Re: Type Transcriptions

[Shark8 <OneWingedShark@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1109 bytes --]

On 21-Apr-14 13:42, Jeffrey Carter wrote:
> On 04/21/2014 11:51 AM, Shark8 wrote:
>>
>> I suppose the better way to phrase it is:
>> "Is it better to strictly stick to the spec, letting generic
>> instantiation take
>> care of these 'extra-typal' parameters, *OR* doing things the natural
>> Ada way,
>> and worrying about 'fixing-up' things in, say, 'Input and 'Output
>> operations."
>
> My vote is for the Ada way, as long as the functionality is correct.
> Making it clear, easy to read and understand, and correct are the
> important things.

To be fair, going the generic-package route isn't necessarily /not/ the 
Ada way -- it's just not, IIUC, all that common. There's an interesting 
Technical Report [attached] on how generic-packages and records share a 
/lot/ of underlying ideas that are identical.

This would, however require some sort of mechanism to keep sessions 
associated with the Security_Parameters; this sort of situation is where 
something like package-pointers would be an interesting solution -- 
IIRC, Randy mentioned the idea as one of them as AI ideas that had been 
rejected.

[-- Attachment #2: www.cs.dartmouth.edu/reports/TR86-104.pdf --]
[-- Type: application/pdf, Size: 763193 bytes --]