May 2010 - Pharo-dev - lists.pharo.org

[Pharo-project] [update 1.1] #11370

by StÃ©phane Ducasse

11370 ----- - Issue 2030: Remove some of the Shadowed variables. Thanks alex. - Issue 2464: Fix Character >> #storeBinaryOn:. Thanks ppm - Issue 2463: timeToRunWithoutGC from squeak ideas. - Issue 2148: Fix trait composition test in Monticello. Thanks Squeak/Alex Stef

May 22, 2010

Re: [Pharo-project] Found senders of #isSystemDefined in Nile

by StÃ©phane Ducasse

Yes we should just update Nile This is a clear example of the fact that since we do not record refactoring external packages can easily degrade On May 22, 2010, at 6:26 PM, Alexandre Bergel wrote: > I have the feeling that all these uniClasses are stone age. All traces should be removed. > > Cheers, > Alexandre > > > On 22 May 2010, at 13:19, Mariano Abel Coca wrote: > >> Hi, >> >> I've found a few senders of #isSystemDefined when trying to run the pharo dev test suite, and I don't know what is the expected behavior. >> >> I've read in: http://www.mail-archive.com/pharo-project@lists.gforge.inria.fr/msg12075.ht… that the idea is to remove the message and all it's senders, but I don't know why is used here in Nile. >> >> The senders are in NSSmartRefStream, inside of #uniClassInstVarsRefs:, #uniClasesDo: and #beginInstance:size: >> >> Any hint on this will be appreciated. >> >> Cheers, >> >> Mariano. >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

May 22, 2010

Re: [Pharo-project] Status of Rome?

by StÃ©phane Ducasse

On May 22, 2010, at 6:11 PM, Igor Stasenko wrote: > On 22 May 2010 18:58, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >>> >>> Err.. what you mean by pharo? >> >> Well I was wondering if you were doing the same as juan.... >> Juan showed me better rendering written in SMalltalk code than calling openGL/cairo... >> for antialiasing so I'm eager to see morphic30 >> > > well, you can do sophisticated AA by implementing shader programs. > It will run on GPU, means veeeery fast. > And if you would ask me, what i prefer - quality or speed, i will > definitely take speed. > > Because, when you need to render thousands of glyphs on a screen > (which is a usual scenario > for text-based editors/viewers), you have to sacrifice the quality anyways. > Otherwise, wait for next 10 years till hardware will be able to render > it withot crawling like a hog. I think that this is indeed good to have something. Now if we have both then even better > > Actually, i would like to cooperate with Juan about this. I think that > AA techniques (as well as many others), > could be moved to GPU, freeing CPU from this time consuming task. > >>> I making a calls to OpenGL, using NativeBoost's OpenGL bindings. >>> All stuff is coded in smalltalk, so sure thing its does things in pharo :) >>> >>>> >>>> On May 22, 2010, at 4:54 PM, Lukas Renggli wrote: >>>> >>>>> That looks cool. I am looking forward seeing that stuff in action :-) >>>>> >>>>> Lukas >>>>> >>>>> 2010/5/22 Igor Stasenko <siguctua(a)gmail.com>: >>>>>> Good. >>>>>> >>>>>> I am now trying to implement a vector(truetype) font rendering. >>>>>> >>>>>> Just finished a little GPU fragment shader snippet, which renders a >>>>>> quad bezier curve, with antialiasing included :) >>>>>> >>>>>> You supplying 3 points, which defines a quadratic curve to GPU, and it >>>>>> does the rest. >>>>>> >>>>>> This is a basis for smooth true-type font rendering, since it contains >>>>>> a contours >>>>>> consisting from line segments or quadratic bezier segments. >>>>>> >>>>>> -- >>>>>> Best regards, >>>>>> Igor Stasenko AKA sig. >>>>>> >>>>>> _______________________________________________ >>>>>> Pharo-project mailing list >>>>>> Pharo-project(a)lists.gforge.inria.fr >>>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> Lukas Renggli >>>>> www.lukas-renggli.ch >>>>> >>>>> _______________________________________________ >>>>> Pharo-project mailing list >>>>> Pharo-project(a)lists.gforge.inria.fr >>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>> >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> Pharo-project(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>> >>> >>> >>> >>> -- >>> Best regards, >>> Igor Stasenko AKA sig. >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > > > -- > Best regards, > Igor Stasenko AKA sig. > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

May 22, 2010

[Pharo-project] Do TTFonts prevent image startup?

by StÃ©phane Ducasse

http://code.google.com/p/pharo/issues/detail?id=2457 Issue 2457: Do TTFonts prevent image startup We should check if this is a problem for Pharo too. See the thread "Please, help me. My squeak image not started" on squeak-dev: http://lists.squeakfoundation.org/pipermail/squeak-dev/2010-May/150846.html

May 22, 2010

Re: [Pharo-project] Found senders of #isSystemDefined in Nile

by Alexandre Bergel

I have the feeling that all these uniClasses are stone age. All traces should be removed. Cheers, Alexandre On 22 May 2010, at 13:19, Mariano Abel Coca wrote: > Hi, > > I've found a few senders of #isSystemDefined when trying to run the pharo dev test suite, and I don't know what is the expected behavior. > > I've read in: http://www.mail-archive.com/pharo-project@lists.gforge.inria.fr/msg12075.ht… that the idea is to remove the message and all it's senders, but I don't know why is used here in Nile. > > The senders are in NSSmartRefStream, inside of #uniClassInstVarsRefs:, #uniClasesDo: and #beginInstance:size: > > Any hint on this will be appreciated. > > Cheers, > > Mariano. > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

May 22, 2010

[Pharo-project] Found senders of #isSystemDefined in Nile

by Mariano Abel Coca

Hi, I've found a few senders of #isSystemDefined when trying to run the pharo dev test suite, and I don't know what is the expected behavior. I've read in: http://www.mail-archive.com/pharo-project@lists.gforge.inria.fr/msg12075.ht… the idea is to remove the message and all it's senders, but I don't know why is used here in Nile. The senders are in NSSmartRefStream, inside of #uniClassInstVarsRefs:, #uniClasesDo: and #beginInstance:size: Any hint on this will be appreciated. Cheers, Mariano.

May 22, 2010

Re: [Pharo-project] Status of Rome?

by Igor Stasenko

On 22 May 2010 19:03, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: > Using the TTFontReader? > Iirc, that parses the contour data, even an example of rendering it with > ballon drawBezier.... somewhere :) > Yes, i will use this data directly, in same way as balloon does, and it will be rasterized by GPU. > Cheers, > Henry > > Den 22. mai 2010 kl. 17.42 skrev Igor Stasenko <siguctua(a)gmail.com>: > >> On 22 May 2010 18:34, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >>> >>> me too. >>> Now igor do you plan to do it in pharo or by calling opengl? >>> >> >> Err.. what you mean by pharo? >> I making a calls to OpenGL, using NativeBoost's OpenGL bindings. >> All stuff is coded in smalltalk, so sure thing its does things in pharo :) >> >>> >>> On May 22, 2010, at 4:54 PM, Lukas Renggli wrote: >>> >>>> That looks cool. I am looking forward seeing that stuff in action :-) >>>> >>>> Lukas >>>> >>>> 2010/5/22 Igor Stasenko <siguctua(a)gmail.com>: >>>>> >>>>> Good. >>>>> >>>>> I am now trying to implement a vector(truetype) font rendering. >>>>> >>>>> Just finished a little GPU fragment shader snippet, which renders a >>>>> quad bezier curve, with antialiasing included :) >>>>> >>>>> You supplying 3 points, which defines a quadratic curve to GPU, and it >>>>> does the rest. >>>>> >>>>> This is a basis for smooth true-type font rendering, since it contains >>>>> a contours >>>>> consisting from line segments or quadratic bezier segments. >>>>> >>>>> -- >>>>> Best regards, >>>>> Igor Stasenko AKA sig. >>>>> >>>>> _______________________________________________ >>>>> Pharo-project mailing list >>>>> Pharo-project(a)lists.gforge.inria.fr >>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>>> >>>> >>>> >>>> >>>> -- >>>> Lukas Renggli >>>> www.lukas-renggli.ch >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> Pharo-project(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >> >> >> >> -- >> Best regards, >> Igor Stasenko AKA sig. >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project -- Best regards, Igor Stasenko AKA sig.

May 22, 2010

Re: [Pharo-project] Status of Rome?

by Igor Stasenko

On 22 May 2010 18:58, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> >> Err.. what you mean by pharo? > > Well I was wondering if you were doing the same as juan.... > Juan showed me better rendering written in SMalltalk code than calling openGL/cairo... > for antialiasing so I'm eager to see morphic30 > well, you can do sophisticated AA by implementing shader programs. It will run on GPU, means veeeery fast. And if you would ask me, what i prefer - quality or speed, i will definitely take speed. Because, when you need to render thousands of glyphs on a screen (which is a usual scenario for text-based editors/viewers), you have to sacrifice the quality anyways. Otherwise, wait for next 10 years till hardware will be able to render it withot crawling like a hog. Actually, i would like to cooperate with Juan about this. I think that AA techniques (as well as many others), could be moved to GPU, freeing CPU from this time consuming task. >> I making a calls to OpenGL, using NativeBoost's OpenGL bindings. >> All stuff is coded in smalltalk, so sure thing its does things in pharo :) >> >>> >>> On May 22, 2010, at 4:54 PM, Lukas Renggli wrote: >>> >>>> That looks cool. I am looking forward seeing that stuff in action :-) >>>> >>>> Lukas >>>> >>>> 2010/5/22 Igor Stasenko <siguctua(a)gmail.com>: >>>>> Good. >>>>> >>>>> I am now trying to implement a vector(truetype) font rendering. >>>>> >>>>> Just finished a little GPU fragment shader snippet, which renders a >>>>> quad bezier curve, with antialiasing included :) >>>>> >>>>> You supplying 3 points, which defines a quadratic curve to GPU, and it >>>>> does the rest. >>>>> >>>>> This is a basis for smooth true-type font rendering, since it contains >>>>> a contours >>>>> consisting from line segments or quadratic bezier segments. >>>>> >>>>> -- >>>>> Best regards, >>>>> Igor Stasenko AKA sig. >>>>> >>>>> _______________________________________________ >>>>> Pharo-project mailing list >>>>> Pharo-project(a)lists.gforge.inria.fr >>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>>> >>>> >>>> >>>> >>>> -- >>>> Lukas Renggli >>>> www.lukas-renggli.ch >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> Pharo-project(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >> >> >> >> -- >> Best regards, >> Igor Stasenko AKA sig. >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- Best regards, Igor Stasenko AKA sig.

May 22, 2010

Re: [Pharo-project] Status of Rome?

by Henrik Johansen

Using the TTFontReader? Iirc, that parses the contour data, even an example of rendering it with ballon drawBezier.... somewhere :) Cheers, Henry Den 22. mai 2010 kl. 17.42 skrev Igor Stasenko <siguctua(a)gmail.com>: > On 22 May 2010 18:34, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> w > rote: >> me too. >> Now igor do you plan to do it in pharo or by calling opengl? >> > > Err.. what you mean by pharo? > I making a calls to OpenGL, using NativeBoost's OpenGL bindings. > All stuff is coded in smalltalk, so sure thing its does things in > pharo :) > >> >> On May 22, 2010, at 4:54 PM, Lukas Renggli wrote: >> >>> That looks cool. I am looking forward seeing that stuff in >>> action :-) >>> >>> Lukas >>> >>> 2010/5/22 Igor Stasenko <siguctua(a)gmail.com>: >>>> Good. >>>> >>>> I am now trying to implement a vector(truetype) font rendering. >>>> >>>> Just finished a little GPU fragment shader snippet, which renders a >>>> quad bezier curve, with antialiasing included :) >>>> >>>> You supplying 3 points, which defines a quadratic curve to GPU, >>>> and it >>>> does the rest. >>>> >>>> This is a basis for smooth true-type font rendering, since it >>>> contains >>>> a contours >>>> consisting from line segments or quadratic bezier segments. >>>> >>>> -- >>>> Best regards, >>>> Igor Stasenko AKA sig. >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> Pharo-project(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>> >>> >>> >>> >>> -- >>> Lukas Renggli >>> www.lukas-renggli.ch >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > > > -- > Best regards, > Igor Stasenko AKA sig. > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

May 22, 2010

Re: [Pharo-project] Status of Rome?

by StÃ©phane Ducasse

> > Err.. what you mean by pharo? Well I was wondering if you were doing the same as juan.... Juan showed me better rendering written in SMalltalk code than calling openGL/cairo... for antialiasing so I'm eager to see morphic30 > I making a calls to OpenGL, using NativeBoost's OpenGL bindings. > All stuff is coded in smalltalk, so sure thing its does things in pharo :) > >> >> On May 22, 2010, at 4:54 PM, Lukas Renggli wrote: >> >>> That looks cool. I am looking forward seeing that stuff in action :-) >>> >>> Lukas >>> >>> 2010/5/22 Igor Stasenko <siguctua(a)gmail.com>: >>>> Good. >>>> >>>> I am now trying to implement a vector(truetype) font rendering. >>>> >>>> Just finished a little GPU fragment shader snippet, which renders a >>>> quad bezier curve, with antialiasing included :) >>>> >>>> You supplying 3 points, which defines a quadratic curve to GPU, and it >>>> does the rest. >>>> >>>> This is a basis for smooth true-type font rendering, since it contains >>>> a contours >>>> consisting from line segments or quadratic bezier segments. >>>> >>>> -- >>>> Best regards, >>>> Igor Stasenko AKA sig. >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> Pharo-project(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>> >>> >>> >>> >>> -- >>> Lukas Renggli >>> www.lukas-renggli.ch >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > > > -- > Best regards, > Igor Stasenko AKA sig. > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

May 22, 2010