[
https://issues.apache.org/jira/browse/BIGTOP-1536?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14307498#comment-14307498
]
RJ Nowling edited comment on BIGTOP-1536 at 2/5/15 4:39 PM:
------------------------------------------------------------
[~jayunit100] For the stats class, I think we should be explicit. What do you
think about adding a totalTransactionCount : long field and dropping the
string?. For the integration test, can you add some small assertions such as
ensuring that the number of customers and stores output from the ETL matches
the input to the generator?
I'd fine with starting with this as a stub and adding json output + more
queries in BIGTOP-1653.
was (Author: rnowling):
[~jayunit100] For the stats class, I think we should be explicit. What do you
think about adding a totalTransactionCount : long field and dropping the
string?. For the integration test, can you add some small assertions such as
ensuring that the number of users and stores output from the ETL matches the
input to the generator?
I'd fine with starting with this as a stub and adding json output + more
queries in BIGTOP-1653.
> BPS Spark : Unit test upgrade + Basic Sales Analytics Example.
> --------------------------------------------------------------
>
> Key: BIGTOP-1536
> URL: https://issues.apache.org/jira/browse/BIGTOP-1536
> Project: Bigtop
> Issue Type: Improvement
> Components: blueprints
> Reporter: RJ Nowling
> Assignee: RJ Nowling
> Attachments: BIGTOP-1536.patch, BIGTOP-1536.patch, BIGTOP-1536.patch,
> firstpass.patch, spark_driver_output.patch
>
>
> Using the Spark data generator and ETL script (BIGTOP-1535), add a simple
> Spark sales analytics example that computes basic stats such as:
> * Number of sales per category per month or quarter
> * Top selling items in each category per month or quarter
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
