[
https://issues.apache.org/jira/browse/SQOOP-2937?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15306468#comment-15306468
]
Attila Szabo commented on SQOOP-2937:
-------------------------------------
Hi all,
[~z17] many thanks for your highlights. I think generally you're absolutely
right, that smaller changes/patches are easier to review, and these should be
reflected in subtasks. I have only one concerns I'd like to raise here: in case
of Sqoop1 connectors it's not always very easy to separate things into smaller
tasks/patches/commits, for example because "splitting" the existing
functionality would result in creating two new instead of the original
solution, but none of them would stand without the other one. So regarding to
that I think we always have to consider if it makes sense to have a series of
different change, or one bigger squashed patch/commit. But as I told: generally
your approach should be the default.
[[email protected]] I'd like to ask the following things:
- First of all, assign the JIRA ticket to you, as you're the one who is working
on it (and have a working solution AFAIR).
- Updload the GDG related patch file to this ticket
- Please open a review ticket (on reviews.apache.org) with the patch file, and
link the review ticket to this JIRA ticket (you could do that by
More->Link->Web link). You could add me to that ticket as a reviewer (and
optionally please add [~z17] too, I think it would be a good opportunity for
her to start with Sqoop1 development, anfter she's already quite experienced
with Sqoop2 )
- Please press "submit patch" button
- Please open a new ticket for the tape related changes, and do the very same
process on that one too (patch file, review ticket, etc.). Assign that ticket
to you as well.
- Optionally (if it is feasible on your side): Please open another ticket as
the parent ticket of this, and convert the two previous ones to subtasks and
link them together,
Cheers,
M.
> Sqoop mainframe module does not support sequential data sets, GDG, or data
> sets on tape
> ---------------------------------------------------------------------------------------
>
> Key: SQOOP-2937
> URL: https://issues.apache.org/jira/browse/SQOOP-2937
> Project: Sqoop
> Issue Type: Improvement
> Components: connectors
> Affects Versions: 1.4.6
> Environment: mainframe
> Reporter: Chris
> Priority: Minor
> Labels: patch
> Fix For: 1.4.7
>
> Original Estimate: 168h
> Remaining Estimate: 168h
>
> The mainframe module (SQOOP-1272) currently supports partitioned data sets.
> I propose adding the following improvements:-
> - sequential data set support
> - GDG data set support
> - data sets stored on tape.
> I propose the following parameters to be added:-
> 1. --datasettype <p|s|g> (p = partitioned data set, s = sequential, g = GDG)
> 2. --tape <true|false> (true = data set is on tape, false = data set is not
> on tape)
> Let me know if this should be all in one issue or separate issues
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
