| 1 | //===--- Language.h -------------------------------------------- -*- 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 CLANG_PSEUDO_LANGUAGE_H |
|---|
| 10 | #define CLANG_PSEUDO_LANGUAGE_H |
|---|
| 11 | |
|---|
| 12 | #include "clang-pseudo/Token.h" |
|---|
| 13 | #include "clang-pseudo/grammar/Grammar.h" |
|---|
| 14 | #include "clang-pseudo/grammar/LRTable.h" |
|---|
| 15 | |
|---|
| 16 | namespace clang { |
|---|
| 17 | namespace pseudo { |
|---|
| 18 | class ForestNode; |
|---|
| 19 | class TokenStream; |
|---|
| 20 | class LRTable; |
|---|
| 21 | |
|---|
| 22 | struct GuardParams { |
|---|
| 23 | llvm::ArrayRef<const ForestNode *> RHS; |
|---|
| 24 | const TokenStream &Tokens; |
|---|
| 25 | // FIXME: use the index of Tokens. |
|---|
| 26 | SymbolID Lookahead; |
|---|
| 27 | }; |
|---|
| 28 | // A guard restricts when a grammar rule can be used. |
|---|
| 29 | // |
|---|
| 30 | // The GLR parser will use the guard to determine whether a rule reduction will |
|---|
| 31 | // be conducted. For example, e.g. a guard may allow the rule |
|---|
| 32 | // `virt-specifier := IDENTIFIER` only if the identifier's text is 'override`. |
|---|
| 33 | // |
|---|
| 34 | // Return true if the guard is satisfied. |
|---|
| 35 | using RuleGuard = llvm::function_ref<bool(const GuardParams &)>; |
|---|
| 36 | |
|---|
| 37 | // A recovery strategy determines a region of code to skip when parsing fails. |
|---|
| 38 | // |
|---|
| 39 | // For example, given `class-def := CLASS IDENT { body [recover=Brackets] }`, |
|---|
| 40 | // if parsing fails while attempting to parse `body`, we may skip up to the |
|---|
| 41 | // matching `}` and assume everything between was a `body`. |
|---|
| 42 | // |
|---|
| 43 | // The provided index is the token where the skipped region begins. |
|---|
| 44 | // Returns the (excluded) end of the range, or Token::Invalid for no recovery. |
|---|
| 45 | using RecoveryStrategy = |
|---|
| 46 | llvm::function_ref<Token::Index(Token::Index Start, const TokenStream &)>; |
|---|
| 47 | |
|---|
| 48 | // Specify a language that can be parsed by the pseduoparser. |
|---|
| 49 | struct Language { |
|---|
| 50 | Grammar G; |
|---|
| 51 | LRTable Table; |
|---|
| 52 | |
|---|
| 53 | // Binding extension ids to corresponding implementations. |
|---|
| 54 | llvm::DenseMap<RuleID, RuleGuard> Guards; |
|---|
| 55 | llvm::DenseMap<ExtensionID, RecoveryStrategy> RecoveryStrategies; |
|---|
| 56 | |
|---|
| 57 | // FIXME: add clang::LangOptions. |
|---|
| 58 | // FIXME: add default start symbols. |
|---|
| 59 | }; |
|---|
| 60 | |
|---|
| 61 | } // namespace pseudo |
|---|
| 62 | } // namespace clang |
|---|
| 63 | |
|---|
| 64 | #endif // CLANG_PSEUDO_LANGUAGE_H |
|---|
| 65 | |
|---|
