[BlueOnyx:15080] Re: YUM updates for 5106R/5107R/5108R

[Michael Stauber]

Hi Hisao,

> But, as you know with original codes, Japanese use entered through GUI
> stored into CODB using EUC-JP. So I changed to UTF-8 when I released
> at major version up.

Yeah, I had seen your 5200R code where you made that transition. But I
might have missed the part where you changed the CODB-submit from EUC-JP
to UTF-8.

As it is at the moment Japanese text such as this ...

私は休暇中だ。後で私に連絡してください。

... gets stored by CODB as this:

get 169 . Email
[...]
102 DATA vacationMsg =
"\347\247\201\343\201\257\344\274\221\346\232\207\344\270\255\343\201\240\343\200\202\345\276\214\343\201\247\347\247\201\343\201\253\351\200\243\347\265\241\343\201\227\343\201\246\343\201\217\343\201\240\343\201\225\343\201\204\343\200\202"
[...]

I'm not even sure what charset that might relate to.

> If you are OK, I’ll check it out and take care of this with you.

That would be much appreciated, Hisao.

During this last update I took the approach of editing
/usr/sausalito/ui/libPhp/I18n.php.

The functions get(), getHml(), getJs() (and the related interpolate
functions) now convert all locales into UTF-8. Regardless of the
encoding they were in before.

There is also a new function called I18n::Utf8Encode() that can be used
to turn strings into UTF-8.

However, I did not change how data is stored in CODB. Which might
contribute to the problem.

-- 
With best regards

Michael Stauber