For Review: Change 20080430-hqm-w Summary: changes to get lzpix to compile

Change 20080430-hqm-w by hqm <at> badtzmaru.home on 2008-04-30 19:10:59 EDT
     in /Users/hqm/openlaszlo/trunk5
     for http://svn.openlaszlo.org/openlaszlo/trunk

Summary: changes to get lzpix to compile

New Features:

Bugs Fixed:

Technical Reviewer: ptw
QA Reviewer: max
Doc Reviewer: (pending)

Documentation:

Release Notes:

Details:

+ make services/LzTrack.lzs compile in the swf9 LFC

note: I had to add these instance vars:
   var __LZdeferDelegates = false;
   var __LZdeleted = false;
because sendEvent was expecting to be able to access them when
sending and event to LzTrack

+ add LzNode-compatible constructors to LzSelectionManager and  
LzDataSelectionManager

Re: what is __LZUserClassPlacementObject, and who is looking up a class for that tag?

what is __LZUserClassPlacementObject, and who is looking up a class for that tag?

For Review: Change 20080430-hqm-w Summary: changes to get lzpix to compile

updated Change 20080430-hqm-w by hqm <at> badtzmaru.home on 2008-04-30  
19:10:59 EDT
     in /Users/hqm/openlaszlo/trunk5
     for http://svn.openlaszlo.org/openlaszlo/trunk

Summary: changes to get lzpix to compile

New Features:

Bugs Fixed:

Technical Reviewer: ptw
QA Reviewer: max
Doc Reviewer: (pending)

Documentation:

Release Notes:

Details:

+ make services/LzTrack.lzs compile in the swf9 LFC

note: I had to add these instance vars:
   var __LZdeferDelegates = false;
   var __LZdeleted = false;
because sendEvent was expecting to be able to access them when
sending and event to LzTrack

+ add LzNode-compatible constructors to LzSelectionManager and  
LzDataSelectionManager

+ removed what looks like an unused var erronesouly being cast to  
LzView in LzAnimatorGroup

+ Needed to make the "ignore" arg to the binder function and optional  
arg, for somer reason.
Maybe we should look at what code is being compiled to call this with  
no args.

+ fix up various layout methods to take optional args if they can be  
called by event handlers

+ app.lzx: fix up various callers that were too few args
  Note: we ought to rename the "stop" method defined in there as it  
conflicts with the LzView.stop() method

+ demos/lzpix/classes/photo.lzx
add some missing var declarations
+ ditto  demos/lzpix/classes/draggedphotos.lzx

+ LzNode, unify the ConstructorMap["__LZUserClassPlacementObject"]  
hackage

Tests:

Files:
M      data/LzParam.lzs
M      WEB-INF/lps/lfc/core/LzNode.lzs
M      WEB-INF/lps/lfc/services/LzTrack.lzs
M      WEB-INF/lps/lfc/services/Library.lzs
M      WEB-INF/lps/lfc/core/dummyclasses.js
M      WEB-INF/lps/lfc/helpers/LzDataSelectionManager.lzs
M      WEB-INF/lps/lfc/helpers/LzSelectionManager.lzs
M      WEB-INF/lps/lfc/controllers/LzAnimatorGroup.lzs
M      WEB-INF/lps/server/src/org/openlaszlo/compiler/NodeModel.java
M      lps/components/utils/layouts/stableborderlayout.lzx
M      demos/lzpix/app.lzx
M      demos/lzpix/classes/search.lzx
M      demos/lzpix/classes/dataman.lzx
M      demos/lzpix/classes/clipboardinterior.lzx
M      demos/lzpix/classes/md5.js
M      demos/lzpix/classes/photo.lzx
M      demos/lzpix/classes/draggedphotos.lzx
M      demos/lzpix/classes/classes.lzx
M      demos/lzpix/classes/albumlayout.lzx
M      demos/lzpix/classes/pivotlayout.lzx
M      demos/lzpix/classes/clipboard.lzx
M      demos/lzpix/classes/favorites.lzx
M      demos/lzpix/views/tools.lzx
M      demos/lzpix/views/mybutton.lzx

Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20080430-hqm-w.tar

For Review: Change 20080430-maxcarlson-e Summary: Add 'cachebitmap' attribute to view

Change 20080430-maxcarlson-e by maxcarlson <at> Roboto on 2008-04-30  
22:51:46 PDT
     in /Users/maxcarlson/openlaszlo/trunk-clean
     for http://svn.openlaszlo.org/openlaszlo/trunk

Summary: Add 'cachebitmap' attribute to view

New Features:

Bugs Fixed: LPP-5904 - Add bitmap caching API

Technical Reviewer: promanik
QA Reviewer: hminsky
Doc Reviewer: (pending)

Documentation:

Release Notes: Setting the cachebitmap attribute to true will tunr on  
bitmap caching for views.  Note that text defaults to  
cachebitmap='true'.  This currently works only in Flash.

Details: lfc.lzx - Add setter method and attribute name.

swf/LzTextSprite.as - Default  cachebitmap to true instead of setting  
directly.

swf/LzSprite.as - Set bitmapcaching capability to true.  Turn off  
bitmap caching for clipped views, and warn in $debug mode.  Add  
setBitmapCache()  and getContext() implementations.

dhtml/LzSprite.js - Set bitmapcaching capability to false.  Add  
getContext() placeholder method.

swf9/LzSprite.as - Set bitmapcaching capability to true.  Add  
setBitmapCache()  and getContext() implementations.  Update getMCRef()  
to return the proper reference.

LaszloView.lzs - Add cachebitmap class attribute and setter.

drawview.lzx - Use inherited cachebitmap property.  Default  
cachebitmap to true.  Use sprite.getContext() API to get graphics  
context.  Rely on sprite warnings when clipping is on and cachebitmap  
is true.

Tests: /test/drawing/drawing.lzx runs as it did before in swf 7,8,9  
and dhtml.

Files:
M      WEB-INF/lps/schema/lfc.lzx
M      WEB-INF/lps/lfc/kernel/swf/LzTextSprite.as
M      WEB-INF/lps/lfc/kernel/swf/LzSprite.as
M      WEB-INF/lps/lfc/kernel/dhtml/LzSprite.js
M      WEB-INF/lps/lfc/kernel/swf9/LzSprite.as
M      WEB-INF/lps/lfc/views/LaszloView.lzs
M      lps/components/extensions/drawview.lzx

Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20080430-maxcarlson-e.tar

[RESENT] One thing you could do in the conversion script

[RESENT: mail to max <at> openlaszlo.org was rejected]

When splitting a <method event= name=> into a handler and a method, if  
the original did not take any arguments give it args="ignore".

Re: For Review: Change 20080430-hqm-w Summary: changes to get lzpix to compile

Approved.

But please file Jira Bugs for each of these issues you raise:

1)
> note: I had to add these instance vars:
>  var __LZdeferDelegates = false;
>  var __LZdeleted = false;

It sounds like we should move __LZdeferDelegates and __LZdeleted up to  
LzMiniNode (should that be called DOMNode or something?) and say that  
the first argument to LzDelegate has to be an LzMiniNode.  Delegates  
and Events have more of a protocol now than they used to.

2)
> + Needed to make the "ignore" arg to the binder function and  
> optional arg, for somer reason.
> Maybe we should look at what code is being compiled to call this  
> with no args.

This is surely a bug that we need to track down.  The binder protocol  
is totally internal, so there should not be any need of an optional  
argument -- the contract between the caller and callee is fixed.

3)
> Note: we ought to rename the "stop" method defined in there as it  
> conflicts with the LzView.stop() method

Perhaps assign to Josh?

On 2008-04-30, at 19:24 EDT, Henry Minsky wrote:

> Change 20080430-hqm-w by hqm <at> badtzmaru.home on 2008-04-30 19:10:59 EDT
>    in /Users/hqm/openlaszlo/trunk5
>    for http://svn.openlaszlo.org/openlaszlo/trunk
>
> Summary: changes to get lzpix to compile
>
> New Features:
>
> Bugs Fixed:
>
> Technical Reviewer: ptw
> QA Reviewer: max
> Doc Reviewer: (pending)
>
> Documentation:
>
> Release Notes:
>
> Details:
>
>
> + make services/LzTrack.lzs compile in the swf9 LFC
>
> note: I had to add these instance vars:
>  var __LZdeferDelegates = false;
>  var __LZdeleted = false;
> because sendEvent was expecting to be able to access them when
> sending and event to LzTrack
>
>
>
>
> + add LzNode-compatible constructors to LzSelectionManager and  
> LzDataSelectionManager
>
> + removed what looks like an unused var erronesouly being cast to  
> LzView in LzAnimatorGroup
>
> + Needed to make the "ignore" arg to the binder function and  
> optional arg, for somer reason.
> Maybe we should look at what code is being compiled to call this  
> with no args.
>
>
> + fix up various layout methods to take optional args if they can be  
> called by event handlers
>
> + app.lzx: fix up various callers that were too few args
> Note: we ought to rename the "stop" method defined in there as it  
> conflicts with the LzView.stop() method
>
> + demos/lzpix/classes/photo.lzx
> add some missing var declarations
> + ditto  demos/lzpix/classes/draggedphotos.lzx
>
>
>
>
> Tests:
>
> Files:
> M      WEB-INF/lps/lfc/services/LzTrack.lzs
> M      WEB-INF/lps/lfc/services/Library.lzs
> M      WEB-INF/lps/lfc/core/dummyclasses.js
> M      WEB-INF/lps/lfc/helpers/LzDataSelectionManager.lzs
> M      WEB-INF/lps/lfc/helpers/LzSelectionManager.lzs
> M      WEB-INF/lps/lfc/controllers/LzAnimatorGroup.lzs
> M      WEB-INF/lps/server/src/org/openlaszlo/compiler/NodeModel.java
> M      lps/components/utils/layouts/stableborderlayout.lzx
> M      demos/lzpix/app.lzx
> M      demos/lzpix/classes/search.lzx
> M      demos/lzpix/classes/dataman.lzx
> M      demos/lzpix/classes/clipboardinterior.lzx
> M      demos/lzpix/classes/md5.js
> M      demos/lzpix/classes/photo.lzx
> M      demos/lzpix/classes/draggedphotos.lzx
> M      demos/lzpix/classes/classes.lzx
> M      demos/lzpix/classes/albumlayout.lzx
> M      demos/lzpix/classes/pivotlayout.lzx
> M      demos/lzpix/classes/clipboard.lzx
> M      demos/lzpix/classes/favorites.lzx
> M      demos/lzpix/views/tools.lzx
> M      demos/lzpix/views/mybutton.lzx
>
> Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20080430-hqm-w.tar

For Review: Change 20080501-lou-a Summary: interim checkin for 4.2a3 release notes

Change 20080501-lou-a by lou <at> loumac.local on 2008-05-01 09:19:22 AST
     in /Users/lou/src/svn/openlaszlo/trunk
     for http://svn.openlaszlo.org/openlaszlo/trunk

Summary: interim checkin for 4.2a3 release notes

Bugs Fixed: LPP-5903 (partial)

Technical Reviewer: (pending)
QA Reviewer: (pending)
Doc Reviewer: (pending)

Details: add new sections for 4.2a3 new features (pending) and fixed  
bugs. May need to add more bugs if and when they get fixed.

Tests: visual verify

Files:
M      docs/release-notes.html

Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20080501-lou- 
a.tar

Re: what is __LZUserClassPlacementObject, and who is looking up a class for that tag?

You have conditionalized the check in makeChild to require that tags  
be implemented by Class objects, not Function.  Is this a strict  
requirement in AS3?  Is Class not a subclass of Function?

Maybe this is the way we want to go.  If so, yes, you will have to  
make this kludge a class.  We ought to unify the test in makeChild so  
that you can't accidentally write a non-class tag implementation in  
the sloppier run times...

On 2008-04-30, at 22:42 EDT, Henry Minsky wrote:

> Oh  I see what is happening, in NodeModel.java; the fake-out  
> placement node
> is a tag whose constructor stuffs the placement
> value into the parent class.
>
> Hmm, well this ought to work in swf9 as well, but something
> seems to not work when I just use a function per the UserClass.lzs  
> code:
>
> ConstructorMap.__LZUserClassPlacementObject = function
> __LZUserClassPlacementObject ( parent , placement ){
>    parent.defaultplacement = placement;
> };
>
> Maybe I better try using a real class instead of a constructor  
> function.
>
>
>
> On Wed, Apr 30, 2008 at 10:35 PM, Henry Minsky <hminsky <at> laszlosystems.com 
> >
> wrote:
>
>> In swf9 running the lzpix app, I get these warnings printed to the
>> debugger:
>>
>> [trace] Error:  Class for tag  __LZUserClassPlacementObject  has  
>> not been
>> defined yet function Function() {}
>> [trace] Error:  Class for tag  __LZUserClassPlacementObject  has  
>> not been
>> defined yet function Function() {}
>> [trace] setShowHandCursor not currently implemented in swf9.
>> [trace] Error:  Class for tag  __LZUserClassPlacementObject  has  
>> not been
>> defined yet function Function() {}
>>
>> Anyone know what is trying to look these up ? I see there is  
>> something in
>> UserClass.lzs which defines
>> an entry in the constructormap, which won't be present in swf9, but  
>> who is
>> actually looking this up at runtime?
>>
>>
>> --
>> Henry Minsky
>> Software Architect
>> hminsky <at> laszlosystems.com
>>
>>
>
>
> -- 
> Henry Minsky
> Software Architect
> hminsky <at> laszlosystems.com

Re: more state hair, reference to "classroot" from a state method

Clearly that's not acceptable.  We have to make this work without the  
LZX user having to add lot's of 'this's.

LPP-5624 is the placeholder for an eventual solution to the problem.   
Feel free to comment there.

On 2008-04-30, at 17:12 EDT, Philip Romanik wrote:

> Hi Henry,
>
> When I was playing around with the contactlist demo, I had to add  
> 'this' all over the place to make these errors go away.
>
>
>
>
>
>> photo.lzx:
>> This code in the lzpix app causes some compile time errors like this:
>>
>> [/tmp/lzswf9/lzgen49044/$lzc$class_state_$$2E$2E$2F$2E$2E$2Fdemos 
>> $2Flzpix$2Fclasses$2Fphoto$2Elzx_163_54.as: 10] line 163: Error:  
>> Access of undefined property classroot.
>>
>>
>>        <view name="intparent" clip="true"  
>> x="$once{ classroot.border }"
>>              y="$once{ classroot.border }">
>>            <view name="interior" bgcolor="0xd9d9d9" stretches="both">
>>                <state apply="${classroot.doesdrag}">
>>                    <handler name="onmousedown">
>>                        classroot.onmousedown.sendEvent();
>>                    </handler>
>>                    <handler name="onmouseup">
>>                        classroot.onmouseup.sendEvent();
>>                    </handler>
>>                </state>
>>            </view>
>>        </view>
>>
>>
>> because "classroot" is not an instance var of LzState. Both the  
>> constraint and it's dependency function
>> need to reference "classroot".
>>
>> Making these into references to  "this.classroot" makes the  
>> compiler happy, and presumably
>> that would work at runtime because "this" would be bound  
>> dynamically to the instance which owns the state?
>> yikes.
>>
>> --
>> Henry Minsky
>> Software Architect
>> <mailto:hminsky <at> laszlosystems.com>hminsky <at> laszlosystems.com