[
https://issues.apache.org/jira/browse/CASSANDRA-9232?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14556912#comment-14556912
]
Tyler Hobbs commented on CASSANDRA-9232:
----------------------------------------
I think the best option is to properly expose the keywords in the driver and
use that in cqlsh so that we can avoid maintaining multiple lists. (The driver
will need to stay updated for {{DESCRIBE}} statements anyway.)
I suppose we should also consider tracking the keywords by Cassandra version so
that when we add new reserved keywords, we don't treat them as reserved in
older Cassandra versions. However, I don't think it's necessarily a bad thing
to preemptively treat keywords as reserved, since users will need to deal with
that before upgrading anyway. So, maybe always using the "newest" keyword list
is okay. What do you think?
The rest of your cleanup looks good to me so far.
> "timestamp" is considered as a reserved keyword in cqlsh completion
> -------------------------------------------------------------------
>
> Key: CASSANDRA-9232
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9232
> Project: Cassandra
> Issue Type: Bug
> Reporter: Michaël Figuière
> Assignee: Stefania
> Priority: Trivial
> Labels: cqlsh
> Fix For: 3.x, 2.1.x
>
>
> cqlsh seems to treat "timestamp" as a reserved keyword when used as an
> identifier:
> {code}
> cqlsh:ks1> create table t1 (int int primary key, ascii ascii, bigint bigint,
> blob blob, boolean boolean, date date, decimal decimal, double double, float
> float, inet inet, text text, time time, timestamp timestamp, timeuuid
> timeuuid, uuid uuid, varchar varchar, varint varint);
> {code}
> Leads to the following completion when building an {{INSERT}} statement:
> {code}
> cqlsh:ks1> insert into t1 (int,
> "timestamp" ascii bigint blob boolean date
> decimal double float inet text time
> timeuuid uuid varchar varint
> {code}
> "timestamp" is a keyword but not a reserved one and should therefore not be
> proposed as a quoted string. It looks like this error happens only for
> timestamp. Not a big deal of course, but it might be worth reviewing the
> keywords treated as reserved in cqlsh, especially with the many changes
> introduced in 3.0.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
