[KoCo-CVS] [Commit] KoreanCodecs/korean/python iso_2022_kr.py
Brought to you by:
perky
Menu
▾
▴
[KoCo-CVS] [Commit] KoreanCodecs/korean/python iso_2022_kr.py
|
From: Hye-Shik C. <pe...@us...> - 2003-01-10 06:09:31
|
perky 03/01/09 22:09:31
Modified: korean/python iso_2022_kr.py
Log:
Change to reimplemented code that can handle ksc5601 designated on G0 area.
Revision Changes Path
1.13 +52 -82 KoreanCodecs/korean/python/iso_2022_kr.py
Index: iso_2022_kr.py
===================================================================
RCS file: /cvsroot/koco/KoreanCodecs/korean/python/iso_2022_kr.py,v
retrieving revision 1.12
retrieving revision 1.13
diff -u -r1.12 -r1.13
--- iso_2022_kr.py 9 Jan 2003 21:35:49 -0000 1.12
+++ iso_2022_kr.py 10 Jan 2003 06:09:31 -0000 1.13
@@ -17,42 +17,40 @@
# along with KoreanCodecs; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
#
-# $Id: iso_2022_kr.py,v 1.12 2003/01/09 21:35:49 perky Exp $
+# $Id: iso_2022_kr.py,v 1.13 2003/01/10 06:09:31 perky Exp $
#
import codecs
-from korean.mappings import ksc5601_hangul
-encmap_hangul = ksc5601_hangul.encoding_map
-decmap_hangul = ksc5601_hangul.decoding_map
-encmap_ideo, decmap_ideo = {}, {}
-encmap_misc, decmap_misc = {}, {}
-US_ASCII = 1
-KSC5601_1987 = 2
+KSC5601_CODEC = 'korean.euc-kr'
+
+US_ASCII = 1
+KSC5601_1987 = 2
+
+G0, G1 = 0, 1 # iso-2022-kr doesn't handle G2 and G3 area.
CHARSETS = {
- "\033(B": US_ASCII,
- "\033$)C": KSC5601_1987,
+ "\033(B": (G0, US_ASCII),
+ "\033)B": (G1, US_ASCII),
+ "\033$(C": (G0, KSC5601_1987),
+ "\033$)C": (G1, KSC5601_1987),
}
SI = '\x0f'
SO = '\x0e'
ESC = '\033'
-DESIGNATIONS = {}
+DESIGNATION_MARK = {}
for k, v in CHARSETS.items():
- DESIGNATIONS[v] = k
-
-# StreamReader was adopted from Tamito KAJIYAMA's iso-2022-jp codec.
+ DESIGNATION_MARK[v] = k
class Codec(codecs.Codec):
# Unicode to character buffer
def encode(self, data, errors='strict'):
- global encmap_ideo, encmap_misc
-
if errors not in ('strict', 'ignore', 'replace'):
raise ValueError, "unknown error handling"
buffer = []
- new_charset = charset = US_ASCII
+ designation = [US_ASCII, US_ASCII]
+ new_designation = designation[:]
new_shiftstate = shiftstate = 0
for c in data:
if c in ('\n', '\r'):
@@ -60,49 +58,30 @@
if c < u'\u0080':
new_shiftstate = 0
+ new_designation[0] = US_ASCII
s = c.encode("ascii", errors)
- elif encmap_hangul.has_key(c):
- new_charset = KSC5601_1987
- new_shiftstate = 1
- s = encmap_hangul[c]
else:
- if not encmap_misc:
- from korean.mappings import ksc5601_misc
- encmap_misc = ksc5601_misc.encoding_map
- if encmap_misc.has_key(c):
- new_charset = KSC5601_1987
- new_shiftstate = 1
- s = encmap_misc[c]
- else:
- if not encmap_ideo:
- from korean.mappings import ksc5601_ideograph
- encmap_ideo = ksc5601_ideograph.encoding_map
- if encmap_ideo.has_key(c):
- new_charset = KSC5601_1987
- new_shiftstate = 1
- s = encmap_ideo[c]
- elif errors == 'replace':
- new_charset = KSC5601_1987
- new_shiftstate = 1
- s = '\xa1\xa1'
- elif errors == 'strict':
- raise UnicodeError, "cannot map \\u%04x to ISO-2022-KR" % ord(c)
- else:
- continue
-
- if charset != new_charset:
- charset = new_charset
- buffer.append(DESIGNATIONS[charset])
- if new_shiftstate != shiftstate:
+ new_shiftstate = 1
+ new_designation[1] = KSC5601_1987
+ s = c.encode('korean.euc_kr', errors)
+
+ if designation[0] != new_designation[0]:
+ buffer.append(DESIGNATION_MARK[(G0, new_designation[0])])
+ designation[0] = new_designation[0]
+ if designation[1] != new_designation[1]:
+ buffer.append(DESIGNATION_MARK[(G1, new_designation[1])])
+ designation[1] = new_designation[1]
+ if shiftstate != new_shiftstate:
+ buffer.append([SI, SO][new_shiftstate])
shiftstate = new_shiftstate
- buffer.append([SI, SO][shiftstate])
if shiftstate:
s = chr(ord(s[0])&0x7F) + chr(ord(s[1])&0x7F)
buffer.append(s)
+
if shiftstate:
buffer.append(SI)
- #buffer.append(DESIGNATIONS[US_ASCII])
+
return (''.join(buffer), len(data))
# character buffer to Unicode
@@ -114,10 +93,12 @@
buffer = []
data = str(data) # character buffer compatible object
size = len(data)
- charset = US_ASCII
- shiftstate = 0
- escstart = -1
- p = 0
+
+ designation = [US_ASCII, KSC5601_1987]
+ shiftstate = 0
+ escstart = -1
+ p = 0
+
while p < size:
if data[p] in ('\n', '\r'):
shiftstate = 0
@@ -127,8 +108,10 @@
escstr = data[escstart:p+1]
if CHARSETS.has_key(escstr):
charset = CHARSETS[escstr]
+ designation[charset[0]] = charset[1]
elif errors == 'strict':
- raise UnicodeError, "unsupported charset found: %s" % repr(data[escstart:p+1])
+ raise UnicodeError, "unsupported charset found: " \
+ + repr(data[escstart:p+1])
escstart = -1
p += 1
elif data[p] == SO:
@@ -141,39 +124,24 @@
escstart = p
p += 1
else:
- if not shiftstate and (
- charset == US_ASCII or data[p] < '\x80'): # ascii
+ if (ord(data[p]) | (shiftstate and 0x80 or 0x00)) >= 0x80:
+ codearea = G1
+ else:
+ codearea = G0
+
+ if designation[codearea] == US_ASCII:
buffer.append(unicode(data[p], "ascii", errors))
p += 1
- else:
+ elif ord(data[p]) & 0x7F >= 0x20: # KSC5601_1987
c = data[p:p+2]
p += 2
if len(c) == 2:
c = chr(ord(c[0])|0x80) + chr(ord(c[1])|0x80)
- if decmap_hangul.has_key(c):
- buffer.append(decmap_hangul[c])
- continue
-
- if not decmap_misc:
- from korean.mappings import ksc5601_misc
- decmap_misc = ksc5601_misc.decoding_map
- if decmap_misc.has_key(c):
- buffer.append(decmap_misc[c])
- continue
-
- if not decmap_ideo:
- from korean.mappings import ksc5601_ideograph
- decmap_ideo = ksc5601_ideograph.decoding_map
- if decmap_ideo.has_key(c):
- buffer.append(decmap_ideo[c])
- continue
+ buffer.append(unicode(c, KSC5601_CODEC, errors))
+ else: # control characters
+ buffer.append(unichr(ord(data[p]) & 0x7F))
+ p += 1
- if errors == 'replace':
- buffer.append(u'\uFFFD') # REPLACEMENT CHARACTER
- elif errors == 'strict':
- raise UnicodeError, "unexpected byte 0x%02x%02x found" % tuple(map(ord, c))
- # XXX: only 1byte?
-
return (u''.join(buffer), len(data))
class StreamWriter(Codec, codecs.StreamWriter):
@@ -188,3 +156,5 @@
def getregentry():
return (Codec().encode,Codec().decode,StreamReader,StreamWriter)
+
+# ex: ts=8 sts=4 et
|
