Re: [DISCUSS] Changing our general mode of development?

[Justin Mclean]

Hi,

+1 and I’ll go further to suggest if a checkin break a number of applications 
out their in the wild we considering reverting it if it can’t be corrected 
quickly.

Thanks,
Justin

Re: [FlexJS] Simple App Not Running

[Christofer Dutz]

Hi Yishay,

Well I would guess that everyone is having the same problem. Due to latest 
refactoring things are broken and not even the Jenkins build is working.
Even if it is providing new versions of the Framework libraries, these 
currently can’t be used to build applications till the issues are fixed.

Chris

Am 28.03.17, 10:53 schrieb "yishayw" :

Same problem with a fresh nightly. I anyone else getting this?



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Simple-App-Not-Running-tp60825p60826.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [DISCUSS] Changing our general mode of development?

[Harbs]

I still have not managed to get maven setup correctly.

> On Mar 28, 2017, at 12:46 PM, Christofer Dutz  
> wrote:
> 
> Hi Harbs,
> (this time replying to the right name ;-) )
> 
> I usually simply make sure I update my repos and do the full maven build with 
> tests and examples locally before pushing … I guess this is sufficient 
> protection against most problems. In IntelliJ that’s two clicks and a cup of 
> coffee or whatever beverage you prefer.
> 
> Chris
> 
> 
> Am 28.03.17, 11:41 schrieb "Harbs" :
> 
>+1.
> 
>I think it’s OK to develop however we might be comfortable on a feature 
> branch, but we definitely want an approved procedure which must be done 
> before committing to the develop branch.
> 
>Harbs
> 
>> On Mar 28, 2017, at 12:29 PM, Christofer Dutz  
>> wrote:
>> 
>> Hi,
>> 
>> For the last months, we have seen a huge increase in people working on the 
>> FlexJS and people working on first applications using FlexJS. I think we 
>> should discuss how we can make sure we don’t have interruptions like the 
>> current one in the future.
>> 
>> One point that has been causing pain in the past, was that some people are 
>> using Ant and some are using Maven. Maven is quite a bit more restrictive 
>> than Ant and it builds a lot more and tests a lot more. Just as an example 
>> in contrast to the Ant build the Maven build builds all Examples and it also 
>> tests some of them to be runnable in a browser. The Ant build only builds 
>> the framework and most of the latest problems only pop up if you build an 
>> application. It has occurred several times that Changed failed the Maven 
>> build but didn’t fail the Ant build … just because the Ant build doesn’t 
>> build everything. We could avoid this problem if people would not simply 
>> ignore build failures reported by the ASF Jenkins, which is taking care of 
>> the Maven build. It is currently setup to give feedback within an hour or so.
>> 
>> Sometimes the “fix” was to exclude a module in Maven. This usually had the 
>> side-effect of the RAT plugin failing after that because it now finds files 
>> without Apache headers. A quick solution to that problem is to log-in to the 
>> ASF Jenkins and to click on “wipe workspace” of that build. After that this 
>> type of problem should go away immediately.
>> 
>> Another point was that sometimes people work together on a larger 
>> refactoring and check-in stuff to develop in order to share code. We should 
>> start using feature branches for this. This has currently not been happening 
>> at all. I have setup everything that if you create a branch IN ALL 3 REPOS 
>> with a name “feature/{somename}” (but the same “somename” in all three ;-) ) 
>>  the ASF Jenkins will setup a Job for that which builds all parts in one go 
>> and give you immediate feedback on the state of your branch. Feature 
>> branches that are not “blue” should not be merged back to develop.
>> 
>> One last pattern I have encountered was people reporting stuff like: “I have 
>> been working on X and have almost finished ... I know it will break Y, but 
>> I’ll push my changes and fix Y after that” … keep in mind: By breaking Y 
>> everyone working on FlexJS is forced to stop working so I will probably veto 
>> every suggestion I encounter on the list that has a similar pattern.
>> 
>> FlexJS has matured and we are approaching a 1.0, but we also must mature the 
>> way we develop or we will hurt early adopters and people willing to help get 
>> FlexJS to shape. We want enterprise users to use our stuff, then we must 
>> start working in an enterprise-acceptable way.
>> 
>> Keep up the awesome work and lets just get a little more awesome ;-)
>> 
>> Chris
>> 
>> 
>> 
> 
> 
> 

Re: [DISCUSS] Changing our general mode of development?

[yishayw]

To be fair, I don't see a lot of instances where changes are pushed into dev,
knowing they will break things. I agree this should be avoided. I'm still
not sure how our current problem could have been avoided. It'll be clearer
once we've figured out what actually happened.



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/DISCUSS-Changing-our-general-mode-of-development-tp60830p60836.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] Summary of Changes

[Peter Ent]

I forgot to add that "align-items:center" is not supported on the SWF side
yet. The Flexbox has a number of properties that I still have to implement
on the flex side.

On 3/28/17, 8:11 AM, "Peter Ent"  wrote:

>The HorizontalLayout and VerticalLayout do not center their content
>because the equivalent code on the JS side does not do that (at least it
>did not do that for me in the tests I ran). I think setting margin:auto is
>supposed to help with that.
>
>On the other hand, you can switch to HorizontalFlexLayout and
>VerticalFlexLayout. Then on the Group or Container CSS, do the following:
>
>.styleForVerticalFlexLayout {
>align-items: center;
>}
>
>.styleForHorizontalFlexLayout {
>align-items: center;
>}
>
>This is still a learning experience both understanding the complexities of
>CSS and building an system that generates clean and robust HTML/CSS/JS.
>
>Thank you for your patience.
>
>—peter
>
>On 3/27/17, 4:46 PM, "Harbs"  wrote:
>
>>I probably need to examine the new “flex” layouts.
>>
>>Is there a way to have content centered using those?
>>
>>> On Mar 27, 2017, at 11:35 PM, Harbs  wrote:
>>> 
>>> Better, but I still have some problems (there’s probably more):
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> >> marginRight="auto"/>
>>> 
>>> 
>>> 
>>> 
>>> >> width="30" height="30"
>>>click="undo_clickHandler(event)" id="undoButton"
>>>src="assets/images/icons/0726-undo.svg">
>>> 
>>> 
>>> 
>>> 
>>> >> width="30" height="30"
>>>click="redo_clickHandler(event)" id="redoButton"
>>>src="assets/images/icons/0727-redo.svg">
>>> 
>>> 
>>> 
>>> 
>>> >> width="30" height="30"
>>>click="zoomin_clickHandler(event)"
>>>src="assets/images/icons/0806-zoom-in.svg"/>
>>> >> width="30" height="30"
>>>click="zoomout_clickHandler(event)"
>>>src="assets/images/icons/0807-zoom-out.svg"/>
>>> >> width="30" height="30"
>>>click="fitButton_clickHandler(event)"
>>>src="assets/images/icons/0843-expand.svg"/>
>>> >> width="30" height="30"
>>>click="previewButton_clickHandler(event)"
>>>src="assets/images/icons/0786-file-preview-white.svg"/>
>>> 
>>> >> width="30" height="30"
>>>click="finishButton_clickHandler()"
>>>src="assets/images/icons/0821-check.svg"/>
>>> 
>>> >> width="30" height="30"
>>>click="cancelButton_clickHandler()"
>>>src="assets/images/icons/0822-cross2.svg"/>
>>> 
>>> 
>>> 1. This used to create a centered group of buttons. Now, the container
>>>has a height of 0 and the buttons don’t show up.
>>> 
>>> 2.
>>> >> id="scrollContainer"
>>>width="100%" height="100%">
>>> 
>>> 
>>> 
>>> >> id="designContainer"
>>>className="_holder" y="0">
>>> 
>>> >> marginLeft="auto" marginRight="auto"/>
>>> 
>>> 
>>> 
>>> 
>>> 
>>> This used to create a scrolling div that was centered in the the
>>>window. It now is aligned left, and I don’t know if it scrolls.
>>> 
>>> There’s other issues, and I’ll see tomorrow what I can work around.
>>> 
 On Mar 27, 2017, at 11:08 PM, Peter Ent  wrote:
 
 Hi,
 
 I just pushed a change to UIBase to set position="absolute" when
setting x
 or y. I think this is perfectly safe and if someone does set x and y
and
 then tries to use a layout where that would be a conflict, they will
get
 have to avoid setting those properties.
 
 I figured this would eventually happen. Let's see if this fixes the
issue.
 
 —peter
 
 
 On 3/27/17, 2:58 PM, "Harbs"  wrote:
 
> Peter,
> 
> I just tried loading my app with your changes, and everything is
>totally
> borked. We rely a lot on absolute positioning and transformations.
> 
> We really need 

Re: [DISCUSS] Changing our general mode of development?

[Harbs]

I do think that development should be done almost exclusively on feature 
branches.

If the build status for a feature branch can be verified on the server (like 
you set it up) that’s ideal because it does not require using specifically ant 
or maven locally.

> On Mar 28, 2017, at 12:59 PM, Harbs  wrote:
> 
> I still have not managed to get maven setup correctly.
> 
>> On Mar 28, 2017, at 12:46 PM, Christofer Dutz  
>> wrote:
>> 
>> Hi Harbs,
>> (this time replying to the right name ;-) )
>> 
>> I usually simply make sure I update my repos and do the full maven build 
>> with tests and examples locally before pushing … I guess this is sufficient 
>> protection against most problems. In IntelliJ that’s two clicks and a cup of 
>> coffee or whatever beverage you prefer.
>> 
>> Chris
>> 
>> 
>> Am 28.03.17, 11:41 schrieb "Harbs" :
>> 
>>   +1.
>> 
>>   I think it’s OK to develop however we might be comfortable on a feature 
>> branch, but we definitely want an approved procedure which must be done 
>> before committing to the develop branch.
>> 
>>   Harbs
>> 
>>> On Mar 28, 2017, at 12:29 PM, Christofer Dutz  
>>> wrote:
>>> 
>>> Hi,
>>> 
>>> For the last months, we have seen a huge increase in people working on the 
>>> FlexJS and people working on first applications using FlexJS. I think we 
>>> should discuss how we can make sure we don’t have interruptions like the 
>>> current one in the future.
>>> 
>>> One point that has been causing pain in the past, was that some people are 
>>> using Ant and some are using Maven. Maven is quite a bit more restrictive 
>>> than Ant and it builds a lot more and tests a lot more. Just as an example 
>>> in contrast to the Ant build the Maven build builds all Examples and it 
>>> also tests some of them to be runnable in a browser. The Ant build only 
>>> builds the framework and most of the latest problems only pop up if you 
>>> build an application. It has occurred several times that Changed failed the 
>>> Maven build but didn’t fail the Ant build … just because the Ant build 
>>> doesn’t build everything. We could avoid this problem if people would not 
>>> simply ignore build failures reported by the ASF Jenkins, which is taking 
>>> care of the Maven build. It is currently setup to give feedback within an 
>>> hour or so.
>>> 
>>> Sometimes the “fix” was to exclude a module in Maven. This usually had the 
>>> side-effect of the RAT plugin failing after that because it now finds files 
>>> without Apache headers. A quick solution to that problem is to log-in to 
>>> the ASF Jenkins and to click on “wipe workspace” of that build. After that 
>>> this type of problem should go away immediately.
>>> 
>>> Another point was that sometimes people work together on a larger 
>>> refactoring and check-in stuff to develop in order to share code. We should 
>>> start using feature branches for this. This has currently not been 
>>> happening at all. I have setup everything that if you create a branch IN 
>>> ALL 3 REPOS with a name “feature/{somename}” (but the same “somename” in 
>>> all three ;-) )  the ASF Jenkins will setup a Job for that which builds all 
>>> parts in one go and give you immediate feedback on the state of your 
>>> branch. Feature branches that are not “blue” should not be merged back to 
>>> develop.
>>> 
>>> One last pattern I have encountered was people reporting stuff like: “I 
>>> have been working on X and have almost finished ... I know it will break Y, 
>>> but I’ll push my changes and fix Y after that” … keep in mind: By breaking 
>>> Y everyone working on FlexJS is forced to stop working so I will probably 
>>> veto every suggestion I encounter on the list that has a similar pattern.
>>> 
>>> FlexJS has matured and we are approaching a 1.0, but we also must mature 
>>> the way we develop or we will hurt early adopters and people willing to 
>>> help get FlexJS to shape. We want enterprise users to use our stuff, then 
>>> we must start working in an enterprise-acceptable way.
>>> 
>>> Keep up the awesome work and lets just get a little more awesome ;-)
>>> 
>>> Chris
>>> 
>>> 
>>> 
>> 
>> 
>> 
> 

Re: [FlexJS] Simple App Not Running

[yishayw]

My mistake, the problem is not with event but with goog.events.Event being
undefined but with org.apache.flex.core.Application being undefined. So it
looks like the same problem Harbs was having [1]

[1]
http://apache-flex-development.247.n4.nabble.com/Falcon-build-failing-td60790.html



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Simple-App-Not-Running-tp60825p60833.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [DISCUSS] Changing our general mode of development?

[Christofer Dutz]

Hi Harbs,

On a system that you can build with Ant all you should need to do, is to 
download Maven in a recent version, unpack is somewhere and add the “bin” 
directory to your systems path. I made sure that the one environment variable 
the build requires is the same as for Ant.

Regarding the Feature branches: I have thought of adding “Jenkinsfile”s to the 
other repos to allow only branching framework if your changes only affect 
framework or framework and typedefs, if you want to do some new typedef stuff 
(or all three, if you want to do changes in the compiler). Currently you would 
need to setup branches in all 3 repos and only the branch in “compiler” would 
trigger the job setup. I didn’t bother implementing the others as I noticed no 
one had ever used this feature except me. In IntelliJ all I usually do is to 
enable synchronized branches. So, whenever I create a new branch, this is done 
for all repos and if I do a pull, commit and push this is done synchronously 
for all 3 repos. So it’s actually no more work at all.

@yishayw I’m not explicitly referring to this instance … there have been 
several occasions where a similar pattern had been pursued and I just wanted to 
make everyone aware that It’s not just the build these changes might be 
breaking, but also every build that relies on SNAPSHOTs.

Chris



Am 28.03.17, 12:04 schrieb "Harbs" :

I do think that development should be done almost exclusively on feature 
branches.

If the build status for a feature branch can be verified on the server 
(like you set it up) that’s ideal because it does not require using 
specifically ant or maven locally.

> On Mar 28, 2017, at 12:59 PM, Harbs  wrote:
> 
> I still have not managed to get maven setup correctly.
> 
>> On Mar 28, 2017, at 12:46 PM, Christofer Dutz 
 wrote:
>> 
>> Hi Harbs,
>> (this time replying to the right name ;-) )
>> 
>> I usually simply make sure I update my repos and do the full maven build 
with tests and examples locally before pushing … I guess this is sufficient 
protection against most problems. In IntelliJ that’s two clicks and a cup of 
coffee or whatever beverage you prefer.
>> 
>> Chris
>> 
>> 
>> Am 28.03.17, 11:41 schrieb "Harbs" :
>> 
>>   +1.
>> 
>>   I think it’s OK to develop however we might be comfortable on a 
feature branch, but we definitely want an approved procedure which must be done 
before committing to the develop branch.
>> 
>>   Harbs
>> 
>>> On Mar 28, 2017, at 12:29 PM, Christofer Dutz 
 wrote:
>>> 
>>> Hi,
>>> 
>>> For the last months, we have seen a huge increase in people working on 
the FlexJS and people working on first applications using FlexJS. I think we 
should discuss how we can make sure we don’t have interruptions like the 
current one in the future.
>>> 
>>> One point that has been causing pain in the past, was that some people 
are using Ant and some are using Maven. Maven is quite a bit more restrictive 
than Ant and it builds a lot more and tests a lot more. Just as an example in 
contrast to the Ant build the Maven build builds all Examples and it also tests 
some of them to be runnable in a browser. The Ant build only builds the 
framework and most of the latest problems only pop up if you build an 
application. It has occurred several times that Changed failed the Maven build 
but didn’t fail the Ant build … just because the Ant build doesn’t build 
everything. We could avoid this problem if people would not simply ignore build 
failures reported by the ASF Jenkins, which is taking care of the Maven build. 
It is currently setup to give feedback within an hour or so.
>>> 
>>> Sometimes the “fix” was to exclude a module in Maven. This usually had 
the side-effect of the RAT plugin failing after that because it now finds files 
without Apache headers. A quick solution to that problem is to log-in to the 
ASF Jenkins and to click on “wipe workspace” of that build. After that this 
type of problem should go away immediately.
>>> 
>>> Another point was that sometimes people work together on a larger 
refactoring and check-in stuff to develop in order to share code. We should 
start using feature branches for this. This has currently not been happening at 
all. I have setup everything that if you create a branch IN ALL 3 REPOS with a 
name “feature/{somename}” (but the same “somename” in all three ;-) )  the ASF 
Jenkins will setup a Job for that which builds all parts in one go and give you 
immediate feedback on the state of your branch. Feature branches that are not 
“blue” should not be merged back to develop.
>>> 
>>> One last pattern I have encountered was people reporting stuff like: “I 
have been working on X and have almost finished ... I know it will 

Re: [FlexJS] Summary of Changes

[Peter Ent]

I just built another test where I nested Panels in Panels in Panels (since
they are compound components) and found some of the layouts still not
working right on the JS side.

It looks to me like the problem is not setting the style attributes right
using Flexbox. While I spent a good deal of time trying to understand
this, I think the algorithmic approach isn't applying the styles correctly
in all cases. I'm trying to figure out exactly why my structure isn't
styled correctly. Please stay tuned…

-peter

On 3/27/17, 5:48 PM, "Greg Dove"  wrote:

>I've just done a sweep through one project fixing our 'borked' stuff, I
>guess the latest change might re-'bork' some of the fixes, but I think at
>least these changes should be easier to address.
>
>Sometimes I needed to swap a Container to a Group and other times not,
>because of the relative/absolute changes.
>I found it easier to migrate a lot of the layout stuff to pure css for now
>using flexbox, thinking that doing so might insulate us from further
>changes in the coded layout support. We would have the ability to swap
>back
>to beads etc later if we want to have that in the mxml, once things have
>become stable.
>
>
>On Tue, Mar 28, 2017 at 9:52 AM, Peter Ent  wrote:
>
>> Does the HTML look OK - the structure. Is there anything missing? You
>> should see a simplified nesting of DIVs. If that's the case, then maybe
>> there is more work to do with the layouts.
>>
>> The topmost Container, outer controls, doesn't look like it has a
>>layout,
>> so with Container that should default to BasicLayout. That's one thing
>>to
>> check.
>>
>> Does the Container with id scrollContainer have a parent somewhere with
>>a
>> fixed width and height? The auto margins should still work for HTML.
>> Mostly the work is on the SWF side; the JS code was "cleaned up" so
>>maybe
>> that's the problem. I took out much of the algorithmic code for Vertical
>> and Horizontal layout since I believe HTML would respond property to the
>> addition of children, removal of children, resizing, etc. I did have,
>>if I
>> remember correctly, to add "white-space:nowrap" to the
>>HorizontalLayout; I
>> don't know that will affect you.
>>
>> Would you mind tinker with the HTML DOM in the browser to see if you can
>> get the look you want? That will help determine if the layout code needs
>> to be adjusted.
>>
>> Thanks and so sorry for the problems. *must use feature branch next
>>time*
>> —peter
>>
>> On 3/27/17, 4:35 PM, "Harbs"  wrote:
>>
>> >Better, but I still have some problems (there’s probably more):
>> >
>> >   
>> >   
>> >   > backgroundColor="0x44"/>
>> >   
>> >   
>> >   
>> >   > marginLeft="auto" marginRight="auto"/>
>> >   
>> >   
>> >   
>> >   
>> >   > >click="undo_clickHandler(event)" id="undoButton"
>> >src="assets/images/icons/0726-undo.svg">
>> >   
>> >   
>> >   
>> >   
>> >   > >click="redo_clickHandler(event)" id="redoButton"
>> >src="assets/images/icons/0727-redo.svg">
>> >   
>> >   
>> >   
>> >   
>> >   > >click="zoomin_clickHandler(event)"
>> >src="assets/images/icons/0806-zoom-in.svg"/>
>> >   > >click="zoomout_clickHandler(event)"
>> >src="assets/images/icons/0807-zoom-out.svg"/>
>> >   > >click="fitButton_clickHandler(event)"
>> >src="assets/images/icons/0843-expand.svg"/>
>> >   > >click="previewButton_clickHandler(event)"
>> >src="assets/images/icons/0786-file-preview-white.svg"/>
>> >   
>> >   > >click="finishButton_clickHandler()"
>> >src="assets/images/icons/0821-check.svg"/>
>> >   
>> >   > >click="cancelButton_clickHandler()"
>> >src="assets/images/icons/0822-cross2.svg"/>
>> >   
>> >
>> >1. This used to create a centered group of buttons. Now, the container
>> >has a height of 0 and the buttons don’t show up.
>> >
>> >2.
>> >   >className="designOuterContainer"
>> id="scrollContainer"
>> >width="100%" height="100%">
>> >   
>> >   
>> >   
>> >   > 

Re: [FlexJS] Summary of Changes

[Peter Ent]

The HorizontalLayout and VerticalLayout do not center their content
because the equivalent code on the JS side does not do that (at least it
did not do that for me in the tests I ran). I think setting margin:auto is
supposed to help with that.

On the other hand, you can switch to HorizontalFlexLayout and
VerticalFlexLayout. Then on the Group or Container CSS, do the following:

.styleForVerticalFlexLayout {
align-items: center;
}

.styleForHorizontalFlexLayout {
align-items: center;
}

This is still a learning experience both understanding the complexities of
CSS and building an system that generates clean and robust HTML/CSS/JS.

Thank you for your patience.

—peter

On 3/27/17, 4:46 PM, "Harbs"  wrote:

>I probably need to examine the new “flex” layouts.
>
>Is there a way to have content centered using those?
>
>> On Mar 27, 2017, at 11:35 PM, Harbs  wrote:
>> 
>> Better, but I still have some problems (there’s probably more):
>> 
>>  
>>  
>>  
>>  
>>  
>>  
>>  > marginRight="auto"/>
>>  
>>  
>>  
>>  
>>  > width="30" height="30"
>>click="undo_clickHandler(event)" id="undoButton"
>>src="assets/images/icons/0726-undo.svg">
>>  
>>  
>>  
>>  
>>  > width="30" height="30"
>>click="redo_clickHandler(event)" id="redoButton"
>>src="assets/images/icons/0727-redo.svg">
>>  
>>  
>>  
>>  
>>  > width="30" height="30"
>>click="zoomin_clickHandler(event)"
>>src="assets/images/icons/0806-zoom-in.svg"/>
>>  > width="30" height="30"
>>click="zoomout_clickHandler(event)"
>>src="assets/images/icons/0807-zoom-out.svg"/>
>>  > width="30" height="30"
>>click="fitButton_clickHandler(event)"
>>src="assets/images/icons/0843-expand.svg"/>
>>  > width="30" height="30"
>>click="previewButton_clickHandler(event)"
>>src="assets/images/icons/0786-file-preview-white.svg"/>
>>  
>>  > width="30" height="30"
>>click="finishButton_clickHandler()"
>>src="assets/images/icons/0821-check.svg"/>
>>  
>>  > width="30" height="30"
>>click="cancelButton_clickHandler()"
>>src="assets/images/icons/0822-cross2.svg"/>
>>  
>> 
>> 1. This used to create a centered group of buttons. Now, the container
>>has a height of 0 and the buttons don’t show up.
>> 
>> 2.
>>  > id="scrollContainer"
>>width="100%" height="100%">
>>  
>>  
>>  
>>  > id="designContainer"
>>className="_holder" y="0">
>>  
>>  > marginLeft="auto" marginRight="auto"/>
>>  
>>  
>>  
>>  
>> 
>> This used to create a scrolling div that was centered in the the
>>window. It now is aligned left, and I don’t know if it scrolls.
>> 
>> There’s other issues, and I’ll see tomorrow what I can work around.
>> 
>>> On Mar 27, 2017, at 11:08 PM, Peter Ent  wrote:
>>> 
>>> Hi,
>>> 
>>> I just pushed a change to UIBase to set position="absolute" when
>>>setting x
>>> or y. I think this is perfectly safe and if someone does set x and y
>>>and
>>> then tries to use a layout where that would be a conflict, they will
>>>get
>>> have to avoid setting those properties.
>>> 
>>> I figured this would eventually happen. Let's see if this fixes the
>>>issue.
>>> 
>>> —peter
>>> 
>>> 
>>> On 3/27/17, 2:58 PM, "Harbs"  wrote:
>>> 
 Peter,
 
 I just tried loading my app with your changes, and everything is
totally
 borked. We rely a lot on absolute positioning and transformations.
 
 We really need the old behavior in some components.
 
 Is there any components which work the same as they used to?
 
 Harbs
 
> On Mar 26, 2017, at 6:55 PM, Peter Ent  wrote:
> 
> For the time being, the Tour main view should have a width and a
>height:
> 
> 
> 
> Then in 

[DISCUSS] Changing our general mode of development?

[Christofer Dutz]

Hi,

For the last months, we have seen a huge increase in people working on the 
FlexJS and people working on first applications using FlexJS. I think we should 
discuss how we can make sure we don’t have interruptions like the current one 
in the future.

One point that has been causing pain in the past, was that some people are 
using Ant and some are using Maven. Maven is quite a bit more restrictive than 
Ant and it builds a lot more and tests a lot more. Just as an example in 
contrast to the Ant build the Maven build builds all Examples and it also tests 
some of them to be runnable in a browser. The Ant build only builds the 
framework and most of the latest problems only pop up if you build an 
application. It has occurred several times that Changed failed the Maven build 
but didn’t fail the Ant build … just because the Ant build doesn’t build 
everything. We could avoid this problem if people would not simply ignore build 
failures reported by the ASF Jenkins, which is taking care of the Maven build. 
It is currently setup to give feedback within an hour or so.

Sometimes the “fix” was to exclude a module in Maven. This usually had the 
side-effect of the RAT plugin failing after that because it now finds files 
without Apache headers. A quick solution to that problem is to log-in to the 
ASF Jenkins and to click on “wipe workspace” of that build. After that this 
type of problem should go away immediately.

Another point was that sometimes people work together on a larger refactoring 
and check-in stuff to develop in order to share code. We should start using 
feature branches for this. This has currently not been happening at all. I have 
setup everything that if you create a branch IN ALL 3 REPOS with a name 
“feature/{somename}” (but the same “somename” in all three ;-) )  the ASF 
Jenkins will setup a Job for that which builds all parts in one go and give you 
immediate feedback on the state of your branch. Feature branches that are not 
“blue” should not be merged back to develop.

One last pattern I have encountered was people reporting stuff like: “I have 
been working on X and have almost finished ... I know it will break Y, but I’ll 
push my changes and fix Y after that” … keep in mind: By breaking Y everyone 
working on FlexJS is forced to stop working so I will probably veto every 
suggestion I encounter on the list that has a similar pattern.

FlexJS has matured and we are approaching a 1.0, but we also must mature the 
way we develop or we will hurt early adopters and people willing to help get 
FlexJS to shape. We want enterprise users to use our stuff, then we must start 
working in an enterprise-acceptable way.

Keep up the awesome work and lets just get a little more awesome ;-)

Chris

Re: [DISCUSS] Changing our general mode of development?

[Harbs]

+1.

I think it’s OK to develop however we might be comfortable on a feature branch, 
but we definitely want an approved procedure which must be done before 
committing to the develop branch.

Harbs

> On Mar 28, 2017, at 12:29 PM, Christofer Dutz  
> wrote:
> 
> Hi,
> 
> For the last months, we have seen a huge increase in people working on the 
> FlexJS and people working on first applications using FlexJS. I think we 
> should discuss how we can make sure we don’t have interruptions like the 
> current one in the future.
> 
> One point that has been causing pain in the past, was that some people are 
> using Ant and some are using Maven. Maven is quite a bit more restrictive 
> than Ant and it builds a lot more and tests a lot more. Just as an example in 
> contrast to the Ant build the Maven build builds all Examples and it also 
> tests some of them to be runnable in a browser. The Ant build only builds the 
> framework and most of the latest problems only pop up if you build an 
> application. It has occurred several times that Changed failed the Maven 
> build but didn’t fail the Ant build … just because the Ant build doesn’t 
> build everything. We could avoid this problem if people would not simply 
> ignore build failures reported by the ASF Jenkins, which is taking care of 
> the Maven build. It is currently setup to give feedback within an hour or so.
> 
> Sometimes the “fix” was to exclude a module in Maven. This usually had the 
> side-effect of the RAT plugin failing after that because it now finds files 
> without Apache headers. A quick solution to that problem is to log-in to the 
> ASF Jenkins and to click on “wipe workspace” of that build. After that this 
> type of problem should go away immediately.
> 
> Another point was that sometimes people work together on a larger refactoring 
> and check-in stuff to develop in order to share code. We should start using 
> feature branches for this. This has currently not been happening at all. I 
> have setup everything that if you create a branch IN ALL 3 REPOS with a name 
> “feature/{somename}” (but the same “somename” in all three ;-) )  the ASF 
> Jenkins will setup a Job for that which builds all parts in one go and give 
> you immediate feedback on the state of your branch. Feature branches that are 
> not “blue” should not be merged back to develop.
> 
> One last pattern I have encountered was people reporting stuff like: “I have 
> been working on X and have almost finished ... I know it will break Y, but 
> I’ll push my changes and fix Y after that” … keep in mind: By breaking Y 
> everyone working on FlexJS is forced to stop working so I will probably veto 
> every suggestion I encounter on the list that has a similar pattern.
> 
> FlexJS has matured and we are approaching a 1.0, but we also must mature the 
> way we develop or we will hurt early adopters and people willing to help get 
> FlexJS to shape. We want enterprise users to use our stuff, then we must 
> start working in an enterprise-acceptable way.
> 
> Keep up the awesome work and lets just get a little more awesome ;-)
> 
> Chris
> 
> 
> 

Re: [DISCUSS] Changing our general mode of development?

[Carlos Rovira]

+1 to all of this.

For people looking at the other's work it more easy to follow git graphs if
people make it's own branch.
I suscribe all Chris's email and agree with all.

So hope folks embrace the maven way of building and abandone ANT since
there's no benefit at all in that build system.

Thanks Chris for bringing this :)



2017-03-28 12:04 GMT+02:00 Harbs :

> I do think that development should be done almost exclusively on feature
> branches.
>
> If the build status for a feature branch can be verified on the server
> (like you set it up) that’s ideal because it does not require using
> specifically ant or maven locally.
>
> > On Mar 28, 2017, at 12:59 PM, Harbs  wrote:
> >
> > I still have not managed to get maven setup correctly.
> >
> >> On Mar 28, 2017, at 12:46 PM, Christofer Dutz <
> christofer.d...@c-ware.de> wrote:
> >>
> >> Hi Harbs,
> >> (this time replying to the right name ;-) )
> >>
> >> I usually simply make sure I update my repos and do the full maven
> build with tests and examples locally before pushing … I guess this is
> sufficient protection against most problems. In IntelliJ that’s two clicks
> and a cup of coffee or whatever beverage you prefer.
> >>
> >> Chris
> >>
> >>
> >> Am 28.03.17, 11:41 schrieb "Harbs" :
> >>
> >>   +1.
> >>
> >>   I think it’s OK to develop however we might be comfortable on a
> feature branch, but we definitely want an approved procedure which must be
> done before committing to the develop branch.
> >>
> >>   Harbs
> >>
> >>> On Mar 28, 2017, at 12:29 PM, Christofer Dutz <
> christofer.d...@c-ware.de> wrote:
> >>>
> >>> Hi,
> >>>
> >>> For the last months, we have seen a huge increase in people working on
> the FlexJS and people working on first applications using FlexJS. I think
> we should discuss how we can make sure we don’t have interruptions like the
> current one in the future.
> >>>
> >>> One point that has been causing pain in the past, was that some people
> are using Ant and some are using Maven. Maven is quite a bit more
> restrictive than Ant and it builds a lot more and tests a lot more. Just as
> an example in contrast to the Ant build the Maven build builds all Examples
> and it also tests some of them to be runnable in a browser. The Ant build
> only builds the framework and most of the latest problems only pop up if
> you build an application. It has occurred several times that Changed failed
> the Maven build but didn’t fail the Ant build … just because the Ant build
> doesn’t build everything. We could avoid this problem if people would not
> simply ignore build failures reported by the ASF Jenkins, which is taking
> care of the Maven build. It is currently setup to give feedback within an
> hour or so.
> >>>
> >>> Sometimes the “fix” was to exclude a module in Maven. This usually had
> the side-effect of the RAT plugin failing after that because it now finds
> files without Apache headers. A quick solution to that problem is to log-in
> to the ASF Jenkins and to click on “wipe workspace” of that build. After
> that this type of problem should go away immediately.
> >>>
> >>> Another point was that sometimes people work together on a larger
> refactoring and check-in stuff to develop in order to share code. We should
> start using feature branches for this. This has currently not been
> happening at all. I have setup everything that if you create a branch IN
> ALL 3 REPOS with a name “feature/{somename}” (but the same “somename” in
> all three ;-) )  the ASF Jenkins will setup a Job for that which builds all
> parts in one go and give you immediate feedback on the state of your
> branch. Feature branches that are not “blue” should not be merged back to
> develop.
> >>>
> >>> One last pattern I have encountered was people reporting stuff like:
> “I have been working on X and have almost finished ... I know it will break
> Y, but I’ll push my changes and fix Y after that” … keep in mind: By
> breaking Y everyone working on FlexJS is forced to stop working so I will
> probably veto every suggestion I encounter on the list that has a similar
> pattern.
> >>>
> >>> FlexJS has matured and we are approaching a 1.0, but we also must
> mature the way we develop or we will hurt early adopters and people willing
> to help get FlexJS to shape. We want enterprise users to use our stuff,
> then we must start working in an enterprise-acceptable way.
> >>>
> >>> Keep up the awesome work and lets just get a little more awesome ;-)
> >>>
> >>> Chris
> >>>
> >>>
> >>>
> >>
> >>
> >>
> >
>
>


-- 



Carlos Rovira

Director General

M: +34 607 22 60 05

http://www.codeoscopic.com

http://www.avant2.es


Este mensaje se dirige exclusivamente a su destinatario y puede contener
información privilegiada o confidencial. Si ha recibido este mensaje por
error, le rogamos que nos lo comunique inmediatamente por esta misma vía y
proceda a su destrucción.

De la 

Re: [FlexJS] Simple App Not Running

[Christofer Dutz]

Hi Yishay,

The Ant build doesn’t build a single Application so I would say if you build 
with Maven without examples this will succeed too, but that doesn’t mean that 
the build is correct. (See the DISCUSS thread I just opened for details)

Chris

Am 28.03.17, 11:21 schrieb "Harbs" :

The ant build of the framework is working.

> On Mar 28, 2017, at 12:11 PM, Christofer Dutz  
wrote:
> 
> Hi Yishay,
> 
> Well I would guess that everyone is having the same problem. Due to 
latest refactoring things are broken and not even the Jenkins build is working.
> Even if it is providing new versions of the Framework libraries, these 
currently can’t be used to build applications till the issues are fixed.
> 
> Chris
> 
> Am 28.03.17, 10:53 schrieb "yishayw" :
> 
>Same problem with a fresh nightly. I anyone else getting this?
> 
> 
> 
>--
>View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Simple-App-Not-Running-tp60825p60826.html
>Sent from the Apache Flex Development mailing list archive at 
Nabble.com.
> 
> 

Re: [DISCUSS] Changing our general mode of development?

[Christofer Dutz]

Hi Harbs,
(this time replying to the right name ;-) )

I usually simply make sure I update my repos and do the full maven build with 
tests and examples locally before pushing … I guess this is sufficient 
protection against most problems. In IntelliJ that’s two clicks and a cup of 
coffee or whatever beverage you prefer.

Chris


Am 28.03.17, 11:41 schrieb "Harbs" :

+1.

I think it’s OK to develop however we might be comfortable on a feature 
branch, but we definitely want an approved procedure which must be done before 
committing to the develop branch.

Harbs

> On Mar 28, 2017, at 12:29 PM, Christofer Dutz  
wrote:
> 
> Hi,
> 
> For the last months, we have seen a huge increase in people working on 
the FlexJS and people working on first applications using FlexJS. I think we 
should discuss how we can make sure we don’t have interruptions like the 
current one in the future.
> 
> One point that has been causing pain in the past, was that some people 
are using Ant and some are using Maven. Maven is quite a bit more restrictive 
than Ant and it builds a lot more and tests a lot more. Just as an example in 
contrast to the Ant build the Maven build builds all Examples and it also tests 
some of them to be runnable in a browser. The Ant build only builds the 
framework and most of the latest problems only pop up if you build an 
application. It has occurred several times that Changed failed the Maven build 
but didn’t fail the Ant build … just because the Ant build doesn’t build 
everything. We could avoid this problem if people would not simply ignore build 
failures reported by the ASF Jenkins, which is taking care of the Maven build. 
It is currently setup to give feedback within an hour or so.
> 
> Sometimes the “fix” was to exclude a module in Maven. This usually had 
the side-effect of the RAT plugin failing after that because it now finds files 
without Apache headers. A quick solution to that problem is to log-in to the 
ASF Jenkins and to click on “wipe workspace” of that build. After that this 
type of problem should go away immediately.
> 
> Another point was that sometimes people work together on a larger 
refactoring and check-in stuff to develop in order to share code. We should 
start using feature branches for this. This has currently not been happening at 
all. I have setup everything that if you create a branch IN ALL 3 REPOS with a 
name “feature/{somename}” (but the same “somename” in all three ;-) )  the ASF 
Jenkins will setup a Job for that which builds all parts in one go and give you 
immediate feedback on the state of your branch. Feature branches that are not 
“blue” should not be merged back to develop.
> 
> One last pattern I have encountered was people reporting stuff like: “I 
have been working on X and have almost finished ... I know it will break Y, but 
I’ll push my changes and fix Y after that” … keep in mind: By breaking Y 
everyone working on FlexJS is forced to stop working so I will probably veto 
every suggestion I encounter on the list that has a similar pattern.
> 
> FlexJS has matured and we are approaching a 1.0, but we also must mature 
the way we develop or we will hurt early adopters and people willing to help 
get FlexJS to shape. We want enterprise users to use our stuff, then we must 
start working in an enterprise-acceptable way.
> 
> Keep up the awesome work and lets just get a little more awesome ;-)
> 
> Chris
> 
> 
> 

Re: Falcon build failing

[Harbs]

Yes. That’s what happened, and I got past that problem.

My problem now is not related to the tests.

I’m having two problems:
1. org.apache.flex.core.Application is not being added as a dependency to the 
main app.
2. Any app which uses BinaryData fails to compile.
This app: https://paste.apache.org/HobL 
outputs: https://paste.apache.org/B2je 

If I comment out BinaryData, I get output, but the dependencies of the main app 
looks like this (no org.apache.flex.core.Application dependency). I get a 
runtime error when calling goog.base unless I add the dependency manually.

// generated by FalconJX
goog.addDependency('../../../test_project.js', ['test_project'], 
['org.apache.flex.core.IBeadModel', 'org.apache.flex.events.ValueChangeEvent', 
'org.apache.flex.core.IFlexJSElement', 
'org.apache.flex.html.beads.layouts.BasicLayout', 
'org.apache.flex.utils.MXMLDataInterpreter', 
'org.apache.flex.core.ValuesManager', 'org.apache.flex.core.IBeadController', 
'org.apache.flex.utils.Timer', 'org.apache.flex.events.BrowserEvent', 
'org.apache.flex.core.IMeasurementBead', 
'org.apache.flex.events.utils.EventUtils', 'org.apache.flex.core.View', 
'org.apache.flex.events.ValueEvent', 'org.apache.flex.utils.CSSUtils', 
'org.apache.flex.utils.Language', 'org.apache.flex.html.Label', 
'org.apache.flex.states.ItemAndDescriptor', 
'org.apache.flex.html.beads.GroupView', 'org.apache.flex.events.ElementEvents', 
'org.apache.flex.utils.StringPadder', 'org.apache.flex.utils.EffectTimer', 
'org.apache.flex.core.SimpleStatesImpl', 
'org.apache.flex.states.SetEventHandler', 
'org.apache.flex.utils.StringTrimmer', 'org.apache.flex.utils.StringUtil', 
'org.apache.flex.core.IContainer', 'org.apache.flex.states.AddItems', 
'org.apache.flex.core.SimpleCSSValuesImpl', 
'org.apache.flex.states.SetProperty', 
'org.apache.flex.html.beads.layouts.HorizontalFlexLayout']);

Harbs

> On Mar 28, 2017, at 11:34 AM, Christofer Dutz  
> wrote:
> 
> Just an edjucated guess from my side … 
> 
> There is a cycle in the dependencies. Maven has problems with this and simply 
> fails to start. Ant doesn’t do this sort of check.
> If you do a wipe all, eventually the prerequisites of your test get wiped and 
> therefore you can’t run the test so you would have to 
> Build once without tests and then your test could run … is that possibly 
> what’s happening?
> 
> Chris
> 
> Am 28.03.17, 10:29 schrieb "Harbs" :
> 
>It’s really weird: I can’t see anything in my app which relies on 
> Uint8Array and  Uint32Array, and I’m still getting this error. :-(
> 
>> On Mar 28, 2017, at 11:21 AM, Harbs  wrote:
>> 
>> It was Josh’s tests that were tripping me up. The build fails after doing a 
>> wipe-all.
>> 
>> I’ve been having other problems with Falcon, and the latest logging seems to 
>> shed some light on what’s wrong:
>> 
>> [java] Dependencies calculated for 'org.apache.flex.states.SetProperty'
>> [java] org.apache.flex.states.SetProperty depends on 
>> org.apache.flex.core.IDocument
>> [java] No GoogDep for Uint8Array
>> [java] No GoogDep for Uint32Array
>> [java] 
>> org.apache.flex.compiler.internal.graph.GoogDepsWriter.getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.clients.MXMLJSC.main(MXMLJSC.java:176)
>> 
>> It looks like an app which depends on BinaryData will fail. I’m guessing 
>> that for some reason falcon is trying to add requires for native JS APIs.
>> 
>> Harbs
>> 
>>> On Mar 27, 2017, at 7:13 AM, Alex Harui >> > wrote:
>>> 
>>> 
>>> 
>>> On 3/26/17, 1:19 PM, "Harbs" >> > wrote:
>>> 
 I’m having trouble building Falcon.
 
 It looks like it’s failing because a text in FlexJSTestBase.java is
 failing. It’s looking for a file
 flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
>>> 
>>> That output has been there for months.  It isn't the cause of a failure.
>>> You have to find the test results and check the output in there.
>>> 
>>> -Alex
>>> 
>> 
> 
> 
> 

Re: Falcon build failing

[Greg Dove]

Harbs, I saw that error a short while ago, but I am not sure it is related
to the native classes.

Dependencies calculated for 'org.apache.flex.states.SetProperty'
org.apache.flex.states.SetProperty depends on org.apache.flex.core.IDocument
was all the appeared immediately before I saw it.

If you can avoid using remove-circulars (if that is an option for you) then
things might still work.

On Tue, Mar 28, 2017 at 9:29 PM, Harbs  wrote:

> It’s really weird: I can’t see anything in my app which relies on
> Uint8Array and  Uint32Array, and I’m still getting this error. :-(
>
> > On Mar 28, 2017, at 11:21 AM, Harbs  wrote:
> >
> > It was Josh’s tests that were tripping me up. The build fails after
> doing a wipe-all.
> >
> > I’ve been having other problems with Falcon, and the latest logging
> seems to shed some light on what’s wrong:
> >
> >  [java] Dependencies calculated for 'org.apache.flex.states.
> SetProperty'
> >  [java] org.apache.flex.states.SetProperty depends on
> org.apache.flex.core.IDocument
> >  [java] No GoogDep for Uint8Array
> >  [java] No GoogDep for Uint32Array
> >  [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
> clients.MXMLJSC.main(MXMLJSC.java:176)
> >
> > It looks like an app which depends on BinaryData will fail. I’m guessing
> that for some reason falcon is trying to add requires for native JS APIs.
> >
> > Harbs
> >
> >> On Mar 27, 2017, at 7:13 AM, Alex Harui > wrote:
> >>
> >>
> >>
> >> On 3/26/17, 1:19 PM, "Harbs" > wrote:
> >>
> >>> I’m having trouble building Falcon.
> >>>
> >>> It looks like it’s failing because a text in FlexJSTestBase.java is
> >>> failing. It’s looking for a file
> >>> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
> >>
> >> That output has been there for months.  It isn't the cause of a failure.
> >> You have to find the test results and check the output in there.
> >>
> >> -Alex
> >>
> >
>
>

Re: Falcon build failing

[Christofer Dutz]

Just an edjucated guess from my side … 

There is a cycle in the dependencies. Maven has problems with this and simply 
fails to start. Ant doesn’t do this sort of check.
If you do a wipe all, eventually the prerequisites of your test get wiped and 
therefore you can’t run the test so you would have to 
Build once without tests and then your test could run … is that possibly what’s 
happening?

Chris

Am 28.03.17, 10:29 schrieb "Harbs" :

It’s really weird: I can’t see anything in my app which relies on 
Uint8Array and  Uint32Array, and I’m still getting this error. :-(

> On Mar 28, 2017, at 11:21 AM, Harbs  wrote:
> 
> It was Josh’s tests that were tripping me up. The build fails after doing 
a wipe-all.
> 
> I’ve been having other problems with Falcon, and the latest logging seems 
to shed some light on what’s wrong:
> 
>  [java] Dependencies calculated for 
'org.apache.flex.states.SetProperty'
>  [java] org.apache.flex.states.SetProperty depends on 
org.apache.flex.core.IDocument
>  [java] No GoogDep for Uint8Array
>  [java] No GoogDep for Uint32Array
>  [java] 
org.apache.flex.compiler.internal.graph.GoogDepsWriter.getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.clients.MXMLJSC.main(MXMLJSC.java:176)
> 
> It looks like an app which depends on BinaryData will fail. I’m guessing 
that for some reason falcon is trying to add requires for native JS APIs.
> 
> Harbs
> 
>> On Mar 27, 2017, at 7:13 AM, Alex Harui > wrote:
>> 
>> 
>> 
>> On 3/26/17, 1:19 PM, "Harbs" > wrote:
>> 
>>> I’m having trouble building Falcon.
>>> 
>>> It looks like it’s failing because a text in FlexJSTestBase.java is
>>> failing. It’s looking for a file
>>> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
>> 
>> That output has been there for months.  It isn't the cause of a failure.
>> You have to find the test results and check the output in there.
>> 
>> -Alex
>> 
> 

Re: [FlexJS] Simple App Not Running

[yishayw]

Stack trace reveals that in 

org.apache.flex.events.Event.js

in line 33:
goog.inherits(org.apache.flex.events.Event, goog.events.Event);

goog.events.Event is undefined, even though line 13 is

goog.require('goog.events.Event');




--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Simple-App-Not-Running-tp60825p60828.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] Simple App Not Running

[Harbs]

The ant build of the framework is working.

> On Mar 28, 2017, at 12:11 PM, Christofer Dutz  
> wrote:
> 
> Hi Yishay,
> 
> Well I would guess that everyone is having the same problem. Due to latest 
> refactoring things are broken and not even the Jenkins build is working.
> Even if it is providing new versions of the Framework libraries, these 
> currently can’t be used to build applications till the issues are fixed.
> 
> Chris
> 
> Am 28.03.17, 10:53 schrieb "yishayw" :
> 
>Same problem with a fresh nightly. I anyone else getting this?
> 
> 
> 
>--
>View this message in context: 
> http://apache-flex-development.247.n4.nabble.com/FlexJS-Simple-App-Not-Running-tp60825p60826.html
>Sent from the Apache Flex Development mailing list archive at Nabble.com.
> 
> 

[FlexJX][FalconJX]

[Greg Dove]

Hi Alex, if you have time, perhaps you can shed some light on this?

I have an mxml component using states, that implements an Interface, 
IFormSequence.

the generated CLASS_INFO looks like this:
FLEXJS_CLASS_INFO = { names: [{ name: 'ActionForm', qName: 
'components.forms.ActionForm', kind: 'class'  }], interfaces: 
[components.forms.IFormSequence] };

and the goog.require for the interface is there at the top
goog.require('components.forms.IFormSequence');

I notice that the 
goog.addDependency('../../../components/forms/IFormSequence.js', 
['components.forms.IFormSequence'], []);

appears after the 
goog.addDependency('../../../components/forms/ActionForm.js', 
['components.forms.ActionForm'], ['org.apache.flex.html.Group']);
in the index.html.
Perhaps the order here is not important and goog resolves everything somehow, 
but I did wonder if the dependency chain in for ActionForm.js above should also 
contain the interface (that's a question, not a suggestion - I don't know what 
is right here). At the moment there is an issue where there is an interface 
check on startup when states or styles or something like this are being 
initialized (it is actually a check for a different interface, but because the 
'interfaces' array is defined and has length in CLASS_INFO it looks there and 
tries to follow follow the inheritance chain on the undefined Interface 
reference, causing mayhem). Because the reference inside the CLASS_INFO 
evaluates to undefined, I know the dependencies are not loading in the right 
order. 

This was working last week so either it is something random/unlrelated 
elsewhere in the project that has changed the order, or the compiler has 
changed perhaps. Could your changes to goog.requires ordering have something to 
do with this, do you think?

I have not checked in the compiler to see if mxml "implements' is being treated 
the same as a regular actionscript 'implements' or if there are different code 
paths internally that might create differences for mxml based output vs regular 
AS, which was another thing I thought about here, because I think 'implements' 
is not often used in mxml. I'm happy to do that if you think that might warrant 
investigation, but figured I would check on whether you thought the other 
changes could be affecting this first.

Also, fyi, I was not using removeCirculars (although I don't *think* there are 
any circular references here and that part has not changed since it worked 
previously). fyi, when I tried remove circulars I got a compilation fail at 
org.apache.flex.compiler.internal.graph.GoogDepsWriter.getListOfFiles(GoogDepsWriter.java:103)

[FlexJS] Simple App Not Running

[yishayw]

After updating falcon this [1] app, gives the following error:

Uncaught TypeError: Cannot read property 'prototype' of undefined
at Object.goog.inherits (base.js:2408)
at a_project.js:106

base.js:2408 looks like this:

  tempCtor.prototype = parentCtor.prototype;
 
and parentCtor is marked as undefined.

I'm going to try the same app on the latest nightly build and see if I'm
having the same problem.

[1] https://paste.apache.org/Rmzt



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Simple-App-Not-Running-tp60825.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: Falcon build failing

[Harbs]

It was Josh’s tests that were tripping me up. The build fails after doing a 
wipe-all.

I’ve been having other problems with Falcon, and the latest logging seems to 
shed some light on what’s wrong:

 [java] Dependencies calculated for 'org.apache.flex.states.SetProperty'
 [java] org.apache.flex.states.SetProperty depends on 
org.apache.flex.core.IDocument
 [java] No GoogDep for Uint8Array
 [java] No GoogDep for Uint32Array
 [java] 
org.apache.flex.compiler.internal.graph.GoogDepsWriter.getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.clients.MXMLJSC.main(MXMLJSC.java:176)

It looks like an app which depends on BinaryData will fail. I’m guessing that 
for some reason falcon is trying to add requires for native JS APIs.

Harbs

> On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
> 
> 
> 
> On 3/26/17, 1:19 PM, "Harbs"  wrote:
> 
>> I’m having trouble building Falcon.
>> 
>> It looks like it’s failing because a text in FlexJSTestBase.java is
>> failing. It’s looking for a file
>> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
> 
> That output has been there for months.  It isn't the cause of a failure.
> You have to find the test results and check the output in there.
> 
> -Alex
> 

Re: Falcon build failing

[Harbs]

It’s really weird: I can’t see anything in my app which relies on Uint8Array 
and  Uint32Array, and I’m still getting this error. :-(

> On Mar 28, 2017, at 11:21 AM, Harbs  wrote:
> 
> It was Josh’s tests that were tripping me up. The build fails after doing a 
> wipe-all.
> 
> I’ve been having other problems with Falcon, and the latest logging seems to 
> shed some light on what’s wrong:
> 
>  [java] Dependencies calculated for 'org.apache.flex.states.SetProperty'
>  [java] org.apache.flex.states.SetProperty depends on 
> org.apache.flex.core.IDocument
>  [java] No GoogDep for Uint8Array
>  [java] No GoogDep for Uint32Array
>  [java] 
> org.apache.flex.compiler.internal.graph.GoogDepsWriter.getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.clients.MXMLJSC.main(MXMLJSC.java:176)
> 
> It looks like an app which depends on BinaryData will fail. I’m guessing that 
> for some reason falcon is trying to add requires for native JS APIs.
> 
> Harbs
> 
>> On Mar 27, 2017, at 7:13 AM, Alex Harui > > wrote:
>> 
>> 
>> 
>> On 3/26/17, 1:19 PM, "Harbs" > > wrote:
>> 
>>> I’m having trouble building Falcon.
>>> 
>>> It looks like it’s failing because a text in FlexJSTestBase.java is
>>> failing. It’s looking for a file
>>> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
>> 
>> That output has been there for months.  It isn't the cause of a failure.
>> You have to find the test results and check the output in there.
>> 
>> -Alex
>> 
> 

Re: [FlexJS] Simple App Not Running

[yishayw]

Same problem with a fresh nightly. I anyone else getting this?



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Simple-App-Not-Running-tp60825p60826.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] Summary of Changes

[piotrz]

Peter,

Could you look into TourJS, maybe there you also found some issue with
layout there. I'm really interesting how to resolve it there.

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Summary-of-Changes-tp60709p60845.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: Falcon build failing

[Josh Tynjala]

I have moved my new tests into the Ant integration tests, so they should no
longer block you.

- Josh

On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:

> It was Josh’s tests that were tripping me up. The build fails after doing
> a wipe-all.
>
> I’ve been having other problems with Falcon, and the latest logging seems
> to shed some light on what’s wrong:
>
>  [java] Dependencies calculated for 'org.apache.flex.states.
> SetProperty'
>  [java] org.apache.flex.states.SetProperty depends on
> org.apache.flex.core.IDocument
>  [java] No GoogDep for Uint8Array
>  [java] No GoogDep for Uint32Array
>  [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
> clients.MXMLJSC.main(MXMLJSC.java:176)
>
> It looks like an app which depends on BinaryData will fail. I’m guessing
> that for some reason falcon is trying to add requires for native JS APIs.
>
> Harbs
>
> > On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
> >
> >
> >
> > On 3/26/17, 1:19 PM, "Harbs"  wrote:
> >
> >> I’m having trouble building Falcon.
> >>
> >> It looks like it’s failing because a text in FlexJSTestBase.java is
> >> failing. It’s looking for a file
> >> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
> >
> > That output has been there for months.  It isn't the cause of a failure.
> > You have to find the test results and check the output in there.
> >
> > -Alex
> >
>
>

Re: Falcon build failing

[Harbs]

Thanks, but I’m having bigger problems. Hopefully Alex will look into the other 
issues…

> On Mar 28, 2017, at 4:55 PM, Josh Tynjala  wrote:
> 
> I have moved my new tests into the Ant integration tests, so they should no
> longer block you.
> 
> - Josh
> 
> On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:
> 
>> It was Josh’s tests that were tripping me up. The build fails after doing
>> a wipe-all.
>> 
>> I’ve been having other problems with Falcon, and the latest logging seems
>> to shed some light on what’s wrong:
>> 
>> [java] Dependencies calculated for 'org.apache.flex.states.
>> SetProperty'
>> [java] org.apache.flex.states.SetProperty depends on
>> org.apache.flex.core.IDocument
>> [java] No GoogDep for Uint8Array
>> [java] No GoogDep for Uint32Array
>> [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
>> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
>> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
>> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
>> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
>> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
>> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
>> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
>> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
>> clients.MXMLJSC.main(MXMLJSC.java:176)
>> 
>> It looks like an app which depends on BinaryData will fail. I’m guessing
>> that for some reason falcon is trying to add requires for native JS APIs.
>> 
>> Harbs
>> 
>>> On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
>>> 
>>> 
>>> 
>>> On 3/26/17, 1:19 PM, "Harbs"  wrote:
>>> 
 I’m having trouble building Falcon.
 
 It looks like it’s failing because a text in FlexJSTestBase.java is
 failing. It’s looking for a file
 flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
>>> 
>>> That output has been there for months.  It isn't the cause of a failure.
>>> You have to find the test results and check the output in there.
>>> 
>>> -Alex
>>> 
>> 
>> 

Re: [FlexJS] Summary of Changes

[Peter Ent]

I just pushed a change to the tour code. It should be better. We should be
able to do this with the regular horizontal and vertical layouts with %
sizing, I think, and not just with Flexbox layouts.

One thing I changed was to remove the grow="1" on the layouts. I put that
in there to make it easy when you wanted everything the same size; maybe
it should have been equalSizes=true to be clearer. What happened was that
the height of the button bar (which I set to an explicit 40 just be
certain) was overridden with flex-grow:1. I also gave some other items
flex-grow:1.

I hope it is better.

‹peter

On 3/28/17, 9:22 AM, "piotrz"  wrote:

>Peter,
>
>Could you look into TourJS, maybe there you also found some issue with
>layout there. I'm really interesting how to resolve it there.
>
>Piotr
>
>
>
>-
>Apache Flex PMC
>piotrzarzyck...@gmail.com
>--
>View this message in context:
>https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle
>x-development.247.n4.nabble.com%2FFlexJS-Summary-of-Changes-tp60709p60
>845.html=02%7C01%7C%7C7bf7b688a84b4f6fe97708d475de9039%7Cfa7b1b5a7b34
>438794aed2c178decee1%7C0%7C0%7C636263046144396713=8blALd9qJE6NYqnhdx
>ex8aXxNXeY2ZK0xZpyj95yTU8%3D=0
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] Summary of Changes

[Peter Ent]

Here is a link to a Pasteboard [1] that should help. It is a set of nested
Panels. If you run this on both SWF and HTML, you'll see differences due
to the fact that the SWF side does not yet handle all of the CSS.

The HTML should be a nesting of hideously colored boxes with their content
centered. 

Using Flexbox, you center along the major axis using
justify-content:center and along the second axis using align-items:center.
The justify-content style is how the collection of children along the
major axis are positioned and spaced while align-items positions the
elements of the collection within the box in the second dimension.

Another thing to keep in mind with Flexbox is the size of individual
items. Start off with specifying no size at all. This will expand the item
along the second axis. In the vertical, it will make each item as wide as
the div while in the horizontal it will make each item as tall as the div.

If you give a size to the secondary dimension (width for vertical box and
height for horizontal box), those values should be honored. See that the
buttons for the leftSide Panel have a width specified as 70px.

The flex-grow and flex-shrink styles apply to individual items for the
main axis. You'll see that in the rightSide Panel that the items (Groups)
have flex-grow:1; since they have the same grow value they expand to fill
the space equally. Flexbox supports different grow values but the SWF-side
version supports only flex-grow of 1 or 0. The 0 means to honor the
element's given size for that dimension.

I hope I explained that clearly enough! You can see why this is taking a
bit longer to sort out.

—peter

[1] https://paste.apache.org/cgoQ

On 3/27/17, 5:48 PM, "Greg Dove"  wrote:

>I've just done a sweep through one project fixing our 'borked' stuff, I
>guess the latest change might re-'bork' some of the fixes, but I think at
>least these changes should be easier to address.
>
>Sometimes I needed to swap a Container to a Group and other times not,
>because of the relative/absolute changes.
>I found it easier to migrate a lot of the layout stuff to pure css for now
>using flexbox, thinking that doing so might insulate us from further
>changes in the coded layout support. We would have the ability to swap
>back
>to beads etc later if we want to have that in the mxml, once things have
>become stable.
>
>
>On Tue, Mar 28, 2017 at 9:52 AM, Peter Ent  wrote:
>
>> Does the HTML look OK - the structure. Is there anything missing? You
>> should see a simplified nesting of DIVs. If that's the case, then maybe
>> there is more work to do with the layouts.
>>
>> The topmost Container, outer controls, doesn't look like it has a
>>layout,
>> so with Container that should default to BasicLayout. That's one thing
>>to
>> check.
>>
>> Does the Container with id scrollContainer have a parent somewhere with
>>a
>> fixed width and height? The auto margins should still work for HTML.
>> Mostly the work is on the SWF side; the JS code was "cleaned up" so
>>maybe
>> that's the problem. I took out much of the algorithmic code for Vertical
>> and Horizontal layout since I believe HTML would respond property to the
>> addition of children, removal of children, resizing, etc. I did have,
>>if I
>> remember correctly, to add "white-space:nowrap" to the
>>HorizontalLayout; I
>> don't know that will affect you.
>>
>> Would you mind tinker with the HTML DOM in the browser to see if you can
>> get the look you want? That will help determine if the layout code needs
>> to be adjusted.
>>
>> Thanks and so sorry for the problems. *must use feature branch next
>>time*
>> —peter
>>
>> On 3/27/17, 4:35 PM, "Harbs"  wrote:
>>
>> >Better, but I still have some problems (there’s probably more):
>> >
>> >   
>> >   
>> >   > backgroundColor="0x44"/>
>> >   
>> >   
>> >   
>> >   > marginLeft="auto" marginRight="auto"/>
>> >   
>> >   
>> >   
>> >   
>> >   > >click="undo_clickHandler(event)" id="undoButton"
>> >src="assets/images/icons/0726-undo.svg">
>> >   
>> >   
>> >   
>> >   
>> >   > >click="redo_clickHandler(event)" id="redoButton"
>> >src="assets/images/icons/0727-redo.svg">
>> >   
>> >   
>> >   
>> >   
>> >   > >click="zoomin_clickHandler(event)"
>> >src="assets/images/icons/0806-zoom-in.svg"/>
>> >

Re: Falcon build failing

[Alex Harui]

Hmm.   Looks like I broke something.  Even DataBindingExample didn't run.
I just pushed a change to fix DataBindingExample and hopefully all other
apps.  Please pull flex-falcon and try again.

Sorry about the delay,
-Alex

On 3/28/17, 7:13 AM, "Harbs"  wrote:

>Thanks, but I’m having bigger problems. Hopefully Alex will look into the
>other issues…
>
>> On Mar 28, 2017, at 4:55 PM, Josh Tynjala  wrote:
>> 
>> I have moved my new tests into the Ant integration tests, so they
>>should no
>> longer block you.
>> 
>> - Josh
>> 
>> On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:
>> 
>>> It was Josh’s tests that were tripping me up. The build fails after
>>>doing
>>> a wipe-all.
>>> 
>>> I’ve been having other problems with Falcon, and the latest logging
>>>seems
>>> to shed some light on what’s wrong:
>>> 
>>> [java] Dependencies calculated for 'org.apache.flex.states.
>>> SetProperty'
>>> [java] org.apache.flex.states.SetProperty depends on
>>> org.apache.flex.core.IDocument
>>> [java] No GoogDep for Uint8Array
>>> [java] No GoogDep for Uint32Array
>>> [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
>>> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
>>> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
>>> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
>>> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
>>> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
>>> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
>>> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
>>> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
>>> clients.MXMLJSC.main(MXMLJSC.java:176)
>>> 
>>> It looks like an app which depends on BinaryData will fail. I’m
>>>guessing
>>> that for some reason falcon is trying to add requires for native JS
>>>APIs.
>>> 
>>> Harbs
>>> 
 On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
 
 
 
 On 3/26/17, 1:19 PM, "Harbs"  wrote:
 
> I’m having trouble building Falcon.
> 
> It looks like it’s failing because a text in FlexJSTestBase.java is
> failing. It’s looking for a file
> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
 
 That output has been there for months.  It isn't the cause of a
failure.
 You have to find the test results and check the output in there.
 
 -Alex
 
>>> 
>>> 
>

Re: [FlexJS] Simple App Not Running

[Alex Harui]

Should be fixed with the latest change to flex-falcon.  Please sync up and
try again.

Thanks,
-Alex

On 3/28/17, 2:37 AM, "yishayw"  wrote:

>My mistake, the problem is not with event but with goog.events.Event being
>undefined but with org.apache.flex.core.Application being undefined. So it
>looks like the same problem Harbs was having [1]
>
>[1]
>https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle
>x-development.247.n4.nabble.com%2FFalcon-build-failing-td60790.html
>ta=02%7C01%7C%7C7de55c6bb6e142c6b39108d475bf3a08%7Cfa7b1b5a7b34438794aed2c
>178decee1%7C0%7C0%7C636262911551290968=1ZfQzLu0BQfKi8n06b%2Bzcxw760J
>Z%2Fz3xIVbQBf45M0U%3D=0
>
>
>
>--
>View this message in context:
>https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle
>x-development.247.n4.nabble.com%2FFlexJS-Simple-App-Not-Running-tp6082
>5p60833.html=02%7C01%7C%7C7de55c6bb6e142c6b39108d475bf3a08%7Cfa7b1b5a
>7b34438794aed2c178decee1%7C0%7C0%7C636262911551300972=Ul89yd0lZgMWTM
>Yy078la4eUr6%2FHn0qmJw6Mp8vSz64%3D=0
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS, Maven] Local build failing occasionally

[Alex Harui]

Hi Piotr,

In this log, on line 2247 is the first error.  After that, everything is
broken.

I haven't looked at the Graphics build since the PixelBender features were
added, but I'm not clear that the .pbj files should be compiled.

HTH,
-Alex

On 3/28/17, 11:56 AM, "piotrz"  wrote:

>Hi Alex,
>
>Pasting full log [1]. I see earlier some warning, but not sure whether
>they
>can cause such fail. I'm going through this logs one by one.
>
>1805 bytes written to
>D:\flex_sdk\Sources\flex-asjs\frameworks\projects\XML\target\XML-0.8.0-SNA
>PSHOT.swc
>in 2.009 seconds
>C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airgl
>obal-20.0.swc
>Warning: The definition XML depended on by __AS3__.vec.Vector$object in
>the
>SWC
>C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airgl
>obal-20.0.swc
>could not be found
>
>
>C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airgl
>obal-20.0.swc
>Warning: The definition XMLList depended on by __AS3__.vec.Vector$object
>in
>the SWC
>C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airgl
>obal-20.0.swc
>could not be found
>
>[1] 
>https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpaste.apa
>che.org%2FItmA=02%7C01%7C%7Cf257976a7a4b48be3e5e08d4760d533d%7Cfa7b1b
>5a7b34438794aed2c178decee1%7C0%7C0%7C636263246981137373=A%2B2jWktBlR
>H90ZO2QSSffl9UJ62zFppLZuIE2nv9x48%3D=0
>
>Piotr
>
>
>
>-
>Apache Flex PMC
>piotrzarzyck...@gmail.com
>--
>View this message in context:
>https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle
>x-development.247.n4.nabble.com%2FFlexJS-Maven-Local-build-failing-occ
>asionally-tp60859p60862.html=02%7C01%7C%7Cf257976a7a4b48be3e5e08d4760
>d533d%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636263246981147385
>a=OgV8g%2BmLx%2FOlinRHzbDqoneDUOc6Rmp8lFGzjGOgVS0%3D=0
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJX][FalconJX]

[Greg Dove]

Thanks, this was used in a local Log utility which is used for indirect
prefixing with info/warn etc. iirc I could not simply use trace
cross-platform for this approach, I think there were problems with
trace.apply for js (I would need to check, but this is what I recall) so I
ended up referencing console.log directly for js

I changed console.log to window.console.log  and that still gave 'No
GoogDep for window'
I think this use should be rare and trace will normally be used, so I added
'window' to NativeUtils and all was well. So I will commit this change.



On Wed, Mar 29, 2017 at 8:21 AM, Alex Harui  wrote:

>
>
> On 3/28/17, 12:12 PM, "Greg Dove"  wrote:
>
> >
> >FWIW I tried with remove-circulars again but that broke on 'No GoogDep for
> >console' this time, but I guess you are still working on this.
>
> Actually, I would prefer that folks who hit "No GoogDep" fix it themselves
> by adding what they need to NativeUtils.java.  Although for console, I
> think you could also use "window.console".
>
> Thanks,
> -Alex
>
>

Re: Falcon build failing

[Harbs]

Looking at you change, I doubt it’ll fix my problem because I WAS using 
remove-circulars, but I’m trying anyway…

> On Mar 28, 2017, at 6:50 PM, Alex Harui  wrote:
> 
> Hmm.   Looks like I broke something.  Even DataBindingExample didn't run.
> I just pushed a change to fix DataBindingExample and hopefully all other
> apps.  Please pull flex-falcon and try again.
> 
> Sorry about the delay,
> -Alex
> 
> On 3/28/17, 7:13 AM, "Harbs"  wrote:
> 
>> Thanks, but I’m having bigger problems. Hopefully Alex will look into the
>> other issues…
>> 
>>> On Mar 28, 2017, at 4:55 PM, Josh Tynjala  wrote:
>>> 
>>> I have moved my new tests into the Ant integration tests, so they
>>> should no
>>> longer block you.
>>> 
>>> - Josh
>>> 
>>> On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:
>>> 
 It was Josh’s tests that were tripping me up. The build fails after
 doing
 a wipe-all.
 
 I’ve been having other problems with Falcon, and the latest logging
 seems
 to shed some light on what’s wrong:
 
[java] Dependencies calculated for 'org.apache.flex.states.
 SetProperty'
[java] org.apache.flex.states.SetProperty depends on
 org.apache.flex.core.IDocument
[java] No GoogDep for Uint8Array
[java] No GoogDep for Uint32Array
[java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
 getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
 compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
 MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
 clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
 flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
 org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
 MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
 staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
 clients.MXMLJSC.main(MXMLJSC.java:176)
 
 It looks like an app which depends on BinaryData will fail. I’m
 guessing
 that for some reason falcon is trying to add requires for native JS
 APIs.
 
 Harbs
 
> On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
> 
> 
> 
> On 3/26/17, 1:19 PM, "Harbs"  wrote:
> 
>> I’m having trouble building Falcon.
>> 
>> It looks like it’s failing because a text in FlexJSTestBase.java is
>> failing. It’s looking for a file
>> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
> 
> That output has been there for months.  It isn't the cause of a
> failure.
> You have to find the test results and check the output in there.
> 
> -Alex
> 
 
 
>> 
> 

Re: Falcon build failing

[Alex Harui]

There are lots of issues being dealt with so I don't know which one you
are referring to.  I'm still working on the exception in getListOfFiles.

-Alex

On 3/28/17, 9:42 AM, "Harbs"  wrote:

>Looking at you change, I doubt it’ll fix my problem because I WAS using
>remove-circulars, but I’m trying anyway…
>
>> On Mar 28, 2017, at 6:50 PM, Alex Harui  wrote:
>> 
>> Hmm.   Looks like I broke something.  Even DataBindingExample didn't
>>run.
>> I just pushed a change to fix DataBindingExample and hopefully all other
>> apps.  Please pull flex-falcon and try again.
>> 
>> Sorry about the delay,
>> -Alex
>> 
>> On 3/28/17, 7:13 AM, "Harbs"  wrote:
>> 
>>> Thanks, but I’m having bigger problems. Hopefully Alex will look into
>>>the
>>> other issues…
>>> 
 On Mar 28, 2017, at 4:55 PM, Josh Tynjala 
wrote:
 
 I have moved my new tests into the Ant integration tests, so they
 should no
 longer block you.
 
 - Josh
 
 On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:
 
> It was Josh’s tests that were tripping me up. The build fails after
> doing
> a wipe-all.
> 
> I’ve been having other problems with Falcon, and the latest logging
> seems
> to shed some light on what’s wrong:
> 
>[java] Dependencies calculated for 'org.apache.flex.states.
> SetProperty'
>[java] org.apache.flex.states.SetProperty depends on
> org.apache.flex.core.IDocument
>[java] No GoogDep for Uint8Array
>[java] No GoogDep for Uint32Array
>[java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
> clients.MXMLJSC.main(MXMLJSC.java:176)
> 
> It looks like an app which depends on BinaryData will fail. I’m
> guessing
> that for some reason falcon is trying to add requires for native JS
> APIs.
> 
> Harbs
> 
>> On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
>> 
>> 
>> 
>> On 3/26/17, 1:19 PM, "Harbs"  wrote:
>> 
>>> I’m having trouble building Falcon.
>>> 
>>> It looks like it’s failing because a text in FlexJSTestBase.java is
>>> failing. It’s looking for a file
>>> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
>> 
>> That output has been there for months.  It isn't the cause of a
>> failure.
>> You have to find the test results and check the output in there.
>> 
>> -Alex
>> 
> 
> 
>>> 
>> 
>

Re: Falcon build failing

[Alex Harui]

OK.  I think I've fixed the getListOfFiles problem.

The Maven examples build seems to be turning on remove-circulars where the
Ant build doesn't.  That might need to be fixed up, but actually it helps
test the code paths in the dependency management.

-Alex

On 3/28/17, 9:54 AM, "Alex Harui"  wrote:

>There are lots of issues being dealt with so I don't know which one you
>are referring to.  I'm still working on the exception in getListOfFiles.
>
>-Alex
>
>On 3/28/17, 9:42 AM, "Harbs"  wrote:
>
>>Looking at you change, I doubt it’ll fix my problem because I WAS using
>>remove-circulars, but I’m trying anyway…
>>
>>> On Mar 28, 2017, at 6:50 PM, Alex Harui  wrote:
>>> 
>>> Hmm.   Looks like I broke something.  Even DataBindingExample didn't
>>>run.
>>> I just pushed a change to fix DataBindingExample and hopefully all
>>>other
>>> apps.  Please pull flex-falcon and try again.
>>> 
>>> Sorry about the delay,
>>> -Alex
>>> 
>>> On 3/28/17, 7:13 AM, "Harbs"  wrote:
>>> 
 Thanks, but I’m having bigger problems. Hopefully Alex will look into
the
 other issues…
 
> On Mar 28, 2017, at 4:55 PM, Josh Tynjala 
>wrote:
> 
> I have moved my new tests into the Ant integration tests, so they
> should no
> longer block you.
> 
> - Josh
> 
> On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:
> 
>> It was Josh’s tests that were tripping me up. The build fails after
>> doing
>> a wipe-all.
>> 
>> I’ve been having other problems with Falcon, and the latest logging
>> seems
>> to shed some light on what’s wrong:
>> 
>>[java] Dependencies calculated for 'org.apache.flex.states.
>> SetProperty'
>>[java] org.apache.flex.states.SetProperty depends on
>> org.apache.flex.core.IDocument
>>[java] No GoogDep for Uint8Array
>>[java] No GoogDep for Uint32Array
>>[java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
>> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
>> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
>> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
>> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
>> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
>> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
>> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
>> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
>> clients.MXMLJSC.main(MXMLJSC.java:176)
>> 
>> It looks like an app which depends on BinaryData will fail. I’m
>> guessing
>> that for some reason falcon is trying to add requires for native JS
>> APIs.
>> 
>> Harbs
>> 
>>> On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
>>> 
>>> 
>>> 
>>> On 3/26/17, 1:19 PM, "Harbs"  wrote:
>>> 
 I’m having trouble building Falcon.
 
 It looks like it’s failing because a text in FlexJSTestBase.java
is
 failing. It’s looking for a file
 flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
>>> 
>>> That output has been there for months.  It isn't the cause of a
>>> failure.
>>> You have to find the test results and check the output in there.
>>> 
>>> -Alex
>>> 
>> 
>> 
 
>>> 
>>
>

Re: [FlexJX][FalconJX]

[Alex Harui]

On 3/28/17, 12:59 AM, "Greg Dove"  wrote:

>Hi Alex, if you have time, perhaps you can shed some light on this?
>
>I have an mxml component using states, that implements an Interface,
>IFormSequence.
>
>the generated CLASS_INFO looks like this:
>FLEXJS_CLASS_INFO = { names: [{ name: 'ActionForm', qName:
>'components.forms.ActionForm', kind: 'class'  }], interfaces:
>[components.forms.IFormSequence] };
>
>and the goog.require for the interface is there at the top
>goog.require('components.forms.IFormSequence');
>
>I notice that the 
>goog.addDependency('../../../components/forms/IFormSequence.js',
>['components.forms.IFormSequence'], []);
>
>appears after the 
>goog.addDependency('../../../components/forms/ActionForm.js',
>['components.forms.ActionForm'], ['org.apache.flex.html.Group']);
>in the index.html.
>Perhaps the order here is not important and goog resolves everything
>somehow, but I did wonder if the dependency chain in for ActionForm.js
>above should also contain the interface (that's a question, not a
>suggestion - I don't know what is right here). At the moment there is an
>issue where there is an interface check on startup when states or styles
>or something like this are being initialized (it is actually a check for
>a different interface, but because the 'interfaces' array is defined and
>has length in CLASS_INFO it looks there and tries to follow follow the
>inheritance chain on the undefined Interface reference, causing mayhem).
>Because the reference inside the CLASS_INFO evaluates to undefined, I
>know the dependencies are not loading in the right order.

AFAIK, the order of addDependency not important.  The addDependency calls
appear to build a lookup table that doesn't get used until evaluating the
first goog.require.

Without -remove-circulars, any class or interface reference that is
actually used in the output should have a goog.require for it.  In the
compiler, the usedNames list keeps track of which classes and interfaces
are in the output.  Type references do not always end up in the output.
For example, if the only reference in AS is "var foo:IMyInterface" then
IMyInterface doesn't go in the output because we don't generate
type-checking code.

Starting last week, With -remove-circulars, only references via extends,
implements, and static initializers go in the dependency list and retain
goog.requires in the output.  But in this case, the interface should still
be in the list.

Anyway, the list of dependencies was broken until about two commits ago.
So try again and see if it works.
And not handling native JS types in static initializers was causing the
exception in GetListOFFiles until my most recent commit.

So pull down everything and see if the interface appears in the
addDependency list again (without remove-circulars).

HTH,
-Alex

Re: [DISCUSS] Changing our general mode of development?

[OmPrakash Muppirala]

The general approach should be to first merge dev branch into the feature
branch and test if any examples are broken.

Then ask for volunteers to test the feature branch and report issues.

After a reasonable amount of time, the feature branch can ne merged into
dev.

This will not solve all the issues, but major breakages can be contained
this way.

At my work, for feature to dev merges,  we have a Pull Request attached to
a code review and an automated build.  But given that we are open source we
need to rely on self discipline and community testing.

Thanks,
Om


On Mar 28, 2017 4:37 AM, "Carlos Rovira" 
wrote:

+1 to all of this.

For people looking at the other's work it more easy to follow git graphs if
people make it's own branch.
I suscribe all Chris's email and agree with all.

So hope folks embrace the maven way of building and abandone ANT since
there's no benefit at all in that build system.

Thanks Chris for bringing this :)



2017-03-28 12:04 GMT+02:00 Harbs :

> I do think that development should be done almost exclusively on feature
> branches.
>
> If the build status for a feature branch can be verified on the server
> (like you set it up) that’s ideal because it does not require using
> specifically ant or maven locally.
>
> > On Mar 28, 2017, at 12:59 PM, Harbs  wrote:
> >
> > I still have not managed to get maven setup correctly.
> >
> >> On Mar 28, 2017, at 12:46 PM, Christofer Dutz <
> christofer.d...@c-ware.de> wrote:
> >>
> >> Hi Harbs,
> >> (this time replying to the right name ;-) )
> >>
> >> I usually simply make sure I update my repos and do the full maven
> build with tests and examples locally before pushing … I guess this is
> sufficient protection against most problems. In IntelliJ that’s two clicks
> and a cup of coffee or whatever beverage you prefer.
> >>
> >> Chris
> >>
> >>
> >> Am 28.03.17, 11:41 schrieb "Harbs" :
> >>
> >>   +1.
> >>
> >>   I think it’s OK to develop however we might be comfortable on a
> feature branch, but we definitely want an approved procedure which must be
> done before committing to the develop branch.
> >>
> >>   Harbs
> >>
> >>> On Mar 28, 2017, at 12:29 PM, Christofer Dutz <
> christofer.d...@c-ware.de> wrote:
> >>>
> >>> Hi,
> >>>
> >>> For the last months, we have seen a huge increase in people working on
> the FlexJS and people working on first applications using FlexJS. I think
> we should discuss how we can make sure we don’t have interruptions like
the
> current one in the future.
> >>>
> >>> One point that has been causing pain in the past, was that some people
> are using Ant and some are using Maven. Maven is quite a bit more
> restrictive than Ant and it builds a lot more and tests a lot more. Just
as
> an example in contrast to the Ant build the Maven build builds all
Examples
> and it also tests some of them to be runnable in a browser. The Ant build
> only builds the framework and most of the latest problems only pop up if
> you build an application. It has occurred several times that Changed
failed
> the Maven build but didn’t fail the Ant build … just because the Ant build
> doesn’t build everything. We could avoid this problem if people would not
> simply ignore build failures reported by the ASF Jenkins, which is taking
> care of the Maven build. It is currently setup to give feedback within an
> hour or so.
> >>>
> >>> Sometimes the “fix” was to exclude a module in Maven. This usually had
> the side-effect of the RAT plugin failing after that because it now finds
> files without Apache headers. A quick solution to that problem is to
log-in
> to the ASF Jenkins and to click on “wipe workspace” of that build. After
> that this type of problem should go away immediately.
> >>>
> >>> Another point was that sometimes people work together on a larger
> refactoring and check-in stuff to develop in order to share code. We
should
> start using feature branches for this. This has currently not been
> happening at all. I have setup everything that if you create a branch IN
> ALL 3 REPOS with a name “feature/{somename}” (but the same “somename” in
> all three ;-) )  the ASF Jenkins will setup a Job for that which builds
all
> parts in one go and give you immediate feedback on the state of your
> branch. Feature branches that are not “blue” should not be merged back to
> develop.
> >>>
> >>> One last pattern I have encountered was people reporting stuff like:
> “I have been working on X and have almost finished ... I know it will
break
> Y, but I’ll push my changes and fix Y after that” … keep in mind: By
> breaking Y everyone working on FlexJS is forced to stop working so I will
> probably veto every suggestion I encounter on the list that has a similar
> pattern.
> >>>
> >>> FlexJS has matured and we are approaching a 1.0, but we also must
> mature the way we develop or we will hurt early adopters and people
willing
> to help get 

Re: [FlexJS] MDL - Project is failing to compile, help!

[Alex Harui]

Hi Santanu,

Nobody else has reported your problem recently.

Again, I think it is time for you to learn to build the repos and do your
development that way so you are not blocked when things get unstable.
Then you could roll back as far as you need to.

Thanks,
-Alex

On 3/28/17, 9:17 PM, "sankar"  wrote:

>Hi,
>
>I see some other build problem posts came up lately.. I am not sure if the
>problem I described are similar to the problems others faced in those
>posts,
>though. But I'm almost out of business! This latest source really ruin the
>whole development here.
>
>Can I get any archive copy of these whole things maybe back from January?
>
>Thanks!
>
>
>
>--
>View this message in context:
>https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle
>x-development.247.n4.nabble.com%2FFlexJS-MDL-Project-is-failing-to-com
>pile-help-tp60668p60871.html=02%7C01%7C%7C8ab03b159bae40ad76ea08d4765
>bacb3%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636263583490920156
>a=UYilx0JyvJht0ZaCRSotwL8Y%2BU4A8dpUgLUP2%2BeVuzs%3D=0
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: Falcon build failing

[Alex Harui]

Yep, but I just added it about 10 minutes ago.

-Alex

On 3/28/17, 9:50 PM, "Harbs"  wrote:

>I’m getting this with a different project:
>No GoogDep for NaN
>
>I guess NaN needs to be added too?
>
>> On Mar 28, 2017, at 8:49 PM, Harbs  wrote:
>> 
>> Progress!
>> 
>> It looks like the only problem I now have is the missing dependency on
>>org.apache.flex.core.Application
>> 
>>> On Mar 28, 2017, at 8:00 PM, Alex Harui  wrote:
>>> 
>>> OK.  I think I've fixed the getListOfFiles problem.
>>> 
>>> The Maven examples build seems to be turning on remove-circulars where
>>>the
>>> Ant build doesn't.  That might need to be fixed up, but actually it
>>>helps
>>> test the code paths in the dependency management.
>>> 
>>> -Alex
>>> 
>>> On 3/28/17, 9:54 AM, "Alex Harui"  wrote:
>>> 
 There are lots of issues being dealt with so I don't know which one
you
 are referring to.  I'm still working on the exception in
getListOfFiles.
 
 -Alex
 
 On 3/28/17, 9:42 AM, "Harbs"  wrote:
 
> Looking at you change, I doubt it’ll fix my problem because I WAS
>using
> remove-circulars, but I’m trying anyway…
> 
>> On Mar 28, 2017, at 6:50 PM, Alex Harui  wrote:
>> 
>> Hmm.   Looks like I broke something.  Even DataBindingExample didn't
>> run.
>> I just pushed a change to fix DataBindingExample and hopefully all
>> other
>> apps.  Please pull flex-falcon and try again.
>> 
>> Sorry about the delay,
>> -Alex
>> 
>> On 3/28/17, 7:13 AM, "Harbs"  wrote:
>> 
>>> Thanks, but I’m having bigger problems. Hopefully Alex will look
>>>into
>>> the
>>> other issues…
>>> 
 On Mar 28, 2017, at 4:55 PM, Josh Tynjala 
 wrote:
 
 I have moved my new tests into the Ant integration tests, so they
 should no
 longer block you.
 
 - Josh
 
 On Tue, Mar 28, 2017 at 1:21 AM, Harbs 
wrote:
 
> It was Josh’s tests that were tripping me up. The build fails
>after
> doing
> a wipe-all.
> 
> I’ve been having other problems with Falcon, and the latest
>logging
> seems
> to shed some light on what’s wrong:
> 
>  [java] Dependencies calculated for 'org.apache.flex.states.
> SetProperty'
>  [java] org.apache.flex.states.SetProperty depends on
> org.apache.flex.core.IDocument
>  [java] No GoogDep for Uint8Array
>  [java] No GoogDep for Uint32Array
>  [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
> 
>compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
> clients.MXMLJSC.main(MXMLJSC.java:176)
> 
> It looks like an app which depends on BinaryData will fail. I’m
> guessing
> that for some reason falcon is trying to add requires for native
>JS
> APIs.
> 
> Harbs
> 
>> On Mar 27, 2017, at 7:13 AM, Alex Harui 
>>wrote:
>> 
>> 
>> 
>> On 3/26/17, 1:19 PM, "Harbs"  wrote:
>> 
>>> I’m having trouble building Falcon.
>>> 
>>> It looks like it’s failing because a text in
>>>FlexJSTestBase.java
>>> is
>>> failing. It’s looking for a file
>>> flex-asjs/frameworks/as/basic-manifest.xml which does not
>>>exist.
>> 
>> That output has been there for months.  It isn't the cause of a
>> failure.
>> You have to find the test results and check the output in there.
>> 
>> -Alex
>> 
> 
> 
>>> 
>> 
> 
 
>>> 
>> 
>

Re: [FlexJS] Anyone working on the build problems?

[Alex Harui]

I have disabled the ListExample and made a few other fixes and the Maven
build is now building the framework and other examples and is now failing
in the IntegrationTests with a Selenium version mismatch error.  Hopefully
Christofer or someone else can figure that out.  I'm going off to do other
things.

Thanks,
-Alex

On 3/27/17, 6:19 PM, "Josh Tynjala"  wrote:

>Yes, after the Maven build failed, I saw that the other MXML tests were
>excluded, and I excluded my new ones in the same place. The Maven side has
>no problems with my changes anymore. Just gotta tweak the Ant build a bit.
>
>If we find the time to come up with better tests in Miami, that would be
>cool.
>
>I have some ideas for how to run my new tests without the asjs dependency.
>Basically, it involves creating a couple of custom AS/MXML classes to
>reference in MXML instead of real components from the framework. The exact
>classes referenced in MXML don't matter for my tests. It was simply
>convenient to follow the patterns of the existing tests to get it written
>quickly. These custom classes could probably be used to improve some of
>those other tests too.
>
>- Josh
>
>On Mar 27, 2017 1:04 PM, "Christofer Dutz" 
>wrote:
>
>Hi Josh,
>
>I guess in that case it was one of the tests which I excluded from the
>Maven build … so if you add another it will automatically be run, as Maven
>doesn’t know it should exclude it. If you have a look in the pom.xml you
>will see the surefire plugin (runs unit tests) and the failsafe plugin
>(runs integration tests). Here a hand full of tests are excluded because
>they need a refactoring to not introduce a cycle or need to be moved to
>the
>framework module (I would prefer the first option)
>
>Guess it would be cool if we figured out a way to refactor the tests so we
>can run them in Maven … eventually doing so over a beer in Miami would be
>a
>good invest of time? What do you think ;-)
>
>Chris
>
>Am 27.03.17, 17:43 schrieb "Josh Tynjala" :
>
>I didn't structure the tests this way. I followed the existing tests
>that
>were already there.
>
>- Josh
>
>On Mar 27, 2017 1:04 AM, "Christofer Dutz" 
>wrote:
>
>> Hi Josh,
>>
>> Tests shouldn’t require a directory structure … especially not
>outside of
>> the current maven modules directory. This type of refactoring is
>what
>> consumed most of my time while migrating Falcon from Ant to Maven.
>Having a
>> look at the current problem, it seems the problems are related to
>its
>> dependency to FlexJSTestBase, which seems to require a set of
>environment
>> variables and expects other directories. This dependency to
>framework
>would
>> have introduced a cycle, which is also the reason I couldn’t
>refactor
>the
>> tests without completely re-writing most of them.
>>
>> We should address the FlexJsTestBase tests and make sure we are able
>to
>> test what they should be testing without introducing a cyclic
>dependency.
>>
>> For now, excluding them from the testsuite is probably the best
>option.
>>
>> Chris
>>
>>
>> Am 26.03.17, 23:19 schrieb "Josh Tynjala" :
>>
>> 1. I don't think this part is actually an issue (other than
>being
>> annoying
>> noise in the output). I also see this warning and my build
>finishes
>> successfully.
>>
>> 2/3. I think these new tests that I wrote rely on flex-asjs in
>some way
>> (since they're parsing MXML), and maybe it's having trouble
>finding
>> that
>> project on your machine. I'll see if I can move them to the
>correct
>> part of
>> the build where it runs the optional "dependent" tests.
>>
>> - Josh
>>
>> On Sun, Mar 26, 2017 at 1:53 PM, Harbs 
>wrote:
>>
>> > There seemed to be three things causing me build problems:
>> > 1. 
>compiler-jx/src/test/java/org/apache/flex/compiler/internal/
>> test/FlexJSTestBase.java
>> > A reference to frameworks/as/basic-manifest.xml which does not
>seem
>> to
>> > exist.
>> > 2. 
>compiler-jx/src/test/java/org/apache/flex/compiler/internal/
>> > codegen/mxml/sourcemaps/TestSourceMapMXMLScript.java The tests
>are
>> > failing.
>> > 3. 
>compiler-jx/src/test/java/org/apache/flex/compiler/internal/
>> > codegen/mxml/sourcemaps/TestSourceMapMXMLEvents.java The tests
>are
>> > failing.
>> >
>> > Removing the test in 1 and the latter two files allowed Falcon
>and
>> asjs to
>> > build on my machine.
>> >
>> > Harbs
>> >
>> > > On Mar 26, 2017, at 11:26 PM, Harbs 
>wrote:
>> > >
>> > > I’m also having trouble building, but it’s failing at a
>totally
>> > different 

Re: Falcon build failing

[Harbs]

OK. Why is this happening now?

> On Mar 29, 2017, at 7:52 AM, Alex Harui  wrote:
> 
> Yep, but I just added it about 10 minutes ago.
> 
> -Alex
> 
> On 3/28/17, 9:50 PM, "Harbs"  wrote:
> 
>> I’m getting this with a different project:
>> No GoogDep for NaN
>> 
>> I guess NaN needs to be added too?
>> 
>>> On Mar 28, 2017, at 8:49 PM, Harbs  wrote:
>>> 
>>> Progress!
>>> 
>>> It looks like the only problem I now have is the missing dependency on
>>> org.apache.flex.core.Application
>>> 
 On Mar 28, 2017, at 8:00 PM, Alex Harui  wrote:
 
 OK.  I think I've fixed the getListOfFiles problem.
 
 The Maven examples build seems to be turning on remove-circulars where
 the
 Ant build doesn't.  That might need to be fixed up, but actually it
 helps
 test the code paths in the dependency management.
 
 -Alex
 
 On 3/28/17, 9:54 AM, "Alex Harui"  wrote:
 
> There are lots of issues being dealt with so I don't know which one
> you
> are referring to.  I'm still working on the exception in
> getListOfFiles.
> 
> -Alex
> 
> On 3/28/17, 9:42 AM, "Harbs"  wrote:
> 
>> Looking at you change, I doubt it’ll fix my problem because I WAS
>> using
>> remove-circulars, but I’m trying anyway…
>> 
>>> On Mar 28, 2017, at 6:50 PM, Alex Harui  wrote:
>>> 
>>> Hmm.   Looks like I broke something.  Even DataBindingExample didn't
>>> run.
>>> I just pushed a change to fix DataBindingExample and hopefully all
>>> other
>>> apps.  Please pull flex-falcon and try again.
>>> 
>>> Sorry about the delay,
>>> -Alex
>>> 
>>> On 3/28/17, 7:13 AM, "Harbs"  wrote:
>>> 
 Thanks, but I’m having bigger problems. Hopefully Alex will look
 into
 the
 other issues…
 
> On Mar 28, 2017, at 4:55 PM, Josh Tynjala 
> wrote:
> 
> I have moved my new tests into the Ant integration tests, so they
> should no
> longer block you.
> 
> - Josh
> 
> On Tue, Mar 28, 2017 at 1:21 AM, Harbs 
> wrote:
> 
>> It was Josh’s tests that were tripping me up. The build fails
>> after
>> doing
>> a wipe-all.
>> 
>> I’ve been having other problems with Falcon, and the latest
>> logging
>> seems
>> to shed some light on what’s wrong:
>> 
>> [java] Dependencies calculated for 'org.apache.flex.states.
>> SetProperty'
>> [java] org.apache.flex.states.SetProperty depends on
>> org.apache.flex.core.IDocument
>> [java] No GoogDep for Uint8Array
>> [java] No GoogDep for Uint32Array
>> [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
>> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
>> 
>> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
>> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
>> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
>> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
>> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
>> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
>> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
>> clients.MXMLJSC.main(MXMLJSC.java:176)
>> 
>> It looks like an app which depends on BinaryData will fail. I’m
>> guessing
>> that for some reason falcon is trying to add requires for native
>> JS
>> APIs.
>> 
>> Harbs
>> 
>>> On Mar 27, 2017, at 7:13 AM, Alex Harui 
>>> wrote:
>>> 
>>> 
>>> 
>>> On 3/26/17, 1:19 PM, "Harbs"  wrote:
>>> 
 I’m having trouble building Falcon.
 
 It looks like it’s failing because a text in
 FlexJSTestBase.java
 is
 failing. It’s looking for a file
 flex-asjs/frameworks/as/basic-manifest.xml which does not
 exist.
>>> 
>>> That output has been there for months.  It isn't the cause of a
>>> failure.
>>> You have to find the test results and check the output in there.
>>> 
>>> -Alex
>>> 
>> 
>> 
 
>>> 
>> 
> 
 
>>> 
>> 
> 

Re: Falcon build failing

[Harbs]

Can I assume that any types in external swcs will not have this issue?

> On Mar 29, 2017, at 8:11 AM, Alex Harui  wrote:
> 
> 
> 
> On 3/28/17, 9:58 PM, "Harbs"  wrote:
> 
>> OK. Why is this happening now?
> 
> In order to handle the reordering of dependencies, I needed to collect
> types used in static initializers.  The code is throwing a big net over
> all sorts of identifiers.  I suppose I need to make it smarter, not sure.
> 
> But it could just be that NativeUtils should contain all of these JS
> "types".  Although I'm a bit surprised to see NaN in there.
> 
> -Alex
> 

Re: Falcon build failing

[Alex Harui]

On 3/28/17, 10:40 PM, "Harbs"  wrote:

>Can I assume that any types in external swcs will not have this issue?

Don't know.  I don't think I've had to add anything like that so far.  If
you are running into that, then I need to adjust the logic.

-Alex

>
>> On Mar 29, 2017, at 8:11 AM, Alex Harui  wrote:
>> 
>> 
>> 
>> On 3/28/17, 9:58 PM, "Harbs"  wrote:
>> 
>>> OK. Why is this happening now?
>> 
>> In order to handle the reordering of dependencies, I needed to collect
>> types used in static initializers.  The code is throwing a big net over
>> all sorts of identifiers.  I suppose I need to make it smarter, not
>>sure.
>> 
>> But it could just be that NativeUtils should contain all of these JS
>> "types".  Although I'm a bit surprised to see NaN in there.
>> 
>> -Alex
>> 
>

Re: Falcon build failing

[Harbs]

I’m getting this with a different project:
No GoogDep for NaN

I guess NaN needs to be added too?

> On Mar 28, 2017, at 8:49 PM, Harbs  wrote:
> 
> Progress!
> 
> It looks like the only problem I now have is the missing dependency on 
> org.apache.flex.core.Application
> 
>> On Mar 28, 2017, at 8:00 PM, Alex Harui  wrote:
>> 
>> OK.  I think I've fixed the getListOfFiles problem.
>> 
>> The Maven examples build seems to be turning on remove-circulars where the
>> Ant build doesn't.  That might need to be fixed up, but actually it helps
>> test the code paths in the dependency management.
>> 
>> -Alex
>> 
>> On 3/28/17, 9:54 AM, "Alex Harui"  wrote:
>> 
>>> There are lots of issues being dealt with so I don't know which one you
>>> are referring to.  I'm still working on the exception in getListOfFiles.
>>> 
>>> -Alex
>>> 
>>> On 3/28/17, 9:42 AM, "Harbs"  wrote:
>>> 
 Looking at you change, I doubt it’ll fix my problem because I WAS using
 remove-circulars, but I’m trying anyway…
 
> On Mar 28, 2017, at 6:50 PM, Alex Harui  wrote:
> 
> Hmm.   Looks like I broke something.  Even DataBindingExample didn't
> run.
> I just pushed a change to fix DataBindingExample and hopefully all
> other
> apps.  Please pull flex-falcon and try again.
> 
> Sorry about the delay,
> -Alex
> 
> On 3/28/17, 7:13 AM, "Harbs"  wrote:
> 
>> Thanks, but I’m having bigger problems. Hopefully Alex will look into
>> the
>> other issues…
>> 
>>> On Mar 28, 2017, at 4:55 PM, Josh Tynjala 
>>> wrote:
>>> 
>>> I have moved my new tests into the Ant integration tests, so they
>>> should no
>>> longer block you.
>>> 
>>> - Josh
>>> 
>>> On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:
>>> 
 It was Josh’s tests that were tripping me up. The build fails after
 doing
 a wipe-all.
 
 I’ve been having other problems with Falcon, and the latest logging
 seems
 to shed some light on what’s wrong:
 
  [java] Dependencies calculated for 'org.apache.flex.states.
 SetProperty'
  [java] org.apache.flex.states.SetProperty depends on
 org.apache.flex.core.IDocument
  [java] No GoogDep for Uint8Array
  [java] No GoogDep for Uint32Array
  [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
 getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
 compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
 MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
 clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
 flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
 org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
 MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
 staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
 clients.MXMLJSC.main(MXMLJSC.java:176)
 
 It looks like an app which depends on BinaryData will fail. I’m
 guessing
 that for some reason falcon is trying to add requires for native JS
 APIs.
 
 Harbs
 
> On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
> 
> 
> 
> On 3/26/17, 1:19 PM, "Harbs"  wrote:
> 
>> I’m having trouble building Falcon.
>> 
>> It looks like it’s failing because a text in FlexJSTestBase.java
>> is
>> failing. It’s looking for a file
>> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
> 
> That output has been there for months.  It isn't the cause of a
> failure.
> You have to find the test results and check the output in there.
> 
> -Alex
> 
 
 
>> 
> 
 
>>> 
>> 
> 

Re: Falcon build failing

[Alex Harui]

On 3/28/17, 9:58 PM, "Harbs"  wrote:

>OK. Why is this happening now?

In order to handle the reordering of dependencies, I needed to collect
types used in static initializers.  The code is throwing a big net over
all sorts of identifiers.  I suppose I need to make it smarter, not sure.

But it could just be that NativeUtils should contain all of these JS
"types".  Although I'm a bit surprised to see NaN in there.

-Alex

Re: git commit: [flex-asjs] [refs/heads/develop] - disable ListExample for now

[Piotr Zarzycki]

Alex,

I just tried to fix ListExample and I was able to get it build, but for
some reason I have report in JS that something is missing. [1] I will be
looking into more.

[1] https://paste.apache.org/wWrN

Piotr

2017-03-29 6:30 GMT+02:00 :

> Repository: flex-asjs
> Updated Branches:
>   refs/heads/develop 16ee6cdbc -> 3dc3eb3fb
>
>
> disable ListExample for now
>
>
> Project: http://git-wip-us.apache.org/repos/asf/flex-asjs/repo
> Commit: http://git-wip-us.apache.org/repos/asf/flex-asjs/commit/3dc3eb3f
> Tree: http://git-wip-us.apache.org/repos/asf/flex-asjs/tree/3dc3eb3f
> Diff: http://git-wip-us.apache.org/repos/asf/flex-asjs/diff/3dc3eb3f
>
> Branch: refs/heads/develop
> Commit: 3dc3eb3fb581972b180bcbf68fe1141414bc7e47
> Parents: 16ee6cd
> Author: Alex Harui 
> Authored: Tue Mar 28 21:30:35 2017 -0700
> Committer: Alex Harui 
> Committed: Tue Mar 28 21:30:35 2017 -0700
>
> --
>  examples/flexjs/pom.xml | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> --
>
>
> http://git-wip-us.apache.org/repos/asf/flex-asjs/blob/
> 3dc3eb3f/examples/flexjs/pom.xml
> --
> diff --git a/examples/flexjs/pom.xml b/examples/flexjs/pom.xml
> index 0831f2f..d49f223 100644
> --- a/examples/flexjs/pom.xml
> +++ b/examples/flexjs/pom.xml
> @@ -50,7 +50,7 @@
>  FlexTeamPage_MDL
>  FlexWebsiteStatsViewer
>  HelloWorld
> -ListExample
> +
>  MapSearch
>  
>  MDLDynamicTabsExample
>
>


-- 

Greetings
Piotr Zarzycki

Flex/AIR/.NET Developer

mobile: +48 880 859 557
e-mail: piotrzarzyck...@gmail.com
skype: zarzycki10

LinkedIn: http://www.linkedin.com/piotrzarzycki

Re: [FlexJS] MDL - Project is failing to compile, help!

[sankar]

Hi,

I see some other build problem posts came up lately.. I am not sure if the
problem I described are similar to the problems others faced in those posts,
though. But I'm almost out of business! This latest source really ruin the
whole development here. 

Can I get any archive copy of these whole things maybe back from January?

Thanks!



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-MDL-Project-is-failing-to-compile-help-tp60668p60871.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: Falcon build failing

[Harbs]

Progress!

It looks like the only problem I now have is the missing dependency on 
org.apache.flex.core.Application

> On Mar 28, 2017, at 8:00 PM, Alex Harui  wrote:
> 
> OK.  I think I've fixed the getListOfFiles problem.
> 
> The Maven examples build seems to be turning on remove-circulars where the
> Ant build doesn't.  That might need to be fixed up, but actually it helps
> test the code paths in the dependency management.
> 
> -Alex
> 
> On 3/28/17, 9:54 AM, "Alex Harui"  wrote:
> 
>> There are lots of issues being dealt with so I don't know which one you
>> are referring to.  I'm still working on the exception in getListOfFiles.
>> 
>> -Alex
>> 
>> On 3/28/17, 9:42 AM, "Harbs"  wrote:
>> 
>>> Looking at you change, I doubt it’ll fix my problem because I WAS using
>>> remove-circulars, but I’m trying anyway…
>>> 
 On Mar 28, 2017, at 6:50 PM, Alex Harui  wrote:
 
 Hmm.   Looks like I broke something.  Even DataBindingExample didn't
 run.
 I just pushed a change to fix DataBindingExample and hopefully all
 other
 apps.  Please pull flex-falcon and try again.
 
 Sorry about the delay,
 -Alex
 
 On 3/28/17, 7:13 AM, "Harbs"  wrote:
 
> Thanks, but I’m having bigger problems. Hopefully Alex will look into
> the
> other issues…
> 
>> On Mar 28, 2017, at 4:55 PM, Josh Tynjala 
>> wrote:
>> 
>> I have moved my new tests into the Ant integration tests, so they
>> should no
>> longer block you.
>> 
>> - Josh
>> 
>> On Tue, Mar 28, 2017 at 1:21 AM, Harbs  wrote:
>> 
>>> It was Josh’s tests that were tripping me up. The build fails after
>>> doing
>>> a wipe-all.
>>> 
>>> I’ve been having other problems with Falcon, and the latest logging
>>> seems
>>> to shed some light on what’s wrong:
>>> 
>>>   [java] Dependencies calculated for 'org.apache.flex.states.
>>> SetProperty'
>>>   [java] org.apache.flex.states.SetProperty depends on
>>> org.apache.flex.core.IDocument
>>>   [java] No GoogDep for Uint8Array
>>>   [java] No GoogDep for Uint32Array
>>>   [java] org.apache.flex.compiler.internal.graph.GoogDepsWriter.
>>> getListOfFiles(GoogDepsWriter.java:103)org.apache.flex.
>>> compiler.internal.codegen.mxml.flexjs.MXMLFlexJSPublisher.publish(
>>> MXMLFlexJSPublisher.java:311)org.apache.flex.compiler.
>>> clients.MXMLJSC.compile(MXMLJSC.java:455)org.apache.
>>> flex.compiler.clients.MXMLJSC._mainNoExit(MXMLJSC.java:313)
>>> org.apache.flex.compiler.clients.MXMLJSC.mainNoExit(
>>> MXMLJSC.java:270)org.apache.flex.compiler.clients.MXMLJSC.
>>> staticMainNoExit(MXMLJSC.java:232)org.apache.flex.compiler.
>>> clients.MXMLJSC.main(MXMLJSC.java:176)
>>> 
>>> It looks like an app which depends on BinaryData will fail. I’m
>>> guessing
>>> that for some reason falcon is trying to add requires for native JS
>>> APIs.
>>> 
>>> Harbs
>>> 
 On Mar 27, 2017, at 7:13 AM, Alex Harui  wrote:
 
 
 
 On 3/26/17, 1:19 PM, "Harbs"  wrote:
 
> I’m having trouble building Falcon.
> 
> It looks like it’s failing because a text in FlexJSTestBase.java
> is
> failing. It’s looking for a file
> flex-asjs/frameworks/as/basic-manifest.xml which does not exist.
 
 That output has been there for months.  It isn't the cause of a
 failure.
 You have to find the test results and check the output in there.
 
 -Alex
 
>>> 
>>> 
> 
 
>>> 
>> 
> 

Re: git commit: [flex-falcon] [refs/heads/develop] - Try to fix getListOfFiles exception. If you get this error and the log says 'No GoogDep for xxx' and 'xxx' is a native Javascript type, add it to N

[Harbs]

Do we need to also add Map and WeakMap, etc.?

> On Mar 28, 2017, at 7:59 PM, aha...@apache.org wrote:
> 
> Repository: flex-falcon
> Updated Branches:
>  refs/heads/develop fce80f5f4 -> 639c5f2c6
> 
> 
> Try to fix getListOfFiles exception.  If you get this error and the log says 
> 'No GoogDep for xxx' and 'xxx' is a native Javascript type, add it to 
> NativeUtils.java
> 
> 
> Project: http://git-wip-us.apache.org/repos/asf/flex-falcon/repo
> Commit: http://git-wip-us.apache.org/repos/asf/flex-falcon/commit/639c5f2c
> Tree: http://git-wip-us.apache.org/repos/asf/flex-falcon/tree/639c5f2c
> Diff: http://git-wip-us.apache.org/repos/asf/flex-falcon/diff/639c5f2c
> 
> Branch: refs/heads/develop
> Commit: 639c5f2c612864461c8a02ed43a14c285daa11ae
> Parents: fce80f5
> Author: Alex Harui 
> Authored: Tue Mar 28 09:59:02 2017 -0700
> Committer: Alex Harui 
> Committed: Tue Mar 28 09:59:02 2017 -0700
> 
> --
> .../java/org/apache/flex/compiler/utils/NativeUtils.java  | 10 ++
> 1 file changed, 10 insertions(+)
> --
> 
> 
> http://git-wip-us.apache.org/repos/asf/flex-falcon/blob/639c5f2c/compiler-jx/src/main/java/org/apache/flex/compiler/utils/NativeUtils.java
> --
> diff --git 
> a/compiler-jx/src/main/java/org/apache/flex/compiler/utils/NativeUtils.java 
> b/compiler-jx/src/main/java/org/apache/flex/compiler/utils/NativeUtils.java
> index 7245a17..b2396c1 100644
> --- 
> a/compiler-jx/src/main/java/org/apache/flex/compiler/utils/NativeUtils.java
> +++ 
> b/compiler-jx/src/main/java/org/apache/flex/compiler/utils/NativeUtils.java
> @@ -90,9 +90,19 @@ public class NativeUtils
> 
> Event("Event"),
> Array("Array"),
> +Float32Array("Float32Array"),
> +Float64Array("Float64Array"),
> +Int8Array("Int8Array"),
> +Int16Array("Int16Array"),
> +Int32Array("Int32Array"),
> +Uint8Array("Uint8Array"),
> +Uint8ClampedArray("Uint8ClampedArray"),
> +Uint16Array("Uint16Array"),
> +Uint32Array("Uint32Array"),
> Date("Date"),
> Math("Math"),
> Error("Error"),
> +RangeError("RangeError"),
> Boolean("Boolean"),
> decodeURI("decodeURI"),
> decodeURIComponent("decodeURIComponent"),
> 

[FlexJS, Maven] Local build failing occasionally

[piotrz]

Hi All,

I'm struggling with FlexJS Maven build "failing". It's about 80% for me.
Errors which I'm getting look like related with build of Graphics - once I
get this stacktrace [1] in the other modules I'm getting warnings and
errors, [2] even that overall compilation indicates that everythings has
been build ok. 

I didn't report it earlier cause it occurs occasionally.

Any thoughts about that ?

[1] https://paste.apache.org/Qx6e  
[2] https://paste.apache.org/k0cG

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-Local-build-failing-occasionally-tp60859.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [DISCUSS] Changing our general mode of development?

[Peter Ent]

+1 to this. 

Having pushed my changes to develop and causing a bit of mayhem, using a
feature branch would have been the right thing to do. I'm currently trying
to get the Charts package to work which will entail a little bit more
change to HTML project, and I will put this into a features branch so
everyone can try it out first before it gets merged into develop.

We need to clean FlexJS up - I believe there are lots of unused files and
the HTML project is way too big. So keep that in mind for the near future
if you want do that.

Regards,
Peter Ent

On 3/28/17, 12:49 PM, "omup...@gmail.com on behalf of OmPrakash Muppirala"
 wrote:

>The general approach should be to first merge dev branch into the feature
>branch and test if any examples are broken.
>
>Then ask for volunteers to test the feature branch and report issues.
>
>After a reasonable amount of time, the feature branch can ne merged into
>dev.
>
>This will not solve all the issues, but major breakages can be contained
>this way.
>
>At my work, for feature to dev merges,  we have a Pull Request attached to
>a code review and an automated build.  But given that we are open source
>we
>need to rely on self discipline and community testing.
>
>Thanks,
>Om
>
>
>On Mar 28, 2017 4:37 AM, "Carlos Rovira" 
>wrote:
>
>+1 to all of this.
>
>For people looking at the other's work it more easy to follow git graphs
>if
>people make it's own branch.
>I suscribe all Chris's email and agree with all.
>
>So hope folks embrace the maven way of building and abandone ANT since
>there's no benefit at all in that build system.
>
>Thanks Chris for bringing this :)
>
>
>
>2017-03-28 12:04 GMT+02:00 Harbs :
>
>> I do think that development should be done almost exclusively on feature
>> branches.
>>
>> If the build status for a feature branch can be verified on the server
>> (like you set it up) that¹s ideal because it does not require using
>> specifically ant or maven locally.
>>
>> > On Mar 28, 2017, at 12:59 PM, Harbs  wrote:
>> >
>> > I still have not managed to get maven setup correctly.
>> >
>> >> On Mar 28, 2017, at 12:46 PM, Christofer Dutz <
>> christofer.d...@c-ware.de> wrote:
>> >>
>> >> Hi Harbs,
>> >> (this time replying to the right name ;-) )
>> >>
>> >> I usually simply make sure I update my repos and do the full maven
>> build with tests and examples locally before pushing Š I guess this is
>> sufficient protection against most problems. In IntelliJ that¹s two
>>clicks
>> and a cup of coffee or whatever beverage you prefer.
>> >>
>> >> Chris
>> >>
>> >>
>> >> Am 28.03.17, 11:41 schrieb "Harbs" :
>> >>
>> >>   +1.
>> >>
>> >>   I think it¹s OK to develop however we might be comfortable on a
>> feature branch, but we definitely want an approved procedure which must
>>be
>> done before committing to the develop branch.
>> >>
>> >>   Harbs
>> >>
>> >>> On Mar 28, 2017, at 12:29 PM, Christofer Dutz <
>> christofer.d...@c-ware.de> wrote:
>> >>>
>> >>> Hi,
>> >>>
>> >>> For the last months, we have seen a huge increase in people working
>>on
>> the FlexJS and people working on first applications using FlexJS. I
>>think
>> we should discuss how we can make sure we don¹t have interruptions like
>the
>> current one in the future.
>> >>>
>> >>> One point that has been causing pain in the past, was that some
>>people
>> are using Ant and some are using Maven. Maven is quite a bit more
>> restrictive than Ant and it builds a lot more and tests a lot more. Just
>as
>> an example in contrast to the Ant build the Maven build builds all
>Examples
>> and it also tests some of them to be runnable in a browser. The Ant
>>build
>> only builds the framework and most of the latest problems only pop up if
>> you build an application. It has occurred several times that Changed
>failed
>> the Maven build but didn¹t fail the Ant build Š just because the Ant
>>build
>> doesn¹t build everything. We could avoid this problem if people would
>>not
>> simply ignore build failures reported by the ASF Jenkins, which is
>>taking
>> care of the Maven build. It is currently setup to give feedback within
>>an
>> hour or so.
>> >>>
>> >>> Sometimes the ³fix² was to exclude a module in Maven. This usually
>>had
>> the side-effect of the RAT plugin failing after that because it now
>>finds
>> files without Apache headers. A quick solution to that problem is to
>log-in
>> to the ASF Jenkins and to click on ³wipe workspace² of that build. After
>> that this type of problem should go away immediately.
>> >>>
>> >>> Another point was that sometimes people work together on a larger
>> refactoring and check-in stuff to develop in order to share code. We
>should
>> start using feature branches for this. This has currently not been
>> happening at all. I have setup everything that if you create a branch IN
>> ALL 3 REPOS with a name 

Re: [FlexJS, Maven] Local build failing occasionally

[Alex Harui]

Hi Piotr,

These links don't seem to be to full logs.  Both seem to indicate a
problem earlier in the log.  Please check the full log for errors.  I
think sometimes the compiler doesn't get Maven to stop.

-Alex

On 3/28/17, 11:35 AM, "piotrz"  wrote:

>Hi All,
>
>I'm struggling with FlexJS Maven build "failing". It's about 80% for me.
>Errors which I'm getting look like related with build of Graphics - once I
>get this stacktrace [1] in the other modules I'm getting warnings and
>errors, [2] even that overall compilation indicates that everythings has
>been build ok. 
>
>I didn't report it earlier cause it occurs occasionally.
>
>Any thoughts about that ?
>
>[1] 
>https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpaste.apa
>che.org%2FQx6e=02%7C01%7C%7C5baca9a4441e437f847308d4760a4efe%7Cfa7b1b
>5a7b34438794aed2c178decee1%7C0%7C0%7C636263234023032243=FMAaqxj5yafx
>3%2FVrMFIQKB5GshXy79JSj8Zzi28%2FKJg%3D=0
>[2] 
>https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpaste.apa
>che.org%2Fk0cG=02%7C01%7C%7C5baca9a4441e437f847308d4760a4efe%7Cfa7b1b
>5a7b34438794aed2c178decee1%7C0%7C0%7C636263234023032243=zSSlalTZzm7W
>R0iD%2BdDjobwQVK0SbE16KMeMHRpWTMk%3D=0
>
>Piotr
>
>
>
>-
>Apache Flex PMC
>piotrzarzyck...@gmail.com
>--
>View this message in context:
>https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle
>x-development.247.n4.nabble.com%2FFlexJS-Maven-Local-build-failing-occ
>asionally-tp60859.html=02%7C01%7C%7C5baca9a4441e437f847308d4760a4efe%
>7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636263234023032243=4rm%
>2F%2BNvP9XSYvK9n46RvHLBg0Vu20Y1MoIJbq5Okk6Q%3D=0
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: Falcon build failing

[piotrz]

Hi,

I think I have same problem but with TourJS and Maven build. [1] :)

[1] https://paste.apache.org/JWPJ

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/Falcon-build-failing-tp60790p60861.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS, Maven] Local build failing occasionally

[piotrz]

Hi Alex,

Pasting full log [1]. I see earlier some warning, but not sure whether they
can cause such fail. I'm going through this logs one by one.

1805 bytes written to
D:\flex_sdk\Sources\flex-asjs\frameworks\projects\XML\target\XML-0.8.0-SNAPSHOT.swc
in 2.009 seconds
C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airglobal-20.0.swc
Warning: The definition XML depended on by __AS3__.vec.Vector$object in the
SWC
C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airglobal-20.0.swc
could not be found


C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airglobal-20.0.swc
Warning: The definition XMLList depended on by __AS3__.vec.Vector$object in
the SWC
C:\Users\piotr\.m2\repository\com\adobe\air\framework\airglobal\20.0\airglobal-20.0.swc
could not be found

[1] https://paste.apache.org/ItmA

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-Local-build-failing-occasionally-tp60859p60862.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [DISCUSS] Changing our general mode of development?

[Alex Harui]

Hi Christofer, 

Thanks for bringing up this topic.  Yes, we should be working in branches
more.

To me, the problem is in getting folks to try your branch.  The "dual"
branch has been out there for weeks now.  How many of you have tried it?
Right now, it seems like the only way to get people to try it is to push
it to develop.  Unfortunately, we don't have enough tests to prove that
someone won't get broken.  We don't want to discourage contributions by
making the process take too long, but we don't want to delay everyone else
by breaking stuff.  Apache projects aren't a business where a manager can
direct people to use the branch right away.  We want a volunteer with a
spare evening to be able to propose a change and not have to wait days for
approval because we're all busy doing other things.  We can have 3 folks
try the branch and like it and a 4th who just doesn't have time to test it
before it gets merged ends up getting broken.  IOW, we have to come up
with a process that compensates for asynchronicity and volunteerism.


So while we should be using branches more, maybe we also have to get
better at defending ourselves against breakage.  I think there are some
common scenarios.

1) Assuming you are working in a local branch, one common breakage is when
you merge back to develop locally and find that your changes no longer
work due to some other changes in the same repo.
2) Another common breakage is when you are working in flex-asjs and a
change to flex-falcon breaks flex-asjs.  This happened recently.  IMO, in
this scenario you can detach the head of flex-falcon using "git checkout
".
3) Are most folks working from the repos or are folks relying on nightly
builds?  Maybe we need to archive nightlies so you can roll back to them
if needed.

What else can we do to make it easier to use a different branch, and/or
get unblocked when something breaks?

Thoughts?
-Alex

On 3/28/17, 11:13 AM, "Peter Ent"  wrote:

>+1 to this. 
>
>Having pushed my changes to develop and causing a bit of mayhem, using a
>feature branch would have been the right thing to do. I'm currently trying
>to get the Charts package to work which will entail a little bit more
>change to HTML project, and I will put this into a features branch so
>everyone can try it out first before it gets merged into develop.
>
>We need to clean FlexJS up - I believe there are lots of unused files and
>the HTML project is way too big. So keep that in mind for the near future
>if you want do that.
>
>Regards,
>Peter Ent
>
>On 3/28/17, 12:49 PM, "omup...@gmail.com on behalf of OmPrakash Muppirala"
> wrote:
>
>>The general approach should be to first merge dev branch into the feature
>>branch and test if any examples are broken.
>>
>>Then ask for volunteers to test the feature branch and report issues.
>>
>>After a reasonable amount of time, the feature branch can ne merged into
>>dev.
>>
>>This will not solve all the issues, but major breakages can be contained
>>this way.
>>
>>At my work, for feature to dev merges,  we have a Pull Request attached
>>to
>>a code review and an automated build.  But given that we are open source
>>we
>>need to rely on self discipline and community testing.
>>
>>Thanks,
>>Om
>>
>>
>>On Mar 28, 2017 4:37 AM, "Carlos Rovira" 
>>wrote:
>>
>>+1 to all of this.
>>
>>For people looking at the other's work it more easy to follow git graphs
>>if
>>people make it's own branch.
>>I suscribe all Chris's email and agree with all.
>>
>>So hope folks embrace the maven way of building and abandone ANT since
>>there's no benefit at all in that build system.
>>
>>Thanks Chris for bringing this :)
>>
>>
>>
>>2017-03-28 12:04 GMT+02:00 Harbs :
>>
>>> I do think that development should be done almost exclusively on
>>>feature
>>> branches.
>>>
>>> If the build status for a feature branch can be verified on the server
>>> (like you set it up) that¹s ideal because it does not require using
>>> specifically ant or maven locally.
>>>
>>> > On Mar 28, 2017, at 12:59 PM, Harbs  wrote:
>>> >
>>> > I still have not managed to get maven setup correctly.
>>> >
>>> >> On Mar 28, 2017, at 12:46 PM, Christofer Dutz <
>>> christofer.d...@c-ware.de> wrote:
>>> >>
>>> >> Hi Harbs,
>>> >> (this time replying to the right name ;-) )
>>> >>
>>> >> I usually simply make sure I update my repos and do the full maven
>>> build with tests and examples locally before pushing Š I guess this is
>>> sufficient protection against most problems. In IntelliJ that¹s two
>>>clicks
>>> and a cup of coffee or whatever beverage you prefer.
>>> >>
>>> >> Chris
>>> >>
>>> >>
>>> >> Am 28.03.17, 11:41 schrieb "Harbs" :
>>> >>
>>> >>   +1.
>>> >>
>>> >>   I think it¹s OK to develop however we might be comfortable on a
>>> feature branch, but we definitely want an approved procedure which must
>>>be
>>> done before 

Re: [FlexJX][FalconJX]

[Greg Dove]

Thanks Alex,

I confirm this fixed things with the problem I described. I rebuilt Falcon
using maven, and after that, compiling without remove-circulars does not
have the issue I experienced prior.

It was important to clean the local project first (I only mention this
because I often forget to do this myself and others reading this may need
to remember to do the same).

FWIW I tried with remove-circulars again but that broke on 'No GoogDep for
console' this time, but I guess you are still working on this.

BTW I expect to be able to be more active on the project in the coming
weeks, but I won't be able to do very much specifically within the next
week or two because of looming client deadlines.
I will add tickets as I notice things (I can see some additional fixes in
the reflection area for instance, just have not investigated enough yet to
articulate them)


On Wed, Mar 29, 2017 at 6:13 AM, Alex Harui  wrote:

>
>
> On 3/28/17, 12:59 AM, "Greg Dove"  wrote:
>
> >Hi Alex, if you have time, perhaps you can shed some light on this?
> >
> >I have an mxml component using states, that implements an Interface,
> >IFormSequence.
> >
> >the generated CLASS_INFO looks like this:
> >FLEXJS_CLASS_INFO = { names: [{ name: 'ActionForm', qName:
> >'components.forms.ActionForm', kind: 'class'  }], interfaces:
> >[components.forms.IFormSequence] };
> >
> >and the goog.require for the interface is there at the top
> >goog.require('components.forms.IFormSequence');
> >
> >I notice that the
> >goog.addDependency('../../../components/forms/IFormSequence.js',
> >['components.forms.IFormSequence'], []);
> >
> >appears after the
> >goog.addDependency('../../../components/forms/ActionForm.js',
> >['components.forms.ActionForm'], ['org.apache.flex.html.Group']);
> >in the index.html.
> >Perhaps the order here is not important and goog resolves everything
> >somehow, but I did wonder if the dependency chain in for ActionForm.js
> >above should also contain the interface (that's a question, not a
> >suggestion - I don't know what is right here). At the moment there is an
> >issue where there is an interface check on startup when states or styles
> >or something like this are being initialized (it is actually a check for
> >a different interface, but because the 'interfaces' array is defined and
> >has length in CLASS_INFO it looks there and tries to follow follow the
> >inheritance chain on the undefined Interface reference, causing mayhem).
> >Because the reference inside the CLASS_INFO evaluates to undefined, I
> >know the dependencies are not loading in the right order.
>
> AFAIK, the order of addDependency not important.  The addDependency calls
> appear to build a lookup table that doesn't get used until evaluating the
> first goog.require.
>
> Without -remove-circulars, any class or interface reference that is
> actually used in the output should have a goog.require for it.  In the
> compiler, the usedNames list keeps track of which classes and interfaces
> are in the output.  Type references do not always end up in the output.
> For example, if the only reference in AS is "var foo:IMyInterface" then
> IMyInterface doesn't go in the output because we don't generate
> type-checking code.
>
> Starting last week, With -remove-circulars, only references via extends,
> implements, and static initializers go in the dependency list and retain
> goog.requires in the output.  But in this case, the interface should still
> be in the list.
>
> Anyway, the list of dependencies was broken until about two commits ago.
> So try again and see if it works.
> And not handling native JS types in static initializers was causing the
> exception in GetListOFFiles until my most recent commit.
>
> So pull down everything and see if the interface appears in the
> addDependency list again (without remove-circulars).
>
> HTH,
> -Alex
>
>

Re: [FlexJX][FalconJX]

[Alex Harui]

On 3/28/17, 12:12 PM, "Greg Dove"  wrote:

>
>FWIW I tried with remove-circulars again but that broke on 'No GoogDep for
>console' this time, but I guess you are still working on this.

Actually, I would prefer that folks who hit "No GoogDep" fix it themselves
by adding what they need to NativeUtils.java.  Although for console, I
think you could also use "window.console".

Thanks,
-Alex

Re: [FlexJX][FalconJX]

[Alex Harui]

Sounds good.  Thanks.

-Alex

On 3/28/17, 12:58 PM, "Greg Dove"  wrote:

>Thanks, this was used in a local Log utility which is used for indirect
>prefixing with info/warn etc. iirc I could not simply use trace
>cross-platform for this approach, I think there were problems with
>trace.apply for js (I would need to check, but this is what I recall) so I
>ended up referencing console.log directly for js
>
>I changed console.log to window.console.log  and that still gave 'No
>GoogDep for window'
>I think this use should be rare and trace will normally be used, so I
>added
>'window' to NativeUtils and all was well. So I will commit this change.
>
>
>
>On Wed, Mar 29, 2017 at 8:21 AM, Alex Harui  wrote:
>
>>
>>
>> On 3/28/17, 12:12 PM, "Greg Dove"  wrote:
>>
>> >
>> >FWIW I tried with remove-circulars again but that broke on 'No GoogDep
>>for
>> >console' this time, but I guess you are still working on this.
>>
>> Actually, I would prefer that folks who hit "No GoogDep" fix it
>>themselves
>> by adding what they need to NativeUtils.java.  Although for console, I
>> think you could also use "window.console".
>>
>> Thanks,
>> -Alex
>>
>>

Re: [FlexJS, Maven] Local build failing occasionally

[Alex Harui]

I believe that Maven runs several COMPCs in parallel, so you will see
mixed output which can be hard to understand.

My guess is that the PBJ files that are downloaded with the new
PixelBender dependency are seen by Maven as a source file to build.  I
think they should be skipped.

-Alex

On 3/28/17, 1:16 PM, "piotrz"  wrote:

>I was going farther by that, but I don't really understand why errors in
>this
>logs, are around XML module not in Graphics.
>
>Graphics need ByteArray. Maybe Harbs or Yishay shed some light on this
>one.
>
>Piotr
>
>
>
>-
>Apache Flex PMC
>piotrzarzyck...@gmail.com
>--
>View this message in context:
>https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle
>x-development.247.n4.nabble.com%2FFlexJS-Maven-Local-build-failing-occ
>asionally-tp60859p60868.html=02%7C01%7C%7C6d49ba668358446cf03808d4761
>87b9c%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636263294906585664
>a=C8JWkY01%2FLmhIzg5jny1dYKaXKcfPLLkEphsLhE9%2B4U%3D=0
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS, Maven] Local build failing occasionally

[piotrz]

I was going farther by that, but I don't really understand why errors in this
logs, are around XML module not in Graphics. 

Graphics need ByteArray. Maybe Harbs or Yishay shed some light on this one.

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-Local-build-failing-occasionally-tp60859p60868.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.