For me, the number of projects I have to deal with which still us ns_share
is heinous.  I just wish there were a drop-in replacement but there just isn't.

Kris


On Tue, 23 Apr 2002 09:41:49 +0100, Harry Moreau <[EMAIL PROTECTED]> wrote:

>On 22 Apr, Michael A. Cleverly wrote:
>> On Thu, 18 Apr 2002, Rusty Brooks wrote:
>>
>>> The point being, if I had a method to bring one of the shared variables into
>>> the current context, like ns_share did, then all other procedures downstream
>>> could be blissfully unaware if they were shared or not.
>>
>> How about using variable read/write traces to mimic the old ns_share
>> behaviour using nsv's? Here's a quick lunch time first stab at it:
>>
>
>ns_share is implemented using traces in AOLserver 3.x.
>This accounts for why it's a lot slower than nsv's.
>
>I haven't benchmarked it for a while, but in the 3.0 beta timeframe
>(when ns_shares included some tcl code for their implementation)
>there was an order of magnitude difference in the speed of ns_share and nsv's.
>
>--
>--Harry Moreau----------------

Reply via email to