Rewrite of IBM doublebyte charsets (original) (raw)

Ulf Zibis Ulf.Zibis at gmx.de
Thu May 21 23:57:05 UTC 2009

Am 22.05.2009 00:30, Xueming Shen schrieb:

Ulf Zibis wrote:

In IBM933.map you state:

# Warning: # (2) we "should" have an entry # 25 000a # in IBM933.nr (b->c only tables) as other ebcdic # mappings do, but the "old" implementation actually # maps \u000a to 25. Keep it old behavior for now. I think we shouldn't stick on old behaviour here, as there obviously was an error in old code, but nobody filed a bug until now. It's a tough call, and I always try to avoid the touch call:-) I believe this brain-damage 0x15, 0x25 -> 000a -> 0x15, 0085->0x15 mapping is the result of the "fix" we made for #4159519, as the workaround solution for the "what is the real new line on ebcdic system" problem, while none of the official ebcdic<->unicode mapping tables from IBM/MSFT do NOT have this hack documented/recorded at all. So we might want to re-check the soundness of this fix, which was made 10 years ago, sometime.

So IBM933 mapping or all other IBMxxx mappings should be corrected in future ?

-Ulf

More information about the core-libs-dev mailing list