| 1 | //===--- FuzzySymbolIndex.h - Lookup symbols for autocomplete ---*- 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 | #ifndef LLVM_CLANG_TOOLS_EXTRA_INCLUDE_FIXER_FUZZY_SYMBOL_INDEX_H |
|---|
| 10 | #define |
|---|
| 11 | |
|---|
| 12 | #include "SymbolIndex.h" |
|---|
| 13 | #include "find-all-symbols/SymbolInfo.h" |
|---|
| 14 | #include "llvm/ADT/SmallString.h" |
|---|
| 15 | #include "llvm/ADT/StringRef.h" |
|---|
| 16 | #include "llvm/Support/Error.h" |
|---|
| 17 | #include <string> |
|---|
| 18 | #include <vector> |
|---|
| 19 | |
|---|
| 20 | namespace clang { |
|---|
| 21 | namespace include_fixer { |
|---|
| 22 | |
|---|
| 23 | // A FuzzySymbolIndex retrieves top-level symbols matching a query string. |
|---|
| 24 | // |
|---|
| 25 | // It refines the contract of SymbolIndex::search to do fuzzy matching: |
|---|
| 26 | // - symbol names are tokenized: "unique ptr", "string ref". |
|---|
| 27 | // - query must match prefixes of symbol tokens: [upt] |
|---|
| 28 | // - if the query has multiple tokens, splits must match: [StR], not [STr]. |
|---|
| 29 | // Helpers for tokenization and regex matching are provided. |
|---|
| 30 | // |
|---|
| 31 | // Implementations may choose to truncate results, refuse short queries, etc. |
|---|
| 32 | class FuzzySymbolIndex : public SymbolIndex { |
|---|
| 33 | public: |
|---|
| 34 | // Loads the specified clang-include-fixer database and returns an index serving it. |
|---|
| 35 | static llvm::Expected<std::unique_ptr<FuzzySymbolIndex>> |
|---|
| 36 | createFromYAML(llvm::StringRef File); |
|---|
| 37 | |
|---|
| 38 | // Helpers for implementing indexes: |
|---|
| 39 | |
|---|
| 40 | // Transforms a symbol name or query into a sequence of tokens. |
|---|
| 41 | // - URLHandlerCallback --> [url, handler, callback] |
|---|
| 42 | // - snake_case11 --> [snake, case, 11] |
|---|
| 43 | // - _WTF$ --> [wtf] |
|---|
| 44 | static std::vector<std::string> tokenize(llvm::StringRef Text); |
|---|
| 45 | |
|---|
| 46 | // Transforms query tokens into an unanchored regexp to match symbol tokens. |
|---|
| 47 | // - [fe f] --> /f(\w* )?e\w* f/, matches [fee fie foe]. |
|---|
| 48 | static std::string queryRegexp(const std::vector<std::string> &Tokens); |
|---|
| 49 | }; |
|---|
| 50 | |
|---|
| 51 | } // namespace include_fixer |
|---|
| 52 | } // namespace clang |
|---|
| 53 | |
|---|
| 54 | #endif // LLVM_CLANG_TOOLS_EXTRA_INCLUDE_FIXER_FUZZY_SYMBOL_INDEX_H |
|---|
| 55 | |
|---|
