1. Here is a ticket since we’re on the same page: 
https://issues.apache.org/jira/browse/IGNITE-6242 
<https://issues.apache.org/jira/browse/IGNITE-6242>
2. Ok, let’s wait for a while. Predicting that this will be supported on day 
due to users’ feedback :)

—
Denis

> On Aug 31, 2017, at 12:49 AM, Vladimir Ozerov <[email protected]> wrote:
> 
> 1) *At least*.- I am ok, makes sense
> 2) *Ideally* - strong no. Product should not have subtle behavior without a
> strong reason. I do not see justification for this change.
> 
> On Thu, Aug 31, 2017 at 4:19 AM, Dmitriy Setrakyan <[email protected]>
> wrote:
> 
>> I think Denis' idea makes sense.
>> 
>> Vladimir, what do you think?
>> 
>> D.
>> 
>> On Wed, Aug 30, 2017 at 2:57 AM, Denis Magda <[email protected]> wrote:
>> 
>>> Igniters,
>>> 
>>> That’s one more feedback about CREATE TABLE usage in practice.
>>> 
>>> The command automatically creates an IgniteCache naming it
>>> SQL_PUBLIC_{TABLE}. So, if a Person table is created you’ll have
>>> SQL_PUBLIC_PERSON cache in the cluster.
>>> 
>>> Honestly, if you keep to SQL APIs the cache name won’t bother you but as
>>> soon as key-value, compute, service grid APIs are needed the cache name
>>> will be used here and there looking bizarre.
>>> 
>>> Let me propose the following usability improvements until our user does
>>> this:
>>> 
>>> - *At least*
>>> Give a way to pass the cache name into WITH clause parameters set
>>> 
>>> - *Ideally*
>>> Support the above and if the cache already exists use it instead of
>>> creating a new one. This might help to resolve another issue brought up
>>> here: http://apache-ignite-developers.2346864.n4.nabble.com/
>>> CREATE-TABLE-usage-from-Java-API-NET-C-td21455.html
>>> 
>>> What do you think? If everyone is ok, I’ll file a JIRA ticket.
>>> 
>>> —
>>> Denis
>> 
>> 
>> 

Reply via email to