crates/regex-automata/data/tests/fowler/fowler-to-toml

#!/usr/bin/env python

from __future__ import absolute_import, division, print_function
import argparse
import os.path as path


def read_tests(f):
    basename, _ = path.splitext(path.basename(f))
    tests = []
    prev_pattern = None

    for lineno, line in enumerate(open(f), 1):
        fields = list(filter(None, map(str.strip, line.split('\t'))))
        if not (4 <= len(fields) <= 5) \
           or 'E' not in fields[0] or fields[0][0] == '#':
            continue

        terse_opts, pat, text, sgroups = fields[0:4]
        groups = []  # groups as integer ranges
        if sgroups == 'NOMATCH':
            groups = []
        elif ',' in sgroups:
            noparen = map(lambda s: s.strip('()'), sgroups.split(')('))
            for g in noparen:
                s, e = map(str.strip, g.split(','))
                groups.append([int(s), int(e)])
                break
        else:
            # This skips tests that should result in an error.
            # There aren't many, so I think we can just capture those
            # manually. Possibly fix this in future.
            continue

        opts = []
        if text == "NULL":
            text = ""
        if pat == 'SAME':
            pat = prev_pattern
        if '$' in terse_opts:
            pat = pat.encode('utf-8').decode('unicode_escape')
            text = text.encode('utf-8').decode('unicode_escape')
            text = text.encode('unicode_escape').decode('utf-8')
            opts.append('escaped')
        else:
            opts.append('escaped')
            text = text.encode('unicode_escape').decode('utf-8')
        if 'i' in terse_opts:
            opts.append('case-insensitive')

        pat = pat.encode('unicode_escape').decode('utf-8')
        pat = pat.replace('\\\\', '\\')
        tests.append({
            'name': '"%s%d"' % (basename, lineno),
            'options': repr(opts),
            'pattern': "'''%s'''" % pat,
            'input': "'''%s'''" % text,
            'matches': str(groups),
        })
        prev_pattern = pat
    return tests


if __name__ == '__main__':
    parser = argparse.ArgumentParser(
        description='Generate match tests from an AT&T POSIX test file.')
    aa = parser.add_argument
    aa('datfile', help='A dat AT&T POSIX test file.')
    args = parser.parse_args()

    tests = read_tests(args.datfile)
    for t in tests:
        print('[[tests]]')
        for k, v in t.items():
            print('%s = %s' % (k, v))
        print('')