Re: [PATCH COMMITTED] Switch IDNA implementation to libidn2 [BZ #19728] [BZ #19729] [BZ #22247]

[Adhemerval Zanella <adhemerval dot zanella at linaro dot org>]

On 23/05/2018 14:38, Carlos O'Donell wrote:
> On 05/23/2018 01:12 PM, Adhemerval Zanella wrote:
>> On 23/05/2018 13:55, Florian Weimer wrote:
>>> On 05/23/2018 06:44 PM, H.J. Lu wrote:
>>>> On Fedora 28, I got
>>>>
>>>> FAIL: resolv/tst-resolv-ai_idn FAIL:
>>>> resolv/tst-resolv-ai_idn-latin1
>>>
>>> See the NEWS file.  You need libidn 2.0.5 (and later) to avoid test
>>> failures.  The mirror push for that is currently under way, and you
>>> should be able to install the new version tomorrow.
>>
>> Shouldn't we in this case mark the tests as unsupported instead of
>> failing?
>  
> $0.02. No. It should fail to indicate you have bugs in your system libidn.
> 
> Unsupported would be if you didn't have a libidn present at all (and the
> test looks for this too and returns FAIL_UNSUPPORTED).
> 

So in this case we should at least document this better, commit message does
not have anything regarding it and NEWS update just states "[...] libidn2 
version 2.0.5 or later is recommended. [...]" which does not really state
why 2.0.5 is preferred.  

Also, reading the newer tests does give any indication why 2.0.5 is required 
to fully compliance nor they check libidn2 version.  To one involved in glibc
developments it should be straightforward to relate possible test-suite to
related error, but I still think that indicating that system tests libidn is 
the culprit nor glibc itself is a better error output.