lrama 0.5.7 → 0.5.9

data/lib/lrama/version.rb

@@ -1,3 +1,3 @@
 module Lrama
-  VERSION = "0.5.7".freeze
+  VERSION = "0.5.9".freeze
 end

data/parser.y

@@ -1,5 +1,10 @@
 class Lrama::Parser
+  expect 7
+
+  token C_DECLARATION CHARACTER IDENT_COLON IDENTIFIER INTEGER STRING TAG
+
 rule
+
   input: prologue_declarations bison_declarations "%%" grammar epilogue_opt
 
   prologue_declarations: # empty
@@ -7,14 +12,12 @@ rule
 
   prologue_declaration: "%{"
                           {
-                            @lexer.status = :c_declaration
-                            @lexer.end_symbol = '%}'
+                            begin_c_declaration("%}")
                             @grammar.prologue_first_lineno = @lexer.line
                           }
                         C_DECLARATION
                           {
-                            @lexer.status = :initial
-                            @lexer.end_symbol = nil
+                            end_c_declaration
                           }
                         "%}"
                           {
@@ -44,15 +47,25 @@ rule
                            @grammar.parse_param = @grammar.build_code(:parse_param, token).token_code.s_value
                          }
                        }
+                   | "%code" IDENTIFIER "{"
+                       {
+                         begin_c_declaration("}")
+                       }
+                     C_DECLARATION
+                       {
+                         end_c_declaration
+                       }
+                     "}"
+                       {
+                         @grammar.add_percent_code(id: val[1], code: val[4])
+                       }
                    | "%initial-action" "{"
                        {
-                         @lexer.status = :c_declaration
-                         @lexer.end_symbol = '}'
+                         begin_c_declaration("}")
                        }
                      C_DECLARATION
                        {
-                         @lexer.status = :initial
-                         @lexer.end_symbol = nil
+                         end_c_declaration
                        }
                      "}"
                        {
@@ -62,13 +75,11 @@ rule
 
   grammar_declaration: "%union" "{"
                          {
-                           @lexer.status = :c_declaration
-                           @lexer.end_symbol = '}'
+                           begin_c_declaration("}")
                          }
                        C_DECLARATION
                          {
-                           @lexer.status = :initial
-                           @lexer.end_symbol = nil
+                           end_c_declaration
                          }
                        "}"
                          {
@@ -77,24 +88,20 @@ rule
                      | symbol_declaration
                      | "%destructor" "{"
                          {
-                           @lexer.status = :c_declaration
-                           @lexer.end_symbol = '}'
+                           begin_c_declaration("}")
                          }
                        C_DECLARATION
                          {
-                           @lexer.status = :initial
-                           @lexer.end_symbol = nil
+                           end_c_declaration
                          }
-                         "}" generic_symlist
+                       "}" generic_symlist
                      | "%printer" "{"
                          {
-                           @lexer.status = :c_declaration
-                           @lexer.end_symbol = '}'
+                           begin_c_declaration("}")
                          }
                        C_DECLARATION
                          {
-                           @lexer.status = :initial
-                           @lexer.end_symbol = nil
+                           end_c_declaration
                          }
                        "}" generic_symlist
                          {
@@ -102,13 +109,11 @@ rule
                          }
                      | "%error-token" "{"
                          {
-                           @lexer.status = :c_declaration
-                           @lexer.end_symbol = '}'
+                           begin_c_declaration("}")
                          }
                        C_DECLARATION
                          {
-                           @lexer.status = :initial
-                           @lexer.end_symbol = nil
+                           end_c_declaration
                          }
                        "}" generic_symlist
                          {
@@ -216,13 +221,11 @@ rule
 
   params: params "{"
             {
-              @lexer.status = :c_declaration
-              @lexer.end_symbol = '}'
+              begin_c_declaration("}")
             }
           C_DECLARATION
             {
-              @lexer.status = :initial
-              @lexer.end_symbol = nil
+              end_c_declaration
             }
           "}"
             {
@@ -230,13 +233,11 @@ rule
             }
         | "{"
             {
-              @lexer.status = :c_declaration
-              @lexer.end_symbol = '}'
+              begin_c_declaration("}")
             }
           C_DECLARATION
             {
-              @lexer.status = :initial
-              @lexer.end_symbol = nil
+              end_c_declaration
             }
           "}"
             {
@@ -273,7 +274,7 @@ rule
   rules: id_colon named_ref_opt ":" rhs_list
            {
              lhs = val[0]
-             lhs.alias = val[1]
+             lhs.alias_name = val[1]
              val[3].each {|hash|
                @grammar.add_rule(lhs: lhs, rhs: hash[:rhs], lineno: hash[:lineno])
              }
@@ -291,56 +292,53 @@ rule
 
   rhs: /* empty */
          {
+           reset_precs
+           result = []
+         }
+     | "%empty"
+         {
+           reset_precs
            result = []
-           @prec_seen = false
-           @code_after_prec = false
          }
      | rhs symbol named_ref_opt
          {
            token = val[1]
-           val[1].alias = val[2]
+           token.alias_name = val[2]
            result = val[0].append(token)
          }
+     | rhs "?"
+          {
+            token = Lrama::Lexer::Token::Parameterizing.new(s_value: val[1])
+            result = val[0].append(token)
+          }
+     | rhs "+"
+          {
+            token = Lrama::Lexer::Token::Parameterizing.new(s_value: val[1])
+            result = val[0].append(token)
+          }
+     | rhs "*"
+          {
+             token = Lrama::Lexer::Token::Parameterizing.new(s_value: val[1])
+             result = val[0].append(token)
+          }
      | rhs "{"
          {
            if @prec_seen
              raise "Multiple User_code after %prec" if @code_after_prec
              @code_after_prec = true
            end
-           @lexer.status = :c_declaration
-           @lexer.end_symbol = '}'
+           begin_c_declaration("}")
          }
        C_DECLARATION
          {
-           @lexer.status = :initial
-           @lexer.end_symbol = nil
+           end_c_declaration
          }
        "}" named_ref_opt
          {
            token = val[3]
-           token.alias = val[6]
+           token.alias_name = val[6]
            result = val[0].append(token)
          }
-     | "{"
-         {
-           if @prec_seen
-             raise "Multiple User_code after %prec" if @code_after_prec
-             @code_after_prec = true
-           end
-           @lexer.status = :c_declaration
-           @lexer.end_symbol = '}'
-         }
-       C_DECLARATION
-         {
-           @lexer.status = :initial
-           @lexer.end_symbol = nil
-         }
-       "}" named_ref_opt
-         {
-           token = val[2]
-           token.alias = val[5]
-           result = [token]
-         }
      | rhs "%prec" symbol
          {
            sym = @grammar.find_symbol_by_id!(val[2])
@@ -356,14 +354,12 @@ rule
   epilogue_opt: # empty
               | "%%"
                   {
-                    @lexer.status = :c_declaration
-                    @lexer.end_symbol = '\Z'
+                    begin_c_declaration('\Z')
                     @grammar.epilogue_first_lineno = @lexer.line + 1
                   }
                 C_DECLARATION
                   {
-                    @lexer.status = :initial
-                    @lexer.end_symbol = nil
+                    end_c_declaration
                     @grammar.epilogue = val[2].s_value
                   }
 
@@ -380,15 +376,17 @@ rule
   generic_symlist_item: symbol
                       | TAG
 
-  string_as_id: STRING { result = Lrama::Lexer::Token.new(type: Lrama::Lexer::Token::Ident, s_value: val[0]) }
+  string_as_id: STRING { result = Lrama::Lexer::Token::Ident.new(s_value: val[0]) }
 end
 
 ---- inner
 
 include Lrama::Report::Duration
 
-def initialize(text)
+def initialize(text, path, debug = false)
   @text = text
+  @path = path
+  @yydebug = debug
 end
 
 def parse
@@ -396,8 +394,8 @@ def parse
     @lexer = Lrama::Lexer.new(@text)
     @grammar = Lrama::Grammar.new
     @precedence_number = 0
+    reset_precs
     do_parse
-    @grammar.extract_references
     @grammar.prepare
     @grammar.compute_nullable
     @grammar.compute_first_set
@@ -411,6 +409,38 @@ def next_token
 end
 
 def on_error(error_token_id, error_value, value_stack)
-  raise ParseError, sprintf("\n%d:%d: parse error on value %s (%s)",
-                            @lexer.line, @lexer.column, error_value.inspect, token_to_str(error_token_id) || '?')
+  if error_value.respond_to?(:line) && error_value.respond_to?(:column)
+    line = error_value.line
+    first_column = error_value.column
+  else
+    line = @lexer.line
+    first_column = @lexer.head_column
+  end
+
+  raise ParseError, <<~ERROR
+    #{@path}:#{line}:#{first_column}: parse error on value #{error_value.inspect} (#{token_to_str(error_token_id) || '?'})
+    #{@text.split("\n")[line - 1]}
+    #{carrets(first_column)}
+  ERROR
+end
+
+private
+
+def reset_precs
+  @prec_seen = false
+  @code_after_prec = false
+end
+
+def begin_c_declaration(end_symbol)
+  @lexer.status = :c_declaration
+  @lexer.end_symbol = end_symbol
+end
+
+def end_c_declaration
+  @lexer.status = :initial
+  @lexer.end_symbol = nil
+end
+
+def carrets(first_column)
+  ' ' * (first_column + 1) + '^' * (@lexer.column - first_column)
 end

data/rbs_collection.lock.yaml

@@ -2,7 +2,7 @@
 sources:
 - type: git
   name: ruby/gem_rbs_collection
-  revision: 95ad664324500c9eec78569b45da98c65a27a511
+  revision: 2de2d4535caba275f3b8533684aab110d921f553
   remote: https://github.com/ruby/gem_rbs_collection.git
   repo_dir: gems
 path: ".gem_rbs_collection"
@@ -11,6 +11,18 @@ gems:
   version: '0'
   source:
     type: stdlib
+- name: fileutils
+  version: '0'
+  source:
+    type: stdlib
+- name: rake
+  version: '13.0'
+  source:
+    type: git
+    name: ruby/gem_rbs_collection
+    revision: 2de2d4535caba275f3b8533684aab110d921f553
+    remote: https://github.com/ruby/gem_rbs_collection.git
+    repo_dir: gems
 - name: stackprof
   version: '0.2'
   source:

data/sample/calc.y

@@ -15,10 +15,12 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <ctype.h>
+%}
 
+%code provides {
 static int yylex(YYSTYPE *val, YYLTYPE *loc);
 static int yyerror(YYLTYPE *loc, const char *str);
-%}
+}
 
 %union {
     int val;

data/sample/parse.y

@@ -4,10 +4,14 @@
 
 %{
 // Prologue
+%}
+
+%code provides {
+
 static enum yytokentype yylex(YYSTYPE *lval, YYLTYPE *yylloc);
 static void yyerror(YYLTYPE *yylloc, const char *msg);
 
-%}
+}
 
 %expect 0
 %define api.pure

data/sig/lrama/grammar/percent_code.rbs

@@ -0,0 +1,10 @@
+module Lrama
+  class Grammar
+    class PercentCode
+      attr_reader id: Lexer::Token::Ident
+      attr_reader code: Lexer::Token::UserCode
+
+      def initialize: (Lexer::Token::Ident id, Lexer::Token::UserCode code) -> void
+    end
+  end
+end

data/sig/lrama/grammar/reference.rbs

@@ -0,0 +1,22 @@
+module Lrama
+  class Grammar
+    class Reference
+      # TODO: Replace untyped referring_symbol with (Grammar::Symbol|Lexer::Token)
+      attr_accessor type: Symbol
+      attr_accessor value: (String|Integer)
+      attr_accessor ex_tag: Lexer::Token?
+      attr_accessor first_column: Integer
+      attr_accessor last_column: Integer
+      attr_accessor referring_symbol: untyped
+      attr_accessor position_in_rhs: Integer?
+
+      def initialize: (
+        type: Symbol, value: (String|Integer), ex_tag: Lexer::Token?,
+        first_column: Integer, last_column: Integer,
+        referring_symbol: untyped, position_in_rhs: Integer?
+      ) -> void
+
+      def tag: () -> untyped
+    end
+  end
+end

data/sig/lrama/grammar.rbs

@@ -0,0 +1,5 @@
+module Lrama
+  class Grammar
+    def numberize_references: (Lexer::Token lhs, Array[Lexer::Token] rhs, Array[Reference]) -> void
+  end
+end

data/sig/lrama/lexer/token/char.rbs

@@ -0,0 +1,8 @@
+module Lrama
+  class Lexer
+    class Token
+      class Char < Token
+      end
+    end
+  end
+end

data/sig/lrama/lexer/token/ident.rbs

@@ -0,0 +1,8 @@
+module Lrama
+  class Lexer
+    class Token
+      class Ident < Token
+      end
+    end
+  end
+end

data/sig/lrama/lexer/token/parameterizing.rbs

@@ -0,0 +1,8 @@
+module Lrama
+  class Lexer
+    class Token
+      class Parameterizing < Token
+      end
+    end
+  end
+end

data/sig/lrama/lexer/token/tag.rbs

@@ -0,0 +1,8 @@
+module Lrama
+  class Lexer
+    class Token
+      class Tag < Token
+      end
+    end
+  end
+end

data/sig/lrama/lexer/token/user_code.rbs

@@ -0,0 +1,9 @@
+module Lrama
+  class Lexer
+    class Token
+      class UserCode < Token
+        attr_accessor references: Array[[Symbol, (String|Integer), Token?, Integer, Integer]]
+      end
+    end
+  end
+end

data/sig/lrama/lexer/token.rbs

@@ -0,0 +1,17 @@
+module Lrama
+  class Lexer
+    class Token
+      attr_accessor s_value: String
+      attr_accessor alias_name: String
+      attr_accessor line: Integer
+      attr_accessor column: Integer
+      attr_accessor referred: bool
+
+      def initialize: (?s_value: String, ?alias_name: String) -> void
+
+      def to_s: () -> String
+      def referred_by?: (String string) -> bool
+      def ==: (Token other) -> bool
+    end
+  end
+end

data/template/bison/_yacc.h

@@ -12,7 +12,7 @@
 #if YYDEBUG && !defined(yydebug)
 extern int yydebug;
 #endif
-  <%-# b4_percent_code_get([[requires]]). %code is not supported -%>
+<%= output.percent_code("requires") %>
 
   <%-# b4_token_enums_defines -%>
 /* Token kinds.  */
@@ -64,7 +64,7 @@ struct YYLTYPE
 int yyparse (<%= output.parse_param %>);
 
 
-  <%-# b4_percent_code_get([[provides]]). %code is not supported -%>
+<%= output.percent_code("provides") %>
   <%-# b4_cpp_guard_close([b4_spec_mapped_header_file]) -%>
     <%- if output.spec_mapped_header_file -%>
 #endif /* !<%= output.b4_cpp_guard__b4_spec_mapped_header_file %>  */

data/template/bison/yacc.c

@@ -68,8 +68,6 @@
 #define YYPULL 1
 
 
-
-
 <%# b4_user_pre_prologue -%>
 /* First part of user prologue.  */
 #line <%= output.aux.prologue_first_lineno %> "<%= output.grammar_file_path %>"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: lrama
 version: !ruby/object:Gem::Version
-  version: 0.5.7
+  version: 0.5.9
 platform: ruby
 authors:
 - Yuichiro Kaneko
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-10-23 00:00:00.000000000 Z
+date: 2023-11-05 00:00:00.000000000 Z
 dependencies: []
 description: LALR (1) parser generator written by Ruby
 email:
@@ -47,6 +47,7 @@ files:
 - lib/lrama/grammar/auxiliary.rb
 - lib/lrama/grammar/code.rb
 - lib/lrama/grammar/error_token.rb
+- lib/lrama/grammar/percent_code.rb
 - lib/lrama/grammar/precedence.rb
 - lib/lrama/grammar/printer.rb
 - lib/lrama/grammar/reference.rb
@@ -55,7 +56,11 @@ files:
 - lib/lrama/grammar/union.rb
 - lib/lrama/lexer.rb
 - lib/lrama/lexer/token.rb
-- lib/lrama/lexer/token/type.rb
+- lib/lrama/lexer/token/char.rb
+- lib/lrama/lexer/token/ident.rb
+- lib/lrama/lexer/token/parameterizing.rb
+- lib/lrama/lexer/token/tag.rb
+- lib/lrama/lexer/token/user_code.rb
 - lib/lrama/option_parser.rb
 - lib/lrama/options.rb
 - lib/lrama/output.rb
@@ -84,7 +89,15 @@ files:
 - sample/parse.y
 - sig/lrama/bitmap.rbs
 - sig/lrama/digraph.rbs
-- sig/lrama/lexer/token/type.rbs
+- sig/lrama/grammar.rbs
+- sig/lrama/grammar/percent_code.rbs
+- sig/lrama/grammar/reference.rbs
+- sig/lrama/lexer/token.rbs
+- sig/lrama/lexer/token/char.rbs
+- sig/lrama/lexer/token/ident.rbs
+- sig/lrama/lexer/token/parameterizing.rbs
+- sig/lrama/lexer/token/tag.rbs
+- sig/lrama/lexer/token/user_code.rbs
 - sig/lrama/report/duration.rbs
 - sig/lrama/report/profile.rbs
 - sig/lrama/warning.rbs

data/lib/lrama/lexer/token/type.rb

@@ -1,8 +0,0 @@
-module Lrama
-  class Lexer
-    class Token < Struct.new(:type, :s_value, :alias, keyword_init: true)
-      class Type < Struct.new(:id, :name, keyword_init: true)
-      end
-    end
-  end
-end

data/sig/lrama/lexer/token/type.rbs

@@ -1,17 +0,0 @@
-module Lrama
-  class Lexer
-    class Token
-      attr_accessor type: Type
-      attr_accessor s_value: String
-      attr_accessor alias: String
-
-      def initialize: (?type: Type, ?s_value: String, ?alias: String) -> void
-      class Type
-        attr_accessor id: Integer
-        attr_accessor name: String
-
-        def initialize: (?id: Integer, ?name: String) -> void
-      end
-    end
-  end
-end