Totally agree that we should not shoot ourselves in the foot again.
The branches are complete and I have a fix in place on the branch.
I added a tags and branches directory to the release dir
release/
tags/
1_1_0 <- is what we released
branches/
1_1_0 <- is what I've fixed and and we will build our
release from
Once we are sure that we have a good build we can create a new tag
release/tags/1_1_0_1
I've run the simple app with the 3 constituent jar files and myfaces-
all and what I've tested is working.
TTFN,
TTFN,
-bd-
On Sep 23, 2005, at 11:24 AM, Sean Schofield wrote:
OK just found this thread from yesterday afternoon. Man if you leave
your computer for a few hours I guess you miss a lot of action!
We definitely should run the TCK before releasing and we should test
everything (for real this time) and vote. This wil take about a week
but we rushed things last time and look what it got us into ...
I agree the problem is urgent but somehow we managed to shoot
ourselves in the foot again with the release. This time it was
announcing the TCK passage before we had a release candidate. This
created the sense of urgency to get the new release out and
apprarently we did not test things very thoroughly.
In the meantime we could put something on the website explaining that
myfaces-all.jar doesn't work and to use impl, api and tomahawk instead
for now. We can also promise an upcoming release soon. Perhaps we
could do a release candidate later today once Bill is finished with
the branch and fixes.
sean
On 9/23/05, Bill Dudney <[EMAIL PROTECTED]> wrote:
checked into the trunk
Thanks Martin!
-bd-
On Sep 23, 2005, at 6:16 AM, Martin Marinschek wrote:
You are right - no question.
we should definitely rerun the TCK if possible.
Oliver is on holidays, remains Manfred to do so...
Manfred?
;)
regards,
Martin
On 9/23/05, Bill Dudney <[EMAIL PROTECTED]> wrote:
Hi All,
This is the essence of what I fixed. I've rebuilt and used the
myfaces-all.jar in the the 'simple.war' test application and
everything I tested worked.
Instead of 'dist-all' use 'release'.
The only problem I see with this approach is that we could
introduce
unintended consequences by releasing from the trunk (see my
response
to Sean just moments ago). We can't be 100% sure that no one has
committed a TCK breaking change since the release went out so in
theory we should rerun the tests before pushing out a new
release. If
we make the change in a branch then we can more easily assert that
nothing has changed except the build file.
My $0.02 worth and I'm not against a release from the trunk its
just
another level of risk.
TTFN,
-bd-
Theoretically we should be doing a patch to the release branch and
releasing from there (with the -Dskip.sandbox=true that Bruno
mentions).
On Sep 23, 2005, at 5:57 AM, Bruno Aranda wrote:
Remember to release without the sandbox. In theory you should use
'ant
-Dskip.sandbox=true dist-all', but that is the origin of the
bug. I
cannot look at it right now...
Bruno
2005/9/23, Martin Marinschek <[EMAIL PROTECTED]>:
Bill,
please go ahead and check in what you have got now. If you do it
anytime during the next hour, I will go ahead and prepare a new
release from this so that we get this faces-config problem out
of the
world as fast as possible!
regards,
Martin
On 9/22/05, Bill Dudney <[EMAIL PROTECTED]> wrote:
I have a lot of other changes related to the cactus/cargo stuff.
I don't have a problem checking in all that but there has not
been
much discussion about it yet. I'd rather see some discussion
happen
before I check in the rest of the changes.
TTFN,
-bd-
On Sep 22, 2005, at 1:38 PM, Mike Kienenberger wrote:
It's got to be checked into the trunk at some point. I don't
see any
advantages in waiting.
On 9/22/05, Bill Dudney <[EMAIL PROTECTED]> wrote:
OTOH, I could checkin everything including the cactus & cargo
changes
and that would fix things on the trunk.
Thoughts?
-bd-
On Sep 22, 2005, at 1:25 PM, Bill Dudney wrote:
No I've not checked it in yet because I'm waiting for
discussion on
the idea of making a 1_1_0 branch that we could do the
emergency
release from.
TTNF,
-bd-
On Sep 22, 2005, at 1:20 PM, Mike Kienenberger wrote:
Is this supposed to be fixed in svn?
I just did a complete checkout about an hour ago, and built
it with
"ant dist-all" and I'm getting the error below if I use
myfaces-
all,
but not if I use myfaces.api, myfaces.impl, and tomahawk
separately.
java.lang.NoClassDefFoundError: org/apache/myfaces/config/
FacesConfigurator
at
org.apache.myfaces.webapp.StartupServletContextListener.init
Fa
ce
s
(StartupServletContextListener.java:63)
at
org.apache.myfaces.webapp.StartupServletContextListener.cont
ex
tI
nit
ia
lized(StartupServletContextListener.java:46)
Or is this a case of the cure being worse than the disease?
-Mike
On 9/22/05, Bill Dudney <[EMAIL PROTECTED]> wrote:
Since this is simply a build problem perhaps we could copy
the tag
1_1_0 to a branch and update only the build.xml file then
build a
new
release.
This would be simple, quick and more or less safe.
Would we have to rerun the TCK? As long as we trust our
tag it
should
be fine. Correct?
TTFN,
-bd-
On Sep 22, 2005, at 12:08 PM, Werner Punz wrote:
Bruno Aranda wrote:
This is causing many troubles to users now as we can see
from
questions in the list.
faces-config.xml is not included in the build because of
MYFACES-598,
when passing -Dskip.sandbox to the build process the file
is not
included. I think this is important enough and we
should do
something
about this. I don't know if it is possible to include the
file
in the
current release but I don't think it is good to
development to
touch
an already released thing.
My opinion here is that we should release a maintenance
release
1.1.1
as soon as possible. Waiting for thoughts on this.
Regards,
Bruno
I cannot vote yet, but due to the fact that many newbys
probably
will try to use the
new release, due to the non beta status and TCK status,
and this
is really a showstopper for many, I´d say get the fix
out ASAP!
--
http://www.irian.at
Your JSF powerhouse -
JSF Trainings in English and German
--
http://www.irian.at
Your JSF powerhouse -
JSF Trainings in English and German
