metron-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Casey Stella <>
Subject Re: [GitHub] incubator-metron issue #293: METRON-473 Add LENGTH() To Stellar
Date Thu, 06 Oct 2016 19:18:30 GMT
The purpose of validate is to ensure the statement makes syntactic sense
from a stellar perspective and it's only called when statements change in
the config.
On Thu, Oct 6, 2016 at 21:15 Casey Stella <> wrote:

> Oh no validate is not called at runtime for every packet! That's called at
> stellar statement input time (e.g. Config pushes to zookeeper)
> On Thu, Oct 6, 2016 at 21:12 ottobackwards <> wrote:
> Github user ottobackwards commented on the issue:
>     I took this jira to get some start of an idea about Stellar, and after
> debugging through it to track down to exitVariable to find the argument
> resolution and then back up to the validate x->null I would say I got what
> I bargained for.
>     I am not sure how I would answer your question on validation, I don't
> know Stellar well enough.  My experience and intuition tells me that
> running the 'execute' and using an exception or error case as validation is
> not very efficient at runtime, if in fact this is what happens at runtime
> as well.  Better to have real metadata on the function and just validating
> the metadata and they syntax of the query I would think.   *A
> logical/runtime error does not mean an invalid expression*.  This method is
> in effect equating them which I think is incorrect ( unless I am mistaking
> the intent of validate()).
> ---
> 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 is enabled but not working, please
> contact infrastructure at or file a JIRA ticket
> with INFRA.
> ---

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message