Re: ocaml-dns- quick ocaml question

[Anil Madhavapeddy <anil@xxxxxxxxxx>]

So cstruct shouldnt be a constraint; it's trivial to retarget to polymorphic 
variants. There is a very slight performance hit, but I don't think that 
matters.

GADTs: I don't see how they help here; did you have anything in mind Raphael, 
or was it a general "new shiny feature"? :-)

-anil

On 18 Jun 2012, at 18:42, Richard Mortier <Richard.Mortier@xxxxxxxxxxxxxxxx> 
wrote:

> 
> On 18 Jun 2012, at 18:38, Raphael Proust wrote:
> 
>> On Mon, Jun 18, 2012 at 5:26 PM, Richard Mortier
>> <Richard.Mortier@xxxxxxxxxxxxxxxx> wrote:
>>> i have two types, rr_type and q_type. rr_type is a subset of q_type, 
>>> extended it by about 5 elements. both are currently implemented in 
>>> Dns.Packet as cenums which are translated to variant types.
>>> 
>>> i've just realised that (of course?) this means that the last definition 
>>> (rr_type) overrides the first (q_type) as they have overlapping elements.
>>> 
>>> is there a nice/idiomatic way around this without either going to 
>>> polymorphic variant types (that's what the `T types are called, right?) or 
>>> just defining the cenum q_type with rr_type aliased to it (for slightly 
>>> better readability), and using that everywhere?
>>> 
>>> eg., should i do something like create submodules Qtype and RRtype of 
>>> Packet, and define cenum t in each?
>> 
>> 
>> I see several possibilities:
>> 
>> - use polymorphic variants (yes that's what `Ts are called)
> 
> unfortunately cstruct targets variants only at the moment.
> 
>> - use one variant type and add assertion checks or other dynamic failure 
>> (bad!)
> 
> ack- i thought that didn't "feel" right :)
> 
>> - use several submodules
>> 
>> - prefix the variants (very common) (feels similar to submodules but prevents
>> "open"-induced ambiguities)
>> type q =
>> | Q_Foo
>> | Q_Bar of int
>> type rr =
>> | RR_Foo
>> | RR_Bar of int
>> | RR_Baz of float
> 
> ok- i'll probably go with one of these options then.  (likely the latter.)
> 
> was just wondering if there was a "better" way.
> 
>> - use GADTs (requires OCaml 4.00 (still in beta)) (I am unsure of the proper
>> syntax, check the manual if need be)
>> type q
>> type rr
>> type 'a t =
>> | Foo : 'a t
>> | Bar : int -> 'a t
>> | Baz : float -> q t (* This branch forces the ['a] to be instantiated with
>>                         [q], thus preventing anyuse with functions requiring
>>                         [rr t]
>>                       *)
> 
> interesting!  but possibly a little advanced for me for now :)
> 
> thanks!
> 
> -- 
> Cheers,
> 
> R.
> 
> 
> 
> 
> This message and any attachment are intended solely for the addressee and may 
> contain confidential information. If you have received this message in error, 
> please send it back to me, and immediately delete it.   Please do not use, 
> copy or disclose the information contained in this message or in any 
> attachment.  Any views or opinions expressed by the author of this email do 
> not necessarily reflect the views of the University of Nottingham.
> 
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system:
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.