1 //===--- TokenAnalyzer.h - Analyze Token Streams ----------------*- 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 declares an abstract TokenAnalyzer, and associated helper 11 /// classes. TokenAnalyzer can be extended to generate replacements based on 12 /// an annotated and pre-processed token stream. 13 /// 14 //===----------------------------------------------------------------------===// 15 16 #ifndef LLVM_CLANG_LIB_FORMAT_TOKENANALYZER_H 17 #define LLVM_CLANG_LIB_FORMAT_TOKENANALYZER_H 18 19 #include "AffectedRangeManager.h" 20 #include "FormatTokenLexer.h" 21 #include "TokenAnnotator.h" 22 23 namespace clang { 24 namespace format { 25 26 class Environment { 27 public: 28 // This sets up an virtual file system with file \p FileName containing the 29 // fragment \p Code. Assumes that \p Code starts at \p FirstStartColumn, 30 // that the next lines of \p Code should start at \p NextStartColumn, and 31 // that \p Code should end at \p LastStartColumn if it ends in newline. 32 // See also the documentation of clang::format::internal::reformat. 33 Environment(StringRef Code, StringRef FileName, unsigned FirstStartColumn = 0, 34 unsigned NextStartColumn = 0, unsigned LastStartColumn = 0); 35 36 FileID getFileID() const { return ID; } 37 38 SourceManager &getSourceManager() const { return SM; } 39 40 ArrayRef<CharSourceRange> getCharRanges() const { return CharRanges; } 41 42 // Returns the column at which the fragment of code managed by this 43 // environment starts. 44 unsigned getFirstStartColumn() const { return FirstStartColumn; } 45 46 // Returns the column at which subsequent lines of the fragment of code 47 // managed by this environment should start. 48 unsigned getNextStartColumn() const { return NextStartColumn; } 49 50 // Returns the column at which the fragment of code managed by this 51 // environment should end if it ends in a newline. 52 unsigned getLastStartColumn() const { return LastStartColumn; } 53 54 // Returns nullptr and prints a diagnostic to stderr if the environment 55 // can't be created. 56 static std::unique_ptr<Environment> make(StringRef Code, StringRef FileName, 57 ArrayRef<tooling::Range> Ranges, 58 unsigned FirstStartColumn = 0, 59 unsigned NextStartColumn = 0, 60 unsigned LastStartColumn = 0); 61 62 private: 63 // This is only set if constructed from string. 64 std::unique_ptr<SourceManagerForFile> VirtualSM; 65 66 // This refers to either a SourceManager provided by users or VirtualSM 67 // created for a single file. 68 SourceManager &SM; 69 FileID ID; 70 71 SmallVector<CharSourceRange, 8> CharRanges; 72 unsigned FirstStartColumn; 73 unsigned NextStartColumn; 74 unsigned LastStartColumn; 75 }; 76 77 class TokenAnalyzer : public UnwrappedLineConsumer { 78 public: 79 TokenAnalyzer(const Environment &Env, const FormatStyle &Style); 80 81 std::pair<tooling::Replacements, unsigned> 82 process(bool SkipAnnotation = false); 83 84 protected: 85 virtual std::pair<tooling::Replacements, unsigned> 86 analyze(TokenAnnotator &Annotator, 87 SmallVectorImpl<AnnotatedLine *> &AnnotatedLines, 88 FormatTokenLexer &Tokens) = 0; 89 90 void consumeUnwrappedLine(const UnwrappedLine &TheLine) override; 91 92 void finishRun() override; 93 94 FormatStyle Style; 95 LangOptions LangOpts; 96 // Stores Style, FileID and SourceManager etc. 97 const Environment &Env; 98 // AffectedRangeMgr stores ranges to be fixed. 99 AffectedRangeManager AffectedRangeMgr; 100 SmallVector<SmallVector<UnwrappedLine, 16>, 2> UnwrappedLines; 101 encoding::Encoding Encoding; 102 }; 103 104 } // end namespace format 105 } // end namespace clang 106 107 #endif 108