[plt-scheme] contract library reoganization

Thanks! -- Matthias


On Sep 17, 2007, at 1:12 PM, Robby Findler wrote:

> It seems that this change is too problematic for now. I've reverted  
> it.
>
> Robby
>
> On 9/17/07, Robby Findler <robby at cs.uchicago.edu> wrote:
>> No, no, you can (in all versions) safely require both (lib  
>> "class.ss")
>> and (lib "contract.ss") and get the same set of identifiers. Its only
>> a problem if you require one or the other.
>>
>> Robby
>>
>> On 9/17/07, Richard Cobbe <cobbe at ccs.neu.edu> wrote:
>>> On Mon, Sep 17, 2007 at 11:15:36AM -0500, Robby Findler wrote:
>>>> I've re-organized the contract library so that it no longer exports
>>>> class & object-related contracts, instead those are exported by  
>>>> (lib
>>>> "class.ss").
>>>>
>>>> This is currently breaking a few planet packages, so if people need
>>>> more time to fix this, I can switch it back temporarily. (The fix
>>>> should be simple: just add (require (lib "class.ss")) to the  
>>>> file in
>>>> question and upload a new planet package).
>>>
>>> I don't think this affects me or my planet packages directly, but  
>>> there is
>>> an important issue here.
>>>
>>> Does this change correspond to a MzScheme version number change?   
>>> I.e.,
>>> in 371.2 the class contracts are in (lib "contract.ss") but in 371.3
>>> they're in (lib "class.ss") or something like that.
>>>
>>> If the changes don't line up like this, then planet package  
>>> authors are
>>> kind of stuck.  There's no value, AFAICT, that they can specify  
>>> in the new
>>> version of the planet package that works in all cases.
>>>
>>> Richard
>>> _________________________________________________
>>>   For list-related administrative tasks:
>>>   http://list.cs.brown.edu/mailman/listinfo/plt-scheme
>>>
>>
> _________________________________________________
>   For list-related administrative tasks:
>   http://list.cs.brown.edu/mailman/listinfo/plt-scheme