Hi team,
I have been working on FLEX-34230 (StageText skins prevents garbage collection
of component with TextInputs and TextAreas)
and it appears that GC is prevented by the internal pool of StageText used in
StageText-based skins (precisely in old StyleableStageText and new
ScrollableStageTex
Digging further into TextInput code, I made some curious findings (at least
for me):
StageText is not GCed because StyleableStageText sets event listeners on
changing, change, enter, events etc... when the skin is attached (in
TextInputBase partAdded).
2) these event listeners should be remov
Don't know this code that well, but when would you trigger detaching of
the skin? Doing it on removing from stage would not be right in many
cases since the widget could come back on stage later.
-Alex
On 4/22/14 4:44 AM, "Maurice Amsellem" wrote:
>Digging further into TextInput code, I made
> Doing it on removing from stage would not be right in many cases since the
> widget could come back on stage later?
I don't know the code well either, and I don't have any definite answer to this
question.
So you are probably right to not detach the skin when the widget is removed
from stage.
I would think that most components do not use a pool of subcomponents, so
detaching skins does not have to be part of the lifecycle. IOW, the
component and its skin and its subcomponents are all available for GC at
the time the last reference to the component is broken. References from
the skin b
>so detaching skins does not have to be part of the lifecycle.
I agree with that, that's why I was asking about removing listeners, rather
than detaching skins. Is that the same ?
IOW, do you mean that explicitly removing listeners from the skin to the
component shouldn't be part of the compon
Never mind my last comment.
I will use an interface such as IProxiedStageTextWrapper, something like
"IPooled".
-Message d'origine-
De : Maurice Amsellem [mailto:maurice.amsel...@systar.com]
Envoyé : mardi 22 avril 2014 19:07
À : dev@flex.apache.org
Objet : RE: Question about mobile S
On 4/22/14 10:07 AM, "Maurice Amsellem"
wrote:
>>so detaching skins does not have to be part of the lifecycle.
>I agree with that, that's why I was asking about removing listeners,
>rather than detaching skins. Is that the same ?
>IOW, do you mean that explicitly removing listeners from the sk
>When I look at SkinnableTextBase.partAdded it looks like it is adding a
>listener to the 'textDisplay'. I assume that 'textDisplay' isn't a StageText
>in a pool. If that's true, >the SkinnableTextBase.as is correct.
>I would expect that 'textDisplay' is a StageTextInputSkin and internally it
I think I understand your description, but I'm surprised that each skin
instance doesn't have its own ScrollableStageText. I would think only the
StageText instances are pooled. It seems ok to use removeFromStage to cut
any references between the StageText and the ScrollableStageText since a
Skin
Hi,
If there's no objections I'll send this out in 12 hours or so. Note that
the CMS buildbot is still not working, so changes I've made to the website
to support this document are not even on staging and but are in SVN. If
CMS is still broken, I will hold the announcement until the site is
upda
Yeah! :) I will definitely after publish spread the word on some Polish
forums! :)
Piotr
-
Apache Flex Committer
piotrzarzyck...@gmail.com
--
View this message in context:
http://apache-flex-development.247.n4.nabble.com/DRAFT-Announcement-FlexJS-and-FalconJX-0-0-1-Released-tp37124p371
Hi Alex,
I've just tried install Flex SDK 12.0 and Nightly build and there is some
problem with downloading OSMF.
Log file - osmf link is broken.
Version 3.0.0 (windows)
Using Locale:pl_PL
Fetched the SDK download mirror URL from the CGI.
AIR version 4.0
Flash Player version 12.0
Creating Apache
If you hit the link from your browser, does it work? It worked for me.
-Alex
On 4/22/14 12:36 PM, "piotrz" wrote:
>Hi Alex,
>
>I've just tried install Flex SDK 12.0 and Nightly build and there is some
>problem with downloading OSMF.
>
>Log file - osmf link is broken.
>
>Version 3.0.0 (windows)
Yes did it. Looks like there is something wrong on my PC.
Thanks Alex
Piotr
-
Apache Flex Committer
piotrzarzyck...@gmail.com
--
View this message in context:
http://apache-flex-development.247.n4.nabble.com/DISCUSS-Discuss-Release-Apache-Flex-SDK-Installer-3-0-RC7-tp35588p37128.html
On Apr 22, 2014 12:08 PM, "Alex Harui" wrote:
>
> Hi,
>
> If there's no objections I'll send this out in 12 hours or so. Note that
> the CMS buildbot is still not working, so changes I've made to the website
> to support this document are not even on staging and but are in SVN. If
> CMS is still
On 4/22/14 1:36 PM, "OmPrakash Muppirala" wrote:
>On Apr 22, 2014 12:08 PM, "Alex Harui" wrote:
>>
>> Hi,
>>
>> If there's no objections I'll send this out in 12 hours or so. Note
>>that
>> the CMS buildbot is still not working, so changes I've made to the
>>website
>> to support this documen
Hi,
> A simple way to try these releases is to use a pre-release of the
> InstallApacheFlex 3.0 application which is available at the following url:
> https://dist.apache.org/repos/dist/dev/flex/installer/3.0/rc7/binaries/
Should we have a non mirrored link in the announcement?
Justin
On 4/22/14 1:46 PM, "Justin Mclean" wrote:
>Hi,
>
>> A simple way to try these releases is to use a pre-release of the
>> InstallApacheFlex 3.0 application which is available at the following
>>url:
>> https://dist.apache.org/repos/dist/dev/flex/installer/3.0/rc7/binaries/
>
>Should we have a n
Hi,
> What would be the alternative?
Not sure, but you may want to consider implications/bandwidth that may consume
or even check with INFRA. See 1.
Justin
1. https://www.apache.org/dev/mirrors.html
On 4/22/14 2:45 PM, "Justin Mclean" wrote:
>Hi,
>
>> What would be the alternative?
>
>Not sure, but you may want to consider implications/bandwidth that may
>consume or even check with INFRA. See 1.
I will be very happy if the installer gets enough traffic to cause a
bandwidth issue. I hope
Hi,
> I also suggested pushing the installer to the release server and mirrors
> and holding the announcement
If it's not an official release I don't think you can put it in the release
area. Perhaps there another non release area that is mirrored?
Justin
On 4/22/14 2:59 PM, "Justin Mclean" wrote:
>Hi,
>
>> I also suggested pushing the installer to the release server and mirrors
>> and holding the announcement
>If it's not an official release I don't think you can put it in the
>release area. Perhaps there another non release area that is mirror
HI,
> It is official in that I can add my vote and send the results. We've only
> delayed doing so in case too many folks ran into the locale issue.
Which IMO is a serious issue - given (from google analytics) that we currently
have 150-200 installs a day and 70% of those are in non en_US local
HI,
BTW we just gone past 7000 installs since releasing 4.12.
Justin
On 4/22/14 3:43 PM, "Justin Mclean" wrote:
>HI,
>
>> It is official in that I can add my vote and send the results. We've
>>only
>> delayed doing so in case too many folks ran into the locale issue.
>
>Which IMO is a serious issue - given (from google analytics) that we
>currently have 150-200
Hi,
I also note from google analytics that with the new installer we had 150 odd
installs with > 50% resulting in failure. It's still 50% failure if I only look
at the last 2 weeks (40 installs) which I assume would be the last RC.
Justin
On 4/22/14 3:57 PM, "Justin Mclean" wrote:
>Hi,
>
>I also note from google analytics that with the new installer we had 150
>odd installs with > 50% resulting in failure. It's still 50% failure if I
>only look at the last 2 weeks (40 installs) which I assume would be the
>last RC.
That's intere
Hi,
> That's interesting. And nobody has lodged a complaint?
IMO people rarely complain when things go wrong - it takes effort. I'd guess
some of those failures will be windows permission issues as that seems to crop
but a bit.
> I wonder why so many failures.
Perhaps we should include the st
> I'm surprised that each skin instance doesn't have its own
> ScrollableStageText. I would think only the StageText instances are pooled.
Actually, this is the case: only StageText are cached.
SST acts as a *key* in the reverse dictionary (SST => ST).
Stage text is released (and put into the
Hi,
Looks like you can somehow select no product as this has been logged (5 times):
/track-installer.html?failure=true&label=&version=&os=windows
In the last 2 month for Flex 4.12 on windows we had 21 successes and 29
failures, for Flex 4.12 on mac we've had 12 successes and 0 failures so it d
Is there an significant difference in user experience or function between
versions that you think might be responsible for a higher failure rate?
Otherwise I'd assume the rate is similar.
Sent via the PANTECH Discover, an AT&T 4G LTE smartphone.
Justin Mclean wrote:
Hi,
> That's interesting.
On 4/22/14 5:27 PM, "Justin Mclean" wrote:
>Hi,
>
>Looks like you can somehow select no product as this has been logged (5
>times):
>/track-installer.html?failure=true&label=&version=&os=windows
>
>In the last 2 month for Flex 4.12 on windows we had 21 successes and 29
>failures, for Flex 4.1
Hi,
> Interesting. No failures on Mac. I'll run some installs on my Windows
> box tonight.
JFYI - there was a couple of failures on mac for other packages.
Justin
I ran 5 installs on my Windows box. All succeeded. However if I had some
other apps open and macafee on-access scanning on, I noticed that the
scrollbar was not so smooth during some tasks and once I even saw "not
responding". It might be that folks are seeing this and bailing. I've
run many in
I keep finding references to Watson in the comments of the SDK such as:
//workaround for Watson bug 1896186. FlashPlayer requres that file
//names be encoded on Macintosh, but not on Windows. Grouped this
//bug with FlashPlayer Watson bug 1899687
In the Sherlock Holmes series Watson he
Hi,
Tried out the installer again (a few times) on windows 7 (in a VM) and all
seemed to work reasonably well - well no errors anyway.
Run on windows XP (in a VM) and got this:
Version 3.0.0 (windows)
Using Locale:en_AU
Fetched the SDK download mirror URL from the CGI.
AIR version 13.0
Flash P
Watson is the internal Adobe bug base for the runtimes (not Flex). No way
we'd get access to the whole thing. However, the public bug base is
supposed to have references back to the internal bug so you might find
some of them in the public bug base.
-Alex
On 4/22/14 11:09 PM, "jude" wrote:
>I
Do you think the delay to get the config file is worse on Windows than Mac?
-Alex
On 4/22/14 11:12 PM, "Justin Mclean" wrote:
>Hi,
>
>Tried out the installer again (a few times) on windows 7 (in a VM) and
>all seemed to work reasonably well - well no errors anyway.
>
>Run on windows XP (in a V
Hi,
> Do you think the delay to get the config file is worse on Windows than Mac?
Certainly seem so but perhaps not a fair test as I'm running windows in a VM.
On OSX I get text showing in 3-4 seconds after application launch, the first
Next button takes 2-3 seconds so that means I can still cl
40 matches
Mail list logo
