| 1 | //===--- TokenConcatenation.h - Token Concatenation Avoidance ---*- 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 the TokenConcatenation class. |
| 10 | // |
| 11 | //===----------------------------------------------------------------------===// |
| 12 | |
| 13 | #ifndef LLVM_CLANG_LEX_TOKENCONCATENATION_H |
| 14 | #define LLVM_CLANG_LEX_TOKENCONCATENATION_H |
| 15 | |
| 16 | #include "clang/Basic/TokenKinds.h" |
| 17 | |
| 18 | namespace clang { |
| 19 | class Preprocessor; |
| 20 | class Token; |
| 21 | |
| 22 | /// TokenConcatenation class, which answers the question of |
| 23 | /// "Is it safe to emit two tokens without a whitespace between them, or |
| 24 | /// would that cause implicit concatenation of the tokens?" |
| 25 | /// |
| 26 | /// For example, it emitting two identifiers "foo" and "bar" next to each |
| 27 | /// other would cause the lexer to produce one "foobar" token. Emitting "1" |
| 28 | /// and ")" next to each other is safe. |
| 29 | /// |
| 30 | class TokenConcatenation { |
| 31 | const Preprocessor &PP; |
| 32 | |
| 33 | enum AvoidConcatInfo { |
| 34 | /// By default, a token never needs to avoid concatenation. Most tokens |
| 35 | /// (e.g. ',', ')', etc) don't cause a problem when concatenated. |
| 36 | aci_never_avoid_concat = 0, |
| 37 | |
| 38 | /// aci_custom_firstchar - AvoidConcat contains custom code to handle this |
| 39 | /// token's requirements, and it needs to know the first character of the |
| 40 | /// token. |
| 41 | aci_custom_firstchar = 1, |
| 42 | |
| 43 | /// aci_custom - AvoidConcat contains custom code to handle this token's |
| 44 | /// requirements, but it doesn't need to know the first character of the |
| 45 | /// token. |
| 46 | aci_custom = 2, |
| 47 | |
| 48 | /// aci_avoid_equal - Many tokens cannot be safely followed by an '=' |
| 49 | /// character. For example, "<<" turns into "<<=" when followed by an =. |
| 50 | aci_avoid_equal = 4 |
| 51 | }; |
| 52 | |
| 53 | /// TokenInfo - This array contains information for each token on what |
| 54 | /// action to take when avoiding concatenation of tokens in the AvoidConcat |
| 55 | /// method. |
| 56 | char TokenInfo[tok::NUM_TOKENS]; |
| 57 | public: |
| 58 | TokenConcatenation(const Preprocessor &PP); |
| 59 | |
| 60 | bool AvoidConcat(const Token &PrevPrevTok, |
| 61 | const Token &PrevTok, |
| 62 | const Token &Tok) const; |
| 63 | |
| 64 | private: |
| 65 | /// IsIdentifierStringPrefix - Return true if the spelling of the token |
| 66 | /// is literally 'L', 'u', 'U', or 'u8'. |
| 67 | bool IsIdentifierStringPrefix(const Token &Tok) const; |
| 68 | }; |
| 69 | } // end clang namespace |
| 70 | |
| 71 | #endif |
| 72 |