build 50588 strange font failure?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

build 50588 strange font failure?

Ben Coman
Not sure if this is a once-off, but just reporting since its a bit strange.

Into a newly minted 50588 image I had recently loaded
Slice-Issue-17373...BenComan.12.mcz
which however only introduces new code with no modification to
existing code (except minor re-factoring of some tests)

I was happily editing a method in the Browser when the problem
occured.  Its okay if I type within an existing word, but with space
either side of the cursor, typing cause a primitive failure per the
stack below.   "....." notes some values.

This occurs even from a new Browser window, on any method.

In system settings, unticking and reticking [Use Free type...]
causes red-pane-of-death on the problematic code panes with similar
stack as below,
but the problem is fixed in newly opened Browser windows.  (btw the
font had changed)

Array class(Object)>>primitiveFailed:
Array class(Object)>>primitiveFailed
Array class(Behavior)>>basicNew:
Array class>>new:
   .....sizeRequested --> nil
FreeTypeFace(FT2Face)>>charmaps
   .....numCharmaps --> nil
FreeTypeFont>>getLinearWidthOf:
   .....self face charmaps --> a FreeTypeFace with numCharmaps --> nil
[ self getLinearWidthOf: aCharacter ] in FreeTypeFont>>linearWidthOf:
in Block: [ self getLinearWidthOf: aCharacter ]
FreeTypeCache>>atFont:charCode:type:ifAbsentPut:
FreeTypeFont>>linearWidthOf:
FreeTypeFont>>widthOf:
LogicalFont>>widthOf:
RubCompositionScanner(RubCharacterScanner)>>setFont
RubCompositionScanner>>setFont
RubCompositionScanner>>setStopConditions
RubCompositionScanner>>endOfRun
RubCompositionScanner>>composeFrom:inRectangle:firstLine:leftSide:rightSide:
RubTextComposer>>composeEachRectangleIn:
RubTextComposer>>composeAllRectangles:
RubTextComposer>>composeOneLine
RubTextComposer>>composeAllLines
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:textStyle:text:container:
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:
RubTextComposer>>recomposeFrom:to:delta:
[ self composer recomposeFrom: start to: stop delta: delta ] in
RubParagraph>>recomposeFrom:to:delta: in Block: [ self composer
recomposeFrom: start to: stop delt...etc...
RubParagraph>>disableDrawingWhile:
RubParagraph>>recomposeFrom:to:delta:
Message>>sendTo:
[ aMessage sendTo: next ] in
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta: in Block: [ aMessage sendTo: next ]
BlockClosure>>on:do:
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta:

In case its useful, I uploaded a fuel dump of the stack here...
http://wikisend.com/download/365684/fuel.zip
expires in 7 days, zip password = pharo

/home/ben/.local/share/Pharo/VMs/spur/pharo-vm/pharo
CoInterpreter VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
StackToRegisterMappingCogit VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
https://github.com/estebanlm/pharo-vm.git Commit:
21ec004cce7d26010c18d357c805a0e1a4ffe376 Date: 2016-01-14 11:42:33
+0100 By: Esteban Lorenzano <[hidden email]> Jenkins build #498

cheers -ben

Reply | Threaded
Open this post in threaded view
|

Re: build 50588 strange font failure?

Nicolai Hess-3-2


2016-02-18 15:12 GMT+01:00 Ben Coman <[hidden email]>:
Not sure if this is a once-off, but just reporting since its a bit strange.

Into a newly minted 50588 image I had recently loaded
Slice-Issue-17373...BenComan.12.mcz
which however only introduces new code with no modification to
existing code (except minor re-factoring of some tests)

I was happily editing a method in the Browser when the problem
occured.  Its okay if I type within an existing word, but with space
either side of the cursor, typing cause a primitive failure per the
stack below.   "....." notes some values.

This occurs even from a new Browser window, on any method.

In system settings, unticking and reticking [Use Free type...]
causes red-pane-of-death on the problematic code panes with similar
stack as below,
but the problem is fixed in newly opened Browser windows.  (btw the
font had changed)

Array class(Object)>>primitiveFailed:
Array class(Object)>>primitiveFailed
Array class(Behavior)>>basicNew:
Array class>>new:
   .....sizeRequested --> nil
FreeTypeFace(FT2Face)>>charmaps
   .....numCharmaps --> nil
FreeTypeFont>>getLinearWidthOf:
   .....self face charmaps --> a FreeTypeFace with numCharmaps --> nil

I see this issue quite often, not related to any special image version (happens since spur/ffi changes).
Happens randomly, most of the time it is enough to reopen the image (of course, you lost all changes).

 
[ self getLinearWidthOf: aCharacter ] in FreeTypeFont>>linearWidthOf:
in Block: [ self getLinearWidthOf: aCharacter ]
FreeTypeCache>>atFont:charCode:type:ifAbsentPut:
FreeTypeFont>>linearWidthOf:
FreeTypeFont>>widthOf:
LogicalFont>>widthOf:
RubCompositionScanner(RubCharacterScanner)>>setFont
RubCompositionScanner>>setFont
RubCompositionScanner>>setStopConditions
RubCompositionScanner>>endOfRun
RubCompositionScanner>>composeFrom:inRectangle:firstLine:leftSide:rightSide:
RubTextComposer>>composeEachRectangleIn:
RubTextComposer>>composeAllRectangles:
RubTextComposer>>composeOneLine
RubTextComposer>>composeAllLines
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:textStyle:text:container:
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:
RubTextComposer>>recomposeFrom:to:delta:
[ self composer recomposeFrom: start to: stop delta: delta ] in
RubParagraph>>recomposeFrom:to:delta: in Block: [ self composer
recomposeFrom: start to: stop delt...etc...
RubParagraph>>disableDrawingWhile:
RubParagraph>>recomposeFrom:to:delta:
Message>>sendTo:
[ aMessage sendTo: next ] in
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta: in Block: [ aMessage sendTo: next ]
BlockClosure>>on:do:
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta:

In case its useful, I uploaded a fuel dump of the stack here...
http://wikisend.com/download/365684/fuel.zip
expires in 7 days, zip password = pharo

/home/ben/.local/share/Pharo/VMs/spur/pharo-vm/pharo
CoInterpreter VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
StackToRegisterMappingCogit VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
https://github.com/estebanlm/pharo-vm.git Commit:
21ec004cce7d26010c18d357c805a0e1a4ffe376 Date: 2016-01-14 11:42:33
+0100 By: Esteban Lorenzano <[hidden email]> Jenkins build #498

cheers -ben


Reply | Threaded
Open this post in threaded view
|

Re: build 50588 strange font failure?

EstebanLM
it is reported… something weird is going on with the FT plugin and I still cannot catch it… but I’m on it.

Esteban

On 18 Feb 2016, at 15:26, Nicolai Hess <[hidden email]> wrote:



2016-02-18 15:12 GMT+01:00 Ben Coman <[hidden email]>:
Not sure if this is a once-off, but just reporting since its a bit strange.

Into a newly minted 50588 image I had recently loaded
Slice-Issue-17373...BenComan.12.mcz
which however only introduces new code with no modification to
existing code (except minor re-factoring of some tests)

I was happily editing a method in the Browser when the problem
occured.  Its okay if I type within an existing word, but with space
either side of the cursor, typing cause a primitive failure per the
stack below.   "....." notes some values.

This occurs even from a new Browser window, on any method.

In system settings, unticking and reticking [Use Free type...]
causes red-pane-of-death on the problematic code panes with similar
stack as below,
but the problem is fixed in newly opened Browser windows.  (btw the
font had changed)

Array class(Object)>>primitiveFailed:
Array class(Object)>>primitiveFailed
Array class(Behavior)>>basicNew:
Array class>>new:
   .....sizeRequested --> nil
FreeTypeFace(FT2Face)>>charmaps
   .....numCharmaps --> nil
FreeTypeFont>>getLinearWidthOf:
   .....self face charmaps --> a FreeTypeFace with numCharmaps --> nil

I see this issue quite often, not related to any special image version (happens since spur/ffi changes).
Happens randomly, most of the time it is enough to reopen the image (of course, you lost all changes).

 
[ self getLinearWidthOf: aCharacter ] in FreeTypeFont>>linearWidthOf:
in Block: [ self getLinearWidthOf: aCharacter ]
FreeTypeCache>>atFont:charCode:type:ifAbsentPut:
FreeTypeFont>>linearWidthOf:
FreeTypeFont>>widthOf:
LogicalFont>>widthOf:
RubCompositionScanner(RubCharacterScanner)>>setFont
RubCompositionScanner>>setFont
RubCompositionScanner>>setStopConditions
RubCompositionScanner>>endOfRun
RubCompositionScanner>>composeFrom:inRectangle:firstLine:leftSide:rightSide:
RubTextComposer>>composeEachRectangleIn:
RubTextComposer>>composeAllRectangles:
RubTextComposer>>composeOneLine
RubTextComposer>>composeAllLines
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:textStyle:text:container:
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:
RubTextComposer>>recomposeFrom:to:delta:
[ self composer recomposeFrom: start to: stop delta: delta ] in
RubParagraph>>recomposeFrom:to:delta: in Block: [ self composer
recomposeFrom: start to: stop delt...etc...
RubParagraph>>disableDrawingWhile:
RubParagraph>>recomposeFrom:to:delta:
Message>>sendTo:
[ aMessage sendTo: next ] in
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta: in Block: [ aMessage sendTo: next ]
BlockClosure>>on:do:
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta:

In case its useful, I uploaded a fuel dump of the stack here...
http://wikisend.com/download/365684/fuel.zip
expires in 7 days, zip password = pharo

/home/ben/.local/share/Pharo/VMs/spur/pharo-vm/pharo
CoInterpreter VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
StackToRegisterMappingCogit VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
https://github.com/estebanlm/pharo-vm.git Commit:
21ec004cce7d26010c18d357c805a0e1a4ffe376 Date: 2016-01-14 11:42:33
+0100 By: Esteban Lorenzano <[hidden email]> Jenkins build #498

cheers -ben

Reply | Threaded
Open this post in threaded view
|

Re: build 50588 strange font failure?

monty-3

Saving the image and clicking something before it completes triggers it for me
 

Sent: Thursday, February 18, 2016 at 10:49 AM
From: "Esteban Lorenzano" <[hidden email]>
To: "Pharo Development List" <[hidden email]>
Subject: Re: [Pharo-dev] build 50588 strange font failure?

it is reported… something weird is going on with the FT plugin and I still cannot catch it… but I’m on it.
 
Esteban
 

On 18 Feb 2016, at 15:26, Nicolai Hess <[hidden email]> wrote: 
 
2016-02-18 15:12 GMT+01:00 Ben Coman <[hidden email][[hidden email]]>:Not sure if this is a once-off, but just reporting since its a bit strange.

Into a newly minted 50588 image I had recently loaded
Slice-Issue-17373...BenComan.12.mcz
which however only introduces new code with no modification to
existing code (except minor re-factoring of some tests)

I was happily editing a method in the Browser when the problem
occured.  Its okay if I type within an existing word, but with space
either side of the cursor, typing cause a primitive failure per the
stack below.   "....." notes some values.

This occurs even from a new Browser window, on any method.

In system settings, unticking and reticking [Use Free type...]
causes red-pane-of-death on the problematic code panes with similar
stack as below,
but the problem is fixed in newly opened Browser windows.  (btw the
font had changed)

Array class(Object)>>primitiveFailed:
Array class(Object)>>primitiveFailed
Array class(Behavior)>>basicNew:
Array class>>new:
   .....sizeRequested --> nil
FreeTypeFace(FT2Face)>>charmaps
   .....numCharmaps --> nil
FreeTypeFont>>getLinearWidthOf:
   .....self face charmaps --> a FreeTypeFace with numCharmaps --> nil
 
I see this issue quite often, not related to any special image version (happens since spur/ffi changes).
Happens randomly, most of the time it is enough to reopen the image (of course, you lost all changes).
 [ self getLinearWidthOf: aCharacter ] in FreeTypeFont>>linearWidthOf:
in Block: [ self getLinearWidthOf: aCharacter ]
FreeTypeCache>>atFont:charCode:type:ifAbsentPut:
FreeTypeFont>>linearWidthOf:
FreeTypeFont>>widthOf:
LogicalFont>>widthOf:
RubCompositionScanner(RubCharacterScanner)>>setFont
RubCompositionScanner>>setFont
RubCompositionScanner>>setStopConditions
RubCompositionScanner>>endOfRun
RubCompositionScanner>>composeFrom:inRectangle:firstLine:leftSide:rightSide:
RubTextComposer>>composeEachRectangleIn:
RubTextComposer>>composeAllRectangles:
RubTextComposer>>composeOneLine
RubTextComposer>>composeAllLines
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:textStyle:text:container:
RubTextComposer>>composeLinesFrom:to:delta:into:priorLines:atY:
RubTextComposer>>recomposeFrom:to:delta:
[ self composer recomposeFrom: start to: stop delta: delta ] in
RubParagraph>>recomposeFrom:to:delta: in Block: [ self composer
recomposeFrom: start to: stop delt...etc...
RubParagraph>>disableDrawingWhile:
RubParagraph>>recomposeFrom:to:delta:
Message>>sendTo:
[ aMessage sendTo: next ] in
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta: in Block: [ aMessage sendTo: next ]
BlockClosure>>on:do:
RubOpeningClosingDelimiterDecorator(RubParagraphDecorator)>>doesNotUnderstand:
#recomposeFrom:to:delta:

In case its useful, I uploaded a fuel dump of the stack here...
http://wikisend.com/download/365684/fuel.zip[http://wikisend.com/download/365684/fuel.zip]
expires in 7 days, zip password = pharo

/home/ben/.local/share/Pharo/VMs/spur/pharo-vm/pharo
CoInterpreter VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
StackToRegisterMappingCogit VMMaker.oscog-eem.1630 uuid:
2ed025ea-f400-4440-8e8b-5aa46d06c9ab Jan 14 2016
https://github.com/estebanlm/pharo-vm.git[https://github.com/estebanlm/pharo-vm.git] Commit:
21ec004cce7d26010c18d357c805a0e1a4ffe376 Date: 2016-01-14 11:42:33
+0100 By: Esteban Lorenzano <[hidden email][[hidden email]]> Jenkins build #498

cheers -ben