[
https://issues.apache.org/jira/browse/CLIMATE-634?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14542393#comment-14542393
]
ASF GitHub Bot commented on CLIMATE-634:
----------------------------------------
Github user kwhitehall commented on the pull request:
https://github.com/apache/climate/pull/196#issuecomment-101771065
@darth-pr From the latter comments on @huikyole PR#197 it would appear that
the request is to deprecate this and a number of other functions with another
function called temporal_subset. I haven't fully reviewed the code towards that
end - of replacing the other functions which were recently mentioned. I had
only reviewed it wrt to replacing this function, which seemed neither here nor
there. THIS tix was indicating that RIGHT now if you try to use the function
utils.calc_climatology_season for a season that is defined between years as
Dec-Jan-Feb (DJF), it WILL throw an error because the code incorrectly attempts
to subset a DS Object as opposed to the DS.values array. BUT if you used the
same function for a season that did not fall between years, e.g. Jun-Jul-Aug.
It would work. On this PR discussion then ensued about if the subsetting that
occurs in the DJF case actually works, and I provided a case using a 3D dataset
representing the DS.values variable and a 1D array representing the DS.times
variable to demonstrate that the existing method does work. That said, the
current PR does not contain the update for the DS.time variable (this would be
an addition to the utils.calc_climatology_season and should be on a different
tix in my opinion, but it was an excellent suggestion by @huikyole and I
decided to demonstrate in the example that without overhauling the code, we can
get it done - but it is not currently a part of this PR, I will create a
separate tix depending on how this ends). As for the slide on #197 I'm still
trying to figure out what that is about. As far as I can tell, the stuff in the
column under my Github Handle are related to the "unittest" to demonstrate the
code extracts the correct portions of a 3D array (like in a DS.values array)
and the corresponding 1D array (like in the times array). I also indicated the
same example using PR#197 with OCW DS will work here.
@MJJoyce @darth-pr I hope that sheds some light on the connectivity
between the two PRs from my perspective
> fix calc_climatology_monthly
> -----------------------------
>
> Key: CLIMATE-634
> URL: https://issues.apache.org/jira/browse/CLIMATE-634
> Project: Apache Open Climate Workbench
> Issue Type: Bug
> Reporter: Kim Whitehall
> Assignee: Kim Whitehall
>
> Bug if start month > end mth causes calc_climatology_season to fail
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
