Coverage for python/lsst/pipe/base/pipeline_graph/_dataset_types.py: 49%
81 statements
« prev ^ index » next coverage.py v7.3.0, created at 2023-08-31 09:39 +0000
« prev ^ index » next coverage.py v7.3.0, created at 2023-08-31 09:39 +0000
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__ = ("DatasetTypeNode",)
25import dataclasses
26from collections.abc import Collection
27from typing import TYPE_CHECKING, Any
29import networkx
30from lsst.daf.butler import DatasetRef, DatasetType, DimensionGraph, Registry, StorageClass
31from lsst.daf.butler.registry import MissingDatasetTypeError
33from ._exceptions import DuplicateOutputError
34from ._nodes import NodeKey, NodeType
36if TYPE_CHECKING:
37 from ._edges import ReadEdge, WriteEdge
40@dataclasses.dataclass(frozen=True, eq=False)
41class DatasetTypeNode:
42 """A node in a pipeline graph that represents a resolved dataset type.
44 Notes
45 -----
46 A dataset type node represents a common definition of the dataset type
47 across the entire graph - it is never a component, and the storage class is
48 the registry dataset type's storage class or (if there isn't one) the one
49 defined by the producing task.
51 Dataset type nodes are intentionally not equality comparable, since there
52 are many different (and useful) ways to compare these objects with no clear
53 winner as the most obvious behavior.
54 """
56 dataset_type: DatasetType
57 """Common definition of this dataset type for the graph.
58 """
60 is_initial_query_constraint: bool
61 """Whether this dataset should be included as a constraint in the initial
62 query for data IDs in QuantumGraph generation.
64 This is only `True` for dataset types that are overall regular inputs, and
65 only if none of those input connections had ``deferQueryConstraint=True``.
66 """
68 is_prerequisite: bool
69 """Whether this dataset type is a prerequisite input that must exist in
70 the Registry before graph creation.
71 """
73 producing_edge: WriteEdge | None
74 """The edge to the task that produces this dataset type."""
76 consuming_edges: Collection[ReadEdge]
77 """The edges to tasks that consume this dataset type."""
79 @classmethod
80 def _from_edges(
81 cls, key: NodeKey, xgraph: networkx.MultiDiGraph, registry: Registry, previous: DatasetTypeNode | None
82 ) -> DatasetTypeNode:
83 """Construct a dataset type node from its edges.
85 Parameters
86 ----------
87 key : `NodeKey`
88 Named tuple that holds the dataset type and serves as the node
89 object in the internal networkx graph.
90 xgraph : `networkx.MultiDiGraph`
91 The internal networkx graph.
92 registry : `lsst.daf.butler.Registry`
93 Registry client for the data repository. Only used to get
94 dataset type definitions and the dimension universe.
95 previous : `DatasetTypeNode` or `None`
96 Previous node for this dataset type.
98 Returns
99 -------
100 node : `DatasetTypeNode`
101 Node consistent with all edges pointing to it and the data
102 repository.
103 """
104 try:
105 dataset_type = registry.getDatasetType(key.name)
106 is_registered = True
107 except MissingDatasetTypeError:
108 dataset_type = None
109 is_registered = False
110 if previous is not None and previous.dataset_type == dataset_type:
111 # This node was already resolved (with exactly the same edges
112 # contributing, since we clear resolutions when edges are added or
113 # removed). The only thing that might have changed was the
114 # definition in the registry, and it didn't.
115 return previous
116 is_initial_query_constraint = True
117 is_prerequisite: bool | None = None
118 producer: str | None = None
119 producing_edge: WriteEdge | None = None
120 # Iterate over the incoming edges to this node, which represent the
121 # output connections of tasks that write this dataset type; these take
122 # precedence over the inputs in determining the graph-wide dataset type
123 # definition (and hence which storage class we register when using the
124 # graph to register dataset types). There should only be one such
125 # connection, but we won't necessarily have checked that rule until
126 # here. As a result there can be at most one iteration of this loop.
127 for _, _, producing_edge in xgraph.in_edges(key, data="instance"):
128 assert producing_edge is not None, "Should only be None if we never loop."
129 if producer is not None:
130 raise DuplicateOutputError(
131 f"Dataset type {key.name!r} is produced by both {producing_edge.task_label!r} "
132 f"and {producer!r}."
133 )
134 producer = producing_edge.task_label
135 dataset_type = producing_edge._resolve_dataset_type(dataset_type, universe=registry.dimensions)
136 is_prerequisite = False
137 is_initial_query_constraint = False
138 consuming_edge: ReadEdge
139 consumers: list[str] = []
140 consuming_edges = list(
141 consuming_edge for _, _, consuming_edge in xgraph.out_edges(key, data="instance")
142 )
143 # Put edges that are not component datasets before any edges that are.
144 consuming_edges.sort(key=lambda consuming_edge: consuming_edge.component is not None)
145 for consuming_edge in consuming_edges:
146 dataset_type, is_initial_query_constraint, is_prerequisite = consuming_edge._resolve_dataset_type(
147 current=dataset_type,
148 universe=registry.dimensions,
149 is_initial_query_constraint=is_initial_query_constraint,
150 is_prerequisite=is_prerequisite,
151 is_registered=is_registered,
152 producer=producer,
153 consumers=consumers,
154 )
155 consumers.append(consuming_edge.task_label)
156 assert dataset_type is not None, "Graph structure guarantees at least one edge."
157 assert is_prerequisite is not None, "Having at least one edge guarantees is_prerequisite is known."
158 return DatasetTypeNode(
159 dataset_type=dataset_type,
160 is_initial_query_constraint=is_initial_query_constraint,
161 is_prerequisite=is_prerequisite,
162 producing_edge=producing_edge,
163 consuming_edges=tuple(consuming_edges),
164 )
166 @property
167 def name(self) -> str:
168 """Name of the dataset type.
170 This is always the parent dataset type, never that of a component.
171 """
172 return self.dataset_type.name
174 @property
175 def key(self) -> NodeKey:
176 """Key that identifies this dataset type in internal and exported
177 networkx graphs.
178 """
179 return NodeKey(NodeType.DATASET_TYPE, self.dataset_type.name)
181 @property
182 def dimensions(self) -> DimensionGraph:
183 """Dimensions of the dataset type."""
184 return self.dataset_type.dimensions
186 @property
187 def storage_class_name(self) -> str:
188 """String name of the storage class for this dataset type."""
189 return self.dataset_type.storageClass_name
191 @property
192 def storage_class(self) -> StorageClass:
193 """Storage class for this dataset type."""
194 return self.dataset_type.storageClass
196 @property
197 def is_calibration(self) -> bool:
198 """Whether this dataset type can be included in
199 `~lsst.daf.butler.CollectionType.CALIBRATION` collections.
200 """
201 return self.dataset_type.isCalibration()
203 def __repr__(self) -> str:
204 return f"{self.name} ({self.storage_class_name}, {self.dimensions})"
206 def generalize_ref(self, ref: DatasetRef) -> DatasetRef:
207 """Convert a `~lsst.daf.butler.DatasetRef` with the dataset type
208 associated with some task to one with the common dataset type defined
209 by this node.
211 Parameters
212 ----------
213 ref : `lsst.daf.butler.DatasetRef`
214 Reference whose dataset type is convertible to this node's, either
215 because it is a component with the node's dataset type as its
216 parent, or because it has a compatible storage class.
218 Returns
219 -------
220 ref : `lsst.daf.butler.DatasetRef`
221 Reference with exactly this node's dataset type.
222 """
223 if ref.isComponent():
224 ref = ref.makeCompositeRef()
225 if ref.datasetType.storageClass_name != self.dataset_type.storageClass_name:
226 return ref.overrideStorageClass(self.dataset_type.storageClass_name)
227 return ref
229 def _to_xgraph_state(self) -> dict[str, Any]:
230 """Convert this node's attributes into a dictionary suitable for use
231 in exported networkx graphs.
232 """
233 return {
234 "dataset_type": self.dataset_type,
235 "is_initial_query_constraint": self.is_initial_query_constraint,
236 "is_prerequisite": self.is_prerequisite,
237 "dimensions": self.dataset_type.dimensions,
238 "storage_class_name": self.dataset_type.storageClass_name,
239 "bipartite": NodeType.DATASET_TYPE.bipartite,
240 }