L2N Parsing

After running klayout's LayoutToNetlist extraction, parse_l2n() converts the result into a clean, JSON-serializable Python dictionary. This makes it easy to inspect, store, or post-process extraction results without working directly with klayout's internal objects.

Functions

Function	Returns	Purpose
`parse_l2n(l2n, ...)`	`dict`	Convert L2N to a structured dict
`l2n_to_json(l2n, ...)`	`str`	Convenience wrapper returning a JSON string

Both accept the same keyword arguments.

Basic usage

from klayout import db as kdb
from kfnetlist.extract import parse_l2n, l2n_to_json

# After extraction:
result = parse_l2n(l2n)

print(result["top_circuit"])    # name of the top cell
print(result["dbu"])            # database unit in microns
print(result["layers"])         # list of layer dicts
print(result["circuits"])       # dict of circuit name -> circuit data

Each circuit entry contains:

pins — top-level port names on the circuit
subcircuits — child instances with name, circuit_ref, and transform
nets — connectivity, with name, pins, subcircuit_pins, and optionally layer_to_polygons / layer_to_holes for per-net shape data

Parameters

Parameter	Type	Default	Description
`l2n`	`kdb.LayoutToNetlist`	(required)	Completed L2N extraction
`flatten`	`bool`	`False`	Collapse hierarchy into the top cell
`include_layers`	`Sequence[LayerInfo] \\| None`	`None`	Keep only nets on these layers
`exclude_layers`	`Sequence[LayerInfo] \\| None`	`None`	Drop nets only on these layers
`include_instances`	`Sequence[str] \\| None`	`None`	Keep only these subcircuit refs
`exclude_instances`	`Sequence[str] \\| None`	`None`	Remove these subcircuit refs

Note

Layer and instance filtering are ignored when flatten=True, since the flattened netlist is detached from the L2N shape data.

Flattening

When flatten=True, all sub-circuits are collapsed into the top cell. The resulting dict has a single entry in circuits and no per-net layer geometry:

flat = parse_l2n(l2n, flatten=True)
assert list(flat["circuits"].keys()) == ["TOP"]

JSON output

l2n_to_json() is a thin wrapper that calls parse_l2n() and serializes the result with json.dumps():

json_str = l2n_to_json(l2n, indent=2)

Topic	Where
Electrical extraction	Electrical L2N
Short detection on L2N results	Short Detection
Extraction pipeline overview	Overview