On 1/31/17, 4:11 AM, "Christofer Dutz" wrote:
>Hi,
>
>There has been a change in reflection between 1.7 and 1.8 (I guess mostly
>related to unit-testing) that the order in which methods are listed via
>reflection is no longer deterministic in 1.8.
>
>So is this problem not occurring in 1.7 and
I'm also running Java 1.8, and I could reproduce.
- Josh
On Jan 31, 2017 2:29 AM, "Harbs" wrote:
> Apparently, Yishay was using Java 1.7 and I was using Java 1.8.
>
> Yishay just removed 1.7 so it’s now compiling using 1.8 and he’s now
> getting the error.
>
> So apparently the order changes be
It seems so, but I committed a temporary fix which seems to resolve it.
> On Jan 31, 2017, at 2:11 PM, Christofer Dutz
> wrote:
>
> So is this problem not occurring in 1.7 and is sometimes in 1.8?
Hi,
There has been a change in reflection between 1.7 and 1.8 (I guess mostly
related to unit-testing) that the order in which methods are listed via
reflection is no longer deterministic in 1.8.
So is this problem not occurring in 1.7 and is sometimes in 1.8? (Sorry haven’t
been following th
Apparently, Yishay was using Java 1.7 and I was using Java 1.8.
Yishay just removed 1.7 so it’s now compiling using 1.8 and he’s now getting
the error.
So apparently the order changes between 1.7 and 1.8…
> On Jan 31, 2017, at 3:12 AM, Alex Harui wrote:
>
> Are you using a different version o
Yes. It is a blocker for me.
I’m not sure what’s different. Yishay is on Windows and I’m on Mac. I recently
upgraded from Maverick to Yosemite. Maybe that changed something. Don’t know.
If the multiple-target will solve this problem, good. For now, I just pushed a
change to check for all three
On 1/30/17, 3:33 PM, "Harbs" wrote:
>
> [java] resolved: SWC: [DefinitionPromise "XML", DefinitionPromise
>"QName", DefinitionPromise "XMLList"]
> [java] XML
> [java] QName
> [java] XMLList
>
>
>So, it looks like the order of the classes are indeed XML, QName, XMLList.
>
>Maybe,
I did this and the result was:
[java]
/Users/harbs/Documents/ApacheFlex/frameworks/libs/player/11.1/playerglobal.swc
which is the same file I sent you.
> On Jan 31, 2017, at 12:17 AM, Alex Harui wrote:
>
> You can output
> cu.getAbsoluteFilename().
To me, the problem looks like the output being fed into postProcess. There is
no goog.require in that string.
When the code loops through the lines, it does not find XML as a foundRequire
and it does not add it because there is no goog.require for the XML.
Interestingly, on systems which output
I added some additional output:
System.out.println("resolved: " +
flexJSProject.resolveQNameToCompilationUnit(usedName));
And in FlexJSProject.isExternalLinkage():
for (String oneqname : qnames)
{
System.out.println(oneqname);
}
[java]
Hmm. The one you sent is the same as the one I have. Maybe I dreamt
about what I saw earlier. Let's go back to verifying how the code works.
In MXMLFlexJSEmitter.java in postProcess(), there is a code block like
this:
for (String usedName :usedNames) {
if (!foundRequires.contains(usedName)) {
Hmm. IIRC, there was a playerglobal.swc that was packaged differently
than the others, but I can't find it.
Harbs, can you send me off-list your playerglobal.swc?
/Users/harbs/Documents/ApacheFlex/frameworks/libs/player/11.1/playerglobal.
swc
Thanks,
-Alex
On 1/30/17, 7:46 AM, "Alex Harui" wr
Crap. You are hitting https://issues.apache.org/jira/browse/FLEX-35203
Which I thought I'd fixed but I guess not.
I'll push a fix today.
-Alex
On 1/30/17, 1:39 AM, "Harbs" wrote:
>FWIW, Yishay’s falcon is from about a month ago. He’s getting (two)
>goog.requires for XML. I’m using an ant bui
FWIW, Yishay’s falcon is from about a month ago. He’s getting (two)
goog.requires for XML. I’m using an ant build (and not Maven) in case that
makes a difference.
> On Jan 30, 2017, at 11:27 AM, Harbs wrote:
>
> I added a bunch of System.outs to see what’s going on:
> http://pastebin.com/Wr4xM
I added a bunch of System.outs to see what’s going on:
http://pastebin.com/Wr4xMDaT
The way it looks to me is that there’s no goog.require being added for XML. So
although it sees XML in the asEmitter used names, it’s not found as a require.
Even if it would, without the goog.require, the code w
Hmm. Well it sees that XML is a usedName.
You can see on commits@ my last changes to MXMLFlexJSEmitter.java. I have
to stop working for tonight, so it will probably be most efficient if you
sprinkle more System.out.println calls throughout that postProcess()
method to see why the requires for XM
Sorry about that.
Here’s the output for the test:
http://pastebin.com/f8BpVwDJ
> On Jan 30, 2017, at 3:34 AM, Alex Harui wrote:
>
> The test project looks like it compiled against the nightly build and thus
> didn't emit the extra output? When you compiled DataBindingExample, it
> had the ext
The test project looks like it compiled against the nightly build and thus
didn't emit the extra output? When you compiled DataBindingExample, it
had the extra output and compiled against the repo.
-Alex
On 1/29/17, 1:12 PM, "Harbs" wrote:
>Here’s what I get for the test project:
>http://past
Here’s what I get for DataBindingExample:
[echo] Compiling DataBindingExample.js
[echo] FLEX_HOME: /Users/harbs/Documents/ApacheFlex/flex-asjs
[echo] FALCONJX_HOME:
/Users/harbs/Documents/ApacheFlex/flex-falcon/compiler-jx
[echo] GOOG_HOME:
/Users/harbs/Documents/ApacheFlex/f
Here’s what I get for the test project:
http://pastebin.com/XXCdPYxi
> On Jan 29, 2017, at 8:13 PM, Alex Harui wrote:
>
> I just pushed a compiler change to output more information. For
> DataBindingExample, it output:
>
> [java] MyInitialView as: [MyInitialView,
> org.apache.flex.events.Custo
On the sample project or DataBindingExample?
> On Jan 29, 2017, at 8:13 PM, Alex Harui wrote:
>
> I just pushed a compiler change to output more information. For
> DataBindingExample, it output:
>
> [java] MyInitialView as: [MyInitialView,
> org.apache.flex.events.CustomEvent, org.apache.flex.
I just pushed a compiler change to output more information. For
DataBindingExample, it output:
[java] MyInitialView as: [MyInitialView,
org.apache.flex.events.CustomEvent, org.apache.flex.html.RadioButton,
models.MyModel, org.apache.flex.events.ValueChangeEvent]
[java] MyInitialView mxml: [o
The problem does not seem to only be that the XML files are not copied. The
problem seems to be that the dependencies are not being calculated.
Here’s the js that the mani class puts out:
http://pastebin.com/bdsbTjVD
That’s coming from this:
http://ns.adobe.com/mxml/2009";
xmlns:js="library
This [1] is my successful build output for what it's worth.
[1] https://paste.apache.org/W6DC
--
View this message in context:
http://apache-flex-development.247.n4.nabble.com/XML-files-not-copying-tp58425p58774.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.
Here’s the output:
http://pastebin.com/tKzYsZj5
And here’s the ant script(s):
http://pastebin.com/Bwm6jph5
http://pastebin.com/MbQFPNQL
Yishay and I have been trying to reproduce this and it’s working for him, but
not for me. Something odd is going on.
Harbs
> On Jan 20, 2017, at 8:11 PM, Ale
Works for me. What console output do you get?
On 1/20/17, 8:53 AM, "Alex Harui" wrote:
>
>
>On 1/20/17, 2:41 AM, "Harbs" wrote:
>
>>Any comments on this?
>
>Sorry, I assumed it was some VSCode issue so I never clicked on the link.
>I will take a look.
>
>-Alex
>
On 1/20/17, 2:41 AM, "Harbs" wrote:
>Any comments on this?
Sorry, I assumed it was some VSCode issue so I never clicked on the link.
I will take a look.
-Alex
Any comments on this?
Harbs
> On Jan 18, 2017, at 9:55 PM, Harbs wrote:
>
> Please see this discussion on Github.[1]
>
> The thing is, it seems to work with my ant build, but I’m not sure why.
>
> I can post my ant script, but it’s really mostly just copied from the
> examples.
>
> [1]https
Please see this discussion on Github.[1]
The thing is, it seems to work with my ant build, but I’m not sure why.
I can post my ant script, but it’s really mostly just copied from the examples.
[1]https://github.com/BowlerHatLLC/vscode-nextgenas/issues/49#issuecomment-273528593
29 matches
Mail list logo
