[Bug locale/30645] New: gb18030.c GB18030-2022 Remove Part of Character Set Correspondence

[Bug locale/30645] New: gb18030.c GB18030-2022 Remove Part of Character Set Correspondence

[kiraskyler at 163 dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=30645

            Bug ID: 30645
           Summary: gb18030.c GB18030-2022 Remove Part of Character Set
                    Correspondence
           Product: glibc
           Version: unspecified
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: locale
          Assignee: unassigned at sourceware dot org
          Reporter: kiraskyler at 163 dot com
  Target Milestone: ---

Created attachment 14976
  --> https://sourceware.org/bugzilla/attachment.cgi?id=14976&action=edit
GB 18030 deleted 9 CJK compatible Chinese characters

![GB18030-2022](https://openstd.samr.gov.cn/bzgk/gb/newGbInfo?hcno=A1931A578FE14957104988029B0833D3)

The third edition of GB 18030 deleted 9 CJK compatible Chinese characters
included in the 2000 and 2005 editions. These 9 CJK compatible Chinese
characters have clear corresponding CJK unified Chinese characters

gb18030: fd9c -> ucs array index: 5d5c, ucs code: f92c
gb18030: fd9d -> ucs array index: 5d5d, ucs code: f979
gb18030: fd9e -> ucs array index: 5d5e, ucs code: f995
gb18030: fd9f -> ucs array index: 5d5f, ucs code: f9e7
gb18030: fda0 -> ucs array index: 5d60, ucs code: f9f1
gb18030: fe40 -> ucs array index: 5dc0, ucs code: fa0c
gb18030: fe41 -> ucs array index: 5dc1, ucs code: fa0d
gb18030: fe47 -> ucs array index: 5dc7, ucs code: fa18
gb18030: fe49 -> ucs array index: 5dc9, ucs code: fa20

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug locale/30645] gb18030.c GB18030-2022 Remove Part of Character Set Correspondence

[kiraskyler at 163 dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=30645

--- Comment #1 from kiraskyler <kiraskyler at 163 dot com> ---
files were affected
- glibc/iconvdata/gb18030.c __twobyte_to_ucs
- glibc/iconvdata/gbk.c __gbk_to_ucs

There is also a logic bug here

When idx=5dd1, it can be found in the array__ Twobyte_ To_ If the result is
found in UCS, it will not enter this if


- glibc/iconvdata/gb18030.c：24207
```
/* A two-byte character */                                    \
idx = (ch - 0x81) * 192 + (ch2 - 0x40);                       \
                                                                              \
if (idx > 0x5E7E                                                      \
  || (ch = __twobyte_to_ucs[idx],                             \
ch == 0 && *inptr != '\0'))                           \
{                                                             \
  /* Handle a few special cases.  */                          \
  if (idx == 0x5dd1)                                          \
    ch = 0x20087;                                             \
  else if (idx == 0x5dd2)                                     \
    ch = 0x20089;                                             \
  else if (idx == 0x5dd3)                                     \
    ch = 0x200cc;                                             \
  else if (idx == 0x5dec)                                     \
    ch = 0x215D7;                                             \
  else if (idx == 0x5df6)                                     \
    ch = 0x2298F;                                             \
  else if (idx == 0x5e11)                                     \
    ch = 0x241FE;                                             \
else                                                          \
{                                                             \
  /* This is an illegal character.  */                \
  STANDARD_FROM_LOOP_ERR_HANDLER (2);                 \
}
```                                                           \

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug locale/30645] gb18030.c GB18030-2022 Remove Part of Character Set Correspondence

[lijianglin2 at huawei dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=30645

lijianglin <lijianglin2 at huawei dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |lijianglin2 at huawei dot com

--- Comment #2 from lijianglin <lijianglin2 at huawei dot com> ---
For details, see https://sourceware.org/bugzilla/show_bug.cgi?id=30243

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug locale/30645] gb18030.c GB18030-2022 Remove Part of Character Set Correspondence

[kiraskyler at 163 dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=30645

--- Comment #3 from kiraskyler <kiraskyler at 163 dot com> ---
(In reply to lijianglin from comment #2)
> For details, see https://sourceware.org/bugzilla/show_bug.cgi?id=30243



This is not the same issue

-- 
You are receiving this mail because:
You are on the CC list for the bug.