1 | //===--- FormatTokenLexer.h - Format C++ code ----------------*- 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 | /// \file |
10 | /// This file contains FormatTokenLexer, which tokenizes a source file |
11 | /// into a token stream suitable for ClangFormat. |
12 | /// |
13 | //===----------------------------------------------------------------------===// |
14 | |
15 | #ifndef LLVM_CLANG_LIB_FORMAT_FORMATTOKENLEXER_H |
16 | #define LLVM_CLANG_LIB_FORMAT_FORMATTOKENLEXER_H |
17 | |
18 | #include "Encoding.h" |
19 | #include "FormatToken.h" |
20 | #include "clang/Basic/SourceLocation.h" |
21 | #include "clang/Basic/SourceManager.h" |
22 | #include "clang/Format/Format.h" |
23 | #include "llvm/ADT/MapVector.h" |
24 | #include "llvm/Support/Regex.h" |
25 | |
26 | #include <stack> |
27 | |
28 | namespace clang { |
29 | namespace format { |
30 | |
31 | enum LexerState { |
32 | NORMAL, |
33 | TEMPLATE_STRING, |
34 | TOKEN_STASHED, |
35 | }; |
36 | |
37 | class FormatTokenLexer { |
38 | public: |
39 | FormatTokenLexer(const SourceManager &SourceMgr, FileID ID, unsigned Column, |
40 | const FormatStyle &Style, encoding::Encoding Encoding); |
41 | |
42 | ArrayRef<FormatToken *> lex(); |
43 | |
44 | const AdditionalKeywords &getKeywords() { return Keywords; } |
45 | |
46 | private: |
47 | void tryMergePreviousTokens(); |
48 | |
49 | bool tryMergeLessLess(); |
50 | bool tryMergeNSStringLiteral(); |
51 | bool tryMergeJSPrivateIdentifier(); |
52 | bool tryMergeCSharpVerbatimStringLiteral(); |
53 | bool tryMergeCSharpKeywordVariables(); |
54 | bool tryMergeCSharpNullConditionals(); |
55 | bool tryMergeCSharpDoubleQuestion(); |
56 | |
57 | bool tryMergeTokens(ArrayRef<tok::TokenKind> Kinds, TokenType NewType); |
58 | |
59 | // Returns \c true if \p Tok can only be followed by an operand in JavaScript. |
60 | bool precedesOperand(FormatToken *Tok); |
61 | |
62 | bool canPrecedeRegexLiteral(FormatToken *Prev); |
63 | |
64 | // Tries to parse a JavaScript Regex literal starting at the current token, |
65 | // if that begins with a slash and is in a location where JavaScript allows |
66 | // regex literals. Changes the current token to a regex literal and updates |
67 | // its text if successful. |
68 | void tryParseJSRegexLiteral(); |
69 | |
70 | // Handles JavaScript template strings. |
71 | // |
72 | // JavaScript template strings use backticks ('`') as delimiters, and allow |
73 | // embedding expressions nested in ${expr-here}. Template strings can be |
74 | // nested recursively, i.e. expressions can contain template strings in turn. |
75 | // |
76 | // The code below parses starting from a backtick, up to a closing backtick or |
77 | // an opening ${. It also maintains a stack of lexing contexts to handle |
78 | // nested template parts by balancing curly braces. |
79 | void handleTemplateStrings(); |
80 | |
81 | void tryParsePythonComment(); |
82 | |
83 | bool tryMerge_TMacro(); |
84 | |
85 | bool tryMergeConflictMarkers(); |
86 | |
87 | FormatToken *getStashedToken(); |
88 | |
89 | FormatToken *getNextToken(); |
90 | |
91 | FormatToken *FormatTok; |
92 | bool IsFirstToken; |
93 | std::stack<LexerState> StateStack; |
94 | unsigned Column; |
95 | unsigned TrailingWhitespace; |
96 | std::unique_ptr<Lexer> Lex; |
97 | const SourceManager &SourceMgr; |
98 | FileID ID; |
99 | const FormatStyle &Style; |
100 | IdentifierTable IdentTable; |
101 | AdditionalKeywords Keywords; |
102 | encoding::Encoding Encoding; |
103 | llvm::SpecificBumpPtrAllocator<FormatToken> Allocator; |
104 | // Index (in 'Tokens') of the last token that starts a new line. |
105 | unsigned FirstInLineIndex; |
106 | SmallVector<FormatToken *, 16> Tokens; |
107 | |
108 | llvm::SmallMapVector<IdentifierInfo *, TokenType, 8> Macros; |
109 | |
110 | bool FormattingDisabled; |
111 | |
112 | llvm::Regex MacroBlockBeginRegex; |
113 | llvm::Regex MacroBlockEndRegex; |
114 | |
115 | void readRawToken(FormatToken &Tok); |
116 | |
117 | void resetLexer(unsigned Offset); |
118 | }; |
119 | |
120 | } // namespace format |
121 | } // namespace clang |
122 | |
123 | #endif |
124 | |