refactor(parser): split expression and control flow parsing

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

1 files changed, 237 insertions, 0 deletions

diff --git a/lib/Parser/ParserExpression.cc b/lib/Parser/ParserExpression.cc
new file mode 100644
index 0000000..3e2665f
--- /dev/null
+++ b/lib/Parser/ParserExpression.cc
@@ -0,0 +1,237 @@
+#include "cps/Parser.h"
+#include "cps/FunctionAST.h"
+#include <cstdio>
+
+using namespace cps;
+
+int Parser::GetTokPrecedence() {
+    if (BinopPrecedence.find(CurTok) == BinopPrecedence.end()) return -1;
+    int TokPrec = BinopPrecedence[CurTok];
+    if (TokPrec <= 0) return -1;
+    return TokPrec;
+}
+
+
+std::unique_ptr<ExprAST> Parser::ParseNumberExpr() {
+    if (CurTok == tok_number_real) {
+        auto Result = std::make_unique<RealExprAST>(Lex.RealVal);
+        getNextToken();
+        return Result;
+    }
+    auto Result = std::make_unique<IntegerExprAST>(Lex.NumVal);
+    getNextToken();
+    return Result;
+}
+
+std::unique_ptr<ExprAST> Parser::ParseIdentifierExpr() {
+    std::string IdName = Lex.IdentifierStr;
+    int Line = Lex.getLine();
+    getNextToken();
+
+    if (CurTok == '(') {
+        getNextToken();
+        auto Args = ParseExprList(')', true);
+        if (!Args) return nullptr;
+        if (CurTok == '[' || CurTok == '.' || CurTok == '^') {
+            fprintf(stderr, "error: record/array/pointer access on a function result is not supported at line %d\n",
+                    Lex.getLine());
+            return nullptr;
+        }
+        return std::make_unique<CallExprAST>(IdName, std::move(*Args), Line);
+    }
+
+    std::vector<DesignatorAccess> Accesses;
+    if (!ParseDesignatorSuffix(Accesses)) return nullptr;
+
+    // Degrade rule: shapes that existed before designators keep their nodes.
+    if (Accesses.empty())
+        return std::make_unique<VariableExprAST>(IdName);
+    if (Accesses.size() == 1 && Accesses[0].Kind == AccessKind::Index)
+        return std::make_unique<ArrayAccessExprAST>(IdName, std::move(Accesses[0].Indices), Line);
+    return std::make_unique<DesignatorExprAST>(IdName, std::move(Accesses), Line);
+}
+
+// Postfix accessor chain after an identifier: '[' exprs ']' | '.' IDENT | '^'
+bool Parser::ParseDesignatorSuffix(std::vector<DesignatorAccess> &Out) {
+    while (true) {
+        if (CurTok == '[') {
+            getNextToken();
+            auto Indices = ParseExprList(']', false);
+            if (!Indices) return false;
+            DesignatorAccess A;
+            A.Kind = AccessKind::Index;
+            A.Indices = std::move(*Indices);
+            Out.push_back(std::move(A));
+        } else if (CurTok == '.') {
+            getNextToken();
+            if (CurTok != tok_identifier) {
+                fprintf(stderr, "error: Expected field name after '.' at line %d\n", Lex.getLine());
+                return false;
+            }
+            DesignatorAccess A;
+            A.Kind = AccessKind::Field;
+            A.FieldName = Lex.IdentifierStr;
+            Out.push_back(std::move(A));
+            getNextToken();
+        } else if (CurTok == '^') {
+            getNextToken();
+            DesignatorAccess A;
+            A.Kind = AccessKind::Deref;
+            Out.push_back(std::move(A));
+        } else {
+            return true;
+        }
+    }
+}
+
+
+std::unique_ptr<ExprAST> Parser::ParseParenExpr() {
+    getNextToken();
+    bool IsType = CurTok == tok_integer_kw || CurTok == tok_real_kw ||
+        CurTok == tok_boolean_kw || CurTok == tok_string_kw || CurTok == tok_char_kw ||
+        CurTok == tok_c_int || CurTok == tok_c_char || CurTok == tok_c_string ||
+        CurTok == tok_c_double ||
+        (CurTok == tok_identifier && TypeNames.count(Lex.IdentifierStr));
+    if (IsType) {
+        std::string TypeName = ParseTypeName();
+        if (TypeName.empty()) return nullptr;
+        if (CurTok != ')') {
+            fprintf(stderr, "error: Expected ')' after cast type at line %d\n", Lex.getLine());
+            return nullptr;
+        }
+        getNextToken();
+        auto Operand = ParseUnary();
+        if (!Operand) return nullptr;
+        return std::make_unique<CastExprAST>(TypeName, std::move(Operand));
+    }
+    auto V = ParseExpression();
+    if (!V) return nullptr;
+    if (CurTok != ')') {
+        fprintf(stderr, "error: expected ')'\n");
+        return nullptr;
+    }
+    getNextToken();
+    return V;
+}
+
+std::unique_ptr<ExprAST> Parser::ParsePrimary() {
+    switch (CurTok) {
+    case tok_identifier: return ParseIdentifierExpr();
+    case tok_number_int: return ParseNumberExpr();
+    case tok_number_real: return ParseNumberExpr();
+    case tok_string_literal: {
+        auto Res = std::make_unique<StringExprAST>(Lex.StringVal);
+        getNextToken();
+        return Res;
+    }
+    case tok_char_literal: {
+        auto Res = std::make_unique<CharExprAST>(Lex.CharVal);
+        getNextToken();
+        return Res;
+    }
+    case tok_true: {
+        getNextToken();
+        return std::make_unique<BooleanExprAST>(true);
+    }
+    case tok_false: {
+        getNextToken();
+        return std::make_unique<BooleanExprAST>(false);
+    }
+    case '(':            return ParseParenExpr();
+    default:
+        fprintf(stderr, "error: unknown token '%c' (%d) at line %d when expecting an expression\n",
+                (char)CurTok, CurTok, Lex.getLine());
+        return nullptr;
+    }
+}
+
+std::unique_ptr<ExprAST> Parser::ParseUnary() {
+    if (CurTok == tok_not) {
+        getNextToken();
+        auto Operand = ParseUnary();
+        if (!Operand) return nullptr;
+        return std::make_unique<UnaryExprAST>(tok_not, std::move(Operand));
+    }
+
+    // ^<designator> — address-of; the operand must be an lvalue shape.
+    if (CurTok == '^') {
+        int Line = Lex.getLine();
+        getNextToken();
+        if (CurTok != tok_identifier) {
+            fprintf(stderr, "error: '^' (address-of) requires a variable, array element or record field at line %d\n",
+                    Line);
+            return nullptr;
+        }
+        std::string BaseName = Lex.IdentifierStr;
+        getNextToken();
+        if (CurTok == '(') {
+            fprintf(stderr, "error: cannot take the address of a function result at line %d\n", Line);
+            return nullptr;
+        }
+        std::vector<DesignatorAccess> Accesses;
+        if (!ParseDesignatorSuffix(Accesses)) return nullptr;
+        auto Target = std::make_unique<DesignatorExprAST>(BaseName, std::move(Accesses), Line);
+        return std::make_unique<AddrOfExprAST>(std::move(Target));
+    }
+
+    if (CurTok == '-') {
+        getNextToken();
+        auto Operand = ParseUnary();
+        if (!Operand) return nullptr;
+        return std::make_unique<UnaryExprAST>('-', std::move(Operand));
+    }
+
+    return ParsePrimary();
+}
+
+std::unique_ptr<ExprAST> Parser::ParseBinOpRHS(int ExprPrec, std::unique_ptr<ExprAST> LHS) {
+    while (true) {
+        int TokPrec = GetTokPrecedence();
+        if (TokPrec < ExprPrec) return LHS;
+
+        int BinOp = CurTok;
+        int Line = Lex.getLine();
+        getNextToken();
+
+        auto RHS = ParseUnary();
+        if (!RHS) return nullptr;
+
+        int NextPrec = GetTokPrecedence();
+        if (TokPrec < NextPrec) {
+            RHS = ParseBinOpRHS(TokPrec + 1, std::move(RHS));
+            if (!RHS) return nullptr;
+        }
+
+        LHS = std::make_unique<BinaryExprAST>(BinOp, std::move(LHS), std::move(RHS), Line);
+    }
+}
+
+std::unique_ptr<ExprAST> Parser::ParseExpression() {
+    auto LHS = ParseUnary();
+    if (!LHS) return nullptr;
+    return ParseBinOpRHS(0, std::move(LHS));
+}
+
+
+std::optional<std::vector<std::unique_ptr<ExprAST>>> Parser::ParseExprList(char CloseDelim, bool AllowEmpty) {
+    std::vector<std::unique_ptr<ExprAST>> Elems;
+    if (AllowEmpty && CurTok == CloseDelim) {
+        getNextToken();
+        return Elems;
+    }
+
+    while (true) {
+        auto E = ParseExpression();
+        if (!E) return std::nullopt;
+        Elems.push_back(std::move(E));
+
+        if (CurTok == CloseDelim) break;
+        if (CurTok != ',') {
+            fprintf(stderr, "error: Expected ',' or '%c' at line %d\n", CloseDelim, Lex.getLine());
+            return std::nullopt;
+        }
+        getNextToken();
+    }
+    getNextToken();
+    return Elems;
+}