Re: [Zh-l10n] Fwd: Bad alignment in CJK and Farsi translations of hello
by Vladimir 'φ-coder/phcoder' Serbinenko
On 29.03.2012 04:41, KanRu Chen wrote:
> Hi,
>
> I think this should be handled by the program not the translator?
> Because the translator won't be able to know where the translations will
> be displayed.
Yes, I'd agree but I'm not involved in project in question so I'd
suggest to take the issue to bug-hello(a)gnu.org
>
> kanru
>
> "Vladimir 'φ-coder/phcoder' Serbinenko" <phcoder(a)gmail.com> writes:
>
>> Japanese and both variants of Chinese translations don't consider the
>> fact that CJK characters take 2 columns:
>> phcoder@debian.x201.phnet:00:37:15:~/compile$ LANGUAGE=ja hello -n
>> ┏━━━━━━━━━━┓
>> ┃世界よ、こんにちは!┃
>> ┗━━━━━━━━━━┛
>> phcoder@debian.x201.phnet:00:41:31:~/compile$ LANGUAGE=zh_CN hello -n
>> ┌─────┐
>> │世界你好!│
>> └─────┘
>> phcoder@debian.x201.phnet:00:43:54:~/compile$ LANGUAGE=zh_TW hello -n
>> ┌───────┐
>> │哈囉,大家好!│
>> └───────┘
>>
>> Farsi case seems to be related to the number of columns occupied by lam-alif ligature (you need one ═ more):
>> phcoder@debian.x201.phnet:00:42:26:~$ LANGUAGE=fa hello -n
>> ╔═════════════╗
>> ║ سلام بر همه! ║
>> ╚═════════════╝
>>
>> To understand copy-pastes use monospace font.
--
Regards
Vladimir 'φ-coder/phcoder' Serbinenko