1//===--- ParsedAST.h - Building translation units ----------------*- 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 exposes building a file as if it were open in clangd, and defines
10// the ParsedAST structure that holds the results.
11//
12// This is similar to a clang -fsyntax-only run that produces a clang AST, but
13// we have several customizations:
14// - preamble handling
15// - capturing diagnostics for later access
16// - running clang-tidy checks
17//
18//===----------------------------------------------------------------------===//
19
20#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANGD_PARSEDAST_H
21#define LLVM_CLANG_TOOLS_EXTRA_CLANGD_PARSEDAST_H
22
23#include "CollectMacros.h"
24#include "Compiler.h"
25#include "Diagnostics.h"
26#include "Headers.h"
27#include "Preamble.h"
28#include "clang-include-cleaner/Record.h"
29#include "support/Path.h"
30#include "clang/Frontend/FrontendAction.h"
31#include "clang/Lex/Preprocessor.h"
32#include "clang/Tooling/Syntax/Tokens.h"
33#include "llvm/ADT/ArrayRef.h"
34#include "llvm/ADT/StringRef.h"
35#include <memory>
36#include <optional>
37#include <string>
38#include <vector>
39
40namespace clang {
41class HeuristicResolver;
42class Sema;
43namespace clangd {
44
45/// Stores and provides access to parsed AST.
46class ParsedAST {
47public:
48 /// Attempts to run Clang and store the parsed AST.
49 /// If \p Preamble is non-null it is reused during parsing.
50 /// This function does not check if preamble is valid to reuse.
51 static std::optional<ParsedAST>
52 build(llvm::StringRef Filename, const ParseInputs &Inputs,
53 std::unique_ptr<clang::CompilerInvocation> CI,
54 llvm::ArrayRef<Diag> CompilerInvocationDiags,
55 std::shared_ptr<const PreambleData> Preamble);
56
57 ParsedAST(ParsedAST &&Other);
58 ParsedAST &operator=(ParsedAST &&Other);
59
60 ~ParsedAST();
61
62 ParsedAST(const ParsedAST &Other) = delete;
63 ParsedAST &operator=(const ParsedAST &Other) = delete;
64
65 /// Note that the returned ast will not contain decls from the preamble that
66 /// were not deserialized during parsing. Clients should expect only decls
67 /// from the main file to be in the AST.
68 ASTContext &getASTContext();
69 const ASTContext &getASTContext() const;
70
71 Sema &getSema();
72
73 Preprocessor &getPreprocessor();
74 std::shared_ptr<Preprocessor> getPreprocessorPtr();
75 const Preprocessor &getPreprocessor() const;
76
77 SourceManager &getSourceManager() {
78 return getASTContext().getSourceManager();
79 }
80 const SourceManager &getSourceManager() const {
81 return getASTContext().getSourceManager();
82 }
83
84 const LangOptions &getLangOpts() const {
85 return getASTContext().getLangOpts();
86 }
87
88 /// This function returns top-level decls present in the main file of the AST.
89 /// The result does not include the decls that come from the preamble.
90 /// (These should be const, but RecursiveASTVisitor requires Decl*).
91 ArrayRef<Decl *> getLocalTopLevelDecls();
92 ArrayRef<const Decl *> getLocalTopLevelDecls() const;
93
94 llvm::ArrayRef<Diag> getDiagnostics() const;
95
96 /// Returns the estimated size of the AST and the accessory structures, in
97 /// bytes. Does not include the size of the preamble.
98 std::size_t getUsedBytes() const;
99 const IncludeStructure &getIncludeStructure() const;
100
101 /// Gets all macro references (definition, expansions) present in the main
102 /// file, including those in the preamble region.
103 const MainFileMacros &getMacros() const;
104 /// Gets all pragma marks in the main file.
105 const std::vector<PragmaMark> &getMarks() const;
106 /// Tokens recorded while parsing the main file.
107 /// (!) does not have tokens from the preamble.
108 const syntax::TokenBuffer &getTokens() const { return Tokens; }
109 /// Returns the PramaIncludes for preamble + main file includes.
110 const include_cleaner::PragmaIncludes &getPragmaIncludes() const;
111
112 /// Returns the version of the ParseInputs this AST was built from.
113 llvm::StringRef version() const { return Version; }
114
115 /// Returns the path passed by the caller when building this AST.
116 PathRef tuPath() const { return TUPath; }
117
118 /// Returns the version of the ParseInputs used to build Preamble part of this
119 /// AST. Might be std::nullopt if no Preamble is used.
120 std::optional<llvm::StringRef> preambleVersion() const;
121
122 const HeuristicResolver *getHeuristicResolver() const {
123 return Resolver.get();
124 }
125
126private:
127 ParsedAST(PathRef TUPath, llvm::StringRef Version,
128 std::shared_ptr<const PreambleData> Preamble,
129 std::unique_ptr<CompilerInstance> Clang,
130 std::unique_ptr<FrontendAction> Action, syntax::TokenBuffer Tokens,
131 MainFileMacros Macros, std::vector<PragmaMark> Marks,
132 std::vector<Decl *> LocalTopLevelDecls, std::vector<Diag> Diags,
133 IncludeStructure Includes, include_cleaner::PragmaIncludes PI);
134 Path TUPath;
135 std::string Version;
136 // In-memory preambles must outlive the AST, it is important that this member
137 // goes before Clang and Action.
138 std::shared_ptr<const PreambleData> Preamble;
139 // We store an "incomplete" FrontendAction (i.e. no EndSourceFile was called
140 // on it) and CompilerInstance used to run it. That way we don't have to do
141 // complex memory management of all Clang structures on our own. (They are
142 // stored in CompilerInstance and cleaned up by
143 // FrontendAction.EndSourceFile).
144 std::unique_ptr<CompilerInstance> Clang;
145 std::unique_ptr<FrontendAction> Action;
146 /// Tokens recorded after the preamble finished.
147 /// - Includes all spelled tokens for the main file.
148 /// - Includes expanded tokens produced **after** preamble.
149 /// - Does not have spelled or expanded tokens for files from preamble.
150 syntax::TokenBuffer Tokens;
151
152 /// All macro definitions and expansions in the main file.
153 MainFileMacros Macros;
154 // Pragma marks in the main file.
155 std::vector<PragmaMark> Marks;
156 // Diags emitted while parsing this AST (including preamble and compiler
157 // invocation).
158 std::vector<Diag> Diags;
159 // Top-level decls inside the current file. Not that this does not include
160 // top-level decls from the preamble.
161 std::vector<Decl *> LocalTopLevelDecls;
162 IncludeStructure Includes;
163 include_cleaner::PragmaIncludes PI;
164 std::unique_ptr<HeuristicResolver> Resolver;
165};
166
167} // namespace clangd
168} // namespace clang
169
170#endif // LLVM_CLANG_TOOLS_EXTRA_CLANGD_PARSEDAST_H
171