On Tue, Mar 11, 2014 at 11:15 PM, Alex Harui wrote:
>
>
> On 3/11/14 10:26 PM, "OmPrakash Muppirala" wrote:
>
> >Doesn't Mustella need bitmap comparison? That would be almost impossible
> >to get to work in HTML. Selenium probably is a better option for these
> >kind of things.
> Well, lately
On 3/11/14 11:42 PM, "Justin Mclean" wrote:
>Hi,
>
>> But at runtime the actual code that runs is written in C. And that's
>>why
>> every Flex SWF I know of can be compiled with playerglobal.swc in the
>> external library path. No code from playerglobal.swc gets linked in to
>> the SWF yet th
Yes, we very much need to be able to test against various browsers (both
old - IE8 comes to mind - and new). Any solution that doesn't allow
automated testing against multiple browsers on multiple platforms seems a
less optimal choice. The various JS, CSS and maybe SVG implementations just
differ t
Hi,
> But at runtime the actual code that runs is written in C. And that's why
> every Flex SWF I know of can be compiled with playerglobal.swc in the
> external library path. No code from playerglobal.swc gets linked in to
> the SWF yet those SWFs run fine.
That may be the case and sorry again
On 3/11/14 11:01 PM, "Sean Thayne" wrote:
>If you ran the JS tests inside of an air app, you could generate bitmaps
>for comparison.. But, that wouldn't really work to test different
>browsers,
>which is another thing FlexJS will probably need. To test everything in
>all
>major browsers.
True.
On 3/11/14 10:40 PM, "Justin Mclean" wrote:
>Hi,
>
>> I don't doubt that, but aren't they stubs and not actual
>>implementations?
>They are actually implemented in AS in the information I've found. Not
>eveything is marked native. You can add EventDispatcher to that short
>list. [1]
My understa
On 3/11/14 10:26 PM, "OmPrakash Muppirala" wrote:
>Doesn't Mustella need bitmap comparison? That would be almost impossible
>to get to work in HTML. Selenium probably is a better option for these
>kind of things.
Well, lately I've been wishing we'd done display-list compares instead of
bitmap
If you ran the JS tests inside of an air app, you could generate bitmaps
for comparison.. But, that wouldn't really work to test different browsers,
which is another thing FlexJS will probably need. To test everything in all
major browsers.
-Sean Thayne
On Tue, Mar 11, 2014 at 11:26 PM, OmPrakas
Hi,
> I don't doubt that, but aren't they stubs and not actual implementations?
They are actually implemented in AS in the information I've found. Not
eveything is marked native. You can add EventDispatcher to that short list. [1]
I would have to actually decompile the player global swf to know
Doesn't Mustella need bitmap comparison? That would be almost impossible
to get to work in HTML. Selenium probably is a better option for these
kind of things.
Another option is that perhaps we could build a runner in AIR that loads
the html stuff in a HTML component. That we could reuse the bi
On 3/11/14 10:09 PM, "Justin Mclean" wrote:
>Hi,
>
>> I thought byte code from playerglobal did not actually get linked into
>>any SWF.
>
>OT to this threadI know but may be useful to someone. After a little
>searching URLLoader, Vector, Rectangle, describeType and setTimeout are
>all in there
Hi,
> I thought byte code from playerglobal did not actually get linked into any
> SWF.
OT to this threadI know but may be useful to someone. After a little searching
URLLoader, Vector, Rectangle, describeType and setTimeout are all in there, may
be other classes and static methods.
Justin
When I looked back when we were in the incubator, no code from
playerglobal was being linked in. That was important to being able to
call it a build tool instead of a required dependency.
Maybe Adobe just pushes the playerglobal.swc when they slipstream a player
update. At least, that's my guess
Hi folks,
I just checked in enough code to get the Button and CheckBox test from the
SDK's checkintest to run under FlexJS. There's another half-dozen or so
tests we should probably finish converting since we have those components
in the FlexJS framework. Getting some sort of checkintest running
Hi,
> I thought byte code from playerglobal did not actually get linked into any
> SWF.
Think there some code that does. Can't recall exactly what but the was a thread
about it way back on this mailing list.
> Actually, that's sort of important from a dependencies point of
> view. What are the
I would definitely wait. IIRC, we did make an attempt to not return
[object Object] in these itemToLabel calls.
-Alex
On 3/11/14 9:18 PM, "Justin Mclean" wrote:
>Hi,
>
>I know it's easy to "fix" - the question is was it broken in the first
>place and is the current behaviour more correct? (Des
On 3/11/14 9:12 PM, "Justin Mclean" wrote:
>Hi,
>
>> PlayerGlobal.swc was the only difference?
>Yes that was the only difference. All other files were identical in both
>and neither instal had an extra or missing file.
>
>> Whether it is cached or not is in the hands of the installer.xml in the
Hi,
I know it's easy to "fix" - the question is was it broken in the first place
and is the current behaviour more correct? (Despite it being a regression?)
And if we do "fix" is it worth making a point release or waiting until 4.13?
Thanks,
Justin
Ah, that test was in a later commit.
Is the cheap fix as simple as:
if (dateField == null) return " ";
-Alex
On 3/11/14 4:46 PM, "Justin Mclean" wrote:
>Hi ,
>
>Sorry my mistake data is not set to null if dataField is null only if the
>dataField doesn't exist.
>
>if (dataField != null
Hi,
> PlayerGlobal.swc was the only difference?
Yes that was the only difference. All other files were identical in both and
neither instal had an extra or missing file.
> Whether it is cached or not is in the hands of the installer.xml in the
> 4.12.0 release. We could change that for 4.13.
So
Piotrz,
Sorry to ding you on your first commit, but the team page here:
http://flex.apache.org/about-people.html looks out of shape. Also, the
list is supposed to be in alphabetical order (based on first name)
Can you please take a look?
Thanks,
Om
On Tue, Mar 11, 2014 at 3:27 PM, wrote:
>
PlayerGlobal.swc was the only difference? That's good to know.
Whether it is cached or not is in the hands of the installer.xml in the
4.12.0 release. We could change that for 4.13.
-Alex
On 3/11/14 5:05 PM, "Justin Mclean" wrote:
>Hi,
>
>Re difference between cached and non chached I got t
Hi,
From the Apache Flex blog comments:
Awesome improvements guys! Thanks! But it seems that the MaskedTextInput aren't
working :(
Justin
Hi,
Re difference between cached and non chached I got this:
diff -qrs ./InstallerTestCached ./InstallerTestNoCache | grep -v "identical"
Files ./InstallerTestCached/frameworks/libs/player/12.0/playerglobal.swc and
./InstallerTestNoCache/frameworks/libs/player/12.0/playerglobal.swc differ
Comp
Hi ,
Sorry my mistake data is not set to null if dataField is null only if the
dataField doesn't exist.
if (dataField != null)
{
if (dataField in data)
data = data[dataField];
else
data = null;
}
But the real question is if dataField is null shoul
Strange. If data is set to null, then in the try block aren't you calling
toString() on null? That should throw instead of return [object Object].
-Alex
On 3/11/14 4:00 PM, "Justin Mclean" wrote:
>Hi,
>
>> I didn't actually run the code, but I cannot figure out how it would
>> return [object
Now it's work fine
thanks
Jean
--
View this message in context:
http://apache-flex-development.247.n4.nabble.com/Installer-for-4-12-0-on-Mac-tp35651p35658.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.
Hi,
> I didn't actually run the code, but I cannot figure out how it would
> return [object object]. If dataField is null, wouldn't it set data = null?
It first gets set to null inside the if typeof(data) == "object" but then fall
through to the return return data.toString() in the try catch bl
Actually, I'm getting the error too. I think Apache is having server
issues. Please be patient and try again later.
Thanks,
-Alex
On 3/11/14 2:50 PM, "Maurice Amsellem" wrote:
>Can you right-click on the Installer, select "Verbose Log" and try again,
>then send the log content.
>
>Maurice
>
Tried RC7 installer.
Couldn't reproduce my RC6 error - when config load fails, button labels
are populated and log window pops open with error message. Much better :)
--
Lee Burrows
ActionScripter
Can you right-click on the Installer, select "Verbose Log" and try again, then
send the log content.
Maurice
-Message d'origine-
De : nitrobear [mailto:j...@netinfluence.ch]
Envoyé : mardi 11 mars 2014 22:29
À : dev@flex.apache.org
Objet : Re: Installer for 4.12.0 on Mac
Hello,
I try
Hello,
I try this version, and same problem :
Version 3.0.0 (mac)
Using Locale:fr_CH
Could not fetch the SDK download mirror URL from the CGI. Going to try the
GeoIP route.
Unable to clean up temporary installation directories
Installation aborted
Jean
--
View this message in context:
http
Hi ntrobear.
One of the solution could be to try a new release candidate Flex Installer
3.0
https://dist.apache.org/repos/dist/dev/flex/installer/3.0/rc7/binaries/
After installation you could back to us with some feedback about that
process.
Piotr
-
Flex/Air developer.
piotrzarzyck...
I try to install the sdk via installer on my Mac, 10.9.2, and I have this
error
Version 2.7.0 (mac)
Could not fetch the SDK download mirror URL from the CGI. Going to try the
GeoIP route.
Unable to clean up temporary installation directories
Installation aborted
AIR version 4.0
Flash Player versio
>I can't figure out the tooltip thing. Could it be from an app underneath the
>installer?
It's an icon on Windows desktop.
Weird. The tooltip also appears when any other app is in the front, but the
cursor is just above that desktop icon.
Windows' bug?
Anyway, it's not the installer ;-)
Regard
On Tue, Mar 11, 2014 at 12:28 PM, Alex Harui wrote:
> I can't figure out the tooltip thing. Could it be from an app underneath
> the installer?
>
> I can't reproduce getting stuck by unplugging during the download of the
> SDK. I've gotten it to partially download and then get stuck in the MD5
Hi,
Many thanks guys for a warm Welcome!
I am really happy and proud to be a part of this wonderful Team! :)
Couple of things about me:
- My main experience is around desktop applications
- I am also responsible for the Fabrication Framework (utility for the
PureMVC framework)
- I have some smal
I can't figure out the tooltip thing. Could it be from an app underneath
the installer?
I can't reproduce getting stuck by unplugging during the download of the
SDK. I've gotten it to partially download and then get stuck in the MD5
calculation on Windows 7 for VMWare.
Right now, my sense is th
On Tue, Mar 11, 2014 at 11:55 AM, Alex Harui wrote:
> There is already a no cache bit in the config file that the Installer
> appears to use correctly, at least in my testing. If you ran a FlexJS
> install, it will cache that zip. I haven't had time to adjust the FlexJS
> install to not cache t
There is already a no cache bit in the config file that the Installer
appears to use correctly, at least in my testing. If you ran a FlexJS
install, it will cache that zip. I haven't had time to adjust the FlexJS
install to not cache that download. Could that be what is happening?
-Alex
On 3/1
I see another serious problem (sorry)
After caching is enabled, it seems like the nightly builds are being cached
as well. For example, under the
.../LocalStore/job/flex-falcon/lastSuccessfulBuild/artifact/out directory,
there is a apache-flex-falconjx-0.0.1-bin.zip. If this gets built for
every
On 3/12/2014 1:31 AM, Alex Harui wrote:
If you pulled from my people dir before 11-Mar-2014 05:10 (in Europe) then
you got a buggy version.
very likely (though no idea what tz that server is in).
You can delete the "in" dir. Om is referring to a different folder. The
installer creates a tem
I'll do a check later. On Mac the compare goes quickly. I don't know if
file times get recorded in the zip and cause a checksum difference or not.
On 3/11/14 11:15 AM, "OmPrakash Muppirala" wrote:
>One issue that has been bothering me, though. I tried an install without
>cache and then with c
On 3/11/14 11:18 AM, "Paul Hastings" wrote:
>On 3/12/2014 1:02 AM, Alex Harui wrote:
>> Hmm. So you couldn't reproduce the "unfinished" install? I guess we'll
>> wait and see if anyone else can reproduce it. Thanks for trying again.
>> I
>
>i used the file from your people dir dated "11-Mar-
On 3/12/2014 1:02 AM, Alex Harui wrote:
Hmm. So you couldn't reproduce the "unfinished" install? I guess we'll
wait and see if anyone else can reproduce it. Thanks for trying again. I
i used the file from your people dir dated "11-Mar-2014 05:10" which i'm not
sure is the same one that ble
One issue that has been bothering me, though. I tried an install without
cache and then with cache on. Then, I zipped both the resulting SDK
directories and did a checksum on them. They turned out to be different.
Is this expected?
A full directory compare was taking way way too long for me to
I just tried RC7 and I cant repro the unfinished install either. Looks
like it is fixed.
Thanks,
Om
On Tue, Mar 11, 2014 at 11:02 AM, Alex Harui wrote:
> Hmm. So you couldn't reproduce the "unfinished" install? I guess we'll
> wait and see if anyone else can reproduce it. Thanks for trying
On Tue, Mar 11, 2014 at 10:56 AM, Paul Hastings wrote:
> On 3/11/2014 9:05 PM, Alex Harui wrote:
>
> Can you try again but right-click and enable verbose logging? I just did
>> an install in my Windows VM and it worked. The log will be big so please
>> post somewhere or email me directly.
>>
>
>
Hmm. So you couldn't reproduce the "unfinished" install? I guess we'll
wait and see if anyone else can reproduce it. Thanks for trying again. I
don't need to see the log if it was successful.
Regarding sizes, can you do a quick compare of the two folders to see what
the extra files are? Did y
On 3/11/2014 9:05 PM, Alex Harui wrote:
Can you try again but right-click and enable verbose logging? I just did
an install in my Windows VM and it worked. The log will be big so please
post somewhere or email me directly.
well the file marked 11-Mar-2014 05:10 installed A-ok. don't suppose y
Best thing to do is probably to uninstall, then re-install from here:
https://dist.apache.org/repos/dist/dev/flex/installer/3.0/rc7/binaries/
Thanks,
-Alex
On 3/11/14 10:25 AM, "Paul Hastings" wrote:
>On 3/11/2014 11:11 PM, Alex Harui wrote:
>> Yeah, there was definitely a bug in the dmg/exe
On 3/11/2014 11:11 PM, Alex Harui wrote:
Yeah, there was definitely a bug in the dmg/exe I put on my people.a.o
folder, but it should have been fixed in the RC7 (which should also have
the version in the log). Plus, it just worked for me on Win 7, but it is
a VMWare VM.
yes that's the build i
Ah, the empty catch change.
I didn't actually run the code, but I cannot figure out how it would
return [object object]. If dataField is null, wouldn't it set data = null?
BTW, I didn't think of this at the time, but sometimes, folks create
proxy-based objects that don't implement property enume
Not surprising. The question is whether you truly care if null is
supposed to be a separate value than undefined.
On 3/11/14 12:17 AM, "Justin Mclean" wrote:
>Hi,
>
>> I won't veto it (unless you use == instead of ===). I think
>> all undefined checks should use ===. Otherwise it would match
Yeah, there was definitely a bug in the dmg/exe I put on my people.a.o
folder, but it should have been fixed in the RC7 (which should also have
the version in the log). Plus, it just worked for me on Win 7, but it is
a VMWare VM.
-Alex
On 3/11/14 8:48 AM, "OmPrakash Muppirala" wrote:
>Paul's i
It is (was) common to use * for a three-value boolean where undefined
means "not-specified, check somewhere else"
On 3/11/14 3:08 AM, "Justin Mclean" wrote:
>Hi,
>
>> I would say to use a proper data type. Would like to think we didn't
>>blindly use * for things just for the sake of convenience
Paul's issue looks very similar to what I saw yesterday, except mine was
with FlexJS.
Om
On Mar 11, 2014 7:05 AM, "Alex Harui" wrote:
> Paul,
>
> Can you try again but right-click and enable verbose logging? I just did
> an install in my Windows VM and it worked. The log will be big so please
>
Congrats Piotr!
-Sugan
-Original Message-
From: Kessler CTR Mark J [mailto:mark.kessler@usmc.mil]
Sent: 11 March 2014 02:48 PM
To: dev@flex.apache.org
Subject: RE: Please welcome Piotr Zarzycki as the newest Apache Flex
committer
Congrats and welcome :)
-Mark
-Original Message
Paul,
Can you try again but right-click and enable verbose logging? I just did
an install in my Windows VM and it worked. The log will be big so please
post somewhere or email me directly.
Thanks,
-Alex
On 3/11/14 2:14 AM, "Paul Hastings" wrote:
>windows 7, installer appears to hang on last "
On 11/03/14 12:55, Justin Mclean wrote:
Does anyone else think this is a little strange given FP 13.0 is still in beta?
v13 of the Player is more like 12.1. Look at the 'new features' on Labs
- basically none.
And was v12.0 so much different to v11.whatever ?
My guess is they've just stripped
Well if you're referring " encourage IT organizations to thoroughly test our
version 13" part. It would make me question what large changes they are
concerned about. Thoroughly testing would be synonymous with has a good chance
of breaking something.
-Mark
-Original Message-
From: J
Hi,
Just saw this:
http://blogs.adobe.com/flashplayer/2014/03/upcoming-changes-to-flash-players-extended-support-release.html
Does anyone else think this is a little strange given FP 13.0 is still in beta?
Thanks,
Justin
Congrats and welcome :)
-Mark
-Original Message-
From: Justin Mclean [mailto:jus...@classsoftware.com]
Sent: Tuesday, March 11, 2014 4:17 AM
To: dev@flex.apache.org
Subject: Please welcome Piotr Zarzycki as the newest Apache Flex committer
Hi,
I am pleased to announce that the Project
Now's a good time to get Git setup on my new box at work then :-)
Tom
On 11/03/14 10:18, Alexander Doroshko wrote:
Probably that's because OSMF.swc file in [Flex SDK]/frameworks/libs
has uppercased name since Flex SDK 4.12, but the corresponding entry
in flex-config.xml file still contains low
Hi,
I think the issue is with OSMF 2.0 there's no public download link to the .swc
only the .swf, so it's not obtained by the installer.
Justin
Probably that's because OSMF.swc file in [Flex SDK]/frameworks/libs has
uppercased name since Flex SDK 4.12, but the corresponding entry in
flex-config.xml file still contains lowercased name.
On 11.03.2014 13:40, Tom Chiverton wrote:
As soon as I try to use an RSL for the framework files my pa
Hi,
> I would say to use a proper data type. Would like to think we didn't blindly
> use * for things just for the sake of convenience.
That was my very first though but it's part of a public API so we can't easily
change that.
Thanks,
Justin
>I think all undefined checks should use ===.
Thanks Alex for the tip, I didn't know that.
I think my code needs some review ;-)
Maurice
-Message d'origine-
De : Alex Harui [mailto:aha...@adobe.com]
Envoyé : mardi 11 mars 2014 08:00
À : dev@flex.apache.org
Objet : Re: ADG Bug of the
I would say to use a proper data type. Would like to think we didn't blindly
use * for things just for the sake of convenience.
-Mark
-Original Message-
From: Alex Harui [mailto:aha...@adobe.com]
Sent: Tuesday, March 11, 2014 2:32 AM
To: dev@flex.apache.org
Subject: Re: ADG Bug of the
As soon as I try to use an RSL for the framework files my passing
-static-link-runtime-shared-libraries=false to MXMLC, compile fails with
/home/tchiverton/workspace/apache-flex-4.12/apache-flex-sdk-4.12.0-bin/frameworks/flex-config.xml(411):
Error: unable to open 'libs/osmf.swc'
and there's
windows 7, installer appears to hang on last "block" in the UI (i let it sit for
almost 2 hours). on a whim i tried to compile using what was "installed" & FB
threw out of memory errors (WAG install never really completed).
log:
Version 3.0.0 (windows)
Using Locale:en_US
Fetched the SDK downlo
Welcome aboard, Piotr!!
__
Jose Barragan
Senior Software Engineer
Codeoscopic
+34 912 94 80 80
http://www.codeoscopic.com
On 11 Mar 2014, at 09:45, Carlos Rovira wrote:
> Congrats Piotr!,
>
> Welcome aboard :)
>
> C.
>
>
> 2014-03-11 9:24 GMT+01:00 Maurice Amsellem :
>
>
Test results of RC7.
Summary:
Connection errors are handled correctly at early steps (resource bundles,
config files), but not during downloads ( at least not in all places).
Details below:
Windows 7, locale=FR
General:
Minor: "Install Log" button tool tip says "Display downloaded apps", and
Congrats Piotr!,
Welcome aboard :)
C.
2014-03-11 9:24 GMT+01:00 Maurice Amsellem :
> Congratulations Piotr.
>
> I am happy to see you on board :-)
>
> Maurice
>
> -Message d'origine-
> De : Christofer Dutz [mailto:christofer.d...@c-ware.de]
> Envoyé : mardi 11 mars 2014 09:22
> À : dev
I think Maurice's proposition as a much cleaner message for the user on the
"Installation progress step" is a great solution for this!
The rest of my proposition such as "restart" and "back" button could wait
for the future releases. I'm going to rise jira ticket for that and try to
do it if Alex
Alex,
This is just you trying to beat the record Om and I set with the first
release of the installer, right?
7 down, 3 to go...
;-)
EdB
On Tue, Mar 11, 2014 at 6:21 AM, Alex Harui wrote:
> This is the discussion thread.
>
> This RC contains lots of small changes to try to handle inabilit
Congratulations Piotr.
I am happy to see you on board :-)
Maurice
-Message d'origine-
De : Christofer Dutz [mailto:christofer.d...@c-ware.de]
Envoyé : mardi 11 mars 2014 09:22
À : dev@flex.apache.org
Objet : AW: Please welcome Piotr Zarzycki as the newest Apache Flex committer
Congrat
About what to do when the installation aborts ( copied from RC6 thread)
Piotr proposes to retry when the connection is recovered (or at least have a
retry button).
I understand and agree (dixit Alex) that's way more work that we want to put in
the installer.
IMO, the problem is that it's not cl
Congrats Piotr :-)
Von: Justin Mclean
Gesendet: Dienstag, 11. März 2014 09:17
An: dev@flex.apache.org
Betreff: Please welcome Piotr Zarzycki as the newest Apache Flex committer
Hi,
I am pleased to announce that the Project Management Committee (PMC) for
Hi,
I am pleased to announce that the Project Management Committee (PMC) for Apache
Flex has invited Piotr Zarzycki to become a committer and that he has accepted.
Piotr has raised several (good quality) JIRA issues, been active and helpful on
the mailing list, has reviewed several release cand
>I think it should be enabled after exception, because when I get back
>my internet connection I could try to install Flex once again without
>restart installer.
ALEX: >That is way more work than I would like to take on.
I think it's not clear to the users "Installation aborted" means you have
Hi,
On the day it was released:
4.10 had 309 installs
4.11 had 352 installs
4.12 had 401 installs
(The 4.12 number may be higher as I'm not 100% sure Google Analytics 100% upto
date).
Justin
Hi,
> What change was this? I'll try to take a look tomorrow.
https://github.com/apache/flex-sdk/commit/a7109bab8409cb60cc0b64de07ce01bad875b701
Justin
Hi,
> I won't veto it (unless you use == instead of ===). I think
> all undefined checks should use ===. Otherwise it would match null.
JFYI There probably 100+ places in the SDK were we use == undefined rather than
== undefined. From a quick search it look like === undefined is used is a few
Hi,
> I don't know how much you'll actually save after the JIT
No idea either - plus it very likely this is not exactly a bottle neck in the
ADG code.
But the original code did offend me as it's meaning was IMO less than clear.
Thanks,
Justin
What change was this? I'll try to take a look tomorrow. Time to pack it
in for today.
On 3/10/14 11:45 PM, "Justin Mclean" wrote:
>Hi,
>
>> If you've got a fix ready to go, might as well check it in.
>
>The original behaviour was a little broken in that it wouldn't call
>toString at all, so I
There are plenty of places to improve SDK code. If that's your itch,
start scratching. I don't know how much you'll actually save after the
JIT, but hey, I won't veto it (unless you use == instead of ===). I think
all undefined checks should use ===. Otherwise it would match null.
-Alex
On 3/
87 matches
Mail list logo
