Add LAC (Lookahead Correction) support

NEWS.md

@@ -2,6 +2,27 @@
 
 ## Lrama 0.8.0 (2026-xx-xx)
 
+### LAC (Lookahead Correction) Support
+
+Added support for LAC (Lookahead Correction), which improves syntax error reporting by providing more accurate error messages. LAC can be enabled using the `%define parse.lac full` directive in grammar files.
+
+Key features:
+- More precise "expected token" lists in syntax error messages
+- Early error detection (before default reductions)
+- Better handling of `%nonassoc` operators
+- Compatible with existing LALR parser generation
+
+Example usage:
+```yacc
+%define parse.lac full
+%define parse.error verbose
+
+%%
+expr: expr '+' expr
+    | NUMBER
+    ;
+```
+
 ## Lrama 0.7.1 (2025-12-24)
 
 ### Optimize IELR

README.md

@@ -30,7 +30,10 @@ Lrama (pronounced in the same way as the noun “llama” in English) is LALR (1
   * b4_locations_if is always true
   * b4_pure_if is always true
   * b4_pull_if is always false
-  * b4_lac_if is always false
+* LAC (Lookahead Correction)
+  * Improves syntax error reporting with more accurate "expected token" messages
+  * Enable with `%define parse.lac full`
+  * Compatible with LALR parser generation
 * Error Tolerance parser
   * Subset of [Repairing Syntax Errors in LR Parsers (Corchuelo et al.)](https://idus.us.es/bitstream/handle/11441/65631/Repairing%20syntax%20errors.pdf) algorithm is supported
 * Parameterizing rules

lib/lrama/grammar.rb

@@ -277,6 +277,7 @@ def validate!
       validate_no_precedence_for_nterm!
       validate_rule_lhs_is_nterm!
       validate_duplicated_precedence!
+      validate_parse_lac!
     end
 
     # @rbs (Grammar::Symbol sym) -> Array[Rule]
@@ -304,6 +305,16 @@ def ielr_defined?
       @define.key?('lr.type') && @define['lr.type'] == 'ielr'
     end
 
+    # @rbs () -> String
+    def parse_lac
+      @define['parse.lac'] || 'none'
+    end
+
+    # @rbs () -> bool
+    def lac_enabled?
+      parse_lac == 'full'
+    end
+
     private
 
     # @rbs () -> void
@@ -599,5 +610,17 @@ def validate_duplicated_precedence!
     def set_locations
       @locations = @locations || @rules.any? {|rule| rule.contains_at_reference? }
     end
+
+    # @rbs () -> void
+    def validate_parse_lac!
+      return unless @define.key?('parse.lac')
+
+      value = @define['parse.lac']
+      valid_values = ['none', 'full']
+
+      unless valid_values.include?(value)
+        raise "Invalid value for parse.lac: '#{value}'. Expected 'none' or 'full'."
+      end
+    end
   end
 end

lib/lrama/output.rb

@@ -372,6 +372,10 @@ def aux
       @grammar.aux
     end
 
+    def lac_enabled?
+      @grammar.lac_enabled?
+    end
+
     def int_array_to_string(ary)
       last = ary.count - 1
 

spec/fixtures/integration/lac.y

@@ -0,0 +1,46 @@
+%{
+#include <stdio.h>
+#include <stdlib.h>
+
+int yylex(YYSTYPE *yylval, YYLTYPE *yylloc);
+void yyerror(YYLTYPE *yylloc, const char *s);
+%}
+
+%define parse.lac full
+%define parse.error verbose
+%locations
+
+%token NUMBER
+
+%left '+' '-'
+%left '*' '/'
+
+%%
+
+program: expr
+       ;
+
+expr: expr '+' expr
+    | expr '-' expr
+    | expr '*' expr
+    | expr '/' expr
+    | '(' expr ')'
+    | NUMBER
+    ;
+
+%%
+
+void yyerror(YYLTYPE *yylloc, const char *s) {
+  (void)yylloc;
+  fprintf(stderr, "Error: %s\n", s);
+}
+
+int yylex(YYSTYPE *yylval, YYLTYPE *yylloc) {
+  (void)yylval;
+  (void)yylloc;
+  return 0;
+}
+
+int main() {
+  return yyparse();
+}

spec/fixtures/integration/lac_accurate_tokens.l

@@ -0,0 +1,48 @@
+%option noinput nounput noyywrap never-interactive bison-bridge bison-locations
+
+%{
+#include <stdio.h>
+#include <stdlib.h>
+#include "lac_accurate_tokens.h"
+%}
+
+%%
+
+"if"    { return IF; }
+"then"  { return THEN; }
+"else"  { return ELSE; }
+"while" { return WHILE; }
+"do"    { return DO; }
+
+[a-zA-Z_][a-zA-Z0-9_]* {
+  (void)yylval;
+  (void)yylloc;
+  return ID;
+}
+
+[0-9]+ {
+  (void)yylval;
+  (void)yylloc;
+  return NUM;
+}
+
+[+\-*/] {
+  return yytext[0];
+}
+
+[\n|\r\n] {
+  return(YYEOF);
+}
+
+[[:space:]] {}
+
+<<EOF>> {
+  return(YYEOF);
+}
+
+. {
+  fprintf(stderr, "Illegal character '%s'\n", yytext);
+  return(YYEOF);
+}
+
+%%

spec/fixtures/integration/lac_accurate_tokens.y

@@ -0,0 +1,65 @@
+%{
+#include <stdio.h>
+#include "lac_accurate_tokens.h"
+#include "lac_accurate_tokens-lexer.h"
+
+static int yyerror(YYLTYPE *loc, const char *str);
+%}
+
+%define parse.lac full
+%define parse.error verbose
+
+%token IF THEN ELSE WHILE DO ID NUM
+
+%left '+' '-'
+%left '*' '/'
+
+%locations
+
+%%
+
+program: stmt_list
+       ;
+
+stmt_list: stmt
+         | stmt_list stmt
+         ;
+
+stmt: if_stmt
+    | while_stmt
+    | expr
+    ;
+
+if_stmt: IF expr THEN stmt
+       | IF expr THEN stmt ELSE stmt
+       ;
+
+while_stmt: WHILE expr DO stmt
+          ;
+
+expr: expr '+' expr
+    | expr '-' expr
+    | expr '*' expr
+    | expr '/' expr
+    | NUM
+    | ID
+    ;
+
+%%
+
+static int yyerror(YYLTYPE *loc, const char *str) {
+  (void)loc;
+  printf("Error: %s\n", str);
+  return 0;
+}
+
+int main(int argc, char *argv[]) {
+  if (argc == 2) {
+    yy_scan_string(argv[1]);
+  }
+
+  if (yyparse()) {
+    return 1;
+  }
+  return 0;
+}

spec/fixtures/integration/lac_default_reduction.l

@@ -0,0 +1,42 @@
+%option noinput nounput noyywrap never-interactive bison-bridge bison-locations
+
+%{
+#include <stdio.h>
+#include <stdlib.h>
+#include "lac_default_reduction.h"
+%}
+
+%%
+
+[a-zA-Z_][a-zA-Z0-9_]* {
+  (void)yylval;
+  (void)yylloc;
+  return ID;
+}
+
+[0-9]+ {
+  (void)yylval;
+  (void)yylloc;
+  return NUM;
+}
+
+[=+*;] {
+  return yytext[0];
+}
+
+[\n|\r\n] {
+  return(YYEOF);
+}
+
+[[:space:]] {}
+
+<<EOF>> {
+  return(YYEOF);
+}
+
+. {
+  fprintf(stderr, "Illegal character '%s'\n", yytext);
+  return(YYEOF);
+}
+
+%%

spec/fixtures/integration/lac_default_reduction.y

@@ -0,0 +1,48 @@
+%{
+#include <stdio.h>
+#include "lac_default_reduction.h"
+#include "lac_default_reduction-lexer.h"
+
+static int yyerror(YYLTYPE *loc, const char *str);
+%}
+
+%define parse.lac full
+%define parse.error verbose
+
+%token ID NUM
+
+%locations
+
+%%
+
+program: stmt
+       ;
+
+stmt: ID '=' expr ';'
+    | expr ';'
+    ;
+
+expr: expr '+' expr
+    | expr '*' expr
+    | NUM
+    | ID
+    ;
+
+%%
+
+static int yyerror(YYLTYPE *loc, const char *str) {
+  (void)loc;
+  printf("Error: %s\n", str);
+  return 0;
+}
+
+int main(int argc, char *argv[]) {
+  if (argc == 2) {
+    yy_scan_string(argv[1]);
+  }
+
+  if (yyparse()) {
+    return 1;
+  }
+  return 0;
+}