db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kim Haase (JIRA)" <j...@apache.org>
Subject [jira] Updated: (DERBY-3010) WITH DEFAULT value documentation needs improvement
Date Tue, 09 Oct 2007 19:52:52 GMT

     [ https://issues.apache.org/jira/browse/DERBY-3010?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Kim Haase updated DERBY-3010:

    Attachment: DERBY-3010.zip

Attaching DERBY-3010.zip and DERBY-3010.diff. I incorporated your suggested changes to rrefsqlj30540.dita,
Dag, with some tweaks to eliminate passive verbs and inconsistencies. Please make sure I didn't
change the meaning in an unintended way.

I have one question: whether the difference between "must" and "should" in the new text is
meant to be significant. The first two new lines say "must"; the third one says "has to be",
which I assume means "must". But the last one says "should". Would it be correct to change
"should" to "must"? In specifications, "should" and "must" have different meanings -- I think
if something is a "must" then an error occurs if the user disobeys the rule, but if something
is a "should" no error occurs. So I'm wondering which is correct in this case.

I also made a minor change to the column-definition section of the ALTER TABLE topic, rrefsqlj81859.dita,
including a cross-reference to rrefsqlj30540.dita.

> WITH DEFAULT value documentation needs improvement
> --------------------------------------------------
>                 Key: DERBY-3010
>                 URL: https://issues.apache.org/jira/browse/DERBY-3010
>             Project: Derby
>          Issue Type: Improvement
>          Components: Documentation
>    Affects Versions:,,,,,,,,
>            Reporter: Dag H. Wanvik
>            Assignee: Kim Haase
>            Priority: Minor
>         Attachments: defaultConstantExpression.txt, DERBY-3010.diff, DERBY-3010.zip
> The descrition in the reference manual for DEFAULT value option, leads
> one to assume that the value may be specified as a constant
> expression, e.g.  as (0), whereas Derby (in compliance with the SQL
> standard) is more restrictive.
> In the specifications for CREATE and ALTER TABLE (rrefsqlj81859.dita
> and rrefsqlj30540.dita), "ConstantExpression" is used:
> column-definition: 
> :
> [ [ WITH ] DEFAULT { ConstantExpression | NULL }
>   | generated-column-spec ] 
> :
> This non-terminal symbol is explained in rrefsqlj30540.dita:
> "For the definition of a default value, a ConstantExpression is an
> expression that does not refer to any table. It can include constants,
> date-time special registers, current schemas, users, and null."
> Perhaps one could change the name to ConstantValue and explain it
> as:
> "For the definition of a default value, a ConstantValue means literals,
> date-time special registers, current schemas, users, and null."
> Or better, one could be more explicit and spell out the options in
> more detail..

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message