cjbj wrote:
> Andy Hassall <andy_at_andyh.co.uk> wrote in message news:<uoii105kpeugqftpbnhqc3egi6tov7dma8_at_4ax.com>...
>
>>On Thu, 29 Jan 2004 08:12:26 -0800, Daniel Morgan <damorgan_at_x.washington.edu>
>>wrote:
>>
>>
>>>Impossible. You can kludge your way around it on a case by case basis
>>>and probably get anything to work ... at least until Oracle releases a
>>>patch that breaks the kludge. But the question I am asking is why, given
>>>all of the characters on the keyboard, anyone would intentionally design
>>>anything using a restricted keyword? Just change to something else and
>>>the entire problem disappears.
>>>
>>
>> [snip]
>>
>>>I'm encouraging you to never use keywords for anything other than their
>>>intended usage.
>>
>> But the OP posted an issue with an @ in a _literal string_, and a PL/SQL
>>_comment_. So you're saying do not use any string present in V$RESERVED_WORDS
>>in a literal string or a PL/SQL comment? That's rather limiting isn't it, given
>>that useful things like spaces are in that view?
>>
>> It's clearly a bug in SQL*Plus (it has no right parsing anything other than
>>substitution variables inside strings or comments, certainly not running them
>>as an SQL*Plus command unless prefixed with the SQLPREFIX character); so how is
>>it a fault on the part of the OP?
>>
>> e.g.
>>
>>SQL> set sqlprefix ~
>>SQL> begin
>> 2 /*
>> 3 @none of sqlplus's business
>>SP2-0310: unable to open file "none.sql"
>> 3 start start.sql
>> 4 ~start start.sql
>>SP2-0310: unable to open file "start.sql"
>> 4
>>
>> (Unless you can find a reference in the manual that says the @ command will be
>>run even when in SQL or PL/SQL mode, regardless of the fact it doesn't make
>>sense - in which I'll stand corrected, but baffled).
>>
>> Keywords shouldn't be used for identifiers, but this isn't about identifiers.
>
>
> SQL*Plus has a very basic command parser. There are some other listed
> "features" at:
>
> http://otn.oracle.com/support/tech/sql_plus/htdocs/runtime2.html#minus_sign
>
> I expect the "@" issue has been a "feature" of SQL*Plus for the last 20 years.
>
> -- CJ
SQL*Plus is irrelevant to the fact that '@' is a keyword.
Rather than pontificating ... run the SQL statement I supplied.
--
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damorgan_at_x.washington.edu
(replace 'x' with a 'u' to reply)
Received on Fri Jan 30 2004 - 00:15:49 CST
