ima-preparation/bug-2022/label_feature_vectors.py

#!/usr/bin/env python3
import glob
import os
import re

import javalang
import pandas as pd

# God class if:
# |M(C)| > E(M) + 6*V(M)
# (number of methods greater than average across all classes plus 6 times the
# standard deviation)

DIR: str = os.path.dirname(os.path.realpath(__file__))
LABEL_DIR: str = DIR + '/defects4j/framework/projects/Closure/modified_classes'
IN_DIR: str = DIR + '/metrics'
OUT_DIR: str = DIR + '/metrics'


def clean_output():
    filelist = glob.glob(OUT_DIR + '/*_labeled.csv')
    for f in filelist:
        os.remove(f)


def get_is_buggy_set() -> set[str]:
    lines: set[str] = set()

    filelist = glob.glob(LABEL_DIR + '/*.src')
    for f in filelist:
        with open(f, 'r') as file_obj:
            for line in file_obj.readlines():
                lines.add(line.strip().strip('\n'))

    return lines


def main():
    is_buggy_set = get_is_buggy_set()
    df = pd.read_csv(IN_DIR + '/feature_vectors.csv')
    df['buggy'] = df['class_name'].apply(
        lambda fqdn: 1 if fqdn in is_buggy_set else 0)
    df.to_csv(OUT_DIR + '/feature_vectors_labeled.csv', index=False)


if __name__ == '__main__':
    main()