Hi Lukas: There are 4 tests that were failing in Windows and I debugged them. The tests are the 4 tests of the class OBKeyBindingsTest
The problem in windows is this. In your tests, you do: modifier: mod keycode: code ^ KeyboardEvent new setType: 'keystroke' buttons: mod position: 123 @ 456 keyValue: code charCode: code hand: nil stamp: nil As you can see, you don't set a codeScan (it is nil). In Paragraph >> dispatchOnKeyEvent: keyEvent with: typeAheadStream You do "OSPlatform current virtualKey: keyEvent scanCode." In mac, everything is ok as it uses the method OSPlatform >> virtualKey: virtualKeyCode "Subclass responsibility to override if necessary" ^nil But in Windows, it uses: virtualKey: virtualKeyCode "Win32Platform virtualKey: $C charCode" (virtualKeyCode <= 90 "$Z charCode" and: [virtualKeyCode >= 65 "$A "]) ifFalse: [^nil]. "#($a $b $c $d $e $f $g $h $i $j $k $l $m $n $o $p $q $r $s $t $u $v $w $x $y $z)" ^(#($a nil $c $d nil $f $g nil nil nil nil $l $m $n nil $p nil nil $s nil nil $v nil $x nil $z) at: virtualKeyCode-64) ifNotNil: [:char | char charCode] There, is a DNU as nil (virtualKeyCode) dnu: <= So, the solution I think is to set a scanCode in the keyEvents you generate in the tests. I don't know which number is correct for each type of event, but I tried just with 0 and tests are green. modifier: mod keycode: code ^ KeyboardEvent new setType: 'keystroke' buttons: mod position: 123 @ 456 keyValue: code charCode: code hand: nil stamp: nil; scanCode: 0 What do you think? is there a better fix ? Cheers Mariano _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
Thank you Mariano for debugging this.
It looks to me that the variable scanCode is indeed only used on Windows, so I guess it is save to change that. I've committed the fix you suggested: Name: OB-Tests-Morphic-lr.25 Author: lr Time: 27 January 2010, 10:20:05 pm UUID: 0987660a-6184-490a-abc0-8d23352c334f Ancestors: OB-Tests-Morphic-lr.24 - added the suggested fix of mariano for the tests (windows only) Lukas 2010/1/27 Mariano Martinez Peck <[hidden email]>: > Hi Lukas: There are 4 tests that were failing in Windows and I debugged > them. The tests are the 4 tests of the class OBKeyBindingsTest > > The problem in windows is this. In your tests, you do: > > modifier: mod keycode: code > ^ KeyboardEvent new > setType: 'keystroke' > buttons: mod > position: 123 @ 456 > keyValue: code > charCode: code > hand: nil > stamp: nil > > As you can see, you don't set a codeScan (it is nil). > > In Paragraph >> dispatchOnKeyEvent: keyEvent with: typeAheadStream > > You do "OSPlatform current virtualKey: keyEvent scanCode." > > In mac, everything is ok as it uses the method > > OSPlatform >> virtualKey: virtualKeyCode > "Subclass responsibility to override if necessary" > ^nil > > > But in Windows, it uses: > > virtualKey: virtualKeyCode > "Win32Platform virtualKey: $C charCode" > > (virtualKeyCode <= 90 "$Z charCode" > and: [virtualKeyCode >= 65 "$A "]) > ifFalse: [^nil]. > "#($a $b $c $d $e $f $g $h $i $j $k $l $m $n $o $p $q $r $s $t $u $v $w > $x $y $z)" > > ^(#($a nil $c $d nil $f $g nil nil nil nil $l $m $n nil $p nil nil $s > nil nil $v nil $x nil $z) > at: virtualKeyCode-64) ifNotNil: [:char | char charCode] > > > There, is a DNU as nil (virtualKeyCode) dnu: <= > > So, the solution I think is to set a scanCode in the keyEvents you generate > in the tests. I don't know which number is correct for each type of event, > but I tried just with 0 and tests are green. > > modifier: mod keycode: code > ^ KeyboardEvent new > setType: 'keystroke' > buttons: mod > position: 123 @ 456 > keyValue: code > charCode: code > hand: nil > stamp: nil; > scanCode: 0 > > > What do you think? is there a better fix ? > > Cheers > > Mariano > > > > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- Lukas Renggli http://www.lukas-renggli.ch _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
On Wed, Jan 27, 2010 at 10:20 PM, Lukas Renggli <[hidden email]> wrote: Thank you Mariano for debugging this. You are welcome. Step by step Pharo tests are getting greener :) It looks to me that the variable scanCode is indeed only used on Thanks!!! Name: OB-Tests-Morphic-lr.25 _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
Free forum by Nabble | Edit this page |