[ 
https://issues.apache.org/jira/browse/DERBY-6103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13635987#comment-13635987
 ] 

Dag H. Wanvik commented on DERBY-6103:
--------------------------------------

Thanks, Kim! The syntax looks correct now. We can probably defer factoring out 
WindowSpecification to a separate topic since its currently trivial. Just two 
more nits and we should be done:

ROW_NUMBER function topic:

"You can specify a WindowName in order to use a WINDOW clause in a 
SelectExpression. "

This sentence, while correct, sounds a bit rationale inverted: One would use a 
WindowName iff one chose to use an explicit WindowClause in a SelectExpression 
rather than using an implicit (inlined) window specification. Currently, this 
possibility doesn't give you much, admittedly, but it's legal SQL syntax.

Window Clause topic:

"Currently, the only valid WindowSpecification is a set of empty parentheses 
(())."

Correct, but we should add that this means that the effective window is the 
entire result set - as you do say in the ROW_NUMBER function topic.


                
> Improve documentation of ROW_NUMBER function
> --------------------------------------------
>
>                 Key: DERBY-6103
>                 URL: https://issues.apache.org/jira/browse/DERBY-6103
>             Project: Derby
>          Issue Type: Improvement
>          Components: Documentation
>    Affects Versions: 10.10.1.1
>            Reporter: Dag H. Wanvik
>            Assignee: Kim Haase
>            Priority: Minor
>         Attachments: DERBY-6103-2.diff, DERBY-6103-2.stat, DERBY-6103-2.zip, 
> DERBY-6103-3.diff, DERBY-6103-3.zip, DERBY-6103.diff, DERBY-6103.stat, 
> DERBY-6103.zip
>
>
> The current documentation doesn't mention that we allow the window definition 
> to be explicit.
> The restriction on WHERE clause no longer applies, nor does the sentence 
> about ORDER BY not being allowed in subqueries. Cf. rreffuncrownumber.dita. 
> Also, the documentation mentions that we don't support LIMIT, but not that we 
> now support the FETCH/OFFSET clause, which is often a better option than 
> ROW_NUMBER.

--
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: http://www.atlassian.com/software/jira

Reply via email to