Re: 0.9.6 Release

[Piotr Zarzycki]

Hi Alex,

Thank you! I will be working in release next week, so if anyone have
anything to add there is some time.

Thanks,
Piotr

On Tue, Jun 18, 2019, 7:28 AM Alex Harui  wrote:

> Mx tests are working locally for me with this last change to
> ListItemRenderer.
>
> So, the question for everyone is:  is there anything else we want to put
> in before trying to cut a release?
>
> We might need to change the ASDoc app after Greg's last change.  We might
> want to bang on ASDoc for a bit and see if anything major needs fixing
> before cutting a release.
>
> The only other thing I thought of was trying to upgrade Selenium so we can
> use a recent FireFox version.  Right now we are stuck way back on FF47, but
> I've also heard that more recent FF versions are less friendly to Flash, so
> you have to use some other browser to test Flash.  That's ok for building
> the Royale SDK since the Flash tests are run in FlashPlayerDebugger and the
> JS tests are run in a browser like FF.
>
> I'm tempted to say we should punt the inject_html in modules problem to
> the next release.
>
> I will be adding more details to
> https://github.com/apache/royale-asjs/wiki/Release-Manager-Notes
> But there should be enough there to get an RC out for vote.
>
> Thanks,
> -Alex
>
> On 6/17/19, 11:06 AM, "Alex Harui"  wrote:
>
> I just saw ASDoc work well enough for me to start with
> mx.binding.BindabilityInfo and switch to mx.charts.AxisRenderer.
>
> There might be JSON parsing issues with other classes.  Run the
> bin/js-debug version, check the console if things don't load and see what
> file failed to load and why.
>
> I will take a look at MXTests next.
>
> -Alex
>
> On 6/17/19, 8:08 AM, "Alex Harui"  wrote:
>
> ASDoc is not fully working for me.  I've been working on that the
> past several days.  It took quite a while to fix the JSON output from new
> and interesting ASDoc patterns.  I'm now looking into failures due to
> changes in coercion rules in the JS output.  ASDoc was trying to save time
> by using plain objects that were isomorphic to a class instead of actual
> class instances.
>
> MXTests are failing on the CI server and probably locally but I
> haven't gotten that far.
>
> I would think we want these things to work, but you are the RM so
> it is up to you.  Volunteers are welcome to help with these and other tasks.
>
> -Alex
>
> On 6/17/19, 12:44 AM, "Piotr Zarzycki" 
> wrote:
>
> Hi Alex,
>
> I don't see any answer here - can I start with trying your
> instruction and
> start release ?
>
> Thanks,
> Piotr
>
> śr., 12 cze 2019 o 10:12 Carlos Rovira <
> carlosrov...@apache.org> napisał(a):
>
> > Hi Alex,
> > just build from scratch right now and seems maven and ant
> was ok
> >
> > El mié., 12 jun. 2019 a las 10:02, Alex Harui
> ()
> > escribió:
> >
> > > The merge is complete, and the main builds seem to be
> successful, except
> > > for Maven on royale-asjs which was failing before the
> merge.
> > >
> > > I'm still catching up on other things, but I think there
> are some bugs we
> > > need to fix before cutting the release?  I think ASDoc may
> not be fully
> > > working?  What else?
> > >
> > > -Alex
> > >
> > > On 6/10/19, 11:05 AM, "Piotr Zarzycki" <
> piotrzarzyck...@gmail.com>
> > wrote:
> > >
> > > Hi Alex,
> > >
> > > Many thanks for that! I will try to be RM. I will have
> some dedicated
> > > time
> > > for that. I will wait for your instruction and merge
> to develop.
> > >
> > > Thanks,
> > > Piotr
> > >
> > >
> > > On Mon, Jun 10, 2019, 7:31 PM Alex Harui
> 
> > > wrote:
> > >
> > > > Well, that turned out to be much more time-consuming
> than I
> > > expected, but
> > > > we can now create identical release artifacts on Mac
> and Win.  I am
> > > hopeful
> > > > this effort will pay off not only now in having
> other folks
> > generate
> > > > releases, but also in the future if signed binaries
> become a
> > > requirement.
> > > >
> > > > There continues to be a lot of distractions in my
> life that can
> > cause
> > > > delays, but I hope to merge the release_practice
> branches into
> > > develop over
> > > > the next day or two and figure out where in the wiki
> to document
> > the
> > > > release process.  So, now is the time for one or
> more people to

Jenkins build is back to normal : royale-asjs_MXTests #862

[Apache Royale CI Server]

See 

Re: 0.9.6 Release

[Alex Harui]

Mx tests are working locally for me with this last change to ListItemRenderer.

So, the question for everyone is:  is there anything else we want to put in 
before trying to cut a release?

We might need to change the ASDoc app after Greg's last change.  We might want 
to bang on ASDoc for a bit and see if anything major needs fixing before 
cutting a release.

The only other thing I thought of was trying to upgrade Selenium so we can use 
a recent FireFox version.  Right now we are stuck way back on FF47, but I've 
also heard that more recent FF versions are less friendly to Flash, so you have 
to use some other browser to test Flash.  That's ok for building the Royale SDK 
since the Flash tests are run in FlashPlayerDebugger and the JS tests are run 
in a browser like FF.

I'm tempted to say we should punt the inject_html in modules problem to the 
next release.

I will be adding more details to 
https://github.com/apache/royale-asjs/wiki/Release-Manager-Notes
But there should be enough there to get an RC out for vote.

Thanks,
-Alex

On 6/17/19, 11:06 AM, "Alex Harui"  wrote:

I just saw ASDoc work well enough for me to start with 
mx.binding.BindabilityInfo and switch to mx.charts.AxisRenderer.

There might be JSON parsing issues with other classes.  Run the 
bin/js-debug version, check the console if things don't load and see what file 
failed to load and why.

I will take a look at MXTests next.

-Alex

On 6/17/19, 8:08 AM, "Alex Harui"  wrote:

ASDoc is not fully working for me.  I've been working on that the past 
several days.  It took quite a while to fix the JSON output from new and 
interesting ASDoc patterns.  I'm now looking into failures due to changes in 
coercion rules in the JS output.  ASDoc was trying to save time by using plain 
objects that were isomorphic to a class instead of actual class instances.

MXTests are failing on the CI server and probably locally but I haven't 
gotten that far.

I would think we want these things to work, but you are the RM so it is 
up to you.  Volunteers are welcome to help with these and other tasks.

-Alex

On 6/17/19, 12:44 AM, "Piotr Zarzycki"  
wrote:

Hi Alex,

I don't see any answer here - can I start with trying your 
instruction and
start release ?

Thanks,
Piotr

śr., 12 cze 2019 o 10:12 Carlos Rovira  
napisał(a):

> Hi Alex,
> just build from scratch right now and seems maven and ant was ok
>
> El mié., 12 jun. 2019 a las 10:02, Alex Harui 
()
> escribió:
>
> > The merge is complete, and the main builds seem to be 
successful, except
> > for Maven on royale-asjs which was failing before the merge.
> >
> > I'm still catching up on other things, but I think there are 
some bugs we
> > need to fix before cutting the release?  I think ASDoc may not 
be fully
> > working?  What else?
> >
> > -Alex
> >
> > On 6/10/19, 11:05 AM, "Piotr Zarzycki" 

> wrote:
> >
> > Hi Alex,
> >
> > Many thanks for that! I will try to be RM. I will have some 
dedicated
> > time
> > for that. I will wait for your instruction and merge to 
develop.
> >
> > Thanks,
> > Piotr
> >
> >
> > On Mon, Jun 10, 2019, 7:31 PM Alex Harui 

> > wrote:
> >
> > > Well, that turned out to be much more time-consuming than 
I
> > expected, but
> > > we can now create identical release artifacts on Mac and 
Win.  I am
> > hopeful
> > > this effort will pay off not only now in having other 
folks
> generate
> > > releases, but also in the future if signed binaries 
become a
> > requirement.
> > >
> > > There continues to be a lot of distractions in my life 
that can
> cause
> > > delays, but I hope to merge the release_practice branches 
into
> > develop over
> > > the next day or two and figure out where in the wiki to 
document
> the
> > > release process.  So, now is the time for one or more 
people to
> step
> > up to
> > > be the RMs for 0.9.6 and help debug and improve the 
process.
> > >
> > > I am going to try very hard not to "own" the process.  If 
something
> > goes
> > > wrong, I am going to ask others to try to debug and fix 
it 

Jenkins build is back to normal : royale-asjs #2359

[Apache Royale CI Server]

See 

Build failed in Jenkins: royale-asjs #2358

[Apache Royale CI Server]

See 


--
[...truncated 707.90 KB...]
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

download-dependency-zip:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

check-dependency:
 [echo] checking for C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx/lib//slf4j-api.jar

download-dependency:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

echo-project-jar:
   [delete] Deleting: C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources\project.properties
 [echo] ${INFO_DOWNLOADING_FILE_FROM}
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

download-apache-license:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

download-other-license:
  [get] Getting: 
https://raw.githubusercontent.com/qos-ch/slf4j/master/LICENSE.txt
  [get] To: C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx\lib\slf4j-api-LICENSE.txt
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

double-check-file:
 [echo] ${env.ROYALE_DOWNLOAD_CACHE}
 [echo] Need file: ${still_no_file}

get-from-cache-if-needed:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

fail-if-not-found:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

download-dependency-jar:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

check-cache:

download-jar:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

get-if-not-cached:
  [get] Getting: 
https://repo1.maven.org/maven2/org/slf4j/slf4j-api/1.7.25/slf4j-api-1.7.25.jar
  [get] To: C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx\lib\slf4j-api.jar
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-tasks\src\main\resources
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx
 [echo] ROYALE_COMPILER_HOME is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\compiler-jx

double-check-file:
 [echo] ${env.ROYALE_DOWNLOAD_CACHE}
 [echo] Need file: ${still_no_file}

get-from-cache-if-needed:
 [echo] basedir is C:\Program Files 
(x86)\Jenkins\workspace\royale-compiler\royaleunit-ant-ta

Build failed in Jenkins: royale-asjs_MXTests #861

[Apache Royale CI Server]

See 


--
[...truncated 2.34 MB...]
[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

Suggestion to request Royale addition to Facebook's 'flash migration recommendations'

[Greg Dove]

fyi I got a ping today from Facebook today about flash migrations to html5.

If anyone knows how to contact them internally to request that they update
their options, this page [1] could do with a Royale mention, I believe. In
case you're wondering, yes, I know of cases where Flex was used in the past
for Facebook games... but even if it is not simply for porting old Flex
apps, I think the possible use of some js externs for a game engine could
make Royale a good option for new facebook projects... Anyhow I'm just
sharing this in case someone knows how to influence updates to their
developer info page and we can get a Royale mention...

1. https://developers.facebook.com/docs/games/gamesonfacebook/webtech

Re: [royale-asjs] 04/04: suppress implicit coercions because we cheat and use JSON objects instead of actual instances

[Greg Dove]

Hi Alex,

My apologies, but I may have messed this directive up a bit with the
earlier change from 'negative' to 'positive' semantics in the config
settings. I just added compiler tests for the output variations and
adjusted the logic for these checks to make them pass, so now the following
should work properly:

@royalesuppresscompleximplicitcoercion false
means 'don't suppress the coercion' (the coercion is on by default). This
would be unusual to use, but could be to used to locally override a
compilation config setting which suppresses it everywhere, so that it is
still output inside a particular method, for example. It sounds like it is
the opposite of the intent in your commit comment, but I guess you tried
this because the boolean toggles were not working correctly/intuitively.


if you want it suppressed locally, you should be able to use 'true' or use
it with no additional specifier, otherwise you can also use it in a similar
way to @royaleignorecoercion with specific types. This should work
correctly now after my fix.

for example:
 var classData:ASDocClass = masterData["filterData"][className];

you can use:
@royalesuppresscompleximplicitcoercion ASDocClass

but the following should (after the fix) suppress all implicit coercions in
the annotated method scope  :
@royalesuppresscompleximplicitcoercion
@royalesuppresscompleximplicitcoercion true

If you want it off completely for the current build, it is best done via
the config setting for the compilation, and then you need no annotations
anywhere.
I did provide some information about the config settings recently in a
post, but I'm going to write this stuff up this week and add it to the
docs. Sorry I should have done that already. I will get it done this week.

Greg


On Tue, Jun 18, 2019 at 6:02 AM  wrote:

> This is an automated email from the ASF dual-hosted git repository.
>
> aharui pushed a commit to branch develop
> in repository https://gitbox.apache.org/repos/asf/royale-asjs.git
>
> commit a09ea1e3a835226ea880ca2bfa7161e11ece22a5
> Author: Alex Harui 
> AuthorDate: Mon Jun 17 11:01:52 2019 -0700
>
> suppress implicit coercions because we cheat and use JSON objects
> instead of actual instances
> ---
>  examples/royale/ASDoc/src/main/royale/models/ASDocModel.as | 9 +
>  1 file changed, 9 insertions(+)
>
> diff --git a/examples/royale/ASDoc/src/main/royale/models/ASDocModel.as
> b/examples/royale/ASDoc/src/main/royale/models/ASDocModel.as
> index bb530d7..493b46e 100644
> --- a/examples/royale/ASDoc/src/main/royale/models/ASDocModel.as
> +++ b/examples/royale/ASDoc/src/main/royale/models/ASDocModel.as
> @@ -493,6 +493,9 @@ package models
>  }
>  }
>
> +/**
> + * @royalesuppresscompleximplicitcoercion false
> + */
>  private function addIfNeededAndMakeAttributes(arr:Array,
> data:ASDocClassEvents):void
>  {
> var n:int = arr.length;
> @@ -528,6 +531,9 @@ package models
> arr.push(data);
>  }
>
> +/**
> + * @royalesuppresscompleximplicitcoercion false
> + */
> private function addAttributes(dest:ASDocClassEvents,
> src:ASDocClassEvents):void
> {
> if (!src.tags) return;
> @@ -884,6 +890,9 @@ package models
>  filterPackageList();
>  }
>
> +/**
> + * @royalesuppresscompleximplicitcoercion false
> + */
>  public function filterByTags(className:String):Boolean
>  {
>  var classData:ASDocClass =
> masterData["filterData"][className];
>
>

Build failed in Jenkins: royale-asjs_MXTests #860

[Apache Royale CI Server]

See 


Changes:

[aharui] ASDoc found these small issues

[aharui] get asdoc to build with most if not all projects

[aharui] use parent instead of itemrendererparent which is now ListView instead

[aharui] suppress implicit coercions because we cheat and use JSON objects

--
[...truncated 2.34 MB...]
[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

Re: 0.9.6 Release

[Alex Harui]

I just saw ASDoc work well enough for me to start with 
mx.binding.BindabilityInfo and switch to mx.charts.AxisRenderer.

There might be JSON parsing issues with other classes.  Run the bin/js-debug 
version, check the console if things don't load and see what file failed to 
load and why.

I will take a look at MXTests next.

-Alex

On 6/17/19, 8:08 AM, "Alex Harui"  wrote:

ASDoc is not fully working for me.  I've been working on that the past 
several days.  It took quite a while to fix the JSON output from new and 
interesting ASDoc patterns.  I'm now looking into failures due to changes in 
coercion rules in the JS output.  ASDoc was trying to save time by using plain 
objects that were isomorphic to a class instead of actual class instances.

MXTests are failing on the CI server and probably locally but I haven't 
gotten that far.

I would think we want these things to work, but you are the RM so it is up 
to you.  Volunteers are welcome to help with these and other tasks.

-Alex

On 6/17/19, 12:44 AM, "Piotr Zarzycki"  wrote:

Hi Alex,

I don't see any answer here - can I start with trying your instruction 
and
start release ?

Thanks,
Piotr

śr., 12 cze 2019 o 10:12 Carlos Rovira  
napisał(a):

> Hi Alex,
> just build from scratch right now and seems maven and ant was ok
>
> El mié., 12 jun. 2019 a las 10:02, Alex Harui 
()
> escribió:
>
> > The merge is complete, and the main builds seem to be successful, 
except
> > for Maven on royale-asjs which was failing before the merge.
> >
> > I'm still catching up on other things, but I think there are some 
bugs we
> > need to fix before cutting the release?  I think ASDoc may not be 
fully
> > working?  What else?
> >
> > -Alex
> >
> > On 6/10/19, 11:05 AM, "Piotr Zarzycki" 
> wrote:
> >
> > Hi Alex,
> >
> > Many thanks for that! I will try to be RM. I will have some 
dedicated
> > time
> > for that. I will wait for your instruction and merge to develop.
> >
> > Thanks,
> > Piotr
> >
> >
> > On Mon, Jun 10, 2019, 7:31 PM Alex Harui 

> > wrote:
> >
> > > Well, that turned out to be much more time-consuming than I
> > expected, but
> > > we can now create identical release artifacts on Mac and Win. 
 I am
> > hopeful
> > > this effort will pay off not only now in having other folks
> generate
> > > releases, but also in the future if signed binaries become a
> > requirement.
> > >
> > > There continues to be a lot of distractions in my life that 
can
> cause
> > > delays, but I hope to merge the release_practice branches into
> > develop over
> > > the next day or two and figure out where in the wiki to 
document
> the
> > > release process.  So, now is the time for one or more people 
to
> step
> > up to
> > > be the RMs for 0.9.6 and help debug and improve the process.
> > >
> > > I am going to try very hard not to "own" the process.  If 
something
> > goes
> > > wrong, I am going to ask others to try to debug and fix it 
first
> > because it
> > > is in the project's best interests for others to truly 
understand
> > how this
> > > stuff works.
> > >
> > > Thanks,
> > > -Alex
> > >
> > > On 5/23/19, 9:54 AM, "Alex Harui" 
> wrote:
> > >
> > > It has turned out to be harder than expected to get the 
same
> > binaries
> > > on Mac and Win.  I now have the identical binaries for
> > royale-compiler and
> > > royale-typedefs Maven artifacts and am starting on 
royale-asjs.  I
> > might
> > > get lucky and the changes that fixed royale-typedefs SWCs will
> > magically
> > > get the royale-asjs SWCs to match.  Then we have to make the 
Ant
> > artifacts
> > > match.
> > >
> > > There have been a lot of distractions in my non-work life 
which
> > has
> > > also impeded progress.  I hope to make much progress this 
coming
> > week and
> > > if we're lucky, I will be asking for a volunteer (or 
volunteers) to
> > test
> > > drive all of this stuff and be the RM for 0.9.6.  I am not 
going to
> > be the
> > > RM.
> > >
> > > -Alex
> > >
> > > On 5/23/1

Re: 0.9.6 Release

[Carlos Rovira]

Hi,

I think both ASDocs and MX are important. The first since is part of the
docs. In fact I switched focus to Modules since I was having problems with
ASDocs as I'm referring to each permalink from Royale Docs, that is one of
the main things we put in priority in our way to 1.0. MX in the other hand
is something others are waiting for so I think is expected in 0.9.6

My next actions will be to end Module example (although I'm still solving
some technical issues) and then go back to royale docs



El lun., 17 jun. 2019 a las 17:12, Piotr Zarzycki (<
piotrzarzyck...@gmail.com>) escribió:

> I will try to run MX Tests locally and see what is happening.
>
> pon., 17 cze 2019 o 17:08 Alex Harui 
> napisał(a):
>
> > ASDoc is not fully working for me.  I've been working on that the past
> > several days.  It took quite a while to fix the JSON output from new and
> > interesting ASDoc patterns.  I'm now looking into failures due to changes
> > in coercion rules in the JS output.  ASDoc was trying to save time by
> using
> > plain objects that were isomorphic to a class instead of actual class
> > instances.
> >
> > MXTests are failing on the CI server and probably locally but I haven't
> > gotten that far.
> >
> > I would think we want these things to work, but you are the RM so it is
> up
> > to you.  Volunteers are welcome to help with these and other tasks.
> >
> > -Alex
> >
> > On 6/17/19, 12:44 AM, "Piotr Zarzycki" 
> wrote:
> >
> > Hi Alex,
> >
> > I don't see any answer here - can I start with trying your
> instruction
> > and
> > start release ?
> >
> > Thanks,
> > Piotr
> >
> > śr., 12 cze 2019 o 10:12 Carlos Rovira 
> > napisał(a):
> >
> > > Hi Alex,
> > > just build from scratch right now and seems maven and ant was ok
> > >
> > > El mié., 12 jun. 2019 a las 10:02, Alex Harui
> > ()
> > > escribió:
> > >
> > > > The merge is complete, and the main builds seem to be successful,
> > except
> > > > for Maven on royale-asjs which was failing before the merge.
> > > >
> > > > I'm still catching up on other things, but I think there are some
> > bugs we
> > > > need to fix before cutting the release?  I think ASDoc may not be
> > fully
> > > > working?  What else?
> > > >
> > > > -Alex
> > > >
> > > > On 6/10/19, 11:05 AM, "Piotr Zarzycki" <
> piotrzarzyck...@gmail.com>
> > > wrote:
> > > >
> > > > Hi Alex,
> > > >
> > > > Many thanks for that! I will try to be RM. I will have some
> > dedicated
> > > > time
> > > > for that. I will wait for your instruction and merge to
> > develop.
> > > >
> > > > Thanks,
> > > > Piotr
> > > >
> > > >
> > > > On Mon, Jun 10, 2019, 7:31 PM Alex Harui
> > 
> > > > wrote:
> > > >
> > > > > Well, that turned out to be much more time-consuming than I
> > > > expected, but
> > > > > we can now create identical release artifacts on Mac and
> > Win.  I am
> > > > hopeful
> > > > > this effort will pay off not only now in having other folks
> > > generate
> > > > > releases, but also in the future if signed binaries become
> a
> > > > requirement.
> > > > >
> > > > > There continues to be a lot of distractions in my life that
> > can
> > > cause
> > > > > delays, but I hope to merge the release_practice branches
> > into
> > > > develop over
> > > > > the next day or two and figure out where in the wiki to
> > document
> > > the
> > > > > release process.  So, now is the time for one or more
> people
> > to
> > > step
> > > > up to
> > > > > be the RMs for 0.9.6 and help debug and improve the
> process.
> > > > >
> > > > > I am going to try very hard not to "own" the process.  If
> > something
> > > > goes
> > > > > wrong, I am going to ask others to try to debug and fix it
> > first
> > > > because it
> > > > > is in the project's best interests for others to truly
> > understand
> > > > how this
> > > > > stuff works.
> > > > >
> > > > > Thanks,
> > > > > -Alex
> > > > >
> > > > > On 5/23/19, 9:54 AM, "Alex Harui"  >
> > > wrote:
> > > > >
> > > > > It has turned out to be harder than expected to get the
> > same
> > > > binaries
> > > > > on Mac and Win.  I now have the identical binaries for
> > > > royale-compiler and
> > > > > royale-typedefs Maven artifacts and am starting on
> > royale-asjs.  I
> > > > might
> > > > > get lucky and the changes that fixed royale-typedefs SWCs
> > will
> > > > magically
> > > > > get the royale-asjs SWCs to match.  Then we have to make
> the
> > Ant
> > > > artifacts
> > > > > match.
> > > > >
> > > > > There have been a lot of distractions in my non-work
> > life w

Re: Compiler Performance (was Re: Problem with Vectors)

[Alex Harui]

Folks are welcome to try various incremental compilation ideas.  I'm not sure 
you can guarantee a benefit in all cases when compiling from the command-line 
because you still may have to parse every file that is a dependency or 
encode/save/reload some sort of state cache.  There should be ways to get a 
benefit when compiling from an IDE because the IDE is your state cache and the 
compiler has some hooks for that.  But the old Flex MXMLC algorithm had bugs so 
I'm not sure I would borrow from that.

BTW, I ran across another jumble which is how much code is run in 
IdentifierNode.resolve.  I have often wondered why it must be so complex and 
whether passing context around would help.  Or prioritizing project scope class 
lookups over multiname lookups.

-Alex

On 6/17/19, 1:09 AM, "yishayw"  wrote:

Without knowing much about the compiler, what about incremental compilation?
Can gains be made by checking for source update timestamps and making a
decision as to whether or not to transpile based on that?



--
Sent from: 
https://nam04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-royale-development.20373.n8.nabble.com%2F&data=02%7C01%7Caharui%40adobe.com%7Cdb2b5c9e69a64cb8137a08d6f2fb2063%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636963557726154820&sdata=DorMV4v6H%2Bh4rcLM4Xp%2FydqX335SGqhwvN2UtGLvCaQ%3D&reserved=0

Re: 0.9.6 Release

[Piotr Zarzycki]

I will try to run MX Tests locally and see what is happening.

pon., 17 cze 2019 o 17:08 Alex Harui  napisał(a):

> ASDoc is not fully working for me.  I've been working on that the past
> several days.  It took quite a while to fix the JSON output from new and
> interesting ASDoc patterns.  I'm now looking into failures due to changes
> in coercion rules in the JS output.  ASDoc was trying to save time by using
> plain objects that were isomorphic to a class instead of actual class
> instances.
>
> MXTests are failing on the CI server and probably locally but I haven't
> gotten that far.
>
> I would think we want these things to work, but you are the RM so it is up
> to you.  Volunteers are welcome to help with these and other tasks.
>
> -Alex
>
> On 6/17/19, 12:44 AM, "Piotr Zarzycki"  wrote:
>
> Hi Alex,
>
> I don't see any answer here - can I start with trying your instruction
> and
> start release ?
>
> Thanks,
> Piotr
>
> śr., 12 cze 2019 o 10:12 Carlos Rovira 
> napisał(a):
>
> > Hi Alex,
> > just build from scratch right now and seems maven and ant was ok
> >
> > El mié., 12 jun. 2019 a las 10:02, Alex Harui
> ()
> > escribió:
> >
> > > The merge is complete, and the main builds seem to be successful,
> except
> > > for Maven on royale-asjs which was failing before the merge.
> > >
> > > I'm still catching up on other things, but I think there are some
> bugs we
> > > need to fix before cutting the release?  I think ASDoc may not be
> fully
> > > working?  What else?
> > >
> > > -Alex
> > >
> > > On 6/10/19, 11:05 AM, "Piotr Zarzycki" 
> > wrote:
> > >
> > > Hi Alex,
> > >
> > > Many thanks for that! I will try to be RM. I will have some
> dedicated
> > > time
> > > for that. I will wait for your instruction and merge to
> develop.
> > >
> > > Thanks,
> > > Piotr
> > >
> > >
> > > On Mon, Jun 10, 2019, 7:31 PM Alex Harui
> 
> > > wrote:
> > >
> > > > Well, that turned out to be much more time-consuming than I
> > > expected, but
> > > > we can now create identical release artifacts on Mac and
> Win.  I am
> > > hopeful
> > > > this effort will pay off not only now in having other folks
> > generate
> > > > releases, but also in the future if signed binaries become a
> > > requirement.
> > > >
> > > > There continues to be a lot of distractions in my life that
> can
> > cause
> > > > delays, but I hope to merge the release_practice branches
> into
> > > develop over
> > > > the next day or two and figure out where in the wiki to
> document
> > the
> > > > release process.  So, now is the time for one or more people
> to
> > step
> > > up to
> > > > be the RMs for 0.9.6 and help debug and improve the process.
> > > >
> > > > I am going to try very hard not to "own" the process.  If
> something
> > > goes
> > > > wrong, I am going to ask others to try to debug and fix it
> first
> > > because it
> > > > is in the project's best interests for others to truly
> understand
> > > how this
> > > > stuff works.
> > > >
> > > > Thanks,
> > > > -Alex
> > > >
> > > > On 5/23/19, 9:54 AM, "Alex Harui" 
> > wrote:
> > > >
> > > > It has turned out to be harder than expected to get the
> same
> > > binaries
> > > > on Mac and Win.  I now have the identical binaries for
> > > royale-compiler and
> > > > royale-typedefs Maven artifacts and am starting on
> royale-asjs.  I
> > > might
> > > > get lucky and the changes that fixed royale-typedefs SWCs
> will
> > > magically
> > > > get the royale-asjs SWCs to match.  Then we have to make the
> Ant
> > > artifacts
> > > > match.
> > > >
> > > > There have been a lot of distractions in my non-work
> life which
> > > has
> > > > also impeded progress.  I hope to make much progress this
> coming
> > > week and
> > > > if we're lucky, I will be asking for a volunteer (or
> volunteers) to
> > > test
> > > > drive all of this stuff and be the RM for 0.9.6.  I am not
> going to
> > > be the
> > > > RM.
> > > >
> > > > -Alex
> > > >
> > > > On 5/23/19, 9:39 AM, "Piotr Zarzycki" <
> > piotrzarzyck...@gmail.com
> > > >
> > > > wrote:
> > > >
> > > > Hi Alex,
> > > >
> > > > It's been a while since you have started effort with
> > > automating
> > > > build.
> > > > Where are you with that ? Are we closer to started
> 0.9.6.
> > Do
> > > you
> > > > need any
> > > > help with this ?
> > >

Re: 0.9.6 Release

[Alex Harui]

ASDoc is not fully working for me.  I've been working on that the past several 
days.  It took quite a while to fix the JSON output from new and interesting 
ASDoc patterns.  I'm now looking into failures due to changes in coercion rules 
in the JS output.  ASDoc was trying to save time by using plain objects that 
were isomorphic to a class instead of actual class instances.

MXTests are failing on the CI server and probably locally but I haven't gotten 
that far.

I would think we want these things to work, but you are the RM so it is up to 
you.  Volunteers are welcome to help with these and other tasks.

-Alex

On 6/17/19, 12:44 AM, "Piotr Zarzycki"  wrote:

Hi Alex,

I don't see any answer here - can I start with trying your instruction and
start release ?

Thanks,
Piotr

śr., 12 cze 2019 o 10:12 Carlos Rovira  napisał(a):

> Hi Alex,
> just build from scratch right now and seems maven and ant was ok
>
> El mié., 12 jun. 2019 a las 10:02, Alex Harui ()
> escribió:
>
> > The merge is complete, and the main builds seem to be successful, except
> > for Maven on royale-asjs which was failing before the merge.
> >
> > I'm still catching up on other things, but I think there are some bugs 
we
> > need to fix before cutting the release?  I think ASDoc may not be fully
> > working?  What else?
> >
> > -Alex
> >
> > On 6/10/19, 11:05 AM, "Piotr Zarzycki" 
> wrote:
> >
> > Hi Alex,
> >
> > Many thanks for that! I will try to be RM. I will have some 
dedicated
> > time
> > for that. I will wait for your instruction and merge to develop.
> >
> > Thanks,
> > Piotr
> >
> >
> > On Mon, Jun 10, 2019, 7:31 PM Alex Harui 
> > wrote:
> >
> > > Well, that turned out to be much more time-consuming than I
> > expected, but
> > > we can now create identical release artifacts on Mac and Win.  I 
am
> > hopeful
> > > this effort will pay off not only now in having other folks
> generate
> > > releases, but also in the future if signed binaries become a
> > requirement.
> > >
> > > There continues to be a lot of distractions in my life that can
> cause
> > > delays, but I hope to merge the release_practice branches into
> > develop over
> > > the next day or two and figure out where in the wiki to document
> the
> > > release process.  So, now is the time for one or more people to
> step
> > up to
> > > be the RMs for 0.9.6 and help debug and improve the process.
> > >
> > > I am going to try very hard not to "own" the process.  If 
something
> > goes
> > > wrong, I am going to ask others to try to debug and fix it first
> > because it
> > > is in the project's best interests for others to truly understand
> > how this
> > > stuff works.
> > >
> > > Thanks,
> > > -Alex
> > >
> > > On 5/23/19, 9:54 AM, "Alex Harui" 
> wrote:
> > >
> > > It has turned out to be harder than expected to get the same
> > binaries
> > > on Mac and Win.  I now have the identical binaries for
> > royale-compiler and
> > > royale-typedefs Maven artifacts and am starting on royale-asjs.  I
> > might
> > > get lucky and the changes that fixed royale-typedefs SWCs will
> > magically
> > > get the royale-asjs SWCs to match.  Then we have to make the Ant
> > artifacts
> > > match.
> > >
> > > There have been a lot of distractions in my non-work life 
which
> > has
> > > also impeded progress.  I hope to make much progress this coming
> > week and
> > > if we're lucky, I will be asking for a volunteer (or volunteers) 
to
> > test
> > > drive all of this stuff and be the RM for 0.9.6.  I am not going 
to
> > be the
> > > RM.
> > >
> > > -Alex
> > >
> > > On 5/23/19, 9:39 AM, "Piotr Zarzycki" <
> piotrzarzyck...@gmail.com
> > >
> > > wrote:
> > >
> > > Hi Alex,
> > >
> > > It's been a while since you have started effort with
> > automating
> > > build.
> > > Where are you with that ? Are we closer to started 0.9.6.
> Do
> > you
> > > need any
> > > help with this ?
> > >
> > > Thanks,
> > > Piotr
> > >
> > > wt., 2 kwi 2019 o 19:30 Alex Harui
>  > >
> > > napisał(a):
> > >
> > > > Update:
> > > >
> > > > In order to make verification of binary release packages
> > created
> > > on the
> > > > server easi

Build failed in Jenkins: royale-asjs_MXTests #859

[Apache Royale CI Server]

See 


--
[...truncated 2.34 MB...]
[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

[mxmlc] using source file: 

Re: Compiler Performance (was Re: Problem with Vectors)

[yishayw]

Without knowing much about the compiler, what about incremental compilation?
Can gains be made by checking for source update timestamps and making a
decision as to whether or not to transpile based on that?



--
Sent from: http://apache-royale-development.20373.n8.nabble.com/

Re: 0.9.6 Release

[Piotr Zarzycki]

Hi Alex,

I don't see any answer here - can I start with trying your instruction and
start release ?

Thanks,
Piotr

śr., 12 cze 2019 o 10:12 Carlos Rovira  napisał(a):

> Hi Alex,
> just build from scratch right now and seems maven and ant was ok
>
> El mié., 12 jun. 2019 a las 10:02, Alex Harui ()
> escribió:
>
> > The merge is complete, and the main builds seem to be successful, except
> > for Maven on royale-asjs which was failing before the merge.
> >
> > I'm still catching up on other things, but I think there are some bugs we
> > need to fix before cutting the release?  I think ASDoc may not be fully
> > working?  What else?
> >
> > -Alex
> >
> > On 6/10/19, 11:05 AM, "Piotr Zarzycki" 
> wrote:
> >
> > Hi Alex,
> >
> > Many thanks for that! I will try to be RM. I will have some dedicated
> > time
> > for that. I will wait for your instruction and merge to develop.
> >
> > Thanks,
> > Piotr
> >
> >
> > On Mon, Jun 10, 2019, 7:31 PM Alex Harui 
> > wrote:
> >
> > > Well, that turned out to be much more time-consuming than I
> > expected, but
> > > we can now create identical release artifacts on Mac and Win.  I am
> > hopeful
> > > this effort will pay off not only now in having other folks
> generate
> > > releases, but also in the future if signed binaries become a
> > requirement.
> > >
> > > There continues to be a lot of distractions in my life that can
> cause
> > > delays, but I hope to merge the release_practice branches into
> > develop over
> > > the next day or two and figure out where in the wiki to document
> the
> > > release process.  So, now is the time for one or more people to
> step
> > up to
> > > be the RMs for 0.9.6 and help debug and improve the process.
> > >
> > > I am going to try very hard not to "own" the process.  If something
> > goes
> > > wrong, I am going to ask others to try to debug and fix it first
> > because it
> > > is in the project's best interests for others to truly understand
> > how this
> > > stuff works.
> > >
> > > Thanks,
> > > -Alex
> > >
> > > On 5/23/19, 9:54 AM, "Alex Harui" 
> wrote:
> > >
> > > It has turned out to be harder than expected to get the same
> > binaries
> > > on Mac and Win.  I now have the identical binaries for
> > royale-compiler and
> > > royale-typedefs Maven artifacts and am starting on royale-asjs.  I
> > might
> > > get lucky and the changes that fixed royale-typedefs SWCs will
> > magically
> > > get the royale-asjs SWCs to match.  Then we have to make the Ant
> > artifacts
> > > match.
> > >
> > > There have been a lot of distractions in my non-work life which
> > has
> > > also impeded progress.  I hope to make much progress this coming
> > week and
> > > if we're lucky, I will be asking for a volunteer (or volunteers) to
> > test
> > > drive all of this stuff and be the RM for 0.9.6.  I am not going to
> > be the
> > > RM.
> > >
> > > -Alex
> > >
> > > On 5/23/19, 9:39 AM, "Piotr Zarzycki" <
> piotrzarzyck...@gmail.com
> > >
> > > wrote:
> > >
> > > Hi Alex,
> > >
> > > It's been a while since you have started effort with
> > automating
> > > build.
> > > Where are you with that ? Are we closer to started 0.9.6.
> Do
> > you
> > > need any
> > > help with this ?
> > >
> > > Thanks,
> > > Piotr
> > >
> > > wt., 2 kwi 2019 o 19:30 Alex Harui
>  > >
> > > napisał(a):
> > >
> > > > Update:
> > > >
> > > > In order to make verification of binary release packages
> > created
> > > on the
> > > > server easier, I have made changes to our build scripts
> and
> > > tools to try to
> > > > generate reproducible binaries.  I've seen two different
> > builds
> > > compare on
> > > > my Mac.   The next challenge will be to see if the server
> > can
> > > build a
> > > > package on Windows that will compare on Mac.
> > > >
> > > > One of the changes I needed to make is to JBurg.  The
> > version of
> > > JBurg we
> > > > use generates method names including a hash that doesn't
> > > reproduce the same
> > > > name each time.  I have changes to JBurg ready, however
> > JBurg is
> > > currently
> > > > under CPL which is category B.  We only need one file, we
> > don't
> > > need or
> > > > want all of JBurg at this time.  The one JBurg file is
> > jointly
> > > owned by
> > > > Adobe and Tom Harwood.  I've contact Tom and he will be
> > filing
> > > an ICLA and
> > > > has given me permission to commit the lines he owns in
> > that one
> > > file.
> > > >
> > > > This is the revision of the