Github user eliaslevy commented on the issue:
https://github.com/apache/flink/pull/3477
I created [FLINK-6419](https://issues.apache.org/jira/browse/FLINK-6419).
---
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
Github user kl0u commented on the issue:
https://github.com/apache/flink/pull/3477
Hi @eliaslevy , @dawidwys is right, in fact if the state's name is "a" and
it has 2 matching events, there will be two returned keys "a_0" and "a_1". This
is definitely counterintuitive and the reason
Github user dawidwys commented on the issue:
https://github.com/apache/flink/pull/3477
Right now if the pattern name is "a" the events will be returned with keys:
"a[0]", "a[1]" and so on, but agree it may be counterintuitive. Please file a
JIRA for it.
---
If your project is set
Github user eliaslevy commented on the issue:
https://github.com/apache/flink/pull/3477
Am I missing something or is there no way to get access to access to
multiple events matched by these new quantifiers? The
`PatternSelectFunction.select` takes an argument of `Map` and
Github user dawidwys commented on the issue:
https://github.com/apache/flink/pull/3477
Great! Thanks for your help @kl0u !
---
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
Github user kl0u commented on the issue:
https://github.com/apache/flink/pull/3477
Merged this! Thanks for the work @dawidwys.
Can you close this PR and the JIRA?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If
Github user kl0u commented on the issue:
https://github.com/apache/flink/pull/3477
Hi @dawidwys , I have 2 comments for this PR:
1) in the `findFinalStateAfterProceed()` in the `NFA` you should also check
the filter condition as you traverse the `PROCEED` edges towards the end.
Github user dawidwys commented on the issue:
https://github.com/apache/flink/pull/3477
I've updated the PR with a solution for the previously failing test. Also
tried to add some more comments. @kl0u I think it is ready for another round of
review :)
---
If your project is set up
Github user kl0u commented on the issue:
https://github.com/apache/flink/pull/3477
Sounds good! Thanks @dawidwys
---
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
Github user dawidwys commented on the issue:
https://github.com/apache/flink/pull/3477
Hi, I've addressed the style comments. Still have some doubts about the
`Preconditions`. I will try to add more comments on the `NFACompiler` with the
implementation that will handle the failing
Github user kl0u commented on the issue:
https://github.com/apache/flink/pull/3477
Hi @dawidwys , the following produces no output, which should not be the
case. The same holds if you change the pattern to `oneOrMore`.
```
public void testZeroOrMore() {
Github user kl0u commented on the issue:
https://github.com/apache/flink/pull/3477
Thanks @dawidwys I will have a look now.
---
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
Github user dawidwys commented on the issue:
https://github.com/apache/flink/pull/3477
Thanks @kl0u for your comments, I've changed all code style specific.
Some comments were addressing a behaviour that one `Pattern` can be
translated into multiple "equivalent" `States`
Github user kl0u commented on the issue:
https://github.com/apache/flink/pull/3477
Hi @dawidwys , I will start reviewing your PR now.
I would say that for now it makes sense to have it with exceptions and not
change the `Pattern` hierarchy. The reason is:
1) not sure
14 matches
Mail list logo