gmane.comp.java.openlaszlo.devel

Re: using TLF framework in edittext loses focus?

augusto callejas — 2012-05-23T00:28:17

i found a *terrible* way to work around this issue, but it works.  i moved the field.setSelection call to a delegate, which allows it to work:

===
                        var del = new lz.Delegate(this, "selectEnd");
                        lz.Timer.addTimer(del, 1);
                </method>
                <method name="selectEnd" args="ignore">
                        this.field.setSelection(this.value.length);
                </method>
===

this should help narrow down what the problem is with the TLF framework.

augusto.

On May 22, 2012, at 4:00 PM, augusto callejas wrote:

Re: using TLF framework in edittext loses focus?

augusto callejas — 2012-05-22T23:00:40

ps-  i'm using a nightly build of openlaszlo (5.0.x-19545)

On May 22, 2012, at 3:58 PM, augusto callejas wrote:

using TLF framework in edittext loses focus?

augusto callejas — 2012-05-22T22:58:19

hi-

i'm writing a custom edittext class that formats my text based on input.  included here is a simplified version that separates text by dashes.
the screenshot shows two instances of my class, one with "hasdirectionallayout" not set and another with "hasdirectionallayout" set to true:

  http://i.imgur.com/AoBSN.png

the first editext instance shows the proper behavior: you type a character, the character is displayed and appends a "-", and sets the selection position at the end of the field, where you can continue typing more characters.

the second edittext instance shows incorrect behavior: you type a character, the character is displayed and appends a "-", but it doesn't set the selection position at the end of the field - it loses focus, so you can't type into the edittext unless you click back into it.

two points to make:

1) setting "hasdirectionallayout=true" changes the behavior of the edittext, which seems wrong.
2) my general approach of overriding "setValue" in order to format my text may not be optimal 

i'm interested in the answer to point 1 more than point 2 - unless there is a different approach to point 2 that still allows me to use "hasdirectionallayout=true".

sample code below.

thanks,
augusto.

====
<canvas height="100%" width="100%">
        <simplelayout axis="y" spacing="10"/>

        <class name="dashes" extends="edittext">
                <attribute name="process" type="boolean" value="true"/>
                <method name="setValue" args="t, isinitvalue=null">
                        if (t == null || t == '') {
                                return;
                        }
                        if (!process) {
                                super.setValue(t, isinitvalue);
                                return;
                        }

                        t += "-"; 
                        setAttribute('process', false);
                        setAttribute('text', t);
                        setAttribute('process', true);

                        if (this['field'] != null)
                                field.setSelection(t.length);
                </method>
        </class>

        <text>works:</text>
        <dashes/>
        <text>doesn't work:</text>
        <dashes hasdirectionallayout="true"/>
</canvas>
====

Re: [Laszlo-user] Microphone.getEnhancedMicrophone(); throws compiler error

Raju Bitter — 2012-05-22T08:35:20

Sebastian,

getEnhancedMicrophone() works with the swf11 runtime, I've tried this code:
<canvas debug="true">

  <class name="micro">
    <passthrough when="$as3">
       import flash.media.Microphone;
       import flash.media.MicrophoneEnhancedOptions;
    </passthrough>
    <attribute name="mic" value="null" />
    <handler name="oninit">
       if ($swf11) {
         this.mic = Microphone.getEnhancedMicrophone();
 var options:MicrophoneEnhancedOptions = this.mic.enhancedOptions;
         Debug.inspect(options);
       } else if ($swf10) {
         Debug.warn("No support for Microphone.getEnhancedMicrophone()
in swf10 runtime");
       }
    </handler>
  </class>

  <micro />

</canvas>

For anyone interested in testing the new functionality, check this
Adobe article:
http://www.adobe.com/devnet/flashplayer/articles/acoustic-echo-cancellation.html

- Raju

Re: Mage+ / Magento: Another example of an enterprise open source software fork

2012-05-16T18:12:44

Funny, we just did create a Magento plugin for OpenMeetings.
Now we need might need to do the same for Mage+ :)

Sebastian

2012/5/12 Raju Bitter <r.bitter.mailinglists< at >googlemail.com>

Safari 5.1.7 for Lion and Snow Leopard disableout-of-date Flash Players

Raju Bitter — 2012-05-14T22:57:28

http://support.apple.com/kb/HT5271
Safari 5.1.7 for OS X Lion and Safari 5.1.7 for OS X Snow Leopard
disable out-of-date versions of Adobe Flash Player.

Out-of-date versions of Adobe Flash Player do not include the latest
security updates and will be disabled to help keep your Mac secure. If
Safari 5.1.7 detects an out-of-date version of Flash Player on your
system, you will see a dialog informing you that Flash Player has been
disabled. The dialog provides the option to go directly to Adobe's
website, where you can download and install an updated version of
Flash Player.

IRC chat channel for OpenLaszlo community?

Raju Bitter — 2012-05-14T09:01:58

Do any of you use IRC at all? We could use the IRC channel
irc://irc.freenode.org/#openlaszlo for project/community members.
Laszlo originally used irc://irc.freenode.org/#laszlo

but I'd prefer using the channel #openlaszlo instead.

- Raju

Mage+ / Magento: Another example of an enterprise open source software fork

Raju Bitter — 2012-05-12T10:55:33

The Magento community just forked the Magento e-commerce system under
the new name Mage+:
http://www.mageplus.org/

Back in 2010, eBay acquired 49% of Magento, and last year they
acquired the remaining 51%. This year the Magento CTO left the
company, and here are some of the reasons for his decision:
"I will answer and give much more information over the next few weeks
as to the reasons I had to leave Magento. As for now I can say that
(very) short term I would not be worried about the ‘open’ part of
Magento, but as I have learned eBay and the folks at X.commerce don’t
really understand the meaning of open and have a hard time explaining
and defining it to them selves and to others. As such, long term, it
would be very interesting to see if Magento will continue to stay open
in the manor the people behind Magento and I meant it to be."
http://techcrunch.com/2012/04/12/recently-departed-magento-cto-and-co-founder-ebay-doesnt-understand-the-meaning-of-open/

I have the same feeling about Laszlo, since the founder and former CTO
David Temkin left the company: The management in the past years has
shown that they don't understand the meaning of open source, and how
an open source project should be run to foster business. And Critical
Path doesn't seem to have any connection to open source
software/projects. Try searching the CP.net website for "open source",
no results...

Re: Nightly builds & DHTML

augusto callejas — 2012-05-11T18:16:31

i verified this as an error on my side.

thanks,
augusto.

On May 10, 2012, at 3:53 AM, Raju Bitter wrote:

Re: Preliminary release of Apache Flex 4.8 availablefor download

Raju Bitter — 2012-05-11T13:37:50

OpenLaszlo as an Apache Software Foundation project: That definitely
would help the OpenLaszlo project, but going through the legal process
of contributing a project like OpenLaszlo to Apache is A LOT of work.
The Adobe legal department seemed to be very busy - including
high-level management - to go through all steps for a legal clearing
to contribute the code. In Adobe's case, that took the company 3-4
months, but I read that the total size of the SVN repository was more
than 10 GB.

And there are some tricky things: Apache Flex, for example, is not
allowed to include the playerglobal.swc for the Flex SDK in their code
base project tree. That means, for future releases of Apache Flex,
they need to use a script to download the playerglobal.swc from a
separate location and add it to the local system using an environment
variable PLAYERGLOBAL. The code for automatic embedding of TTF fonts
can not be contributed to Apache Flex, for legal reasons.

If OpenLaszlo would become an ASF project, those limitations would
apply to OpenLaszlo as well, making it more difficult to provide the
out-of-the-box experience you currently have.

Still, the advantage of being an Apache project would be tremendous.

- Raju

Re: Preliminary release of Apache Flex 4.8 availablefor download

Raju Bitter — 2012-05-11T13:10:13

Yes, that would be great. OpenLaszlo as an ASF project. ;-)

Here's LZPix compiled to SWF11 with Apache Flex...
http://www.flickr.com/photos/semanticmemories/7176235824/in/photostream/lightbox/

Re: Preliminary release of Apache Flex 4.8 availablefor download

Bret Simister — 2012-05-11T13:06:34

Nice job Raju!

On May 11, 2012, at 6:02 AM, P T Withington <ptw< at >pobox.com> wrote:

Re: Preliminary release of Apache Flex 4.8 availablefor download

P T Withington — 2012-05-11T13:02:32

Neat!

Now you just need to get Apache to pick up OpenLaszlo.  :)

On 2012-05-11, at 08:58, Raju Bitter wrote:

Re: Preliminary release of Apache Flex 4.8 availablefor download

Raju Bitter — 2012-05-11T12:58:53

I did a quick test run with Apache Flex 4.8, and I can build the LFC,
compile the LZPix app - everything seems to be working fine. That's
promising, since OpenLaszlo should be using Apache Flex in the future.

- Raju

Preliminary release of Apache Flex 4.8 available fordownload

Raju Bitter — 2012-05-11T12:42:43

The Apache Flex project has created a preliminary release of Apache
Flex 4.8, which should be compatible with the Adobe Flex 4.6 SDK:
http://people.apache.org/~cframpton/ApacheFlexRC/

Apache Flex 4.8.0 is a parity release from Apache of Adobe Flex 4.6.

Re: Nightly builds & DHTML

Raju Bitter — 2012-05-10T10:53:39

I tested with both 19597 and 19599, and don't see any errors either
when building the LPS, or when compiling and running a DHTML app.

- Raju

Re: Nightly builds & DHTML

Raju Bitter — 2012-05-10T10:50:39

Is that an error message you get when building LPS?

If that's an error message shown in the JavaScript console when
running the application, make sure that you clear your  browser cache
- or the browser might load the cached version of the LFC.

- Raju

inputtext and bidirectional text bug (openlaszlo5.0.x-19545)

augusto callejas — 2012-05-10T05:32:03

hi-

i want to enable bidirectional support on an input field via a constraint, but i'm unable to.  consider this code:

====
<canvas width="100%" height="100%">
        <attribute name="flag" type="boolean" value="true"/>
        <simplelayout axis="y" spacing="10"/>
        <inputtext hasdirectionallayout="true" direction="rtl" bgcolor="green" textalign="right"/>
        <inputtext hasdirectionallayout="${canvas.flag}" direction="rtl" bgcolor="red" textalign="right"/>
</canvas>
====

this code displays two inputtext views, one green and one red.  the green one has "hasdirectionallayout" set to true,
while the green one has "hasdirectionallayout" set to a constraint, which equals true.  they should both act the same,
but they don't:

green: while typing text, the blinking cursor stays to the left of the text (expected behavior)
red: while typing type, the blinking cursor stays to the right of the text (unexpected behavior)

if others believe this is a bug, i can file a jira issue on it.  if others have any temporary work-arounds to this bug, tips
would be appreciated.

thanks,
augusto.

Nightly builds & DHTML

augusto callejas — 2012-05-10T00:27:18

hi-

i noticed that it was mentioned that DHTML generation was broken, and then fixed in the discussion here:

  http://forum.openlaszlo.org/showthread.php?p=48430

i did a subversion checkout of the latest (Last Changed Rev: 19597) and build the server from source, but i still get the
following error:

  Uncaught TypeError: Cannot read property 'Object' of undefined

is there a specific subversion revision where this was fixed?

thanks,
augusto.

Bug within Flex compiler integration when including application specific SWCs

Raju Bitter — 2012-05-08T17:21:47

There's a small bug within SWF9External.java, where the Flex
compilation process is looking for a possible "flexlib" folder in the
wrong directory. If that would be fixed, SWCs would not have to be
added to the WEB-INF/flexlib folder, but could be in a "flexlib"
subfolder relative to the canvas LZX file (main application file).
http://jira.openlaszlo.org/jira/browse/LPP-10183

I only need to solve one problem for that: is there a way to access
the absolute path of the folder containing the canvas LZX file from
within the SWF9External class?

-  Raju

Re: Correct value for < at >subtopic in swf11 kernel

Raju Bitter — 2012-05-07T22:18:03

That's what I guessed, thanks Tucker!

- Raju