myfaces-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Bruno Aranda <brunoara...@gmail.com>
Subject Re: Outstanding webapp issues
Date Wed, 06 Jul 2005 14:56:08 GMT
Poor Mathias, he is under a lot of pressure. I would let him do for a
while, at least, there are a few other issues that must also be solved
in the meanwhile. It is always time to rollback. I would make the
decision wether to rollback or not when arrives the time to announce
the end of the reorganization. I would keep the patch applied for a
while, because it allows people to test it heavily. Everyone of us use
the dataTable on a different case...

Regards,

Bruno


2005/7/6, Sean Schofield <sean.schofield@gmail.com>:
> Manfred,
> 
> Its my fault for committing.  Mathias just submitted a patch for
> people to play with.  I did a quick test but I thought it would be
> easier for us to comment on code in SVN instead of trying to keep
> track of a series of patches.
> 
> Do you want me to roll it back?  I think there have been some good
> improvements in the implementation.  My vote would be to let Mathias
> continue working on it as long as he has time.  We can wait a few days
> ... If this is a problem though I will roll back and we can let him
> work offline.
> 
> sean
> 
> 
> On 7/6/05, Manfred Geiler <manfred.geiler@gmail.com> wrote:
> > Ok, looks like I have to jump into the "bad cop" role now.  :-(
> >
> > DataTable is a very very delicate component. I know there were some
> > long outstanding issues, that need to be fixed.
> > BUT, DataTable supported some important things that only occured in
> > very complex scenarious. This makes the code look awful and patchy in
> > some places (escpecially for the nested table stuff), but all those
> > things were extensively reconsidered and tested, believe me. MyFaces
> > DataTable solved many issues that RI DataTable had for a long time
> > (and still has I think).
> > Please do not get me wrong. I know that the current situation, where
> > the original author (=me ;-) currently does not have enough time to
> > dig deep into the implemention, is not desirable. Sorry for that.
> > Therefore I really appreciate the fantastic work and help of Sean and
> > Mathias and I try to phrase careful enough to not intimidate someone.
> > ;-)
> > So, please be very very careful with the DataTable. Mathias, if you
> > have really made it to get DataTable easier and clearer, that would
> > really be great, but please: test, test, test. There will be many
> > sleeping issues that you might not be aware of by now, believe me. I
> > remember the hard days (and nights) when I wrote the original
> > DataTable component.
> >
> > No need for panic of course. If you are not able to fix all issues
> > with features that have worked already (eg. preserveDataTable!) in
> > reasonable time, we should decide about a rollback scenario. This
> > would take out the pressure and would give us more time for a more
> > relaxed discussion.
> >
> > Anyway, thanks Mathias for dealing with this "hot iron" and please
> > keep us up to date.
> >
> > Thanks,
> > -Manfred
> >
> > PS. I would not really be the right cast for a "bad cop", would I?  :-)
> >
> >
> >
> > 2005/7/6, Martin Marinschek <martin.marinschek@gmail.com>:
> > > I have just tried out the simple examples - the dataTable
> > > implementation seems not to be working anymore.
> > >
> > > Now I don't know if this is an issue with the new dataTable
> > > implementation or with the svn reorg, but we should definitely try to
> > > fix that.
> > >
> > > regards,
> > >
> > > Martin
> > >
> > > On 7/6/05, Bruno Aranda <brunoaranda@gmail.com> wrote:
> > > > Two more build issues (easy ones). I write them here as a reminder to
> > > > create a patch later (if no one creates one before, feel free to do
> > > > so):
> > > >
> > > > - When building only a subproject with 'ant dist', associated files
> > > > (*.tld's etc) do not make into the jar as the build.properties
> > > > (situated in the subproject parent folder) is not read.
> > > > - It is missing the dependency on sandbox.jar in the general
> > > > dependencies, as if you want to create an example backing bean that
> > > > use a class from the sandbox, the class is not found. This has
> > > > happened to me while preparing an example for the Jurgen's schedule
> > > > component. Easy, just include the dependency. I could have done that
> > > > in the time of writing this lines...
> > > >
> > > > Regards,
> > > >
> > > > Bruno
> > > >
> > > > 2005/7/5, Sean Schofield <sean.schofield@gmail.com>:
> > > > > Thanks to Bruno for helping me test the webapps after the new build.
> > > > > Here is a list of outstanding issues that he gave me:
> > > > >
> > > > > Bruno's List
> > > > > =========
> > > > >
> > > > > 08:55:25,671 ERROR [PropertyResolverImpl]
> > > > > org.apache.myfaces.examples.misc.Country
> > > > > javax.faces.el.PropertyNotFoundException: Bean:
> > > > > org.apache.myfaces.examples.misc.Country, property: cities
> > > > >
> > > > > It seems that it is looking for the bean
> > > > > org.apache.myfaces.examples.misc.Country when it should look for
> > > > > SimpleCountry.
> > > > >
> > > > > - sandbox.war: The [SOURCE] link does not work because the source
code
> > > > > servlet system is not present. (I think you already know that one
;-).
> > > > >
> > > > > - tiles.war: Not working, but I don't know too much about tiles.
It is
> > > > > looking for a missing renderer (UIViewRoot?).
> > > > >
> > > > > I don't know which is the exact state of the reorg, but it seems
to me
> > > > > that it is almost complete. Have you applied the patch for the
> > > > > AddResource thing, or have you thought an alternative solution? We
> > > > > could also ask Sylvain as he could shed more light on this.
> > > > >
> > > > > And there is also a trivial issue: there is a link to the rssTicker
in
> > > > > the standard examples.
> > > > >
> > > > > * * *
> > > > >
> > > > > Also, there are a few other issues that I intentionally skipped over
> > > > > for now.  The WAP example is not being built (nor is the WAP code.)
> > > > > The dynamic code generation is not being done.  We should probably
> > > > > tweak the build scripts to do this.  Ideally, this can be done in
a
> > > > > manner similar to the rest of the build.  That is, you can run foo-all
> > > > > from the top level build dir or foo from the subproject.
> > > > >
> > > > > If someone wants to help with patches to any of these problems that
> > > > > would be helpful.  Just post to the dev list so we all know which
> > > > > issues have been addressed.  For now I don't think we need JIRA since
> > > > > we really should be able to finish these off quickly and they are
top
> > > > > priority.  If feel they should go in JIRA that is fine also.
> > > > >
> > > > > sean
> > > > >
> > > >
> > >
> >
>

Mime
View raw message