ok, give me a bit to run through it.
-Mark
On Mon, Oct 14, 2013 at 10:35 AM, Alex Harui wrote:
> IMO, no. In fact you are encouraged to test with RC1 to see if there are
> any other things broken that need to be addressed in an RC2.
>
> On 10/14/13 7:20 AM, "Mark Kessler" wrote:
>
> >So does
On 10/14/13 2:06 AM, "Erik de Bruin" wrote:
>> I think (even though it's a little painful) we have go with the tests
>>passing as tagged (even though they are not part of the source release)
>>otherwise a new RC is needed. Otherwise there's no way of reproducing
>>the test results accurately. I
IMO, no. In fact you are encouraged to test with RC1 to see if there are
any other things broken that need to be addressed in an RC2.
On 10/14/13 7:20 AM, "Mark Kessler" wrote:
>So does this mean I should wait for rc2 before testing? since I'm a little
>late here atm.
>
>-Mark
So does this mean I should wait for rc2 before testing? since I'm a little
late here atm.
-Mark
> I think (even though it's a little painful) we have go with the tests passing
> as tagged (even though they are not part of the source release) otherwise a
> new RC is needed. Otherwise there's no way of reproducing the test results
> accurately. In this case carrying over the previous votes w
Hi,
> My point is that if I get a working copy from the tag, I don't get the
> updated Mustella tests.
And that's a valid point.
> That means that if I run the Mustella tests that comes with the code from the
> tag, I get failures.
There would be errors yes.
> I don't want failures on an RC. In
Thanks for the clarification.
Maurice
-Message d'origine-
De : Justin Mclean [mailto:jus...@classsoftware.com]
Envoyé : lundi 14 octobre 2013 10:04
À : dev@flex.apache.org
Objet : Re: [DISCUSSION] Apache Flex 4.11.0 RC 1
Hi,
Just to clear up any confusion, RC1 was tagged as &q
Hi,
> IMHO, to get a clean RC2, we need a new branch of develop
Currently release and develop are identical so all we need to do is wait for a
full pass of the CI and then make RC2?
> Since that likely won't happen, and I don't want to actively oppose
> this release anymore, I've changed my vote
Ok, one more before I will actually shut up about this:
My point is that if I get a working copy from the tag, I don't get the
updated Mustella tests. That means that if I run the Mustella tests
that comes with the code from the tag, I get failures. I don't want
failures on an RC. In order to 'fix
Hi,
> Actually, there is a tag labeled "apache-flex-sdk-4.11.0RC1" on the git repo,
> that is dated at 10/05/2013.
> Does this one correspond to the current RC1, in terms of content?
Yep 100% but it doesn't include the update to the tests to make them pass.
Thanks,
Justin
Hi,
> Procedurally, I don't think you can cherry pick parts of the code to
> get something that 'works'.
And cherry picking is a painful to do anyway, I'd like to avoid it if we can.
> I think that you should be able to go to git, say "give me the code as it was
> when we cut RC1"
That exactly w
Hi,
Just to clear up any confusion, RC1 was tagged as "apache-flex-sdk-4.11.0RC1"
and that is 100% identical to what's in the source release candidate. This
could be checked by making a source zip and comparing the hashes. The have been
changes made in both develop and release since then but th
undi 14 octobre 2013 09:16
À : dev@flex.apache.org
Objet : RE: [DISCUSSION] Apache Flex 4.11.0 RC 1
ok
-Message d'origine-
De : Erik de Bruin [mailto:e...@ixsoftware.nl] Envoyé : lundi 14 octobre 2013
09:00 À : dev@flex.apache.org Objet : Re: [DISCUSSION] Apache Flex 4.11.0 RC 1
N
ok
-Message d'origine-
De : Erik de Bruin [mailto:e...@ixsoftware.nl]
Envoyé : lundi 14 octobre 2013 09:00
À : dev@flex.apache.org
Objet : Re: [DISCUSSION] Apache Flex 4.11.0 RC 1
Nope, it does contain the contents of that branch at the time the VOTE was
called.
EdB
On Mon, O
>>The RC hasn't changed. All fixes made since the RC failed the tests
>>(and the changes to the tests) are NOT part of the release that we are
>>currently voting on. What makes you change your vote?
> Maybe I missed something, but I am unaware of any flaws in the contents of
> the RC1 source packag
gt; Maurice
>
> -Message d'origine-
> De : Erik de Bruin [mailto:e...@ixsoftware.nl]
> Envoyé : lundi 14 octobre 2013 08:49
> À : dev@flex.apache.org
> Objet : Re: [DISCUSSION] Apache Flex 4.11.0 RC 1
>
> That fix isn't in the RC currently being VOTE-d on...
>
On 10/13/13 11:53 PM, "Erik de Bruin" wrote:
>Alex,
>
>The RC hasn't changed. All fixes made since the RC failed the tests
>(and the changes to the tests) are NOT part of the release that we are
>currently voting on. What makes you change your vote?
Maybe I missed something, but I am unaware of
gt; Maurice
>
> -Message d'origine-
> De : Justin Mclean [mailto:jus...@classsoftware.com] Envoyé : lundi 14
> octobre 2013 07:24 À : dev@flex.apache.org Objet : Re: [DISCUSSION]
> Apache Flex 4.11.0 RC 1
>
> Hi,
>
> Thanks for voting +1 Alex, but there has to b
Alex,
The RC hasn't changed. All fixes made since the RC failed the tests
(and the changes to the tests) are NOT part of the release that we are
currently voting on. What makes you change your vote?
EdB
On Mon, Oct 14, 2013 at 7:03 AM, Alex Harui wrote:
> For some reason, the 3rd test passed
--Message d'origine-
> De : Justin Mclean [mailto:jus...@classsoftware.com]
> Envoyé : lundi 14 octobre 2013 07:24
> À : dev@flex.apache.org
> Objet : Re: [DISCUSSION] Apache Flex 4.11.0 RC 1
>
> Hi,
>
> Thanks for voting +1 Alex, but there has to be something
SSION] Apache Flex 4.11.0 RC 1
Hi,
Thanks for voting +1 Alex, but there has to be something wrong with it, it
would be unusual for first RC to pass :-)
I need to look at the mobile skins in a little more detail and may actually
vote -1 depending on what that turns up. The mobile DPI test pass
Hi,
I've done all the pre-voting tests on RC1 (sig check, compile,
checkintests, build for IDE, loaded in FB, compiled several small and
large projects) and it all seems to check out... However (yes, here we
go again), since there have been several commits made to both develop
AND release branch s
Hi,
Thanks for voting +1 Alex, but there has to be something wrong with it, it
would be unusual for first RC to pass :-)
I need to look at the mobile skins in a little more detail and may actually
vote -1 depending on what that turns up. The mobile DPI test pass which is a
start.
As we're not
For some reason, the 3rd test passed tonight, and the two
mx/states/Transition failures on Windows which did not pass this morning
passed tonight.
So now, I can submit my vote.
On 10/13/13 9:58 PM, "Justin Mclean" wrote:
>Hi,
>
>> Anyway, I fixed the tests and they passed for me on both Mac and
Hi,
> Anyway, I fixed the tests and they passed for me on both Mac and Win.
All good - so in theory we have no failing tests now right?
Does the 3rd test you had failing (and doesn't for me) also have the same embed
font issue?
Thanks,
Justin
I'm guessing they were developed on Windows so they passed there if the
player handled the device fonts the same way, which happens often, but not
all the time.
But why we started seeing these now, I don't know. It is a small test so
maybe it gets skipped on occasion.
Anyway, I fixed the tests a
Hi,
> Turns out it's a new test SWF and doesn't have fonts embedded. So yes,
> these two were faulty tests.
So how could they pass on windows or before? Seems odd they would suddenly
start failing for no reason.
Thanks,
Justin
Turns out it's a new test SWF and doesn't have fonts embedded. So yes,
these two were faulty tests.
-Alex
On 10/13/13 8:44 PM, "Alex Harui" wrote:
>Just starting to dig. I'll keep the FTETextField idea in mind. I thought
>you said it passed for you? How did you get it to fail this time? Di
Hi,
> Just starting to dig. I'll keep the FTETextField idea in mind. I thought
> you said it passed for you? How did you get it to fail this time? Did
> you change player versions or something?
It was passing for me over he weekend, but I've changed player versions since
then.
Justin
Just starting to dig. I'll keep the FTETextField idea in mind. I thought
you said it passed for you? How did you get it to fail this time? Did
you change player versions or something?
-Alex
On 10/13/13 7:08 PM, "Justin Mclean" wrote:
>Hi,
>
>> [java]
Hi,
> [java] =
> [java]Failed:
> [java] =
> [java]
> components/AdvancedDataGrid/Styles/advanceddatagrid_styles_FLEX_32848
> style_textSelectedColor_singleRow Failed Compa
Hi,
Can only reproduce 2 here when running 11.9.
[java] =
[java] Failed:
[java] =
[java]
components/AdvancedDataGrid/Styles/advanceddatagrid_styles_FLEX_32848
style_t
Actually, 3 failures on Mac
[java]
components/AdvancedDataGrid/Styles/advanceddatagrid_styles_FLEX_32848
style_textSelectedColor_singleRow Failed CompareBitmap(body:step 3)
compare returned-4
[java]
components/AdvancedDataGrid/Styles/advanceddatagrid_styles_FLEX_32848
style_textSelect
Hi,
Looks there there's a couple of issues that need to be sorted - try this out:
http://ns.adobe.com/mxml/2009";
xmlns:s="library://ns.adobe.com/flex/spark"
applicationDPI="640">
Hi,
> I don't think Combobox is supposed to work with mobile.
Ah well that could be my issue then - was a just quick test I put to together
to see if there's any obvious issues.
You might want to double check button sizes (they are possible too large at 640
dpi) and the popup window close but
On Oct 13, 2013 4:33 PM, "Justin Mclean" wrote:
>
> Hi,
>
> Tested on OSX.
> -MD5 and sigs check out (the new check_sigs.sh in build can help out here)
> - Can compile source code (OSX/Java 1.7)
> - README, RELEASE_NOTES and NOTICE all in good shape.
> - Rat report says no bad licences.
> - Turned
Hi,
Tested on OSX.
-MD5 and sigs check out (the new check_sigs.sh in build can help out here)
- Can compile source code (OSX/Java 1.7)
- README, RELEASE_NOTES and NOTICE all in good shape.
- Rat report says no bad licences.
- Turned into usable SDK via build scripts (not installer). Added AIR 3.9.
HI,
> OK, my mustella runs have finished. I have two mx/states/Transition
> failures on Windows and two ADG failures on Mac.
Mind posting what the failure are? The ADG tests run clean here.
Given that there very little difference with windows and OS in terms of SDK
code it more likely to be a
OK, my mustella runs have finished. I have two mx/states/Transition
failures on Windows and two ADG failures on Mac.
I'll try to investigate them this evening if nobody beats me to it.
-Alex
On 10/11/13 5:16 PM, "Justin Mclean" wrote:
>Hi,
>
>Please place all discussion around the release can
Sorry, read into the source a bit more -- looks like it is the font
support, not Batik.
On Sun, Oct 13, 2013 at 9:57 AM, Nicholas Kwiatkowski wrote:
> During a build from source with the ANT script, we are still prompting for
> acceptance for a non-compatible license for what seems to be Batik.
During a build from source with the ANT script, we are still prompting for
acceptance for a non-compatible license for what seems to be Batik. We then
proceed to download Batik from an Apache web server. Is Batik still
covered by the Adobe license? (I think the prompt is for Batik -- the
prompt d
Hi,
> - when I start the installer to download 4.11 RC1, the default choice is on
> AIR 3.8 / FP 11.8, although AIR 3.9 / FP 11.9 are in the list.
> Is that intended?
It's intentional as we're not fully tested 3.9/11.9 yet.
> - I am testing 4.11 RC1 with AIR 3.8 / FP11.8, so that they are some
a focus on
DataGrids
=> OK, no regression
Regards,
Maurice
-Message d'origine-
De : Maurice Amsellem [mailto:maurice.amsel...@systar.com]
Envoyé : samedi 12 octobre 2013 18:50
À : dev@flex.apache.org
Objet : RE: [DISCUSSION] Apache Flex 4.11.0 RC 1
Hi,
- when I start the
test with AIR 3.9 / FP 11.9, let me know.
Maurice
-Message d'origine-
De : Justin Mclean [mailto:jus...@classsoftware.com]
Envoyé : samedi 12 octobre 2013 10:46
À : dev@flex.apache.org
Objet : Re: [DISCUSSION] Apache Flex 4.11.0 RC 1
Hi,
> I never said I don't build the
Hi,
> I never said I don't build the binary myself.
Ah sorry some confusion there I has assumed by this:
"The make scripts never worked for me on my Windows machine."
And other bits of the conversation that you weren't compiling it, my mistake.
> I build sdk binary and load the Installer via th
On Sat, Oct 12, 2013 at 12:08 AM, Justin Mclean wrote:
> Hi,
>
> > I agree with all this. The part: "and test the resulting executable on
> > their own platform" is where the problem arises. The binary that our ant
> > build creates is useless unless we get all the other dependencies.
> Obviousl
Hi,
> I agree with all this. The part: "and test the resulting executable on
> their own platform" is where the problem arises. The binary that our ant
> build creates is useless unless we get all the other dependencies.
Obviously you can't test it fully without the 3rd party dependancies. So ye
On Fri, Oct 11, 2013 at 9:20 PM, Justin Mclean wrote:
> Hi,
>
> > I dont think it is a matter of PMC vs. non-PMC members. AFAIK, Windows
> > users need to use the Installer to test the SDKs with existing Flex
> > projects.
>
> What the issue with the constructFlexForIDE.bat or
> makeApacheFlexfor
Hi,
> I dont think it is a matter of PMC vs. non-PMC members. AFAIK, Windows
> users need to use the Installer to test the SDKs with existing Flex
> projects.
What the issue with the constructFlexForIDE.bat or makeApacheFlexforIDE.bat
scripts?
> What are you testing then?
The release as defi
On Fri, Oct 11, 2013 at 8:32 PM, Justin Mclean wrote:
> Hi,
>
> > I prefer the second option. This is how we tested previous releases.
> Not everyone tests it that way. It's certainly easier to do it that way if
> you don't have a complication environment set up, and it's certainly easier
> for n
Hi,
> I prefer the second option. This is how we tested previous releases.
Not everyone tests it that way. It's certainly easier to do it that way if you
don't have a complication environment set up, and it's certainly easier for non
PMC members to test it that way.
The official vote is on the
On Fri, Oct 11, 2013 at 7:11 PM, Justin Mclean wrote:
> Hi,
>
> > I dont think it is a valid test of the RC if the Installer is downloading
> > the wrong OSMF swc.
> I only put it there as a convenience to people who can't compile the SDK.
> People who have binding votes shouldn't be using the ins
Hi,
> I dont think it is a valid test of the RC if the Installer is downloading
> the wrong OSMF swc.
I only put it there as a convenience to people who can't compile the SDK.
People who have binding votes shouldn't be using the installer :-)
> In that case, you should probably revert the chang
On Fri, Oct 11, 2013 at 6:55 PM, Justin Mclean wrote:
> Hi,
>
> > Did we have to make any code changes to make OSMF 2.0 work with the Flex
> > SDK?
> No but it certainly requires more testing. IMO I'd hold off on changing
> the installer, unless other people think otherwise.
>
>
I dont think it is
Hi,
> Did we have to make any code changes to make OSMF 2.0 work with the Flex
> SDK?
No but it certainly requires more testing. IMO I'd hold off on changing the
installer, unless other people think otherwise.
Thanks,
Justin
On Fri, Oct 11, 2013 at 6:39 PM, Justin Mclean wrote:
> Hi,
>
> > Should we update the Installer config to point to OSMF 2.0 swc for the
> RC1?
> Issue there is we really need older version for 4.9 and 4.10 instals and
> newer for 4.11. Not sure the installer can do that without a code change
> an
Hi,
> Should we update the Installer config to point to OSMF 2.0 swc for the RC1?
Issue there is we really need older version for 4.9 and 4.10 instals and newer
for 4.11. Not sure the installer can do that without a code change and a VOTE.
Justin
There's a problem. The SDK's downloads.xml was updated to OSMF 2.0. But
the Installer is still looking at the earlier version.
Should we update the Installer config to point to OSMF 2.0 swc for the RC1?
Thanks,
Om
On Fri, Oct 11, 2013 at 6:30 PM, OmPrakash Muppirala
wrote:
> On Fri, Oct 11,
On Fri, Oct 11, 2013 at 6:18 PM, Justin Mclean wrote:
> Hi,
>
> > Need a clarification about the voting process. Are we going to carry
> votes
> > across RCs as suggested in this thread [1] ?
>
> If the changes are relatively minor between RC I think we'll can carry
> votes over. IF this is the c
Hi,
> Need a clarification about the voting process. Are we going to carry votes
> across RCs as suggested in this thread [1] ?
If the changes are relatively minor between RC I think we'll can carry votes
over. IF this is the case I'll mention it in the VOTE for the RC, people can
then vote -1
Hi,
To make it easier for people to test who are unable to compile the SDK I've
added RC1 to as an option in the installer. Download and give it a go!
Thanks,
Justin
Need a clarification about the voting process. Are we going to carry votes
across RCs as suggested in this thread [1] ?
Thanks,
Om
[1]
http://apache-flex-development.247.n4.nabble.com/Ideas-for-dealing-with-less-VOTEs-per-RC-td29767.html
On Fri, Oct 11, 2013 at 5:16 PM, Justin Mclean wrote
Hi,
Please place all discussion around the release candidate here and not in the
[VOTE] thread.
Thanks,
Justin
63 matches
Mail list logo
