Re: [Gm2] CASE: gm2-4.1.2 of 2011-11-08 bug (or possibly philosophical difference)

[Gaius Mulley]

john o goyo <address@hidden> writes:

> On 13-Nov-11, at 4:11 p, Manfred Hollstein wrote:
>
>> On Sun, 13 Nov 2011, 21:44:48 +0100, john o goyo wrote:
>>> On 13-Nov-11, at 6:45 a, Gaius Mulley wrote:
>>>> [...]
>>>> Hi Gary,
>>>>
>>>> yes I think this is why I didn't skip the test in gm2 1.0.  But I
>>>> concede that in huge legacy code bases it could be painful to add
>>>> ELSE
>>>> to each variant record.  As you say it is also easy to disable the
>>>> check
>>>> anyway.
>>>
>>> Legacy code is exactly my problem.  There are a lot of such cases --
>>> not just with variant records but also incomplete case selections in
>>> code.
>>
>> FWIW, but to be honest, I really hate to implement some workarounds
>> in a
>> tool whose purpose should just be to compile a properly defined
>> language
>> into machine code... I normally try to _educate_ every person who
>> contacts me wrt/ "could the compiler not do this to save me from doing
>> that?", that (s)he should fix the anomalies in their own code, because
>> the next time, they'll complain about other non-portabilities...
>
> This is not a question of workarounds.  PIM3 allows incomplete CASE
> statements.  If gm2 is to allow PIM3 syntax, then it must allow
> incomplete CASE statements.
>
> As for "anomalies" in existing code, the legacy code I mention
> followed PIM3 and compiled under PIM3 compilers -- including the ETH
> compilers.  My intention is not to edit the code, it is to compile
> it.  If someday, I decide to actually maintain the code or use it for
> some other purpose, then I will modify it to conform to ISO but not
> before.
>
> john
>
>>
>>> john
>>
>> Just my 2 ¢ ;-)
>>
>> Cheers..
>>
>> l8er
>> manfred

Hi,

just to say I'm currently working on a fix to disable the check by value
but to include a check that all values are type compatible.  Hopefully
the fix will be available in a couple of days,

regards,
Gaius