gnustep-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Incompatible compiler option fexec-charset


From: Richard Frith-Macdonald
Subject: Re: Incompatible compiler option fexec-charset
Date: Wed, 7 Dec 2011 15:57:52 +0000

On 7 Dec 2011, at 15:45, Wolfgang Lux wrote:

> David Chisnall wrote:
> 
>> On 7 Dec 2011, at 15:32, Richard Frith-Macdonald wrote:
>> 
>>> No .. the idea is to let them work in whatever environment they like 
>>> (latin-2 is a perfectly good example), but have the *binary*  they produce 
>>> contain UTF-8 encoded strings so that the running executable will display 
>>> the correct characters.
>> 
>> But that's not what you're testing.  Your source file contains some random 
>> encoding (latin-1, I believe).  The compiler will then interpret this as 
>> being whatever the current locale encoding is.  You are then specifying that 
>> the output is UTF-8.  So it is relying on a conversion from 
>> latin1-interpreted-as-something-random to UTF-8.  
> 
> That is not true. The configure script sets LANG and LC_ALL to 
> en_US.ISO-8859-1 before compiling the test, which is the canonical way to 
> select the source locale. So in this respect I think Richard's test is 
> correct.

Yes ... but your Ubuntu system seems to illustrate a problem ... if a system 
does not support the standard latin1 locale, the test won't work (though the 
fallback where we use -finput-charset=ISO-8859-1 should work for modern 
versions of gcc).   If something as common as Ubuntu doesn't have latin1 
support (presumably that's to do with the way it's configured) then we need 
some way to deal with that :-(


reply via email to

[Prev in Thread] Current Thread [Next in Thread]