate(nextDataId2, partial=False, log=log) 

276 if predicate(dataId3): 

277 datasets[self._datasetType].append(FileDataset(refs=[DatasetRef(self._datasetType, dataId3)], 

278 path=path)) 

279 

280 def translate(self, dataId2: dict, *, partial: bool = False, log: Log) -> Optional[DataCoordinate]: 

281 # Docstring inherited from PathElementHandler. 

282 rawDataId3 = self._translator(dataId2, partial=partial, log=log) 

283 if partial: 

284 return DataCoordinate.standardize(rawDataId3, universe=self._datasetType.dimensions.universe) 

285 else: 

286 return DataCoordinate.standardize(rawDataId3, graph=self._datasetType.dimensions)