1 //===- llvm/Support/Unicode.h - Unicode character properties -*- C++ -*-=====//
3 // The LLVM Compiler Infrastructure
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
8 //===----------------------------------------------------------------------===//
10 // This file defines functions that allow querying certain properties of Unicode
13 //===----------------------------------------------------------------------===//
15 #include "llvm/ADT/StringRef.h"
21 enum ColumnWidthErrors {
22 ErrorInvalidUTF8 = -2,
23 ErrorNonPrintableCharacter = -1
26 /// Determines if a character is likely to be displayed correctly on the
27 /// terminal. Exact implementation would have to depend on the specific
28 /// terminal, so we define the semantic that should be suitable for generic case
29 /// of a terminal capable to output Unicode characters.
31 /// All characters from the Unicode code point range are considered printable
33 /// * C0 and C1 control character ranges;
34 /// * default ignorable code points as per 5.21 of
35 /// http://www.unicode.org/versions/Unicode6.2.0/UnicodeStandard-6.2.pdf
36 /// except for U+00AD SOFT HYPHEN, as it's actually displayed on most
38 /// * format characters (category = Cf);
39 /// * surrogates (category = Cs);
40 /// * unassigned characters (category = Cn).
41 /// \return true if the character is considered printable.
42 bool isPrintable(int UCS);
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
49 /// The implementation defines it in a way that is expected to be compatible
50 /// with a generic Unicode-capable terminal.
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);
60 } // namespace unicode