# -*- coding: utf-8 -*- # # misaka.rb - a "美坂" compatible Shiori module for ninix # Copyright (C) 2002 by Tamito KAJIYAMA # Copyright (C) 2002, 2003 by MATSUMURA Namihiko # Copyright (C) 2002-2019 by Shyouzou Sugitani # # This program is free software; you can redistribute it and/or modify it # under the terms of the GNU General Public License (version 2) as # published by the Free Software Foundation. It is distributed in the # hope that it will be useful, but WITHOUT ANY WARRANTY; without even the # implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR # PURPOSE. See the GNU General Public License for more details. # # TODO: # error recovery # - ${foo} # - $(foo) ? require "stringio" require_relative "../home" require_relative "../logging" begin require 'charlock_holmes' rescue LoadError CharlockHolmes = nil end module Misaka class MisakaError < StandardError # ValueError # pass end def self.lexical_error(path: nil, position: nil) if path.nil? and position.nil? error = 'lexical error' elsif path.nil? at = ('line ' + position[0].to_s + ', column ' + position[1].to_s) error = ('lexical error at ' + at.to_s) elsif position.nil? error = ('lexical error in ' + path.to_s) else at = ('line ' + position[0].to_s + ', column ' + position[1].to_s) error = ('lexical error at ' + at.to_s + ' in ' + path.to_s) end fail MisakaError.new(error) end def self.syntax_error(message, path: nil, position: nil) if path.nil? and position.nil? error = ('syntax error (' + message.to_s + ')') elsif path.nil? at = ('line ' + position[0].to_s + ', column ' + position[1].to_s) error = ('syntax error at ' + at.to_s + ' (' + message.to_s{1} + ')') elsif position.nil? error = ('syntax error in ' + path.to_s + ' (' + message.to_s{1} + ')') else at = ('line ' + position[0].to_s + ', column ' + position[1].to_s) error = ('syntax error at ' + at.to_s + ' in ' + path.to_s + ' (' + message.to_s + ')') end fail MisakaError.new(error) end def self.list_dict(top_dir) buf = [] begin filelist, debug, error = Misaka.read_misaka_ini(top_dir) rescue #except Exception as e: filelist = [] end for filename in filelist buf << File.join(top_dir, filename) end return buf end def self.read_misaka_ini(top_dir) path = File.join(top_dir, 'misaka.ini') filelist = [] debug = 0 error = 0 open(path) do |f| while true line = f.gets break if line.nil? line = line.strip() next if line.empty? or line.start_with?('//') case line when 'dictionaries' line = f.gets if line.strip() != '{' Misaka.syntax_error('expected an open brace', :path => path) end while true line = f.gets Misaka.syntax_error('unexpected end of file', :path => path) if line.nil? line = line.strip() break if line == '}' next if line.empty? or line.start_with?('//') filelist << Home.get_normalized_path(line) end when 'debug,0' debug = 0 when 'debug,1' debug = 1 when 'error,0' error = 0 when 'error,1' error = 1 when 'propertyhandler,0' #pass when 'propertyhandler,1' #pass else Logging::Logging.debug("misaka.rb: unknown directive '#{line}'") end end end return filelist, debug, error end ### LEXER ### TOKEN_NEWLINE = 1 TOKEN_WHITESPACE = 2 TOKEN_OPEN_BRACE = 3 TOKEN_CLOSE_BRACE = 4 TOKEN_OPEN_PAREN = 5 TOKEN_CLOSE_PAREN = 6 TOKEN_OPEN_BRACKET = 7 TOKEN_CLOSE_BRACKET = 8 TOKEN_DOLLAR = 9 TOKEN_COMMA = 11 TOKEN_SEMICOLON = 12 TOKEN_OPERATOR = 13 TOKEN_DIRECTIVE = 14 TOKEN_TEXT = 15 class Lexer attr_reader :buffer Re_comment = Regexp.new('\A[ \t]*//[^\r\n]*') Re_newline = Regexp.new('\A(\r\n|\r|\n)') Patterns = [ [TOKEN_NEWLINE, Re_newline], [TOKEN_WHITESPACE, Regexp.new('\A[ \t]+')], [TOKEN_OPEN_BRACE, Regexp.new('\A{')], [TOKEN_CLOSE_BRACE, Regexp.new('\A}')], [TOKEN_OPEN_PAREN, Regexp.new('\A\(')], [TOKEN_CLOSE_PAREN, Regexp.new('\A\)')], [TOKEN_OPEN_BRACKET, Regexp.new('\A\[')], [TOKEN_CLOSE_BRACKET, Regexp.new('\A\]')], [TOKEN_DOLLAR, Regexp.new('\A\$')], [TOKEN_COMMA, Regexp.new('\A,')], [TOKEN_SEMICOLON, Regexp.new('\A;')], [TOKEN_OPERATOR, Regexp.new('\A([!=]=|[<>]=?|=[<>]|&&|\|\||\+\+|--|[+\-*/]?=|[+\-*/%\^])')], [TOKEN_DIRECTIVE, Regexp.new('\A#[_A-Za-z][_A-Za-z0-9]*')], #[TOKEN_TEXT, Regexp.new("[!&|]|(\\[,\"]|[\x01\x02#'.0-9:?@A-Z\\_`a-z~]|[\x80-\xff].)+")], [TOKEN_TEXT, Regexp.new("\\A((\"[^\"]*\")|[!&|]|(\\\\[,\"]|[\x01\x02#'.0-9:?@A-Z\\\\_`a-z~\"]|[\\u0080-\\uffff]+)+)")], ] Token_names = { TOKEN_WHITESPACE => 'whitespace', TOKEN_NEWLINE => 'a newline', TOKEN_OPEN_BRACE => 'an open brace', TOKEN_CLOSE_BRACE => 'a close brace', TOKEN_OPEN_PAREN => 'an open paren', TOKEN_CLOSE_PAREN => 'a close paren', TOKEN_OPEN_BRACKET => 'an open bracket', TOKEN_CLOSE_BRACKET => 'a close bracket', TOKEN_DOLLAR => 'a dollar', TOKEN_COMMA => 'a comma', TOKEN_SEMICOLON => 'a semicolon', TOKEN_OPERATOR => 'an operator', TOKEN_DIRECTIVE => 'an directive', TOKEN_TEXT => 'text', } def initialize(charset: 'CP932') @buffer = [] @charset = charset end def _match(data, line, column, path) temp = data.clone while temp.length > 0 if column.zero? match = Re_comment.match(temp) unless match.nil? column = (column + match[0].length) temp = match.post_match next end end break_flag = false for token, pattern in Patterns match = pattern.match(temp) unless match.nil? lexeme = match[0] if token == TOKEN_TEXT and \ lexeme.start_with?('"') and lexeme.end_with?('"') unless lexeme.include?('$') @buffer << [token, lexeme[1..-2], [line, column]] else # XXX line, column = _match(lexeme[1..-2], line, column + 1, path) column += 1 end else @buffer << [token, lexeme, [line, column]] end temp = match.post_match break_flag = true break end end unless break_flag ###print(temp[0..100 - 1]) Misaka.lexical_error(:path => path, :position => [line, column]) end if token == TOKEN_NEWLINE line = (line + 1) column = 0 else column = (column + lexeme.length) end end return line, column end def read(f, path: nil) line = 1 column = 0 data = f.read() line, column = _match(data, line, column, path) @buffer << [TOKEN_NEWLINE, '', [line, column]] @path = path end def get_position @position end def pop begin token, lexeme, @position = @buffer.shift rescue #except IndexError: Misaka.syntax_error('unexpected end of file', :path => @path) end ###print(token, repr(lexeme)) return token, lexeme end def pop_check(expected) token, lexeme = pop() if token != expected Misaka.syntax_error(['exptected ', Token_names[expected], ', but returns ', Token_names[token]].join(''), :path => @path, :position => get_position()) end return lexeme end def look_ahead(index: 0) begin token, lexeme, position = @buffer[index] rescue #except IndexError: Misaka.syntax_error('unexpected end of file', :path => @path) end return token, lexeme end def skip_space(accept_eof: false) while not @buffer.empty? token, lexeme = look_ahead() return false unless [TOKEN_NEWLINE, TOKEN_WHITESPACE].include?(token) pop() end unless accept_eof Misaka.syntax_error('unexpected end of file', :path => @path) end return true end def skip_line while not @buffer.empty? token, lexeme = pop() break if token == TOKEN_NEWLINE end end end ### PARSER ### NODE_TEXT = 1 NODE_STRING = 2 NODE_VARIABLE = 3 NODE_INDEXED = 4 NODE_ASSIGNMENT = 5 NODE_FUNCTION = 6 NODE_IF = 7 NODE_CALC = 8 NODE_AND_EXPR = 9 NODE_OR_EXPR = 10 NODE_COMP_EXPR = 11 NODE_ADD_EXPR = 12 NODE_MUL_EXPR = 13 NODE_POW_EXPR = 14 class Parser def initialize(charset) @lexer = Lexer.new(:charset => charset) @charset = charset end def read(f, path: nil) @lexer.read(f, :path => path) @path = path end def get_dict common = nil groups = [] # skip junk tokens at the beginning of the file while true if @lexer.skip_space(:accept_eof => true) return common, groups end token, lexeme = @lexer.look_ahead() if token == TOKEN_DIRECTIVE and lexeme == '#_Common' or \ token == TOKEN_DOLLAR break end @lexer.skip_line() end token, lexeme = @lexer.look_ahead() if token == TOKEN_DIRECTIVE and lexeme == '#_Common' common = get_common() if @lexer.skip_space(:accept_eof => true) return common, groups end end while true groups << get_group() if @lexer.skip_space(:accept_eof => true) return common, groups end end return nil, [] end def get_common @lexer.pop() token, lexeme = @lexer.look_ahead() if token == TOKEN_WHITESPACE @lexer.pop() end @lexer.pop_check(TOKEN_NEWLINE) condition = get_brace_expr() @lexer.pop_check(TOKEN_NEWLINE) return condition end def get_group # get group name buf = [] @lexer.pop_check(TOKEN_DOLLAR) while true token, lexeme = @lexer.look_ahead() ##if token == TOKEN_TEXT or \ ## token == TOKEN_OPERATOR and \ ## ['+', '-', '*', '/', '%'].include?(lexeme) ## XXX unless [TOKEN_NEWLINE, TOKEN_COMMA, TOKEN_SEMICOLON].include?(token) token, lexeme = @lexer.pop() buf << unescape(lexeme) else break end end if buf.empty? Misaka.syntax_error('null identifier', :path => @path, :position => @lexer.get_position()) end name = ['$', buf.join('')].join('') # get group parameters parameters = [] while true token, lexeme = @lexer.look_ahead() if token == TOKEN_WHITESPACE @lexer.pop() end token, lexeme = @lexer.look_ahead() if token == TOKEN_NEWLINE break elsif [TOKEN_COMMA, TOKEN_SEMICOLON].include?(token) @lexer.pop() else Misaka.syntax_error('expected a delimiter', :path => @path, :position => @lexer.get_position()) end token, lexeme = @lexer.look_ahead() if token == TOKEN_WHITESPACE @lexer.pop() end token, lexeme = @lexer.look_ahead() break if token == TOKEN_NEWLINE token, lexeme = @lexer.look_ahead() if token == TOKEN_OPEN_BRACE parameters << get_brace_expr() elsif token == TOKEN_TEXT token, lexeme = @lexer.pop() parameters << [NODE_TEXT, unescape(lexeme)] else Misaka.syntax_error('expected a parameter or brace expression', :path => @path, :position => @lexer.get_position()) end end # get sentences @lexer.pop_check(TOKEN_NEWLINE) sentences = [] while true break if @lexer.skip_space(:accept_eof => true) token, lexeme = @lexer.look_ahead() break if token == TOKEN_DOLLAR sentence = get_sentence() next if sentence.nil? sentences << sentence end return [name, parameters, sentences] end def get_sentence token, lexeme = @lexer.look_ahead() if token == TOKEN_OPEN_BRACE token, lexeme = @lexer.look_ahead(:index => 1) if token == TOKEN_NEWLINE @lexer.pop_check(TOKEN_OPEN_BRACE) token, lexeme = @lexer.look_ahead() if token == TOKEN_WHITESPACE @lexer.pop() end @lexer.pop_check(TOKEN_NEWLINE) line = [] while true @lexer.skip_space() token, lexeme = @lexer.look_ahead() break if token == TOKEN_CLOSE_BRACE for node in get_line() line << node end @lexer.pop_check(TOKEN_NEWLINE) end @lexer.pop_check(TOKEN_CLOSE_BRACE) token, lexeme = @lexer.look_ahead() if token == TOKEN_WHITESPACE @lexer.pop() end else begin line = get_line() # beginning with brace expression rescue # except MisakaError as error: Logging::Logging.debug(error) @lexer.skip_line() return nil end end else begin line = get_line() rescue => error # except MisakaError as error: Logging::Logging.debug(error) @lexer.skip_line() return nil end end @lexer.pop_check(TOKEN_NEWLINE) return line end def is_whitespace(node) return (node[0] == NODE_TEXT and node[1].strip().empty?) end def unescape(text) text = text.gsub('\\,', ',') text = text.gsub('\\"', '"') return text end def get_word buf = [] while true token, lexeme = @lexer.look_ahead() if token == TOKEN_TEXT token, lexeme = @lexer.pop() if unescape(lexeme).start_with?('"') and \ unescape(lexeme).end_with?('"') buf << [NODE_STRING, unescape(lexeme)[1..-2]] else buf << [NODE_TEXT, unescape(lexeme)] end elsif [TOKEN_WHITESPACE, TOKEN_DOLLAR].include?(token) token, lexeme = @lexer.pop() buf << [NODE_TEXT, lexeme] elsif token == TOKEN_OPEN_BRACE buf << get_brace_expr() else break end end # strip whitespace at the beginning and/or end of line if not buf.empty? and is_whitespace(buf[0]) buf.delete_at(0) end if not buf.empty? and is_whitespace(buf[-1]) buf.delete_at(-1) end return buf end def get_line buf = [] while true token, lexeme = @lexer.look_ahead() if [TOKEN_NEWLINE, TOKEN_CLOSE_BRACE].include?(token) break elsif token == TOKEN_TEXT token, lexeme = @lexer.pop() buf << [NODE_TEXT, unescape(lexeme)] elsif [TOKEN_WHITESPACE, TOKEN_DOLLAR, TOKEN_OPEN_PAREN, TOKEN_CLOSE_PAREN, TOKEN_OPEN_BRACKET, TOKEN_CLOSE_BRACKET, TOKEN_OPERATOR, TOKEN_COMMA, TOKEN_SEMICOLON, TOKEN_DIRECTIVE].include?(token) token, lexeme = @lexer.pop() buf << [NODE_TEXT, lexeme] elsif token == TOKEN_OPEN_BRACE buf << get_brace_expr() else fail RuntimeError.new('should not reach here') end end # strip whitespace at the beginning and/or end of line if not buf.empty? and is_whitespace(buf[0]) buf.delete_at(0) end if not buf.empty? and is_whitespace(buf[-1]) buf.delete_at(-1) end return buf end def get_brace_expr @lexer.pop_check(TOKEN_OPEN_BRACE) @lexer.skip_space() @lexer.pop_check(TOKEN_DOLLAR) @lexer.skip_space() # get identifier (function or variable) nodelist = [[NODE_TEXT, '$']] while true token, lexeme = @lexer.look_ahead() if token == TOKEN_TEXT or \ (token == TOKEN_OPERATOR and ['+', '-', '*', '/', '%'].include?(lexeme)) token, lexeme = @lexer.pop() nodelist << [NODE_TEXT, unescape(lexeme)] elsif token == TOKEN_OPEN_BRACE nodelist << get_brace_expr() else break end end if nodelist.length == 1 Misaka.syntax_error('null identifier', :path => @path, :position => @lexer.get_position()) end @lexer.skip_space() token, lexeme = @lexer.look_ahead() if token == TOKEN_OPEN_PAREN # function name = nodelist.map {|node| node[1] }.join('') if name == '$if' cond_expr = get_cond_expr() then_clause = [[NODE_TEXT, 'true']] else_clause = [[NODE_TEXT, 'false']] @lexer.skip_space() token, lexeme = @lexer.look_ahead() if token == TOKEN_OPEN_BRACE @lexer.pop_check(TOKEN_OPEN_BRACE) @lexer.skip_space() then_clause = [] while true line = get_line() for node in line then_clause << node end token, lexem = @lexer.pop() break if token == TOKEN_CLOSE_BRACE fail "assert" unless token == TOKEN_NEWLINE end @lexer.skip_space() token, lexeme = @lexer.look_ahead() if token == TOKEN_TEXT and lexeme == 'else' @lexer.pop() @lexer.skip_space() @lexer.pop_check(TOKEN_OPEN_BRACE) @lexer.skip_space() else_clause = [] while true line = get_line() for node in line else_clause << node end token, lexem = @lexer.pop() break if token == TOKEN_CLOSE_BRACE fail "assert" unless token == TOKEN_NEWLINE end elsif token == TOKEN_OPEN_BRACE ## XXX @lexer.pop_check(TOKEN_OPEN_BRACE) @lexer.skip_space() else_clause = [] while true line = get_line() for node in line else_clause << node end token, lexem = @lexer.pop() break if token == TOKEN_CLOSE_BRACE fail "assert" unless token == TOKEN_NEWLINE end else else_clause = [[NODE_TEXT, '']] end end node = [NODE_IF, cond_expr, then_clause, else_clause] elsif name == '$calc' node = [NODE_CALC, get_add_expr()] else @lexer.pop_check(TOKEN_OPEN_PAREN) @lexer.skip_space() args = [] token, lexeme = @lexer.look_ahead() if token != TOKEN_CLOSE_PAREN while true args << get_argument() @lexer.skip_space() token, lexeme = @lexer.look_ahead() break if token != TOKEN_COMMA @lexer.pop() @lexer.skip_space() end end @lexer.pop_check(TOKEN_CLOSE_PAREN) node = [NODE_FUNCTION, name, args] end else # variable token, lexeme = @lexer.look_ahead() if token == TOKEN_OPERATOR operator = @lexer.pop_check(TOKEN_OPERATOR) case operator when '=', '+=', '-=', '*=', '/=' @lexer.skip_space() value = get_add_expr() when '++' operator = '+=' value = [[NODE_TEXT, '1']] when '--' operator = '-=' value = [[NODE_TEXT, '1']] else Misaka.syntax_error(['bad operator ', operator].join(''), :path => @path, :position => @lexer.get_position()) end node = [NODE_ASSIGNMENT, nodelist, operator, value] elsif token == TOKEN_OPEN_BRACKET @lexer.pop_check(TOKEN_OPEN_BRACKET) @lexer.skip_space() index = get_word() @lexer.skip_space() @lexer.pop_check(TOKEN_CLOSE_BRACKET) node = [NODE_INDEXED, nodelist, index] else node = [NODE_VARIABLE, nodelist] end end @lexer.skip_space() @lexer.pop_check(TOKEN_CLOSE_BRACE) return node end def get_argument buf = [] while true token, lexeme = @lexer.look_ahead() case token when TOKEN_TEXT token, lexeme = @lexer.pop() if unescape(lexeme).start_with?('"') and \ unescape(lexeme).end_with?('"') buf << [NODE_STRING, unescape(lexeme)[1..-2]] else buf << [NODE_TEXT, unescape(lexeme)] end when TOKEN_WHITESPACE, TOKEN_DOLLAR, TOKEN_OPEN_BRACKET, TOKEN_CLOSE_BRACKET, TOKEN_OPERATOR, TOKEN_SEMICOLON token, lexeme = @lexer.pop() buf << [NODE_TEXT, lexeme] when TOKEN_OPEN_BRACE buf << get_brace_expr() when TOKEN_NEWLINE @lexer.skip_space() else break end end # strip whitespace at the beginning and/or end of line if not buf.empty? and is_whitespace(buf[0]) buf.delete_at(0) end if not buf.empty? and is_whitespace(buf[-1]) buf.delete_at(-1) end return buf end def get_cond_expr @lexer.pop_check(TOKEN_OPEN_PAREN) @lexer.skip_space() or_expr = get_or_expr() @lexer.skip_space() @lexer.pop_check(TOKEN_CLOSE_PAREN) return or_expr end def get_or_expr buf = [NODE_OR_EXPR] buf << get_and_expr() while true @lexer.skip_space() token, lexeme = @lexer.look_ahead() break unless token == TOKEN_OPERATOR and lexeme == '||' @lexer.pop() @lexer.skip_space() buf << get_and_expr() end if buf.length > 2 return buf else return buf[1] end end def get_and_expr buf = [NODE_AND_EXPR] buf << get_sub_expr() while true @lexer.skip_space() token, lexeme = @lexer.look_ahead() break unless token == TOKEN_OPERATOR and lexeme == '&&' @lexer.pop() @lexer.skip_space() buf << get_sub_expr() end if buf.length > 2 return buf else return buf[1] end end def get_sub_expr token, lexeme = @lexer.look_ahead() if token == TOKEN_OPEN_PAREN return get_cond_expr() else return get_comp_expr() end end def get_comp_expr buf = [NODE_COMP_EXPR] buf << get_add_expr() @lexer.skip_space() token, lexeme = @lexer.look_ahead() if token == TOKEN_OPERATOR and \ ['==', '!=', '<', '<=', '=<', '>', '>=', '=>'].include?(lexeme) if lexeme == '=<' lexeme = '<=' elsif lexeme == '=>' lexeme = '>=' end buf << lexeme @lexer.pop() @lexer.skip_space() buf << get_add_expr() end if buf.length == 2 buf << '==' buf << [[NODE_TEXT, 'true']] end return buf end def get_add_expr buf = [NODE_ADD_EXPR] buf << get_mul_expr() while true @lexer.skip_space() token, lexeme = @lexer.look_ahead() break unless token == TOKEN_OPERATOR and ['+', '-'].include?(lexeme) buf << lexeme @lexer.pop() @lexer.skip_space() buf << get_mul_expr() end if buf.length > 2 return [buf] else return buf[1] end end def get_mul_expr buf = [NODE_MUL_EXPR] buf << get_pow_expr() while true @lexer.skip_space() token, lexeme = @lexer.look_ahead() break unless token == TOKEN_OPERATOR and ['*', '/', '%'].include?(lexeme) buf << lexeme @lexer.pop() @lexer.skip_space() buf << get_pow_expr() end if buf.length > 2 return [buf] else return buf[1] end end def get_pow_expr buf = [NODE_POW_EXPR] buf << get_unary_expr() while true @lexer.skip_space() token, lexeme = @lexer.look_ahead() break unless token == TOKEN_OPERATOR and lexeme == '^' @lexer.pop() @lexer.skip_space() buf << get_unary_expr() end if buf.length > 2 return [buf] else return buf[1] end end def get_unary_expr token, lexeme = @lexer.look_ahead() if token == TOKEN_OPERATOR and ['+', '-'].include?(lexeme) buf = [NODE_ADD_EXPR, [[NODE_TEXT, '0']], lexeme] @lexer.pop() @lexer.skip_space() buf << get_unary_expr() return [buf] else return get_factor() end end def get_factor token, lexeme = @lexer.look_ahead() if token == TOKEN_OPEN_PAREN @lexer.pop_check(TOKEN_OPEN_PAREN) @lexer.skip_space() add_expr = get_add_expr() @lexer.skip_space() @lexer.pop_check(TOKEN_CLOSE_PAREN) return add_expr else return get_word() end end # for debug def dump_list(nodelist, depth: 0) for node in nodelist dump_node(node, depth) end end def dump_node(node, depth) indent = ' ' * depth case node[0] when NODE_TEXT print([indent, 'TEXT'].join(''), " ", \ ['"', node[1], '"'].join(''), "\n") when NODE_STRING print([indent, 'STRING'].join(''), " ", \ ['"', node[1], '"'].join(''), "\n") when NODE_VARIABLE print([indent, 'VARIABLE'].join(''), "\n") dump_list(node[1], :depth => depth + 1) when NODE_INDEXED print([indent, 'INDEXED'].join(''), "\n") dump_list(node[1], :depth => depth + 1) print([indent, 'index'].join(''), "\n") dump_list(node[2], :depth => depth + 1) when NODE_ASSIGNMENT print([indent, 'ASSIGNMENT'].join(''), "\n") dump_list(node[1], :depth => depth + 1) print([indent, 'op'].join(''), " ", node[2], "\n") dump_list(node[3], :depth => depth + 1) when NODE_FUNCTION print([indent, 'FUNCTION'].join(''), " ", node[1], "\n") for i in 0..node[2].length-1 print([indent, 'args[' + i.to_s + ']'].join(''), "\n") dump_list(node[2][i], :depth => depth + 1) end when NODE_IF print([indent, 'IF'].join(''), "\n") dump_node(node[1], depth + 1) print([indent, 'then'].join(''), "\n") dump_list(node[2], :depth => depth + 1) print([indent, 'else'].join(''), "\n") dump_list(node[3], :depth => depth + 1) when NODE_CALC print([indent, 'CALC'].join(''), "\n") dump_list(node[1], :depth => depth + 1) when NODE_AND_EXPR print([indent, 'AND_EXPR'].join(''), "\n") for i in 1..node.length-1 dump_node(node[i], depth + 1) end when NODE_OR_EXPR print([indent, 'OR_EXPR'].join(''), "\n") for i in 1..node.length-1 dump_node(node[i], depth + 1) end when NODE_COMP_EXPR print([indent, 'COMP_EXPR'].join(''), "\n") dump_list(node[1], :depth => depth + 1) print([indent, 'op'].join(''), " ", node[2], "\n") dump_list(node[3], :depth => depth + 1) when NODE_ADD_EXPR print([indent, 'ADD_EXPR'].join(''), "\n") dump_list(node[1], :depth => depth + 1) for i in 2.step(node.length-1, 2) print([indent, 'op'].join(''), " ", node[i], "\n") dump_list(node[i + 1], :depth => depth + 1) end when NODE_MUL_EXPR print([indent, 'MUL_EXPR'].join(''), "\n") dump_list(node[1], :depth => depth + 1) for i in 2.step(node.length-1, 2) print([indent, 'op'].join(''), " ", node[i], "\n") dump_list(node[i + 1], :depth => depth + 1) end when NODE_POW_EXPR print([indent, 'POW_EXPR'].join(''), "\n") dump_list(node[1], :depth => depth + 1) for i in 2..node.length-1 print([indent, 'op ^'].join(''), "\n") dump_list(node[i], :depth => depth + 1) end else print(node, "\n") fail RuntimeError.new('should not reach here') end end end # <<< syntax >>> # dict := sp ( common sp )? ( group sp )* # sp := ( NEWLINE | WHITESPACE )* # common := "#_Common" NEWLINE brace_expr NEWLINE # group := group_name ( delimiter ( STRING | brace_expr ) )* # ( delimiter )? NEWLINE ( sentence )* # group_name := DOLLAR ( TEXT )+ # delimiter := ( WHITESPACE )? ( COMMA | SEMICOLON ) ( WHITESPACE )? # sentence := line NEWLINE | # OPEN_BRACE NEWLINE ( line NEWLINE )* CLOSE_BRACE NEWLINE # word := ( TEXT | WHITESPACE | DOLLAR | brace_expr | string )* # line := ( TEXT | WHITESPACE | DOLLAR | brace_expr | QUOTE | # OPEN_PAREN | CLOSE_PAREN | OPEN_BRACKET | CLOSE_BRACKET | # OPERATOR | COMMA | SEMICOLON | DIRECTIVE )* # string := QUOTE ( # TEXT | WHITESPACE | DOLLAR | OPEN_BRACE | CLOSE_BRACE | # OPEN_PAREN | CLOSE_PAREN | OPEN_BRACKET | CLOSE_BRACKET | # OPERATOR | COMMA | SEMICOLON | DIRECTIVE )* # QUOTE # brace_expr := variable | indexed | assignment | increment | # function | if | calc # variable := OPEN_BRACE sp var_name sp CLOSE_BRACE # indexed := OPEN_BRACE sp var_name sp # OPEN_BRACKET sp word sp CLOSE_BRACKET sp CLOSE_BRACE # var_name := DOLLAR ( TEXT | brace_expr )+ # assignment := OPEN_BRACE sp var_name sp assign_ops sp add_expr sp CLOSE_BRACE # assign_ops := "=" | "+=" | "-=" | "*=" | "/=" # increment := OPEN_BRACE sp var_name sp inc_ops sp CLOSE_BRACE # inc_ops := "+" "+" | "-" "-" # function := OPEN_BRACE sp DOLLAR sp ( TEXT )+ sp OPEN_PAREN # ( sp argument ( sp COMMA sp argument )* sp )? # CLOSE_PAREN sp CLOSE_BRACE # argument := ( TEXT | WHITESPACE | DOLLAR | brace_expr | string | # OPEN_BRACKET | CLOSE_BRACKET | OPERATOR | SEMICOLON )* # if := OPEN_BRACE sp DOLLAR sp "if" sp cond_expr # ( sp OPEN_BRACE ( sp line )* sp CLOSE_BRACE ( sp "else" # sp OPEN_BRACE ( sp line )* sp CLOSE_BRACE )? sp )? # sp CLOSE_BRACE # calc := OPEN_BRACE sp DOLLAR sp "calc" sp add_expr sp CLOSE_BRACE # cond_expr := OPEN_PAREN sp or_expr sp CLOSE_PAREN # or_expr := and_expr ( sp "||" sp and_expr )* # and_expr := sub_expr ( sp "&&" sp sub_expr )* # sub_expr := comp_expr | cond_expr # comp_expr := add_expr ( sp comp_op sp add_expr )? # comp_op := "==" | "!=" | "<" | "<=" | ">" | ">=" # add_expr := mul_expr (sp add_op sp mul_expr)* # add_op := "+" | "-" # mul_expr := pow_expr (sp mul_op sp pow_expr)* # mul_op := "*" | "/" | "%" # pow_expr := unary_expr (sp pow_op sp unary_expr)* # pow_op := "^" # unary_expr := unary_op sp unary_expr | factor # unary_op := "+" | "-" # factor := OPEN_PAREN sp add_expr sp CLOSE_PAREN | word ### INTERPRETER ### class Group def initialize(misaka, name, item_list: nil) @misaka = misaka @name = name if item_list.nil? ## FIXME @list = [] else @list = item_list end end def __len__ @list.length end def __getitem__(index) @list[index] end def get return nil if @list.empty? return @list.sample end def copy(name) MisakaArray.new(@misaka, name, @list) end end class NonOverlapGroup < Group def initialize(misaka, name, item_list: nil) super(misaka, name, :item_list => item_list) @indexes = [] end def get return nil if @list.empty? if @indexes.empty? @indexes = Array(0..@list.length-1) end i = Array(0..@indexes.length-1).sample index = @indexes.delete_at(i) return @list[index] end end class SequentialGroup < Group def initialize(misaka, name, item_list: nil) super(misaka, name, :item_list => item_list) @indexes = [] end def get return nil if @list.empty? if @indexes.empty? @indexes = Array(0..@list.length-1) end index = @indexes.shift return @list[index] end end class MisakaArray < Group def append(s) @list << [[NODE_TEXT, s.to_s]] end def index(s) for i in 0..@list.length-1 if s == @misaka.expand(@list[i]) return i end end return -1 end def pop return nil if @list.empty? i = Array(0..@list.length-1).sample item = @list.delete_at(i) return item end def popmatchl(s) buf = [] for i in 0..@list.length-1 t = @misaka.expand(@list[i]) if t.start_with?(s) buf << i end end return nil if buf.empty? i = buf.sample item = @list.delete_at(i) return item end end TYPE_SCHOLAR = 1 TYPE_ARRAY = 2 class Shiori DBNAME = 'misaka.db' def initialize(dll_name) @dll_name = dll_name @charset = 'CP932' end def use_saori(saori) @saori = saori end def find(top_dir, dll_name) result = 0 if not Misaka.list_dict(top_dir).empty? result = 210 elsif File.file?(File.join(top_dir, 'misaka.ini')) result = 200 end return result end def show_description Logging::Logging.info( "Shiori: MISAKA compatible module for ninix\n" \ " Copyright (C) 2002 by Tamito KAJIYAMA\n" \ " Copyright (C) 2002, 2003 by MATSUMURA Namihiko\n" \ " Copyright (C) 2002-2019 by Shyouzou Sugitani") end def reset @dict = {} @variable = {} @constant = {} @misaka_debug = 0 @misaka_error = 0 @reference = [nil, nil, nil, nil, nil, nil, nil, nil] @mouse_move_count = {} @random_talk = -1 @otherghost = [] @communicate = ['', ''] reset_request() end def load(dir: Dir::getwd) @misaka_dir = dir @dbpath = File.join(@misaka_dir, DBNAME) @saori_library = SaoriLibrary.new(@saori, @misaka_dir) reset() begin filelist, debug, error = Misaka.read_misaka_ini(@misaka_dir) rescue #except IOError: Logging::Logging.debug('cannot read misaka.ini') return 0 rescue #except MisakaError as error: Logging::Logging.debug(error) return 0 end @misaka_debug = debug @misaka_error = error global_variables = [] global_constants = [] # charset auto-detection unless CharlockHolmes.nil? detector = CharlockHolmes::EncodingDetector.new for filename in filelist path = File.join(@misaka_dir, filename) begin f = open(path, 'rb') rescue #except IOError: Logging::Logging.debug('cannot read ' + filename.to_s) next end ext = File.extname(filename) if ext == '.__1' result = detector.detect(crypt(f.read())) else result = detector.detect(f.read()) end f.close() if result[:confidence] > 98 and \ result[:encoding] != 'ISO-8859-1' # XXX @charset = result[:encoding] if @charset == 'Shift_JIS' @charset = 'CP932' # XXX end Logging::Logging.debug("CharlockHolmes(misaka.rb): '" + @charset.to_s + "'") break end end end for filename in filelist path = File.join(@misaka_dir, filename) basename = File.basename(filename, ".*") ext = File.extname(filename) begin if ext == '.__1' # should read lines as bytes f = open(path, 'rb') else f = open(path, :encoding => @charset + ":utf-8") end rescue #except IOError: Logging::Logging.debug('cannot read ' + filename.to_s) next end if ext == '.__1' f = StringIO.new(crypt(f.read())) end begin variables, constants = read(f, :path => path) rescue => error #except MisakaError as error: Logging::Logging.debug(error) next end global_variables |= variables global_constants |= constants end eval_globals(global_variables, false) load_database() eval_globals(global_constants, true) if @dict.include?('$_OnRandomTalk') @random_talk = 0 end return 1 end def crypt(data) return data.chars.map {|c| (c.ord ^ 0xff).chr }.join('').force_encoding(@charset).encode("UTF-8", :invalid => :replace, :undef => :replace) end def eval_globals(sentences, constant) for sentence in sentences for node in sentence if node[0] == NODE_ASSIGNMENT eval_assignment(node[1], node[2], node[3], :constant => constant) elsif node[0] == NODE_FUNCTION eval_function(node[1], node[2]) end end end end def read(f, path: nil) parser = Parser.new(@charset) parser.read(f, :path => path) common, dic = parser.get_dict() variables = [] constants = [] for name, parameters, sentences in dic if sentences.nil? next elsif name == '$_Variable' variables |= sentences next elsif name == '$_Constant' constants |= sentences next end _GroupClass = Group conditions = [] conditions << common unless common.nil? for node in parameters if node[0] == NODE_IF conditions << node elsif node[0] == NODE_TEXT and node[1] == 'nonoverlap' _GroupClass = NonOverlapGroup elsif node[0] == NODE_TEXT and node[1] == 'sequential' _GroupClass = SequentialGroup else #pass # ignore unknown parameters end end group = _GroupClass.new(self, name, :item_list => sentences) if @dict.include?(name) grouplist = @dict[name] else grouplist = @dict[name] = [] end grouplist << [group, conditions] end return variables, constants end def strip_newline(line) line.chomp end def load_database begin open(@dbpath) do |f| while true begin line = f.gets rescue #except UnicodeError: Logging::Logging.debug( 'misaka.rb: malformed database (ignored)') break end break if line.nil? header = strip_newline(line).split(nil, -1) if header[0] == 'SCHOLAR' name = header[1] value = strip_newline(f.gets) @variable[name] = [TYPE_SCHOLAR, value] elsif header[0] == 'ARRAY' begin size = Integer(header[1]) rescue #except ValueError: Logging::Logging.debug( 'misaka.rb: malformed database (ignored)') break end name = header[2] array = MisakaArray.new(self, name) for _ in 0..size-1 value = strip_newline(f.gets) array << value end @variable[name] = [TYPE_ARRAY, array] else fail RuntimeError.new('should not reach here') end end end rescue #except IOError: return end end def save_database begin open(@dbpath, 'w') do |f| for name in @variable.keys value_type, value = @variable[name] if value_type == TYPE_SCHOLAR next if value == '' f.write('SCHOLAR ' + name.to_s + "\n" + value.to_s + "\n") elsif value_type == TYPE_ARRAY f.write('ARRAY ' + value.length.to_s + ' ' + name.to_s + "\n") for item in value f.write(expand(item).to_s + "\n") end else fail RuntimeError.new('should not reach here') end end end rescue #except IOError: Logging::Logging.debug('misaka.rb: cannot write database (ignored)') return end end def unload save_database() @saori_library.unload() return 1 end def reset_request @req_command = '' @req_protocol = '' @req_key = [] @req_header = {} end def request(req_string) header = req_string.split(/\r?\n/, 0) line = header.shift unless line.nil? line = line.force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace).strip() req_list = line.split(nil, -1) if req_list.length >= 2 @req_command = req_list[0].strip() @req_protocol = req_list[1].strip() end for line in header line = line.force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace).strip() next if line.empty? next unless line.include?(':') key, value = line.split(':', 2) key = key.strip() begin value = Integer(value.strip()) rescue value = value.strip() end @req_key << key @req_header[key] = value end end # FIXME ref0 = get_ref(0) ref1 = get_ref(1) ref2 = get_ref(2) ref3 = get_ref(3) ref4 = get_ref(4) ref5 = get_ref(5) ref6 = get_ref(6) ref7 = get_ref(7) event = @req_header['ID'] script = nil if event == 'otherghostname' n = 0 refs = [] while true ref = get_ref(n) unless ref.nil? refs << ref else break end n += 1 end @otherghost = [] for ref in refs name, s0, s1 = ref.split(1.chr, 3) @otherghost << [name, s0, s1] end end if event == 'OnMouseMove' key = [ref3, ref4] # side, part if @mouse_move_count.include?(key) count = @mouse_move_count[key] else count = 0 end @mouse_move_count[key] = count + 5 end if event == 'OnCommunicate' @communicate = [ref0, ref1] script = get('$_OnGhostCommunicateReceive', :default => nil) elsif event == 'charset' script = @charset else @reference = [ref0, ref1, ref2, ref3, ref4, ref5, ref6, ref7] script = get(['$', event].join(''), :default => nil) end if event == 'OnSecondChange' if @random_talk.zero? reset_random_talk_interval() elsif @random_talk > 0 @random_talk -= 1 if @random_talk.zero? script = get('$_OnRandomTalk', :default => nil) end end end unless script.nil? script = script.strip() else script = '' end reset_request() response = ["SHIORI/3.0 200 OK\r\n", "Sender: Misaka\r\n", "Charset: ", @charset.encode(@charset, :invalid => :replace, :undef => :replace), "\r\n", "Value: ", script.encode(@charset, :invalid => :replace, :undef => :replace), "\r\n"].join('') to = eval_variable([[NODE_TEXT, '$to']]) unless to.nil? @variable.delete('$to') response = [response, "Reference0: ", to.encode(@charset, :invalid => :replace, :undef => :replace), "\r\n"].join('') end response = [response, "\r\n"].join('') return response end def get_ref(num) key = ['Reference', num.to_s].join('') return @req_header[key] end # internal def reset_random_talk_interval interval = 0 if @variable.include?('$_talkinterval') value_type, value = @variable['$_talkinterval'] else value_type, value = TYPE_SCHOLAR, '' end if value_type == TYPE_SCHOLAR begin interval = [Integer(value), 0].max rescue #except ValueError: #pass end end @random_talk = (interval * Array(5..15).sample / 10.0).to_i end def get(name, default: '') grouplist = @dict[name] return default if grouplist.nil? for group, conditions in grouplist if eval_and_expr(conditions) == 'true' return expand(group.get()) end end return default end def expand(nodelist) return '' if nodelist.nil? buf = [] for node in nodelist buf << eval_(node) end return buf.join('') end def expand_args(nodelist) tmp = expand(nodelist) if tmp.start_with?('$') return eval_variable(nodelist) else return tmp end end def eval_(node) case node[0] when NODE_TEXT result = node[1] when NODE_STRING result = node[1] when NODE_VARIABLE result = eval_variable(node[1]) when NODE_INDEXED result = eval_indexed(node[1], node[2]) when NODE_ASSIGNMENT result = eval_assignment(node[1], node[2], node[3]) when NODE_FUNCTION result = eval_function(node[1], node[2]) when NODE_IF result = eval_if(node[1], node[2], node[3]) when NODE_CALC result = eval_calc(node[1]) when NODE_COMP_EXPR result = eval_comp_expr(node[1], node[2], node[3]) when NODE_AND_EXPR result = eval_and_expr(node[1..-1]) when NODE_OR_EXPR result = eval_or_expr(node[1..-1]) when NODE_ADD_EXPR result = eval_add_expr(node[1..-1]) when NODE_MUL_EXPR result = eval_mul_expr(node[1..-1]) when NODE_POW_EXPR result = eval_pow_expr(node[1..-1]) else ##print(node) fail RuntimeError.new('should not reach here') end return result end SYSTEM_VARIABLES = [ # current date and time '$year', '$month', '$day', '$hour', '$minute', '$second', # day of week (0 = Sunday) '$dayofweek', # ghost uptime '$elapsedhour', '$elapsedminute', '$elapsedsecond', # OS uptime '$elapsedhouros', '$elapsedminuteos', '$elapsedsecondos', # OS accumlative uptime '$elapsedhourtotal', '$elapsedminutetotal', '$elapsedsecondtotal', # system information '$os.version', '$os.name', '$os.phisicalmemorysize', '$os.freememorysize', '$os.totalmemorysize', '$cpu.vendorname', '$cpu.name', '$cpu.clockcycle', # number of days since the last network update '$daysfromlastupdate', # number of days since the first boot '$daysfromfirstboot', # name of the ghost with which it has communicated ##'$to', '$sender', '$lastsentence', '$otherghostlist', ] def eval_variable(name) now = Time.now name = expand(name) case name when '$year' result = now.year when '$month' result = now.month when '$day' result = now.day when '$hour' result = now.hour when '$minute' result = now.min when '$second' result = now.sec when '$dayofweek' result = now.wday when '$lastsentence' result = @communicate[1] when '$otherghostlist' unless @otherghost.empty? ghost = @otherghost.sample result = ghost[0] else result = '' end when '$sender' result = @communicate[0] when *SYSTEM_VARIABLES result = '' when *@dict.keys result = get(name) when *@constant.keys value_type, value = @constant[name] if value_type == TYPE_ARRAY result = expand(value.get()) else result = value.to_s end when *@variable.keys value_type, value = @variable[name] if value_type == TYPE_ARRAY result = expand(value.get()) else result = value.to_s end else result = '' end return result end def eval_indexed(name, index) name = expand(name) index = expand(index) begin index = Integer(index) rescue #except ValueError: index = 0 end case name when '$otherghostlist' group = [] for ghost in @otherghost group << ghost[0] end when *SYSTEM_VARIABLES return '' when *@dict.keys grouplist = @dict[name] group, constraints = grouplist[0] when *@constant.keys return '' when *@variable.keys value_type, group = @variable[name] return '' if value_type != TYPE_ARRAY else return '' end return '' if index < 0 or index >= group.length return expand(group[index]) end def eval_assignment(name, operator, value, constant: false) name = expand(name) value = expand(value) if ['+=', '-=', '*=', '/='].include?(operator) begin operand = Integer(value) rescue #except ValueError: operand = 0 end if @constant.include?(name) value_type, value = @constant[name] else value_type, value = nil, '' end if value_type.nil? if @variable.include?(name) value_type, value = @variable[name] else value_type, value = TYPE_SCHOLAR, '' end end return '' if value_type == TYPE_ARRAY begin value = Integer(value) rescue #except ValueError: value = 0 end if operator == '+=' value += operand elsif operator == '-=' value -= operand elsif operator == '*=' value *= operand elsif operator == '/=' and not operand.zero? value /= operand value = value.to_i end end if constant or @constant.include?(name) @constant[name] = [TYPE_SCHOLAR, value] else @variable[name] = [TYPE_SCHOLAR, value] end if name == '$_talkinterval' reset_random_talk_interval() end return '' end def eval_comp_expr(operand1, operator, operand2) value1 = expand(operand1) value2 = expand(operand2) begin operand1 = Integer(value1) operand2 = Integer(value2) rescue #except ValueError: operand1 = value1 operand2 = value2 end if (operator == '==' and operand1 == operand2) or \ (operator == '!=' and operand1 != operand2) or \ (operator == '<' and operand1 < operand2) or \ (operator == '<=' and operand1 <= operand2) or \ (operator == '>' and operand1 > operand2) or \ (operator == '>=' and operand1 >= operand2) return 'true' end return 'false' end def eval_and_expr(conditions) for condition in conditions boolean = eval_(condition) if boolean.strip() != 'true' return 'false' end end return 'true' end def eval_or_expr(conditions) for condition in conditions boolean = eval_(condition) if boolean.strip() == 'true' return 'true' end end return 'false' end def eval_add_expr(expression) value = expand(expression[0]) begin value = Integer(value) rescue #except ValueError: value = 0 end for i in 1.step(expression.length-1, 2) operand = expand(expression[i + 1]) begin operand = Integer(operand) rescue #except ValueError: operand = 0 end if expression[i] == '+' value += operand elsif expression[i] == '-' value -= operand end end return value.to_s end def eval_mul_expr(expression) value = expand(expression[0]) begin value = Integer(value) rescue #except ValueError: value = 0 end for i in 1.step(expression.length-1, 2) operand = expand(expression[i + 1]) begin operand = Integer(operand) rescue #except ValueError: operand = 0 end if expression[i] == '*' value *= operand elsif expression[i] == '/' and not operand.zero? value /= operand value = value.to_i elsif expression[i] == '%' and not operand.zero? value = (value % operand) end end return value.to_s end def eval_pow_expr(expression) value = expand(expression[-1]) begin value = Integer(value) rescue #except ValueError: value = 0 end for i in 1..expression.length-1 operand = expand(expression[-i - 1]) begin operand = Integer(operand) rescue #except ValueError: operand = 0 end value = (operand**value) end return value.to_s end def eval_if(condition, then_clause, else_clause) boolean = eval_(condition) if boolean.strip() == 'true' return expand(then_clause) else return expand(else_clause) end end def eval_calc(expression) expand(expression) end def eval_function(name, args) function = SYSTEM_FUNCTIONS[name] return '' if function.nil? return method(function).call(args) end def is_number(s) return (s and s.chars.map {|c| '0123456789'.include?(c) }.all?) end def split(s) buf = [] i, j = 0, s.length while i < j if s[i].ord < 0x80 buf << s[i] i += 1 else buf << s[i..i + 1] i += 2 end end return buf end def exec_reference(args) return '' if args.length != 1 n = expand_args(args[0]) if '01234567'.include?(n) value = @reference[n.to_i] return value.to_s unless value.nil? end return '' end def exec_random(args) return '' if args.length != 1 n = expand_args(args[0]) begin return Array(0..Integer(n)-1).sample.to_s rescue #except ValueError: return '' # n < 1 or not a number end end def exec_choice(args) return '' if args.empty? return expand_args(args.sample) end def exec_getvalue(args) return '' if args.length != 2 begin n = Integer(expand_args(args[1])) rescue #except ValueError: return '' end if n < 0 return '' end value_list = expand_args(args[0]).split(',', 0) begin return value_list[n] rescue #except IndexError: return '' end end def exec_search(args) namelist = [] for arg in args name = expand_args(arg) if name.strip() namelist << name end end return '' if namelist.empty? keylist = [] for key in keys() # dict, variable, constant break_flag = false for name in namelist unless key.include?(name) break_flag = true break end end unless break_flag keylist << key end end return '' if keylist.empty? return eval_variable([[NODE_TEXT, keylist.sample]]) end def keys buf = @dict.keys() buf |= @variable.keys() buf |= @constant.keys() return buf end def exec_backup(args) if args.empty? save_database() end return '' end def exec_getmousemovecount(args) if args.length == 2 side = expand_args(args[0]) part = expand_args(args[1]) begin key = [Integer(side), part] rescue #except ValueError: #pass else if @mouse_move_count.include?(key) return @mouse_move_count[key].to_s else return 0 end end end return '' end def exec_resetmousemovecount(args) if args.length == 2 side = expand_args(args[0]) part = expand_args(args[1]) begin key = [Integer(side), part] rescue #except ValueError: pass else @mouse_move_count[key] = 0 end end return '' end def exec_substring(args) return '' if args.length != 3 value = expand_args(args[2]) begin count = Integer(value) rescue #except ValueError: return '' end return '' if count < 0 value = expand_args(args[1]) begin offset = Integer(value) rescue #except ValueError: return '' end return '' if offset < 0 s = expand_args(args[0]).encode(@charset, :invalid => :replace, :undef => :replace) return s[offset..offset + count - 1].force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace) end def exec_substringl(args) return '' if args.length != 2 value = expand_args(args[1]) begin count = Integer(value) rescue #except ValueError: return '' end return '' if count < 0 s = expand_args(args[0]).encode(@charset, :invalid => :replace, :undef => :replace) return s[0..count-1].force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace) end def exec_substringr(args) return '' if args.length != 2 value = expand_args(args[1]) begin count = Integer(value) rescue #except ValueError: return '' end return '' if count < 0 s = expand_args(args[0]).encode(@charset, :invalid => :replace, :undef => :replace) return s[s.length - count..-1].force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace) end def exec_substringfirst(args) return '' if args.length != 1 buf = expand_args(args[0]) return '' if buf.empty? return buf[0] end def exec_substringlast(args) return '' if args.length != 1 buf = expand_args(args[0]) return '' if buf.empty? return buf[-1] end def exec_length(args) return '' if args.length != 1 return expand_args(args[0]).encode(@charset, :invalid => :replace, :undef => :replace).length.to_s end Katakana2hiragana = { 'ァ' => 'ぁ', 'ア' => 'あ', 'ィ' => 'ぃ', 'イ' => 'い', 'ゥ' => 'ぅ', 'ウ' => 'う', 'ェ' => 'ぇ', 'エ' => 'え', 'ォ' => 'ぉ', 'オ' => 'お', 'カ' => 'か', 'ガ' => 'が', 'キ' => 'き', 'ギ' => 'ぎ', 'ク' => 'く', 'グ' => 'ぐ', 'ケ' => 'け', 'ゲ' => 'げ', 'コ' => 'こ', 'ゴ' => 'ご', 'サ' => 'さ', 'ザ' => 'ざ', 'シ' => 'し', 'ジ' => 'じ', 'ス' => 'す', 'ズ' => 'ず', 'セ' => 'せ', 'ゼ' => 'ぜ', 'ソ' => 'そ', 'ゾ' => 'ぞ', 'タ' => 'た', 'ダ' => 'だ', 'チ' => 'ち', 'ヂ' => 'ぢ', 'ッ' => 'っ', 'ツ' => 'つ', 'ヅ' => 'づ', 'テ' => 'て', 'デ' => 'で', 'ト' => 'と', 'ド' => 'ど', 'ナ' => 'な', 'ニ' => 'に', 'ヌ' => 'ぬ', 'ネ' => 'ね', 'ノ' => 'の', 'ハ' => 'は', 'バ' => 'ば', 'パ' => 'ぱ', 'ヒ' => 'ひ', 'ビ' => 'び', 'ピ' => 'ぴ', 'フ' => 'ふ', 'ブ' => 'ぶ', 'プ' => 'ぷ', 'ヘ' => 'へ', 'ベ' => 'べ', 'ペ' => 'ぺ', 'ホ' => 'ほ', 'ボ' => 'ぼ', 'ポ' => 'ぽ', 'マ' => 'ま', 'ミ' => 'み', 'ム' => 'む', 'メ' => 'め', 'モ' => 'も', 'ャ' => 'ゃ', 'ヤ' => 'や', 'ュ' => 'ゅ', 'ユ' => 'ゆ', 'ョ' => 'ょ', 'ヨ' => 'よ', 'ラ' => 'ら', 'リ' => 'り', 'ル' => 'る', 'レ' => 'れ', 'ロ' => 'ろ', 'ヮ' => 'ゎ', 'ワ' => 'わ', 'ヰ' => 'ゐ', 'ヱ' => 'ゑ', 'ヲ' => 'を', 'ン' => 'ん', 'ヴ' => 'う゛', } def exec_hiraganacase(args) return '' if args.length != 1 buf = [] string = expand_args(args[0]).force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace) #string = args[0].force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace) for c in string.chars if Katakana2hiragana.include?(c) c = Katakana2hiragana[c] else #pass end buf << c end return buf.join('') end def exec_isequallastandfirst(args) return 'false' if args.length != 2 buf0 = expand_args(args[0]) buf1 = expand_args(args[1]) if not buf0.empty? and not buf1.empty? and buf0[-1] == buf1[0] return 'true' else return 'false' end end def exec_append(args) if args.length == 2 name = expand_args(args[0]) if not name.empty? and name.start_with?('$') if @variable.include?(name) value_type, value = @variable[name] else value_type, value = TYPE_SCHOLAR, '' end if value_type == TYPE_SCHOLAR value_type = TYPE_ARRAY array = MisakaArray.new(self, name) if value != '' array.append(value) end value = array end member_list = expand_args(args[1]).split(1.chr, 0) for member in member_list value.append(member) end @variable[name] = [value_type, value] end end return '' end def exec_stringexists(args) return '' if args.length != 2 name = expand_args(args[0]) if @variable.include?(name) value_type, value = @variable[name] else value_type, value = TYPE_SCHOLAR, '' end return '' if value_type == TYPE_SCHOLAR if value.index(expand_args(args[1])).nil? # value.index(expand_args(args[1])) < 0 return 'false' else return 'true' end end def exec_copy(args) return '' if args.length != 2 src_name = expand_args(args[0]) return '' unless not src_name.empty? and src_name.start_with?('$') new_name = expand_args(args[1]) return '' unless not new_name.empty? and new_name.start_with?('$') source = nil if @dict.include?(src_name) grouplist = @dict[src_name] source, conditions = grouplist[0] elsif @variable.include?(src_name) value_type, value = @variable[src_name] if value_type == TYPE_ARRAY source = value end end if source.nil? value = MisakaArray.new(self, new_name) else value = source.copy(new_name) end @variable[new_name] = [TYPE_ARRAY, value] return '' end def exec_pop(args) if args.length == 1 name = expand_args(args[0]) if @variable.include?(name) value_type, value = @variable[name] else value_type, value = TYPE_SCHOLAR, '' end if value_type == TYPE_ARRAY return expand(value.pop()) end end return '' end def exec_popmatchl(args) if args.length == 2 name = expand_args(args[0]) if @variable.include?(name) value_type, value = @variable[name] else value_type, value = TYPE_SCHOLAR, '' end if value_type == TYPE_ARRAY return expand(value.popmatchl(expand_args(args[1]))) end end return '' end def exec_index(args) if args.length == 2 pos = expand_args(args[1]).find(expand_args(args[0])) if pos > 0 s = expand_args(args[0])[pos] pos = s.encode(@charset, :invalid => :replace, :undef => :replace).length end return pos.to_s end return '' end def exec_insentence(args) return '' if args.empty? s = expand_args(args[0]) for i in 1..args.length-1 unless s.include?(expand_args(args[i])) return 'false' end end return 'true' end def exec_substringw(args) return '' if args.length != 3 value = expand_args(args[2]) begin count = Integer(value) rescue #except ValueError: return '' end if count < 0 return '' end value = expand_args(args[1]) begin offset = Integer(value) rescue #except ValueError: return '' end return '' if offset < 0 buf = expand_args(args[0]) return buf[offset..offset + count - 1] end def exec_substringwl(args) return '' if args.length != 2 value = expand_args(args[1]) begin count = Integer(value) rescue #except ValueError: return '' end return '' if count < 0 buf = expand_args(args[0]) return buf[0..count-1] end def exec_substringwr(args) return '' if args.length != 2 value = expand_args(args[1]) begin count = Integer(value) rescue #except ValueError: return '' end return '' if count < 0 buf = expand_args(args[0]) return buf[buf.length - count..-1] end Prefixes = ['さん', 'ちゃん', '君', 'くん', '様', 'さま', '氏', '先生'] def exec_adjustprefix(args) return '' if args.length != 2 s = expand_args(args[0]) for prefix in Prefixes if s.end_with?(prefix) s = [s[0..-prefix.length-1], expand_args(args[1])].join('') break end end return s end def exec_count(args) return '' if args.length != 1 name = expand_args(args[0]) begin value_type, value = @variable[name] if value_type == TYPE_SCHOLAR return '1' end return value.length.to_s rescue #except KeyError: return '-1' end end def exec_inlastsentence(args) return '' if args.empty? s = @communicate[1] for i in 1..args.length-1 unless s.include?(expand_args(args[i])) return 'false' end end return 'true' end def saori(args) saori_statuscode = '' saori_header = [] saori_value = {} saori_protocol = '' req = ["EXECUTE SAORI/1.0\r\n", "Sender: MISAKA\r\n", "SecurityLevel: local\r\n", "Charset: ", @charset.encode(@charset, :invalid => :replace, :undef => :replace), "\r\n"].join('') for i in 1..args.length-1 req = [req, 'Argument', i.to_s.encode(@charset, :invalid => :replace, :undef => :replace), ': ', expand_args(args[i]).encode(@charset, :invalid => :replace, :undef => :replace), "\r\n"].join('') end req = [req, "\r\n"].join('') response = @saori_library.request(expand_args(args[0]), req) header = response.split(/\r?\n/, 0) unless header.empty? line = header.shift line = line.force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace).strip() if line.include?(' ') saori_protocol, saori_statuscode = line.split(' ', 2) saori_protocol.strip! saori_statuscode.strip! end for line in header line = line.force_encoding(@charset).encode('utf-8', :invalid => :replace, :undef => :replace).strip() next if line.empty? nexy unless line.include?(':') key, value = line.split(':', 2) key.strip! value.strip! unless key.empty? saori_header << key saori_value[key] = value end end end if saori_value.include?('Result') return saori_value['Result'] else return '' end end def load_saori(args) return '' if args.empty? result = @saori_library.load(expand_args(args[0]), @misaka_dir) if result.zero? return '' else result.to_s end end def unload_saori(args) return '' if args.empty? result = @saori_library.unload(:name => expand_args(args[0])) if result.zero? return '' else return result.to_s end end def exec_isghostexists(args) result = 'false' if args.length != 1 if @otherghost.length > 0 result = 'true' @variable['$to'] = [TYPE_SCHOLAR, @otherghost.sample[0]] end else for ghost in @otherghost if ghost[0] == expand_args(args[0]) result = 'true' break end end end return result end end SYSTEM_FUNCTIONS = { '$reference' => 'exec_reference', '$random' => 'exec_random', '$choice' => 'exec_choice', '$getvalue' => 'exec_getvalue', '$inlastsentence' => 'exec_inlastsentence', '$isghostexists' => 'exec_isghostexists', '$search' => 'exec_search', '$backup' => 'exec_backup', '$getmousemovecount' => 'exec_getmousemovecount', '$resetmousemovecount' => 'exec_resetmousemovecount', '$substring' => 'exec_substring', '$substringl' => 'exec_substringl', '$substringr' => 'exec_substringr', '$substringfirst' => 'exec_substringfirst', '$substringlast' => 'exec_substringlast', '$length' => 'exec_length', '$hiraganacase' => 'exec_hiraganacase', '$isequallastandfirst' => 'exec_isequallastandfirst', '$append' => 'exec_append', '$stringexists' => 'exec_stringexists', '$copy' => 'exec_copy', '$pop' => 'exec_pop', '$popmatchl' => 'exec_popmatchl', '$index' => 'exec_index', '$insentence' => 'exec_insentence', '$substringw' => 'exec_substringw', '$substringwl' => 'exec_substringwl', '$substringwr' => 'exec_substringwr', '$adjustprefix' => 'exec_adjustprefix', '$count' => 'exec_count', '$saori' => 'saori', '$loadsaori' => 'load_saori', '$unloadsaori' => 'unload_saori', } class SaoriLibrary def initialize(saori, top_dir) @saori_list = {} @saori = saori end def load(name, top_dir) result = 0 head, name = File.split(name.gsub('\\', '/')) # XXX: don't encode here top_dir = File.join(top_dir, head) unless @saori.nil? or @saori_list.include?(name) module_ = @saori.request(name) unless module_.nil? @saori_list[name] = module_ end end if @saori_list.include?(name) result = @saori_list[name].load(:dir => top_dir) end return result end def unload(name: nil) unless name.nil? name = File.split(name.gsub('\\', '/'))[-1] # XXX: don't encode here if @saori_list.include?(name) @saori_list[name].unload() @saori_list.delete(name) end else for key in @saori_list.keys @saori_list[key].unload() end end return nil end def request(name, req) result = '' # FIXME name = File.split(name.gsub('\\', '/'))[-1] # XXX: don't encode here if not name.empty? and @saori_list.include?(name) result = @saori_list[name].request(req) end return result end end end