Github user ChengXiangLi closed the pull request at:
https://github.com/apache/flink/pull/1414
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature i
Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-161232136
Alright, I merged it. Could you please close it if github doesn't do so
automatically.
Thanks for the work!
---
If your project is set up for it, you can repl
Github user ChengXiangLi commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-161153580
Let's reuse the count aggregation now, as it's better for performance now.
I would revisit this after NULL value handling is enabled.
---
If your project is set up
Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-160956025
Ah, I see you changed it to my (wrong) suggestion. @ChengXiangLi Do you
want to change it back to your initial version (plus maybe the fix for the
naming and cleanExpr)
Github user twalthr commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-160949963
Looks good. No objecttions.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have
Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-160879522
Oh, you are way ahead of me. :sweat_smile: You are right, keep it that way
since we will probably have support for Null values in the future.
I'll merge it late
Github user ChengXiangLi commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-160822216
`a.count` and `b.count` may not the same all the times(for example is `a`
contains NULL values), that's why i try to aggregate it twice. But NULL value
is not allow
Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-160153085
It's good that you found the problem and added a test. :smile:
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub a
Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/1414#issuecomment-160153032
Hi,
the reason it was implemented as it was is to eliminate common
subexpressions. For example, if you wrote:
```
select(a.count, b.count)
```
t
GitHub user ChengXiangLi opened a pull request:
https://github.com/apache/flink/pull/1414
[FLINK-3087] [Table-API] support multi count in aggregation
The `Literal(1)` is used as the IntermediateField of `Count` aggregation,
so multi `Count` looks the same in `ExpandAggregations`.
Y
10 matches
Mail list logo
