Coverage for python/lsst/pipe/base/executionButlerBuilder.py: 15%
Shortcuts 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
Shortcuts 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 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 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
23__all__ = ("buildExecutionButler", )
25import io
27from collections import defaultdict
28import itertools
29from typing import Callable, DefaultDict, Mapping, Optional, Set, Tuple, Iterable, List, Union
31from lsst.daf.butler import (DatasetRef, DatasetType, Butler, DataCoordinate, ButlerURI, Config)
32from lsst.daf.butler.core.utils import getClassOf
33from lsst.daf.butler.transfers import RepoExportContext
34from lsst.daf.butler.core.repoRelocation import BUTLER_ROOT_TAG
36from .graph import QuantumGraph, QuantumNode
37from .pipeline import PipelineDatasetTypes
39DataSetTypeMap = Mapping[DatasetType, Set[DataCoordinate]]
42def _accumulate(
43 graph: QuantumGraph,
44 dataset_types: PipelineDatasetTypes,
45) -> Tuple[Set[DatasetRef], DataSetTypeMap]:
46 # accumulate the DatasetRefs that will be transferred to the execution
47 # registry
49 # exports holds all the existing data that will be migrated to the
50 # execution butler
51 exports: Set[DatasetRef] = set()
53 # inserts is the mapping of DatasetType to dataIds for what is to be
54 # inserted into the registry. These are the products that are expected
55 # to be produced during processing of the QuantumGraph
56 inserts: DefaultDict[DatasetType, Set[DataCoordinate]] = defaultdict(set)
58 # Add inserts for initOutputs (including initIntermediates); these are
59 # defined fully by their DatasetType, because they have no dimensions, and
60 # they are by definition not resolved. initInputs are part of Quantum and
61 # that's the only place the graph stores the dataset IDs, so we process
62 # them there even though each Quantum for a task has the same ones.
63 for dataset_type in itertools.chain(dataset_types.initIntermediates, dataset_types.initOutputs):
64 inserts[dataset_type].add(DataCoordinate.makeEmpty(dataset_type.dimensions.universe))
66 n: QuantumNode
67 for quantum in (n.quantum for n in graph):
68 for attrName in ("initInputs", "inputs", "outputs"):
69 attr: Mapping[DatasetType, Union[DatasetRef, List[DatasetRef]]] = getattr(quantum, attrName)
71 for type, refs in attr.items():
72 # This if block is because init inputs has a different
73 # signature for its items
74 if not isinstance(refs, list):
75 refs = [refs]
76 # iterate over all the references, if it has an id, it
77 # means it exists and should be exported, if not it should
78 # be inserted into the new registry
79 for ref in refs:
80 if ref.id is not None:
81 exports.add(ref)
82 else:
83 if ref.isComponent():
84 # We can't insert a component, and a component will
85 # be part of some other upstream dataset, so it
86 # should be safe to skip them here
87 continue
88 inserts[type].add(ref.dataId)
89 return exports, inserts
92def _discoverCollections(butler: Butler, collections: Iterable[str]) -> set[str]:
93 # Recurse through any discovered collections to make sure all collections
94 # are exported. This exists because I ran into a situation where some
95 # collections were not properly being discovered and exported. This
96 # method may be able to be removed in the future if collection export
97 # logic changes
98 collections = set(collections)
99 while True:
100 discoveredCollections = set(butler.registry.queryCollections(collections, flattenChains=True,
101 includeChains=True))
102 if len(discoveredCollections) > len(collections):
103 collections = discoveredCollections
104 else:
105 break
106 return collections
109def _export(butler: Butler, collections: Optional[Iterable[str]], exports: Set[DatasetRef],
110 inserts: DataSetTypeMap) -> io.StringIO:
111 # This exports the datasets that exist in the input butler using
112 # daf butler objects, however it reaches in deep and does not use the
113 # public methods so that it can export it to a string buffer and skip
114 # disk access.
115 yamlBuffer = io.StringIO()
116 # Yaml is hard coded, since the class controls both ends of the
117 # export/import
118 BackendClass = getClassOf(butler._config["repo_transfer_formats", "yaml", "export"])
119 backend = BackendClass(yamlBuffer)
120 exporter = RepoExportContext(butler.registry, butler.datastore, backend, directory=None, transfer=None)
121 exporter.saveDatasets(exports)
123 # Need to ensure that the dimension records for outputs are
124 # transferred.
125 for _, dataIds in inserts.items():
126 exporter.saveDataIds(dataIds)
128 # Look for any defined collection, if not get the defaults
129 if collections is None:
130 collections = butler.registry.defaults.collections
132 # look up all collections associated with those inputs, this follows
133 # all chains to make sure everything is properly exported
134 for c in _discoverCollections(butler, collections):
135 exporter.saveCollection(c)
136 exporter._finish()
138 # reset the string buffer to the beginning so the read operation will
139 # actually *see* the data that was exported
140 yamlBuffer.seek(0)
141 return yamlBuffer
144def _setupNewButler(butler: Butler, outputLocation: ButlerURI, dirExists: bool) -> Butler:
145 # Set up the new butler object at the specified location
146 if dirExists:
147 # Remove the existing table, if the code got this far and this exists
148 # clobber must be true
149 executionRegistry = outputLocation.join("gen3.sqlite3")
150 if executionRegistry.exists():
151 executionRegistry.remove()
152 else:
153 outputLocation.mkdir()
155 # Copy the existing butler config, modifying the location of the
156 # registry to the specified location.
157 # Preserve the root path from the existing butler so things like
158 # file data stores continue to look at the old location.
159 config = Config(butler._config)
160 config["root"] = outputLocation.geturl()
161 config["allow_put_of_predefined_dataset"] = True
162 config["registry", "db"] = "sqlite:///<butlerRoot>/gen3.sqlite3"
164 # Remove any namespace that may be set in main registry.
165 config.pop(("registry", "namespace"), None)
167 # record the current root of the datastore if it is specified relative
168 # to the butler root
169 if config.get(("datastore", "root")) == BUTLER_ROOT_TAG:
170 config["datastore", "root"] = butler._config.configDir.geturl()
171 config["datastore", "trust_get_request"] = True
173 # Requires that we use the dimension configuration from the original
174 # butler and not use the defaults.
175 config = Butler.makeRepo(root=outputLocation, config=config,
176 dimensionConfig=butler.registry.dimensions.dimensionConfig,
177 overwrite=True, forceConfigRoot=False)
179 # Return a newly created butler
180 return Butler(config, writeable=True)
183def _import(yamlBuffer: io.StringIO,
184 newButler: Butler,
185 inserts: DataSetTypeMap,
186 run: str,
187 butlerModifier: Optional[Callable[[Butler], Butler]]
188 ) -> Butler:
189 # This method takes the exports from the existing butler, imports
190 # them into the newly created butler, and then inserts the datasets
191 # that are expected to be produced.
193 # import the existing datasets using "split" mode. "split" is safe
194 # because execution butler is assumed to be able to see all the file
195 # locations that the main datastore can see. "split" supports some
196 # absolute URIs in the datastore.
197 newButler.import_(filename=yamlBuffer, format="yaml", reuseIds=True, transfer="split")
199 # If there is modifier callable, run it to make necessary updates
200 # to the new butler.
201 if butlerModifier is not None:
202 newButler = butlerModifier(newButler)
204 # Register datasets to be produced and insert them into the registry
205 for dsType, dataIds in inserts.items():
206 newButler.registry.registerDatasetType(dsType)
207 newButler.registry.insertDatasets(dsType, dataIds, run)
209 return newButler
212def buildExecutionButler(butler: Butler,
213 graph: QuantumGraph,
214 outputLocation: Union[str, ButlerURI],
215 run: str,
216 *,
217 clobber: bool = False,
218 butlerModifier: Optional[Callable[[Butler], Butler]] = None,
219 collections: Optional[Iterable[str]] = None
220 ) -> Butler:
221 r"""buildExecutionButler is a function that is responsible for exporting
222 input `QuantumGraphs` into a new minimal `~lsst.daf.butler.Butler` which
223 only contains datasets specified by the `QuantumGraph`. These datasets are
224 both those that already exist in the input `~lsst.daf.butler.Butler`, and
225 those that are expected to be produced during the execution of the
226 `QuantumGraph`.
228 Parameters
229 ----------
230 butler : `lsst.daf.butler.Bulter`
231 This is the existing `~lsst.daf.butler.Butler` instance from which
232 existing datasets will be exported. This should be the
233 `~lsst.daf.butler.Butler` which was used to create any `QuantumGraphs`
234 that will be converted with this object.
235 graph : `QuantumGraph`
236 Graph containing nodes that are to be exported into an execution
237 butler
238 outputLocation : `str` or `~lsst.daf.butler.ButlerURI`
239 URI Location at which the execution butler is to be exported. May be
240 specified as a string or a ButlerURI instance.
241 run : `str` optional
242 The run collection that the exported datasets are to be placed in. If
243 None, the default value in registry.defaults will be used.
244 clobber : `bool`, Optional
245 By default a butler will not be created if a file or directory
246 already exists at the output location. If this is set to `True`
247 what is at the location will be deleted prior to running the
248 export. Defaults to `False`
249 butlerModifier : `~typing.Callable`, Optional
250 If supplied this should be a callable that accepts a
251 `~lsst.daf.butler.Butler`, and returns an instantiated
252 `~lsst.daf.butler.Butler`. This callable may be used to make any
253 modifications to the `~lsst.daf.butler.Butler` desired. This
254 will be called after importing all datasets that exist in the input
255 `~lsst.daf.butler.Butler` but prior to inserting Datasets expected
256 to be produced. Examples of what this method could do include
257 things such as creating collections/runs/ etc.
258 collections : `~typing.Iterable` of `str`, Optional
259 An iterable of collection names that will be exported from the input
260 `~lsst.daf.butler.Butler` when creating the execution butler. If not
261 supplied the `~lsst.daf.butler.Butler`\ 's `~lsst.daf.butler.Registry`
262 default collections will be used.
264 Returns
265 -------
266 executionButler : `lsst.daf.butler.Butler`
267 An instance of the newly created execution butler
269 Raises
270 ------
271 FileExistsError
272 Raised if something exists in the filesystem at the specified output
273 location and clobber is `False`
274 NotADirectoryError
275 Raised if specified output URI does not correspond to a directory
276 """
277 # We know this must refer to a directory.
278 outputLocation = ButlerURI(outputLocation, forceDirectory=True)
280 # Do this first to Fail Fast if the output exists
281 if (dirExists := outputLocation.exists()) and not clobber:
282 raise FileExistsError("Cannot create a butler at specified location, location exists")
283 if not outputLocation.isdir():
284 raise NotADirectoryError("The specified output URI does not appear to correspond to a directory")
286 # Gather all DatasetTypes from the Python and check any that already exist
287 # in the registry for consistency. This does not check that all dataset
288 # types here exist, because they might want to register dataset types
289 # later. It would be nice to also check that, but to that we would need to
290 # be told whether they plan to register dataset types later (DM-30845).
291 dataset_types = PipelineDatasetTypes.fromPipeline(graph.iterTaskGraph(), registry=butler.registry)
293 exports, inserts = _accumulate(graph, dataset_types)
294 yamlBuffer = _export(butler, collections, exports, inserts)
296 newButler = _setupNewButler(butler, outputLocation, dirExists)
298 return _import(yamlBuffer, newButler, inserts, run, butlerModifier)