| blob | 5b4b3950c6a1d0c61dc226c01f35087c3a4904b8 |
1 // -*- C++ -*-
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 //===----------------------------------------------------------------------===//
10 // WARNING, this entire header is generated by
11 // utils/generate_width_estimation_table.py
12 // DO NOT MODIFY!
14 // UNICODE, INC. LICENSE AGREEMENT - DATA FILES AND SOFTWARE
15 //
16 // See Terms of Use <https://www.unicode.org/copyright.html>
17 // for definitions of Unicode Inc.'s Data Files and Software.
18 //
19 // NOTICE TO USER: Carefully read the following legal agreement.
20 // BY DOWNLOADING, INSTALLING, COPYING OR OTHERWISE USING UNICODE INC.'S
21 // DATA FILES ("DATA FILES"), AND/OR SOFTWARE ("SOFTWARE"),
22 // YOU UNEQUIVOCALLY ACCEPT, AND AGREE TO BE BOUND BY, ALL OF THE
23 // TERMS AND CONDITIONS OF THIS AGREEMENT.
24 // IF YOU DO NOT AGREE, DO NOT DOWNLOAD, INSTALL, COPY, DISTRIBUTE OR USE
25 // THE DATA FILES OR SOFTWARE.
26 //
27 // COPYRIGHT AND PERMISSION NOTICE
28 //
29 // Copyright (c) 1991-2022 Unicode, Inc. All rights reserved.
30 // Distributed under the Terms of Use in https://www.unicode.org/copyright.html.
31 //
32 // Permission is hereby granted, free of charge, to any person obtaining
33 // a copy of the Unicode data files and any associated documentation
34 // (the "Data Files") or Unicode software and any associated documentation
35 // (the "Software") to deal in the Data Files or Software
36 // without restriction, including without limitation the rights to use,
37 // copy, modify, merge, publish, distribute, and/or sell copies of
38 // the Data Files or Software, and to permit persons to whom the Data Files
39 // or Software are furnished to do so, provided that either
40 // (a) this copyright and permission notice appear with all copies
41 // of the Data Files or Software, or
42 // (b) this copyright and permission notice appear in associated
43 // Documentation.
44 //
45 // THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF
46 // ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
47 // WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
48 // NONINFRINGEMENT OF THIRD PARTY RIGHTS.
49 // IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN THIS
50 // NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL
51 // DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
52 // DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
53 // TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
54 // PERFORMANCE OF THE DATA FILES OR SOFTWARE.
55 //
56 // Except as contained in this notice, the name of a copyright holder
57 // shall not be used in advertising or otherwise to promote the sale,
58 // use or other dealings in these Data Files or Software without prior
59 // written authorization of the copyright holder.
61 #ifndef _LIBCPP___FORMAT_WIDTH_ESTIMATION_TABLE_H
62 #define _LIBCPP___FORMAT_WIDTH_ESTIMATION_TABLE_H
64 #include <__algorithm/ranges_upper_bound.h>
65 #include <__config>
66 #include <__cstddef/ptrdiff_t.h>
67 #include <cstdint>
69 #if !defined(_LIBCPP_HAS_NO_PRAGMA_SYSTEM_HEADER)
70 # pragma GCC system_header
71 #endif
73 _LIBCPP_BEGIN_NAMESPACE_STD
75 #if _LIBCPP_STD_VER >= 20
79 /// The entries of the characters with an estimated width of 2.
80 ///
81 /// Contains the entries for [format.string.std]/12
82 /// - Any code point with the East_Asian_Width="W" or East_Asian_Width="F"
83 /// Derived Extracted Property as described by UAX #44
84 /// - U+4DC0 - U+4DFF (Yijing Hexagram Symbols)
85 /// - U+1F300 - U+1F5FF (Miscellaneous Symbols and Pictographs)
86 /// - U+1F900 - U+1F9FF (Supplemental Symbols and Pictographs)
87 ///
88 /// The data is generated from
89 /// - https://www.unicode.org/Public/UCD/latest/ucd/EastAsianWidth.txt
90 /// - The "overrides" in [format.string.std]/12
91 ///
92 /// The format of EastAsianWidth.txt is two fields separated by a semicolon.
93 /// Field 0: Unicode code point value or range of code point values
94 /// Field 1: East_Asian_Width property, consisting of one of the following values:
95 /// "A", "F", "H", "N", "Na", "W"
96 /// - All code points, assigned or unassigned, that are not listed
97 /// explicitly are given the value "N".
98 /// - The unassigned code points in the following blocks default to "W":
99 /// CJK Unified Ideographs Extension A: U+3400..U+4DBF
100 /// CJK Unified Ideographs: U+4E00..U+9FFF
101 /// CJK Compatibility Ideographs: U+F900..U+FAFF
102 /// - All undesignated code points in Planes 2 and 3, whether inside or
103 /// outside of allocated blocks, default to "W":
104 /// Plane 2: U+20000..U+2FFFD
105 /// Plane 3: U+30000..U+3FFFD
106 ///
107 /// The table is similar to the table
108 /// __extended_grapheme_custer_property_boundary::__entries
109 /// which explains the details of these classes. The only difference is this
110 /// table lacks a property, thus having more bits available for the size.
111 ///
112 /// The maximum code point that has an estimated width of 2 is U+3FFFD. This
113 /// value can be encoded in 18 bits. Thus the upper 3 bits of the code point
114 /// are always 0. These 3 bits are used to enlarge the offset range. This
115 /// optimization reduces the table in Unicode 15 from 184 to 104 entries,
116 /// saving 320 bytes.
117 ///
118 /// The data has 2 values:
119 /// - bits [0, 13] The size of the range, allowing 16384 elements.
120 /// - bits [14, 31] The lower bound code point of the range. The upper bound of
121 /// the range is lower bound + size.
231 /// The upper bound entry of EastAsianWidth.txt.
232 ///
233 /// Values greater than this value may have more than 18 significant bits.
234 /// They always have a width of 1. This property makes it possible to store
235 /// the table in its compact form.
238 /// Returns the estimated width of a Unicode code point.
239 ///
240 /// \\pre The code point is a valid Unicode code point.
241 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr int __estimated_width(const char32_t __code_point) noexcept {
242 // Since __table_upper_bound contains the unshifted range do the
243 // comparison without shifting.
247 // When the code-point is less than the first element in the table
248 // the lookup is quite expensive. Since quite some scripts are in
249 // that range, it makes sense to validate that first.
250 // The std_format_spec_string_unicode benchmark gives a measurable
251 // improvement.
255 ptrdiff_t __i = std::ranges::upper_bound(__entries, (__code_point << 14) | 0x3fffu) - __entries;
262 }
268 _LIBCPP_END_NAMESPACE_STD