You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@flex.apache.org by ma...@apache.org on 2013/10/14 22:21:46 UTC
git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix
https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid column resi
Updated Branches:
refs/heads/develop e376a1e5d -> fca2a3812
UNDO FIX https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when scrolled and dataProvider is changed )
UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid column resize and sort bug when releasing mouse outside of headers)
Project: http://git-wip-us.apache.org/repos/asf/flex-sdk/repo
Commit: http://git-wip-us.apache.org/repos/asf/flex-sdk/commit/fca2a381
Tree: http://git-wip-us.apache.org/repos/asf/flex-sdk/tree/fca2a381
Diff: http://git-wip-us.apache.org/repos/asf/flex-sdk/diff/fca2a381
Branch: refs/heads/develop
Commit: fca2a38124a2f93bdf8868b008b845553919279c
Parents: e376a1e
Author: mamsellem <ma...@systar.com>
Authored: Mon Oct 14 22:21:17 2013 +0200
Committer: mamsellem <ma...@systar.com>
Committed: Mon Oct 14 22:21:17 2013 +0200
----------------------------------------------------------------------
.../projects/spark/src/spark/components/Grid.as | 23 +++-----------------
.../spark/components/GridColumnHeaderGroup.as | 4 +---
2 files changed, 4 insertions(+), 23 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/flex-sdk/blob/fca2a381/frameworks/projects/spark/src/spark/components/Grid.as
----------------------------------------------------------------------
diff --git a/frameworks/projects/spark/src/spark/components/Grid.as b/frameworks/projects/spark/src/spark/components/Grid.as
index 847fa1d..9dcfbd3 100644
--- a/frameworks/projects/spark/src/spark/components/Grid.as
+++ b/frameworks/projects/spark/src/spark/components/Grid.as
@@ -269,9 +269,7 @@ public class Grid extends Group implements IDataGridElement, IDataProviderEnhanc
* rowIndex of the caret after a collection refresh event.
*/
private var caretSelectedItem:Object = null;
- private var updateCaretForDataProviderChanged:Boolean = false;
- private var updateCaretForDataProviderChangeLastEvent:CollectionEvent;
-
+
/**
* @private
* True while updateDisplayList is running. Use to disable invalidateSize(),
@@ -4602,13 +4600,8 @@ public class Grid extends Group implements IDataGridElement, IDataProviderEnhanc
if (!variableRowHeight)
setFixedRowHeight(gridDimensions.getRowHeight(0));
- if (updateCaretForDataProviderChanged){
- updateCaretForDataProviderChanged = false;
- updateCaretForDataProviderChange(updateCaretForDataProviderChangeLastEvent);
- updateCaretForDataProviderChangeLastEvent = null;
}
- }
-
+
//--------------------------------------------------------------------------
//
// Methods
@@ -5626,19 +5619,9 @@ public class Grid extends Group implements IDataGridElement, IDataProviderEnhanc
invalidateSize();
invalidateDisplayList();
- if (caretRowIndex != -1) {
- if (event.kind == CollectionEventKind.RESET){
- // defer for reset events
- updateCaretForDataProviderChanged = true;
- updateCaretForDataProviderChangeLastEvent = event;
- invalidateDisplayList();
- }
- else {
+ if (caretRowIndex != -1)
updateCaretForDataProviderChange(event);
- }
- }
-
// Trigger bindings to selectedIndex/selectedCell/selectedItem and the plurals of those.
if (selectionChanged)
dispatchFlexEvent(FlexEvent.VALUE_COMMIT);
http://git-wip-us.apache.org/repos/asf/flex-sdk/blob/fca2a381/frameworks/projects/spark/src/spark/components/GridColumnHeaderGroup.as
----------------------------------------------------------------------
diff --git a/frameworks/projects/spark/src/spark/components/GridColumnHeaderGroup.as b/frameworks/projects/spark/src/spark/components/GridColumnHeaderGroup.as
index bd62162..51827f9 100644
--- a/frameworks/projects/spark/src/spark/components/GridColumnHeaderGroup.as
+++ b/frameworks/projects/spark/src/spark/components/GridColumnHeaderGroup.as
@@ -1064,11 +1064,9 @@ public class GridColumnHeaderGroup extends Group implements IDataGridElement
{
const eventHeaderCP:CellPosition = new CellPosition();
const eventHeaderXY:Point = new Point();
- // mouse can be released outside of header , so don't check cf. https://issues.apache.org/jira/browse/FLEX-33818
- if (event.type != MouseEvent.MOUSE_UP && !eventToHeaderLocations(event, eventHeaderCP, eventHeaderXY))
+ if (!eventToHeaderLocations(event, eventHeaderCP, eventHeaderXY))
return;
-
const eventSeparatorIndex:int = eventHeaderCP.rowIndex;
const eventColumnIndex:int = (eventSeparatorIndex == -1) ? eventHeaderCP.columnIndex : -1;
RE: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix
https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
Posted by Maurice Amsellem <ma...@systar.com>.
> You can try the opening-fb-debugger and then running the debug swf trick. Let me know if you need more details on how to do this.
Yes please.
Also, how is it possible to run only one test case ?
Eg. DataGrid_requireSelection_test001 > requireSelection_test15 instead of the full DataGrid/Properties mini_run?
Or at least, all the test cases in DataGrid_requireSelection_test001?
Thanks for your help.
Maurice
-----Message d'origine-----
De : omuppi1@gmail.com [mailto:omuppi1@gmail.com] De la part de OmPrakash Muppirala
Envoyé : lundi 14 octobre 2013 23:52
À : dev@flex.apache.org
Objet : Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
On Mon, Oct 14, 2013 at 2:42 PM, Maurice Amsellem < maurice.amsellem@systar.com> wrote:
> The WIKI says Mustella can be debugged using fdb.
> Is there a way to debug inside a given test using an IDE debugger
> (such as IDEA or FB), like in FlexUnit ?
>
> Maybe using OneClickMustella?
>
>
Not using OneClicMustella for sure. You can try the opening-fb-debugger and then running the debug swf trick. Let me know if you need more details on how to do this.
Thanks,
Om
>
> Maurice
>
> -----Message d'origine-----
> De : Maurice Amsellem [mailto:maurice.amsellem@systar.com]
> Envoyé : lundi 14 octobre 2013 23:37
> À : dev@flex.apache.org
> Objet : RE: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
> https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> when scrolled and dataProvider is changed ) UNDO Fix
> https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
>
> Thanks, I will read it.
>
> -----Message d'origine-----
> De : omuppi1@gmail.com [mailto:omuppi1@gmail.com] De la part de
> OmPrakash Muppirala Envoyé : lundi 14 octobre 2013 23:35 À :
> dev@flex.apache.orgObjet : Re: git commit: [flex-sdk]
> [refs/heads/develop] - UNDO FIX
> https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> when scrolled and dataProvider is changed ) UNDO Fix
> https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
>
> On Mon, Oct 14, 2013 at 2:31 PM, Maurice Amsellem <
> maurice.amsellem@systar.com> wrote:
>
> > Hi, I have undone the fixes, and the tests are passing, so
> > obviously, the fix have broken the tests.
> >
> > On the other hand, the manual tests I have made tells me that the
> > fixes have actually fixed the bugs they intended to.
> >
> > So maybe the tests were somehow "compensating" for the existing bugs.
> >
> > That's what I am trying to understand.
> >
> > So I am simulating in an external app that I can debug what the
> > tests
> > 14 and 15 were supposed to test, to see if I have missed something.
> >
> > Not easy at all, Mustella insiders remaining largely mysterious to
> > me
> :-(.
> >
>
> Have you seen this wiki:
> https://cwiki.apache.org/confluence/display/FLEX/Mustella+Overview ?
>
> Thanks,
> Om
>
>
> >
> > Maurice
> >
> > -----Message d'origine-----
> > De : Justin Mclean [mailto:justin@classsoftware.com] Envoyé : lundi
> > 14 octobre 2013 23:16 À : dev@flex.apache.org Objet : Re: git commit:
> > [flex-sdk] [refs/heads/develop] - UNDO FIX
> > https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes
> > blank when scrolled and dataProvider is changed ) UNDO Fix
> > https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid
> colum...
> >
> > Hi,
> >
> > Are we sure these caused the test failures? And if so assuming the
> > patches are correct isn't it better to fix the tests?
> >
> > Justin
> >
>
Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818
(Spark Datagrid colum...
Posted by OmPrakash Muppirala <bi...@gmail.com>.
On Mon, Oct 14, 2013 at 2:42 PM, Maurice Amsellem <
maurice.amsellem@systar.com> wrote:
> The WIKI says Mustella can be debugged using fdb.
> Is there a way to debug inside a given test using an IDE debugger (such as
> IDEA or FB), like in FlexUnit ?
>
> Maybe using OneClickMustella?
>
>
Not using OneClicMustella for sure. You can try the opening-fb-debugger
and then running the debug swf trick. Let me know if you need more details
on how to do this.
Thanks,
Om
>
> Maurice
>
> -----Message d'origine-----
> De : Maurice Amsellem [mailto:maurice.amsellem@systar.com]
> Envoyé : lundi 14 octobre 2013 23:37
> À : dev@flex.apache.org
> Objet : RE: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
> https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> when scrolled and dataProvider is changed ) UNDO Fix
> https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
>
> Thanks, I will read it.
>
> -----Message d'origine-----
> De : omuppi1@gmail.com [mailto:omuppi1@gmail.com] De la part de OmPrakash
> Muppirala Envoyé : lundi 14 octobre 2013 23:35 À : dev@flex.apache.orgObjet : Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
> https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> when scrolled and dataProvider is changed ) UNDO Fix
> https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
>
> On Mon, Oct 14, 2013 at 2:31 PM, Maurice Amsellem <
> maurice.amsellem@systar.com> wrote:
>
> > Hi, I have undone the fixes, and the tests are passing, so obviously,
> > the fix have broken the tests.
> >
> > On the other hand, the manual tests I have made tells me that the
> > fixes have actually fixed the bugs they intended to.
> >
> > So maybe the tests were somehow "compensating" for the existing bugs.
> >
> > That's what I am trying to understand.
> >
> > So I am simulating in an external app that I can debug what the tests
> > 14 and 15 were supposed to test, to see if I have missed something.
> >
> > Not easy at all, Mustella insiders remaining largely mysterious to me
> :-(.
> >
>
> Have you seen this wiki:
> https://cwiki.apache.org/confluence/display/FLEX/Mustella+Overview ?
>
> Thanks,
> Om
>
>
> >
> > Maurice
> >
> > -----Message d'origine-----
> > De : Justin Mclean [mailto:justin@classsoftware.com] Envoyé : lundi 14
> > octobre 2013 23:16 À : dev@flex.apache.org Objet : Re: git commit:
> > [flex-sdk] [refs/heads/develop] - UNDO FIX
> > https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> > when scrolled and dataProvider is changed ) UNDO Fix
> > https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid
> colum...
> >
> > Hi,
> >
> > Are we sure these caused the test failures? And if so assuming the
> > patches are correct isn't it better to fix the tests?
> >
> > Justin
> >
>
RE: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix
https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
Posted by Maurice Amsellem <ma...@systar.com>.
The WIKI says Mustella can be debugged using fdb.
Is there a way to debug inside a given test using an IDE debugger (such as IDEA or FB), like in FlexUnit ?
Maybe using OneClickMustella?
Maurice
-----Message d'origine-----
De : Maurice Amsellem [mailto:maurice.amsellem@systar.com]
Envoyé : lundi 14 octobre 2013 23:37
À : dev@flex.apache.org
Objet : RE: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
Thanks, I will read it.
-----Message d'origine-----
De : omuppi1@gmail.com [mailto:omuppi1@gmail.com] De la part de OmPrakash Muppirala Envoyé : lundi 14 octobre 2013 23:35 À : dev@flex.apache.org Objet : Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
On Mon, Oct 14, 2013 at 2:31 PM, Maurice Amsellem < maurice.amsellem@systar.com> wrote:
> Hi, I have undone the fixes, and the tests are passing, so obviously,
> the fix have broken the tests.
>
> On the other hand, the manual tests I have made tells me that the
> fixes have actually fixed the bugs they intended to.
>
> So maybe the tests were somehow "compensating" for the existing bugs.
>
> That's what I am trying to understand.
>
> So I am simulating in an external app that I can debug what the tests
> 14 and 15 were supposed to test, to see if I have missed something.
>
> Not easy at all, Mustella insiders remaining largely mysterious to me :-(.
>
Have you seen this wiki:
https://cwiki.apache.org/confluence/display/FLEX/Mustella+Overview ?
Thanks,
Om
>
> Maurice
>
> -----Message d'origine-----
> De : Justin Mclean [mailto:justin@classsoftware.com] Envoyé : lundi 14
> octobre 2013 23:16 À : dev@flex.apache.org Objet : Re: git commit:
> [flex-sdk] [refs/heads/develop] - UNDO FIX
> https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> when scrolled and dataProvider is changed ) UNDO Fix
> https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
>
> Hi,
>
> Are we sure these caused the test failures? And if so assuming the
> patches are correct isn't it better to fix the tests?
>
> Justin
>
RE: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix
https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
Posted by Maurice Amsellem <ma...@systar.com>.
Thanks, I will read it.
-----Message d'origine-----
De : omuppi1@gmail.com [mailto:omuppi1@gmail.com] De la part de OmPrakash Muppirala
Envoyé : lundi 14 octobre 2013 23:35
À : dev@flex.apache.org
Objet : Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
On Mon, Oct 14, 2013 at 2:31 PM, Maurice Amsellem < maurice.amsellem@systar.com> wrote:
> Hi, I have undone the fixes, and the tests are passing, so obviously,
> the fix have broken the tests.
>
> On the other hand, the manual tests I have made tells me that the
> fixes have actually fixed the bugs they intended to.
>
> So maybe the tests were somehow "compensating" for the existing bugs.
>
> That's what I am trying to understand.
>
> So I am simulating in an external app that I can debug what the tests
> 14 and 15 were supposed to test, to see if I have missed something.
>
> Not easy at all, Mustella insiders remaining largely mysterious to me :-(.
>
Have you seen this wiki:
https://cwiki.apache.org/confluence/display/FLEX/Mustella+Overview ?
Thanks,
Om
>
> Maurice
>
> -----Message d'origine-----
> De : Justin Mclean [mailto:justin@classsoftware.com] Envoyé : lundi 14
> octobre 2013 23:16 À : dev@flex.apache.org Objet : Re: git commit:
> [flex-sdk] [refs/heads/develop] - UNDO FIX
> https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> when scrolled and dataProvider is changed ) UNDO Fix
> https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
>
> Hi,
>
> Are we sure these caused the test failures? And if so assuming the
> patches are correct isn't it better to fix the tests?
>
> Justin
>
Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818
(Spark Datagrid colum...
Posted by OmPrakash Muppirala <bi...@gmail.com>.
On Mon, Oct 14, 2013 at 2:31 PM, Maurice Amsellem <
maurice.amsellem@systar.com> wrote:
> Hi, I have undone the fixes, and the tests are passing, so obviously, the
> fix have broken the tests.
>
> On the other hand, the manual tests I have made tells me that the fixes
> have actually fixed the bugs they intended to.
>
> So maybe the tests were somehow "compensating" for the existing bugs.
>
> That's what I am trying to understand.
>
> So I am simulating in an external app that I can debug what the tests 14
> and 15 were supposed to test, to see if I have missed something.
>
> Not easy at all, Mustella insiders remaining largely mysterious to me :-(.
>
Have you seen this wiki:
https://cwiki.apache.org/confluence/display/FLEX/Mustella+Overview ?
Thanks,
Om
>
> Maurice
>
> -----Message d'origine-----
> De : Justin Mclean [mailto:justin@classsoftware.com]
> Envoyé : lundi 14 octobre 2013 23:16
> À : dev@flex.apache.org
> Objet : Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
> https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank
> when scrolled and dataProvider is changed ) UNDO Fix
> https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
>
> Hi,
>
> Are we sure these caused the test failures? And if so assuming the patches
> are correct isn't it better to fix the tests?
>
> Justin
>
RE: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix
https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
Posted by Maurice Amsellem <ma...@systar.com>.
Hi, I have undone the fixes, and the tests are passing, so obviously, the fix have broken the tests.
On the other hand, the manual tests I have made tells me that the fixes have actually fixed the bugs they intended to.
So maybe the tests were somehow "compensating" for the existing bugs.
That's what I am trying to understand.
So I am simulating in an external app that I can debug what the tests 14 and 15 were supposed to test, to see if I have missed something.
Not easy at all, Mustella insiders remaining largely mysterious to me :-(.
Maurice
-----Message d'origine-----
De : Justin Mclean [mailto:justin@classsoftware.com]
Envoyé : lundi 14 octobre 2013 23:16
À : dev@flex.apache.org
Objet : Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
Hi,
Are we sure these caused the test failures? And if so assuming the patches are correct isn't it better to fix the tests?
Justin
Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX
https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when
scrolled and dataProvider is changed ) UNDO Fix
https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid colum...
Posted by Alex Harui <ah...@adobe.com>.
On 10/14/13 2:16 PM, "Justin Mclean" <ju...@classsoftware.com> wrote:
>Hi,
>
>And if so assuming the patches are correct isn't it better to fix the
>tests?
IMO, that's not a safe assumption to make. And while investigating
whether the patches are correct it is best to revert so it doesn't block
building another RC. Just because the patch fixes the problem in a test
case doesn't mean it couldn't have side effects. We've seen that happen
several times recently.
For example, I was pondering changing my vote on 4.11 back to -1 because
it occurred to me that the scenario in the radiobutton focus ring test may
not be as "unreasonable" as I originally thought. Mark Goldin's thread on
RadioButtonGroup made me realize that people may have multiple sets of
radio buttons all with the same group name if they specify sets of
radiobuttons in different MXML files or template a group in an MXML
component. In theory, there should only be a problem if there is no
selected radiobutton in each set of radiobuttons and I think most folks
should have a default selection in each radiobutton group, but what do
others think?
-Alex
Re: git commit: [flex-sdk] [refs/heads/develop] - UNDO FIX https://issues.apache.org/jira/browse/FLEX-33813 (DataGrid goes blank when scrolled and dataProvider is changed ) UNDO Fix https://issues.apache.org/jira/browse/FLEX-33818 (Spark Datagrid column resi
Posted by Justin Mclean <ju...@classsoftware.com>.
Hi,
Are we sure these caused the test failures? And if so assuming the patches are correct isn't it better to fix the tests?
Justin