On 2011/1/3 Magnus Therning <[email protected]> wrote:
> On Sun, Jan 2, 2011 at 12:20, Peter Simons <[email protected]> wrote:
>> Hi Magnus,
>>
>>  >> the gist of this issue is that "parsec >= 3.0 || == 2.1.*" is translated 
>> to
>>  >> "parsec>=3.0", which is just plain wrong. Now, wouldn't it be easiest to
>>  >> modify cabal2arch so that it translates that specification to 
>> "parsec>=2.1"?
>>  >> A straight-forward algorithm to accomplish that would be to use the 
>> *lowest*
>>  >> version bound in these kinds of alternatives. Right now, cabal2arch 
>> appears
>>  >> to be using the *first* version bound, which is not exactly optimal.
>>  >
>>  > Well, there's an obvious problem with what you suggest, since
>>  > "parsec>=2.1" is plain wrong too!
>>
>> why is that?
>
> Now you make me think that I might have missed something, but surely
> parsec version 2.2 wouldn't satisfy the former, but would satisfy the
> latter.

I think we have reasons to translate that to 'parsec>=2.1' or
'parsec<2.2' (I would personally use even parsec=2.1.0.1)
* we won't upgrade to parsec 2.2 unless that is included in some
Haskell Platform
* we can expect Parsec 2.2 to be mostly API-compatible with 2.1 if it
comes to existence (but the author of the package didn't want to make
this assumption)
* if HP includes Parsec 3, PKGBUILDs will nevertheless be regenerated
to include this new information.

-- 
Rémy.

_______________________________________________
arch-haskell mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/arch-haskell

Reply via email to