Or Class (not String) loaded from the proper ClassLoader
On Mon, Jun 18, 2018 at 3:51 PM Steve Ebersole wrote:
> You can pass a RegionFactory instance.
>
> On Mon, Jun 18, 2018 at 1:48 PM Henri Tremblay
> wrote:
>
>> Hi,
>>
>> A change in the latest Spring Boot version is now setting a real
You can pass a RegionFactory instance.
On Mon, Jun 18, 2018 at 1:48 PM Henri Tremblay
wrote:
> Hi,
>
> A change in the latest Spring Boot version is now setting a real class
> loader when creating a JCache cache manager. Before it used to pass null.
> Just like Hibernate still does.
>
> So
> On 18 Jun 2018, at 17:49, Sanne Grinovero wrote:
>
> I was experimenting with "let's encrypt" but then you told me that I
> wasn't allowed to use that service.
I don’t remember saying that. Knowing me I surely gave you a reason :)
___
Hi,
A change in the latest Spring Boot version is now setting a real class
loader when creating a JCache cache manager. Before it used to pass null.
Just like Hibernate still does.
So right now, we end up with two class managers. One for Hibernate 2nd
level cache and one for Spring cache.
We
On Sat, 16 Jun 2018 at 18:51, Emmanuel Bernard wrote:
>
> Let’s wait quickly for Sanne’s feedback on why he was having certificate
> issues. My experience with let’s encrypt has been stellar so far (httpd).
I was experimenting with "let's encrypt" but then you told me that I
wasn't allowed to
On Mon, Jun 18, 2018, 2:44 AM Yoann Rodiere wrote:
> If by "basic types" you mean *all* basic types, including user-defined
> ones, I think it would make sense. Otherwise it sounds a bit limiting.
>
In 6.0 "user-defined type" means a very different thing generally speaking.
But I mean any
for native SQL I think it makes sense to only support basic types for
parameters binding.
On Mon, 18 Jun 2018 at 08:45, Yoann Rodiere wrote:
> If by "basic types" you mean *all* basic types, including user-defined
> ones, I think it would make sense. Otherwise it sounds a bit limiting.
>
>
If by "basic types" you mean *all* basic types, including user-defined
ones, I think it would make sense. Otherwise it sounds a bit limiting.
There's the case of embedded IDs that might be considered as an exception,
but I'm not sure there's a compelling reason to do so.
By the way... it's not
