[update] 10420

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

[update] 10420

Adrian Lienhard
- Issue 838: Make mouse buttons mapping platform dependent
- Issue 405: Demo mode is unusable
- Issue 1053: Transcript updates from foreground process not displayed  
in Transcript window
- Issue 1063: Class var order in Monticello (bogus dirty packages)
- Issue 1019: 0006579: String>>#findLastOccuranceOfString:startingAt:  
isBuggedWhen: [substring first=substring last (now with deprecation of  
old method)
- Issue 1087: tallyIt nows trigger a complete profiling (also renamed  
tally it to profile it)
- Browser label: do not print 'System Browser: <Class>' but only  
'<Class>'
- Disable window animation for closing

35 issues left for milestone 1.0.

BTW: Yesterday, Pharo was on Y Combinator Hacker News: "Pharo: a  
Smalltalk VM that doesn't look like a toy out of the box" (http://news.ycombinator.com/item?id=787363 
)

___________________
http://www.adrian-lienhard.ch/


_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Reply | Threaded
Open this post in threaded view
|

Re: [update] 10420

Stéphane Ducasse
Thanks adrian.

After working on papers and preparing ESUG I hope to get some times to  
integrate.
        - I would like to integrate CUIS strike fonts (now this is ready) in  
inbox
        - lot of tiny fixes

I'm hesitating about the compiler notification but this is so cool not  
to have pop up
but exception. What other think?
I run all the tests and they seems good.

Stef

On Aug 27, 2009, at 9:43 PM, Adrian Lienhard wrote:

> - Issue 838: Make mouse buttons mapping platform dependent
> - Issue 405: Demo mode is unusable
> - Issue 1053: Transcript updates from foreground process not displayed
> in Transcript window
> - Issue 1063: Class var order in Monticello (bogus dirty packages)
> - Issue 1019: 0006579: String>>#findLastOccuranceOfString:startingAt:
> isBuggedWhen: [substring first=substring last (now with deprecation of
> old method)
> - Issue 1087: tallyIt nows trigger a complete profiling (also renamed
> tally it to profile it)
> - Browser label: do not print 'System Browser: <Class>' but only
> '<Class>'
> - Disable window animation for closing
>
> 35 issues left for milestone 1.0.
>
> BTW: Yesterday, Pharo was on Y Combinator Hacker News: "Pharo: a
> Smalltalk VM that doesn't look like a toy out of the box" (http://news.ycombinator.com/item?id=787363
> )
>
> ___________________
> http://www.adrian-lienhard.ch/
>
>
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project


_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Reply | Threaded
Open this post in threaded view
|

Re: [update] 10420

csrabak
 Em 27/08/2009 18:15, Stéphane Ducasse < [hidden email] >
escreveu:

I saw a lot of reports on Mac, but in a Windows XP SP3 a dev image updated to 10419 when loaded with #10420 starts to pop up dozens of DNUs due ByteString not having between:and: which seems to come from FreeTypeFontProvider>>fontInfoFor:familyName:

Once this happens troubleshooting becomes impossible because _anything_ I attempt to do pops the same DNU!!

I Alt-.'d the image to be able to report this but the image seems not usable until I get a means to start it with some sort of script to circumvent the error.



>  Thanks adrian.
>  After working on papers and preparing ESUG I hope to get some times
> to integrate.   - I would like  to integrate CUIS  strike fonts (now
> this is ready) in inbox - lot of tiny fixes
>  I'm hesitating about the compiler  notification but this is so cool
> not to have  pop up but exception. What other think?   I run all the
> tests and they seems good.
>  Stef
>  On Aug 27, 2009, at 9:43 PM, Adrian Lienhard wrote:
>
> > - Issue 838:  Make mouse buttons mapping  platform dependent Issue
> > - 405: Demo  mode is unusable Issue 1053:  Transcript updates from
> > - foreground process not displayed
> > in Transcript window
> > - Issue 1063: Class var order in Monticello (bogus dirty packages)
> > - Issue                       1019:                       0006579:
> > - String>>#findLastOccuranceOfString:startingAt:
> > isBuggedWhen:   [substring    first=substring   last   (now   with
> > deprecation of  old method) -  Issue 1087: tallyIt nows  trigger a
> > complete profiling (also renamed tally it to profile it) - Browser
> > label:  do not  print 'System  Browser: '  but only  ''  - Disable
> > window animation for closing
> > 35 issues left for milestone 1.0.
> > BTW: Yesterday, Pharo  was on Y Combinator Hacker  News: "Pharo: a
> > Smalltalk  VM  that  doesn't look  like  a  toy  out of  the  box"
> > (http://news.ycombinator.com/item?id=787363 )
> > ___________________ http://www.adrian-lienhard.ch/

_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Reply | Threaded
Open this post in threaded view
|

Re: [update] 10420

Adrian Lienhard
Hmm, I can reproduce the problem by updating from a Pharo 09.08.3  
image and then opening the test runner. The problem is that the array  
stored in the variable 'families' of a FreeTypeFontProvider is screwed  
up. Its 'index' variable points to the string 'DejaVu Sans'.

This reminds me of http://code.google.com/p/pharo/issues/detail?id=405

OK, there was a fix for this but I overlooked that we need to run  
"FreeTypeFontProvider current initialize" as postscript.

So, the current workaround is:

FreeTypeFontProvider current initialize.

I will add that to the next update.

Adrian

On Aug 29, 2009, at 03:26 , [hidden email] wrote:

> Em 27/08/2009 18:15, Stéphane Ducasse < [hidden email] >
> escreveu:
>
> I saw a lot of reports on Mac, but in a Windows XP SP3 a dev image  
> updated to 10419 when loaded with #10420 starts to pop up dozens of  
> DNUs due ByteString not having between:and: which seems to come from  
> FreeTypeFontProvider>>fontInfoFor:familyName:
>
> Once this happens troubleshooting becomes impossible because  
> _anything_ I attempt to do pops the same DNU!!
>
> I Alt-.'d the image to be able to report this but the image seems  
> not usable until I get a means to start it with some sort of script  
> to circumvent the error.
>
>
>
>> Thanks adrian.
>> After working on papers and preparing ESUG I hope to get some times
>> to integrate.   - I would like  to integrate CUIS  strike fonts (now
>> this is ready) in inbox - lot of tiny fixes
>> I'm hesitating about the compiler  notification but this is so cool
>> not to have  pop up but exception. What other think?   I run all the
>> tests and they seems good.
>> Stef
>> On Aug 27, 2009, at 9:43 PM, Adrian Lienhard wrote:
>>
>>> - Issue 838:  Make mouse buttons mapping  platform dependent Issue
>>> - 405: Demo  mode is unusable Issue 1053:  Transcript updates from
>>> - foreground process not displayed
>>> in Transcript window
>>> - Issue 1063: Class var order in Monticello (bogus dirty packages)
>>> - Issue                       1019:                       0006579:
>>> - String>>#findLastOccuranceOfString:startingAt:
>>> isBuggedWhen:   [substring    first=substring   last   (now   with
>>> deprecation of  old method) -  Issue 1087: tallyIt nows  trigger a
>>> complete profiling (also renamed tally it to profile it) - Browser
>>> label:  do not  print 'System  Browser: '  but only  ''  - Disable
>>> window animation for closing
>>> 35 issues left for milestone 1.0.
>>> BTW: Yesterday, Pharo  was on Y Combinator Hacker  News: "Pharo: a
>>> Smalltalk  VM  that  doesn't look  like  a  toy  out of  the  box"
>>> (http://news.ycombinator.com/item?id=787363 )
>>> ___________________ http://www.adrian-lienhard.ch/
>
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project


_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Reply | Threaded
Open this post in threaded view
|

Re: [update] 10420

Damien Cassou
On Sat, Aug 29, 2009 at 1:24 PM, Adrian Lienhard<[hidden email]> wrote:

> Hmm, I can reproduce the problem by updating from a Pharo 09.08.3
> image and then opening the test runner. The problem is that the array
> stored in the variable 'families' of a FreeTypeFontProvider is screwed
> up. Its 'index' variable points to the string 'DejaVu Sans'.
>
> This reminds me of http://code.google.com/p/pharo/issues/detail?id=405
>
> OK, there was a fix for this but I overlooked that we need to run
> "FreeTypeFontProvider current initialize" as postscript.
>
> So, the current workaround is:
>
> FreeTypeFontProvider current initialize.
>
> I will add that to the next update.

We have done that in 10423 with Stéphane. Seems to work fine.

--
Damien Cassou
http://damiencassou.seasidehosting.st

"Lambdas are relegated to relative obscurity until Java makes them
popular by not having them." James Iry

_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project