Coverage for python/lsst/pipe/base/graph/graph.py: 19%
341 statements
« prev ^ index » next coverage.py v6.4.1, created at 2022-06-05 02:33 -0700
« prev ^ index » next coverage.py v6.4.1, created at 2022-06-05 02:33 -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 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__ = ("QuantumGraph", "IncompatibleGraphError")
25import io
26import json
27import lzma
28import os
29import pickle
30import struct
31import time
32import uuid
33import warnings
34from collections import defaultdict, deque
35from itertools import chain
36from types import MappingProxyType
37from typing import (
38 Any,
39 BinaryIO,
40 DefaultDict,
41 Deque,
42 Dict,
43 FrozenSet,
44 Generator,
45 Iterable,
46 List,
47 Mapping,
48 MutableMapping,
49 Optional,
50 Set,
51 Tuple,
52 TypeVar,
53 Union,
54)
56import networkx as nx
57from lsst.daf.butler import DatasetRef, DatasetType, DimensionRecordsAccumulator, DimensionUniverse, Quantum
58from lsst.resources import ResourcePath, ResourcePathExpression
59from lsst.utils.introspection import get_full_type_name
60from networkx.drawing.nx_agraph import write_dot
62from ..connections import iterConnections
63from ..pipeline import TaskDef
64from ._implDetails import DatasetTypeName, _DatasetTracker, _pruner
65from ._loadHelpers import LoadHelper
66from ._versionDeserializers import DESERIALIZER_MAP
67from .quantumNode import BuildId, QuantumNode
69_T = TypeVar("_T", bound="QuantumGraph")
71# modify this constant any time the on disk representation of the save file
72# changes, and update the load helpers to behave properly for each version.
73SAVE_VERSION = 3
75# Strings used to describe the format for the preamble bytes in a file save
76# The base is a big endian encoded unsigned short that is used to hold the
77# file format version. This allows reading version bytes and determine which
78# loading code should be used for the rest of the file
79STRUCT_FMT_BASE = ">H"
80#
81# Version 1
82# This marks a big endian encoded format with an unsigned short, an unsigned
83# long long, and an unsigned long long in the byte stream
84# Version 2
85# A big endian encoded format with an unsigned long long byte stream used to
86# indicate the total length of the entire header.
87STRUCT_FMT_STRING = {1: ">QQ", 2: ">Q"}
89# magic bytes that help determine this is a graph save
90MAGIC_BYTES = b"qgraph4\xf6\xe8\xa9"
93class IncompatibleGraphError(Exception):
94 """Exception class to indicate that a lookup by NodeId is impossible due
95 to incompatibilities
96 """
98 pass
101class QuantumGraph:
102 """QuantumGraph is a directed acyclic graph of `QuantumNode` objects
104 This data structure represents a concrete workflow generated from a
105 `Pipeline`.
107 Parameters
108 ----------
109 quanta : Mapping of `TaskDef` to sets of `Quantum`
110 This maps tasks (and their configs) to the sets of data they are to
111 process.
112 metadata : Optional Mapping of `str` to primitives
113 This is an optional parameter of extra data to carry with the graph.
114 Entries in this mapping should be able to be serialized in JSON.
116 Raises
117 ------
118 ValueError
119 Raised if the graph is pruned such that some tasks no longer have nodes
120 associated with them.
121 """
123 def __init__(
124 self,
125 quanta: Mapping[TaskDef, Set[Quantum]],
126 metadata: Optional[Mapping[str, Any]] = None,
127 pruneRefs: Optional[Iterable[DatasetRef]] = None,
128 ):
129 self._buildGraphs(quanta, metadata=metadata, pruneRefs=pruneRefs)
131 def _buildGraphs(
132 self,
133 quanta: Mapping[TaskDef, Set[Quantum]],
134 *,
135 _quantumToNodeId: Optional[Mapping[Quantum, uuid.UUID]] = None,
136 _buildId: Optional[BuildId] = None,
137 metadata: Optional[Mapping[str, Any]] = None,
138 pruneRefs: Optional[Iterable[DatasetRef]] = None,
139 ) -> None:
140 """Builds the graph that is used to store the relation between tasks,
141 and the graph that holds the relations between quanta
142 """
143 self._metadata = metadata
144 self._buildId = _buildId if _buildId is not None else BuildId(f"{time.time()}-{os.getpid()}")
145 # Data structures used to identify relations between components;
146 # DatasetTypeName -> TaskDef for task,
147 # and DatasetRef -> QuantumNode for the quanta
148 self._datasetDict = _DatasetTracker[DatasetTypeName, TaskDef](createInverse=True)
149 self._datasetRefDict = _DatasetTracker[DatasetRef, QuantumNode]()
151 self._nodeIdMap: Dict[uuid.UUID, QuantumNode] = {}
152 self._taskToQuantumNode: MutableMapping[TaskDef, Set[QuantumNode]] = defaultdict(set)
153 for taskDef, quantumSet in quanta.items():
154 connections = taskDef.connections
156 # For each type of connection in the task, add a key to the
157 # `_DatasetTracker` for the connections name, with a value of
158 # the TaskDef in the appropriate field
159 for inpt in iterConnections(connections, ("inputs", "prerequisiteInputs", "initInputs")):
160 self._datasetDict.addConsumer(DatasetTypeName(inpt.name), taskDef)
162 for output in iterConnections(connections, ("outputs",)):
163 self._datasetDict.addProducer(DatasetTypeName(output.name), taskDef)
165 # For each `Quantum` in the set of all `Quantum` for this task,
166 # add a key to the `_DatasetTracker` that is a `DatasetRef` for one
167 # of the individual datasets inside the `Quantum`, with a value of
168 # a newly created QuantumNode to the appropriate input/output
169 # field.
170 for quantum in quantumSet:
171 if _quantumToNodeId:
172 if (nodeId := _quantumToNodeId.get(quantum)) is None:
173 raise ValueError(
174 "If _quantuMToNodeNumber is not None, all quanta must have an "
175 "associated value in the mapping"
176 )
177 else:
178 nodeId = uuid.uuid4()
180 inits = quantum.initInputs.values()
181 inputs = quantum.inputs.values()
182 value = QuantumNode(quantum, taskDef, nodeId)
183 self._taskToQuantumNode[taskDef].add(value)
184 self._nodeIdMap[nodeId] = value
186 for dsRef in chain(inits, inputs):
187 # unfortunately, `Quantum` allows inits to be individual
188 # `DatasetRef`s or an Iterable of such, so there must
189 # be an instance check here
190 if isinstance(dsRef, Iterable):
191 for sub in dsRef:
192 if sub.isComponent():
193 sub = sub.makeCompositeRef()
194 self._datasetRefDict.addConsumer(sub, value)
195 else:
196 assert isinstance(dsRef, DatasetRef)
197 if dsRef.isComponent():
198 dsRef = dsRef.makeCompositeRef()
199 self._datasetRefDict.addConsumer(dsRef, value)
200 for dsRef in chain.from_iterable(quantum.outputs.values()):
201 self._datasetRefDict.addProducer(dsRef, value)
203 if pruneRefs is not None:
204 # track what refs were pruned and prune the graph
205 prunes: Set[QuantumNode] = set()
206 _pruner(self._datasetRefDict, pruneRefs, alreadyPruned=prunes)
208 # recreate the taskToQuantumNode dict removing nodes that have been
209 # pruned. Keep track of task defs that now have no QuantumNodes
210 emptyTasks: Set[str] = set()
211 newTaskToQuantumNode: DefaultDict[TaskDef, Set[QuantumNode]] = defaultdict(set)
212 # accumulate all types
213 types_ = set()
214 # tracker for any pruneRefs that have caused tasks to have no nodes
215 # This helps the user find out what caused the issues seen.
216 culprits = set()
217 # Find all the types from the refs to prune
218 for r in pruneRefs:
219 types_.add(r.datasetType)
221 # For each of the tasks, and their associated nodes, remove any
222 # any nodes that were pruned. If there are no nodes associated
223 # with a task, record that task, and find out if that was due to
224 # a type from an input ref to prune.
225 for td, taskNodes in self._taskToQuantumNode.items():
226 diff = taskNodes.difference(prunes)
227 if len(diff) == 0:
228 if len(taskNodes) != 0:
229 tp: DatasetType
230 for tp in types_:
231 if (tmpRefs := next(iter(taskNodes)).quantum.inputs.get(tp)) and not set(
232 tmpRefs
233 ).difference(pruneRefs):
234 culprits.add(tp.name)
235 emptyTasks.add(td.label)
236 newTaskToQuantumNode[td] = diff
238 # update the internal dict
239 self._taskToQuantumNode = newTaskToQuantumNode
241 if emptyTasks:
242 raise ValueError(
243 f"{', '.join(emptyTasks)} task(s) have no nodes associated with them "
244 f"after graph pruning; {', '.join(culprits)} caused over-pruning"
245 )
247 # Graph of quanta relations
248 self._connectedQuanta = self._datasetRefDict.makeNetworkXGraph()
249 self._count = len(self._connectedQuanta)
251 # Graph of task relations, used in various methods
252 self._taskGraph = self._datasetDict.makeNetworkXGraph()
254 # convert default dict into a regular to prevent accidental key
255 # insertion
256 self._taskToQuantumNode = dict(self._taskToQuantumNode.items())
258 @property
259 def taskGraph(self) -> nx.DiGraph:
260 """Return a graph representing the relations between the tasks inside
261 the quantum graph.
263 Returns
264 -------
265 taskGraph : `networkx.Digraph`
266 Internal datastructure that holds relations of `TaskDef` objects
267 """
268 return self._taskGraph
270 @property
271 def graph(self) -> nx.DiGraph:
272 """Return a graph representing the relations between all the
273 `QuantumNode` objects. Largely it should be preferred to iterate
274 over, and use methods of this class, but sometimes direct access to
275 the networkx object may be helpful
277 Returns
278 -------
279 graph : `networkx.Digraph`
280 Internal datastructure that holds relations of `QuantumNode`
281 objects
282 """
283 return self._connectedQuanta
285 @property
286 def inputQuanta(self) -> Iterable[QuantumNode]:
287 """Make a `list` of all `QuantumNode` objects that are 'input' nodes
288 to the graph, meaning those nodes to not depend on any other nodes in
289 the graph.
291 Returns
292 -------
293 inputNodes : iterable of `QuantumNode`
294 A list of nodes that are inputs to the graph
295 """
296 return (q for q, n in self._connectedQuanta.in_degree if n == 0)
298 @property
299 def outputQuanta(self) -> Iterable[QuantumNode]:
300 """Make a `list` of all `QuantumNode` objects that are 'output' nodes
301 to the graph, meaning those nodes have no nodes that depend them in
302 the graph.
304 Returns
305 -------
306 outputNodes : iterable of `QuantumNode`
307 A list of nodes that are outputs of the graph
308 """
309 return [q for q, n in self._connectedQuanta.out_degree if n == 0]
311 @property
312 def allDatasetTypes(self) -> Tuple[DatasetTypeName, ...]:
313 """Return all the `DatasetTypeName` objects that are contained inside
314 the graph.
316 Returns
317 -------
318 tuple of `DatasetTypeName`
319 All the data set type names that are present in the graph
320 """
321 return tuple(self._datasetDict.keys())
323 @property
324 def isConnected(self) -> bool:
325 """Return True if all of the nodes in the graph are connected, ignores
326 directionality of connections.
327 """
328 return nx.is_weakly_connected(self._connectedQuanta)
330 def pruneGraphFromRefs(self: _T, refs: Iterable[DatasetRef]) -> _T:
331 r"""Return a graph pruned of input `~lsst.daf.butler.DatasetRef`\ s
332 and nodes which depend on them.
334 Parameters
335 ----------
336 refs : `Iterable` of `DatasetRef`
337 Refs which should be removed from resulting graph
339 Returns
340 -------
341 graph : `QuantumGraph`
342 A graph that has been pruned of specified refs and the nodes that
343 depend on them.
344 """
345 newInst = object.__new__(type(self))
346 quantumMap = defaultdict(set)
347 for node in self:
348 quantumMap[node.taskDef].add(node.quantum)
350 # convert to standard dict to prevent accidental key insertion
351 quantumDict: Dict[TaskDef, Set[Quantum]] = dict(quantumMap.items())
353 newInst._buildGraphs(
354 quantumDict,
355 _quantumToNodeId={n.quantum: n.nodeId for n in self},
356 metadata=self._metadata,
357 pruneRefs=refs,
358 )
359 return newInst
361 def getQuantumNodeByNodeId(self, nodeId: uuid.UUID) -> QuantumNode:
362 """Lookup a `QuantumNode` from an id associated with the node.
364 Parameters
365 ----------
366 nodeId : `NodeId`
367 The number associated with a node
369 Returns
370 -------
371 node : `QuantumNode`
372 The node corresponding with input number
374 Raises
375 ------
376 KeyError
377 Raised if the requested nodeId is not in the graph.
378 """
379 return self._nodeIdMap[nodeId]
381 def getQuantaForTask(self, taskDef: TaskDef) -> FrozenSet[Quantum]:
382 """Return all the `Quantum` associated with a `TaskDef`.
384 Parameters
385 ----------
386 taskDef : `TaskDef`
387 The `TaskDef` for which `Quantum` are to be queried
389 Returns
390 -------
391 frozenset of `Quantum`
392 The `set` of `Quantum` that is associated with the specified
393 `TaskDef`.
394 """
395 return frozenset(node.quantum for node in self._taskToQuantumNode[taskDef])
397 def getNodesForTask(self, taskDef: TaskDef) -> FrozenSet[QuantumNode]:
398 """Return all the `QuantumNodes` associated with a `TaskDef`.
400 Parameters
401 ----------
402 taskDef : `TaskDef`
403 The `TaskDef` for which `Quantum` are to be queried
405 Returns
406 -------
407 frozenset of `QuantumNodes`
408 The `frozenset` of `QuantumNodes` that is associated with the
409 specified `TaskDef`.
410 """
411 return frozenset(self._taskToQuantumNode[taskDef])
413 def findTasksWithInput(self, datasetTypeName: DatasetTypeName) -> Iterable[TaskDef]:
414 """Find all tasks that have the specified dataset type name as an
415 input.
417 Parameters
418 ----------
419 datasetTypeName : `str`
420 A string representing the name of a dataset type to be queried,
421 can also accept a `DatasetTypeName` which is a `NewType` of str for
422 type safety in static type checking.
424 Returns
425 -------
426 tasks : iterable of `TaskDef`
427 `TaskDef` objects that have the specified `DatasetTypeName` as an
428 input, list will be empty if no tasks use specified
429 `DatasetTypeName` as an input.
431 Raises
432 ------
433 KeyError
434 Raised if the `DatasetTypeName` is not part of the `QuantumGraph`
435 """
436 return (c for c in self._datasetDict.getConsumers(datasetTypeName))
438 def findTaskWithOutput(self, datasetTypeName: DatasetTypeName) -> Optional[TaskDef]:
439 """Find all tasks that have the specified dataset type name as an
440 output.
442 Parameters
443 ----------
444 datasetTypeName : `str`
445 A string representing the name of a dataset type to be queried,
446 can also accept a `DatasetTypeName` which is a `NewType` of str for
447 type safety in static type checking.
449 Returns
450 -------
451 `TaskDef` or `None`
452 `TaskDef` that outputs `DatasetTypeName` as an output or None if
453 none of the tasks produce this `DatasetTypeName`.
455 Raises
456 ------
457 KeyError
458 Raised if the `DatasetTypeName` is not part of the `QuantumGraph`
459 """
460 return self._datasetDict.getProducer(datasetTypeName)
462 def tasksWithDSType(self, datasetTypeName: DatasetTypeName) -> Iterable[TaskDef]:
463 """Find all tasks that are associated with the specified dataset type
464 name.
466 Parameters
467 ----------
468 datasetTypeName : `str`
469 A string representing the name of a dataset type to be queried,
470 can also accept a `DatasetTypeName` which is a `NewType` of str for
471 type safety in static type checking.
473 Returns
474 -------
475 result : iterable of `TaskDef`
476 `TaskDef` objects that are associated with the specified
477 `DatasetTypeName`
479 Raises
480 ------
481 KeyError
482 Raised if the `DatasetTypeName` is not part of the `QuantumGraph`
483 """
484 return self._datasetDict.getAll(datasetTypeName)
486 def findTaskDefByName(self, taskName: str) -> List[TaskDef]:
487 """Determine which `TaskDef` objects in this graph are associated
488 with a `str` representing a task name (looks at the taskName property
489 of `TaskDef` objects).
491 Returns a list of `TaskDef` objects as a `PipelineTask` may appear
492 multiple times in a graph with different labels.
494 Parameters
495 ----------
496 taskName : str
497 Name of a task to search for
499 Returns
500 -------
501 result : list of `TaskDef`
502 List of the `TaskDef` objects that have the name specified.
503 Multiple values are returned in the case that a task is used
504 multiple times with different labels.
505 """
506 results = []
507 for task in self._taskToQuantumNode.keys():
508 split = task.taskName.split(".")
509 if split[-1] == taskName:
510 results.append(task)
511 return results
513 def findTaskDefByLabel(self, label: str) -> Optional[TaskDef]:
514 """Determine which `TaskDef` objects in this graph are associated
515 with a `str` representing a tasks label.
517 Parameters
518 ----------
519 taskName : str
520 Name of a task to search for
522 Returns
523 -------
524 result : `TaskDef`
525 `TaskDef` objects that has the specified label.
526 """
527 for task in self._taskToQuantumNode.keys():
528 if label == task.label:
529 return task
530 return None
532 def findQuantaWithDSType(self, datasetTypeName: DatasetTypeName) -> Set[Quantum]:
533 """Return all the `Quantum` that contain a specified `DatasetTypeName`.
535 Parameters
536 ----------
537 datasetTypeName : `str`
538 The name of the dataset type to search for as a string,
539 can also accept a `DatasetTypeName` which is a `NewType` of str for
540 type safety in static type checking.
542 Returns
543 -------
544 result : `set` of `QuantumNode` objects
545 A `set` of `QuantumNode`s that contain specified `DatasetTypeName`
547 Raises
548 ------
549 KeyError
550 Raised if the `DatasetTypeName` is not part of the `QuantumGraph`
552 """
553 tasks = self._datasetDict.getAll(datasetTypeName)
554 result: Set[Quantum] = set()
555 result = result.union(quantum for task in tasks for quantum in self.getQuantaForTask(task))
556 return result
558 def checkQuantumInGraph(self, quantum: Quantum) -> bool:
559 """Check if specified quantum appears in the graph as part of a node.
561 Parameters
562 ----------
563 quantum : `Quantum`
564 The quantum to search for
566 Returns
567 -------
568 `bool`
569 The result of searching for the quantum
570 """
571 for node in self:
572 if quantum == node.quantum:
573 return True
574 return False
576 def writeDotGraph(self, output: Union[str, io.BufferedIOBase]) -> None:
577 """Write out the graph as a dot graph.
579 Parameters
580 ----------
581 output : str or `io.BufferedIOBase`
582 Either a filesystem path to write to, or a file handle object
583 """
584 write_dot(self._connectedQuanta, output)
586 def subset(self: _T, nodes: Union[QuantumNode, Iterable[QuantumNode]]) -> _T:
587 """Create a new graph object that contains the subset of the nodes
588 specified as input. Node number is preserved.
590 Parameters
591 ----------
592 nodes : `QuantumNode` or iterable of `QuantumNode`
594 Returns
595 -------
596 graph : instance of graph type
597 An instance of the type from which the subset was created
598 """
599 if not isinstance(nodes, Iterable):
600 nodes = (nodes,)
601 quantumSubgraph = self._connectedQuanta.subgraph(nodes).nodes
602 quantumMap = defaultdict(set)
604 node: QuantumNode
605 for node in quantumSubgraph:
606 quantumMap[node.taskDef].add(node.quantum)
608 # convert to standard dict to prevent accidental key insertion
609 quantumDict: Dict[TaskDef, Set[Quantum]] = dict(quantumMap.items())
610 # Create an empty graph, and then populate it with custom mapping
611 newInst = type(self)({})
612 newInst._buildGraphs(
613 quantumDict,
614 _quantumToNodeId={n.quantum: n.nodeId for n in nodes},
615 _buildId=self._buildId,
616 metadata=self._metadata,
617 )
618 return newInst
620 def subsetToConnected(self: _T) -> Tuple[_T, ...]:
621 """Generate a list of subgraphs where each is connected.
623 Returns
624 -------
625 result : list of `QuantumGraph`
626 A list of graphs that are each connected
627 """
628 return tuple(
629 self.subset(connectedSet)
630 for connectedSet in nx.weakly_connected_components(self._connectedQuanta)
631 )
633 def determineInputsToQuantumNode(self, node: QuantumNode) -> Set[QuantumNode]:
634 """Return a set of `QuantumNode` that are direct inputs to a specified
635 node.
637 Parameters
638 ----------
639 node : `QuantumNode`
640 The node of the graph for which inputs are to be determined
642 Returns
643 -------
644 set of `QuantumNode`
645 All the nodes that are direct inputs to specified node
646 """
647 return set(pred for pred in self._connectedQuanta.predecessors(node))
649 def determineOutputsOfQuantumNode(self, node: QuantumNode) -> Set[QuantumNode]:
650 """Return a set of `QuantumNode` that are direct outputs of a specified
651 node.
653 Parameters
654 ----------
655 node : `QuantumNode`
656 The node of the graph for which outputs are to be determined
658 Returns
659 -------
660 set of `QuantumNode`
661 All the nodes that are direct outputs to specified node
662 """
663 return set(succ for succ in self._connectedQuanta.successors(node))
665 def determineConnectionsOfQuantumNode(self: _T, node: QuantumNode) -> _T:
666 """Return a graph of `QuantumNode` that are direct inputs and outputs
667 of a specified node.
669 Parameters
670 ----------
671 node : `QuantumNode`
672 The node of the graph for which connected nodes are to be
673 determined.
675 Returns
676 -------
677 graph : graph of `QuantumNode`
678 All the nodes that are directly connected to specified node
679 """
680 nodes = self.determineInputsToQuantumNode(node).union(self.determineOutputsOfQuantumNode(node))
681 nodes.add(node)
682 return self.subset(nodes)
684 def determineAncestorsOfQuantumNode(self: _T, node: QuantumNode) -> _T:
685 """Return a graph of the specified node and all the ancestor nodes
686 directly reachable by walking edges.
688 Parameters
689 ----------
690 node : `QuantumNode`
691 The node for which all ansestors are to be determined
693 Returns
694 -------
695 graph of `QuantumNode`
696 Graph of node and all of its ansestors
697 """
698 predecessorNodes = nx.ancestors(self._connectedQuanta, node)
699 predecessorNodes.add(node)
700 return self.subset(predecessorNodes)
702 def findCycle(self) -> List[Tuple[QuantumNode, QuantumNode]]:
703 """Check a graph for the presense of cycles and returns the edges of
704 any cycles found, or an empty list if there is no cycle.
706 Returns
707 -------
708 result : list of tuple of `QuantumNode`, `QuantumNode`
709 A list of any graph edges that form a cycle, or an empty list if
710 there is no cycle. Empty list to so support if graph.find_cycle()
711 syntax as an empty list is falsy.
712 """
713 try:
714 return nx.find_cycle(self._connectedQuanta)
715 except nx.NetworkXNoCycle:
716 return []
718 def saveUri(self, uri: ResourcePathExpression) -> None:
719 """Save `QuantumGraph` to the specified URI.
721 Parameters
722 ----------
723 uri : convertible to `ResourcePath`
724 URI to where the graph should be saved.
725 """
726 buffer = self._buildSaveObject()
727 path = ResourcePath(uri)
728 if path.getExtension() not in (".qgraph"):
729 raise TypeError(f"Can currently only save a graph in qgraph format not {uri}")
730 path.write(buffer) # type: ignore # Ignore because bytearray is safe to use in place of bytes
732 @property
733 def metadata(self) -> Optional[MappingProxyType[str, Any]]:
734 """ """
735 if self._metadata is None:
736 return None
737 return MappingProxyType(self._metadata)
739 @classmethod
740 def loadUri(
741 cls,
742 uri: ResourcePathExpression,
743 universe: DimensionUniverse,
744 nodes: Optional[Iterable[uuid.UUID]] = None,
745 graphID: Optional[BuildId] = None,
746 minimumVersion: int = 3,
747 ) -> QuantumGraph:
748 """Read `QuantumGraph` from a URI.
750 Parameters
751 ----------
752 uri : convertible to `ResourcePath`
753 URI from where to load the graph.
754 universe: `~lsst.daf.butler.DimensionUniverse`
755 DimensionUniverse instance, not used by the method itself but
756 needed to ensure that registry data structures are initialized.
757 nodes: iterable of `int` or None
758 Numbers that correspond to nodes in the graph. If specified, only
759 these nodes will be loaded. Defaults to None, in which case all
760 nodes will be loaded.
761 graphID : `str` or `None`
762 If specified this ID is verified against the loaded graph prior to
763 loading any Nodes. This defaults to None in which case no
764 validation is done.
765 minimumVersion : int
766 Minimum version of a save file to load. Set to -1 to load all
767 versions. Older versions may need to be loaded, and re-saved
768 to upgrade them to the latest format before they can be used in
769 production.
771 Returns
772 -------
773 graph : `QuantumGraph`
774 Resulting QuantumGraph instance.
776 Raises
777 ------
778 TypeError
779 Raised if pickle contains instance of a type other than
780 QuantumGraph.
781 ValueError
782 Raised if one or more of the nodes requested is not in the
783 `QuantumGraph` or if graphID parameter does not match the graph
784 being loaded or if the supplied uri does not point at a valid
785 `QuantumGraph` save file.
788 Notes
789 -----
790 Reading Quanta from pickle requires existence of singleton
791 DimensionUniverse which is usually instantiated during Registry
792 initialization. To make sure that DimensionUniverse exists this method
793 accepts dummy DimensionUniverse argument.
794 """
795 uri = ResourcePath(uri)
796 # With ResourcePath we have the choice of always using a local file
797 # or reading in the bytes directly. Reading in bytes can be more
798 # efficient for reasonably-sized pickle files when the resource
799 # is remote. For now use the local file variant. For a local file
800 # as_local() does nothing.
802 if uri.getExtension() in (".pickle", ".pkl"):
803 with uri.as_local() as local, open(local.ospath, "rb") as fd:
804 warnings.warn("Pickle graphs are deprecated, please re-save your graph with the save method")
805 qgraph = pickle.load(fd)
806 elif uri.getExtension() in (".qgraph"):
807 with LoadHelper(uri, minimumVersion) as loader:
808 qgraph = loader.load(universe, nodes, graphID)
809 else:
810 raise ValueError("Only know how to handle files saved as `pickle`, `pkl`, or `qgraph`")
811 if not isinstance(qgraph, QuantumGraph):
812 raise TypeError(f"QuantumGraph save file contains unexpected object type: {type(qgraph)}")
813 return qgraph
815 @classmethod
816 def readHeader(cls, uri: ResourcePathExpression, minimumVersion: int = 3) -> Optional[str]:
817 """Read the header of a `QuantumGraph` pointed to by the uri parameter
818 and return it as a string.
820 Parameters
821 ----------
822 uri : convertible to `ResourcePath`
823 The location of the `QuantumGraph` to load. If the argument is a
824 string, it must correspond to a valid `ResourcePath` path.
825 minimumVersion : int
826 Minimum version of a save file to load. Set to -1 to load all
827 versions. Older versions may need to be loaded, and re-saved
828 to upgrade them to the latest format before they can be used in
829 production.
831 Returns
832 -------
833 header : `str` or `None`
834 The header associated with the specified `QuantumGraph` it there is
835 one, else `None`.
837 Raises
838 ------
839 ValueError
840 Raised if `QuantuGraph` was saved as a pickle.
841 Raised if the extention of the file specified by uri is not a
842 `QuantumGraph` extention.
843 """
844 uri = ResourcePath(uri)
845 if uri.getExtension() in (".pickle", ".pkl"):
846 raise ValueError("Reading a header from a pickle save is not supported")
847 elif uri.getExtension() in (".qgraph"):
848 return LoadHelper(uri, minimumVersion).readHeader()
849 else:
850 raise ValueError("Only know how to handle files saved as `qgraph`")
852 def buildAndPrintHeader(self) -> None:
853 """Creates a header that would be used in a save of this object and
854 prints it out to standard out.
855 """
856 _, header = self._buildSaveObject(returnHeader=True)
857 print(json.dumps(header))
859 def save(self, file: BinaryIO) -> None:
860 """Save QuantumGraph to a file.
862 Parameters
863 ----------
864 file : `io.BufferedIOBase`
865 File to write pickle data open in binary mode.
866 """
867 buffer = self._buildSaveObject()
868 file.write(buffer) # type: ignore # Ignore because bytearray is safe to use in place of bytes
870 def _buildSaveObject(self, returnHeader: bool = False) -> Union[bytearray, Tuple[bytearray, Dict]]:
871 # make some containers
872 jsonData: Deque[bytes] = deque()
873 # node map is a list because json does not accept mapping keys that
874 # are not strings, so we store a list of key, value pairs that will
875 # be converted to a mapping on load
876 nodeMap = []
877 taskDefMap = {}
878 headerData: Dict[str, Any] = {}
880 # Store the QauntumGraph BuildId, this will allow validating BuildIds
881 # at load time, prior to loading any QuantumNodes. Name chosen for
882 # unlikely conflicts.
883 headerData["GraphBuildID"] = self.graphID
884 headerData["Metadata"] = self._metadata
886 # counter for the number of bytes processed thus far
887 count = 0
888 # serialize out the task Defs recording the start and end bytes of each
889 # taskDef
890 inverseLookup = self._datasetDict.inverse
891 taskDef: TaskDef
892 # sort by task label to ensure serialization happens in the same order
893 for taskDef in self.taskGraph:
894 # compressing has very little impact on saving or load time, but
895 # a large impact on on disk size, so it is worth doing
896 taskDescription = {}
897 # save the fully qualified name.
898 taskDescription["taskName"] = get_full_type_name(taskDef.taskClass)
899 # save the config as a text stream that will be un-persisted on the
900 # other end
901 stream = io.StringIO()
902 taskDef.config.saveToStream(stream)
903 taskDescription["config"] = stream.getvalue()
904 taskDescription["label"] = taskDef.label
906 inputs = []
907 outputs = []
909 # Determine the connection between all of tasks and save that in
910 # the header as a list of connections and edges in each task
911 # this will help in un-persisting, and possibly in a "quick view"
912 # method that does not require everything to be un-persisted
913 #
914 # Typing returns can't be parameter dependent
915 for connection in inverseLookup[taskDef]: # type: ignore
916 consumers = self._datasetDict.getConsumers(connection)
917 producer = self._datasetDict.getProducer(connection)
918 if taskDef in consumers:
919 # This checks if the task consumes the connection directly
920 # from the datastore or it is produced by another task
921 producerLabel = producer.label if producer is not None else "datastore"
922 inputs.append((producerLabel, connection))
923 elif taskDef not in consumers and producer is taskDef:
924 # If there are no consumers for this tasks produced
925 # connection, the output will be said to be the datastore
926 # in which case the for loop will be a zero length loop
927 if not consumers:
928 outputs.append(("datastore", connection))
929 for td in consumers:
930 outputs.append((td.label, connection))
932 # dump to json string, and encode that string to bytes and then
933 # conpress those bytes
934 dump = lzma.compress(json.dumps(taskDescription).encode())
935 # record the sizing and relation information
936 taskDefMap[taskDef.label] = {
937 "bytes": (count, count + len(dump)),
938 "inputs": inputs,
939 "outputs": outputs,
940 }
941 count += len(dump)
942 jsonData.append(dump)
944 headerData["TaskDefs"] = taskDefMap
946 # serialize the nodes, recording the start and end bytes of each node
947 dimAccumulator = DimensionRecordsAccumulator()
948 for node in self:
949 # compressing has very little impact on saving or load time, but
950 # a large impact on on disk size, so it is worth doing
951 simpleNode = node.to_simple(accumulator=dimAccumulator)
953 dump = lzma.compress(simpleNode.json().encode())
954 jsonData.append(dump)
955 nodeMap.append(
956 (
957 str(node.nodeId),
958 {
959 "bytes": (count, count + len(dump)),
960 "inputs": [str(n.nodeId) for n in self.determineInputsToQuantumNode(node)],
961 "outputs": [str(n.nodeId) for n in self.determineOutputsOfQuantumNode(node)],
962 },
963 )
964 )
965 count += len(dump)
967 headerData["DimensionRecords"] = {
968 key: value.dict() for key, value in dimAccumulator.makeSerializedDimensionRecordMapping().items()
969 }
971 # need to serialize this as a series of key,value tuples because of
972 # a limitation on how json cant do anyting but strings as keys
973 headerData["Nodes"] = nodeMap
975 # dump the headerData to json
976 header_encode = lzma.compress(json.dumps(headerData).encode())
978 # record the sizes as 2 unsigned long long numbers for a total of 16
979 # bytes
980 save_bytes = struct.pack(STRUCT_FMT_BASE, SAVE_VERSION)
982 fmt_string = DESERIALIZER_MAP[SAVE_VERSION].FMT_STRING()
983 map_lengths = struct.pack(fmt_string, len(header_encode))
985 # write each component of the save out in a deterministic order
986 # buffer = io.BytesIO()
987 # buffer.write(map_lengths)
988 # buffer.write(taskDef_pickle)
989 # buffer.write(map_pickle)
990 buffer = bytearray()
991 buffer.extend(MAGIC_BYTES)
992 buffer.extend(save_bytes)
993 buffer.extend(map_lengths)
994 buffer.extend(header_encode)
995 # Iterate over the length of pickleData, and for each element pop the
996 # leftmost element off the deque and write it out. This is to save
997 # memory, as the memory is added to the buffer object, it is removed
998 # from from the container.
999 #
1000 # Only this section needs to worry about memory pressue because
1001 # everything else written to the buffer prior to this pickle data is
1002 # only on the order of kilobytes to low numbers of megabytes.
1003 while jsonData:
1004 buffer.extend(jsonData.popleft())
1005 if returnHeader:
1006 return buffer, headerData
1007 else:
1008 return buffer
1010 @classmethod
1011 def load(
1012 cls,
1013 file: BinaryIO,
1014 universe: DimensionUniverse,
1015 nodes: Optional[Iterable[uuid.UUID]] = None,
1016 graphID: Optional[BuildId] = None,
1017 minimumVersion: int = 3,
1018 ) -> QuantumGraph:
1019 """Read QuantumGraph from a file that was made by `save`.
1021 Parameters
1022 ----------
1023 file : `io.IO` of bytes
1024 File with pickle data open in binary mode.
1025 universe: `~lsst.daf.butler.DimensionUniverse`
1026 DimensionUniverse instance, not used by the method itself but
1027 needed to ensure that registry data structures are initialized.
1028 nodes: iterable of `int` or None
1029 Numbers that correspond to nodes in the graph. If specified, only
1030 these nodes will be loaded. Defaults to None, in which case all
1031 nodes will be loaded.
1032 graphID : `str` or `None`
1033 If specified this ID is verified against the loaded graph prior to
1034 loading any Nodes. This defaults to None in which case no
1035 validation is done.
1036 minimumVersion : int
1037 Minimum version of a save file to load. Set to -1 to load all
1038 versions. Older versions may need to be loaded, and re-saved
1039 to upgrade them to the latest format before they can be used in
1040 production.
1042 Returns
1043 -------
1044 graph : `QuantumGraph`
1045 Resulting QuantumGraph instance.
1047 Raises
1048 ------
1049 TypeError
1050 Raised if pickle contains instance of a type other than
1051 QuantumGraph.
1052 ValueError
1053 Raised if one or more of the nodes requested is not in the
1054 `QuantumGraph` or if graphID parameter does not match the graph
1055 being loaded or if the supplied uri does not point at a valid
1056 `QuantumGraph` save file.
1058 Notes
1059 -----
1060 Reading Quanta from pickle requires existence of singleton
1061 DimensionUniverse which is usually instantiated during Registry
1062 initialization. To make sure that DimensionUniverse exists this method
1063 accepts dummy DimensionUniverse argument.
1064 """
1065 # Try to see if the file handle contains pickle data, this will be
1066 # removed in the future
1067 try:
1068 qgraph = pickle.load(file)
1069 warnings.warn("Pickle graphs are deprecated, please re-save your graph with the save method")
1070 except pickle.UnpicklingError:
1071 with LoadHelper(file, minimumVersion) as loader:
1072 qgraph = loader.load(universe, nodes, graphID)
1073 if not isinstance(qgraph, QuantumGraph):
1074 raise TypeError(f"QuantumGraph pickle file has contains unexpected object type: {type(qgraph)}")
1075 return qgraph
1077 def iterTaskGraph(self) -> Generator[TaskDef, None, None]:
1078 """Iterate over the `taskGraph` attribute in topological order
1080 Yields
1081 ------
1082 taskDef : `TaskDef`
1083 `TaskDef` objects in topological order
1084 """
1085 yield from nx.topological_sort(self.taskGraph)
1087 @property
1088 def graphID(self) -> BuildId:
1089 """Returns the ID generated by the graph at construction time"""
1090 return self._buildId
1092 def __iter__(self) -> Generator[QuantumNode, None, None]:
1093 yield from nx.topological_sort(self._connectedQuanta)
1095 def __len__(self) -> int:
1096 return self._count
1098 def __contains__(self, node: QuantumNode) -> bool:
1099 return self._connectedQuanta.has_node(node)
1101 def __getstate__(self) -> dict:
1102 """Stores a compact form of the graph as a list of graph nodes, and a
1103 tuple of task labels and task configs. The full graph can be
1104 reconstructed with this information, and it preseves the ordering of
1105 the graph ndoes.
1106 """
1107 universe: Optional[DimensionUniverse] = None
1108 for node in self:
1109 dId = node.quantum.dataId
1110 if dId is None:
1111 continue
1112 universe = dId.graph.universe
1113 return {"reduced": self._buildSaveObject(), "graphId": self._buildId, "universe": universe}
1115 def __setstate__(self, state: dict) -> None:
1116 """Reconstructs the state of the graph from the information persisted
1117 in getstate.
1118 """
1119 buffer = io.BytesIO(state["reduced"])
1120 with LoadHelper(buffer, minimumVersion=3) as loader:
1121 qgraph = loader.load(state["universe"], graphID=state["graphId"])
1123 self._metadata = qgraph._metadata
1124 self._buildId = qgraph._buildId
1125 self._datasetDict = qgraph._datasetDict
1126 self._nodeIdMap = qgraph._nodeIdMap
1127 self._count = len(qgraph)
1128 self._taskToQuantumNode = qgraph._taskToQuantumNode
1129 self._taskGraph = qgraph._taskGraph
1130 self._connectedQuanta = qgraph._connectedQuanta
1132 def __eq__(self, other: object) -> bool:
1133 if not isinstance(other, QuantumGraph):
1134 return False
1135 if len(self) != len(other):
1136 return False
1137 for node in self:
1138 if node not in other:
1139 return False
1140 if self.determineInputsToQuantumNode(node) != other.determineInputsToQuantumNode(node):
1141 return False
1142 if self.determineOutputsOfQuantumNode(node) != other.determineOutputsOfQuantumNode(node):
1143 return False
1144 if set(self.allDatasetTypes) != set(other.allDatasetTypes):
1145 return False
1146 return set(self.taskGraph) == set(other.taskGraph)