SOC-Observation-Code/tools/removeiso.py

#!/usr/bin/python3
"""
SPDX-FileCopyrightText: 2023 Ämin Baumeler <amin@indyfac.ch> and
Eleftherios-Ermis Tselentis <eleftheriosermis.tselentis@oeaw.ac.at>

SPDX-License-Identifier: GPL-3.0-or-later
"""
import sys
import getopt
import numpy as np
import networkx as nx

"""Parse command line arguments first."""
helpstr = """Usage: """+sys.argv[0]+""" -f <filename>

Filters out the isomorphic graphs in `filename'."""

FILENAME = False

try:
    opts, args = getopt.getopt(sys.argv[1:], "hf:")
except getopt.GetoptError:
    print(helpstr, file=sys.stderr)
    sys.exit(2)
for opt, arg in opts:
    if opt == '-h':
        print(helpstr)
        sys.exit()
    elif opt == '-f':
        FILENAME = arg

if not FILENAME:
    print(helpstr, file=sys.stderr)
    sys.exit(2)


with open(FILENAME, 'rb') as f:
    num_lines = sum(1 for _ in f)
print(f"Total number of digraphs is {num_lines}", file=sys.stderr)


nonisodigraphs = set()

cnt = 0
fnd = 0
with open(FILENAME, 'r') as f:
    while 1:
        cnt = cnt + 1
        print("\r{:.2f}%".format(100*cnt/num_lines), end='', file=sys.stderr)
        line = f.readline().strip('{}\n')
        if not line:
            break
        G = nx.from_numpy_matrix(np.vectorize(int)(np.matrix([x.split(',') for x in line.split('},{')])), create_using=nx.DiGraph)

        couldbeold = False
        for H in nonisodigraphs:
            # Returns False if graphs are definitely not isomorphic.
            # True does NOT guarantee isomorphism.
            if nx.faster_could_be_isomorphic(G, H):
                couldbeold = True
                break

        if not couldbeold:
            nonisodigraphs.add(G)
            fnd = fnd + 1
            print(f"{{{{{line}}}}}")
            continue

        new = True
        for H in nonisodigraphs:
            if nx.is_isomorphic(G, H):
                new = False
                break
        if new:
            nonisodigraphs.add(G)
            fnd = fnd + 1
            print(f"{{{{{line}}}}}")
    print(f"\r100%  Found {fnd} non-isomoric graphs", file=sys.stderr)