require "singleton"
require "chise/parser"
+require "chise/chisedb"
+require "chise/iconv"
+require "chise/utf8"
+require "chise/ids"
module CHISE
class CharacterFactory # generate Character object and cache them
@chars = {}
end
- def get(char_id)
+ def get(s)
check_max
- mcs = @parser.parse(char_id)
+ mcs = @parser.parse(s)
@chars[mcs] = Character.new(mcs) if @chars[mcs].nil?
@chars[mcs]
end
def check_max
- clear if MAX_CACHE_CHARACTER < @chars.length # clear all caches.
+ clear if MAX_CACHE_CHARACTER < @chars.length # clear all cache
end
end
class Character
- def initialize(char_id=nil)
+ include UTF8Value
+ include IDS_Module
+
+ def initialize(char_id)
+ raise if char_id.nil?
+ raise unless char_id.kind_of?(Integer) # make sure char_id is Integer.
+ raise if char_id < 0 # make sure char_id is positive.
@char_id = char_id
- @attributes = {}
- @check_all_database = false
+ @char_id.freeze
+ # @utf8_mcs = CHISE.i_tou8(@char_id)
+ @utf8_mcs = itou8(@char_id)
+ @utf8_mcs.freeze
+ @feature = {}
+ @check_all_done = nil
end
attr_reader :char_id
+ attr_reader :utf8_mcs
-
-
-
-
-
- def to_i() @char_id end
- def mcs_utf8() Character.u4itou8(@char_id) end
- def mcs_hex() sprintf("%x", @char_id) end
-
- def self.get(char_id) # flyweight pattern
- CharacterFactory.instance.get(char_id)
- end
-
- def normalize_attribute_name(b)
- a = b.dup
- a.gsub!(/_/, "-") #underline\82Í-\82É\92u\8a·
- a.sub!(/-at-/, "@")
- a.sub!(/^map-/, "=>")
- a.sub!(/^to-/, "->")
- a.sub!(/^from-/, "<-")
- a
+ def self.get(s)
+ CharacterFactory.instance.get(s)
end
- def get_char_attribute(b) # XEmacs CHISE compatible API
- a = normalize_attribute_name(b)
- #p [a, b]
- atr = @attributes[a]
- return atr if atr
- atr = check_database(a)
- if atr
- @attributes[a] = atr
- return atr
- end
- return get_char_attribute("="+a) unless a =~ /^=/
- #\93ª\82É=\82ª\82Â\82¢\82Ä\82È\82¢\8fê\8d\87\82Í\82»\82ê\82ª\8fÈ\97ª\82³\82ê\82Ä\82¢\82é\82±\82Æ\82ð\89¼\92è\82µ\82Ä\81A\8dÄ\8bA\82·\82é
- nil
+ def inspect
+ sprintf("Char:%x", @char_id)
end
- def put_char_attribute(b,v)
- a = normalize_attribute_name(b)
- @attributes[a] = v;
- CharDB.instance.put(a, mcs_utf8(), v)
- end
+ def to_s() @utf8_mcs; end
- def char_attribute_alist() check_all_database(); @attributes; end
- def char_attribute_list() check_all_database(); @attributes.keys; end
- alias [] get_char_attribute #\82»\82Ì\97ª\8fÌ
- alias []= put_char_attribute
- alias alist char_attribute_alist
- alias list char_attribute_list
+ def [](f)
+ f = normalize_feature_name(f)
- def method_missing(mid, *args) # ref. ostruct.rb
- mname = mid.id2name
- return get_char_attribute(mname) if args.length == 0
- put_char_attribute(mname.chop, args[0]) if mname =~ /=$/ #\91ã\93ü
- end
+ v = @feature[f]
+ return v if v
+ v = @feature["="+f]
+ return v if v
- def has_attribute?() # \88Ó\96¡\82Ì\82 \82éattribute\82ð\8e\9d\82Á\82Ä\82Ü\82·\82©?
- keys = list
- keys.delete_if {|k|
- k =~ /ids/
- }
- return (keys.length != 0)
- end
-
- def ==(ch)
- return false if ch.nil?
- return false unless ch.is_a? Character
- self.char_id == ch.char_id
- end
-
- def self.u4itou4(num)
- return "" unless num.is_a?(Integer)
- return sprintf("%c%c%c%c", num&0xff, (num >> 8)&0xff, (num >> 16)&0xff, (num >> 24)&0xff) #UCS-4\90\94\92l\82ð\95¶\8e\9a\97ñ\82É\82µ\82Äreturn
- end
-
- def self.u4itou8(char_id) #ucs\82Ì\90\94\92l\82ð\8eó\82¯\82Æ\82è\81AUTF-8\82Ì\95¶\8e\9a\88ê\95¶\8e\9a\82ð\95Ô\82·
- begin
- u4 = Character.u4itou4(char_id)
- u8 = Uconv.u4tou8(u4)
- return u8
- rescue
- #raise ArgumentError, "invalid char_id (#{char_id})", caller(1)
- #print "error\n"
- return ""
+ v = get_feature(f)
+ if v
+ @feature[f] = v
+ return v
end
- end
-
- def check_database(a)
- db = CharDB.instance
- u8 = mcs_utf8()
- v = db.get(a, u8) # u8\82Å\95\\82³\82ê\82é\95¶\8e\9a\82Ìa\83A\83g\83\8a\83r\83\85\81[\83g\82ð\92²\82×\82é\81B
- v
- end
-
- def check_all_database() # \8c»\8dÝ\82Ì@char_id\82©\82ç\81A\95¶\8e\9a\83f\81[\83^\83x\81[\83X\82ð\8eQ\8fÆ\82·\82é
- return if @check_all_database
- return if @char_id.nil?
- db = CharDB.instance
- u8 = mcs_utf8()
- atrs = db.get_all(u8) #u8\82Å\95\\82³\82ê\82é\95¶\8e\9a\82Ì\83A\83g\83\8a\83r\83\85\81[\83g\82ð\91S\95\94\8e\9d\82Á\82Ä\82±\82¢
- atrs.each {|a,v|
- @attributes[a] = v #\82Æ\82©\82¢\82¤\8a´\82¶\82Å\91ã\93ü\82·\82é\82Ì\82Å\82¦\82¦\82©\82È?
- }
- @check_all_database = true #\8fd\82¢\8f\88\97\9d\82È\82Ì\82Å\88ê\89\9echeck\82·\82é
- end
-
- def ucs()
- #p "ucs"
- #ar = %w{ucs ucs-big5 ucs-cdp ucs-cns ucs-jis ucs-ks =>ucs =>ucs* =>ucs-jis}
- #ar = %w{ucs ucs-jis ucs-big5 ucs-cdp ucs-cns ucs-ks =>ucs =>ucs* =>ucs-jis}
- ar = %w{ucs-jis ucs =>ucs-jis}
- #\95À\82Ñ\8f\87\82Í\9c\93\88Ó\93I\82Å\81Aucs-jis\82ð\90æ\82É\8fo\82µ\82Ä\82¢\82é\81B\96{\97\88\82Í\82±\82ê\82à\8ew\92è\82Å\82«\82é\82æ\82¤\82É\82·\82é\82×\82«\81B
- ar.each {|a| #p [a]
- u = get_char_attribute(a)
- return u if u
- }
- nil
- end
-
- #-------------------------------------------------------------------CCS\8aÖ\8cW
- def to_utf8() Uconv.u4tou8(Character.u4itou4(ucs())) end #UTF8\95¶\8e\9a\97ñ\82ð\95Ô\82·
- #alias to_s to_utf8
- alias to_s mcs_utf8
- def map_utf8()
- u = ucs()
- if u.nil? || 0xffff < u
- return to_er()
- else
- return to_utf8()
+ v = get_feature("="+f)
+ if v
+ @feature["="+f] = v
+ return v
end
- end
- alias map_ucs map_utf8
-
- def map_ucs_er()
- u = ucs()
- if u.nil? || 0xffff < u
- return to_er()
- else
- return Character.get(u).to_er()
- end
- end
-
- def to_euc()
- u = ucs()
- return "" if u.nil? || 0xffff < u
- Uconv.u16toeuc(Uconv.u4tou16(Character.u4itou4(ucs())))
- end
-
- def map_euc()
- e = to_euc()
- return e if e != ""
- return to_er()
- end
- def to_sjis()
- u = ucs()
- return "" if u.nil? || 0xffff < u
- Uconv.u16tosjis(Uconv.u4tou16(Character.u4itou4(ucs())))
- end
-
- def map_sjis()
- e = to_sjis()
- return e if e != ""
- return to_er()
- end
-
- def to_er(codesys=nil) #\8eÀ\91Ì\8eQ\8fÆ\82ð\95Ô\82·\81A\8aó\96]\82·\82écodesys\82ª\88ø\90\94(\96¢\8eÀ\91\95)
- return "" if @char_id.nil?
- return sprintf("&#x%04x;", @char_id) if @char_id <= 0xffff
- return sprintf("&#x%05x;", @char_id) if @char_id <= 0xfffff
- EntityReference.each_codesys {|codesys, er_prefix, keta, numtype|
- code = self[codesys]
- next if code.nil?
- return sprintf("&#{er_prefix}%0#{keta}#{numtype};", code)
- }
- return sprintf("&MCS-%08X;", @char_id) #\96{\93\96\82Í\82±\82ê\82Í\96³\82µ\82É\82µ\82½\82¢
+ nil
end
- def to_er_list()
- ar = []
- EntityReference.each_codesys {|codesys, er_prefix, keta, numtype|
- er = to_er(codesys)
- ar << er if er
- }
- ar
+ def []=(k,v)
+ f = normalize_feature_name(k)
+ cd = ChiseDB.instance
+ ft = cd.get_feature(f)
+ ft.set_value(@char_id, v)
+ @feature[f] = v;
end
- def inspect_x()
- return "<>" if @char_id.nil?
- ar = [to_utf8(), to_er().sub(/^&/,"").chop]
- "<"+ar.join(",")+">"
- end
- alias inspect inspect_x
+ def method_missing(mid, *args) # ref. ostruct.rb
+ mname = mid.id2name
- def inspect_all_codesys() #\96¢\8a®\90¬
- #to_er\82ð\91S\82Ä\82Ìcodesys\82É\82¨\82¢\82Ä\8eÀ\8ds\82·\82é\81B\82»\82Ì\8c\8b\89Ê\82ð\83R\83\93\83p\83N\83g\82É\82Ü\82Æ\82ß\82é
- end
+ return self[mname] if args.empty? # get
- def inspect_all()
- ar = [inspect.chop]
- alist.to_a.sort.each {|a, v| ar << "#{a}:#{v}" }
- return ar.join(",")+">"
- end
+ if args.length == 1 && /=\Z/ =~ mname # put
+ self[mname.chop] = args.shift
+ return
+ end
- def dump_all()
- ar = [inspect]
- alist.to_a.sort.each {|a, v| ar << "#{a}:#{v}" }
- return ar.join('\n')+'\n'
+ raise "error"
end
- def get_attributes()
- str = ""
- alist.to_a.sort.each {|a, v|
- str += "#{a}: #{v}\n"
- }
- str
+ def to_er
+ en = EntityReferenceEncoder.new
+ en.to_er(self)
end
- def inspect_ids(hex_flag=false)
- ids = decompose
- ar = []
- ar << (hex_flag ? "x"+mcs_hex : to_utf8)
- if to_s != ids #ids\82ª\95\94\95i\82»\82Ì\82à\82Ì\82¾\82Á\82½\82ç\95\94\95i\92Ç\89Á\82Í\82µ\82È\82¢
- ids.each_char {|ch|
- char = ch.char
- next if char.is_ids?
- if hex_flag then
- ar << "x"+char.mcs_hex
- else
- u = char.to_utf8
- if u != ""
- ar << u
- else
- ar << char.to_er
- end
- end
- }
- end
- return "("+ar.join("\t")+")"
+ def is_idc?
+ 0x2ff0 <= @char_id && @char_id <= 0x2fff
end
- #--------------------------------------------------------------------IDS\8aÖ\8cW
- def glyph_decompose() do_decompose(false) end
- def decompose() do_decompose(true) end
- def do_decompose(check_meaning = true)
- k = self.to_s
- # idss = self["ids"]
- # return idss if idss
- # return k if self.is_basic_kanji? #\8aî\96{\8a¿\8e\9a\82Ístop kanji\82Æ\82·\82é\82¼\82Æ\81B
- if check_meaning
- return self["ids-represent"] if self["ids-represent"] #ids_represent\82ð\8e\9d\82Á\82Ä\82¢\82é\8fê\8d\87\82Í\82»\82Ì\92l\82Æ\82·\82é\81B
- return self["ids-element"] if self["ids-element"] #ids_element\82ð\8e\9d\82Á\82Ä\82¢\82é\8fê\8d\87\82Í\82»\82Ì\92l\82Æ\82·\82é\81B
- idss = self["ids-meaning"]
- return idss if idss && 0 < idss.length && k != idss
- end
- idss = self["ids-aggregated"]
- return idss if idss && 0 < idss.length && k != idss
- idss = self["ids"]
- return idss if idss && 0 < idss.length && k != idss
- return k
- # return k if idss.nil? || idss.length == 0 || k == idss
- # if idss.char_length == 2
- # p ["What???", k, idss, k.inspect_all]
- # #return idssx[1] #\93ñ\8cÂ\96Ú\82¾\82¯\95Ô\82·\82Æ\82©?
- # return k #IDS\82É\93W\8aJ\82·\82é\95û\96@\82ª\96³\82¢\82Æ\81B
- # end
- # return k if k == idss
- # if idss.include?(k) #<C5-4C4D><C6-4A37>\82±\82Ì\93ñ\95¶\8e\9a\82ÌBUG\91Î\8dô
- # #return idss.sub(k, "")
- # return k #IDS\82É\93W\8aJ\82·\82é\95û\96@\82ª\96³\82¢\82Æ\81B
- # end
- # return idss
+ def idc_argument_number
+ return 0 unless is_idc?
+ return 3 if @char_id == 0x2ff2 || @char_id == 0x2ff3
+ return 2
end
- def decompose_all
- pde = ""
- de = self.decompose #\8fo\94\93_
- level = 0
- while true
- pde = de
- de = pde.decompose #\82à\82¤\88ê\93x\95ª\89ð\82ð\82µ\82Ä\82Ý\82é\81B
- break if pde == de #\83\8b\81[\83v\82ð\94²\82¯\82¾\82·
- exit if 10 < level #p ["too many recursive", self]
- level += 1
- end
- return de
- end
+ private
- def decompose_all_nu(level=nil)
- level = 0 if level.nil?
- if 10 < level
- p ["too many recursive", self]
- exit
- end
- de = self.decompose
- return de.decompose_all(level+1) if de != self #\82È\82É\82©\95Ï\89»\82ª\82 \82Á\82½\82©\82ç\8dÄ\8bA
- return de #\82à\82¤\82±\82ê\88È\8fã\95Ï\89»\82Í\96³\82³\82»\82¤\82¾\82¼\82Æ\81B
+ def get_feature(f)
+ cd = ChiseDB.instance
+ cd.load_feature(f, @char_id)
end
- def is_ids?() 0x2ff0 <= @char_id && @char_id <= 0x2fff end
-
- def ids_operator_argc()
- return 0 unless is_ids?
- return 3 if @char_id == 0x2ff2 || @char_id == 0x2ff3
- return 2
+ def normalize_feature_name(a)
+ a = a.gsub(/_/, "-") #underlineは-に置換
+ a = a.sub(/-at-/, "@")
+ a = a.sub(/-plus-/, "+")
+ a = a.sub(/\Amap-/, "=>")
+ a = a.sub(/\Ato-/, "->")
+ a = a.sub(/\Afrom-/, "<-")
+ a
end
end
end