commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Matt Benson <gudnabr...@yahoo.com>
Subject Re: [VOTE] Release JXPath 1.3 based on RC3
Date Fri, 20 Jun 2008 15:04:39 GMT
I do hereby cancel this vote.  I also forgot to update
the release notes with the last bug I entered/fixed in
order to get tests (whose breakage only showed up on
JDK1.3) working again.  Thanks and apologies to those
who checked out this RC.  I think the next one will be
a winner...

-Matt

--- sebb <sebbaz@gmail.com> wrote:

> On 18/06/2008, Matt Benson <gudnabrsam@yahoo.com>
> wrote:
> > Because the findbugs results, as well as the minor
> >  documentation issues that were identified, have
> >  already been addressed in trunk, it probably
> doesn't
> >  make any sense to release 1.3 only to follow it
> >  immediately with 1.3.1 when the content of 1.3.1
> is
> >  already known before 1.3 is released.  Does
> anyone
> >  object to my cancelling this vote and preparing
> RC4?
> >
> 
> OK by me.
> 
> >  Thanks,
> >
> > Matt
> >
> >
> >  --- Matt Benson <gudnabrsam@yahoo.com> wrote:
> >
> >  >
> >  > --- sebb <sebbaz@gmail.com> wrote:
> >  >
> >  > > On 16/06/2008, Matt Benson
> <gudnabrsam@yahoo.com>
> >  > > wrote:
> >  > > >
> >  > > >  --- sebb <sebbaz@gmail.com> wrote:
> >  > > >
> >  > > >
> >  > > > > On 15/06/2008, Oliver Heger
> >  > > >  > <oliver.heger@oliver-heger.de> wrote:
> >  > > >  > > sebb schrieb:
> >  > > >  > >
> >  > > >  > > > On 14/06/2008, Matt Benson
> >  > > >  > <gudnabrsam@yahoo.com> wrote:
> >  > > >  > > >
> >  > > >  > > > >  --- Oliver Heger
> >  > > >  > <oliver.heger@oliver-heger.de> wrote:
> >  > > >  > > > >
> >  > > >  > > > >  > +1
> >  > > >  > > > >  >
> >  > > >  > > > >  > Artifacts look very good. I
> also ran
> >  > > the
> >  > > >  > tests for
> >  > > >  > > > >  > commons configuration
> >  > > >  > > > >  > with the new version
> successfully.
> >  > > >  > > > >  >
> >  > > >  > > > >  > The only thing that makes me a
> bit
> >  > > uneasy
> >  > > >  > is the
> >  > > >  > > > >  > findbugs report showing
> >  > > >  > > > >  > 133 errors. Did you have a look
> at
> >  > > those?
> >  > > >  > > > >  >
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > > I actually didn't, but I don't see
> >  > > anything in
> >  > > >  > there
> >  > > >  > > > >  that really surprises me.  Some
> false
> >  > > >  > positives (e.g.
> >  > > >  > > > >  String ==), some Serialization
> issues
> >  > I
> >  > > knew
> >  > > >  > were
> >  > > >  > > > >  there.  It would be nice to
> attack
> >  > these
> >  > > for
> >  > > >  > another
> >  > > >  > > > >  release.
> >  > > >  > > > >
> >  > > >  > > >
> >  > > >  > > > Certainly some of them need fixing,
> e.g.
> >  > > >  > > >
> >  > > >  > > > Use of non-localized
> String.toUpperCase()
> >  > > or
> >  > > >  > String.toLowerCase
> >  > > >  > > > at
> >  > > >  > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
http://people.apache.org/~mbenson/jxpath-1.3-rc3/site/xref/org/apache/commons/jxpath/ri/model/NodePointer.html#549
> >  > > >  > > > and
> >  > > >  > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
http://people.apache.org/~mbenson/jxpath-1.3-rc3/site/xref/org/apache/commons/jxpath/ri/model/dom/DOMNodePointer.html#330
> >  > > >  > > >
> >  > > >  > > > These should use something like
> >  > > >  > toUpperCase(Locale.ENGLISH).
> >  > > >  > > >
> >  > > >  > > > Might also be worth adding
> exclusions for
> >  > > the
> >  > > >  > bugs that are false
> >  > > >  > > positives...
> >  > > >  > > >
> >  > > >  > > >
> >  > > >  > >  I think if the problems were
> introduced
> >  > > during
> >  > > >  > the work on the 1.3 release,
> >  > > >  > > they really should be addressed.
> However if
> >  > > they
> >  > > >  > live in the code base for a
> >  > > >  > > longer time, they have obviously not
> caused
> >  > > major
> >  > > >  > problems yet, and the
> >  > > >  > > strategy to fix them in the next
> release
> >  > > seems
> >  > > >  > reasonable to me.
> >  > > >  > >
> >  > > >  >
> >  > > >  > Good point.
> >  > > >  >
> >  > > >  > Is this one new?
> >  > > >  >
> >  > > >  > Dead store to collection in
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
org.apache.commons.jxpath.ri.model.beans.CollectionPointer.createPath(JXPathContext)
> >  > > >  >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
http://people.apache.org/~mbenson/jxpath-1.3-rc3/site/xref/org/apache/commons/jxpath/ri/model/beans/CollectionPointer.html#125
> >  > > >  >
> >  > > >  > The code certainly looks odd ...
> >  > > >  >
> >  > > >
> >  > > >
> >  > > > Not new, and from the POV of understanding
> what
> >  > > the
> >  > > >  code there does (grows the underlying
> >  > collection
> >  > > to a
> >  > > >  size such that the index is valid) doesn't
> >  > appear
> >  > > >  problematic.
> >  > >
> >  > > What's confusing is that the collection
> variable
> >  > is
> >  > > assigned, but not used.
> >  > >
> >  >
> >  > Actually the problem here is shadowing.  I'll
> work
> >  > on
> >  > this.  This is a bug, but one that would only
> make a
> >  > difference for arrays as they must be
> reassigned
> >  > while
> >  > a collection would simply be grown.  Not sure
> if
> >  > this
> >  > should be a blocker.  :|
> >  >
> >  > -Matt
> >  >
> >  > > Perhaps remove the assignment to make it
> clearer.
> >  > >
> >  > > >
> >  > > >  > (most of the other dead store reports
> seem to
> >  > > be
> >  > > >  > FPs)
> >  > > >  >
> >  > > >  >
> >  > > >  > If getMessage() is heavily used, then
> this
> >  > one
> >  > > >  > should be fixed:
> >  > > >  >
> >  > > >  > Method
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
org.apache.commons.jxpath.ri.parser.ParseException.getMessage()
> >  > > >  > concatenates strings using + in a loop
> >  > > >
> >  > > >
> >  > > > This is generated code.  It wouldn't hurt
> to fix
> >  > > it,
> >  > > >  since the JXPath parser code is generated
> and
> >  > > then
> >  > > >  saved, but in theory if we ever
> regenerated the
> >  > > parser
> >  > > >  from the JavaCC grammar we'd be starting
> from
> >  > > scratch
> >  > > >  to add back any improvements made.  It
> being
> >  > the
> >  > > case
> >  > > >  that the rightness of making changes to
> >  > generated
> >  > > code
> >  > > >  is in doubt, I certainly wouldn't think
> the
> >  > > release
> >  > > >  should be held up because of it.
> >  > >
> >  > > Sorry, did not notice that was generated
> code.
> >  > > There were plenty of other complaints about
> the
> >  > > generated code which I
> >  > > did ignore.
> >  > >
> >  > > >  So far the only issue I've seen that I
> would
> >  > > think is
> >  > > >  terror-worthy is Phil's discovery of my
> having
> >  > > left in
> >  > > >  some [io] references when I (relatively)
> >  > recently
> >  > > >  cloned its build instructions for a
> much-needed
> >  > > >  update.  He, however, has given his +1.
> >  > > >
> >  > > >  In any event, all these issues should be
> >  > > addressed at
> >  > > >  some point, so I'll go ahead and work on
> these
> >  > in
> >  > > >  trunk.
> >  > > >
> >  > >
> >  > > That's fine by me.
> >  > >
> >  > > >
> >  > > >  -Matt
> >  > > >
> >  > > >
> >  > > >  >
> >  > > >  > >  Oliver
> >  > > >  > >
> >  > > >  > >
> >  > > >  > >
> >  > > >  > >
> >  > > >  > > >
> >  > > >  > > > >  Does your +1 still stand?
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > >  -Matt
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > >  > Oliver
> >  > > >  > > > >  >
> >  > > >  > > > >  > Matt Benson schrieb:
> >  > > >  > > > >  > > Thanks to anyone who reported
> >  > issues
> >  > > with
> >  > > >  > the
> >  > > >  > > > >  > previous
> >  > > >  > > > >  > > two release candidates, and
> >  > > especially to
> >  > > >  > those
> >  > > >  > > > >  > who
> >  > > >  > > > >  > > helped resolve them.
> >  > > >  > > > >  > >
> >  > > >  > > > >  > > The artifacts are here:
> >  > > >  > > > >  > >
> >  > > >  >
> >  > >
> http://people.apache.org/~mbenson/jxpath-1.3-rc3/
> >  > > >  > > > >  > >
> >  > > >  > > > >  > > The tag is here:
> >  > > >  > > > >  > >
> >  > > >  > > > >  >
> >  > > >  > > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
http://svn.apache.org/viewvc/commons/proper/jxpath/tags/JXPATH_1_3_RC3/
> >  > > >  > > > >  > >
> >  > > >  > > > >  > > Site:
> >  > > >  > > > >  > >
> >  > > >  > > > >  >
> >  > > >  > > > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
http://people.apache.org/~mbenson/jxpath-1.3-rc3/site
> >  > > >  > > > >  > >
> >  > > >  > > > >  > > Clirr Report (compared to
> 1.2;
> >  > > one-shot
> >  > > >  > not
> >  > > >  > > > >  > working w/
> >  > > >  > > > >  > > M2)
> >  > > >  > > > >  > >
> >  > > >  > > > >  >
> >  > > >  > > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
http://people.apache.org/~mbenson/jxpath-1.3-rc3/clirr-report.txt
> >  > > >  > > > >  > >
> >  > > >  > > > >  > > I'd be grateful if you can
> make
> >  > time
> >  > > to
> >  > > >  > check the
> >  > > >  > > > >  > > artifacts and cast your vote,
> >  > which
> >  > > will
> >  > > >  > be open
> >  > > >  > > > >  > at
> >  > > >  > > > >  > > least until Friday, June 20.
> >  > > >  > > > >  > >
> >  > > >  > > > >  > > Thanks,
> >  > > >  > > > >  > > Matt
> >  > > >  > > > >  > >
> >  > > >  > > > >  > >
> >  > > >  > > > >  > >
> >  > > >  > > > >  > >
> >  > > >  > > > >  > >
> >  > > >  > > > >  >
> >  > > >  > > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > >  > > > >  > > To unsubscribe, e-mail:
> >  > > >  > > > >  >
> dev-unsubscribe@commons.apache.org
> >  > > >  > > > >  > > For additional commands,
> e-mail:
> >  > > >  > > > >  > dev-help@commons.apache.org
> >  > > >  > > > >  > >
> >  > > >  > > > >  >
> >  > > >  > > > >  >
> >  > > >  > > > >  >
> >  > > >  > > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > >  > > > >  > To unsubscribe, e-mail:
> >  > > >  > > > >  >
> dev-unsubscribe@commons.apache.org
> >  > > >  > > > >  > For additional commands,
> e-mail:
> >  > > >  > > > >  > dev-help@commons.apache.org
> >  > > >  > > > >  >
> >  > > >  > > > >  >
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > >  > > > >  To unsubscribe, e-mail:
> >  > > >  > > dev-unsubscribe@commons.apache.org
> >  > > >  > > > >  For additional commands, e-mail:
> >  > > >  > dev-help@commons.apache.org
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > > >
> >  > > >  > > >
> >  > > >  > > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > >  > > > To unsubscribe, e-mail:
> >  > > >  > > dev-unsubscribe@commons.apache.org
> >  > > >  > > > For additional commands, e-mail:
> >  > > >  > dev-help@commons.apache.org
> >  > > >  > > >
> >  > > >  > > >
> >  > > >  > >
> >  > > >  > >
> >  > > >  > >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > >  > >  To unsubscribe, e-mail:
> >  > > >  > dev-unsubscribe@commons.apache.org
> >  > > >  > >  For additional commands, e-mail:
> >  > > >  > dev-help@commons.apache.org
> >  > > >  > >
> >  > > >  > >
> >  > > >  >
> >  > > >  >
> >  > > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > >  > To unsubscribe, e-mail:
> >  > > >  > dev-unsubscribe@commons.apache.org
> >  > > >  > For additional commands, e-mail:
> >  > > >  > dev-help@commons.apache.org
> >  > > >  >
> >  > > >  >
> >  > > >
> >  > > >
> >  > > >
> >  > > >
> >  > > >
> >  > > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > >  To unsubscribe, e-mail:
> >  > > dev-unsubscribe@commons.apache.org
> >  > > >  For additional commands, e-mail:
> >  > > dev-help@commons.apache.org
> >  > > >
> >  > > >
> >  > >
> >  > >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > > To unsubscribe, e-mail:
> >  > > dev-unsubscribe@commons.apache.org
> >  > > For additional commands, e-mail:
> >  > > dev-help@commons.apache.org
> >  > >
> >  > >
> >  >
> >  >
> >  >
> >  >
> >  >
> >  >
> > 
>
---------------------------------------------------------------------
> >  > To unsubscribe, e-mail:
> >  > dev-unsubscribe@commons.apache.org
> >  > For additional commands, e-mail:
> >  > dev-help@commons.apache.org
> >  >
> >  >
> >
> >
> >
> >
> >
> > 
>
---------------------------------------------------------------------
> >  To unsubscribe, e-mail:
> dev-unsubscribe@commons.apache.org
> >  For additional commands, e-mail:
> dev-help@commons.apache.org
> >
> >
> 
>
---------------------------------------------------------------------
> To unsubscribe, e-mail:
> dev-unsubscribe@commons.apache.org
> For additional commands, e-mail:
> dev-help@commons.apache.org
> 
> 



      

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Mime
View raw message