Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Christofer Dutz]

Hi Alex,

So looking at the download.xml it seems the ant build downloads:
http://www.apache.org/dyn/mirrors/mirrors.cgi/flex/pixelbender/1.0/binaries/apache-flex-sdk-pixel-bender-1.0.0-bin.zip

That should be easy to integrate … I’ll look into that.

Chris

Am 10.03.17, 08:44 schrieb "Alex Harui" :



On 3/9/17, 11:31 PM, "Christofer Dutz"  wrote:

>Ok … so that’s a Flash-Only bead?

I expect there will be a JS version, but it won't use the PBJ files.

>If the pbj files ar “sources”, I would add them to FlexJS. If they have
>to be simply read as pbj files, they belong into src/main/resources, If
>something compiles them to something else, then it belongs into
>src/main/flex …
>But I’ll have a look at that right away.

PBJ files are compiled PBK sources.  Because they are compiled output and
FlexJS is an Apache release, they cannot be placed in the repo and release
packages.  They have to be downloaded.  They could be generated by the
build script by compiling the PBK sources, but the PBK->PBJ compiler is an
Adobe tool and enough folks complained about setting it up that we decided
to create a separate release for it, so the flex-sdk could use the PBJ
files from the binary convenience package and not have to deal with the
Adobe tool.

The FlexJS Ant script has been adjusted to download the PBJ.  Now we need
to find a way to adjust Maven.  I'm now guessing that we should do
something like we did for the externs in compiler-build-tools to download
and place the PBJ files.

Thoughts?
-Alex

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Christofer Dutz]

Ok … commenting my own post again.

So Pixelbender is a native application. If I was to create a plugin to compile 
the pbk to pbj files, we would add a dependency to an installed tool and 
introduce another Environment Variable.
I don’t want to do that for the Maven build. I’m happy that we got rid of 
dependencies like that. 

If the pbj files are somewhat static, an alternative would be to package the 
compiled pbj files into a jar or zip and to manually release that. We would 
then reference these static resources from Maven
As they are our code, we should be allowed to release the binary files. If we 
do it as Apache Flex team, we can use the Apache Nexus and that mirrors stuff 
to Maven Central.

Would you think that’s a good way to do it?

Chris


Am 10.03.17, 08:43 schrieb "Christofer Dutz" :

Ok … so I changed the flexjs-maven-plugins default for 
“failOnCompilerWarnings” from false to true and now the build for Graphics 
fails.

Investigating the problem a little more, it seems as if the “pbk” files are 
source files and these are compiled by pixelbender to pbj files. Is that 
correct?
If yes, then the location for the pbk files is correct, however do we need 
a Maven plugin to compile the pbk to pbj files.

Chris


Am 10.03.17, 08:31 schrieb "Christofer Dutz" :

Ok … so that’s a Flash-Only bead?
If the pbj files ar “sources”, I would add them to FlexJS. If they have 
to be simply read as pbj files, they belong into src/main/resources, If 
something compiles them to something else, then it belongs into src/main/flex …
But I’ll have a look at that right away. 

Chris

Am 09.03.17, 16:37 schrieb "Harbs" :

Yes we do.

Yishay created a BlendBead and the pixel bender files are needed to 
get the blends to render correctly in Flash.

> On Mar 9, 2017, at 5:09 PM, Christofer Dutz 
 wrote:
> 
> Hi Yishay,
> 
> Yeah … but why do we need them in FlexJS now? I was quite happy 
with not needing stuff like that.
> 
> Chris
> 
> 
> Am 09.03.17, 15:27 schrieb "yishayw" :
> 
>PBJ files are a compiled version of a Pixel Bender script that 
can be loaded
>by the Flash Player [1].
> 
>As far as I understand it's a proprietary Adobe format which 
requires a
>proprietary tool, which is why we want to remove them from our 
repo. We're
>using them to implement blend-mode styles in flash.
> 
>I recently added them to downloads.xml, which is a file 
containing a bunch
>of ant tasks for adding external dependencies that need to be 
downloaded.
>We're wondering how the same trick could be done for Maven. 
> 
> 
>[1] http://ncannasse.fr/projects/pbj
> 
> 
> 
>--
>View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-New-Upstream-Dependency-was-Re-git-commit-flex-asjs-refs-heads-develop-Added-Shader-fil-tp60260p60279.html
>Sent from the Apache Flex Development mailing list archive at 
Nabble.com.
> 
> 

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Alex Harui]

On 3/9/17, 11:31 PM, "Christofer Dutz"  wrote:

>Ok … so that’s a Flash-Only bead?

I expect there will be a JS version, but it won't use the PBJ files.

>If the pbj files ar “sources”, I would add them to FlexJS. If they have
>to be simply read as pbj files, they belong into src/main/resources, If
>something compiles them to something else, then it belongs into
>src/main/flex …
>But I’ll have a look at that right away.

PBJ files are compiled PBK sources.  Because they are compiled output and
FlexJS is an Apache release, they cannot be placed in the repo and release
packages.  They have to be downloaded.  They could be generated by the
build script by compiling the PBK sources, but the PBK->PBJ compiler is an
Adobe tool and enough folks complained about setting it up that we decided
to create a separate release for it, so the flex-sdk could use the PBJ
files from the binary convenience package and not have to deal with the
Adobe tool.

The FlexJS Ant script has been adjusted to download the PBJ.  Now we need
to find a way to adjust Maven.  I'm now guessing that we should do
something like we did for the externs in compiler-build-tools to download
and place the PBJ files.

Thoughts?
-Alex

Re: [FlexJS] ASDoc (was Re: [FlexJS] maven faling (WeakMap)

[Christofer Dutz]

+1 For switching, because everything is better than having a legacy dependency 
on the old FDK.
+1 For adding a marker to implement something to produce static content as soon 
as we have the time.

Chris



Am 10.03.17, 07:12 schrieb "Alex Harui" :



On 3/9/17, 9:45 PM, "piotrz"  wrote:

>Hi Alex,
>
> +1 for switching.
>
>I can spent some time to try make your ADoc visually similar to the old
>one.- But for sure it won't be fast. My reference will be
>http://flex.apache.org/asdoc/
>
>It can be nice excercises.

I suppose it won't ever be as fast as static HTML, but we do want to see
how close we can get.  Also, I think it might be possible for the FlexJS
version to have fancier filtering and searching than the static version.

-Alex

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Christofer Dutz]

Ok … so I changed the flexjs-maven-plugins default for “failOnCompilerWarnings” 
from false to true and now the build for Graphics fails.

Investigating the problem a little more, it seems as if the “pbk” files are 
source files and these are compiled by pixelbender to pbj files. Is that 
correct?
If yes, then the location for the pbk files is correct, however do we need a 
Maven plugin to compile the pbk to pbj files.

Chris


Am 10.03.17, 08:31 schrieb "Christofer Dutz" :

Ok … so that’s a Flash-Only bead?
If the pbj files ar “sources”, I would add them to FlexJS. If they have to 
be simply read as pbj files, they belong into src/main/resources, If something 
compiles them to something else, then it belongs into src/main/flex …
But I’ll have a look at that right away. 

Chris

Am 09.03.17, 16:37 schrieb "Harbs" :

Yes we do.

Yishay created a BlendBead and the pixel bender files are needed to get 
the blends to render correctly in Flash.

> On Mar 9, 2017, at 5:09 PM, Christofer Dutz 
 wrote:
> 
> Hi Yishay,
> 
> Yeah … but why do we need them in FlexJS now? I was quite happy with 
not needing stuff like that.
> 
> Chris
> 
> 
> Am 09.03.17, 15:27 schrieb "yishayw" :
> 
>PBJ files are a compiled version of a Pixel Bender script that can 
be loaded
>by the Flash Player [1].
> 
>As far as I understand it's a proprietary Adobe format which 
requires a
>proprietary tool, which is why we want to remove them from our 
repo. We're
>using them to implement blend-mode styles in flash.
> 
>I recently added them to downloads.xml, which is a file containing 
a bunch
>of ant tasks for adding external dependencies that need to be 
downloaded.
>We're wondering how the same trick could be done for Maven. 
> 
> 
>[1] http://ncannasse.fr/projects/pbj
> 
> 
> 
>--
>View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-New-Upstream-Dependency-was-Re-git-commit-flex-asjs-refs-heads-develop-Added-Shader-fil-tp60260p60279.html
>Sent from the Apache Flex Development mailing list archive at 
Nabble.com.
> 
> 

Re: [FlexJS] ASDoc (was Re: [FlexJS] maven faling (WeakMap)

[yishayw]

+1

Sencha docs are eating their own dog food and doing fancy filtering. It
looks nice [1] for Sencha Touch 2.3, but gets slow for ExtJS 6.2 [2].

[1] http://docs.sencha.com/touch/2.3.1/#!/api
[2] http://docs.sencha.com/extjs/6.2.0/



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-ASDoc-was-Re-FlexJS-maven-faling-WeakMap-tp60325p60335.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Christofer Dutz]

Ok … so that’s a Flash-Only bead?
If the pbj files ar “sources”, I would add them to FlexJS. If they have to be 
simply read as pbj files, they belong into src/main/resources, If something 
compiles them to something else, then it belongs into src/main/flex …
But I’ll have a look at that right away. 

Chris

Am 09.03.17, 16:37 schrieb "Harbs" :

Yes we do.

Yishay created a BlendBead and the pixel bender files are needed to get the 
blends to render correctly in Flash.

> On Mar 9, 2017, at 5:09 PM, Christofer Dutz  
wrote:
> 
> Hi Yishay,
> 
> Yeah … but why do we need them in FlexJS now? I was quite happy with not 
needing stuff like that.
> 
> Chris
> 
> 
> Am 09.03.17, 15:27 schrieb "yishayw" :
> 
>PBJ files are a compiled version of a Pixel Bender script that can be 
loaded
>by the Flash Player [1].
> 
>As far as I understand it's a proprietary Adobe format which requires a
>proprietary tool, which is why we want to remove them from our repo. 
We're
>using them to implement blend-mode styles in flash.
> 
>I recently added them to downloads.xml, which is a file containing a 
bunch
>of ant tasks for adding external dependencies that need to be 
downloaded.
>We're wondering how the same trick could be done for Maven. 
> 
> 
>[1] http://ncannasse.fr/projects/pbj
> 
> 
> 
>--
>View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-New-Upstream-Dependency-was-Re-git-commit-flex-asjs-refs-heads-develop-Added-Shader-fil-tp60260p60279.html
>Sent from the Apache Flex Development mailing list archive at 
Nabble.com.
> 
> 

Re: draft announcement for 4.16.0

[Christofer Dutz]

+1

Am 10.03.17, 07:28 schrieb "Alex Harui" :

+1

On 3/9/17, 10:20 PM, "Justin Mclean"  wrote:

>Hi,
>
>once the mirror have caught up I intend to use this as the announcement -
>any feedback?
>
>Thanks,
>Justin
>
>> On 12 Feb 2017, at 10:30 am, Justin Mclean 
>>wrote:
>> 
>> The Apache Flex community is pleased to announce the release of Apache
>> Flex SDK 4.16.0
>> 
>> The Apache Flex SDK is a highly productive, open source application
>> framework for building and maintaining expressive applications that
>>deploy
>> consistently on all major browsers, desktops and devices (including
>> smartphones, tablets and tv).
>> 
>> The Apache Flex SDK 4.16.0 is an update to Apache Flex SDK 4.15.0 that
>> adds support for the latest Flash Player and AIR runtimes and many bug
>>fixes.
>> 
>> For more details, see the release notes at:
>> http://www.apache.org/dist/flex/4.16.0/RELEASE_NOTES
>>
>> 
>> 
>> The Apache Flex SDK is available in source form from the following
>> download page:
>> http://flex.apache.org/download-source.html
>>
>> 
>> It's recommended you install the Apache Flex SDK by using the Apache
>>Flex
>> IDE installer:
>> http://flex.apache.org/installer.html
>>
>> 
>> When downloading from a mirror site, please remember to verify the
>> downloads using signatures or MD5 hashes.
>> 
>> For more information on Apache Flex, visit the project home page:
>> http://flex.apache.org 
>> 
>> Thank you for using Apache Flex,
>> The Apache Flex Community
>

Re: Flex SDK release

[Justin Mclean]

Hi,

> I would add also to this activities post on our blog - It seems to be dead a
> bit.

I think there a JIRA for that as well, if not when It’s back up I’ll add one.

Our whole site really needs fresh content or rather people willing to help out 
with that.

Thanks,
Justin

Re: Flex SDK release

[piotrz]

Hi Justin,

I would add also to this activities post on our blog - It seems to be dead a
bit.

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/Flex-SDK-release-tp60329p60331.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: draft announcement for 4.16.0

[Alex Harui]

+1

On 3/9/17, 10:20 PM, "Justin Mclean"  wrote:

>Hi,
>
>once the mirror have caught up I intend to use this as the announcement -
>any feedback?
>
>Thanks,
>Justin
>
>> On 12 Feb 2017, at 10:30 am, Justin Mclean 
>>wrote:
>> 
>> The Apache Flex community is pleased to announce the release of Apache
>> Flex SDK 4.16.0
>> 
>> The Apache Flex SDK is a highly productive, open source application
>> framework for building and maintaining expressive applications that
>>deploy
>> consistently on all major browsers, desktops and devices (including
>> smartphones, tablets and tv).
>> 
>> The Apache Flex SDK 4.16.0 is an update to Apache Flex SDK 4.15.0 that
>> adds support for the latest Flash Player and AIR runtimes and many bug
>>fixes.
>> 
>> For more details, see the release notes at:
>> http://www.apache.org/dist/flex/4.16.0/RELEASE_NOTES
>>
>> 
>> 
>> The Apache Flex SDK is available in source form from the following
>> download page:
>> http://flex.apache.org/download-source.html
>>
>> 
>> It's recommended you install the Apache Flex SDK by using the Apache
>>Flex
>> IDE installer:
>> http://flex.apache.org/installer.html
>>
>> 
>> When downloading from a mirror site, please remember to verify the
>> downloads using signatures or MD5 hashes.
>> 
>> For more information on Apache Flex, visit the project home page:
>> http://flex.apache.org 
>> 
>> Thank you for using Apache Flex,
>> The Apache Flex Community
>

Flex SDK release

[Justin Mclean]

Hi,

Can any one help out with these activities as part of the Flex SDK release 
[1][2][3]. They are updating the installer script, update the web site and the 
AS docs.

Thanks,
Justin

PS looks like JIRA may of just gone off the air so these links may not work.

1. https://issues.apache.org/jira/browse/FLEX-35156
2. https://issues.apache.org/jira/browse/FLEX-35157
3. https://issues.apache.org/jira/browse/FLEX-35158

Re: [FlexJS] ASDoc (was Re: [FlexJS] maven faling (WeakMap)

[Alex Harui]

On 3/9/17, 9:45 PM, "piotrz"  wrote:

>Hi Alex,
>
> +1 for switching.
>
>I can spent some time to try make your ADoc visually similar to the old
>one.- But for sure it won't be fast. My reference will be
>http://flex.apache.org/asdoc/
>
>It can be nice excercises.

I suppose it won't ever be as fast as static HTML, but we do want to see
how close we can get.  Also, I think it might be possible for the FlexJS
version to have fancier filtering and searching than the static version.

-Alex

Re: draft announcement for 4.16.0

[Justin Mclean]

Hi,

once the mirror have caught up I intend to use this as the announcement - any 
feedback?

Thanks,
Justin

> On 12 Feb 2017, at 10:30 am, Justin Mclean  wrote:
> 
> The Apache Flex community is pleased to announce the release of Apache
> Flex SDK 4.16.0
> 
> The Apache Flex SDK is a highly productive, open source application
> framework for building and maintaining expressive applications that deploy
> consistently on all major browsers, desktops and devices (including
> smartphones, tablets and tv).
> 
> The Apache Flex SDK 4.16.0 is an update to Apache Flex SDK 4.15.0 that
> adds support for the latest Flash Player and AIR runtimes and many bug fixes.
> 
> For more details, see the release notes at:
> http://www.apache.org/dist/flex/4.16.0/RELEASE_NOTES 
> 
> 
> 
> The Apache Flex SDK is available in source form from the following
> download page:
> http://flex.apache.org/download-source.html 
> 
> 
> It's recommended you install the Apache Flex SDK by using the Apache Flex
> IDE installer:
> http://flex.apache.org/installer.html 
> 
> When downloading from a mirror site, please remember to verify the
> downloads using signatures or MD5 hashes.
> 
> For more information on Apache Flex, visit the project home page:
> http://flex.apache.org 
> 
> Thank you for using Apache Flex,
> The Apache Flex Community

Re: [FlexJS] ASDoc (was Re: [FlexJS] maven faling (WeakMap)

[piotrz]

Hi Alex,

 +1 for switching.

I can spent some time to try make your ADoc visually similar to the old
one.- But for sure it won't be fast. My reference will be
http://flex.apache.org/asdoc/

It can be nice excercises.

Piotr




-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-ASDoc-was-Re-FlexJS-maven-faling-WeakMap-tp60325p60326.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

[FlexJS] ASDoc (was Re: [FlexJS] maven faling (WeakMap)

[Alex Harui]

Renaming this thread to get more attention as well.

The Ant build is trying to generate ASDoc and failing because it uses the
Flex SDK compiler to generate ASDoc and it can't handle the new compiler
features we've added to Falcon.

So again, see my questions below.  The ASDoc example is ugly but somewhat
operational [1][2].  Should we switch to using it for now?

Thanks,
-Alex

[1] https://s.apache.org/FlexJSASDoc
[2] https://s.apache.org/FlexJSASDocJS


On 3/5/17, 8:10 PM, "Alex Harui"  wrote:

>How many folks actually use the FlexJS ASDoc?
>
>Do folks have the cycles to make the ASDoc example a bit better?  The dual
>branch is pretty much going to force us to give up on using regular Flex
>to generate our ASDoc.  It might be time to just switch to the ASDoc
>example until Chris or others get the DITA-to-HTML conversion working.  At
>least more of us know how to change MXML and AS in the example than change
>the compiler or tweak the XSLT for DITA.  And it would be eating our own
>dog food.
>
>Thoughts?
>-Alex

[VOTE] [RESULT] Release Apache Flex 4.16 RC3

[justin]

Hi,

The vote passes with 3 +1 votes and no other votes.

Thanks to everyone who tested out this or any of the previous RCs.

Justin

[4.16.0] # Apache Flex SDK 4.16.0 nightly build 254: Successful

[flex . ci . builds]

flex-sdk_release-candidate - Build #254 - Successful

Changes since last build:
No changes

For more information, check the console output at 
http://apacheflexbuild.cloudapp.net:8080/job/flex-sdk_release-candidate/254/.

Re: Just need to vent...

[Harbs]

JustificationStyle is in a separate Text.swc file, and yes, the js files are 
there.

The js files are copied correctly to the bin/js-debug folder, but something is 
wrong with the dependency tree.

There’s quite a few classes that are not recognized. Some are in TLF and some 
are in Text.

The full console output was in that paste.

> On Mar 10, 2017, at 12:20 AM, Alex Harui  wrote:
> 
> 
> 
> On 3/9/17, 12:41 PM, "Harbs"  wrote:
> 
>> I’ve already eliminated those errors by removing circular dependencies.
>> 
>> I moved static functions to utility classes and all inter-class
>> references are using interfaces.
>> 
>> To get you the output, I’d need to revert the state of the repo.
>> 
>> I’m not getting a new class of errors:
>> 
>> https://paste.apache.org/mQIx 
>> 
>> Which outputs this HTML:
>> https://paste.apache.org/1xWp 
>> 
>> BTW, I’ve been using the dependency chain in the HTML to figure out where
>> the interdependencies are. I used the same technique while tracking these
>> issues down in my app. It’s a very effective method. Of course it was
>> much easier in my app than it’s proved to be in TLF…
> 
> Hmm.  Looks like some files like JustificationStyle are not even being
> seen as a dependency by the compiler.  I would first unpack TLF.swc and
> see if the js/out folder even contains a JustificationStyle.js.  If it
> doesn't the build for TLF.swc needs to be fixed to cause
> JustificationStyle.as to be cross-compiled.
> 
> If JustificationStyle.js is in the SWC, then the question is what file
> goog.required it before remove-circulars.  But that usually shows up in
> the console output.
> 
> HTH,
> -Alex
> 

Re: Just need to vent...

[Harbs]

It looks to me like the missing definitions are classes which are only used in 
static methods.

Could it be that Falcon is not properly declaring dependencies in static 
methods?

> On Mar 10, 2017, at 12:28 AM, Harbs  wrote:
> 
> JustificationStyle is in a separate Text.swc file, and yes, the js files are 
> there.
> 
> The js files are copied correctly to the bin/js-debug folder, but something 
> is wrong with the dependency tree.
> 
> There’s quite a few classes that are not recognized. Some are in TLF and some 
> are in Text.
> 
> The full console output was in that paste.
> 
>> On Mar 10, 2017, at 12:20 AM, Alex Harui  wrote:
>> 
>> 
>> 
>> On 3/9/17, 12:41 PM, "Harbs"  wrote:
>> 
>>> I’ve already eliminated those errors by removing circular dependencies.
>>> 
>>> I moved static functions to utility classes and all inter-class
>>> references are using interfaces.
>>> 
>>> To get you the output, I’d need to revert the state of the repo.
>>> 
>>> I’m not getting a new class of errors:
>>> 
>>> https://paste.apache.org/mQIx 
>>> 
>>> Which outputs this HTML:
>>> https://paste.apache.org/1xWp 
>>> 
>>> BTW, I’ve been using the dependency chain in the HTML to figure out where
>>> the interdependencies are. I used the same technique while tracking these
>>> issues down in my app. It’s a very effective method. Of course it was
>>> much easier in my app than it’s proved to be in TLF…
>> 
>> Hmm.  Looks like some files like JustificationStyle are not even being
>> seen as a dependency by the compiler.  I would first unpack TLF.swc and
>> see if the js/out folder even contains a JustificationStyle.js.  If it
>> doesn't the build for TLF.swc needs to be fixed to cause
>> JustificationStyle.as to be cross-compiled.
>> 
>> If JustificationStyle.js is in the SWC, then the question is what file
>> goog.required it before remove-circulars.  But that usually shows up in
>> the console output.
>> 
>> HTH,
>> -Alex
>> 
> 

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Harbs]

Is it still using old falcon artifacts?

> On Mar 9, 2017, at 11:37 PM, Alex Harui  wrote:
> 
> [1] looks like it is still looking for mx.core.ByteArrayAsset instead of
> org.apache.flex.core.ByteArrayAsset
> 
> HTH,
> -Alex
> 
> On 3/9/17, 12:32 PM, "piotrz"  wrote:
> 
>> I just added manually missing *.pbj to the folder
>> "frameworks\projects\Graphics\src\main\flex\org\apache\flex\graphics\utils
>> \shaderClasses\"
>> and it seems that for Maven it's not enough this files, cause graphics is
>> still failing for me. [1]
>> 
>> [1] https://paste.apache.org/zKrJ
>> 
>> 
>> 
>> -
>> Apache Flex PMC
>> piotrzarzyck...@gmail.com
>> --
>> View this message in context:
>> http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-Jenkins-
>> What-s-up-with-framework-build-tp60281p60311.html
>> Sent from the Apache Flex Development mailing list archive at Nabble.com.
> 

Re: Just need to vent...

[Alex Harui]

On 3/9/17, 12:41 PM, "Harbs"  wrote:

>I’ve already eliminated those errors by removing circular dependencies.
>
>I moved static functions to utility classes and all inter-class
>references are using interfaces.
>
>To get you the output, I’d need to revert the state of the repo.
>
>I’m not getting a new class of errors:
>
>https://paste.apache.org/mQIx 
>
>Which outputs this HTML:
>https://paste.apache.org/1xWp 
>
>BTW, I’ve been using the dependency chain in the HTML to figure out where
>the interdependencies are. I used the same technique while tracking these
>issues down in my app. It’s a very effective method. Of course it was
>much easier in my app than it’s proved to be in TLF…

Hmm.  Looks like some files like JustificationStyle are not even being
seen as a dependency by the compiler.  I would first unpack TLF.swc and
see if the js/out folder even contains a JustificationStyle.js.  If it
doesn't the build for TLF.swc needs to be fixed to cause
JustificationStyle.as to be cross-compiled.

If JustificationStyle.js is in the SWC, then the question is what file
goog.required it before remove-circulars.  But that usually shows up in
the console output.

HTH,
-Alex

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Alex Harui]

On 3/9/17, 1:25 PM, "Christofer Dutz"  wrote:

>If it’s our stuff and we are allowed to release it, I could whip up a
>small maven project to release the files as snapshots.
>But I still don’t quite understand why we need them now. Has something
>changed in the Graphics module?
>What are these files? I’d like to understand what they are used for.

These files work with Flash to execute special visual effects.   Yishay
and Harbs want have a FlexJS version of these effects.

>
>Tomorrow / Like the whole last 3 weeks, I’m completely swamped. I hope to
>spare a few hours on Saturday.
>At least on Monday most of the ApacheCon / FlexJS Summit preparation
>stress should be over ;-)

Actually, maybe instead of making it a Maven dependency, we can treat it
like the externs we download.  How do you decide whether something should
be converted into a Maven artifact or just downloaded and unpacked?

Thanks,
-Alex

Re: [FlexJS] Panel

[Peter Ent]

The way things are set up right now, due to the nesting of containers (div
or DisplayObjectContainer), you do get (0,0) at the top. It will usually
be necessary to have nested containers for scrolling purposes if nothing
else.

As I'm going through the code, I don't see that IChrome is actually being
recognized!! It looks like Panel is special-casing it somehow, but right
now Panel is very messed up for me due to changes I've made in my
workspace, so I'm not really certain what is going on it with.

‹peter

On 3/9/17, 2:34 PM, "yishayw"  wrote:

>I think it makes sense to start the coordinates axis below the titlebar.
>In
>ExtJS they have the concept of docked components such as toolbars and
>titles. So I think you can do (I changed syntax to pseudo-mxml)
>
>
>
>
>docked="top"/>
>
>
>In other words, docked items override the container's layout and change
>the
>coordinates axis point of origin.
>
>
>
>--
>View this message in context:
>http://apache-flex-development.247.n4.nabble.com/FlexJS-Panel-tp60285p
>60307.html
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: AW: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Alex Harui]

On 3/9/17, 1:09 PM, "Christofer Dutz"  wrote:

>Are we allowed to do that? Are the pbj files our "code" or resource?

AFAIK, yes we can.  The artifacts we are discussing are an official Apache
release.  A obj file is a compiled file like a .class or .swf file

https://dist.apache.org/repos/dist/release/flex/pixelbender/1.0/


-Alex

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Christofer Dutz]

If it’s our stuff and we are allowed to release it, I could whip up a small 
maven project to release the files as snapshots.
But I still don’t quite understand why we need them now. Has something changed 
in the Graphics module? 
What are these files? I’d like to understand what they are used for.

Tomorrow / Like the whole last 3 weeks, I’m completely swamped. I hope to spare 
a few hours on Saturday.
At least on Monday most of the ApacheCon / FlexJS Summit preparation stress 
should be over ;-)

Chris



Am 09.03.17, 22:09 schrieb "Christofer Dutz" :

Are we allowed to do that? Are the pbj files our "code" or resource?

Chris



Von meinem Samsung Galaxy Smartphone gesendet.


 Ursprüngliche Nachricht 
Von: Alex Harui 
Datum: 09.03.17 19:41 (GMT+01:00)
An: dev@flex.apache.org
Betreff: Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

Looks like Maven is reporting success, but it hasn't succeeded.  Various
SWCs are failing to build because of errors, but the root cause seems to
be the missing .pbj files in the Graphics.swc.

So, I think we need to figure out how to get those pbj files in a Maven
way.  Can we just stick the release artifacts into Maven Central manually?

-Alex

On 3/9/17, 9:50 AM, "carlos.rov...@gmail.com on behalf of Carlos Rovira"
 wrote:

>Hi, just see this thread, and update from repos and build all and seems
>working ok
>I'm a bit off due to work load, so was worried, but seems all is ok...just
>fixed minutes ago? :)
>
>2017-03-09 16:39 GMT+01:00 piotrz :
>
>> Hi Chris!
>>
>> I got you :) Was waiting for the proper thread in order to complain
>>about
>> something. I see after some changes in compiler (maybe) that some
>>modules
>> are failing but in general build didn't fail.
>>
>> Example:
>> http://apacheflexbuild.cloudapp.net:8080/job/flex-
>> asjs%20(maven)/161/console
>>
>> The second thing why this errors are happened ?
>>
>> Thanks,
>> Pete
>>
>>
>>
>>
>>
>> -
>> Apache Flex PMC
>> piotrzarzyck...@gmail.com
>> --
>> View this message in context: http://apache-flex-
>> development.247.n4.nabble.com/FlexJS-Maven-Jenkins-What-
>> s-up-with-framework-build-tp60281p60284.html
>> Sent from the Apache Flex Development mailing list archive at
>>Nabble.com.
>>
>
>
>
>--
>
>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 vigente Ley Orgánica de Protección de Datos (15/1999), le
>comunicamos
>que sus datos forman parte de un fichero cuyo responsable es CODEOSCOPIC
>S.A. La finalidad de dicho tratamiento es facilitar la prestación del
>servicio o información solicitados, teniendo usted derecho de acceso,
>rectificación, cancelación y oposición de sus datos dirigiéndose a
>nuestras
>oficinas c/ Paseo de la Habana 9-11, 28036, Madrid con la documentación
>necesaria.

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Alex Harui]

[1] looks like it is still looking for mx.core.ByteArrayAsset instead of
org.apache.flex.core.ByteArrayAsset

HTH,
-Alex

On 3/9/17, 12:32 PM, "piotrz"  wrote:

>I just added manually missing *.pbj to the folder
>"frameworks\projects\Graphics\src\main\flex\org\apache\flex\graphics\utils
>\shaderClasses\"
>and it seems that for Maven it's not enough this files, cause graphics is
>still failing for me. [1]
>
>[1] https://paste.apache.org/zKrJ
>
>
>
>-
>Apache Flex PMC
>piotrzarzyck...@gmail.com
>--
>View this message in context:
>http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-Jenkins-
>What-s-up-with-framework-build-tp60281p60311.html
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

AW: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Christofer Dutz]

Are we allowed to do that? Are the pbj files our "code" or resource?

Chris



Von meinem Samsung Galaxy Smartphone gesendet.


 Ursprüngliche Nachricht 
Von: Alex Harui 
Datum: 09.03.17 19:41 (GMT+01:00)
An: dev@flex.apache.org
Betreff: Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

Looks like Maven is reporting success, but it hasn't succeeded.  Various
SWCs are failing to build because of errors, but the root cause seems to
be the missing .pbj files in the Graphics.swc.

So, I think we need to figure out how to get those pbj files in a Maven
way.  Can we just stick the release artifacts into Maven Central manually?

-Alex

On 3/9/17, 9:50 AM, "carlos.rov...@gmail.com on behalf of Carlos Rovira"
 wrote:

>Hi, just see this thread, and update from repos and build all and seems
>working ok
>I'm a bit off due to work load, so was worried, but seems all is ok...just
>fixed minutes ago? :)
>
>2017-03-09 16:39 GMT+01:00 piotrz :
>
>> Hi Chris!
>>
>> I got you :) Was waiting for the proper thread in order to complain
>>about
>> something. I see after some changes in compiler (maybe) that some
>>modules
>> are failing but in general build didn't fail.
>>
>> Example:
>> http://apacheflexbuild.cloudapp.net:8080/job/flex-
>> asjs%20(maven)/161/console
>>
>> The second thing why this errors are happened ?
>>
>> Thanks,
>> Pete
>>
>>
>>
>>
>>
>> -
>> Apache Flex PMC
>> piotrzarzyck...@gmail.com
>> --
>> View this message in context: http://apache-flex-
>> development.247.n4.nabble.com/FlexJS-Maven-Jenkins-What-
>> s-up-with-framework-build-tp60281p60284.html
>> Sent from the Apache Flex Development mailing list archive at
>>Nabble.com.
>>
>
>
>
>--
>
>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 vigente Ley Orgánica de Protección de Datos (15/1999), le
>comunicamos
>que sus datos forman parte de un fichero cuyo responsable es CODEOSCOPIC
>S.A. La finalidad de dicho tratamiento es facilitar la prestación del
>servicio o información solicitados, teniendo usted derecho de acceso,
>rectificación, cancelación y oposición de sus datos dirigiéndose a
>nuestras
>oficinas c/ Paseo de la Habana 9-11, 28036, Madrid con la documentación
>necesaria.

Re: Just need to vent...

[Harbs]

I’ve already eliminated those errors by removing circular dependencies. 

I moved static functions to utility classes and all inter-class references are 
using interfaces.

To get you the output, I’d need to revert the state of the repo.

I’m not getting a new class of errors:

https://paste.apache.org/mQIx 

Which outputs this HTML:
https://paste.apache.org/1xWp 

BTW, I’ve been using the dependency chain in the HTML to figure out where the 
interdependencies are. I used the same technique while tracking these issues 
down in my app. It’s a very effective method. Of course it was much easier in 
my app than it’s proved to be in TLF…

> On Mar 9, 2017, at 10:26 PM, Alex Harui  wrote:
> 
> It might be a bug in Google Closure Compiler.  Can you send me the
> bin/js-debug/index.html.  It will have the dependency chain in it.
> 
> Thanks,
> -Alex
> 
> On 3/9/17, 12:43 AM, "Harbs"  wrote:
> 
>> I already using remove-circulars and it’s not helping in this case. If
>> you’re interested here’s the output: https://paste.apache.org/Kd9l
>> 
>> And running the debug in the browser has tons of runtime errors.
>> 
>> I had this problem when porting my app as well. Basically static
>> properties really mess things up, and TLF has lots of static stuff.
>> 
>> I managed to get rid of some issues with the interfaces, but it could be
>> I should have been spending more time getting rid of some of the static
>> properties…
>> 
>>> On Mar 9, 2017, at 3:10 AM, Alex Harui  wrote:
>>> 
>>> 
>>> 
>>> On 3/8/17, 5:05 PM, "Harbs"  wrote:
>>> 
 The only other option is to just drop TLF and that’s way more work (I
 think).
 
 As it is, it cannot be compiled by the Google Compiler.
 
 There’s lots of superclass dependencies on subclasses.
 
 The good news is that I’m making it a lot more reliant on interfaces
 which is probably a good thing.
>>> 
>>> More reliance on interfaces is a good thing, but I don't know why we
>>> can't
>>> just make folks use -remove-circulars.
>>> 
>>> It is "ok" to have superclass dependencies on subclasses as long as they
>>> aren't true circularities.  If they are tests for a particular type,
>>> inside method code, it shouldn't matter.  We just have to get the set of
>>> requires right.  Or is there some case that is truly impossible?
>>> 
>>> Thanks,
>>> -Alex
>>> 
>> 
> 

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[piotrz]

I just added manually missing *.pbj to the folder
"frameworks\projects\Graphics\src\main\flex\org\apache\flex\graphics\utils\shaderClasses\"
and it seems that for Maven it's not enough this files, cause graphics is
still failing for me. [1]

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



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-Jenkins-What-s-up-with-framework-build-tp60281p60311.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: Just need to vent...

[Alex Harui]

It might be a bug in Google Closure Compiler.  Can you send me the
bin/js-debug/index.html.  It will have the dependency chain in it.

Thanks,
-Alex

On 3/9/17, 12:43 AM, "Harbs"  wrote:

>I already using remove-circulars and it’s not helping in this case. If
>you’re interested here’s the output: https://paste.apache.org/Kd9l
>
>And running the debug in the browser has tons of runtime errors.
>
>I had this problem when porting my app as well. Basically static
>properties really mess things up, and TLF has lots of static stuff.
>
>I managed to get rid of some issues with the interfaces, but it could be
>I should have been spending more time getting rid of some of the static
>properties…
>
>> On Mar 9, 2017, at 3:10 AM, Alex Harui  wrote:
>> 
>> 
>> 
>> On 3/8/17, 5:05 PM, "Harbs"  wrote:
>> 
>>> The only other option is to just drop TLF and that’s way more work (I
>>> think).
>>> 
>>> As it is, it cannot be compiled by the Google Compiler.
>>> 
>>> There’s lots of superclass dependencies on subclasses.
>>> 
>>> The good news is that I’m making it a lot more reliant on interfaces
>>> which is probably a good thing.
>> 
>> More reliance on interfaces is a good thing, but I don't know why we
>>can't
>> just make folks use -remove-circulars.
>> 
>> It is "ok" to have superclass dependencies on subclasses as long as they
>> aren't true circularities.  If they are tests for a particular type,
>> inside method code, it shouldn't matter.  We just have to get the set of
>> requires right.  Or is there some case that is truly impossible?
>> 
>> Thanks,
>> -Alex
>> 
>

Re: [FalconJX] regex patterns

[Alex Harui]

I put in some code to handle whitespace chars.  Or do we need to handle
all chars in some unicode ranges?

-Alex

On 3/9/17, 10:42 AM, "Harbs"  wrote:

>Here’s one of them:
>
>static private const anyPrintChar:RegExp =
>/[^\u0009\u000a\u000d\u0020]/g;
>
>outputs:
>
>org.apache.flex.textLayout.elements.SpanElement.anyPrintChar = /[^ 
>
> ]/g;
>
>> On Mar 9, 2017, at 8:20 PM, Harbs  wrote:
>> 
>> Yes. These are from TLF. The strings are evaluated by Java. There’s no
>>error except on the Javascript side because the string literals become
>>non-sensical.
>> 
>> We had a similar issue when porting our app. I think we used
>>String.fromCharCode() to get around the problem there.
>> 
>> Basically, we need some way to tell the compiler that any strings used
>>in RegExp objects should come through bit-for-bit identical and not be
>>evaluated as Strings. I have no idea what’s involved in doing that in
>>Java...
>> 
>>> On Mar 9, 2017, at 7:43 PM, Alex Harui  wrote:
>>> 
>>> Are these from TLF?  What error are you getting?  Falcon compiles TLF
>>>as
>>> part of the integration tests.  Is it the JS output that is broken?
>>> 
>>> Thanks,
>>> -Alex
>>> 
>>> On 3/9/17, 1:18 AM, "Harbs"  wrote:
>>> 
 I’ve come across quite a few regex patterns which break compilation.
 Here’s some examples:
 
 private static const _newLineTabPattern:RegExp =
/[\u0009\u000a\u000d]/g;
 static private const brRegEx:RegExp = /\u2028/;
 private static const _newLineRegex:RegExp = /\u000A|\u000D\u000A?/g;
 public static const anyPrintChar:RegExp =
/[^\u0009\u000a\u000d\u0020]/g;
 public static const attrRegex:RegExp =
 /\s+(\w+)(?:\s*=\s*(".*?"|'.*?'|[\w\.]+))?/sg;
 public static const tagRegex:RegExp =
 
/<(\/?)(\w+)((?:\s+\w+(?:\s*=\s*(?:".*?"|'.*?'|[\w\.]+))?)*)\s*(\/?)>/s
g;
 public static const stripRegex:RegExp =
 /|<\?(".*?"|'.*?'|[^>"']+)*>|"']+)*>/sg;
 
 The last three have to do with unsupported flags, but the first four
 break simply because the compiler evaluates the strings and they
become
 spaces and line breaks, etc.
 
 What can we do to prevent the compiler from killing patterns during
 compilation?
 
 Harbs
>>> 
>> 
>

Re: [FlexJS] Group container

[yishayw]

Carlos Rovira wrote
> Hi Peter, I think this awesome. get rid of the hardcoded styles in
> component classes is such important thing and first point not only in your
> effort of create a good layout strategy, but an important previous step if
> we want to implement theming in FlexJS.

It seems to me theming in FlexJS should be pretty easy, regardless of
whether or not we're using CSS for layouts. You can always swap the layout
bead (and view beads for that matter) in your CSS.




--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Group-container-tp60170p60308.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] Panel

[yishayw]

I think it makes sense to start the coordinates axis below the titlebar. In
ExtJS they have the concept of docked components such as toolbars and
titles. So I think you can do (I changed syntax to pseudo-mxml)







In other words, docked items override the container's layout and change the
coordinates axis point of origin.



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Panel-tp60285p60307.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlaconJX] problems with class references

[Harbs]

> 
> Not sure what you mean by "importer classes have namespace declarations".
> What does that code look like?

This:
internal class BaseTextLayoutImporter extends ConverterBase implements 
ITextImporter
{
private var _ns:Namespace;  // namespace of 
expected in imported/exported content

private var _textFlowNamespace:Namespace; // namespace of the 
TextFlow element against which the namespaces of the following elements are 
validated

public function BaseTextLayoutImporter(nsValue:Namespace, 
config:ImportExportConfiguration)
{
_ns = nsValue;
_config = config;
}

public function get ns(): Namespace
{
return _ns;
}

public class TextLayoutImporter extends BaseTextLayoutImporter 
implements ITextLayoutImporter
{
public function TextLayoutImporter()
{
super(new Namespace("flow", 
"http://ns.adobe.com/textLayout/2008;), defaultConfiguration);
}



> Are there any occurrences of two colons "::" in the source file?

With the exception of CONFIG::debug, this is it:

var beginArrayChild:XML = 
xmlTree..*::BeginMissingElements[0];
var endArrayChild:XML = 
xmlTree..*::EndMissingElements[0];
var textLayoutMarkup:XML = 
xmlTree..*::TextFlow[0];

Harbs

Re: [FalconJX] casting idea

[Harbs]

Huh. I did not know about @flexjsemitcoercion.

Good to know!

> On Mar 9, 2017, at 8:51 PM, Alex Harui  wrote:
> 
> 
> 
> On 3/9/17, 10:44 AM, "Harbs"  wrote:
> 
>> Umm... I guess I was not thinking clearly. (exhaustion can do that…) ;-)
>> I was thinking that the casting happens when compiling the app, but
>> that’s not true as the js from the library is pre-compiled.
>> 
>> However, I don’t think I can globally disable “as” in TLF, and I don’t
>> think we can do that in the FlexJS SDK libs either.
> 
> But we can turn it on where we need it.  Basically, -skipAsCoercions and
> @flexjsemitcoercion assume you don't want 99% of "as" usage and will
> annotate where you need it.  The default assumes you need it and will
> @flexjsignorecoercion where you don't need it.  The framework could
> probably get away with assuming we don't need it.
> 
> HTH,
> -Alex
> 

Re: git commit: [flex-falcon] [refs/heads/develop] - fix interface override checking

[Alex Harui]

OK, there is another thread about Maven discussing this.  These errors are
because upstream SWCs failed because of the missing .pbj files.

-Alex

On 3/9/17, 10:49 AM, "piotrz"  wrote:

>Error example from build [1]
>
>[INFO] --- flexjs-maven-plugin:0.8.0-SNAPSHOT:compile-as
>(default-compile-as) @ Mobile ---
>[INFO] Executing COMPC in tool group Falcon with args:
>[-load-config=C:\Jenkins\workspace\flex-asjs
>(maven)\frameworks\projects\Mobile\target\compile-as-config.xml]
>Loading configuration: C:\Jenkins\workspace\flex-asjs
>(maven)\frameworks\projects\Mobile\target\compile-as-config.xml
>
>C:\Jenkins\workspace\flex-asjs
>(maven)\frameworks\projects\Mobile\src\main\flex\org\apache\cordova\Applic
>ation.as(22):
>col: 9 Warning: Definition org.apache.flex.core.Application could not be
>found.
>
>   import org.apache.flex.core.Application;
>  ^
>
>C:\Jenkins\workspace\flex-asjs
>(maven)\frameworks\projects\Mobile\src\main\flex\org\apache\cordova\Applic
>ation.as(33):
>col: 35 Error: The definition of base class Application was not found.
>
>   public class Application extends org.apache.flex.core.Application
>implements IFlexInfo
>
>
>[1]
>http://apacheflexbuild.cloudapp.net:8080/job/flex-asjs%20(maven)/161/conso
>le
>
>Piotr
>
>
>
>
>-
>Apache Flex PMC
>piotrzarzyck...@gmail.com
>--
>View this message in context:
>http://apache-flex-development.247.n4.nabble.com/Re-git-commit-flex-fa
>lcon-refs-heads-develop-fix-interface-override-checking-tp60274p60303.html
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: git commit: [flex-falcon] [refs/heads/develop] - fix interface override checking

[piotrz]

Error example from build [1]

[INFO] --- flexjs-maven-plugin:0.8.0-SNAPSHOT:compile-as
(default-compile-as) @ Mobile ---
[INFO] Executing COMPC in tool group Falcon with args:
[-load-config=C:\Jenkins\workspace\flex-asjs
(maven)\frameworks\projects\Mobile\target\compile-as-config.xml]
Loading configuration: C:\Jenkins\workspace\flex-asjs
(maven)\frameworks\projects\Mobile\target\compile-as-config.xml

C:\Jenkins\workspace\flex-asjs
(maven)\frameworks\projects\Mobile\src\main\flex\org\apache\cordova\Application.as(22):
col: 9 Warning: Definition org.apache.flex.core.Application could not be
found.

import org.apache.flex.core.Application;
   ^

C:\Jenkins\workspace\flex-asjs
(maven)\frameworks\projects\Mobile\src\main\flex\org\apache\cordova\Application.as(33):
col: 35 Error: The definition of base class Application was not found.

public class Application extends org.apache.flex.core.Application
implements IFlexInfo
  

[1]
http://apacheflexbuild.cloudapp.net:8080/job/flex-asjs%20(maven)/161/console

Piotr




-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/Re-git-commit-flex-falcon-refs-heads-develop-fix-interface-override-checking-tp60274p60303.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FalconJX] casting idea

[Alex Harui]

On 3/9/17, 10:44 AM, "Harbs"  wrote:

>Umm... I guess I was not thinking clearly. (exhaustion can do that…) ;-)
>I was thinking that the casting happens when compiling the app, but
>that’s not true as the js from the library is pre-compiled.
>
>However, I don’t think I can globally disable “as” in TLF, and I don’t
>think we can do that in the FlexJS SDK libs either.

But we can turn it on where we need it.  Basically, -skipAsCoercions and
@flexjsemitcoercion assume you don't want 99% of "as" usage and will
annotate where you need it.  The default assumes you need it and will
@flexjsignorecoercion where you don't need it.  The framework could
probably get away with assuming we don't need it.

HTH,
-Alex

Re: [FalconJX] casting idea

[Harbs]

Umm... I guess I was not thinking clearly. (exhaustion can do that…) ;-) I was 
thinking that the casting happens when compiling the app, but that’s not true 
as the js from the library is pre-compiled.

However, I don’t think I can globally disable “as” in TLF, and I don’t think we 
can do that in the FlexJS SDK libs either.

> On Mar 9, 2017, at 8:30 PM, Alex Harui  wrote:
> 
> 
> 
> On 3/9/17, 10:22 AM, "Harbs"  wrote:
> 
>>> You can turn off all "as" output in the cross-compile via
>>> -skipAsCoercions
>> 
>> Only if you are compiling your own app — and you never rely on “as”
>> casting.
>> 
>> That doesn’t help if you are writing a library.
> 
> I guess I don't understand.  IMO, either your code needs coercion at
> runtime or it doesn't.  Why is a Library different from an App in this
> case?
> 
> Thanks,
> -Alex
> 

Re: [FalconJX] regex patterns

[Harbs]

Here’s one of them:

static private const anyPrintChar:RegExp = /[^\u0009\u000a\u000d\u0020]/g; 

outputs:

org.apache.flex.textLayout.elements.SpanElement.anyPrintChar = /[^  

 ]/g;

> On Mar 9, 2017, at 8:20 PM, Harbs  wrote:
> 
> Yes. These are from TLF. The strings are evaluated by Java. There’s no error 
> except on the Javascript side because the string literals become non-sensical.
> 
> We had a similar issue when porting our app. I think we used 
> String.fromCharCode() to get around the problem there.
> 
> Basically, we need some way to tell the compiler that any strings used in 
> RegExp objects should come through bit-for-bit identical and not be evaluated 
> as Strings. I have no idea what’s involved in doing that in Java...
> 
>> On Mar 9, 2017, at 7:43 PM, Alex Harui  wrote:
>> 
>> Are these from TLF?  What error are you getting?  Falcon compiles TLF as
>> part of the integration tests.  Is it the JS output that is broken?
>> 
>> Thanks,
>> -Alex
>> 
>> On 3/9/17, 1:18 AM, "Harbs"  wrote:
>> 
>>> I’ve come across quite a few regex patterns which break compilation.
>>> Here’s some examples:
>>> 
>>> private static const _newLineTabPattern:RegExp = /[\u0009\u000a\u000d]/g;
>>> static private const brRegEx:RegExp = /\u2028/;
>>> private static const _newLineRegex:RegExp = /\u000A|\u000D\u000A?/g;
>>> public static const anyPrintChar:RegExp = /[^\u0009\u000a\u000d\u0020]/g;
>>> public static const attrRegex:RegExp =
>>> /\s+(\w+)(?:\s*=\s*(".*?"|'.*?'|[\w\.]+))?/sg;
>>> public static const tagRegex:RegExp =
>>> /<(\/?)(\w+)((?:\s+\w+(?:\s*=\s*(?:".*?"|'.*?'|[\w\.]+))?)*)\s*(\/?)>/sg;
>>> public static const stripRegex:RegExp =
>>> /|<\?(".*?"|'.*?'|[^>"']+)*>|"']+)*>/sg;
>>> 
>>> The last three have to do with unsupported flags, but the first four
>>> break simply because the compiler evaluates the strings and they become
>>> spaces and line breaks, etc.
>>> 
>>> What can we do to prevent the compiler from killing patterns during
>>> compilation?
>>> 
>>> Harbs
>> 
> 

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Alex Harui]

Looks like Maven is reporting success, but it hasn't succeeded.  Various
SWCs are failing to build because of errors, but the root cause seems to
be the missing .pbj files in the Graphics.swc.

So, I think we need to figure out how to get those pbj files in a Maven
way.  Can we just stick the release artifacts into Maven Central manually?

-Alex

On 3/9/17, 9:50 AM, "carlos.rov...@gmail.com on behalf of Carlos Rovira"
 wrote:

>Hi, just see this thread, and update from repos and build all and seems
>working ok
>I'm a bit off due to work load, so was worried, but seems all is ok...just
>fixed minutes ago? :)
>
>2017-03-09 16:39 GMT+01:00 piotrz :
>
>> Hi Chris!
>>
>> I got you :) Was waiting for the proper thread in order to complain
>>about
>> something. I see after some changes in compiler (maybe) that some
>>modules
>> are failing but in general build didn't fail.
>>
>> Example:
>> http://apacheflexbuild.cloudapp.net:8080/job/flex-
>> asjs%20(maven)/161/console
>>
>> The second thing why this errors are happened ?
>>
>> Thanks,
>> Pete
>>
>>
>>
>>
>>
>> -
>> Apache Flex PMC
>> piotrzarzyck...@gmail.com
>> --
>> View this message in context: http://apache-flex-
>> development.247.n4.nabble.com/FlexJS-Maven-Jenkins-What-
>> s-up-with-framework-build-tp60281p60284.html
>> Sent from the Apache Flex Development mailing list archive at
>>Nabble.com.
>>
>
>
>
>-- 
>
>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 vigente Ley Orgánica de Protección de Datos (15/1999), le
>comunicamos
>que sus datos forman parte de un fichero cuyo responsable es CODEOSCOPIC
>S.A. La finalidad de dicho tratamiento es facilitar la prestación del
>servicio o información solicitados, teniendo usted derecho de acceso,
>rectificación, cancelación y oposición de sus datos dirigiéndose a
>nuestras
>oficinas c/ Paseo de la Habana 9-11, 28036, Madrid con la documentación
>necesaria.

Re: [FalconJX] casting idea

[Alex Harui]

On 3/9/17, 10:22 AM, "Harbs"  wrote:

>> You can turn off all "as" output in the cross-compile via
>>-skipAsCoercions
>
>Only if you are compiling your own app — and you never rely on “as”
>casting.
>
>That doesn’t help if you are writing a library.

I guess I don't understand.  IMO, either your code needs coercion at
runtime or it doesn't.  Why is a Library different from an App in this
case?

Thanks,
-Alex

Re: [FalconJX] casting idea

[Harbs]

> You can turn off all "as" output in the cross-compile via -skipAsCoercions

Only if you are compiling your own app — and you never rely on “as” casting.

That doesn’t help if you are writing a library.

> Nor do I want to be the person who implements it.

Can’t say I blame you there… ;-)

> On Mar 9, 2017, at 8:14 PM, Alex Harui  wrote:
> 
> You can turn off all "as" output in the cross-compile via -skipAsCoercions
> compiler option.  Then you shouldn't need @flexjsignorecoercion.  And then
> you use @flexjsemitcoercion where you actually need it.
> 
> Introducing new keywords makes me nervous, but I am not opposed to it.
> Nor do I want to be the person who implements it.
> 
> My 2 cents,
> -Alex
> 
> On 3/9/17, 9:37 AM, "Harbs"  wrote:
> 
>> It’s really a drag to constantly put in @flexjsignorecoercion every time
>> I’m using “as” to make the compiler happy. 99% of the time I’m not
>> interested in Language.as being added to the code.
>> 
>> Here’s a thought: Can we add a new operator for non-essential casts?
>> 
>> Something like this:
>> 
>> var foo:Thing = item as Thing would cross-compiler to var foo =
>> Language.as(item,Thing)
>> 
>> but var foo:Thing = item asa Thing would just compile as “var foo =
>> thing;” (and on the SWF side, “asa” would become a regular “as”).
>> 
>> Thoughts?
>> 
>> Harbs
> 

Re: [FalconJX] regex patterns

[Harbs]

Yes. These are from TLF. The strings are evaluated by Java. There’s no error 
except on the Javascript side because the string literals become non-sensical.

We had a similar issue when porting our app. I think we used 
String.fromCharCode() to get around the problem there.

Basically, we need some way to tell the compiler that any strings used in 
RegExp objects should come through bit-for-bit identical and not be evaluated 
as Strings. I have no idea what’s involved in doing that in Java...

> On Mar 9, 2017, at 7:43 PM, Alex Harui  wrote:
> 
> Are these from TLF?  What error are you getting?  Falcon compiles TLF as
> part of the integration tests.  Is it the JS output that is broken?
> 
> Thanks,
> -Alex
> 
> On 3/9/17, 1:18 AM, "Harbs"  wrote:
> 
>> I’ve come across quite a few regex patterns which break compilation.
>> Here’s some examples:
>> 
>> private static const _newLineTabPattern:RegExp = /[\u0009\u000a\u000d]/g;
>> static private const brRegEx:RegExp = /\u2028/;
>> private static const _newLineRegex:RegExp = /\u000A|\u000D\u000A?/g;
>> public static const anyPrintChar:RegExp = /[^\u0009\u000a\u000d\u0020]/g;
>> public static const attrRegex:RegExp =
>> /\s+(\w+)(?:\s*=\s*(".*?"|'.*?'|[\w\.]+))?/sg;
>> public static const tagRegex:RegExp =
>> /<(\/?)(\w+)((?:\s+\w+(?:\s*=\s*(?:".*?"|'.*?'|[\w\.]+))?)*)\s*(\/?)>/sg;
>> public static const stripRegex:RegExp =
>> /|<\?(".*?"|'.*?'|[^>"']+)*>|"']+)*>/sg;
>> 
>> The last three have to do with unsupported flags, but the first four
>> break simply because the compiler evaluates the strings and they become
>> spaces and line breaks, etc.
>> 
>> What can we do to prevent the compiler from killing patterns during
>> compilation?
>> 
>> Harbs
> 

Re: [FalconJX] casting idea

[Alex Harui]

You can turn off all "as" output in the cross-compile via -skipAsCoercions
compiler option.  Then you shouldn't need @flexjsignorecoercion.  And then
you use @flexjsemitcoercion where you actually need it.

Introducing new keywords makes me nervous, but I am not opposed to it.
Nor do I want to be the person who implements it.

My 2 cents,
-Alex

On 3/9/17, 9:37 AM, "Harbs"  wrote:

>It’s really a drag to constantly put in @flexjsignorecoercion every time
>I’m using “as” to make the compiler happy. 99% of the time I’m not
>interested in Language.as being added to the code.
>
>Here’s a thought: Can we add a new operator for non-essential casts?
>
>Something like this:
>
>var foo:Thing = item as Thing would cross-compiler to var foo =
>Language.as(item,Thing)
>
>but var foo:Thing = item asa Thing would just compile as “var foo =
>thing;” (and on the SWF side, “asa” would become a regular “as”).
>
>Thoughts?
>
>Harbs

Re: [FlexJS] Panel

[Alex Harui]

On 3/9/17, 8:49 AM, "Peter Ent"  wrote:

>The idea of "chrome" is that these are components that do not count as
>regular children and have special locations within the component. The
>TitleBar and ControlBar are considered chrome because if you ask for the
>children of the Panel, they are not listed and if you have so many
>children in the Panel that it requires scrolling, these items do not
>scroll. You can consider the fixed scrollbars in the SWF version to be
>chrome pieces too.

Before we go too far into whether we need Ichrome or separate slots, I'm
curious about what coordinate space expectations folks have as well.

In regular Flex, the pixel below the Panel's titlebar is 0,0 instead of 0,
titlebar.height because the content area is its own container.  Who has
used Panel-like UI's in JS and what coordinate space did it use for
content children?

Thanks,
-Alex

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[Carlos Rovira]

Hi, just see this thread, and update from repos and build all and seems
working ok
I'm a bit off due to work load, so was worried, but seems all is ok...just
fixed minutes ago? :)

2017-03-09 16:39 GMT+01:00 piotrz :

> Hi Chris!
>
> I got you :) Was waiting for the proper thread in order to complain about
> something. I see after some changes in compiler (maybe) that some modules
> are failing but in general build didn't fail.
>
> Example:
> http://apacheflexbuild.cloudapp.net:8080/job/flex-
> asjs%20(maven)/161/console
>
> The second thing why this errors are happened ?
>
> Thanks,
> Pete
>
>
>
>
>
> -
> Apache Flex PMC
> piotrzarzyck...@gmail.com
> --
> View this message in context: http://apache-flex-
> development.247.n4.nabble.com/FlexJS-Maven-Jenkins-What-
> s-up-with-framework-build-tp60281p60284.html
> Sent from the Apache Flex Development mailing list archive at Nabble.com.
>



-- 

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 vigente Ley Orgánica de Protección de Datos (15/1999), le comunicamos
que sus datos forman parte de un fichero cuyo responsable es CODEOSCOPIC
S.A. La finalidad de dicho tratamiento es facilitar la prestación del
servicio o información solicitados, teniendo usted derecho de acceso,
rectificación, cancelación y oposición de sus datos dirigiéndose a nuestras
oficinas c/ Paseo de la Habana 9-11, 28036, Madrid con la documentación
necesaria.

Re: [FalconJX] regex patterns

[Alex Harui]

Are these from TLF?  What error are you getting?  Falcon compiles TLF as
part of the integration tests.  Is it the JS output that is broken?

Thanks,
-Alex

On 3/9/17, 1:18 AM, "Harbs"  wrote:

>I’ve come across quite a few regex patterns which break compilation.
>Here’s some examples:
>
>private static const _newLineTabPattern:RegExp = /[\u0009\u000a\u000d]/g;
>static private const brRegEx:RegExp = /\u2028/;
>private static const _newLineRegex:RegExp = /\u000A|\u000D\u000A?/g;
>public static const anyPrintChar:RegExp = /[^\u0009\u000a\u000d\u0020]/g;
>public static const attrRegex:RegExp =
>/\s+(\w+)(?:\s*=\s*(".*?"|'.*?'|[\w\.]+))?/sg;
>public static const tagRegex:RegExp =
>/<(\/?)(\w+)((?:\s+\w+(?:\s*=\s*(?:".*?"|'.*?'|[\w\.]+))?)*)\s*(\/?)>/sg;
>public static const stripRegex:RegExp =
>/|<\?(".*?"|'.*?'|[^>"']+)*>|"']+)*>/sg;
>
>The last three have to do with unsupported flags, but the first four
>break simply because the compiler evaluates the strings and they become
>spaces and line breaks, etc.
>
>What can we do to prevent the compiler from killing patterns during
>compilation?
>
>Harbs

Re: [FlaconJX] problems with class references

[Alex Harui]

On 3/9/17, 12:33 AM, "Harbs"  wrote:

>I have already removed tlf_internal from the entire code base.
>
>I’m not sure where the QName is coming from. The importer classes have
>namespace declarations, maybe that’s throwing a spanner into the works?

Not sure what you mean by "importer classes have namespace declarations".
What does that code look like?

Are there any occurrences of two colons "::" in the source file?

Thanks,
-Alex

[FalconJX] casting idea

[Harbs]

It’s really a drag to constantly put in @flexjsignorecoercion every time I’m 
using “as” to make the compiler happy. 99% of the time I’m not interested in 
Language.as being added to the code.

Here’s a thought: Can we add a new operator for non-essential casts?

Something like this:

var foo:Thing = item as Thing would cross-compiler to var foo = 
Language.as(item,Thing)

but var foo:Thing = item asa Thing would just compile as “var foo = thing;” 
(and on the SWF side, “asa” would become a regular “as”).

Thoughts?

Harbs

Re: git commit: [flex-falcon] [refs/heads/develop] - fix interface override checking

[Alex Harui]

On 3/9/17, 12:51 AM, "Piotr Zarzycki"  wrote:

>Hi Alex,
>
>I'm not sure what lately has changed in the compiler, but in MDL and
>Mobile
>there are a lot of weard errors. Maven build passes, but it shouldn't.
>
>Any thoughts on that?

Hmm.  I just ran the build in flex-asjs develop branch using the
flex-falcon develop branch and everything, including examples, compiled
cleanly.

What errors are you getting?  Make sure you aren't using one of the "dual"
branches against a "develop" branch.  It has to be both "dual" or both
"develop".

HTH,
-Alex

Re: [FlexJS] Panel

[Peter Ent]

The idea of "chrome" is that these are components that do not count as
regular children and have special locations within the component. The
TitleBar and ControlBar are considered chrome because if you ask for the
children of the Panel, they are not listed and if you have so many
children in the Panel that it requires scrolling, these items do not
scroll. You can consider the fixed scrollbars in the SWF version to be
chrome pieces too.

Let's say you wanted to have a special type of title bar that did have
images and other buttons. If your component implements IChrome, then you
could add it to a container that knew how to handle chrome:


   
   


Where your MyTitleBar.mxml might be:



   




Then the ChromedContainer would place the MyTitleBar at the top of the
container outside of the content area. But you could also do this as:



   



   
 
   
   
 


This would accomplish the same thing. The problem with "chrome" is that
you could put the chrome into the middle of the container and then I'm not
sure how the layout algorithm would handle that.

‹peter


On 3/9/17, 11:16 AM, "PKumar"  wrote:

>ChromeContainer should be fine. We should be able to modify the panel
>title
>bar style as header corlor , border etc.
>
>Can Panel title bar contain other controls as image , button , check box?
>
>
>On 09-Mar-2017 9:37 PM, "piotrz [via Apache Flex Development]" <
>ml-node+s247n60286...@n4.nabble.com> wrote:
>
>Hi Peter,
>
>Let's say that you reconstructed Panel using new Group container - How can
>we use there Chrome ?
>
>I would go with ChromedContainer.
>
>Piotr
>Apache Flex PMC
>piotrzarzyck...@gmail.com
>
>
>--
>If you reply to this email, your message will be added to the discussion
>below:
>http://apache-flex-development.247.n4.nabble.com/FlexJS-Panel-
>tp60285p60286.html
>To start a new topic under Apache Flex Development, email
>ml-node+s247n1...@n4.nabble.com
>To unsubscribe from Apache Flex Development, click here
>.jtp?macro=unsubscribe_by_code=1=cHJhc2hha3VtYXJAZ21haWwuY29tfDF
>8LTU0MTcyMzE2NA==>
>.
>NAML
>.jtp?macro=macro_viewer=instant_html%21nabble%3Aemail.naml=nabble.
>naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-na
>bble.view.web.template.NodeNamespace=notify_subscribers%21nabb
>le%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21
>nabble%3Aemail.naml>
>
>
>
>
>--
>View this message in context:
>http://apache-flex-development.247.n4.nabble.com/FlexJS-Panel-tp60285p
>60287.html
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] Panel

[PKumar]

ChromeContainer should be fine. We should be able to modify the panel title
bar style as header corlor , border etc.

Can Panel title bar contain other controls as image , button , check box?


On 09-Mar-2017 9:37 PM, "piotrz [via Apache Flex Development]" <
ml-node+s247n60286...@n4.nabble.com> wrote:

Hi Peter,

Let's say that you reconstructed Panel using new Group container - How can
we use there Chrome ?

I would go with ChromedContainer.

Piotr
Apache Flex PMC
piotrzarzyck...@gmail.com


--
If you reply to this email, your message will be added to the discussion
below:
http://apache-flex-development.247.n4.nabble.com/FlexJS-Panel-
tp60285p60286.html
To start a new topic under Apache Flex Development, email
ml-node+s247n1...@n4.nabble.com
To unsubscribe from Apache Flex Development, click here

.
NAML





--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Panel-tp60285p60287.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] Panel

[piotrz]

Hi Peter,

Let's say that you reconstructed Panel using new Group container - How can
we use there Chrome ?

I would go with ChromedContainer.

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Panel-tp60285p60286.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS] [Maven] [Jenkins] What's up with framework build?

[piotrz]

Hi Chris!

I got you :) Was waiting for the proper thread in order to complain about
something. I see after some changes in compiler (maybe) that some modules
are failing but in general build didn't fail. 

Example:
http://apacheflexbuild.cloudapp.net:8080/job/flex-asjs%20(maven)/161/console

The second thing why this errors are happened ?

Thanks,
Pete





-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-Jenkins-What-s-up-with-framework-build-tp60281p60284.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

[FlexJS] Panel

[Peter Ent]

Hi,

I'm trying to determine if "chrome" is a necessity as part of Container or if a 
specialty "ChromedContainer" should be made (or if at all).

Panel is one component that uses chrome (the TitleBar and ControlBar implement 
IChrome). Is Panel used a lot?

As an alternative, Panel can be re-constructed to look as it does now without 
the use of chrome simply by using Group for the Panel itself and Group for the 
TitleBar and ControlBar and a Container with a ScrollingViewport for the 
content.

Please let me know your thoughts.

Thanks,
Peter Ent
Adobe Systems/Apache Flex Project

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Harbs]

Yes we do.

Yishay created a BlendBead and the pixel bender files are needed to get the 
blends to render correctly in Flash.

> On Mar 9, 2017, at 5:09 PM, Christofer Dutz  wrote:
> 
> Hi Yishay,
> 
> Yeah … but why do we need them in FlexJS now? I was quite happy with not 
> needing stuff like that.
> 
> Chris
> 
> 
> Am 09.03.17, 15:27 schrieb "yishayw" :
> 
>PBJ files are a compiled version of a Pixel Bender script that can be 
> loaded
>by the Flash Player [1].
> 
>As far as I understand it's a proprietary Adobe format which requires a
>proprietary tool, which is why we want to remove them from our repo. We're
>using them to implement blend-mode styles in flash.
> 
>I recently added them to downloads.xml, which is a file containing a bunch
>of ant tasks for adding external dependencies that need to be downloaded.
>We're wondering how the same trick could be done for Maven. 
> 
> 
>[1] http://ncannasse.fr/projects/pbj
> 
> 
> 
>--
>View this message in context: 
> http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-New-Upstream-Dependency-was-Re-git-commit-flex-asjs-refs-heads-develop-Added-Shader-fil-tp60260p60279.html
>Sent from the Apache Flex Development mailing list archive at Nabble.com.
> 
> 

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Christofer Dutz]

Hi Yishay,

Yeah … but why do we need them in FlexJS now? I was quite happy with not 
needing stuff like that.

Chris


Am 09.03.17, 15:27 schrieb "yishayw" :

PBJ files are a compiled version of a Pixel Bender script that can be loaded
by the Flash Player [1].

As far as I understand it's a proprietary Adobe format which requires a
proprietary tool, which is why we want to remove them from our repo. We're
using them to implement blend-mode styles in flash.

I recently added them to downloads.xml, which is a file containing a bunch
of ant tasks for adding external dependencies that need to be downloaded.
We're wondering how the same trick could be done for Maven. 
 

[1] http://ncannasse.fr/projects/pbj



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-New-Upstream-Dependency-was-Re-git-commit-flex-asjs-refs-heads-develop-Added-Shader-fil-tp60260p60279.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

[FlexJS] [Maven] [Jenkins] What's up with framework build?

[Christofer Dutz]

Hi,

So, it seems we can’t build now ... anyone got an idea why? Anyone working on 
fixing that?

Chris

Re: [compc] error # 5? 6? 7? ;-)

[Harbs]

I pulled the changes and built falcon. The fix seems to work. :-)

Thanks!

> On Mar 9, 2017, at 10:30 AM, Harbs  wrote:
> 
> Cool. Thanks.
> 
>> On Mar 9, 2017, at 9:06 AM, Alex Harui  wrote:
>> 
>> OK, pushed a fix for this.  IMO, it was a too-sensitive check in Falcon.
>> It is allowed in SWF.
>> 
>> -Alex
>> 
>> On 3/8/17, 3:18 PM, "Harbs"  wrote:
>> 
>>> FWIW, I’m using FDT, and it does not report an error with the interfaces.
>>> (It’s only reported when I actually compile it.) I don’t know if that’s
>>> an indication that it’s okay with Flash, but I’d guess that yes.
>>> 
 On Mar 9, 2017, at 1:10 AM, Alex Harui  wrote:
 
 
 
 On 3/8/17, 11:29 AM, "Harbs"  wrote:
 
> In my quest to remove circular dependencies, I’ve created a whole slew
> of
> interfaces and I’ve come across an interesting problem:
> 
> This works correctly when dealing with concrete classes. For some
> reason
> when done with interfaces, I get the error above.
> 
> Is this an easy fix?
 
 Hard to say.  One good test would be to see if this works with the
 regular
 Flex SDK MXMLC.  There's also a question about whether the runtime
 allows
 this or not.  I don't see why it shouldn't work.  If it is just that
 Falcon is too picky, that should be "easy".
 
 -Alex
 
>>> 
>> 
> 

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[yishayw]

PBJ files are a compiled version of a Pixel Bender script that can be loaded
by the Flash Player [1].

As far as I understand it's a proprietary Adobe format which requires a
proprietary tool, which is why we want to remove them from our repo. We're
using them to implement blend-mode styles in flash.

I recently added them to downloads.xml, which is a file containing a bunch
of ant tasks for adding external dependencies that need to be downloaded.
We're wondering how the same trick could be done for Maven. 
 

[1] http://ncannasse.fr/projects/pbj



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Maven-New-Upstream-Dependency-was-Re-git-commit-flex-asjs-refs-heads-develop-Added-Shader-fil-tp60260p60279.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS][Maven] New Upstream Dependency (was Re: git commit: [flex-asjs] [refs/heads/develop] - Added Shader files (with compiler argument)

[Christofer Dutz]

Ok … the renaming trick worked ;-)

Could you please explain what these pbj files are?

We could put stuff in maven central by officially releasing them via Apache, 
but are they Apache content? If not we should use Sonatypes OSS repo to 
manually release something (if we are allowed to do that).

Chris



Am 09.03.17, 00:33 schrieb "Alex Harui" :

Changing subject to get Christofer's attention.

I'm thinking that we should simply submit the pixel bender release zips to
Maven Central.  I don't know if the Mavenizer can help, or if there is
some other solution.

For those who haven't been following, the goal is to use .pbj files found
in these artifacts:
https://dist.apache.org/repos/dist/release/flex/pixelbender/1.0/

Thanks,
-Alex


On 3/8/17, 7:07 AM, "yishayw"  wrote:

>Fixed the Ant script and removed the pbj files. Waiting for advice on
>Maven
>(where Jenkins is failing).
>
>
>
>--
>View this message in context:
>http://apache-flex-development.247.n4.nabble.com/Re-git-commit-flex-as
>js-refs-heads-develop-Added-Shader-files-with-compiler-argument-tp60217p60
>254.html
>Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: [FlexJS, MDL] Ability to create dynamic components

[piotrz]

Hi Carlos,

Link to the thread [1] and I will describe it more later today, but from the
thread you can also understand something.

[1]
http://apache-flex-development.247.n4.nabble.com/FlexJS-MDL-Dynamic-Child-Problem-td59595.html

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-MDL-Ability-to-create-dynamic-components-tp60255p60277.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.

[FalconJX] regex patterns

[Harbs]

I’ve come across quite a few regex patterns which break compilation. Here’s 
some examples:

private static const _newLineTabPattern:RegExp = /[\u0009\u000a\u000d]/g;
static private const brRegEx:RegExp = /\u2028/;
private static const _newLineRegex:RegExp = /\u000A|\u000D\u000A?/g;
public static const anyPrintChar:RegExp = /[^\u0009\u000a\u000d\u0020]/g;
public static const attrRegex:RegExp = 
/\s+(\w+)(?:\s*=\s*(".*?"|'.*?'|[\w\.]+))?/sg;
public static const tagRegex:RegExp = 
/<(\/?)(\w+)((?:\s+\w+(?:\s*=\s*(?:".*?"|'.*?'|[\w\.]+))?)*)\s*(\/?)>/sg;
public static const stripRegex:RegExp = 
/|<\?(".*?"|'.*?'|[^>"']+)*>|"']+)*>/sg;

The last three have to do with unsupported flags, but the first four break 
simply because the compiler evaluates the strings and they become spaces and 
line breaks, etc.

What can we do to prevent the compiler from killing patterns during compilation?

Harbs

Re: git commit: [flex-falcon] [refs/heads/develop] - fix interface override checking

[Piotr Zarzycki]

Hi Alex,

I'm not sure what lately has changed in the compiler, but in MDL and Mobile
there are a lot of weard errors. Maven build passes, but it shouldn't.

Any thoughts on that?

Thanks,
Piotr

W dniu czwartek, 9 marca 2017  napisał(a):

> Repository: flex-falcon
> Updated Branches:
>   refs/heads/develop 6e14d6835 -> dad773a21
>
>
> fix interface override checking
>
>
> Project: http://git-wip-us.apache.org/repos/asf/flex-falcon/repo
> Commit: http://git-wip-us.apache.org/repos/asf/flex-falcon/commit/dad773a2
> Tree: http://git-wip-us.apache.org/repos/asf/flex-falcon/tree/dad773a2
> Diff: http://git-wip-us.apache.org/repos/asf/flex-falcon/diff/dad773a2
>
> Branch: refs/heads/develop
> Commit: dad773a2193b5a80aee1e5a5f82d30b82ac19431
> Parents: 6e14d68
> Author: Alex Harui >
> Authored: Wed Mar 8 22:42:14 2017 -0800
> Committer: Alex Harui >
> Committed: Wed Mar 8 22:42:14 2017 -0800
>
> --
>  .../semantics/MethodBodySemanticChecker.java|  7 +++
>  .../src/test/java/as/ASInheritanceTests.java| 59 
>  2 files changed, 66 insertions(+)
> --
>
>
> http://git-wip-us.apache.org/repos/asf/flex-falcon/blob/
> dad773a2/compiler/src/main/java/org/apache/flex/compiler/
> internal/semantics/MethodBodySemanticChecker.java
> --
> diff --git a/compiler/src/main/java/org/apache/flex/compiler/internal/
> semantics/MethodBodySemanticChecker.java b/compiler/src/main/java/org/
> apache/flex/compiler/internal/semantics/MethodBodySemanticChecker.java
> index 1b51727..3904823 100644
> --- a/compiler/src/main/java/org/apache/flex/compiler/internal/semantics/
> MethodBodySemanticChecker.java
> +++ b/compiler/src/main/java/org/apache/flex/compiler/internal/semantics/
> MethodBodySemanticChecker.java
> @@ -2799,7 +2799,14 @@ public class MethodBodySemanticChecker
>  if( conflicts.size() > 0 )
>  {
>  for( IFunctionDefinition overriden : conflicts )
> +{
> +   if ((overriden instanceof SetterDefinition &&
> +   funcDef instanceof GetterDefinition) ||
> +   (overriden instanceof GetterDefinition &&
> +   funcDef instanceof SetterDefinition))
> +   continue;
>  addProblem(new InterfaceMethodOverrideProblem(iNode,
> funcDef.getBaseName(), overriden.getParent().getBaseName()));
> +}
>  }
>
>  }
>
> http://git-wip-us.apache.org/repos/asf/flex-falcon/blob/
> dad773a2/compiler/src/test/java/as/ASInheritanceTests.java
> --
> diff --git a/compiler/src/test/java/as/ASInheritanceTests.java
> b/compiler/src/test/java/as/ASInheritanceTests.java
> index 4d1d393..b5833e1 100644
> --- a/compiler/src/test/java/as/ASInheritanceTests.java
> +++ b/compiler/src/test/java/as/ASInheritanceTests.java
> @@ -83,4 +83,63 @@ public class ASInheritanceTests  extends
> ASFeatureTestsBase{
>
>  compileAndExpectErrors(source, false,false,false, null,"No
> default constructor found in base class A.\n");
>  }
> +
> +@Test
> +public void InterfaceOverrideError()
> +{
> +// all tests can assume that flash.display.Sprite
> +// flash.system.System and flash.events.Event have been imported
> +String[] imports = new String[]
> +{
> +};
> +String[] declarations = new String[]
> +{
> +};
> +String[] testCode = new String[]
> +{
> +
> +};
> +String[] extra = new String[]
> +{
> +"interface A {",
> +"function get text():String;",
> +"}",
> +"interface B extends A {",
> +"function get text():String;",
> +"}"
> +};
> +String source = getAS(imports, declarations, testCode, extra);
> +
> +compileAndExpectErrors(source, false,false,false, null,"Cannot
> override an interface method.  Method text conflicts with a method in base
> interface A.\n");
> +}
> +
> +@Test
> +public void InterfaceOverrideOK()
> +{
> +// all tests can assume that flash.display.Sprite
> +// flash.system.System and flash.events.Event have been imported
> +String[] imports = new String[]
> +{
> +};
> +String[] declarations = new String[]
> +{
> +};
> +String[] testCode = new String[]
> +{
> +
> +};
> +String[] extra = new String[]
> +{

Re: Just need to vent...

[Harbs]

I already using remove-circulars and it’s not helping in this case. If you’re 
interested here’s the output: https://paste.apache.org/Kd9l

And running the debug in the browser has tons of runtime errors.

I had this problem when porting my app as well. Basically static properties 
really mess things up, and TLF has lots of static stuff.

I managed to get rid of some issues with the interfaces, but it could be I 
should have been spending more time getting rid of some of the static 
properties…

> On Mar 9, 2017, at 3:10 AM, Alex Harui  wrote:
> 
> 
> 
> On 3/8/17, 5:05 PM, "Harbs"  wrote:
> 
>> The only other option is to just drop TLF and that’s way more work (I
>> think).
>> 
>> As it is, it cannot be compiled by the Google Compiler.
>> 
>> There’s lots of superclass dependencies on subclasses.
>> 
>> The good news is that I’m making it a lot more reliant on interfaces
>> which is probably a good thing.
> 
> More reliance on interfaces is a good thing, but I don't know why we can't
> just make folks use -remove-circulars.
> 
> It is "ok" to have superclass dependencies on subclasses as long as they
> aren't true circularities.  If they are tests for a particular type,
> inside method code, it shouldn't matter.  We just have to get the set of
> requires right.  Or is there some case that is truly impossible?
> 
> Thanks,
> -Alex
> 

Re: [FlaconJX] problems with class references

[Harbs]

I have already removed tlf_internal from the entire code base.

I’m not sure where the QName is coming from. The importer classes have 
namespace declarations, maybe that’s throwing a spanner into the works?

I don’t see this as a very high priority issue if simple cases work.

> On Mar 9, 2017, at 7:35 AM, Alex Harui  wrote:
> 
> I can't seem to reproduce this in a small test. In theory, the only code
> that outputs "new QName" would be in response to a namespace token like
> mx_internal::someProperty.
> 
> I see in original TLF code the call has "tlf_internal::description", but
> in your example, it is just "description".  Can you delete the output to
> make sure the compiler really is generating "new Qname" from just
> "description"?
> 
> Thanks,
> -Alex
> 
> On 3/8/17, 5:19 PM, "Alex Harui"  wrote:
> 
>> It appears there is more to it.  The class that owns memberType also has a
>> tlf_internal::description field which is confusing the fetching of
>> memberType.description.
>> 
>> Are you going to be leveraging tlf_internal namespaces in the port?  That
>> hasn't been fully debugged and will not generated small code.
>> 
>> -Alex
>> 
>> On 3/8/17, 3:28 PM, "Alex Harui"  wrote:
>> 
>>> 
>>> 
>>> On 3/8/17, 2:48 AM, "Harbs"  wrote:
>>> 
 Please look at this paste:
 https://paste.apache.org/JfGh 
 
 Changing the type of memberType to “*” resolves this issue for me, but
 the question is whether this is a bug which should be fixed, or a
 limitation that requires a work-around.
>>> 
>>> That's a bug.  "public" should never end up in a Qname.  I'll see if I
>>> can
>>> reproduce it.
>>> 
>>> Thanks,
>>> -Alex
>>> 
>> 
> 

Re: [compc] error # 5? 6? 7? ;-)

[Harbs]

Cool. Thanks.

> On Mar 9, 2017, at 9:06 AM, Alex Harui  wrote:
> 
> OK, pushed a fix for this.  IMO, it was a too-sensitive check in Falcon.
> It is allowed in SWF.
> 
> -Alex
> 
> On 3/8/17, 3:18 PM, "Harbs"  wrote:
> 
>> FWIW, I’m using FDT, and it does not report an error with the interfaces.
>> (It’s only reported when I actually compile it.) I don’t know if that’s
>> an indication that it’s okay with Flash, but I’d guess that yes.
>> 
>>> On Mar 9, 2017, at 1:10 AM, Alex Harui  wrote:
>>> 
>>> 
>>> 
>>> On 3/8/17, 11:29 AM, "Harbs"  wrote:
>>> 
 In my quest to remove circular dependencies, I’ve created a whole slew
 of
 interfaces and I’ve come across an interesting problem:
 
 This works correctly when dealing with concrete classes. For some
 reason
 when done with interfaces, I get the error above.
 
 Is this an easy fix?
>>> 
>>> Hard to say.  One good test would be to see if this works with the
>>> regular
>>> Flex SDK MXMLC.  There's also a question about whether the runtime
>>> allows
>>> this or not.  I don't see why it shouldn't work.  If it is just that
>>> Falcon is too picky, that should be "easy".
>>> 
>>> -Alex
>>> 
>> 
>