#38: generate a WfInstance from a .dot file

requirements.txt

@@ -1,6 +1,6 @@
 jsonschema~=3.2.0
 matplotlib>=3.3.0
-networkx>=2.4
+networkx>=3.0
 numpy>=1.19.1
 python-dateutil>=2.8.1
 requests>=2.24.0

wfcommons/common/workflow.py

@@ -192,7 +192,7 @@ def read_dot(self, dot_file_path: Optional[pathlib.Path] = None) -> None:
 
  tasks_map = {}
  for node in graph.nodes(data=True):
- task_name = f"{node[1]['label']}_{node[0]}"
+ task_name = f"{node[1]['label']}_ID{node[0]}"
  task = Task(name=task_name, task_type=TaskType.COMPUTE, runtime=0, task_id=node[0])
  self.add_task(task)
  tasks_map[node[0]] = task_name

wfcommons/wfchef/duplicate.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 #
-# Copyright (c) 2021 The WfCommons Team.
+# Copyright (c) 2021-2024 The WfCommons Team.
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -92,25 +92,38 @@ def duplicate(path: pathlib.Path,
  f"Cannot create synthentic graph with {num_nodes} nodes from base graph with {graph.order()} nodes")
 
  all_microstructures = json.loads(base_path.joinpath("microstructures.json").read_text())
- microstructures, freqs = map(list, zip(*[(ms, ms["frequency"]) for ms_hash, ms in all_microstructures.items()]))
-
- p: List[float] = (np.array(freqs) / np.sum(freqs)).tolist()
- while graph.order() < num_nodes and microstructures:
- i = random.choice(range(len(microstructures)))
- ms = microstructures[i]
- while ms["nodes"]:
- j = random.choice(range(len(ms["nodes"])))
- structure = ms["nodes"][j]
- if graph.order() + len(structure) > num_nodes:
- del ms["nodes"][j]
- else:
+ try:
+ microstructures, freqs = map(list, zip(*[(ms, ms["frequency"]) for ms_hash, ms in all_microstructures.items()]))
+
+ p: List[float] = (np.array(freqs) / np.sum(freqs)).tolist()
+ while graph.order() < num_nodes and microstructures:
+ i = random.choice(range(len(microstructures)))
+ ms = microstructures[i]
+ while ms["nodes"]:
+ j = random.choice(range(len(ms["nodes"])))
+ structure = ms["nodes"][j]
+ if graph.order() + len(structure) > num_nodes:
+ del ms["nodes"][j]
+ else:
+ break
+
+ if not ms["nodes"]: # delete microstructure
+ del microstructures[i]
+ del p[i]
+ continue
+
+ duplicate_nodes(graph, structure)
+
+ except ValueError as e:
+ nodes = graph.copy().nodes
+ while graph.order() < num_nodes:
+ structure = set[str]()
+ if graph.order() + len(nodes) - 2 > num_nodes:
  break
+ for node in nodes:
+ if str(node) not in ["SRC", "DST"]:
+ structure.add(node)
 
- if not ms["nodes"]: # delete microstructure
- del microstructures[i]
- del p[i]
- continue
-
- duplicate_nodes(graph, structure)
+ duplicate_nodes(graph, structure)
 
  return graph

wfcommons/wfchef/find_microstructures.py

@@ -1,26 +1,26 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 #
-# Copyright (c) 2021 The WfCommons Team.
+# Copyright (c) 2021-2024 The WfCommons Team.
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
 
-import networkx as nx
-from hashlib import sha256
-from typing import Union, Set, Optional, Dict, Hashable, List
-from uuid import uuid4
-import pathlib
 import json
-from itertools import product
-from networkx.readwrite import read_gpickle, write_gpickle
+import math
+import networkx as nx
 import numpy as np
+import pathlib
+import pickle
+
+from hashlib import sha256
 from itertools import chain, combinations
-import argparse
-from .utils import create_graph, string_hash, type_hash, combine_hashes, annotate, draw
-import math
+from typing import Union, Set, Optional, Dict, List
+from uuid import uuid4
+
+from .utils import create_graph, combine_hashes, annotate, draw
 
 this_dir = pathlib.Path(__file__).resolve().parent
 
@@ -233,7 +233,8 @@ def save_microstructures(workflow_path: Union[pathlib.Path],
  g_savedir.mkdir(exist_ok=True, parents=True)
 
  base_graph_path = g_savedir.joinpath("base_graph.pickle")
- write_gpickle(graph, str(base_graph_path))
+ with open(base_graph_path, 'wb') as pf:
+ pickle.dump(graph, pf, pickle.HIGHEST_PROTOCOL)
  summary["base_graphs"][graph.name] = {
  "size": graph.size(),
  "order": graph.order()