clang-tools  11.0.0
Quality.h
Go to the documentation of this file.
1 //===--- Quality.h - Ranking alternatives for ambiguous queries --*- 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 /// Some operations such as code completion produce a set of candidates.
10 /// Usually the user can choose between them, but we should put the best options
11 /// at the top (they're easier to select, and more likely to be seen).
12 ///
13 /// This file defines building blocks for ranking candidates.
14 /// It's used by the features directly and also in the implementation of
15 /// indexes, as indexes also need to heuristically limit their results.
16 ///
17 /// The facilities here are:
18 /// - retrieving scoring signals from e.g. indexes, AST, CodeCompletionString
19 /// These are structured in a way that they can be debugged, and are fairly
20 /// consistent regardless of the source.
21 /// - compute scores from scoring signals. These are suitable for sorting.
22 /// - sorting utilities like the TopN container.
23 /// These could be split up further to isolate dependencies if we care.
24 ///
25 //===----------------------------------------------------------------------===//
26 
27 #ifndef LLVM_CLANG_TOOLS_EXTRA_CLANGD_QUALITY_H
28 #define LLVM_CLANG_TOOLS_EXTRA_CLANGD_QUALITY_H
29 
30 #include "ExpectedTypes.h"
31 #include "FileDistance.h"
32 #include "clang/Sema/CodeCompleteConsumer.h"
33 #include "llvm/ADT/ArrayRef.h"
34 #include "llvm/ADT/StringRef.h"
35 #include "llvm/ADT/StringSet.h"
36 #include <algorithm>
37 #include <functional>
38 #include <vector>
39 
40 namespace llvm {
41 class raw_ostream;
42 } // namespace llvm
43 
44 namespace clang {
45 class CodeCompletionResult;
46 
47 namespace clangd {
48 
49 struct Symbol;
50 class URIDistance;
51 
52 // Signals structs are designed to be aggregated from 0 or more sources.
53 // A default instance has neutral signals, and sources are merged into it.
54 // They can be dumped for debugging, and evaluate()d into a score.
55 
56 /// Attributes of a symbol that affect how much we like it.
58  bool Deprecated = false;
59  bool ReservedName = false; // __foo, _Foo are usually implementation details.
60  // FIXME: make these findable once user types _.
61  bool ImplementationDetail = false;
62  unsigned References = 0;
63 
65  Unknown = 0,
75  } Category = Unknown;
76 
77  void merge(const CodeCompletionResult &SemaCCResult);
78  void merge(const Symbol &IndexResult);
79 
80  // Condense these signals down to a single number, higher is better.
81  float evaluate() const;
82 };
83 llvm::raw_ostream &operator<<(llvm::raw_ostream &,
84  const SymbolQualitySignals &);
85 
86 /// Attributes of a symbol-query pair that affect how much we like it.
88  /// The name of the symbol (for ContextWords). Must be explicitly assigned.
89  llvm::StringRef Name;
90  /// 0-1+ fuzzy-match score for unqualified name. Must be explicitly assigned.
91  float NameMatch = 1;
92  /// Lowercase words relevant to the context (e.g. near the completion point).
93  llvm::StringSet<>* ContextWords = nullptr;
94  bool Forbidden = false; // Unavailable (e.g const) or inaccessible (private).
95  /// Whether fixits needs to be applied for that completion or not.
96  bool NeedsFixIts = false;
97  bool InBaseClass = false; // A member from base class of the accessed class.
98 
100  /// These are used to calculate proximity between the index symbol and the
101  /// query.
102  llvm::StringRef SymbolURI;
103  /// FIXME: unify with index proximity score - signals should be
104  /// source-independent.
105  /// Proximity between best declaration and the query. [0-1], 1 is closest.
107 
108  // Scope proximity is only considered (both index and sema) when this is set.
110  llvm::Optional<llvm::StringRef> SymbolScope;
111  // A symbol from sema should be accessible from the current scope.
112  bool SemaSaysInScope = false;
113 
114  // An approximate measure of where we expect the symbol to be used.
120  } Scope = GlobalScope;
121 
122  enum QueryType {
125  } Query = Generic;
126 
127  CodeCompletionContext::Kind Context = CodeCompletionContext::CCC_Other;
128 
129  // Whether symbol is an instance member of a class.
130  bool IsInstanceMember = false;
131 
132  // Whether clang provided a preferred type in the completion context.
133  bool HadContextType = false;
134  // Whether a source completion item or a symbol had a type information.
135  bool HadSymbolType = false;
136  // Whether the item matches the type expected in the completion context.
137  bool TypeMatchesPreferred = false;
138 
139  void merge(const CodeCompletionResult &SemaResult);
140  void merge(const Symbol &IndexResult);
141 
142  // Condense these signals down to a single number, higher is better.
143  float evaluate() const;
144 };
145 llvm::raw_ostream &operator<<(llvm::raw_ostream &,
146  const SymbolRelevanceSignals &);
147 
148 /// Combine symbol quality and relevance into a single score.
149 float evaluateSymbolAndRelevance(float SymbolQuality, float SymbolRelevance);
150 
151 /// TopN<T> is a lossy container that preserves only the "best" N elements.
152 template <typename T, typename Compare = std::greater<T>> class TopN {
153 public:
154  using value_type = T;
155  TopN(size_t N, Compare Greater = Compare())
156  : N(N), Greater(std::move(Greater)) {}
157 
158  // Adds a candidate to the set.
159  // Returns true if a candidate was dropped to get back under N.
160  bool push(value_type &&V) {
161  bool Dropped = false;
162  if (Heap.size() >= N) {
163  Dropped = true;
164  if (N > 0 && Greater(V, Heap.front())) {
165  std::pop_heap(Heap.begin(), Heap.end(), Greater);
166  Heap.back() = std::move(V);
167  std::push_heap(Heap.begin(), Heap.end(), Greater);
168  }
169  } else {
170  Heap.push_back(std::move(V));
171  std::push_heap(Heap.begin(), Heap.end(), Greater);
172  }
173  assert(Heap.size() <= N);
174  assert(std::is_heap(Heap.begin(), Heap.end(), Greater));
175  return Dropped;
176  }
177 
178  // Returns candidates from best to worst.
179  std::vector<value_type> items() && {
180  std::sort_heap(Heap.begin(), Heap.end(), Greater);
181  assert(Heap.size() <= N);
182  return std::move(Heap);
183  }
184 
185 private:
186  const size_t N;
187  std::vector<value_type> Heap; // Min-heap, comparator is Greater.
188  Compare Greater;
189 };
190 
191 /// Returns a string that sorts in the same order as (-Score, Tiebreak), for
192 /// LSP. (The highest score compares smallest so it sorts at the top).
193 std::string sortText(float Score, llvm::StringRef Tiebreak = "");
194 
196  uint32_t NumberOfParameters = 0;
198  CodeCompleteConsumer::OverloadCandidate::CandidateKind Kind =
199  CodeCompleteConsumer::OverloadCandidate::CandidateKind::CK_Function;
200 };
201 llvm::raw_ostream &operator<<(llvm::raw_ostream &,
202  const SignatureQualitySignals &);
203 
204 } // namespace clangd
205 } // namespace clang
206 
207 #endif // LLVM_CLANG_TOOLS_EXTRA_CLANGD_QUALITY_H
clang::clangd::SymbolRelevanceSignals::merge
void merge(const CodeCompletionResult &SemaResult)
Definition: Quality.cpp:297
clang::clangd::SymbolQualitySignals::Deprecated
bool Deprecated
Definition: Quality.h:58
clang::clangd::SymbolQualitySignals::ImplementationDetail
bool ImplementationDetail
Definition: Quality.h:61
llvm
Some operations such as code completion produce a set of candidates.
Definition: YAMLGenerator.cpp:28
clang::clangd::TopN::items
std::vector< value_type > items() &&
Definition: Quality.h:179
clang::clangd::SymbolRelevanceSignals::HadSymbolType
bool HadSymbolType
Definition: Quality.h:135
clang::clangd::SignatureQualitySignals
Definition: Quality.h:195
clang::clangd::SymbolQualitySignals::Category
enum clang::clangd::SymbolQualitySignals::SymbolCategory Category
IndexResult
const Symbol * IndexResult
Definition: CodeComplete.cpp:170
clang::clangd::evaluateSymbolAndRelevance
float evaluateSymbolAndRelevance(float SymbolQuality, float SymbolRelevance)
Combine symbol quality and relevance into a single score.
Definition: Quality.cpp:467
clang::clangd::SymbolRelevanceSignals::SymbolURI
llvm::StringRef SymbolURI
These are used to calculate proximity between the index symbol and the query.
Definition: Quality.h:102
clang::clangd::TopN::TopN
TopN(size_t N, Compare Greater=Compare())
Definition: Quality.h:155
clang::clangd::SymbolQualitySignals::Variable
Definition: Quality.h:66
clang::clangd::SymbolQualitySignals::Unknown
Definition: Quality.h:65
clang::clangd::SymbolRelevanceSignals::FunctionScope
Definition: Quality.h:116
Kind
BindArgumentKind Kind
Definition: AvoidBindCheck.cpp:59
clang::clangd::SymbolQualitySignals::ReservedName
bool ReservedName
Definition: Quality.h:59
clang::clangd::ScopeDistance
Support lookups like FileDistance, but the lookup keys are symbol scopes.
Definition: FileDistance.h:117
clang::clangd::SymbolQualitySignals
Attributes of a symbol that affect how much we like it.
Definition: Quality.h:57
clang::clangd::SymbolRelevanceSignals::Forbidden
bool Forbidden
Definition: Quality.h:94
clang::clangd::SignatureQualitySignals::NumberOfOptionalParameters
uint32_t NumberOfOptionalParameters
Definition: Quality.h:197
clang::clangd::SymbolRelevanceSignals::AccessibleScope
AccessibleScope
Definition: Quality.h:115
clang::clangd::TopN::push
bool push(value_type &&V)
Definition: Quality.h:160
FileDistance.h
clang::clangd::SymbolRelevanceSignals::evaluate
float evaluate() const
Definition: Quality.cpp:351
clang::clangd::SymbolQualitySignals::Constructor
Definition: Quality.h:70
clang::clangd::TopN
TopN<T> is a lossy container that preserves only the "best" N elements.
Definition: Quality.h:152
clang::clangd::SymbolQualitySignals::References
unsigned References
Definition: Quality.h:62
clang::clangd::TopN::value_type
T value_type
Definition: Quality.h:154
clang::clangd::SymbolQualitySignals::Destructor
Definition: Quality.h:71
SemaResult
const CodeCompletionResult * SemaResult
Definition: CodeComplete.cpp:169
clang::clangd::Symbol
The class presents a C++ symbol, e.g.
Definition: Symbol.h:36
clang::clangd::SymbolRelevanceSignals::HadContextType
bool HadContextType
Definition: Quality.h:133
clang::clangd::SignatureQualitySignals::NumberOfParameters
uint32_t NumberOfParameters
Definition: Quality.h:196
ExpectedTypes.h
clang::clangd::SymbolRelevanceSignals::Query
enum clang::clangd::SymbolRelevanceSignals::QueryType Query
clang::clangd::SymbolRelevanceSignals::Scope
enum clang::clangd::SymbolRelevanceSignals::AccessibleScope Scope
clang::clangd::SymbolRelevanceSignals::SemaFileProximityScore
float SemaFileProximityScore
FIXME: unify with index proximity score - signals should be source-independent.
Definition: Quality.h:106
clang::clangd::SymbolQualitySignals::Function
Definition: Quality.h:69
clang::clangd::SymbolRelevanceSignals::FileProximityMatch
URIDistance * FileProximityMatch
Definition: Quality.h:99
clang::clangd::operator<<
llvm::raw_ostream & operator<<(llvm::raw_ostream &OS, const CodeCompletion &C)
Definition: CodeComplete.cpp:1912
clang::clangd::SymbolRelevanceSignals::TypeMatchesPreferred
bool TypeMatchesPreferred
Definition: Quality.h:137
clang::clangd::SymbolRelevanceSignals::ScopeProximityMatch
ScopeDistance * ScopeProximityMatch
Definition: Quality.h:109
clang::clangd::SymbolRelevanceSignals::Name
llvm::StringRef Name
The name of the symbol (for ContextWords). Must be explicitly assigned.
Definition: Quality.h:89
clang::clangd::URIDistance
Definition: FileDistance.h:93
clang::clangd::SymbolQualitySignals::Namespace
Definition: Quality.h:72
clang::clangd::SymbolRelevanceSignals::IsInstanceMember
bool IsInstanceMember
Definition: Quality.h:130
clang::clangd::SymbolRelevanceSignals::SemaSaysInScope
bool SemaSaysInScope
Definition: Quality.h:112
Score
llvm::Optional< float > Score
Definition: FuzzyMatchTests.cpp:48
clang::clangd::SymbolQualitySignals::Keyword
Definition: Quality.h:73
clang::clangd::SymbolRelevanceSignals::NeedsFixIts
bool NeedsFixIts
Whether fixits needs to be applied for that completion or not.
Definition: Quality.h:96
clang::clangd::sortText
std::string sortText(float Score, llvm::StringRef Name)
Returns a string that sorts in the same order as (-Score, Tiebreak), for LSP.
Definition: Quality.cpp:485
clang::clangd::SymbolQualitySignals::Macro
Definition: Quality.h:67
clang::clangd::SymbolRelevanceSignals::ClassScope
Definition: Quality.h:117
clang::clangd::SymbolQualitySignals::Operator
Definition: Quality.h:74
clang::clangd::SymbolQualitySignals::Type
Definition: Quality.h:68
clang::clangd::SymbolRelevanceSignals::SymbolScope
llvm::Optional< llvm::StringRef > SymbolScope
Definition: Quality.h:110
clang
===– Representation.cpp - ClangDoc Representation --------—*- C++ -*-===//
Definition: ApplyReplacements.h:27
clang::clangd::SymbolQualitySignals::SymbolCategory
SymbolCategory
Definition: Quality.h:64
clang::clangd::SymbolRelevanceSignals::ContextWords
llvm::StringSet * ContextWords
Lowercase words relevant to the context (e.g. near the completion point).
Definition: Quality.h:93
clang::clangd::SymbolQualitySignals::evaluate
float evaluate() const
Definition: Quality.cpp:203
clang::clangd::SignatureQualitySignals::Kind
CodeCompleteConsumer::OverloadCandidate::CandidateKind Kind
Definition: Quality.h:198
clang::clangd::SymbolRelevanceSignals::Generic
Definition: Quality.h:124
clang::clangd::SymbolRelevanceSignals::NameMatch
float NameMatch
0-1+ fuzzy-match score for unqualified name. Must be explicitly assigned.
Definition: Quality.h:91
clang::clangd::SymbolRelevanceSignals::CodeComplete
Definition: Quality.h:123
clang::clangd::SymbolRelevanceSignals::InBaseClass
bool InBaseClass
Definition: Quality.h:97
clang::clangd::SymbolRelevanceSignals::GlobalScope
Definition: Quality.h:119
clang::clangd::SymbolRelevanceSignals
Attributes of a symbol-query pair that affect how much we like it.
Definition: Quality.h:87
clang::clangd::Context
A context is an immutable container for per-request data that must be propagated through layers that ...
Definition: Context.h:69
clang::clangd::SymbolRelevanceSignals::QueryType
QueryType
Definition: Quality.h:122
clang::clangd::SymbolRelevanceSignals::FileScope
Definition: Quality.h:118
clang::clangd::SymbolQualitySignals::merge
void merge(const CodeCompletionResult &SemaCCResult)
Definition: Quality.cpp:183