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 "Encoding.h"
21#include "FormatToken.h"
22#include "FormatTokenLexer.h"
23#include "TokenAnnotator.h"
24#include "UnwrappedLineParser.h"
25#include "clang/Basic/Diagnostic.h"
26#include "clang/Basic/DiagnosticOptions.h"
27#include "clang/Basic/FileManager.h"
28#include "clang/Basic/SourceManager.h"
29#include "clang/Format/Format.h"
30#include "llvm/ADT/STLExtras.h"
31#include "llvm/Support/Debug.h"
32
33namespace clang {
34namespace format {
35
36class Environment {
37public:
38  // This sets up an virtual file system with file \p FileName containing the
39  // fragment \p Code. Assumes that \p Code starts at \p FirstStartColumn,
40  // that the next lines of \p Code should start at \p NextStartColumn, and
41  // that \p Code should end at \p LastStartColumn if it ends in newline.
42  // See also the documentation of clang::format::internal::reformat.
43  Environment(StringRef Code, StringRef FileName,
44              ArrayRef<tooling::Range> Ranges, unsigned FirstStartColumn = 0,
45              unsigned NextStartColumn = 0, unsigned LastStartColumn = 0);
46
47  FileID getFileID() const { return ID; }
48
49  const SourceManager &getSourceManager() const { return SM; }
50
51  ArrayRef<CharSourceRange> getCharRanges() const { return CharRanges; }
52
53  // Returns the column at which the fragment of code managed by this
54  // environment starts.
55  unsigned getFirstStartColumn() const { return FirstStartColumn; }
56
57  // Returns the column at which subsequent lines of the fragment of code
58  // managed by this environment should start.
59  unsigned getNextStartColumn() const { return NextStartColumn; }
60
61  // Returns the column at which the fragment of code managed by this
62  // environment should end if it ends in a newline.
63  unsigned getLastStartColumn() const { return LastStartColumn; }
64
65private:
66  // This is only set if constructed from string.
67  std::unique_ptr<SourceManagerForFile> VirtualSM;
68
69  // This refers to either a SourceManager provided by users or VirtualSM
70  // created for a single file.
71  SourceManager &SM;
72  FileID ID;
73
74  SmallVector<CharSourceRange, 8> CharRanges;
75  unsigned FirstStartColumn;
76  unsigned NextStartColumn;
77  unsigned LastStartColumn;
78};
79
80class TokenAnalyzer : public UnwrappedLineConsumer {
81public:
82  TokenAnalyzer(const Environment &Env, const FormatStyle &Style);
83
84  std::pair<tooling::Replacements, unsigned> process();
85
86protected:
87  virtual std::pair<tooling::Replacements, unsigned>
88  analyze(TokenAnnotator &Annotator,
89          SmallVectorImpl<AnnotatedLine *> &AnnotatedLines,
90          FormatTokenLexer &Tokens) = 0;
91
92  void consumeUnwrappedLine(const UnwrappedLine &TheLine) override;
93
94  void finishRun() override;
95
96  FormatStyle Style;
97  // Stores Style, FileID and SourceManager etc.
98  const Environment &Env;
99  // AffectedRangeMgr stores ranges to be fixed.
100  AffectedRangeManager AffectedRangeMgr;
101  SmallVector<SmallVector<UnwrappedLine, 16>, 2> UnwrappedLines;
102  encoding::Encoding Encoding;
103};
104
105} // end namespace format
106} // end namespace clang
107
108#endif
109