aurora-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From John Sirois <jsir...@apache.org>
Subject Re: Review Request 45457: Mitigate error-prone queries by disallowing no-op filters.
Date Tue, 29 Mar 2016 23:20:16 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/45457/#review125978
-----------------------------------------------------------



How would these queries produce incorrect results as compared to previous queries?  Perviously,
a check for the collection being null would have had to have been made by the caller (or a
check for !IsSet) to skip the calls you've added precondition checks to.  The IsSet calls
are now gone, meaning the call sites using those have been semantically adjusted.  I think
this only leaves the callers who pass null of which there should be none due to the semantic
change as well.

- John Sirois


On March 29, 2016, 5:01 p.m., Bill Farner wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/45457/
> -----------------------------------------------------------
> 
> (Updated March 29, 2016, 5:01 p.m.)
> 
> 
> Review request for Aurora, John Sirois and Zameer Manji.
> 
> 
> Bugs: AURORA-1652
>     https://issues.apache.org/jira/browse/AURORA-1652
> 
> 
> Repository: aurora
> 
> 
> Description
> -------
> 
> Short of surveying call sites, i don't know the impact of this change.  However, my sense
is that it's still better to fail fast than produce incorrect results (due to the recent change
in semantics of filtering by an empty iterable).
> 
> 
> Diffs
> -----
> 
>   src/main/java/org/apache/aurora/scheduler/base/Query.java c76b365f43eb6a3b9b0b63a879b43eb04dcd8fac

> 
> Diff: https://reviews.apache.org/r/45457/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Bill Farner
> 
>


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