Hi everyone,
Stephan pointed out that our naming of a generic/blackbox/opaque type in
SQL might be not intuitive for users. As the term ANY rather describes a
"super-class of all types" which is not the case in our type system. Our
current ANY type stands for a type that is just a blackbox wit
Thanks to Timo for bringing up an interesting topic.
Personally, "OPAQUE" doesn't seem very intuitive with respect to types. (It
suits pretty well to glasses, thought. :)) Anyway, could we just use
"UNKNOWN", which is more explicit and true reflects its nature?
Thanks,
Xuefu
On Fri, Oct 18, 201
I prefer OPAQUE compared to ANY because any is often the root object in an
object hierarchy and would indicate to users the wrong thing.
Aljoscha
> On 18. Oct 2019, at 18:41, Xuefu Z wrote:
>
> Thanks to Timo for bringing up an interesting topic.
>
> Personally, "OPAQUE" doesn't seem very int
+1 to rename ANY.
I don't have strong opinion on the new name. I think "OPAQUE" is fine,
because it is introduced in IBM Informix and Oracle.
In Informix, it says [1]:
"An opaque data type is fully encapsulated; the database server does not
know about the internal format of an opaque data type.
T
OPAQUE seems to be a little bit advanced to a lot non-english
speakers (including me). I think Xuefu raised a good alternative:
UNKNOWN. What do you think about it?
Best,
Kurt
On Mon, Oct 21, 2019 at 3:49 PM Aljoscha Krettek
wrote:
> I prefer OPAQUE compared to ANY because any is often the roo
Hi,
IMHO, `UNKNOWN` does not fully reflects the situation here, because the types
are
actually “known” to users, and users just want to leave them out of Flink type
system.
+1 for `RAW`, for it's more intuitive than `OPAQUE`.
Best,
Paul Lam
> 在 2019年10月21日,16:43,Kurt Young 写道:
>
> OPAQUE s
I also think `UNKNOWN` is not suitable here.
Because we already have `UNKNOWN` value in SQL, i.e. the three-valued logic
(TRUE, FALSE, UNKNOWN) of BOOLEAN type.
It will confuse users here, what's the relationship between them.
Best,
Jark
On Mon, 21 Oct 2019 at 17:53, Paul Lam wrote:
> Hi,
>
> I
I would also avoid `UNKNOWN` because of the mentioned reasons.
I'm fine with `RAW`. I will wait another day or two until I conclude the
discussion.
Thanks,
Timo
On 21.10.19 12:23, Jark Wu wrote:
I also think `UNKNOWN` is not suitable here.
Because we already have `UNKNOWN` value in SQL, i.e
+1 to RAW, if there's no better candidate comes up.
Best,
Kurt
On Mon, Oct 21, 2019 at 9:25 PM Timo Walther wrote:
> I would also avoid `UNKNOWN` because of the mentioned reasons.
>
> I'm fine with `RAW`. I will wait another day or two until I conclude the
> discussion.
>
> Thanks,
> Timo
>
>
“OPAQUE” seems a little strange to me.
+ 1 for ‘RAW’.
Best,
Terry Wang
> 2019年10月22日 09:19,Kurt Young 写道:
>
> +1 to RAW, if there's no better candidate comes up.
>
> Best,
> Kurt
>
>
> On Mon, Oct 21, 2019 at 9:25 PM Timo Walther wrote:
>
>> I would also avoid `UNKNOWN` because of the me
+1 for RAW.
I agree that this is clearer than OPAQUE (which I initially proposed).
On Mon, Oct 21, 2019 at 10:33 AM Jark Wu wrote:
>
> +1 to rename ANY.
>
> I don't have strong opinion on the new name. I think "OPAQUE" is fine,
> because it is introduced in IBM Informix and Oracle.
> In Informi
+1 for RAW type, which is easy to guess its actual meaning, even for a new
user.
Best,
Weike
On Tue, Oct 22, 2019 at 5:31 PM David Anderson wrote:
> +1 for RAW.
>
> I agree that this is clearer than OPAQUE (which I initially proposed).
>
> On Mon, Oct 21, 2019 at 10:33 AM Jark Wu wrote:
> >
>
Thanks everyone for the feedback!
We have a clear favorite which is the RAW type. I will make sure that
this change is applied to the Flink code base.
Thanks again,
Timo
On 22.10.19 04:07, Terry Wang wrote:
“OPAQUE” seems a little strange to me.
+ 1 for ‘RAW’.
Best,
Terry Wang
2019年10月
13 matches
Mail list logo