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: svn commit: r372240 - in /db/derby/code/trunk/java: client/org/apache/derby/loc/clientmessages_en.properties shared/org/apache/derby/shared/common/reference/SQLState.java
Date Wed, 25 Jan 2006 19:43:03 GMT
David W. Van Couvering wrote:

> I got confused by your statement because I didn't see 0A003 in the
> table, but then I found this statement further up.  I'm quoting it here
> so that all of us working on SQL States are aware of the restriction.  A
> SQLState "class" is the first two characters of the state string.  A
> SQLState "subclass" are the last three characters of the state string.
> 
> <quote>
> Class values that begin with one of the <digit>s '0', '1', '2', '3', or
> '4' or one of the <simple Latin upper case letter>s
> 'A', 'B', 'C', 'D', 'E', 'F', 'G', or 'H' are returned only for
> conditions defined in ISO/IEC 9075 or in any other
> International Standard. The range of such class values are called
> standard-defined classes. Some such class
> codes are reserved for use by specific International Standards, as
> specified elsewhere in this Clause. Subclass
> values associated with such classes that also begin with one of those 13
> characters are returned only for conditions
> defined in ISO/IEC 9075 or some other International Standard. The range
> of such class values are called standard defined
> classes. Subclass values associated with such classes that begin with
> one of the <digit>s '5', '6', '7', '8',
> or '9' or one of the <simple Latin upper case letter>s 'I', 'J', 'K',
> 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W',
> 'X', 'Y', or 'Z' are reserved for implementation-specified conditions
> and are called implementation-defined subclasses.
> </quote>
> 
> In terms of having separate messages for JDBC2 and JDBC3, I was doing a
> cut-and-paste.  I could change this, I don't have any strong opinions
> about it, but it may very well create a lot of test diffs.  It's
> personally not an itch I'm motivated to scratch.

Since the diffs will occur anyway when you change the SQLState, it could
be combined. Changing the SQLState to be valid does seem to be the itch
your are scratching, given the description of DERBY-400.

Dan.


Mime
View raw message