Checks for "conflicting" node IDs defined multiple times in a DOT file

check_for_conflicting_node_ids.py

#! /usr/bin/env python
#
# Scans through a jumboDBG / LJA output DOT file; looks for cases where
# the same node is "defined" on multiple lines. This can be caused by the
# same truncated node ID being misused across lines.
#
# USAGE:
# ./check_for_conflicting_node_ids.py graph.dot
#
# Note that this assumes that the input graph was output by jumboDBG / LJA --
# that is, all nodes are defined on lines that include the word "filled".
# If this is not the case for your graph, this will not notice certain nodes.

import sys
from collections import Counter

if len(sys.argv) != 2:
    raise ValueError("This program only needs one argument (a DOT file)")

graph_fn = sys.argv[1]
print(f"Looking in DOT file {graph_fn}...")

seen_nodes = []
with open(graph_fn, "r") as f:
    for line in f:
        # HACK -- depends on jumboDBG / LJA always labelling nodes as "style=filled"
        if "filled" in line:
            nodename = line.split(" ")[0]
            seen_nodes.append(nodename)
c = Counter(seen_nodes)
multi = [x for x in c if c[x] > 1]
print(f"Saw {len(seen_nodes):,} node line(s) in the DOT file.")
if len(multi) > 0:
    print(f"{len(multi):,} nodes occur more than 1 time:")
    print(multi)
else:
    print("Phew, no nodes occur more than 1 time.")