Home | History | Annotate | Line # | Download | only in Support 
      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 namespace llvm {
     18 class StringRef;
     19 
     20 namespace sys {
     21 namespace unicode {
     22 
     23 enum ColumnWidthErrors {
     24   ErrorInvalidUTF8 = -2,
     25   ErrorNonPrintableCharacter = -1
     26 };
     27 
     28 /// Determines if a character is likely to be displayed correctly on the
     29 /// terminal. Exact implementation would have to depend on the specific
     30 /// terminal, so we define the semantic that should be suitable for generic case
     31 /// of a terminal capable to output Unicode characters.
     32 ///
     33 /// All characters from the Unicode code point range are considered printable
     34 /// except for:
     35 ///   * C0 and C1 control character ranges;
     36 ///   * default ignorable code points as per 5.21 of
     37 ///     http://www.unicode.org/versions/Unicode6.2.0/UnicodeStandard-6.2.pdf
     38 ///     except for U+00AD SOFT HYPHEN, as it's actually displayed on most
     39 ///     terminals;
     40 ///   * format characters (category = Cf);
     41 ///   * surrogates (category = Cs);
     42 ///   * unassigned characters (category = Cn).
     43 /// \return true if the character is considered printable.
     44 bool isPrintable(int UCS);
     45 
     46 /// Gets the number of positions the UTF8-encoded \p Text is likely to occupy
     47 /// when output on a terminal ("character width"). This depends on the
     48 /// implementation of the terminal, and there's no standard definition of
     49 /// character width.
     50 ///
     51 /// The implementation defines it in a way that is expected to be compatible
     52 /// with a generic Unicode-capable terminal.
     53 ///
     54 /// \return Character width:
     55 ///   * ErrorNonPrintableCharacter (-1) if \p Text contains non-printable
     56 ///     characters (as identified by isPrintable);
     57 ///   * 0 for each non-spacing and enclosing combining mark;
     58 ///   * 2 for each CJK character excluding halfwidth forms;
     59 ///   * 1 for each of the remaining characters.
     60 int columnWidthUTF8(StringRef Text);
     61 
     62 /// Fold input unicode character according the Simple unicode case folding
     63 /// rules.
     64 int foldCharSimple(int C);
     65 
     66 } // namespace unicode
     67 } // namespace sys
     68 } // namespace llvm
     69 
     70 #endif
     71