You are viewing a plain text version of this content. The canonical link for it is here.
Posted to derby-dev@db.apache.org by Samuel Andrew McIntyre <fu...@nonintuitive.com> on 2004/09/28 23:30:33 UTC
release status as of 9/28/04
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello all,
Here's a summary of the movement on the various issues concerning
getting a first Derby release out there.
1 - IP/copyright concerns - It appears this has ballooned into a more
serious issue than previously thought. Discussion of the proper way to
handle the transition of the code to the ASF license and in what way
IBM copyright is retained, if any, is occurring at
licensing@apache.org. I am sure we will hear from the Incubator PMC as
soon as this issue is settled.
2 - Apply all currently pending bug fixes. I have the following list of
submitted patches:
[APPLIED] [PATCH] Optimization of
org.apache.derby.impl.services.uuid.BasicUUID.toByteArray()
[APPLIED] [PATCH] Set Derby's build number to be the subversion
revision number
[PATCH] derby.war file build
[PATCH] Fix to prevent empty log file switches that could cause
recovery failures
[PATCH] minor bugs in dblook
[PATCH] Extension Packaging
[PATCH] retrieveMessages... true by default in ij
[PATCH] Network servlet display only message key
[PATCH] added 3 more parser generated files to the clobber target in
main build.xml
[PATCH] Various fixes to javadoc generation
Please let me know if there are any I am missing here. Note that the
change copyright to ASF patch is not included here because of issue #1.
3 - Now that we have bug tracking, please speak up if you consider any
of the currently logged bugs as a showstopper for the release.
4 - Please comment on what the content and form of the release should
be. I think jars, javadoc and HTML documentation should be included,
since that is what we have at the moment, packaged as a zip and tar.gz
file. Please speak up if you are interested in rpms for this first
release.
5 - Consensus on Derby's upgrade policy/version scheme. There is
currently a vote pending in the "[VOTE] Derby upgrade policy" thread.
Please post other concerns if you have them.
Thanks,
andrew
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (Darwin)
iD8DBQFBWdf/DfB0XauCH7wRAnIpAKCQuibGpX2ECw9ubcLsWYL4GiAyvQCfZUlV
nyIZCZRKcFlZG7bwkjgHUBs=
=tf6x
-----END PGP SIGNATURE-----
Re: release status as of 9/28/04
Posted by Jack Klebanoff <kl...@Mutagen.Net>.
There is one bug fix that was not applied. Being new to Apache I did not
include "[PATCH]" in the subject.
My original email is:
| I would like to submit a fix for a bug in triggers.
|
| The bug is that a trigger of the form:
| create trigger ... values myFunction();
| has no effect. MyFunction is not called even if the trigger is fired.
| Side effects of myFunction do not happen. Derby does not allow a "CALL"
| statement inside a trigger action so a values statement is the only way
| to call a function/procedure in a trigger action.
|
| The cause of the bug is that since the values are not actually used by
| the trigger, the trigger code does not fetch the row(s) returned by the
| trigger action. The fix is simple: change class
| org.apache.derby.impl.sql.execute.GenericTriggerExecutor to fetch (and
| discard) the rows returned by a trigger action.
|
| Please review the change. The diff file is attached.
Index: java/engine/org/apache/derby/impl/sql/execute/GenericTriggerExecutor.java
===================================================================
--- java/engine/org/apache/derby/impl/sql/execute/GenericTriggerExecutor.java (revision 37092)
+++ java/engine/org/apache/derby/impl/sql/execute/GenericTriggerExecutor.java (working copy)
@@ -157,7 +157,18 @@
*/
try
{
- ps.execute(spsActivation, false, false, false);
+ ResultSet rs = ps.execute(spsActivation, false, false, false);
+ if( rs.returnsRows())
+ {
+ // Fetch all the data to ensure that functions in the select list or values statement will
+ // be evaluated and side effects will happen. Why else would the trigger action return
+ // rows, but for side effects?
+ // The result set was opened in ps.execute()
+ while( rs.getNextRow() != null)
+ {
+ }
+ }
+ rs.close();
}
catch (StandardException e)
{
===================================================================
Samuel Andrew McIntyre wrote:
>
> Hello all,
>
> Here's a summary of the movement on the various issues concerning
> getting a first Derby release out there.
>
> 1 - IP/copyright concerns - It appears this has ballooned into a more
> serious issue than previously thought. Discussion of the proper way to
> handle the transition of the code to the ASF license and in what way
> IBM copyright is retained, if any, is occurring at
> licensing@apache.org. I am sure we will hear from the Incubator PMC as
> soon as this issue is settled.
>
> 2 - Apply all currently pending bug fixes. I have the following list
> of submitted patches:
>
> [APPLIED] [PATCH] Optimization of
> org.apache.derby.impl.services.uuid.BasicUUID.toByteArray()
> [APPLIED] [PATCH] Set Derby's build number to be the subversion
> revision number
> [PATCH] derby.war file build
> [PATCH] Fix to prevent empty log file switches that could cause
> recovery failures
> [PATCH] minor bugs in dblook
> [PATCH] Extension Packaging
> [PATCH] retrieveMessages... true by default in ij
> [PATCH] Network servlet display only message key
> [PATCH] added 3 more parser generated files to the clobber target in
> main build.xml
> [PATCH] Various fixes to javadoc generation
>
> Please let me know if there are any I am missing here. Note that the
> change copyright to ASF patch is not included here because of issue #1.
>
> 3 - Now that we have bug tracking, please speak up if you consider any
> of the currently logged bugs as a showstopper for the release.
>
> 4 - Please comment on what the content and form of the release should
> be. I think jars, javadoc and HTML documentation should be included,
> since that is what we have at the moment, packaged as a zip and tar.gz
> file. Please speak up if you are interested in rpms for this first
> release.
>
> 5 - Consensus on Derby's upgrade policy/version scheme. There is
> currently a vote pending in the "[VOTE] Derby upgrade policy" thread.
>
> Please post other concerns if you have them.
>
> Thanks,
> andrew
sync to maven repo, was: release status as of 9/28/04
Posted by Jeremy Boynes <jb...@gluecode.com>.
Daniel John Debrunner wrote:
>
> Samuel Andrew McIntyre wrote:
>
>
>>4 - Please comment on what the content and form of the release should
>>be. I think jars, javadoc and HTML documentation should be included,
>>since that is what we have at the moment, packaged as a zip and tar.gz
>>file. Please speak up if you are interested in rpms for this first
>
> I think there should be a packaging of the release that is just the jar
> files, no documentation included.
>
Please can you incorporate distribution of jars to the maven repository
on ibiblio so that they can easily be used in maven projects. I believe
it is sufficient to physically place the jars in the right location at
the ASF and they will then be automatically rsync'ed to ibiblio.
If you are familiar with maven please ignore the rest; if not this will
explain why this is useful.
When you write a maven project you can declare in your project.xml file
dependencies on external artifacts such as jar files. If those are not
available on your machine, maven will automatically download them once
from a central repository. This defaults to http://www.ibiblio.org/maven
For this to work, maven expects the files to be stored in certain way so
it knows what to look for. The default path is:
${groupId}/${type}s/${artifactId}-${version}.${type}
where the variables are defined in your project.xml file. For example,
for Derby version 1.0 the dependency would be:
<dependency>
<groupId>incubator-derby
<version>1.0
and the resulting path:
derby/jars/incubator-derby-1.0.jar
Once the file is download maven will not check the remote repo again;
there is a special version "SNAPSHOT" intended for nightly builds which
causes maven to compare timestamps every time and download again if the
remote version is newer.
At least, that is how it works now - there are discussion on repository@
about adopting a standard repo layout but IMHO there are enough users
vested already to warrant keeping the existing format.
Thanks
--
Jeremy
Re: release status as of 9/28/04
Posted by Daniel John Debrunner <dj...@debrunners.com>.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Samuel Andrew McIntyre wrote:
>
> 4 - Please comment on what the content and form of the release should
> be. I think jars, javadoc and HTML documentation should be included,
> since that is what we have at the moment, packaged as a zip and tar.gz
> file. Please speak up if you are interested in rpms for this first
release.
I think there should be a packaging of the release that is just the jar
files, no documentation included.
Dan.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFBYZXSIv0S4qsbfuQRAjvnAKDjoAM8f1WJ1SE8/4rMsS9Zv+y9AgCgrZvu
sDqcdTHZy98E+sVUWadtM8M=
=x1Tp
-----END PGP SIGNATURE-----
Re: release status as of 9/28/04
Posted by myrnap <my...@Golux.Com>.
Samuel Andrew McIntyre wrote:
>
> 3 - Now that we have bug tracking, please speak up if you consider any
> of the currently logged bugs as a showstopper for the release.
Jira bug DERBY-24 needs to be fixed.
Myrna