[
https://issues.apache.org/jira/browse/ACCUMULO-2097?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13858922#comment-13858922
]
Josh Elser edited comment on ACCUMULO-2097 at 12/30/13 5:01 PM:
----------------------------------------------------------------
bq. I'm already working on the CSV/TSV part, even though there is not a JIRA
for it yet. Do you want to wait for me to finish that so you can integrate it
in, or do you have different plans?
/me facepalm
I just worked up a bunch of stuff last night too, haha. I was making something
standalone and then going to see how/where it makes sense to merge it in. Lemme
push my stuff to Github for now.
Edit: https://github.com/joshelser/accumulo-csv I got most of the boilerplate
stubbed out last night. I was planning on supporting rfile creation or mutation
writing. I also planned to implement a rudimentary schema support to show how
the non-fixed colfam set for a table can be used and tracked.
was (Author: elserj):
bq. I'm already working on the CSV/TSV part, even though there is not a JIRA
for it yet. Do you want to wait for me to finish that so you can integrate it
in, or do you have different plans?
/me facepalm
I just worked up a bunch of stuff last night too, haha. I was making something
standalone and then going to see how/where it makes sense to merge it in. Lemme
push my stuff to Github for now.
> Add ability to load when from script when MiniAccumuloCluster starts up
> -----------------------------------------------------------------------
>
> Key: ACCUMULO-2097
> URL: https://issues.apache.org/jira/browse/ACCUMULO-2097
> Project: Accumulo
> Issue Type: Improvement
> Components: instamo-archetype
> Reporter: Michael Wall
> Assignee: Josh Elser
> Priority: Minor
>
> It would be helpful to be able load data from a script when the
> MiniAccumuloCluster starts up from the instamo-archetype generated project.
> I was thinking something like the -f option from the shell which simply runs
> commands. My idea here is that it is easy to load that same file on a real
> cluster and have the same data used in testing. I have written something
> like this in the past that parsed the file and created mutations.
> [~elserj] suggested loading data from CSV, which was interesting as well.
> Ideally, this functionality would work for any example in the archetype,
> shell, mapreduce whatever.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
