[BlueOnyx:17646] Re: Q: 5208R with Japanese Character

[Eiji Hamano]

Hi Michael

1.  The "command mode" ;

  For exsample, please look at the following. We login 5208R by SSH
  with UTF-8 mode.
     http://www.hypersys.ne.jp/support/5208R/5208R_SSH_with_UTF8.jpg

  And please look at the other. We login it with EUC mode.
     http://www.hypersys.ne.jp/support/5208R/5208R_SSH_with_EUC.jpg


2.  The login page with Japanese lang ;

  Thank you!

Eiji Hamano



Michael Stauber wrote:
> Hi Eiji,
>
>> 1.  Michael was changed Japanese code of 5106-8R from the EUC to UTF8
>>      with very struggling.
>>      However, at 5208R, the Japanese command mode was back to EUC again.
>>
>>      Why?  Is it because the system was made with other than you ?
>
> Sorry, but I do not understand what you mean with "command mode". That
> probably did not translate correctly into English.
>
> As for EUC and UTF-8? It is like this:
>
> UTF-8 was in its infancy when the initial GUI was created. Back then the
> guys at Cobalt had to use ISO-8859-1 for English and EUC for Japanese.
>
> Since then a lot has changed and UTF-8 can now be used for pretty much
> all languages which are relevant to us. Including Japanese. This
> required a lot of changes in the GUI code. Additionally all language
> files had to be converted from ISO-8859-1 to UTF-8 and the Japanese
> language files had to be converted from EUC-JP to UTF-8.
>
> With that done we can consistently use UTF-8 everywhere, which makes
> code maintenance a lot simpler.
>
>> 2.  The 5208R's login page was broken with Japanese lang mode.
>>
>>    Please see: http://www.hypersys.ne.jp/support/5208R/5208R_login.jpg
>
> OK, I see the problem. Apparently the word for "Login" in Japanese is
> incorrectly displayed. I will take a look at this and will publish a fix
> within the next day or two.
>
> Thank you for bringing this to my attention.
>
>> But only with chrome 42.0 it has no problem!  Hmm..?
>
> It may be that Chrome has some mechanism that is more tolerant with
> incorrect encodings in pages.
>
> Anyway: Many thanks again and I'll publish a fix for this.
>