[
https://issues.apache.org/jira/browse/ASTERIXDB-1496?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15380827#comment-15380827
]
Michael J. Carey commented on ASTERIXDB-1496:
---------------------------------------------
We indeed do - that's a problem right now - it shouldn't be automatically
nullable unless there's a ?.
(Because we specify the type of the field and say ENFORCED - and enforcement of
a type should include enforcement of its optionality. This is a "user model
bug" right now - the type spec for an optional index field should share the
same grammar as the type spec in the datatype definition for any other field.)
> Having an ability to put nullable field information when creating an enforced
> index
> -----------------------------------------------------------------------------------
>
> Key: ASTERIXDB-1496
> URL: https://issues.apache.org/jira/browse/ASTERIXDB-1496
> Project: Apache AsterixDB
> Issue Type: Improvement
> Reporter: Taewoo Kim
> Assignee: Abdullah Alamoudi
>
> It would be nice if we have an ability to put nullable field ("?")
> information when creating an enforced type index to indicate whether the
> field is nullable or not.
> The current syntax
> create index fbSendTimeIdx on FacebookMessages(send-time:datetime) type btree
> enforced;
> The proposed syntax(nullable field)
> create index fbSendTimeIdx on FacebookMessages(send-time:datetime?) type
> btree enforced;
> or
> (non-nullable field)
> create index fbSendTimeIdx on FacebookMessages(send-time:datetime) type btree
> enforced;
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
