tests/gen-casefold-txt.py

   1 #!/usr/bin/env python3
   2 # Copyright (C) 1998, 1999 Tom Tromey
   3 # Copyright (C) 2001 Red Hat Software
   4 #
   5 # This program is free software; you can redistribute it and/or modify
   6 # it under the terms of the GNU General Public License as published by
   7 # the Free Software Foundation; either version 2, or (at your option)
   8 # any later version.
   9 #
  10 # This program is distributed in the hope that it will be useful,
  11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  13 # GNU General Public License for more details.
  14 #
  15 # You should have received a copy of the GNU General Public License
  16 # along with this program; if not, see <http://www.gnu.org/licenses/>.
  17
  18 """
  19 gen-casefold-txt.py - Generate test cases for casefolding from Unicode data.
  20 See http://www.unicode.org/Public/UNIDATA/UnicodeCharacterDatabase.html
  21 Usage:
  22     I consider the output of this program to be unrestricted.
  23     Use it as you will.
  24 """
  25
  26 import sys
  27 import argparse
  28
  29
  30 def main(argv):
  31     parser = argparse.ArgumentParser(
  32         description="Generate test cases for casefolding from Unicode data")
  33     parser.add_argument("UNICODE-VERSION")
  34     parser.add_argument("CaseFolding.txt")
  35     args = parser.parse_args(argv[1:])
  36     version = getattr(args, "UNICODE-VERSION")
  37     filename = getattr(args, "CaseFolding.txt")
  38
  39     print("""\
  40 # Test cases generated from Unicode {} data
  41 # by gen-casefold-txt.py. Do not edit.
  42 #
  43 # Some special hand crafted tests
  44 #
  45 AaBbCc@@\taabbcc@@
  46 #
  47 # Now the automatic tests
  48 #""".format(version))
  49
  50     # Names of fields in the CaseFolding table
  51     CODE, STATUS, MAPPING = range(3)
  52
  53     with open(filename, encoding="utf-8") as fileobj:
  54         for line in fileobj:
  55             # strip comments and skip empty lines
  56             line = line.split("#", 1)[0].strip()
  57             if not line:
  58                 continue
  59
  60             fields = [f.strip() for f in line.split(";", 3)[:3]]
  61             if len(fields) != 3:
  62                 raise SystemExit(
  63                     "Entry for %s has wrong number of fields (%d)" % (
  64                         fields[CODE], len(fields)))
  65
  66             status = fields[STATUS]
  67             # skip simple and Turkic mappings
  68             if status in "ST":
  69                 continue
  70
  71             code = chr(int(fields[CODE], 16))
  72             values = "".join(
  73                 [chr(int(v, 16)) for v in fields[MAPPING].split()])
  74             print("{}\t{}".format(code, values))
  75
  76
  77 if __name__ == "__main__":
  78     sys.exit(main(sys.argv))