cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dave Brosius (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (CASSANDRA-6136) CQL should not allow an empty string as column identifier
Date Thu, 03 Oct 2013 12:12:42 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-6136?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13785028#comment-13785028
] 

Dave Brosius edited comment on CASSANDRA-6136 at 10/3/13 12:10 PM:
-------------------------------------------------------------------

version3 disallows at the parse layer.


was (Author: dbrosius):
version2 disallows at the parse layer.

> CQL should not allow an empty string as column identifier
> ---------------------------------------------------------
>
>                 Key: CASSANDRA-6136
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-6136
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Michaël Figuière
>            Assignee: Dave Brosius
>            Priority: Minor
>         Attachments: 6136.txt, 6136_v2.txt, 6136_v3.txt
>
>
> CQL currently allows users to create a table with an empty string as column identifier:
> {code}
> CREATE TABLE t (k int primary key, "" int);
> {code}
> Which results in the following table:
> {code}
> CREATE TABLE t (
>   k int,
>   "" int,
>   PRIMARY KEY (k)
> ) WITH
>   bloom_filter_fp_chance=0.010000 AND
>   caching='KEYS_ONLY' AND
>   comment='' AND
>   dclocal_read_repair_chance=0.000000 AND
>   gc_grace_seconds=864000 AND
>   index_interval=128 AND
>   read_repair_chance=0.100000 AND
>   replicate_on_write='true' AND
>   populate_io_cache_on_flush='false' AND
>   default_time_to_live=0 AND
>   speculative_retry='NONE' AND
>   memtable_flush_period_in_ms=0 AND
>   compaction={'class': 'SizeTieredCompactionStrategy'} AND
>   compression={'sstable_compression': 'SnappyCompressor'};
> {code}
> Empty strings are not allowed for keyspace and table identifiers though.
> I guess it's just a case that we haven't covered. Of course making it illegal in a future
version would be a breaking change, but nobody serious would manually have chosen such an
identifier...



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message