glibc/locale/gen-translit.py

178 lines
5.1 KiB
Python

#!/usr/bin/python3
# Generate the locale/C-translit.h file.
# Copyright (C) 2018-2019 Free Software Foundation, Inc.
# This file is part of the GNU C Library.
#
# The GNU C Library is free software; you can redistribute it and/or
# modify it under the terms of the GNU Lesser General Public
# License as published by the Free Software Foundation; either
# version 2.1 of the License, or (at your option) any later version.
#
# The GNU C Library is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public
# License along with the GNU C Library; if not, see
# <http://www.gnu.org/licenses/>.
import re
import sys
class StringLiteral:
"Source of a string literal and its decomposition into code points."
def __init__(self, s):
# States:
# 0 regular character sequence
# 1 backslash seen
# 2 in hexadecimal escape sequence
state = 0
result = []
for ch in s:
if state == 0:
if ch == '\\':
state = 1
else:
result.append(ord(ch))
elif state == 1:
if ch in "\\\"":
result.append(ord(ch))
state = 0
elif ch == 'x':
state = 2
result.append(0)
else:
raise ValueError("invalid character {!r} in {!r}".format(
ch, s))
elif state == 2:
if ch in "0123456789abcdefABCDEF":
result[-1] = result[-1] * 16 + int(ch, 16)
else:
if ch == '\\':
state = 1
else:
state = 0
if state == 1:
raise ValueError("trailing backslash in {!r}".format(s))
self.source = s
self.decoded = tuple(result)
class Translit:
"Pair of transliteration and source."
__RE_TRANSLIT = re.compile(
r'^"((?:[^"\\]|\\x[0-9a-fA-F])+)"\s+'
r'"((?:[^"\\]|\\["\\])*)"\s*(?:#.*)?$')
def __init__(self, line):
match = self.__RE_TRANSLIT.match(line)
if not match:
raise IOError("invalid line {}: {!r}".format(
lineno + 1, line))
codepoints, replacement = match.groups()
self.codepoints = StringLiteral(codepoints)
self.replacement = StringLiteral(replacement)
# List of Translit objects.
translits = []
# Read transliterations from standard input.
for lineno, line in enumerate(sys.stdin):
line = line.strip()
# Skip empty lines and comments.
if (not line) or line[0] == '#':
continue
translit = Translit(line)
# Check ordering of codepoints.
if translits \
and translit.codepoints.decoded <= translits[-1].codepoints.decoded:
raise IOError("unexpected codepoint {!r} on line {}: {!r}".format(
translit.codeponts.decoded, lineno + 1, line))
translits.append(translit)
# Generate the C sources.
write = sys.stdout.write
write("#include <stdint.h>\n")
write("#define NTRANSLIT {}\n".format(len(translits)))
write("static const uint32_t translit_from_idx[] =\n{\n ")
col = 2
total = 0
for translit in translits:
if total > 0:
if col + 7 >= 79:
write(",\n ")
col = 2
else:
write(", ")
col += 2
write("{:4}".format(total))
total += len(translit.codepoints.decoded) + 1
col += 4
write("\n};\n")
write("static const wchar_t translit_from_tbl[] =\n ")
col = 1
first = True
for translit in translits:
if first:
first = False
else:
if col + 6 >= 79:
write("\n ")
col = 1
write(" L\"\\0\"")
col += 6
if col > 2 and col + len(translit.codepoints.source) + 4 >= 79:
write("\n ")
col = 2
else:
write(" ")
col += 1
write("L\"{}\"".format(translit.codepoints.source))
col += len(translit.codepoints.source) + 3
write(";\n")
write("static const uint32_t translit_to_idx[] =\n{\n ")
col = 2
total = 0
for translit in translits:
if total > 0:
if col + 7 >= 79:
write(",\n ")
col = 2
else:
write(", ")
col += 2
write("{:4}".format(total))
total += len(translit.replacement.decoded) + 2
col += 4
write("\n};\n")
write("static const wchar_t translit_to_tbl[] =\n ")
col = 1
first = True
for translit in translits:
if first:
first = False
else:
if col + 6 >= 79:
write("\n ")
col = 1
write(" L\"\\0\"")
col += 6
if col > 2 and col + len(translit.replacement.source) + 6 >= 79:
write("\n ")
col = 2
else:
write(" ")
col += 1
write("L\"{}\\0\"".format(translit.replacement.source))
col += len(translit.replacement.source) + 5
write(";\n")