cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sylvain Lebresne (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-3783) Add 'null' support to CQL 3.0
Date Fri, 01 Mar 2013 13:25:14 GMT


Sylvain Lebresne commented on CASSANDRA-3783:

bq. It would also be convenient if CQL3 supported null values for collections.

I'm afraid this is not so simple I'm afraid. Doing that would basically require having an
internal representation of null for every type (because in that case, we would be talking
of actually storing a null value, not just deleting a column). And we don't have an easy candidate
for that internal representation currently. Meaning we would have to change a bit the internal
representation of collection columns (say, add some flags somewhere in the column name that
say "this is a null element"). But I'm not convinced we can do that without breaking backward
compatibility really. In any case, this is a very special case so we should handle that in
a followup ticket if we do handle. But again, not sure how doable that is in practice.

Say I have 
ALTER TABLE metrics ADD valueBySecond list<double>;
Just for the record, this sound like a bad use of a list. Because valueBySecond sound like
something that by design will grow fairly big. Remember that currently you cannot fetch parts
of a collection. And even when we allow fetching part of a collection, we'll only be able
to do it for maps and sets, not for lists. This is typically a case where you'd rather want
to have a composite PK.
> Add 'null' support to CQL 3.0
> -----------------------------
>                 Key: CASSANDRA-3783
>                 URL:
>             Project: Cassandra
>          Issue Type: Sub-task
>          Components: API
>            Reporter: Sylvain Lebresne
>            Assignee: MichaƂ Michalski
>            Priority: Minor
>              Labels: cql3
>             Fix For: 1.2.3
>         Attachments: 3783-v2.patch, 3783-wip-v1.patch
> Dense composite supports adding records where only a prefix of all the component specifying
the key is defined. In other words, with:
> {noformat}
> CREATE TABLE connections (
>    userid int,
>    ip text,
>    port int,
>    protocol text,
>    time timestamp,
>    PRIMARY KEY (userid, ip, port, protocol)
> {noformat}
> you can insert
> {noformat}
> INSERT INTO connections (userid, ip, port, time) VALUES (2, '', 80, 123456789);
> {noformat}
> You cannot however select that column specifically (i.e, without selecting column (2,
'', 80, 'http') for instance).
> This ticket proposes to allow that though 'null', i.e. to allow
> {noformat}
> SELECT * FROM connections WHERE userid = 2 AND ip = '' AND port = 80 AND protocol
= null;
> {noformat}
> It would then also make sense to support:
> {noformat}
> INSERT INTO connections (userid, ip, port, protocol, time) VALUES (2, '',
80, null, 123456789);
> {noformat}
> as an equivalent to the insert query above.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:

View raw message