Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
Refactored based on feedback for some things, based on making what I was
trying for more correct in others.
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
Ok, I'll change it. Feels a little crossing the streams, but we'll see
---
Github user nickwallen commented on the issue:
https://github.com/apache/metron/pull/856
The justification that you mentioned just doesn't seem strong enough to me.
Unless there is more that I am missing.
IMHO We should only use magic commands for things that can't be
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
Do you feel strongly that this should be a Function? @cestella ? I'm not
opposed to changing it if you are. I would like to here some more feedback
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
It did not seem appropriate to me for this to be a stellar function, and
%magic is the other way to execute things from the shell. At the time at least.
---
Github user nickwallen commented on the issue:
https://github.com/apache/metron/pull/856
Why is this validation process driven by a %magic command?
Magics were made for functionality that cannot be implemented directly
within a Stellar execution environment. Often for
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
@nickwallen yeah, we need to cover a bunch a scenarios
---
Github user nickwallen commented on the issue:
https://github.com/apache/metron/pull/856
@ottobackwards Yes, definitely. I like it at a 50k foot level. The only
thing that struck me was the need for the different annotation types. But I
haven't had a chance to dig into it yet.
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
@nickwallen any feedback, does the annotated approach match what you
imagined?
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
Closing to test build in travis
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
Bump?
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
@cestella @simonellistonball
With the new implementation, doing the blob or file check should be a piece
of cake would you prefer it as part of this or as a new issue?
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
https://github.com/apache/metron/pull/856/commits/65278a67a07f1c4c23ab2d95ebb6de92e1cac731
introduces conceptually what @nickwallen and I have been discussing.
I need to think about
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
Per the conversation above, i'm going to take a stab at the attributed
approach.
I think the Stellar Functions should be a separate Jira.
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
I am glad for the interest in this PR, and that it seems to have sparked
some great ideas for continuing on.
What I would like to do is line it up as follows
1. This PR with
Github user cestella commented on the issue:
https://github.com/apache/metron/pull/856
@simonellistonball Agree to the namespace idea. My bad :)
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
@simonellistonball, yes, the namespace should be part of the jira and
interface design
---
Github user ottobackwards commented on the issue:
https://github.com/apache/metron/pull/856
So, the scenario here is checking things that *were* valid when uploaded,
but have been invalidated by external changes ( language changes ). I would
like to keep the magic specific.
Github user simonellistonball commented on the issue:
https://github.com/apache/metron/pull/856
@cestella I would say that proposed validate function has to be very much
in a namespace. It feels like a name that would be much more useful for a
function replacing our current approach
Github user cestella commented on the issue:
https://github.com/apache/metron/pull/856
Also, it might be useful for `%validate_configured_expressions` to take a
file path so you can validate a set of configs on disk (again, if it gets to
zookeeper, zk_load_utils.sh should fail if
Github user cestella commented on the issue:
https://github.com/apache/metron/pull/856
Any chance we can add a `VALIDATE(str, type)` function to the stellar
management functions where str is the json blob string for the config and the
type is the type of config? Generally the goal
24 matches
Mail list logo
