Hi Vladimir,
What you said resonated with me: subtle differences will be present across
catalog types albeit our strong will of making it consistent. And I don't
have an issue with "flexible behavior". We don't have to enforce perfect
consistency of catalog behavior, and I would judge case by case
Hi, Haizhou.
Thanks for bringing this up. Yes, this problem appears similar to what I
explained but with a slightly different manifestation. What is funny is
that you observed it when doing REST integration tests for Spark, and I was
doing the same for Trino (the fully-fledged integration tests ar
Hello Vladimir,
I want to raise that we've been observing similar behavior differences
regarding CREATE OR REPLACE between Hive/Hadoop catalog and REST catalog at
here: https://github.com/apache/iceberg/issues/11109
The context: Iceberg spark integration test has traditionally only included
tests
Hi,
Sure, will do.
*Vladimir Ozerov*
Founder
querifylabs.com
Ср, 23 окт. 2024 г. в 08:50, Jean-Baptiste Onofré :
> I second Ryan here, it would be great to clarify in the
> "implementation notes" section.
>
> Thanks !
> Regards
> JB
>
> On Wed, Oct 23, 2024 at 1:10 AM rdb...@gmail.com wrote:
I second Ryan here, it would be great to clarify in the
"implementation notes" section.
Thanks !
Regards
JB
On Wed, Oct 23, 2024 at 1:10 AM rdb...@gmail.com wrote:
>
> Thanks Vladimir! Would you like to open a PR to make that change? It sounds
> like another good item to put into the "Implement
Thanks Vladimir! Would you like to open a PR to make that change? It sounds
like another good item to put into the "Implementation notes" section.
On Sun, Oct 20, 2024 at 11:41 PM Vladimir Ozerov
wrote:
> Hi Jean-Baptiste,
>
> Agreed. REST spec looks good. I am talking about the general spec, wh
Hi Jean-Baptiste,
Agreed. REST spec looks good. I am talking about the general spec, where it
might be useful to add a hint to engine developers, that CREATE OR REPLACE
semantics in Iceberg is expected to follow slightly different semantics.
This is already broken in Trino: depending on catalog ty
Hi Vladimir,
As Ryan said, it's not a bug: CREATE OR REPLACE can be seen as "CREATE
AND UPDATE" from table format perspective. Specifically for the
properties, it makes sense to not delete the current properties as it
can be used in several use cases (security, tables grouping, ...).
I'm not sure
Hi Ryan,
Thanks for the clarification. Yes, I think my confusion was caused by the
fact that many engines treat CREATE OR REPLACE as a semantic equivalent of
DROP + CREATE, which is performed atomically (e.g., Flink [1]). Table
formats add history on top of that, which is expected to be retained,
Hi Vladimir,
This isn't a bug. The behavior of CREATE OR REPLACE is to replace the data
of a table, but to maintain things like other refs, snapshot history,
permissions (if supported by the catalog), and table properties. Table
properties are replaced if they are set in the operation like `b` in
10 matches
Mail list logo