Re: Adopt a platform today!

[David Chisnall]

On 21 Feb 2011, at 22:39, Riccardo Mottola wrote:

> Hi,
> 
> ON FreeBSD/ia32
> 
> 5478 Passed tests
>  18 Failed tests
>   5 Skipped sets
>   1 Dashed hope

On the same platform with ICU installed, I get:

5589 Passed tests
   3 Failed tests
   3 Dashed hopes
   1 Skipped set

The skipped test appears to be caused by a bug in configure finding OpenSSL and 
then refusing to use it.

The failures are:

--- Running tests in base/NSCalendarDate ---

test01.m:
Failed test:     test01.m:45 ... 12:00am is midnight

This one is weird - it parses 12:00am as 12:00pm.  Not especially interested in 
spending time fixing this because NSCalendarDate is a deprecated API and we now 
have implementations of the more sane classes that replace it (many thanks 
again to Stef for that).

--- Running tests in base/NSTimeZone ---

use.m:
Failed test:     use.m:58 ... Correctly localizes DST time zone name
Expected 'Brasilia Summer Time' and got 'Brasilia Time'
Failed test:     use.m:62 ... Correctly localizes short DST time zone name
Expected 'GMT-02:00' and got 'GMT-03:00'

This looks like this is an ICU bug.  It seems to be ignoring the DST part of 
the request.  If I change America/Sao_Paulo to Europe/London, then it returns 
the correct values (the tests fail, because they're expecting the San Paulo 
versions).  Does this pass for other people?  It's slightly weird, because the 
subsequent tests (which check the daylight saving offset for this time zone) 
are all passing.

Upgrading to ICU 4.6 (I was on 4.4), I now get three fails here:
Failed test:     use.m:54 ... Correctly localizes short time zone name
Expected 'GMT-03:00' and got 'BRT'
Failed test:     use.m:58 ... Correctly localizes DST time zone name
Expected 'Brasilia Summer Time' and got 'Brasilia Time'
Failed test:     use.m:62 ... Correctly localizes short DST time zone name
Expected 'GMT-02:00' and got 'BRT'

If I change this to Europe/London again, the two that expect GMT offsets now 
still get the short name (GMT / BST), but it does correctly respect DST, so I 
think this failure is specific to Brasilia Time.

One thing I noticed is that a lot of the new ICU code is using variable names 
that start with underscores.  The C standard specifically reserves identifiers 
that start with an underscore for the implementation, and any C compiler is 
completely at liberty to reject them or interpret them as something completely 
random without deviating from the spec, so we really should be using them in 
GNUstep, and especially not for things that are not exported as symbols.

This includes instance variables, by the way - we use them in quite a few 
places, which is very bad and something that we should avoid.  If something 
needs to avoid conflicts with other code, use gs_ as the prefix, not _.

David

> 
> 
> the failures appear all date & number formatter and timezone based. I have 
> icu 3.8 installed, so I guess it is too old and it is like not having it.
> 
> Riccardo
> 
> _______________________________________________
> Discuss-gnustep mailing list
> Discuss-gnustep@gnu.org
> http://lists.gnu.org/mailman/listinfo/discuss-gnustep