Coverage for python/lsst/pipe/base/pipeline_graph/_dataset_types.py: 49%
81 statements
« prev ^ index » next coverage.py v7.4.4, created at 2024-04-06 04:05 -0700
« prev ^ index » next coverage.py v7.4.4, created at 2024-04-06 04:05 -0700
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/>.
27from __future__ import annotations
29__all__ = ("DatasetTypeNode",)
31import dataclasses
32from collections.abc import Collection
33from typing import TYPE_CHECKING, Any
35import networkx
36from lsst.daf.butler import DatasetRef, DatasetType, DimensionGroup, Registry, StorageClass
37from lsst.daf.butler.registry import MissingDatasetTypeError
39from ._exceptions import DuplicateOutputError
40from ._nodes import NodeKey, NodeType
42if TYPE_CHECKING:
43 from ._edges import ReadEdge, WriteEdge
46@dataclasses.dataclass(frozen=True, eq=False)
47class DatasetTypeNode:
48 """A node in a pipeline graph that represents a resolved dataset type.
50 Notes
51 -----
52 A dataset type node represents a common definition of the dataset type
53 across the entire graph - it is never a component, and the storage class is
54 the registry dataset type's storage class or (if there isn't one) the one
55 defined by the producing task.
57 Dataset type nodes are intentionally not equality comparable, since there
58 are many different (and useful) ways to compare these objects with no clear
59 winner as the most obvious behavior.
60 """
62 dataset_type: DatasetType
63 """Common definition of this dataset type for the graph.
64 """
66 is_initial_query_constraint: bool
67 """Whether this dataset should be included as a constraint in the initial
68 query for data IDs in QuantumGraph generation.
70 This is only `True` for dataset types that are overall regular inputs, and
71 only if none of those input connections had ``deferQueryConstraint=True``.
72 """
74 is_prerequisite: bool
75 """Whether this dataset type is a prerequisite input that must exist in
76 the Registry before graph creation.
77 """
79 producing_edge: WriteEdge | None
80 """The edge to the task that produces this dataset type."""
82 consuming_edges: Collection[ReadEdge]
83 """The edges to tasks that consume this dataset type."""
85 @classmethod
86 def _from_edges(
87 cls, key: NodeKey, xgraph: networkx.MultiDiGraph, registry: Registry, previous: DatasetTypeNode | None
88 ) -> DatasetTypeNode:
89 """Construct a dataset type node from its edges.
91 Parameters
92 ----------
93 key : `NodeKey`
94 Named tuple that holds the dataset type and serves as the node
95 object in the internal networkx graph.
96 xgraph : `networkx.MultiDiGraph`
97 The internal networkx graph.
98 registry : `lsst.daf.butler.Registry`
99 Registry client for the data repository. Only used to get
100 dataset type definitions and the dimension universe.
101 previous : `DatasetTypeNode` or `None`
102 Previous node for this dataset type.
104 Returns
105 -------
106 node : `DatasetTypeNode`
107 Node consistent with all edges pointing to it and the data
108 repository.
109 """
110 try:
111 dataset_type = registry.getDatasetType(key.name)
112 is_registered = True
113 except MissingDatasetTypeError:
114 dataset_type = None
115 is_registered = False
116 if previous is not None and previous.dataset_type == dataset_type:
117 # This node was already resolved (with exactly the same edges
118 # contributing, since we clear resolutions when edges are added or
119 # removed). The only thing that might have changed was the
120 # definition in the registry, and it didn't.
121 return previous
122 is_initial_query_constraint = True
123 is_prerequisite: bool | None = None
124 producer: str | None = None
125 producing_edge: WriteEdge | None = None
126 # Iterate over the incoming edges to this node, which represent the
127 # output connections of tasks that write this dataset type; these take
128 # precedence over the inputs in determining the graph-wide dataset type
129 # definition (and hence which storage class we register when using the
130 # graph to register dataset types). There should only be one such
131 # connection, but we won't necessarily have checked that rule until
132 # here. As a result there can be at most one iteration of this loop.
133 for _, _, producing_edge in xgraph.in_edges(key, data="instance"):
134 assert producing_edge is not None, "Should only be None if we never loop."
135 if producer is not None:
136 raise DuplicateOutputError(
137 f"Dataset type {key.name!r} is produced by both {producing_edge.task_label!r} "
138 f"and {producer!r}."
139 )
140 producer = producing_edge.task_label
141 dataset_type = producing_edge._resolve_dataset_type(dataset_type, universe=registry.dimensions)
142 is_prerequisite = False
143 is_initial_query_constraint = False
144 consuming_edge: ReadEdge
145 consumers: list[str] = []
146 consuming_edges = list(
147 consuming_edge for _, _, consuming_edge in xgraph.out_edges(key, data="instance")
148 )
149 # Put edges that are not component datasets before any edges that are.
150 consuming_edges.sort(key=lambda consuming_edge: consuming_edge.component is not None)
151 for consuming_edge in consuming_edges:
152 dataset_type, is_initial_query_constraint, is_prerequisite = consuming_edge._resolve_dataset_type(
153 current=dataset_type,
154 universe=registry.dimensions,
155 is_initial_query_constraint=is_initial_query_constraint,
156 is_prerequisite=is_prerequisite,
157 is_registered=is_registered,
158 producer=producer,
159 consumers=consumers,
160 )
161 consumers.append(consuming_edge.task_label)
162 assert dataset_type is not None, "Graph structure guarantees at least one edge."
163 assert is_prerequisite is not None, "Having at least one edge guarantees is_prerequisite is known."
164 return DatasetTypeNode(
165 dataset_type=dataset_type,
166 is_initial_query_constraint=is_initial_query_constraint,
167 is_prerequisite=is_prerequisite,
168 producing_edge=producing_edge,
169 consuming_edges=tuple(consuming_edges),
170 )
172 @property
173 def name(self) -> str:
174 """Name of the dataset type.
176 This is always the parent dataset type, never that of a component.
177 """
178 return self.dataset_type.name
180 @property
181 def key(self) -> NodeKey:
182 """Key that identifies this dataset type in internal and exported
183 networkx graphs.
184 """
185 return NodeKey(NodeType.DATASET_TYPE, self.dataset_type.name)
187 @property
188 def dimensions(self) -> DimensionGroup:
189 """Dimensions of the dataset type."""
190 return self.dataset_type.dimensions.as_group()
192 @property
193 def storage_class_name(self) -> str:
194 """String name of the storage class for this dataset type."""
195 return self.dataset_type.storageClass_name
197 @property
198 def storage_class(self) -> StorageClass:
199 """Storage class for this dataset type."""
200 return self.dataset_type.storageClass
202 @property
203 def is_calibration(self) -> bool:
204 """Whether this dataset type can be included in
205 `~lsst.daf.butler.CollectionType.CALIBRATION` collections.
206 """
207 return self.dataset_type.isCalibration()
209 def __repr__(self) -> str:
210 return f"{self.name} ({self.storage_class_name}, {self.dimensions})"
212 def generalize_ref(self, ref: DatasetRef) -> DatasetRef:
213 """Convert a `~lsst.daf.butler.DatasetRef` with the dataset type
214 associated with some task to one with the common dataset type defined
215 by this node.
217 Parameters
218 ----------
219 ref : `lsst.daf.butler.DatasetRef`
220 Reference whose dataset type is convertible to this node's, either
221 because it is a component with the node's dataset type as its
222 parent, or because it has a compatible storage class.
224 Returns
225 -------
226 ref : `lsst.daf.butler.DatasetRef`
227 Reference with exactly this node's dataset type.
228 """
229 if ref.isComponent():
230 ref = ref.makeCompositeRef()
231 if ref.datasetType.storageClass_name != self.dataset_type.storageClass_name:
232 return ref.overrideStorageClass(self.dataset_type.storageClass_name)
233 return ref
235 def _to_xgraph_state(self) -> dict[str, Any]:
236 """Convert this node's attributes into a dictionary suitable for use
237 in exported networkx graphs.
238 """
239 return {
240 "dataset_type": self.dataset_type,
241 "is_initial_query_constraint": self.is_initial_query_constraint,
242 "is_prerequisite": self.is_prerequisite,
243 "dimensions": self.dataset_type.dimensions,
244 "storage_class_name": self.dataset_type.storageClass_name,
245 "bipartite": NodeType.DATASET_TYPE.bipartite,
246 }