[
https://issues.apache.org/jira/browse/ASTERIXDB-1400?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16899251#comment-16899251
]
Michael J. Carey commented on ASTERIXDB-1400:
---------------------------------------------
Xikui: How hard would it be to do this...? (Borrowing from the file feed,
presumably)
> Loading of non-empty datasets should be supported
> -------------------------------------------------
>
> Key: ASTERIXDB-1400
> URL: https://issues.apache.org/jira/browse/ASTERIXDB-1400
> Project: Apache AsterixDB
> Issue Type: Improvement
> Components: *DB - AsterixDB, AQL - Translator AQL, OTH - Other, STO
> - Storage
> Reporter: Michael J. Carey
> Assignee: Xikui Wang
> Priority: Major
>
> From a user model perspective, we should have a "load more" capability so
> that users can add data into an existing dataset using load-like DDL. Under
> the hood this can essentially be an insert-into-from-external-files feature,
> i.e., it doesn't have to be optimized yet, beyond being executed as a single
> Hyracks job that does a series of inserts (into the dataset and its
> associated indexes) - as from a user's perspective today it is REALLY a pain
> to not be able to add more data after an initial load! As a result we see
> folks having to do work around like loading into some other dataset and then
> doing an insert-into, or abusing feeds, or.... Let's fix this! It can't be
> super-hard, as it would essentially operate like a short-lived datafeed under
> the hood.
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
