FM to BDD

This transformation converts a feature model into a Binary Decision Diagram (BDD) representation. BDDs are efficient for representing and manipulating logical functions, making them ideal for feature model analysis.

flamapy relies on file extensions to identify the required transformation. Ensure you specify the correct file extension for this transformation.

---

Code Examples

from flamapy.metamodels.fm_metamodel.transformations.featureide_reader import FeatureIDEReader
from flamapy.metamodels.bdd_metamodel.transformations.fm_to_bdd import FmToBDD

# Load the feature model from a FeatureIDE file
feature_model = FeatureIDEReader('path/to/feature/model.xml').transform()

# Transform the feature model into a BDD representation
bdd_model = FmToBDD(feature_model).transform()

print(bdd_model)

---

Building large BDDs

The size of a BDD, and the cost of building it, depend on the order of its variables. FmToBDD builds the BDD clause by clause over a feature-tree variable order. With dd built against CUDD it also reorders variables dynamically. A native builder (Logic2BDD on CUDD, by D. Fernández-Amorós, UNED) is in preparation and not part of 2.9. These options can be passed to FmToBDD(...) or set with environment variables:

Option	Environment variable	Values (default first)
builder	FLAMAPY_BDD_BUILDER	python; native and auto are experimental
var_order	FLAMAPY_BDD_VAR_ORDER	tree, force, model
build	FLAMAPY_BDD_BUILD	incremental, monolithic (Python builder)
clause_order	FLAMAPY_BDD_CLAUSE_ORDER	minspan, minmax, model (Python builder)
reorder	FLAMAPY_BDD_REORDER	auto, 1, 0 (Python builder)

bdd_model = FmToBDD(feature_model, builder="python", var_order="force").transform()

The options change how the BDD is built, never the configurations it represents.