1 | //===- llvm/Support/Unicode.h - Unicode character properties -*- C++ -*-=====// |
2 | // |
3 | // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. |
4 | // See https://llvm.org/LICENSE.txt for license information. |
5 | // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception |
6 | // |
7 | //===----------------------------------------------------------------------===// |
8 | // |
9 | // This file defines functions that allow querying certain properties of Unicode |
10 | // characters. |
11 | // |
12 | //===----------------------------------------------------------------------===// |
13 | |
14 | #ifndef LLVM_SUPPORT_UNICODE_H |
15 | #define LLVM_SUPPORT_UNICODE_H |
16 | |
17 | #include "llvm/ADT/SmallString.h" |
18 | #include <optional> |
19 | #include <string> |
20 | |
21 | namespace llvm { |
22 | class StringRef; |
23 | |
24 | namespace sys { |
25 | namespace unicode { |
26 | |
27 | enum ColumnWidthErrors { |
28 | ErrorInvalidUTF8 = -2, |
29 | ErrorNonPrintableCharacter = -1 |
30 | }; |
31 | |
32 | /// Determines if a character is likely to be displayed correctly on the |
33 | /// terminal. Exact implementation would have to depend on the specific |
34 | /// terminal, so we define the semantic that should be suitable for generic case |
35 | /// of a terminal capable to output Unicode characters. |
36 | /// |
37 | /// Printable codepoints are those in the categories L, M, N, P, S and Zs |
38 | /// \return true if the character is considered printable. |
39 | bool isPrintable(int UCS); |
40 | |
41 | // Formatting codepoints are codepoints in the Cf category. |
42 | bool isFormatting(int UCS); |
43 | |
44 | /// Gets the number of positions the UTF8-encoded \p Text is likely to occupy |
45 | /// when output on a terminal ("character width"). This depends on the |
46 | /// implementation of the terminal, and there's no standard definition of |
47 | /// character width. |
48 | /// |
49 | /// The implementation defines it in a way that is expected to be compatible |
50 | /// with a generic Unicode-capable terminal. |
51 | /// |
52 | /// \return Character width: |
53 | /// * ErrorNonPrintableCharacter (-1) if \p Text contains non-printable |
54 | /// characters (as identified by isPrintable); |
55 | /// * 0 for each non-spacing and enclosing combining mark; |
56 | /// * 2 for each CJK character excluding halfwidth forms; |
57 | /// * 1 for each of the remaining characters. |
58 | int columnWidthUTF8(StringRef Text); |
59 | |
60 | /// Fold input unicode character according the Simple unicode case folding |
61 | /// rules. |
62 | int foldCharSimple(int C); |
63 | |
64 | /// Maps the name or the alias of a Unicode character to its associated |
65 | /// codepoints. |
66 | /// The names and aliases are derived from UnicodeData.txt and NameAliases.txt |
67 | /// For compatibility with the semantics of named character escape sequences in |
68 | /// C++, this mapping does an exact match sensitive to casing and spacing. |
69 | /// \return The codepoint of the corresponding character, if any. |
70 | std::optional<char32_t> nameToCodepointStrict(StringRef Name); |
71 | |
72 | struct LooseMatchingResult { |
73 | char32_t CodePoint; |
74 | SmallString<64> Name; |
75 | }; |
76 | |
77 | std::optional<LooseMatchingResult> nameToCodepointLooseMatching(StringRef Name); |
78 | |
79 | struct MatchForCodepointName { |
80 | std::string Name; |
81 | uint32_t Distance = 0; |
82 | char32_t Value = 0; |
83 | }; |
84 | |
85 | SmallVector<MatchForCodepointName> |
86 | nearestMatchesForCodepointName(StringRef Pattern, std::size_t MaxMatchesCount); |
87 | |
88 | } // namespace unicode |
89 | } // namespace sys |
90 | } // namespace llvm |
91 | |
92 | #endif |
93 | |