rexical-1.0.7/0000755000004100000410000000000013523645102013177 5ustar www-datawww-datarexical-1.0.7/test/0000755000004100000410000000000013523645102014156 5ustar www-datawww-datarexical-1.0.7/test/rex-20060125.rb0000644000004100000410000000744313523645102016206 0ustar www-datawww-data#!C:/Program Files/ruby-1.8/bin/ruby # # rex # # Copyright (c) 2005 ARIMA Yasuhiro # # This program is free software. # You can distribute/modify this program under the terms of # the GNU LGPL, Lesser General Public License version 2.1. # For details of LGPL, see the file "COPYING". # ## --------------------------------------------------------------------- REX_OPTIONS = <<-EOT o -o --output-file file name of output [.rb] o -s --stub - append stub code for debug o -i --ignorecase - ignore char case o -C --check-only - syntax check only o - --independent - independent mode o -d --debug - print debug information o -h --help - print this message and quit o - --version - print version and quit o - --copyright - print copyright and quit EOT ## --------------------------------------------------------------------- require 'getoptlong' require 'rex/generator' require 'rex/info' ## --------------------------------------------------------------------- =begin class Rex def initialize end end =end def main $cmd = File.basename($0, ".rb") opt = get_options filename = ARGV[0] rex = Rex::Generator.new(opt) begin rex.grammar_file = filename rex.read_grammar rex.parse if opt['--check-only'] $stderr.puts "syntax ok" return 0 end rex.write_scanner rescue Rex::ParseError, Errno::ENOENT msg = $!.to_s unless /\A\d/ === msg msg[0,0] = ' ' end $stderr.puts "#{$cmd}:#{rex.grammar_file}:#{rex.lineno}:#{msg}" return 1 end return 0 end ## --------------------------------------------------------------------- def get_options tmp = REX_OPTIONS.collect do |line| next if /\A\s*\z/ === line disp, sopt, lopt, takearg, doc = line.strip.split(/\s+/, 5) a = [] a.push lopt unless lopt == '-' a.push sopt unless sopt == '-' a.push takearg == '-' ? GetoptLong::NO_ARGUMENT : GetoptLong::REQUIRED_ARGUMENT a end getopt = GetoptLong.new(*tmp.compact) getopt.quiet = true opt = {} begin getopt.each do |name, arg| raise GetoptLong::InvalidOption, "#{$cmd}: #{name} given twice" if opt.key? name opt[name] = arg.empty? ? true : arg end rescue GetoptLong::AmbigousOption, GetoptLong::InvalidOption, GetoptLong::MissingArgument, GetoptLong::NeedlessArgument usage 1, $!.message end usage if opt['--help'] if opt['--version'] puts "#{$cmd} version #{Rex::VERSION}" exit 0 end if opt['--copyright'] puts "#{$cmd} version #{Rex::VERSION}" puts "#{Rex::Copyright} <#{Rex::Mailto}>" exit 0 end usage(1, 'no grammar file given') if ARGV.empty? usage(1, 'too many grammar files given') if ARGV.size > 1 opt end ## --------------------------------------------------------------------- def usage(status=0, msg=nil ) f = (status == 0 ? $stdout : $stderr) f.puts "#{$cmd}: #{msg}" if msg f.print <<-EOT Usage: #{$cmd} [options] Options: EOT REX_OPTIONS.each do |line| next if line.strip.empty? if /\A\s*\z/ === line f.puts next end disp, sopt, lopt, takearg, doc = line.strip.split(/\s+/, 5) if disp == 'o' sopt = nil if sopt == '-' lopt = nil if lopt == '-' opt = [sopt, lopt].compact.join(',') takearg = nil if takearg == '-' opt = [opt, takearg].compact.join(' ') f.printf "%-27s %s\n", opt, doc end end exit status end ## --------------------------------------------------------------------- main rexical-1.0.7/test/test_generator.rb0000644000004100000410000001424013523645102017531 0ustar www-datawww-datagem "minitest" require 'minitest/autorun' require 'tempfile' require 'rexical' require 'stringio' require 'open3' class TestGenerator < Minitest::Test def test_header_is_written_after_module rex = Rexical::Generator.new( "--independent" => true ) rex.grammar_file = File.join File.dirname(__FILE__), 'assets', 'test.rex' rex.read_grammar rex.parse output = StringIO.new rex.write_scanner output comments = [] output.string.split(/[\n]/).each do |line| comments << line.chomp if line =~ /^#/ end assert_match 'DO NOT MODIFY', comments.join assert_equal '#--', comments.first assert_equal '#++', comments.last end def test_rubocop_security rex = Rexical::Generator.new( "--independent" => true ) rex.grammar_file = File.join File.dirname(__FILE__), 'assets', 'test.rex' rex.read_grammar rex.parse output = Tempfile.new(["rex_output", ".rb"]) begin rex.write_scanner output output.close stdin, stdoe, wait_thr = Open3.popen2e "rubocop --only Security #{output.path}" if ! wait_thr.value.success? fail stdoe.read end ensure output.close output.unlink end end def test_read_non_existent_file rex = Rexical::Generator.new(nil) rex.grammar_file = 'non_existent_file' assert_raises Errno::ENOENT do rex.read_grammar end end def test_scanner_nests_classes source = parse_lexer %q{ module Foo class Baz::Calculator < Bar rule \d+ { [:NUMBER, text.to_i] } \s+ { [:S, text] } end end } assert_match 'Baz::Calculator < Bar', source end def test_scanner_inherits source = parse_lexer %q{ class Calculator < Bar rule \d+ { [:NUMBER, text.to_i] } \s+ { [:S, text] } end } assert_match 'Calculator < Bar', source end def test_scanner_inherits_many_levels source = parse_lexer %q{ class Calculator < Foo::Bar rule \d+ { [:NUMBER, text.to_i] } \s+ { [:S, text] } end } assert_match 'Calculator < Foo::Bar', source end def test_stateful_lexer m = build_lexer %q{ class Foo rule \d { @state = :digit; [:foo, text] } :digit \w { @state = nil; [:w, text] } end } scanner = m::Foo.new scanner.scan_setup('1w1') assert_tokens [ [:foo, '1'], [:w, 'w'], [:foo, '1']], scanner end def test_simple_scanner m = build_lexer %q{ class Calculator rule \d+ { [:NUMBER, text.to_i] } \s+ { [:S, text] } end } calc = m::Calculator.new calc.scan_setup('1 2 10') assert_tokens [[:NUMBER, 1], [:S, ' '], [:NUMBER, 2], [:S, ' '], [:NUMBER, 10]], calc end def test_simple_scanner_with_empty_action m = build_lexer %q{ class Calculator rule \d+ { [:NUMBER, text.to_i] } \s+ # skips whitespaces end } calc = m::Calculator.new calc.scan_setup('1 2 10') assert_tokens [[:NUMBER, 1], [:NUMBER, 2], [:NUMBER, 10]], calc end def test_parses_macros_with_escapes source = parse_lexer %q{ class Foo macro w [\ \t]+ rule {w} { [:SPACE, text] } end } assert source.index('@ss.scan(/[ \t]+/))') end def test_simple_scanner_with_macros m = build_lexer %q{ class Calculator macro digit \d+ rule {digit} { [:NUMBER, text.to_i] } \s+ { [:S, text] } end } calc = m::Calculator.new calc.scan_setup('1 2 10') assert_tokens [[:NUMBER, 1], [:S, ' '], [:NUMBER, 2], [:S, ' '], [:NUMBER, 10]], calc end def test_nested_macros source = parse_lexer %q{ class Calculator macro nonascii [^\0-\177] string "{nonascii}*" rule {string} { [:STRING, text] } end } assert_match '"[^\0-\177]*"', source end def test_more_nested_macros source = parse_lexer %q{ class Calculator macro nonascii [^\0-\177] sing {nonascii}* string "{sing}" rule {string} { [:STRING, text] } end } assert_match '"[^\0-\177]*"', source end def test_changing_state_during_lexing lexer = build_lexer %q{ class Calculator rule a { self.state = :B ; [:A, text] } :B b { self.state = nil ; [:B, text] } end } calc1 = lexer::Calculator.new calc2 = lexer::Calculator.new calc1.scan_setup('aaaaa') calc2.scan_setup('ababa') # Doesn't lex all 'a's assert_raises(lexer::Calculator::ScanError) { tokens(calc1) } # Does lex alternating 'a's and 'b's calc2.scan_setup('ababa') assert_tokens [[:A, 'a'], [:B, 'b'], [:A, 'a'], [:B, 'b'], [:A, 'a']], calc2 end def test_changing_state_is_possible_between_next_token_calls lexer = build_lexer %q{ class Calculator rule a { [:A, text] } :B b { [:B, text] } end } calc = lexer::Calculator.new calc.scan_setup('ababa') assert_equal [:A, 'a'], calc.next_token calc.state = :B assert_equal [:B, 'b'], calc.next_token calc.state = nil assert_equal [:A, 'a'], calc.next_token calc.state = :B assert_equal [:B, 'b'], calc.next_token calc.state = nil assert_equal [:A, 'a'], calc.next_token end def test_match_eos lexer = build_lexer %q{ class Calculator option matcheos rule a { [:A, text] } $ { [:EOF, ""] } :B b { [:B, text] } } calc = lexer::Calculator.new calc.scan_setup("a") assert_equal [:A, 'a'], calc.next_token assert_equal [:EOF, ""], calc.next_token end def parse_lexer(str) rex = Rexical::Generator.new("--independent" => true) out = StringIO.new rex.grammar_lines = StringScanner.new(str) rex.parse rex.write_scanner(out) out.string end def build_lexer(str) mod = Module.new mod.module_eval(parse_lexer(str)) mod end def tokens(scanner) tokens = [] while token = scanner.next_token tokens << token end tokens end def assert_tokens(expected, scanner) assert_equal expected, tokens(scanner) end end rexical-1.0.7/test/rex-20060511.rb0000644000004100000410000000727413523645102016207 0ustar www-datawww-data#!/usr/local/bin/ruby # # rex # # Copyright (c) 2005-2006 ARIMA Yasuhiro # # This program is free software. # You can distribute/modify this program under the terms of # the GNU LGPL, Lesser General Public License version 2.1. # For details of LGPL, see the file "COPYING". # ## --------------------------------------------------------------------- REX_OPTIONS = <<-EOT o -o --output-file file name of output [.rb] o -s --stub - append stub code for debug o -i --ignorecase - ignore char case o -C --check-only - syntax check only o - --independent - independent mode o -d --debug - print debug information o -h --help - print this message and quit o - --version - print version and quit o - --copyright - print copyright and quit EOT ## --------------------------------------------------------------------- require 'getoptlong' require 'rex/generator' require 'rex/info' ## --------------------------------------------------------------------- class RexRunner def run @status = 1 usage 'no grammar file given' if ARGV.empty? usage 'too many grammar files given' if ARGV.size > 1 filename = ARGV[0] rex = Rex::Generator.new(@opt) begin rex.grammar_file = filename rex.read_grammar rex.parse if @opt['--check-only'] $stderr.puts "syntax ok" return 0 end rex.write_scanner @status = 0 rescue Rex::ParseError, Errno::ENOENT msg = $!.to_s unless /\A\d/ === msg msg[0,0] = ' ' end $stderr.puts "#{@cmd}:#{rex.grammar_file}:#{rex.lineno}:#{msg}" ensure exit @status end end def initialize @status = 2 @cmd = File.basename($0, ".rb") tmp = REX_OPTIONS.collect do |line| next if /\A\s*\z/ === line disp, sopt, lopt, takearg, doc = line.strip.split(/\s+/, 5) a = [] a.push lopt unless lopt == '-' a.push sopt unless sopt == '-' a.push takearg == '-' ? GetoptLong::NO_ARGUMENT : GetoptLong::REQUIRED_ARGUMENT a end getopt = GetoptLong.new(*tmp.compact) getopt.quiet = true @opt = {} begin getopt.each do |name, arg| raise GetoptLong::InvalidOption, "#{@cmd}: #{name} given twice" if @opt.key? name @opt[name] = arg.empty? ? true : arg end rescue GetoptLong::AmbigousOption, GetoptLong::InvalidOption, GetoptLong::MissingArgument, GetoptLong::NeedlessArgument usage $!.message end usage if @opt['--help'] if @opt['--version'] puts "#{@cmd} version #{Rex::VERSION}" exit 0 end if @opt['--copyright'] puts "#{@cmd} version #{Rex::VERSION}" puts "#{Rex::Copyright} <#{Rex::Mailto}>" exit 0 end end def usage( msg=nil ) f = $stderr f.puts "#{@cmd}: #{msg}" if msg f.print <<-EOT Usage: #{@cmd} [options] Options: EOT REX_OPTIONS.each do |line| next if line.strip.empty? if /\A\s*\z/ === line f.puts next end disp, sopt, lopt, takearg, doc = line.strip.split(/\s+/, 5) if disp == 'o' sopt = nil if sopt == '-' lopt = nil if lopt == '-' opt = [sopt, lopt].compact.join(',') takearg = nil if takearg == '-' opt = [opt, takearg].compact.join(' ') f.printf "%-27s %s\n", opt, doc end end exit @status end end RexRunner.new.run rexical-1.0.7/test/assets/0000755000004100000410000000000013523645102015460 5ustar www-datawww-datarexical-1.0.7/test/assets/test.rex0000644000004100000410000000026013523645102017155 0ustar www-datawww-datamodule A module B class C < SomethingElse macro w [\s\r\n\f]* rule # [:state] pattern [actions] {w}~={w} { [:INCLUDES, text] } end end end rexical-1.0.7/bin/0000755000004100000410000000000013523645102013747 5ustar www-datawww-datarexical-1.0.7/bin/rex0000644000004100000410000000067513523645102014500 0ustar www-datawww-data#!/usr/bin/env ruby # # rex # # Copyright (c) 2005-2006 ARIMA Yasuhiro # # This program is free software. # You can distribute/modify this program under the terms of # the GNU LGPL, Lesser General Public License version 2.1. # For details of LGPL, see the file "COPYING". # ## --------------------------------------------------------------------- require 'rubygems' require 'rexical' Rexical::Cmd.new.run rexical-1.0.7/CHANGELOG.rdoc0000644000004100000410000000070313523645102015337 0ustar www-datawww-data=== 1.0.7 / 2019-08-06 * Security * prefer File.open to Kernel.open === 1.0.6 * Bug fixes * scanner states work better. Thanks Mat. === 1.0.5 * Bug fixes * Scanners with nested classes work better === 1.0.4 * Bug fixes * Generated tokenizer only tokenizes on pulls === 1.0.3 * Bug fixes * renamed to "Rexical" because someone already has "rex". === 1.0.2 * Bug fixes * Fixed nested macros so that backslashes will work rexical-1.0.7/DOCUMENTATION.en.rdoc0000644000004100000410000001317413523645102016470 0ustar www-datawww-data= REX: Ruby Lex for Racc == About Lexical Scanner Generator used with Racc for Ruby == Usage rex [options] grammarfile -o --output-file filename designate output filename -s --stub append stub main for debugging -i --ignorecase ignore character case -C --check-only only check syntax --independent independent mode -d --debug print debug information -h --help print usage --version print version --copyright print copyright == Default Output Filename The destination file for foo.rex is foo.rex.rb. To use, include the following in the Ruby source code file. require 'foo.rex' == Grammar File Format A definition consists of a header section, a rule section, and a footer section. The rule section includes one or more clauses. Each clause starts with a keyword. Summary: [Header section] "class" Foo ["option" [options] ] ["inner" [methods] ] ["macro" [macro-name regular-expression] ] "rule" [start-state] pattern [actions] "end" [Footer section] === Grammar File Description Example class Foo macro BLANK \s+ DIGIT \d+ rule {BLANK} {DIGIT} { [:NUMBER, text.to_i] } . { [text, text] } end == Header Section ( Optional ) All of the contents described before the definitions in the rule section are copied to the beginning of the output file. == Footer Section ( Optional ) All the contents described after the definitions in the rule section are copied to the end of the output file. == Rule Section The rule section starts at the line beginning with the "class" keyword and ends at the line beginning with the "end" keyword. The class name is specified after the "class" keyword. If a module name is specified, the class will be included in the module. A class that inherits Racc::Parser is generated. === Example of Rule Section Definition class Foo class Bar::Foo == Option Section ( Optional ) This section begins with the "option" keyword. "ignorecase" ignore the character case when pattern matching "stub" append stub main for debugging "independent" independent mode, do not inherit Racc. == Inner Section for User Code ( Optional ) This section begins with the "inner" keyword. The contents defined here are defined by the contents of the class of the generated scanner. == Macro Section ( Optional ) This section begins with the "macro" keyword. A name is assigned to one regular expression. A space character (0x20) can be included by using a backslash \ to escape. === Example of Macro Definition DIGIT \d+ IDENT [a-zA-Z_][a-zA-Z0-9_]* BLANK [\ \t]+ REMIN \/\* REMOUT \*\/ == Rule Section This section begins with the "rule" keyword. [state] pattern [actions] === state: Start State ( Optional ) A start state is indicated by an identifier beginning with ":", a Ruby symbol. If uppercase letters follow the ":", the state becomes an exclusive start state. If lowercase letters follow the ":", the state becomes an inclusive start state. The initial value and the default value of a start state are nil. === pattern: String Pattern A regular expression specifies a character string. A regular expression description may include a macro definition enclosed by curly braces { }. A macro definition is used when the regular expression includes whitespace. === actions: Processing Actions ( Optional ) An action is executed when the pattern is matched. The action defines the process for creating the appropriate token. A token is a two-element array containing a type and a value, or is nil. The following elements can be used to create a token. lineno Line number ( Read Only ) text Matched string ( Read Only ) state Start state ( Read/Write ) The action is a block of Ruby code enclosed by { }. Do not use functions that exit the block and change the control flow. ( return, exit, next, break, ... ) If the action is omitted, the matched character string is discarded, and the process advances to the next scan. === Example of Rule Section Definition {REMIN} { state = :REM ; [:REM_IN, text] } :REM {REMOUT} { state = nil ; [:REM_OUT, text] } :REM (.+)(?={REMOUT}) { [:COMMENT, text] } {BLANK} -?{DIGIT} { [:NUMBER, text.to_i] } {WORD} { [:word, text] } . { [text, text] } == Comments ( Optional ) Any text following a "#" to the end of the line becomes a comment. == Using the Generated Class === scan_setup( str ) Initializes the scanner with the str string argument. This is redefined and used. === scan_str( str ) Parses the string described in the defined grammar. The tokens are stored internally. === scan_file( filename ) Reads in a file described in the defined grammar. The tokens are stored internally. === next_token The tokens stored internally are extracted one by one. When there are no more tokens, nil is returned. == Notice This specification is provisional and may be changed without prior notice. rexical-1.0.7/Rakefile0000644000004100000410000000121313523645102014641 0ustar www-datawww-data# -*- ruby -*- require 'rubygems' require 'hoe' Hoe.plugin :debugging Hoe.plugin :git Hoe.plugins.delete :rubyforge Hoe.spec 'rexical' do self.readme_file = 'README.rdoc' self.history_file = 'CHANGELOG.rdoc' developer('Aaron Patterson', 'aaronp@rubyforge.org') self.extra_rdoc_files = FileList['*.rdoc'] self.extra_dev_deps += [ ["rubocop", "~> 0.74.0"] ] end namespace :gem do namespace :spec do task :dev do File.open("#{HOE.name}.gemspec", 'w') do |f| HOE.spec.version = "#{HOE.version}.#{Time.now.strftime("%Y%m%d%H%M%S")}" f.write(HOE.spec.to_ruby) end end end end # vim: syntax=Ruby rexical-1.0.7/lib/0000755000004100000410000000000013523645102013745 5ustar www-datawww-datarexical-1.0.7/lib/rexical/0000755000004100000410000000000013523645102015374 5ustar www-datawww-datarexical-1.0.7/lib/rexical/rexcmd.rb0000644000004100000410000000672013523645102017210 0ustar www-datawww-data# # rexcmd.rb # # Copyright (c) 2005-2006 ARIMA Yasuhiro # # This program is free software. # You can distribute/modify this program under the terms of # the GNU LGPL, Lesser General Public License version 2.1. # For details of LGPL, see the file "COPYING". # ## --------------------------------------------------------------------- require 'getoptlong' module Rexical class Cmd OPTIONS = <<-EOT o -o --output-file file name of output [.rb] o -s --stub - append stub code for debug o -i --ignorecase - ignore char case o -C --check-only - syntax check only o - --independent - independent mode o - --matcheos - allow match against end of string o -d --debug - print debug information o -h --help - print this message and quit o - --version - print version and quit o - --copyright - print copyright and quit EOT def run @status = 1 usage 'no grammar file given' if ARGV.empty? usage 'too many grammar files given' if ARGV.size > 1 filename = ARGV[0] rex = Rexical::Generator.new(@opt) begin rex.grammar_file = filename rex.read_grammar rex.parse if @opt['--check-only'] $stderr.puts "syntax ok" return 0 end rex.write_scanner @status = 0 rescue Rexical::ParseError, Errno::ENOENT msg = $!.to_s unless /\A\d/ === msg msg[0,0] = ' ' end $stderr.puts "#{@cmd}:#{rex.grammar_file}:#{rex.lineno}:#{msg}" ensure exit @status end end def initialize @status = 2 @cmd = File.basename($0, ".rb") tmp = OPTIONS.lines.collect do |line| next if /\A\s*\z/ === line # disp, sopt, lopt, takearg, doc _, sopt, lopt, takearg, _ = line.strip.split(/\s+/, 5) a = [] a.push lopt unless lopt == '-' a.push sopt unless sopt == '-' a.push takearg == '-' ? GetoptLong::NO_ARGUMENT : GetoptLong::REQUIRED_ARGUMENT a end getopt = GetoptLong.new(*tmp.compact) getopt.quiet = true @opt = {} begin getopt.each do |name, arg| raise GetoptLong::InvalidOption, "#{@cmd}: #{name} given twice" if @opt.key? name @opt[name] = arg.empty? ? true : arg end rescue GetoptLong::AmbigousOption, GetoptLong::InvalidOption, GetoptLong::MissingArgument, GetoptLong::NeedlessArgument usage $!.message end usage if @opt['--help'] if @opt['--version'] puts "#{@cmd} version #{Rexical::VERSION}" exit 0 end if @opt['--copyright'] puts "#{@cmd} version #{Rexical::VERSION}" puts "#{Rexical::Copyright} <#{Rexical::Mailto}>" exit 0 end end def usage( msg=nil ) f = $stderr f.puts "#{@cmd}: #{msg}" if msg f.print <<-EOT Usage: #{@cmd} [options] Options: EOT OPTIONS.each_line do |line| next if line.strip.empty? if /\A\s*\z/ === line f.puts next end disp, sopt, lopt, takearg, doc = line.strip.split(/\s+/, 5) if disp == 'o' sopt = nil if sopt == '-' lopt = nil if lopt == '-' opt = [sopt, lopt].compact.join(',') takearg = nil if takearg == '-' opt = [opt, takearg].compact.join(' ') f.printf "%-27s %s\n", opt, doc end end exit @status end end end rexical-1.0.7/lib/rexical/generator.rb0000644000004100000410000002647213523645102017722 0ustar www-datawww-data# # generator.rb # # Copyright (c) 2005-2006 ARIMA Yasuhiro # # This program is free software. # You can distribute/modify this program under the terms of # the GNU Lesser General Public License version 2 or later. # require 'strscan' module Rexical class ParseError < StandardError ; end class Generator attr_accessor :grammar_file attr_accessor :grammar_lines attr_accessor :scanner_file attr_accessor :module_name attr_accessor :class_name attr_accessor :lineno attr_accessor :rules attr_accessor :exclusive_states attr_accessor :ignorecase attr_accessor :independent attr_accessor :debug def initialize(opts) @lineno = 0 @macro = {} @rules = [] @exclusive_states = [nil] @grammar_lines = nil @scanner_header = "" @scanner_footer = "" @scanner_inner = "" @opt = opts end def add_header( st ) @scanner_header += "#{st}\n" end def add_footer( st ) @scanner_footer += "#{st}\n" end def add_inner( st ) @scanner_inner += "#{st}\n" end def add_option( st ) opts = st.split opts.each do |opt| case opt when /ignorecase/i @opt['--ignorecase'] = true when /stub/i @opt['--stub'] = true when /independent/i @opt['--independent'] = true when /matcheos/i @opt['--matcheos'] = true end end end def add_macro( st ) ss = StringScanner.new(st) ss.scan(/\s+/) key = ss.scan(/\S+/) ss.scan(/\s+/) st = ss.post_match len = st.size ndx = 0 while ndx <= len c = st[ndx,1] ndx += 1 case c when '\\' ndx += 1 next when '#', ' ' ndx -= 1 break end end expr = st[0,ndx] expr.gsub!('\ ', ' ') key = '{' + key + '}' @macro.each_pair do |k, e| expr.gsub!(k) { |m| e } end @macro[key] = expr rescue raise ParseError, "parse error in add_macro:'#{st}'" end def add_rule( rule_state, rule_expr, rule_action=nil ) st = rule_expr.dup @macro.each_pair do |k, e| rule_expr.gsub!(k) { |m| e } end if rule_state.to_s[1,1] =~ /[A-Z]/ @exclusive_states << rule_state unless @exclusive_states.include?(rule_state) exclusive_state = rule_state start_state = nil else exclusive_state = nil start_state = rule_state end rule = [exclusive_state, start_state, rule_expr, rule_action] @rules << rule rescue raise ParseError, "parse error in add_rule:'#{st}'" end def read_grammar @grammar_lines = StringScanner.new File.read(grammar_file) end def next_line @lineno += 1 @grammar_lines.scan_until(/\n/).chomp rescue nil end def parse state1 = :HEAD state2 = nil state3 = nil lastmodes = [] while st = next_line case state1 when :FOOT add_footer st when :HEAD ss = StringScanner.new(st) if ss.scan(/class/) state1 = :CLASS st = ss.post_match.strip @class_name = st else add_header st end when :CLASS s = st.strip next if s.size == 0 or s[0,1] == '#' ss = StringScanner.new(st) if ss.scan(/option.*$/) state2 = :OPTION next end if ss.scan(/inner.*$/) state2 = :INNER next end if ss.scan(/macro.*$/) state2 = :MACRO next end if ss.scan(/rule.*$/) state2 = :RULE next end if ss.scan(/end.*$/) state1 = :FOOT next end case state2 when :OPTION add_option st when :INNER add_inner st when :MACRO add_macro st when :RULE case state3 when nil rule_state, rule_expr, rule_action = parse_rule(st) if rule_action =~ /\s*\{/ lastmodes = parse_action(rule_action, lastmodes) if lastmodes.empty? add_rule rule_state, rule_expr, rule_action else state3 = :CONT rule_action += "\n" end else add_rule rule_state, rule_expr end when :CONT rule_action += "#{st}\n" lastmodes = parse_action(st, lastmodes) if lastmodes.empty? state3 = nil add_rule rule_state, rule_expr, rule_action else end end # case state3 end # case state2 end # case state1 end # while end def parse_rule(st) st.strip! return if st.size == 0 or st[0,1] == '#' ss = StringScanner.new(st) ss.scan(/\s+/) rule_state = ss.scan(/\:\S+/) ss.scan(/\s+/) rule_expr = ss.scan(/\S+/) ss.scan(/\s+/) [rule_state, rule_expr, ss.post_match] end def parse_action(st, lastmodes=[]) modes = lastmodes mode = lastmodes[-1] ss = StringScanner.new(st) until ss.eos? c = ss.scan(/./) case c when '#' if (mode == :brace) or (mode == nil) #p [c, mode, modes] return modes end when '{' if (mode == :brace) or (mode == nil) mode = :brace modes.push mode end when '}' if (mode == :brace) modes.pop mode = modes[0] end when "'" if (mode == :brace) mode = :quote modes.push mode elsif (mode == :quote) modes.pop mode = modes[0] end when '"' if (mode == :brace) mode = :doublequote modes.push mode elsif (mode == :doublequote) modes.pop mode = modes[0] end when '`' if (mode == :brace) mode = :backquote modes.push mode elsif (mode == :backquote) modes.pop mode = modes[0] end end end #p [c, mode, modes] return modes end REX_HEADER = <<-REX_EOT.gsub(/^ {6}/, '') #-- # DO NOT MODIFY!!!! # This file is automatically generated by rex %s # from lexical definition file "%s". #++ REX_EOT REX_UTIL = <<-REX_EOT require 'strscan' class ScanError < StandardError ; end attr_reader :lineno attr_reader :filename attr_accessor :state def scan_setup(str) @ss = StringScanner.new(str) @lineno = 1 @state = nil end def action yield end def scan_str(str) scan_setup(str) do_parse end alias :scan :scan_str def load_file( filename ) @filename = filename File.open(filename, "r") do |f| scan_setup(f.read) end end def scan_file( filename ) load_file(filename) do_parse end REX_EOT REX_STUB = <<-REX_EOT if __FILE__ == $0 exit if ARGV.size != 1 filename = ARGV.shift rex = %s.new begin rex.load_file filename while token = rex.next_token p token end rescue $stderr.printf %s, rex.filename, rex.lineno, $!.message end end REX_EOT def scanner_io unless scanner_file = @opt['--output-file'] scanner_file = grammar_file + ".rb" end File.open(scanner_file, 'wb') end private :scanner_io def write_scanner f = scanner_io flag = "" flag += "i" if @opt['--ignorecase'] f.printf REX_HEADER, Rexical::VERSION, grammar_file unless @opt['--independent'] f.printf "require 'racc/parser'\n" end @scanner_header.each_line do |s| f.print s end if @opt['--independent'] f.puts "class #{@class_name}" else f.puts "class #{@class_name} < Racc::Parser" end f.print REX_UTIL eos_check = @opt["--matcheos"] ? "" : "return if @ss.eos?" ## scanner method f.print <<-REX_EOT def next_token #{eos_check} # skips empty actions until token = _next_token or @ss.eos?; end token end def _next_token text = @ss.peek(1) @lineno += 1 if text == "\\n" token = case @state REX_EOT exclusive_states.each do |es| if es.nil? f.printf <<-REX_EOT when #{(["nil"] + rules.collect{ |rule| rule[1].nil? ? "nil" : rule[1] }).uniq.join(', ')} REX_EOT else f.printf <<-REX_EOT when #{es} REX_EOT end f.printf <<-REX_EOT case REX_EOT rules.each do |rule| exclusive_state, start_state, rule_expr, rule_action = *rule if es == exclusive_state if rule_action if start_state f.print <<-REX_EOT when((state == #{start_state}) and (text = @ss.scan(/#{rule_expr}/#{flag}))) action #{rule_action} REX_EOT else f.print <<-REX_EOT when (text = @ss.scan(/#{rule_expr}/#{flag})) action #{rule_action} REX_EOT end else if start_state f.print <<-REX_EOT when (@state == #{start_state}) && (text = @ss.scan(/#{rule_expr}/#{flag})) ; REX_EOT else f.print <<-REX_EOT when (text = @ss.scan(/#{rule_expr}/#{flag})) ; REX_EOT end end end end # rules.each if @opt["--matcheos"] eos_check = <<-REX_EOT when @@ss.scan(/$/) ; REX_EOT else eos_check = "" end f.print <<-REX_EOT #{eos_check} else text = @ss.string[@ss.pos .. -1] raise ScanError, "can not match: '" + text + "'" end # if REX_EOT end # exclusive_states.each f.print <<-REX_EOT else raise ScanError, "undefined state: '" + state.to_s + "'" end # case state REX_EOT if @opt['--debug'] f.print <<-REX_EOT p token REX_EOT end f.print <<-REX_EOT token end # def _next_token REX_EOT @scanner_inner.each_line do |s| f.print s end f.puts "end # class" @scanner_footer.each_line do |s| f.print s end f.printf REX_STUB, @class_name, '"%s:%d:%s\n"' if @opt['--stub'] f.close end ## def write_scanner end ## class Generator end ## module Rexical rexical-1.0.7/lib/rexical.rb0000644000004100000410000000027213523645102015722 0ustar www-datawww-datarequire "rexical/generator" require "rexical/rexcmd" module Rexical VERSION = "1.0.7" Copyright = "Copyright (c) 2005-2006 ARIMA Yasuhiro" Mailto = "arima.yasuhiro@nifty.com" end rexical-1.0.7/rexical.gemspec0000644000004100000410000000562413523645102016202 0ustar www-datawww-data######################################################### # This file has been automatically generated by gem2tgz # ######################################################### # -*- encoding: utf-8 -*- # stub: rexical 1.0.7 ruby lib Gem::Specification.new do |s| s.name = "rexical".freeze s.version = "1.0.7" s.required_rubygems_version = Gem::Requirement.new(">= 0".freeze) if s.respond_to? :required_rubygems_version= s.require_paths = ["lib".freeze] s.authors = ["Aaron Patterson".freeze] s.date = "2019-08-06" s.description = "Rexical is a lexical scanner generator that is used with\nRacc to generate Ruby programs.\nRexical is written in Ruby.".freeze s.email = ["aaronp@rubyforge.org".freeze] s.executables = ["rex".freeze] s.extra_rdoc_files = ["CHANGELOG.rdoc".freeze, "DOCUMENTATION.en.rdoc".freeze, "DOCUMENTATION.ja.rdoc".freeze, "Manifest.txt".freeze, "README.rdoc".freeze] s.files = ["CHANGELOG.rdoc".freeze, "DOCUMENTATION.en.rdoc".freeze, "DOCUMENTATION.ja.rdoc".freeze, "Manifest.txt".freeze, "README.ja".freeze, "README.rdoc".freeze, "Rakefile".freeze, "bin/rex".freeze, "lib/rexical.rb".freeze, "lib/rexical/generator.rb".freeze, "lib/rexical/rexcmd.rb".freeze, "sample/a.cmd".freeze, "sample/b.cmd".freeze, "sample/c.cmd".freeze, "sample/calc3.racc".freeze, "sample/calc3.rex".freeze, "sample/calc3.rex.rb".freeze, "sample/calc3.tab.rb".freeze, "sample/error1.rex".freeze, "sample/error2.rex".freeze, "sample/sample.html".freeze, "sample/sample.rex".freeze, "sample/sample.rex.rb".freeze, "sample/sample.xhtml".freeze, "sample/sample1.c".freeze, "sample/sample1.rex".freeze, "sample/sample2.bas".freeze, "sample/sample2.rex".freeze, "sample/simple.html".freeze, "sample/simple.xhtml".freeze, "sample/xhtmlparser.racc".freeze, "sample/xhtmlparser.rex".freeze, "test/assets/test.rex".freeze, "test/rex-20060125.rb".freeze, "test/rex-20060511.rb".freeze, "test/test_generator.rb".freeze] s.homepage = "http://github.com/tenderlove/rexical/tree/master".freeze s.licenses = ["MIT".freeze] s.rdoc_options = ["--main".freeze, "README.rdoc".freeze] s.rubygems_version = "2.5.2.1".freeze s.summary = "Rexical is a lexical scanner generator that is used with Racc to generate Ruby programs".freeze if s.respond_to? :specification_version then s.specification_version = 4 if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then s.add_development_dependency(%q.freeze, ["~> 3.18"]) s.add_development_dependency(%q.freeze, ["< 7", ">= 4.0"]) s.add_development_dependency(%q.freeze, ["~> 0.74.0"]) else s.add_dependency(%q.freeze, ["~> 3.18"]) s.add_dependency(%q.freeze, ["< 7", ">= 4.0"]) s.add_dependency(%q.freeze, ["~> 0.74.0"]) end else s.add_dependency(%q.freeze, ["~> 3.18"]) s.add_dependency(%q.freeze, ["< 7", ">= 4.0"]) s.add_dependency(%q.freeze, ["~> 0.74.0"]) end end rexical-1.0.7/README.rdoc0000644000004100000410000000232713523645102015011 0ustar www-datawww-data= Rexical * http://github.com/tenderlove/rexical/tree/master == DESCRIPTION Rexical is a lexical scanner generator that is used with Racc to generate Ruby programs. Rexical is written in Ruby. == SYNOPSIS Several examples of Rexical grammar files are provided in the sample directory. Here is an example of a lexical definition: class Sample macro BLANK [\ \t]+ rule {BLANK} # no action \d+ { [:digit, text.to_i] } \w+ { [:word, text] } \n . { [text, text] } end Here are examples of the command line usage: $ rex sample1.rex --stub $ ruby sample1.rex.rb sample1.c $ rex sample2.rex --stub $ ruby sample2.rex.rb sample2.bas $ racc calc3.racc $ rex calc3.rex $ ruby calc3.tab.rb The description files for lexical analysis in the sample directory are the files ending with the .rex extension. == REQUIREMENTS * ruby version 1.8.x or later. == INSTALL * sudo gem install rexical == LICENSE Rexical is distributed under the terms of the GNU Lesser General Public License version 2. Note that you do NOT need to follow LGPL for your own parser (Rexical outputs). You can provide those files under any licenses you want. rexical-1.0.7/DOCUMENTATION.ja.rdoc0000644000004100000410000001072413523645102016456 0ustar www-datawww-data = REX: Ruby Lex for Racc == 概要 Racc と併用する Ruby 用の字句スキャナ生成ツール。 == 使い方 rex [options] grammarfile -o --output-file filename 出力ファイル名指定 -s --stub デバッグ用の主処理を付加 -i --ignorecase 大文字小文字を区別しない -C --check-only 文法検査のみ --independent 非依存モード -d --debug デバッグ情報表示 -h --help 使い方の説明 --version バージョン表明 --copyright 著作権情報表示 == デフォルトの出力ファイル名 foo.rex について foo.rex.rb を出力する。 以下のように利用されることを想定している。 require 'foo.rex' == 入力ファイル構造 頭部、規則部、脚部の順に定義する。 規則部には、複数の節が含まれる。 各節は、行頭がキーワードで始まる。 概要： [頭部] "class" Foo ["option" [オプション] ] ["inner" [メソッド定義] ] ["macro" [マクロ名 正規表現] ] "rule" [スタート状態] パターン [アクション] "end" [脚部] === 入力ファイル記述例 class Foo macro BLANK \s+ DIGIT \d+ rule {BLANK} {DIGIT} { [:NUMBER, text.to_i] } . { [text, text] } end == 頭部（省略可能） 規則部の定義以前に記述された内容は、すべて出力ファイル冒頭に転記される。 == 脚部（省略可能） 規則部の定義以降に記述された内容は、すべて出力ファイル末尾に転記される。 == 規則部 規則部は "class" キーワードから始まる行から "end" キーワードから始まる 行までである。 "class" キーワードに続けて出力するクラス名を指定する。 モジュール名で修飾すると、モジュール内クラスとなる。 Racc::Parser を継承したクラスを生成する。 === 規則部定義例 class Foo class Bar::Foo == オプション（省略可能） この節は "option" キーワードで始まる。 "ignorecase" 大文字小文字を区別しない。 "stub" デバッグ用の主処理を付加 "independent" 非依存モード。Racc を継承しない。 == 内部ユーザコード（省略可能） この節は "inner" キーワードで始まる。 ここで定義した内容は、生成したスキャナのクラスの内部で定義される。 == マクロ定義（省略可能） この節は "macro" キーワードで始まる。 一綴りの正規表現に名前をつける。 \ でエスケープすることで空白を含めることができる。 === マクロ定義例 DIGIT \d+ IDENT [a-zA-Z_][a-zA-Z0-9_]* BLANK [\ \t]+ REMIN \/\* REMOUT \*\/ == 走査規則 この節は "rule" キーワードで始まる。 [state] pattern [actions] === state: スタート状態（省略可能） スタート状態は ":" を前置する識別子で表される。 続く英字が大文字のとき、排他的スタート状態となる。 小文字のとき、包含的スタート状態となる。 スタート状態の初期値および省略時値は nil である。 === pattern: 文字列パターン 文字列を特定するための正規表現。 正規表現の記述には、括弧で括ったマクロ定義を用いることができる。 空白を含む正規表現を用いるには、マクロを使用する。 === actions: アクション（省略可能） パターンに適合するときアクションは実行される。 適切なトークンを作成する処理を定義する。 トークンは、種別と値の二項を持つ配列、または nil である。 トークンを作成するために以下の要素を利用できる。 lineno 入力行番号 ( Read Only ) text 検出した文字列 ( Read Only ) state スタート状態 ( Read/Write ) アクションは { } で括った Ruby のブロックである。 ブロックを越えて制御の流れを変える機能を使用してはいけない。 ( return, exit, next, break, ... ) アクションが省略されると、適合した文字列は破棄されて次の走査に進む。 === 走査規則定義例 {REMIN} { state = :REM ; [:REM_IN, text] } :REM {REMOUT} { state = nil ; [:REM_OUT, text] } :REM (.+)(?={REMOUT}) { [:COMMENT, text] } {BLANK} -?{DIGIT} { [:NUMBER, text.to_i] } {WORD} { [:word, text] } . { [text, text] } == コメント（省略可能） 各行において "#" から 行末までがコメントになる。 == 生成したクラスの使い方 === scan_setup() スキャナの実行開始時に初期化するためのイベント。 再定義して使用する。 === scan_str( str ) 定義された文法によって記述された文字列を解釈する。 token を内部に保持する。 === scan_file( filename ) 定義された文法によって記述されたファイルを読み込む。 token を内部に保持する。 === next_token 内部に保持する token をひとつずつ取り出す。 最後は nil を返す。 == 注意 本仕様は暫定的であり、予告なく変更される場合がある。 rexical-1.0.7/README.ja0000644000004100000410000000332213523645102014450 0ustar www-datawww-dataRexical README =========== Rexical は Ruby のためのスキャナジェネレータです。 lex の Ruby 版に相当します。 Racc とともに使うように設計されています。 必要環境 -------- * ruby 1.8 以降 インストール ------------ パッケージのトップディレクトリで次のように入力してください。 ($ は通常ユーザ、# はルートのプロンプトです) $ ruby setup.rb config $ ruby setup.rb setup ($ su) # ruby setup.rb install これで通常のパスに Racc がインストールされます。自分の好き なディレクトリにインストールしたいときは、setup.rb config に 各種オプションをつけて実行してください。オプションのリストは $ ruby setup.rb --help で見られます。 テスト ------ sample/ 以下にいくつか Rexical の文法ファイルのサンプルが用意 してあります。以下を実行してください。 $ rex sample1.rex --stub $ ruby sample1.rex.rb sample1.c $ rex sample2.rex --stub $ ruby sample2.rex.rb sample2.bas $ racc calc3.racc $ rex calc3.rex $ ruby calc3.tab.rb Rexical の詳しい文法は doc/ ディレクトリ以下を見てください。 また記述例は sample/ ディレクトリ以下を見てください。 ライセンス ---------- ライセンスは GNU Lesser General Public License (LGPL) version 2 です。ただしユーザが書いた規則ファイルや、Racc がそこから生成した Ruby スクリプトはその対象外です。好きなライセンスで配布してください。 バグなど -------- Rexical を使っていてバグらしき現象に遭遇したら、下記のアドレスまで メールをください。 そのときはできるだけバグを再現できる文法ファイルを付けてください。 ARIMA Yasuhiro arima.yasuhiro@nifty.com http://raa.ruby-lang.org/project/rex/ rexical-1.0.7/Manifest.txt0000644000004100000410000000120713523645102015506 0ustar www-datawww-dataCHANGELOG.rdoc DOCUMENTATION.en.rdoc DOCUMENTATION.ja.rdoc Manifest.txt README.ja README.rdoc Rakefile bin/rex lib/rexical.rb lib/rexical/generator.rb lib/rexical/rexcmd.rb sample/a.cmd sample/b.cmd sample/c.cmd sample/calc3.racc sample/calc3.rex sample/calc3.rex.rb sample/calc3.tab.rb sample/error1.rex sample/error2.rex sample/sample.html sample/sample.rex sample/sample.rex.rb sample/sample.xhtml sample/sample1.c sample/sample1.rex sample/sample2.bas sample/sample2.rex sample/simple.html sample/simple.xhtml sample/xhtmlparser.racc sample/xhtmlparser.rex test/assets/test.rex test/rex-20060125.rb test/rex-20060511.rb test/test_generator.rb rexical-1.0.7/sample/0000755000004100000410000000000013523645102014460 5ustar www-datawww-datarexical-1.0.7/sample/calc3.rex.rb0000644000004100000410000000351213523645102016570 0ustar www-datawww-data# # DO NOT MODIFY!!!! # This file is automatically generated by rex 1.0.0 # from lexical definition file "calc3.rex". # require 'racc/parser' # # calc3.rex # lexical scanner definition for rex # class Calculator3 < Racc::Parser require 'strscan' class ScanError < StandardError ; end attr_reader :lineno attr_reader :filename def scan_setup ; end def action &block yield end def scan_str( str ) scan_evaluate str do_parse end def load_file( filename ) @filename = filename open(filename, "r") do |f| scan_evaluate f.read end end def scan_file( filename ) load_file filename do_parse end def next_token @rex_tokens.shift end def scan_evaluate( str ) scan_setup @rex_tokens = [] @lineno = 1 ss = StringScanner.new(str) state = nil until ss.eos? text = ss.peek(1) @lineno += 1 if text == "\n" case state when nil case when (text = ss.scan(/\s+/)) ; when (text = ss.scan(/\d+/)) @rex_tokens.push action { [:NUMBER, text.to_i] } when (text = ss.scan(/.|\n/)) @rex_tokens.push action { [text, text] } else text = ss.string[ss.pos .. -1] raise ScanError, "can not match: '" + text + "'" end # if else raise ScanError, "undefined state: '" + state.to_s + "'" end # case state end # until ss end # def scan_evaluate end # class if __FILE__ == $0 exit if ARGV.size != 1 filename = ARGV.shift rex = Calculator3.new begin rex.load_file filename while token = rex.next_token p token end rescue $stderr.printf "%s:%d:%s\n", rex.filename, rex.lineno, $!.message end end rexical-1.0.7/sample/xhtmlparser.rex0000644000004100000410000000521213523645102017551 0ustar www-datawww-data# # xhtmlparser.rex # lexical scanner definition for rex # # usage # rex xhtmlparser.rex --stub # ruby xhtmlparser.rex.rb sample.xhtml # class XHTMLParser option ignorecase macro BLANK \s+ TAG_IN \< TAG_OUT \> ETAG_IN \<\/ ETAG_OUT \/\> XTAG_IN \<\? XTAG_OUT \?\> EXT \! REM \-\- EQUAL \= Q1 \' Q2 \" rule # [:state] pattern [actions] {XTAG_IN} { state = :TAG; [:xtag_in, text] } {ETAG_IN} { state = :TAG; [:etag_in, text] } {TAG_IN} { state = :TAG; [:tag_in, text] } :TAG {EXT} { state = :EXT; [:ext, text] } :EXT {REM} { state = :REM; [:rem_in, text] } :EXT {XTAG_OUT} { state = nil; [:xtag_out, text] } :EXT {TAG_OUT} { state = nil; [:tag_out, text] } :EXT .+(?={REM}) { [:exttext, text] } :EXT .+(?={TAG_OUT}) { [:exttext, text] } :EXT .+(?=$) { [:exttext, text] } :EXT \n :REM {REM} { state = :EXT; [:rem_out, text] } :REM .+(?={REM}) { [:remtext, text] } :REM .+(?=$) { [:remtext, text] } :REM \n :TAG {BLANK} :TAG {XTAG_OUT} { state = nil; [:xtag_out, text] } :TAG {ETAG_OUT} { state = nil; [:etag_out, text] } :TAG {TAG_OUT} { state = nil; [:tag_out, text] } :TAG {EQUAL} { [:equal, text] } :TAG {Q1} { state = :Q1; [:quote1, text] } # ' :Q1 {Q1} { state = :TAG; [:quote1, text] } # ' :Q1 [^{Q1}]+(?={Q1}) { [:value, text] } # ' :TAG {Q2} { state = :Q2; [:quote2, text] } # " :Q2 {Q2} { state = :TAG; [:quote2, text] } # " :Q2 [^{Q2}]+(?={Q2}) { [:value, text] } # " :TAG [\w\-]+(?={EQUAL}) { [:attr, text] } :TAG [\w\-]+ { [:element, text] } \s+(?=\S) .*\S(?=\s*{ETAG_IN}) { [:text, text] } .*\S(?=\s*{TAG_IN}) { [:text, text] } .*\S(?=\s*$) { [:text, text] } \s+(?=$) inner end rexical-1.0.7/sample/error1.rex0000644000004100000410000000036713523645102016420 0ustar www-datawww-data# # eooro1.rex # lexical definition sample for rex # class Error1 macro BLANK [\ \t]+ rule {BLANK} # no action \d+ { [:digit, text.to_i] } \w+ { [:word, text] } \n # . { [text, text] } end rexical-1.0.7/sample/calc3.rex0000644000004100000410000000033113523645102016162 0ustar www-datawww-data# # calc3.rex # lexical scanner definition for rex # class Calculator3 macro BLANK \s+ DIGIT \d+ rule {BLANK} {DIGIT} { [:NUMBER, text.to_i] } .|\n { [text, text] } inner end rexical-1.0.7/sample/calc3.racc0000644000004100000410000000147013523645102016301 0ustar www-datawww-data# # A simple calculator, version 3. # class Calculator3 prechigh nonassoc UMINUS left '*' '/' left '+' '-' preclow options no_result_var rule target : exp | /* none */ { 0 } exp : exp '+' exp { val[0] + val[2] } | exp '-' exp { val[0] - val[2] } | exp '*' exp { val[0] * val[2] } | exp '/' exp { val[0] / val[2] } | '(' exp ')' { val[1] } | '-' NUMBER =UMINUS { -(val[1]) } | NUMBER end ---- header ---- # # generated by racc # require 'calc3.rex' ---- inner ---- ---- footer ---- puts 'sample calc' puts '"q" to quit.' calc = Calculator3.new while true print '>>> '; $stdout.flush str = $stdin.gets.strip break if /q/i === str begin p calc.scan_str(str) rescue ParseError puts 'parse error' end end rexical-1.0.7/sample/error2.rex0000644000004100000410000000041013523645102016406 0ustar www-datawww-data# # error2.rex # lexical definition sample for rex # class Error2 macro BLANK [\ \t]+ rule {BLANK} # no action \d+ { [:digit, text.to_i] } \w+ { [:word, text] } \n . { state = :NONDEF ; [text, text] } end rexical-1.0.7/sample/b.cmd0000644000004100000410000000004213523645102015362 0ustar www-datawww-datacall racc xhtmlparser.racc -v %* rexical-1.0.7/sample/sample1.rex0000644000004100000410000000202513523645102016541 0ustar www-datawww-data# # sample1.rex # lexical definition sample for rex # # usage # rex sample1.rex --stub # ruby sample1.rex.rb sample1.c # class Sample1 macro BLANK \s+ REM_IN \/\* REM_OUT \*\/ REM \/\/ rule # [:state] pattern [actions] # remark {REM_IN} { state = :REMS; [:rem_in, text] } :REMS {REM_OUT} { state = nil; [:rem_out, text] } :REMS .*(?={REM_OUT}) { [:remark, text] } {REM} { state = :REM; [:rem_in, text] } :REM \n { state = nil; [:rem_out, text] } :REM .*(?=$) { [:remark, text] } # literal \"[^"]*\" { [:string, text] } # " \'[^']\' { [:character, text] } # ' # skip {BLANK} # no action # numeric \d+ { [:digit, text.to_i] } # identifier \w+ { [:word, text] } . { [text, text] } end rexical-1.0.7/sample/sample.html0000644000004100000410000000146713523645102016637 0ustar www-datawww-data Title

HTML 4.01

rexical-1.0.7/sample/xhtmlparser.racc0000644000004100000410000000243313523645102017665 0ustar www-datawww-data# # xml parser # class XHTMLParser rule target : /* none */ | xml_doc xml_doc : xml_header extra xml_body | xml_header xml_body | xml_body xml_header : xtag_in element attributes xtag_out xml_body : tag_from contents tag_to tag_from : tag_in element attributes tag_out tag_empty : tag_in element attributes etag_out tag_to : etag_in element tag_out attributes : /* none */ | attributes attribute attribute : attr equal quoted quoted : quote1 value quote1 | quote2 value quote2 contents : /* none */ | contents content content : text | extra | tag_from contents tag_to | tag_empty extra : tag_in ext extra_texts tag_out extra_texts : /* none */ | extra_texts rem_in remtexts rem_out | extra_texts exttext remtexts : remtext | remtexts remtext end ---- header ---- # # generated by racc # require 'xhtmlparser.rex' ---- inner ---- ---- footer ---- exit if ARGV.size == 0 filename = ARGV.shift htmlparser = XHTMLParser.new htmlparser.scan_file filename rexical-1.0.7/sample/c.cmd0000644000004100000410000000025113523645102015365 0ustar www-datawww-data:ruby xhtmlparser.tab.rb simple.html %* :ruby xhtmlparser.tab.rb simple.xhtml %* :ruby xhtmlparser.tab.rb sample.html %* ruby xhtmlparser.tab.rb sample.xhtml %* rexical-1.0.7/sample/sample2.rex0000644000004100000410000000144313523645102016545 0ustar www-datawww-data# # sample2.rex # lexical definition sample for rex # # usage # rex sample2.rex --stub # ruby sample2.rex.rb sample2.bas # class Sample2 option ignorecase macro BLANK \s+ REMARK \' # ' rule {REMARK} { state = :REM; [:rem_in, text] } # ' :REM \n { state = nil; [:rem_out, text] } :REM .*(?=$) { [:remark, text] } \"[^"]*\" { [:string, text] } # " {BLANK} # no action INPUT { [:input, text] } PRINT { [:print, text] } \d+ { [:digit, text.to_i] } \w+ { [:word, text] } . { [text, text] } end rexical-1.0.7/sample/a.cmd0000644000004100000410000000004013523645102015357 0ustar www-datawww-datacall rex xhtmlparser.rex -s %* rexical-1.0.7/sample/calc3.tab.rb0000644000004100000410000000706313523645102016545 0ustar www-datawww-data# # DO NOT MODIFY!!!! # This file is automatically generated by racc 1.4.4 # from racc grammar file "calc3.racc". # require 'racc/parser' # # generated by racc # require 'calc3.rex' class Calculator3 < Racc::Parser ##### racc 1.4.4 generates ### racc_reduce_table = [ 0, 0, :racc_error, 1, 11, :_reduce_none, 0, 11, :_reduce_2, 3, 12, :_reduce_3, 3, 12, :_reduce_4, 3, 12, :_reduce_5, 3, 12, :_reduce_6, 3, 12, :_reduce_7, 2, 12, :_reduce_8, 1, 12, :_reduce_none ] racc_reduce_n = 10 racc_shift_n = 19 racc_action_table = [ 7, 8, 9, 10, 6, 18, 3, 4, 11, 5, 7, 8, 9, 10, 3, 4, 13, 5, 3, 4, nil, 5, 3, 4, nil, 5, 3, 4, nil, 5, 3, 4, nil, 5, 7, 8, 7, 8 ] racc_action_check = [ 12, 12, 12, 12, 1, 12, 10, 10, 3, 10, 2, 2, 2, 2, 0, 0, 6, 0, 4, 4, nil, 4, 9, 9, nil, 9, 8, 8, nil, 8, 7, 7, nil, 7, 17, 17, 16, 16 ] racc_action_pointer = [ 8, 4, 7, -1, 12, nil, 16, 24, 20, 16, 0, nil, -3, nil, nil, nil, 33, 31, nil ] racc_action_default = [ -2, -10, -1, -10, -10, -9, -10, -10, -10, -10, -10, -8, -10, 19, -5, -6, -3, -4, -7 ] racc_goto_table = [ 2, 1, nil, nil, 12, nil, nil, 14, 15, 16, 17 ] racc_goto_check = [ 2, 1, nil, nil, 2, nil, nil, 2, 2, 2, 2 ] racc_goto_pointer = [ nil, 1, 0 ] racc_goto_default = [ nil, nil, nil ] racc_token_table = { false => 0, Object.new => 1, :UMINUS => 2, "*" => 3, "/" => 4, "+" => 5, "-" => 6, "(" => 7, ")" => 8, :NUMBER => 9 } racc_use_result_var = false racc_nt_base = 10 Racc_arg = [ racc_action_table, racc_action_check, racc_action_default, racc_action_pointer, racc_goto_table, racc_goto_check, racc_goto_default, racc_goto_pointer, racc_nt_base, racc_reduce_table, racc_token_table, racc_shift_n, racc_reduce_n, racc_use_result_var ] Racc_token_to_s_table = [ '$end', 'error', 'UMINUS', '"*"', '"/"', '"+"', '"-"', '"("', '")"', 'NUMBER', '$start', 'target', 'exp'] Racc_debug_parser = false ##### racc system variables end ##### # reduce 0 omitted # reduce 1 omitted module_eval <<'.,.,', 'calc3.racc', 13 def _reduce_2( val, _values) 0 end .,., module_eval <<'.,.,', 'calc3.racc', 15 def _reduce_3( val, _values) val[0] + val[2] end .,., module_eval <<'.,.,', 'calc3.racc', 16 def _reduce_4( val, _values) val[0] - val[2] end .,., module_eval <<'.,.,', 'calc3.racc', 17 def _reduce_5( val, _values) val[0] * val[2] end .,., module_eval <<'.,.,', 'calc3.racc', 18 def _reduce_6( val, _values) val[0] / val[2] end .,., module_eval <<'.,.,', 'calc3.racc', 19 def _reduce_7( val, _values) val[1] end .,., module_eval <<'.,.,', 'calc3.racc', 20 def _reduce_8( val, _values) -(val[1]) end .,., # reduce 9 omitted def _reduce_none( val, _values) val[0] end end # class Calculator3 puts 'sample calc' puts '"q" to quit.' calc = Calculator3.new while true print '>>> '; $stdout.flush str = $stdin.gets.strip break if /q/i === str begin p calc.scan_str(str) rescue ParseError puts 'parse error' end end rexical-1.0.7/sample/sample.rex0000644000004100000410000000036613523645102016466 0ustar www-datawww-data# # sample.rex # lexical definition sample for rex # class Sample macro BLANK [\ \t]+ rule {BLANK} # no action \d+ { [:digit, text.to_i] } \w+ { [:word, text] } \n . { [text, text] } end rexical-1.0.7/sample/sample1.c0000644000004100000410000000017213523645102016166 0ustar www-datawww-data int main(int argc, char **argv) { /* block remark */ int i = 100; // inline remark printf("hello, world\n"); } rexical-1.0.7/sample/simple.html0000644000004100000410000000010713523645102016635 0ustar www-datawww-data

Hello World.

rexical-1.0.7/sample/sample.rex.rb0000644000004100000410000000366613523645102017076 0ustar www-datawww-data#-- # DO NOT MODIFY!!!! # This file is automatically generated by rex 1.0.5 # from lexical definition file "sample/sample.rex". #++ require 'racc/parser' # # sample.rex # lexical definition sample for rex # class Sample < Racc::Parser require 'strscan' class ScanError < StandardError ; end attr_reader :lineno attr_reader :filename attr_accessor :state def scan_setup(str) @ss = StringScanner.new(str) @lineno = 1 @state = nil end def action yield end def scan_str(str) scan_setup(str) do_parse end alias :scan :scan_str def load_file( filename ) @filename = filename open(filename, "r") do |f| scan_setup(f.read) end end def scan_file( filename ) load_file(filename) do_parse end def next_token return if @ss.eos? # skips empty actions until token = _next_token or @ss.eos?; end token end def _next_token text = @ss.peek(1) @lineno += 1 if text == "\n" token = case @state when nil case when (text = @ss.scan(/[ \t]+/)) ; when (text = @ss.scan(/\d+/)) action { [:digit, text.to_i] } when (text = @ss.scan(/\w+/)) action { [:word, text] } when (text = @ss.scan(/\n/)) ; when (text = @ss.scan(/./)) action { [text, text] } else text = @ss.string[@ss.pos .. -1] raise ScanError, "can not match: '" + text + "'" end # if else raise ScanError, "undefined state: '" + state.to_s + "'" end # case state token end # def _next_token end # class rexical-1.0.7/sample/simple.xhtml0000644000004100000410000000037513523645102017034 0ustar www-datawww-data

XHTML 1.1

rexical-1.0.7/sample/sample2.bas0000644000004100000410000000006713523645102016515 0ustar www-datawww-data' inline remark i = 100 input st print "hello, world" rexical-1.0.7/sample/sample.xhtml0000644000004100000410000000160013523645102017014 0ustar www-datawww-data Title

XHTML 1.1