1 2006-01-23 Atsushi Enomoto <atsushi@ximian.com>
3 * ISO2022JP.cs GB18030Encoding.cs CP936.cs CP51932.cs:
6 2006-01-23 Atsushi Enomoto <atsushi@ximian.com>
9 This was the last one encoding which needed incomplete byte
11 * DbcsEncoding.cs, GB18030Encoding.cs, CP950.cs, CP932.cs,
12 CP936.cs, CP51932.cs, CP949.cs (all):
13 Made GetCharCount() and GetChars() in DbcsDecoder as different
14 internal CheckRange() methods (it is bad design to have those
15 methods as base methods). Use CheckRange() everywhere.
17 2006-01-23 Atsushi Enomoto <atsushi@ximian.com>
20 handle incomplete byte (thus conversion implementation went into
21 CP949Decoder like other DBCS encodings).
23 2006-01-23 Atsushi Enomoto <atsushi@ximian.com>
26 handle incomplete byte (thus conversion implementation went into
27 CP936Decoder like other DBCS encodings).
29 2006-01-20 Atsushi Enomoto <atsushi@ximian.com>
32 Encoding.GetChars() does not handle incomplete bytes but
33 Decoder.GetChars() should do it (which was has vanished
34 when I fixed CP51932 4 days ago). Now it uses Decoder which
35 preserves incomplete byte like CP932 and CP54936.
37 2006-01-20 Atsushi Enomoto <atsushi@ximian.com>
40 Implemented Decoder to handle incomplete bytes correctly. Hold
41 different incomplete bytes for GetChars() and GetCharCount().
42 "reflesh" parameter in 2.0 is now handled correctly.
43 Now GetChars()/GetCharCount() in Encoding are implemented
44 correctly to use Decoder.
46 2006-01-20 Atsushi Enomoto <atsushi@ximian.com>
49 * GB18030Encoding.cs : GetEncoder() and GetDecoder() should be
50 overriden. in the Encoder, refresh parameter should handle
51 incomplete field at its end of processing. Also it should
52 differentiate GetByteCount() and GetChars() for its internal
54 * GB18030Source.cs : handle end index of GB18030Map as inclusive,
55 not exclusive. Reject out-of-range GBX conversion.
57 2006-01-20 Atsushi Enomoto <atsushi@ximian.com>
59 * GB18030Encoding.cs : GetChars() ditto. Fixed bug #77306.
61 2006-01-20 Atsushi Enomoto <atsushi@ximian.com>
63 * CP949.cs : GetChars() exactly the same fix as CP950.
66 2006-01-20 Atsushi Enomoto <atsushi@ximian.com>
68 * CP950.cs : GetChars() should filter out invalid index to internal
69 conversion array. Fixed bug #77274.
71 2006-01-17 Atsushi Enomoto <atsushi@ximian.com>
73 * DbcsEncoding.cs CP950.cs CP932.cs CP936.cs CP51932.cs CP949.cs :
74 Made all as really serializable. Removed references to DbcsConvert
75 and added GetConvert() methods instead.
77 2006-01-17 Atsushi Enomoto <atsushi@ximian.com>
79 * DbcsEncoding.cs ISO2022JP.cs GB18030Encoding.cs CP950.cs
80 CP932.cs CP936.cs CP51932.cs CP949.cs (All) :
81 Marked as [Serializable].
83 2006-01-16 Atsushi Enomoto <atsushi@ximian.com>
86 Implement GetCharCount(); it is too costy to use GetChars()
87 in GetCharCount(). Removed extra code in GetChars(). Check table
88 index range before accessing it by index.
90 2006-01-16 Atsushi Enomoto <atsushi@ximian.com>
93 In GetChars(), insufficient byte is not preserved.
94 Under 1.x profile it is just discarded.
95 Under 2.0 profile it is converted to \u30FB.
96 Fixed GetCharCount() to match GetChars().
98 2006-01-13 Atsushi Enomoto <atsushi@ximian.com>
100 * CP932.cs : GetByteCount() returned insufficient count at PrivateUse
101 area. Fixed bug #77724.
103 (Additionally, reduced extraneous indentation in switch-cases).
105 2006-01-12 Atsushi Enomoto <atsushi@ximian.com>
107 * CP51932.cs : oops, another extra debugging code was remaining.
110 2005-12-06 Atsushi Enomoto <atsushi@ximian.com>
112 Now I can rename new CP936_2.cs file to CP936.cs without svn diff mess.
114 2005-12-06 Atsushi Enomoto <atsushi@ximian.com>
116 * DbcsConvert.cs : added static Gb2312 field for common use.
117 * dbcs-table-generator.cs :
118 Added shortcut options for BIG5 and GB2312. Compressed data a bit.
119 * CP936.cs, CP936_2.cs : GB2312.TXT does not exist anymore on
120 unicode.org and it does not work like MS Codepage 936, so added
121 fresh implementation which is almost the same as CP950.cs.
122 CP936.cs is therefore removed.
123 * GB18030Encoding.cs :
124 switched implementation from Gb2312Convert to (base) DbcsConvert
126 * gb2312-build.sh : added note that we don't use it anymore.
128 with it CP936 can handles CP936-only codepoints as well.
129 * big5.table : compressed a bit.
130 * I18N.CJK.dll.sources : in the meantime, use CP936_2.cs
132 2005-12-06 Atsushi Enomoto <atsushi@ximian.com>
134 * renamed from big5-table-generator.cs to dbcs-table-generator.cs.
136 2005-12-06 Atsushi Enomoto <atsushi@ximian.com>
138 * big5-table-generator.cs : made it to work for GB2312 table
139 generation purpose (different format than gb2312.table.)
141 2005-12-06 Atsushi Enomoto <atsushi@ximian.com>
145 GB18030Source.cs : conversion from Encoding to MonoEncoding and
146 added fallback support in GEtBytes(Impl), as well as others.
148 2005-12-05 Atsushi Enomoto <atsushi@ximian.com>
150 * big5-table-generator.cs : new file to generate "big5.table".
151 * CP950.cs : added fallback support. switched to GetBytesImpl().
152 * DbcsEncoding.cs : switched to MonoEncoding (should have been
153 committed when CP949.cs was committed).
154 * big5.table : updated to match MS CP950.
156 2005-12-05 Atsushi Enomoto <atsushi@ximian.com>
158 * CP949.cs : migration from GetBytes (of Encoding) to GetBytesImpl
159 (of MonoEncoding). Added fallback support.
161 2005-12-05 Atsushi Enomoto <atsushi@ximian.com>
163 * I18N.CJK_test.dll.sources : new file for NUnit test.
164 * Makefile : removed NO_TEST.
166 2005-12-03 Atsushi Enomoto <atsushi@ximian.com>
169 GetCharCount() was returning excessive size. Now it is also based
170 on MonoEncoding, with fallback support.
172 2005-12-03 Atsushi Enomoto <atsushi@ximian.com>
174 * CP932.cs : oops, wrong range.
176 2005-12-03 Atsushi Enomoto <atsushi@ximian.com>
179 in GetBytes() \uFF01-\uFF60 are incorrectly mapped to ASCII
180 characters (they should be mapped to full-width characters).
181 Base type is now MonoEncoding, without adding fallback support.
183 2005-12-02 Atsushi Enomoto <atsushi@ximian.com>
186 Base type is now MonoEncoding (fallbacks are not handled yet).
187 Largely rewritten to handle SI/SO flag (CP50222)
188 and mode transition (all) correctly.
190 2005-11-21 Atsushi Enomoto <atsushi@ximian.com>
192 * CP949.cs : UHC level 2 base number is wrong.
193 Fixed bug #76760 Patch by Hye-Shik Chang.
195 2005-11-16 Atsushi Enomoto <atsushi@ximian.com>
197 * GB18030Encoding.cs : check GB2312 first, and then check GB18030.
198 Fixed GB18030 part of bug #76722.
200 2005-11-16 Atsushi Enomoto <atsushi@ximian.com>
202 * GB2312Convert.cs , CP936.cs, GB18030Encoding.cs : don't handle
203 single byte characters in UcsToGbk. Fixed GB2312 part of #76722.
205 2005-11-01 Atsushi Enomoto <atsushi@ximian.com>
207 * CP949.cs : Now it contains both CP949 (UHC) and CP51949 (EUC) for
208 Korean encoding. This should fix bug #76588.
210 2005-09-25 Atsushi Enomoto <atsushi@ximian.com>
212 * I18N.CJK.dll.sources : needs fix as well.
214 2005-09-25 Atsushi Enomoto <atsushi@ximian.com>
218 So, now "CP50221.cs" is improper, since it contains all 50220-50222.
220 2005-09-25 Atsushi Enomoto <atsushi@ximian.com>
222 * CP50221.cs : actually 50220 and 50222 are also iso-2022-jp encodings.
223 Fixed bug #76139. This file is likely to be renamed.
225 2005-09-18 Atsushi Enomoto <atsushi@ximian.com>
227 * CP50221.cs, GB18030Encoding.cs : because of silly design
228 Common/Manager expects fixed class names CPxxxxx. Fixed bug #76139.
230 2005-08-22 Raja R Harinath <rharinath@novell.com>
232 * Makefile (EXTRA_DISTFILES): Add README.gb18030 and gb18030.table.
234 2005-08-19 Atsushi Enomoto <atsushi@ximian.com>
236 why didn't we have it? ;-)
243 I18N.CJK.dll.sources : added support for GB18030 encoding.
244 * Gb2312Convert.cs : added utility conversion methods.