Re: ASDoc, Routing, Releases

I pushed an Ant script that copies ASDoc from the CI server and copies it to the asdoc folder on the site. I also pushed the ASDoc. I was going to hook this up to a CI build but am having second thoughts. Maybe we should only update after a release. Otherwise, folks may find stuff in ASDoc that

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

Andrew, please merge royale-docs develop branch to master. If you don't have time, I will do it, but I want to make sure you are set up to do so. -Alex On 2/12/18, 10:11 PM, "Alex Harui" wrote: >OK, I fixed up and ran the releasecandidate script and it says it posted

Jenkins build is back to normal : royale-asjs_jsonly #293

See

Build failed in Jenkins: royale-asjs_jsonly #292

See Changes: [aharui] fix rm script [aharui] wrong versions [aharui] final adjustments to npm publishing [aharui] fix up versions --

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

OK, I fixed up and ran the releasecandidate script and it says it posted the bits to npm. I believe all branches are merged and appropriately versioned. On to the next release! -Alex On 2/12/18, 12:29 PM, "omup...@gmail.com on behalf of OmPrakash Muppirala"

Re: Build failed in Jenkins: royale-typedefs #139

Where would the fork live? You might also have to verify that all of the things we patch are available on GH or Git. Seems like a good possibility, -Alex On 2/12/18, 1:46 PM, "Gabe Harbs" wrote: >We could fork the Google externs on Github and update those. If we pull

[OFF TOPIC] Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

Justin, I have a suggestion which I hope you take in the spirit of constructive criticism. Anything you might see and wish to share with this project on a release which is an observation for the future should be shared in a separate thread with an altered subject. FWIW the collective work’s

Re: First Theme in place

Hi Alex, very good questions here. Sincerely I'd like to have responses but this path is completely new for me and need to experiment with it. If someone has examples or urls where this is covered, please share here, since I'm starting right now to see what can be done. As you see, yesterday I

Re: Build failed in Jenkins: royale-typedefs #139

We could fork the Google externs on Github and update those. If we pull from the forked repo, we would be able to quickly make changes that we need and make PRs back to Google. That would also make it easy to pull in changes that are made to Google’s repo. Harbs > On Feb 12, 2018, at 10:00

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

On Mon, Feb 12, 2018 at 12:22 PM, Alex Harui wrote: > The steps that publish NPM are in releasecandidate.xml. It does not muck > with the packages. It looked like your change was to the package.json > that is already in the package. > Yes that is correct. When

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

The steps that publish NPM are in releasecandidate.xml. It does not muck with the packages. It looked like your change was to the package.json that is already in the package. -Alex On 2/12/18, 12:08 PM, "omup...@gmail.com on behalf of OmPrakash Muppirala"

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

On Feb 12, 2018 11:58 AM, "Alex Harui" wrote: On 2/12/18, 10:42 AM, "omup...@gmail.com on behalf of OmPrakash Muppirala" wrote: >On Feb 12, 2018 10:13 AM, "Alex Harui" wrote: > >AIUI,

Re: Build failed in Jenkins: royale-typedefs #139

For js.patch, see the js/build.xml in royale-typedefs and the make_patch target. Lately, I've been thinking about finding some other way of dealing with this. We used patches so we didn't have to host 3rd-party files in our source tree. I think we may now have 3rd-party files elsewhere in our

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

On 2/12/18, 10:42 AM, "omup...@gmail.com on behalf of OmPrakash Muppirala" wrote: >On Feb 12, 2018 10:13 AM, "Alex Harui" wrote: > >AIUI, js-swf did not work via NPM as generated from the source package. I

Re: Build failed in Jenkins: royale-typedefs #139

Where is that and how do I add to the patches? > On Feb 12, 2018, at 7:31 PM, Alex Harui wrote: > > Patch files are auto-generated by the Ant script, verified by a build, > then committed to the src/main/patch/js.patch so that the Maven build can > use it. If we

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

AIUI, js-swf did not work via NPM as generated from the source package. I thought you tweaked it before publishing. Am I wrong about that? -Alex On 2/12/18, 10:10 AM, "omup...@gmail.com on behalf of OmPrakash Muppirala" wrote: >On Feb 12,

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

On Feb 12, 2018 8:24 AM, "Alex Harui" wrote: Voters, Apparently some of you have not quite bought into the new philosophy of "better than last release and not illegal". The npm install that worked in the last release will be broken by this new release. Please

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

On 2/9/18, 3:38 PM, "Justin Mclean" wrote: >Hi, > >> I only found one copyright. Where are the others? > >There’s 2 in the ASFCopyrightAttributionAndLinks.as file I believe. Might >be another I would need to double check. > >Found via: >find . -type f -exec grep

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

I was kind of waiting for you to come back online to see where you stood on these issues. You’re the RM and I’m not interested in causing you extra work. If you had wanted to do another RC, then I would have waited and tested that one. If the npm issue could have been fixed without another RC,

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

My point is that is shouldn't matter when Om fixed the npm issue. It was too late and not as important as getting other stuff done. No need to wait to see how hard the fix would be. We want to adopt the philosophy that by default, every found issue goes to the next release unless it is REALLY

Re: [VOTE] Release Apache Royale 0.9.1 RC1

+1 Package https://dist.apache.org/repos/dist/dev/royale/0.9.1/rc1/apache-royale-0.9.1-src.tar.gz Java 1.8 OS: Mac OS X x86_64 10.12.6 Source kit signatures match: y Source kit builds: y README is ok: y RELEASE_NOTES is ok: y NOTICE is ok: y LICENSE is ok: y No unapproved licenses or archives: y

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

My comments were not to say that I wasn’t voting for the release. I was just documenting my difficulties with the approval script. In fact I was about to vote +1 when I saw that Om fixed the npm issue. I wanted to test the npm release before I sent in my vote. If you want to push it to next

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

+1 - Let's go with release! If npm won't work I can download sdk manually and use it in my IDE. Thanks, Piotr 2018-02-12 17:24 GMT+01:00 Alex Harui : > Voters, > > Apparently some of you have not quite bought into the new philosophy of > "better than last release and

RE: Build failed in Jenkins: royale-typedefs #139

Add, see my later mail. From: Gabe Harbs Sent: Monday, February 12, 2018 6:23 PM To: dev@royale.apache.org Subject: Re: Build failed in Jenkins: royale-typedefs #139 What was it? Did you add a space, or remove a space? > On Feb 12,

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

Voters, Apparently some of you have not quite bought into the new philosophy of "better than last release and not illegal". The cost of a new RC is non-trivial. We have some users who are asking for help to port their apps against a deadline. To me, that is way more important than fixing

Re: Build failed in Jenkins: royale-typedefs #139

What was it? Did you add a space, or remove a space? > On Feb 12, 2018, at 6:11 PM, Yishay Weiss wrote: > > I fixed it for my machine. Hopefully Jenkins will be happy too. > > From: Gabe Harbs > Sent: Monday, February 12, 2018 5:18 PM > To:

Re: Build failed in Jenkins: royale-typedefs #139

Yishay, I did try it the same and it didn't work to me. Your commit helps. Weird. :) Thanks, Piotr 2018-02-12 17:11 GMT+01:00 Yishay Weiss : > I fixed it for my machine. Hopefully Jenkins will be happy too. > > From: Gabe Harbs > Sent:

Jenkins build is back to normal : royale-typedefs #140

See

RE: Build failed in Jenkins: royale-typedefs #139

I fixed it for my machine. Hopefully Jenkins will be happy too. From: Gabe Harbs Sent: Monday, February 12, 2018 5:18 PM To: dev@royale.apache.org Subject: Re: Build failed in Jenkins: royale-typedefs #139 Does anyone understand the

Re: Build failed in Jenkins: royale-typedefs #139

Does anyone understand the patch file syntax enough to guess what’s wrong? > On Feb 12, 2018, at 5:15 PM, Yishay Weiss wrote: > > I have the same result with ant on my machine. It’s probably a windows/mac > thing. > > From: Piotr

RE: Build failed in Jenkins: royale-typedefs #139

I have the same result with ant on my machine. It’s probably a windows/mac thing. From: Piotr Zarzycki Sent: Monday, February 12, 2018 3:13 PM To: dev@royale.apache.org Subject: Re: Build failed in Jenkins: royale-typedefs #139

Re: Build failed in Jenkins: royale-typedefs #139

Locally I have same result. I tried couple of things but nothing help. I see that it is being applied by following command: [DEBUG] Executing command line: [C:\Program Files\Git\cmd\git.exe, apply, --ignore-whitespace, --whitespace=nowarn,

Re: Build failed in Jenkins: royale-typedefs #139

Thanks. > On Feb 12, 2018, at 2:43 PM, Piotr Zarzycki wrote: > > Let me build typedefs locally by Maven will see. > > 2018-02-12 13:36 GMT+01:00 Gabe Harbs : > >> I have no idea why this failed. The patch applied on my machine and line >> 64

Re: Build failed in Jenkins: royale-typedefs #139

Let me build typedefs locally by Maven will see. 2018-02-12 13:36 GMT+01:00 Gabe Harbs : > I have no idea why this failed. The patch applied on my machine and line > 64 is an empty line. > > Maybe it’s a Windows line ending problem? > > Thoughts? > Harbs > > > On Feb 12,

Re: Build failed in Jenkins: royale-typedefs #139

I have no idea why this failed. The patch applied on my machine and line 64 is an empty line. Maybe it’s a Windows line ending problem? Thoughts? Harbs > On Feb 12, 2018, at 2:12 PM, apacheroyal...@gmail.com wrote: > > See >

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

It failed also on ANT [1]. I made for Maven build configuration changes, so project should be build after each commit etc. [1] http://apacheroyaleci.westus2.cloudapp.azure.com:8080/job/royale-typedefs/139/ 2018-02-12 13:04 GMT+01:00 Piotr Zarzycki : > I just fired up

Build failed in Jenkins: royale-typedefs #139

See -- Started by upstream project "royale-compiler" build number 89 originally caused by: Started by timer Building in workspace

Build failed in Jenkins: royale-typedefs #138

See Changes: [harbs] Added overflowX and overflowY -- Started by an SCM change Building in workspace

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

How do we ge royale-typedefs to be built on the Jenkins server? I checked and my changes did not seem to queue the job. Harbs > On Feb 12, 2018, at 1:48 PM, Harbs wrote: > > I think I figured it out. Here’s what I did: > > 1. I added the w3c_css.js file to an unrelated

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

I don’t know about scaleX and rotate, but I was able to get translateY() to work by using @namespace "http://www.w3.org/1999/xhtml”; It’s worth a shot… Harbs > On Feb 12, 2018, at 1:54 PM, Piotr Zarzycki wrote: > > I had problem lately. I did found workaround. It

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

I’m pretty sure that those need to be handled in the compiler. The compiler needs to understand CSS as it parses it. Harbs > On Feb 12, 2018, at 1:48 PM, Piotr Zarzycki wrote: > > Harbs, > > Does that file is also the place where we may add some missing functions >

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

Harbs, Does that file is also the place where we may add some missing functions from CSS ? For example scaleY, rotate etc. ? Thanks, Piotr 2018-02-12 12:17 GMT+01:00 Harbs : > The relevant extern file appears to be: w3c_css.js > > Harbs > > > On Feb 12, 2018, at 1:16 PM,

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

I think I figured it out. Here’s what I did: 1. I added the w3c_css.js file to an unrelated git repo and committed it. 2. I added the properties to the file. 3. I ran git diff on the repo, which gave the the changes. 4. I copied the output into js.patch 5. I changed the path of w3c_css.js in the

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

The relevant extern file appears to be: w3c_css.js Harbs > On Feb 12, 2018, at 1:16 PM, Harbs wrote: > > The externs for CSSStyleDeclaration has `overflow`, but it’s missing > `overflowX` and `overflowY`. > > That’s causing these two lines to fail (only in maven for some

Re: [MAVEN-BUILD] Royale-asjs - Build # 485 - Failure

The externs for CSSStyleDeclaration has `overflow`, but it’s missing `overflowX` and `overflowY`. That’s causing these two lines to fail (only in maven for some reason): (value as UIBase).element.style.overflowY = "auto"; (contentView as UIBase).element.style.overflowY = "auto”; I could change

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

BTW, I noticed that I’m getting an error near the end of my ant build. This does not cause the “general bull” to fail: check-playerglobal-home: [echo] PLAYERGLOBAL_HOME is /Apache/frameworks/libs/player [echo] playerglobal.version is 11.1 [echo] playerglobal.swc is

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

Does this require a new rc or can the script just be updated? > On Feb 12, 2018, at 10:08 AM, OmPrakash Muppirala > wrote: > > I've pushed in a fix for the broken js-swf npm installation process. >

Re: First Theme in place

Thanks for getting started on this. Good to know that Theme is working for you. I think we've used Theme in a few other places already like Flat in case you want to see it working somewhere else. AIUI, Vivid is just one look-and-feel you and others will provide based on a new "sub-DOM" within

Re: vars in CSS

I tried to fix most of the advanced selectors for test cases folks brought up, but things like CSS vars were not tested. You can look at the tests we do have to see what was tested. You can't put in Type Selectors for HTML Elements without specifying the HTML namespace for them. Maybe that's

Re: [DISCUSS] Discuss Release Apache Royale 0.9.1 RC1

I've pushed in a fix for the broken js-swf npm installation process. https://github.com/apache/royale-asjs/commit/b9a1b32975cc755475ade761f36993efc8f1f058 Alex, given that this release is about npm installation (as per the RELEASE_NOTES), we need this fix to go in. Sorry for the trouble :-(

Re: [VOTE] Release Apache Royale 0.9.1 RC1

I will wait to cast my vote for that. > On Feb 12, 2018, at 9:57 AM, OmPrakash Muppirala wrote: > > js-swf npm package seems to have a bug. I will check in a fix to 0.9.1 > branch soon.