libcxx/utils/generate_extended_grapheme_cluster_test.py

   1 #!/usr/bin/env python
   2 # ===----------------------------------------------------------------------===##
   3 #
   4 # Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
   5 # See https://llvm.org/LICENSE.txt for license information.
   6 # SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
   7 #
   8 # ===----------------------------------------------------------------------===##
   9
  10 # The code is based on
  11 # https://github.com/microsoft/STL/blob/main/tools/unicode_properties_parse/grapheme_break_test_data_gen.py
  12 #
  13 # Copyright (c) Microsoft Corporation.
  14 # SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
  15
  16 from pathlib import Path
  17 from dataclasses import dataclass, field
  18 from typing import Optional, TextIO
  19 from array import array
  20 import sys
  21
  22
  23 @dataclass
  24 class BreakTestItem:
  25     code_points: list[int] = field(default_factory=list)
  26     encoded: str = ""
  27     breaks_utf8: list[int] = field(default_factory=list)
  28     breaks_utf16: list[int] = field(default_factory=list)
  29     breaks_utf32: list[int] = field(default_factory=list)
  30
  31
  32 class CommentLine:
  33     pass
  34
  35
  36 class EOF:
  37     pass
  38
  39
  40 def parseBreakTestLine(input: TextIO) -> Optional[BreakTestItem]:
  41     result = BreakTestItem()
  42     code_point = -1
  43     utf8 = 0
  44     utf16 = 0
  45     utf32 = 0
  46
  47     while True:
  48         c = input.read(1)
  49         if c == "\N{DIVISION SIGN}":
  50             # The line starts with a division sign, don't add it to the output.
  51             if code_point != -1:
  52                 result.code_points.append(code_point)
  53                 code_point = -1
  54                 result.breaks_utf8.append(utf8)
  55                 result.breaks_utf16.append(utf16)
  56                 result.breaks_utf32.append(utf32)
  57
  58             assert input.read(1).isspace()
  59             continue
  60         if c == "\N{MULTIPLICATION SIGN}":
  61             assert input.read(1).isspace()
  62             continue
  63         if c.isalnum():
  64             while next := input.read(1):
  65                 if next.isalnum():
  66                     c += next
  67                 else:
  68                     assert next.isspace()
  69                     break
  70             i = int(c, base=16)
  71             if code_point == -1:
  72                 code_point = i
  73
  74             result.encoded += f"\\U{i:08x}"
  75             c = chr(i)
  76             utf8 += c.encode().__len__()
  77             # Since we only care about the number of code units the byte order
  78             # doesn't matter. The byte order is specified to avoid the BOM
  79             utf16 += int(c.encode("utf-16-le").__len__() / 2)
  80             utf32 += int(c.encode("utf-32-le").__len__() / 4)
  81             continue
  82         if c == "#":
  83             input.readline()
  84             return result
  85         if c == "\n":
  86             return result
  87         if c == "":
  88             return None
  89         assert False
  90
  91
  92 cpp_template = """// -*- C++ -*-
  93 //===----------------------------------------------------------------------===//
  94 //
  95 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
  96 // See https://llvm.org/LICENSE.txt for license information.
  97 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
  98 //
  99 //===----------------------------------------------------------------------===//
 100
 101 // WARNING, this entire header is generated by
 102 // utiles/generate_extended_grapheme_cluster_test.py
 103 // DO NOT MODIFY!
 104
 105 // UNICODE, INC. LICENSE AGREEMENT - DATA FILES AND SOFTWARE
 106 //
 107 // See Terms of Use <https://www.unicode.org/copyright.html>
 108 // for definitions of Unicode Inc.'s Data Files and Software.
 109 //
 110 // NOTICE TO USER: Carefully read the following legal agreement.
 111 // BY DOWNLOADING, INSTALLING, COPYING OR OTHERWISE USING UNICODE INC.'S
 112 // DATA FILES ("DATA FILES"), AND/OR SOFTWARE ("SOFTWARE"),
 113 // YOU UNEQUIVOCALLY ACCEPT, AND AGREE TO BE BOUND BY, ALL OF THE
 114 // TERMS AND CONDITIONS OF THIS AGREEMENT.
 115 // IF YOU DO NOT AGREE, DO NOT DOWNLOAD, INSTALL, COPY, DISTRIBUTE OR USE
 116 // THE DATA FILES OR SOFTWARE.
 117 //
 118 // COPYRIGHT AND PERMISSION NOTICE
 119 //
 120 // Copyright (c) 1991-2022 Unicode, Inc. All rights reserved.
 121 // Distributed under the Terms of Use in https://www.unicode.org/copyright.html.
 122 //
 123 // Permission is hereby granted, free of charge, to any person obtaining
 124 // a copy of the Unicode data files and any associated documentation
 125 // (the "Data Files") or Unicode software and any associated documentation
 126 // (the "Software") to deal in the Data Files or Software
 127 // without restriction, including without limitation the rights to use,
 128 // copy, modify, merge, publish, distribute, and/or sell copies of
 129 // the Data Files or Software, and to permit persons to whom the Data Files
 130 // or Software are furnished to do so, provided that either
 131 // (a) this copyright and permission notice appear with all copies
 132 // of the Data Files or Software, or
 133 // (b) this copyright and permission notice appear in associated
 134 // Documentation.
 135 //
 136 // THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF
 137 // ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
 138 // WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 139 // NONINFRINGEMENT OF THIRD PARTY RIGHTS.
 140 // IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN THIS
 141 // NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL
 142 // DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
 143 // DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
 144 // TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
 145 // PERFORMANCE OF THE DATA FILES OR SOFTWARE.
 146 //
 147 // Except as contained in this notice, the name of a copyright holder
 148 // shall not be used in advertising or otherwise to promote the sale,
 149 // use or other dealings in these Data Files or Software without prior
 150 // written authorization of the copyright holder.
 151
 152 #ifndef LIBCXX_TEST_STD_UTILITIES_FORMAT_FORMAT_STRING_FORMAT_STRING_STD_EXTENDED_GRAPHEME_CLUSTER_H
 153 #define LIBCXX_TEST_STD_UTILITIES_FORMAT_FORMAT_STRING_FORMAT_STRING_STD_EXTENDED_GRAPHEME_CLUSTER_H
 154
 155 #include <array>
 156 #include <string_view>
 157 #include <vector>
 158
 159 template <class CharT>
 160 struct data {{
 161   /// The input to parse.
 162   std::basic_string_view<CharT> input;
 163
 164   /// The first code point all extended grapheme clusters in the input.
 165   std::vector<char32_t> code_points;
 166
 167   /// The offset of the last code units of the extended grapheme clusters in the input.
 168   ///
 169   /// The vector has the same number of entries as \\ref code_points.
 170   std::vector<size_t> breaks;
 171 }};
 172
 173 /// The data for UTF-8.
 174 std::array<data<char>, {0}> data_utf8 = {{{{
 175 {1}}}}};
 176
 177 /// The data for UTF-16.
 178 ///
 179 /// Note that most of the data for the UTF-16 and UTF-32 are identical. However
 180 /// since the size of the code units differ the breaks can contain different
 181 /// values.
 182 std::array<data<wchar_t>, {0}> data_utf16 = {{{{
 183 {2}}}}};
 184
 185 /// The data for UTF-8.
 186 ///
 187 /// Note that most of the data for the UTF-16 and UTF-32 are identical. However
 188 /// since the size of the code units differ the breaks can contain different
 189 /// values.
 190 std::array<data<wchar_t>, {0}> data_utf32 = {{{{
 191 {3}}}}};
 192
 193 #endif // LIBCXX_TEST_STD_UTILITIES_FORMAT_FORMAT_STRING_FORMAT_STRING_STD_EXTENDED_GRAPHEME_CLUSTER_H"""
 194
 195 cpp_test_data_line_template = "     {{{}, {{{}}}, {{{}}}}}"
 196
 197
 198 def lineToCppDataLineUtf8(line: BreakTestItem) -> str:
 199     return cpp_test_data_line_template.format(
 200         f'"{line.encoded}"',
 201         ", ".join([str(x) for x in line.code_points]),
 202         ", ".join([str(x) for x in line.breaks_utf8]),
 203     )
 204
 205
 206 def lineToCppDataLineUtf16(line: BreakTestItem) -> str:
 207     return cpp_test_data_line_template.format(
 208         f'L"{line.encoded}"',
 209         ", ".join([str(x) for x in line.code_points]),
 210         ", ".join([str(x) for x in line.breaks_utf16]),
 211     )
 212
 213
 214 def lineToCppDataLineUtf32(line: BreakTestItem) -> str:
 215     return cpp_test_data_line_template.format(
 216         f'L"{line.encoded}"',
 217         ", ".join([str(x) for x in line.code_points]),
 218         ", ".join([str(x) for x in line.breaks_utf32]),
 219     )
 220
 221
 222 """
 223 Generate test data from "GraphemeBreakText.txt"
 224 This file can be downloaded from:
 225 https://www.unicode.org/Public/UCD/latest/ucd/auxiliary/GraphemeBreakTest.txt
 226 This script looks for GraphemeBreakTest.txt in same directory as this script
 227 """
 228
 229
 230 def generate_all() -> str:
 231     test_data_path = Path(__file__)
 232     test_data_path = test_data_path.absolute()
 233     test_data_path = (
 234         test_data_path.parent / "data" / "unicode" / "GraphemeBreakTest.txt"
 235     )
 236     lines = list()
 237     with open(test_data_path, mode="rt", encoding="utf-8") as file:
 238         while line := parseBreakTestLine(file):
 239             if len(line.encoded) > 0:
 240                 lines.append(line)
 241     return cpp_template.format(
 242         len(lines),
 243         ",\n".join(map(lineToCppDataLineUtf8, lines)),
 244         ",\n".join(map(lineToCppDataLineUtf16, lines)),
 245         ",\n".join(map(lineToCppDataLineUtf32, lines)),
 246     )
 247
 248
 249 if __name__ == "__main__":
 250     if len(sys.argv) == 2:
 251         sys.stdout = open(sys.argv[1], "w")
 252     print(generate_all())