external/ton
1
0
Fork 0
mirror of https://github.com/ton-blockchain/ton synced 2025-02-12 11:12:16 +00:00
Code Issues Releases Wiki Activity
ton/tolk/ast-replacer.h

198 lines
10 KiB
C
Raw Permalink Normal View History

[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
/*
This file is part of TON Blockchain Library.
TON Blockchain Library is free software: you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation, either version 2 of the License, or
(at your option) any later version.
TON Blockchain Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License
along with TON Blockchain Library. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include "ast.h"
#include "platform-utils.h"
/*
* A module of implementing traversing a vertex tree and replacing any vertex to another.
* For example, to replace "beginCell()" call to "begin_cell()" in a function body (in V<ast_function>)
* regardless of the place this call is performed, you need to iterate over all the function AST,
* to find ast_function_call(beginCell), create ast_function_call(begin_cell) instead and to replace
* a pointer inside its parent.
* Inheriting from ASTVisitor makes this task quite simple, without any boilerplate.
*
* If you need just to traverse a vertex tree without replacing vertices,
* consider another api: ast-visitor.h.
*/
namespace tolk {
class ASTReplacer {
protected:
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
GNU_ATTRIBUTE_ALWAYS_INLINE static AnyExprV replace_children(const ASTExprLeaf* v) {
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
return v;
}
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
GNU_ATTRIBUTE_ALWAYS_INLINE AnyExprV replace_children(const ASTExprUnary* v) {
auto* v_mutable = const_cast<ASTExprUnary*>(v);
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
v_mutable->child = replace(v_mutable->child);
return v_mutable;
}
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
GNU_ATTRIBUTE_ALWAYS_INLINE AnyExprV replace_children(const ASTExprBinary* v) {
auto* v_mutable = const_cast<ASTExprBinary*>(v);
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
v_mutable->lhs = replace(v->lhs);
v_mutable->rhs = replace(v->rhs);
return v_mutable;
}
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
GNU_ATTRIBUTE_ALWAYS_INLINE AnyExprV replace_children(const ASTExprVararg* v) {
auto* v_mutable = const_cast<ASTExprVararg*>(v);
for (AnyExprV& child : v_mutable->children) {
child = replace(child);
}
return v_mutable;
}
GNU_ATTRIBUTE_ALWAYS_INLINE AnyV replace_children(const ASTStatementUnary* v) {
auto* v_mutable = const_cast<ASTStatementUnary*>(v);
v_mutable->child = replace(v_mutable->child);
return v_mutable;
}
GNU_ATTRIBUTE_ALWAYS_INLINE AnyV replace_children(const ASTStatementVararg* v) {
auto* v_mutable = const_cast<ASTStatementVararg*>(v);
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
for (AnyV& child : v_mutable->children) {
child = replace(child);
}
return v_mutable;
}
public:
virtual ~ASTReplacer() = default;
virtual AnyV replace(AnyV v) = 0;
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
virtual AnyExprV replace(AnyExprV v) = 0;
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
};
class ASTReplacerInFunctionBody : public ASTReplacer {
protected:
using parent = ASTReplacerInFunctionBody;
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
// expressions
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
virtual AnyExprV replace(V<ast_empty_expression> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_parenthesized_expression> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_tensor> v) { return replace_children(v); }
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
virtual AnyExprV replace(V<ast_typed_tuple> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_reference> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_local_var_lhs> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_local_vars_declaration> v) { return replace_children(v); }
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
virtual AnyExprV replace(V<ast_int_const> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_string_const> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_bool_const> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_null_keyword> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_argument> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_argument_list> v) { return replace_children(v); }
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
virtual AnyExprV replace(V<ast_dot_access> v) { return replace_children(v); }
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
virtual AnyExprV replace(V<ast_function_call> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_underscore> v) { return replace_children(v); }
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
virtual AnyExprV replace(V<ast_assign> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_set_assign> v) { return replace_children(v); }
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
virtual AnyExprV replace(V<ast_unary_operator> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_binary_operator> v) { return replace_children(v); }
virtual AnyExprV replace(V<ast_ternary_operator> v) { return replace_children(v); }
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
virtual AnyExprV replace(V<ast_cast_as_operator> v) { return replace_children(v); }
// statements
virtual AnyV replace(V<ast_empty_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_sequence> v) { return replace_children(v); }
virtual AnyV replace(V<ast_return_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_if_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_repeat_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_while_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_do_while_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_throw_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_assert_statement> v) { return replace_children(v); }
virtual AnyV replace(V<ast_try_catch_statement> v) { return replace_children(v); }
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
AnyExprV replace(AnyExprV v) final {
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
switch (v->type) {
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
case ast_empty_expression: return replace(v->as<ast_empty_expression>());
case ast_parenthesized_expression: return replace(v->as<ast_parenthesized_expression>());
[Tolk] v0.6 syntax: `fun`, `import`, `var`, types on the right, etc. Lots of changes, actually. Most noticeable are: - traditional //comments - #include -> import - a rule "import what you use" - ~ found -> !found (for -1/0) - null() -> null - is_null?(v) -> v == null - throw is a keyword - catch with swapped arguments - throw_if, throw_unless -> assert - do until -> do while - elseif -> else if - drop ifnot, elseifnot - drop rarely used operators A testing framework also appears here. All tests existed earlier, but due to significant syntax changes, their history is useless.
2024-10-31 07:11:41 +00:00
case ast_tensor: return replace(v->as<ast_tensor>());
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
case ast_typed_tuple: return replace(v->as<ast_typed_tuple>());
case ast_reference: return replace(v->as<ast_reference>());
case ast_local_var_lhs: return replace(v->as<ast_local_var_lhs>());
case ast_local_vars_declaration: return replace(v->as<ast_local_vars_declaration>());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
case ast_int_const: return replace(v->as<ast_int_const>());
case ast_string_const: return replace(v->as<ast_string_const>());
case ast_bool_const: return replace(v->as<ast_bool_const>());
[Tolk] v0.6 syntax: `fun`, `import`, `var`, types on the right, etc. Lots of changes, actually. Most noticeable are: - traditional //comments - #include -> import - a rule "import what you use" - ~ found -> !found (for -1/0) - null() -> null - is_null?(v) -> v == null - throw is a keyword - catch with swapped arguments - throw_if, throw_unless -> assert - do until -> do while - elseif -> else if - drop ifnot, elseifnot - drop rarely used operators A testing framework also appears here. All tests existed earlier, but due to significant syntax changes, their history is useless.
2024-10-31 07:11:41 +00:00
case ast_null_keyword: return replace(v->as<ast_null_keyword>());
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
case ast_argument: return replace(v->as<ast_argument>());
case ast_argument_list: return replace(v->as<ast_argument_list>());
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
case ast_dot_access: return replace(v->as<ast_dot_access>());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
case ast_function_call: return replace(v->as<ast_function_call>());
case ast_underscore: return replace(v->as<ast_underscore>());
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
case ast_assign: return replace(v->as<ast_assign>());
case ast_set_assign: return replace(v->as<ast_set_assign>());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
case ast_unary_operator: return replace(v->as<ast_unary_operator>());
case ast_binary_operator: return replace(v->as<ast_binary_operator>());
case ast_ternary_operator: return replace(v->as<ast_ternary_operator>());
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
case ast_cast_as_operator: return replace(v->as<ast_cast_as_operator>());
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
default:
throw UnexpectedASTNodeType(v, "ASTReplacerInFunctionBody::replace");
}
}
AnyV replace(AnyV v) final {
switch (v->type) {
case ast_empty_statement: return replace(v->as<ast_empty_statement>());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
case ast_sequence: return replace(v->as<ast_sequence>());
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
case ast_return_statement: return replace(v->as<ast_return_statement>());
case ast_if_statement: return replace(v->as<ast_if_statement>());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
case ast_repeat_statement: return replace(v->as<ast_repeat_statement>());
case ast_while_statement: return replace(v->as<ast_while_statement>());
[Tolk] v0.6 syntax: `fun`, `import`, `var`, types on the right, etc. Lots of changes, actually. Most noticeable are: - traditional //comments - #include -> import - a rule "import what you use" - ~ found -> !found (for -1/0) - null() -> null - is_null?(v) -> v == null - throw is a keyword - catch with swapped arguments - throw_if, throw_unless -> assert - do until -> do while - elseif -> else if - drop ifnot, elseifnot - drop rarely used operators A testing framework also appears here. All tests existed earlier, but due to significant syntax changes, their history is useless.
2024-10-31 07:11:41 +00:00
case ast_do_while_statement: return replace(v->as<ast_do_while_statement>());
case ast_throw_statement: return replace(v->as<ast_throw_statement>());
case ast_assert_statement: return replace(v->as<ast_assert_statement>());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
case ast_try_catch_statement: return replace(v->as<ast_try_catch_statement>());
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
#ifdef TOLK_DEBUG
case ast_asm_body:
throw UnexpectedASTNodeType(v, "ASTReplacer::replace");
#endif
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
default: {
// be very careful, don't forget to handle all statements (not expressions) above!
AnyExprV as_expr = reinterpret_cast<const ASTNodeExpressionBase*>(v);
return replace(as_expr);
}
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
}
}
public:
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
virtual bool should_visit_function(const FunctionData* fun_ref) = 0;
void start_replacing_in_function(const FunctionData* fun_ref, V<ast_function_declaration> v_function) {
replace(v_function->get_body());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
}
};
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
const std::vector<const FunctionData*>& get_all_not_builtin_functions();
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
template<class BodyReplacerT>
[Tolk] Rewrite the type system from Hindley-Milner to static typing FunC's (and Tolk's before this PR) type system is based on Hindley-Milner. This is a common approach for functional languages, where types are inferred from usage through unification. As a result, type declarations are not necessary: () f(a,b) { return a+b; } // a and b now int, since `+` (int, int) While this approach works for now, problems arise with the introduction of new types like bool, where `!x` must handle both int and bool. It will also become incompatible with int32 and other strict integers. This will clash with structure methods, struggle with proper generics, and become entirely impractical for union types. This PR completely rewrites the type system targeting the future. 1) type of any expression is inferred and never changed 2) this is available because dependent expressions already inferred 3) forall completely removed, generic functions introduced (they work like template functions actually, instantiated while inferring) 4) instantiation `<...>` syntax, example: `t.tupleAt<int>(0)` 5) `as` keyword, for example `t.tupleAt(0) as int` 6) methods binding is done along with type inferring, not before ("before", as worked previously, was always a wrong approach)
2024-12-30 15:31:27 +00:00
void replace_ast_of_all_functions() {
BodyReplacerT visitor;
for (const FunctionData* fun_ref : get_all_not_builtin_functions()) {
if (visitor.should_visit_function(fun_ref)) {
visitor.start_replacing_in_function(fun_ref, fun_ref->ast_root->as<ast_function_declaration>());
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
}
}
[Tolk] AST-based semantic analysis, get rid of Expr This is a huge refactoring focusing on untangling compiler internals (previously forked from FunC). The goal is to convert AST directly to Op (a kind of IR representation), doing all code analysis at AST level. Noteable changes: - AST-based semantic kernel includes: registering global symbols, scope handling and resolving local/global identifiers, lvalue/rvalue calc and check, implicit return detection, mutability analysis, pure/impure validity checks, simple constant folding - values of `const` variables are calculated NOT based on CodeBlob, but via a newly-introduced AST-based constant evaluator - AST vertices are now inherited from expression/statement/other; expression vertices have common properties (TypeExpr, lvalue/rvalue) - symbol table is rewritten completely, SymDef/SymVal no longer exist, lexer now doesn't need to register identifiers - AST vertices have references to symbols, filled at different stages of pipeline - the remaining "FunC legacy part" is almost unchanged besides Expr which was fully dropped; AST is converted to Ops (IR) directly
2024-12-16 18:19:45 +00:00
}
[Tolk] Implement AST: intermediate representation of tolk files Now, the whole .tolk file can be loaded as AST tree and then converted to Expr/Op. This gives a great ability to implement AST transformations. In the future, more and more code analysis will be moved out of legacy to AST-level.
2024-10-31 07:03:33 +00:00
} // namespace tolk
Reference in a new issue Copy permalink