db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel John Debrunner <...@apache.org>
Subject Re: [jira] Commented: (DERBY-591) Several builtin functions are incorrectly exposed through the JDBC escaped function syntax {fn <function_call>}
Date Fri, 10 Feb 2006 20:09:08 GMT
Bernt M. Johnsen wrote:
> Moves to the mailing list.
> 
> 
>>>>>>>>>>>>>Daniel John Debrunner (JIRA) wrote (2006-02-10
15:54:59):
>>
>>    [ http://issues.apache.org/jira/browse/DERBY-591?page=comments#action_12365900
] 
> 
> 
>>Daniel John Debrunner commented on DERBY-591:
>>---------------------------------------------
>>
>>Not sure what you are asking here Bernt. This is a bug reporting
>>that several Derby functions are incorrectly exposed thorugh the
>>JDBC escaped function mechanism.  It's not about implementing more
>>escaped functions.
> 
> 
> Sorry if I was unclear. 
> 
> I'll try again: Sateesh addresses those functions in SQL which are
> *not* a part of the JDBC escaped-function-set as defined in the JDBC
> spec (unless I misinterpreted). I want also to address the functions
> defined in the JDBC-spec which have other names (and possibly other
> arguments/argument-syntax) in the SQL standard. They're all
> implemented today as if the JDBC escaped names are equal to the SQL
> names.
> 
> Was that clearer?

Not really, but a little better, I missed you were going the other way. :-)

I don't know what you mean by "address the functions", what do you want
to do with them? Satheesh is "addressing the functions" by removing the
current incorrect and undocumented visibilty of SQL functions through
the JDBC escape. I don't think you mean that you want to remove SQL
functions that exist today, are documented and happen to have the same
functionality as JDBC escaped functions.

Possibly you seem to mean that if there is a JDBC escape function {fn
X()} then there should be a SQL function X(). But that doesn't seem to
agree with your last sentence where you say they are implemented as
though the names are the same.

If you are saying that we should be adding functions to Derby that match
the name of some JDBC escaped functions. I think this needs to be on a
case by case basis, for example I'm not sure why Derby would need a SQL
curtime() function when it already has current time and current_time.

Sorry if I'm being dense.
Dan.



Mime
View raw message