diff options
Diffstat (limited to 'tool/lrama/lib/lrama/lexer/token')
-rw-r--r-- | tool/lrama/lib/lrama/lexer/token/char.rb | 10 | ||||
-rw-r--r-- | tool/lrama/lib/lrama/lexer/token/ident.rb | 10 | ||||
-rw-r--r-- | tool/lrama/lib/lrama/lexer/token/instantiate_rule.rb | 25 | ||||
-rw-r--r-- | tool/lrama/lib/lrama/lexer/token/tag.rb | 14 | ||||
-rw-r--r-- | tool/lrama/lib/lrama/lexer/token/user_code.rb | 79 |
5 files changed, 138 insertions, 0 deletions
diff --git a/tool/lrama/lib/lrama/lexer/token/char.rb b/tool/lrama/lib/lrama/lexer/token/char.rb new file mode 100644 index 0000000000..9e21952c42 --- /dev/null +++ b/tool/lrama/lib/lrama/lexer/token/char.rb @@ -0,0 +1,10 @@ +# frozen_string_literal: true + +module Lrama + class Lexer + class Token + class Char < Token + end + end + end +end diff --git a/tool/lrama/lib/lrama/lexer/token/ident.rb b/tool/lrama/lib/lrama/lexer/token/ident.rb new file mode 100644 index 0000000000..84835c00bc --- /dev/null +++ b/tool/lrama/lib/lrama/lexer/token/ident.rb @@ -0,0 +1,10 @@ +# frozen_string_literal: true + +module Lrama + class Lexer + class Token + class Ident < Token + end + end + end +end diff --git a/tool/lrama/lib/lrama/lexer/token/instantiate_rule.rb b/tool/lrama/lib/lrama/lexer/token/instantiate_rule.rb new file mode 100644 index 0000000000..db7e611c5f --- /dev/null +++ b/tool/lrama/lib/lrama/lexer/token/instantiate_rule.rb @@ -0,0 +1,25 @@ +# frozen_string_literal: true + +module Lrama + class Lexer + class Token + class InstantiateRule < Token + attr_reader :args, :lhs_tag + + def initialize(s_value:, alias_name: nil, location: nil, args: [], lhs_tag: nil) + super s_value: s_value, alias_name: alias_name, location: location + @args = args + @lhs_tag = lhs_tag + end + + def rule_name + s_value + end + + def args_count + args.count + end + end + end + end +end diff --git a/tool/lrama/lib/lrama/lexer/token/tag.rb b/tool/lrama/lib/lrama/lexer/token/tag.rb new file mode 100644 index 0000000000..52dcb50ce7 --- /dev/null +++ b/tool/lrama/lib/lrama/lexer/token/tag.rb @@ -0,0 +1,14 @@ +# frozen_string_literal: true + +module Lrama + class Lexer + class Token + class Tag < Token + # Omit "<>" + def member + s_value[1..-2] or raise "Unexpected Tag format (#{s_value})" + end + end + end + end +end diff --git a/tool/lrama/lib/lrama/lexer/token/user_code.rb b/tool/lrama/lib/lrama/lexer/token/user_code.rb new file mode 100644 index 0000000000..9712208642 --- /dev/null +++ b/tool/lrama/lib/lrama/lexer/token/user_code.rb @@ -0,0 +1,79 @@ +# frozen_string_literal: true + +require "strscan" + +module Lrama + class Lexer + class Token + class UserCode < Token + attr_accessor :tag + + def references + @references ||= _references + end + + private + + def _references + scanner = StringScanner.new(s_value) + references = [] + + until scanner.eos? do + case + when reference = scan_reference(scanner) + references << reference + when scanner.scan(/\/\*/) + scanner.scan_until(/\*\//) + else + scanner.getch + end + end + + references + end + + def scan_reference(scanner) + start = scanner.pos + case + # $ references + # It need to wrap an identifier with brackets to use ".-" for identifiers + when scanner.scan(/\$(<[a-zA-Z0-9_]+>)?\$/) # $$, $<long>$ + tag = scanner[1] ? Lrama::Lexer::Token::Tag.new(s_value: scanner[1]) : nil + return Lrama::Grammar::Reference.new(type: :dollar, name: "$", ex_tag: tag, first_column: start, last_column: scanner.pos) + when scanner.scan(/\$(<[a-zA-Z0-9_]+>)?(\d+)/) # $1, $2, $<long>1 + tag = scanner[1] ? Lrama::Lexer::Token::Tag.new(s_value: scanner[1]) : nil + return Lrama::Grammar::Reference.new(type: :dollar, number: Integer(scanner[2]), index: Integer(scanner[2]), ex_tag: tag, first_column: start, last_column: scanner.pos) + when scanner.scan(/\$(<[a-zA-Z0-9_]+>)?([a-zA-Z_][a-zA-Z0-9_]*)/) # $foo, $expr, $<long>program (named reference without brackets) + tag = scanner[1] ? Lrama::Lexer::Token::Tag.new(s_value: scanner[1]) : nil + return Lrama::Grammar::Reference.new(type: :dollar, name: scanner[2], ex_tag: tag, first_column: start, last_column: scanner.pos) + when scanner.scan(/\$(<[a-zA-Z0-9_]+>)?\[([a-zA-Z_.][-a-zA-Z0-9_.]*)\]/) # $[expr.right], $[expr-right], $<long>[expr.right] (named reference with brackets) + tag = scanner[1] ? Lrama::Lexer::Token::Tag.new(s_value: scanner[1]) : nil + return Lrama::Grammar::Reference.new(type: :dollar, name: scanner[2], ex_tag: tag, first_column: start, last_column: scanner.pos) + + # @ references + # It need to wrap an identifier with brackets to use ".-" for identifiers + when scanner.scan(/@\$/) # @$ + return Lrama::Grammar::Reference.new(type: :at, name: "$", first_column: start, last_column: scanner.pos) + when scanner.scan(/@(\d+)/) # @1 + return Lrama::Grammar::Reference.new(type: :at, number: Integer(scanner[1]), index: Integer(scanner[1]), first_column: start, last_column: scanner.pos) + when scanner.scan(/@([a-zA-Z][a-zA-Z0-9_]*)/) # @foo, @expr (named reference without brackets) + return Lrama::Grammar::Reference.new(type: :at, name: scanner[1], first_column: start, last_column: scanner.pos) + when scanner.scan(/@\[([a-zA-Z_.][-a-zA-Z0-9_.]*)\]/) # @[expr.right], @[expr-right] (named reference with brackets) + return Lrama::Grammar::Reference.new(type: :at, name: scanner[1], first_column: start, last_column: scanner.pos) + + # $: references + when scanner.scan(/\$:\$/) # $:$ + return Lrama::Grammar::Reference.new(type: :index, name: "$", first_column: start, last_column: scanner.pos) + when scanner.scan(/\$:(\d+)/) # $:1 + return Lrama::Grammar::Reference.new(type: :index, number: Integer(scanner[1]), first_column: start, last_column: scanner.pos) + when scanner.scan(/\$:([a-zA-Z_][a-zA-Z0-9_]*)/) # $:foo, $:expr (named reference without brackets) + return Lrama::Grammar::Reference.new(type: :index, name: scanner[1], first_column: start, last_column: scanner.pos) + when scanner.scan(/\$:\[([a-zA-Z_.][-a-zA-Z0-9_.]*)\]/) # $:[expr.right], $:[expr-right] (named reference with brackets) + return Lrama::Grammar::Reference.new(type: :index, name: scanner[1], first_column: start, last_column: scanner.pos) + + end + end + end + end + end +end |