That sounds like a plan then. How many flags values are "free"? Maybe we could 
identify one specifically for some specific kind of cross platform 
serialization method (again I'm thinking json although not necessarily that). 

My thinking is to make it easier for the clients to put together something 
truly cross platform out of the box so to speak and not to rely on the user of 
the library to setup the correct hooks. Unless I minsunderstood what you said :)


-----------------------------------
This message was sent by Blackberry

----- Original Message -----
From: Aaron Stone <[EMAIL PROTECTED]>
To: Kieran Benton
Cc: [email protected] <[email protected]>
Sent: Wed Jan 30 08:25:42 2008
Subject: Re: Cross-client memcached compatibility

I think it would make sense to flag for serialization method -- rather
than trying to pick or define a single serialization method for all
clients, the clients simply flag with the method they used.

Most languages have libraries available to interpret most other
languages' native serialization, so I think we can leverage that via
callbacks and flexible data type flags.

Aaron


On Wed, 2008-01-30 at 07:50 +0000, Kieran Benton wrote:
> 
> How about standardising on something like json? Either that or we're
> basically going to have to put a custom serialization rules together
> each client has to follow?
> 
> 
> -----------------------------------
> This message was sent by Blackberry
> 
> ----- Original Message -----
> From: [EMAIL PROTECTED]
> <[EMAIL PROTECTED]>
> To: dormando <[EMAIL PROTECTED]>
> Cc: [email protected] <[email protected]>
> Sent: Wed Jan 30 06:08:47 2008
> Subject: Re: Cross-client memcached compatibility
> 
> 
> On Jan 29, 2008, at 21:40, dormando wrote:
> 
> > Wonder if we can just put up a list of flag recommendations (and 
> > storage mediums?) and see if it catches on? Or perhaps pisses 
> > someone off enough to provoke discussion?
> 
> 
>         Makes sense to me.
> 
>         The most controversial part would be lists and dictionaries
> since 
> they nest types.
> 
> --
> Dustin Sallings
> 
> 
> 
> 
> 

Reply via email to