March 2014 - Pharo-dev - lists.pharo.org

EyeInspector Updating

by Sven Van Caekenberghe

Please have a look at https://pharo.fogbugz.com/f/cases/13156/EyeInspector-does-not-automatically… this is a proposal for a solution, but I want some comments by those who understand this better than me. More specifically, I do not understand the description getSelection size = 0 ifTrue: guard in EyeInspector>>#refresh and the guard that I added in EyeInspector>>#updateList works, in the sense that it refuses identical updates, but when the list changes, the old contents is already empty, which I don't understand. Sven PS: There is also another issue https://pharo.fogbugz.com/f/cases/13157/EyeTreeInspector-has-no-auto-refres… but that one is open

March 30, 2014

Hotfix (was Re: DateAndTime>>#= slow?)

by Ben Coman

March 30, 2014

Re: [Pharo-dev] VistaCursors are killing my LinuxVM (beCursor primitive failing)

by Igor Stasenko

ok, thanks. https://pharo.fogbugz.com/f/cases/13155/Enable-XRender-support-for-Linux-VMs On 29 March 2014 20:45, Nicolai Hess <nicolaihess(a)web.de> wrote: > 2014-03-29 18:08 GMT+01:00 Igor Stasenko <siguctua(a)gmail.com>: > > >> >> >> On 28 March 2014 22:18, Nicolai Hess <nicolaihess(a)web.de> wrote: >> >>> Yes we have this issue: >>> >>> 12960 <https://pharo.fogbugz.com/default.asp?12960> >>> Primitive 101 (Cursor beCursor) fails on Ubuntu >>> >>> we just need someone who can edit the pharo vm build config. >>> >>> >>> "We just need to enable xrender support in the vm build config" >> i haven't seen anything related to 'xrender support' in configs, despite >> i was one >> of those who made those configs originally.. :) >> can you elaborate, what exactly you talking about? >> >> >> >>> >>> >>> >>> 2014-03-28 18:52 GMT+01:00 phil(a)highoctane.be <phil(a)highoctane.be>: >>> >>> I moved a windows image with VistaCursors to Linux and the beCursor >>>> primitive failed, leading to a white screen opening. >>>> >>>> Maybe this is because the cursors are bigger than the 16x16 form. >>>> >>>> I saw discussions of this being fixed in Squeak. Are we having this >>>> issue in Pharo? >>>> >>>> Phil >>>> >>> >>> >> >> >> -- >> Best regards, >> Igor Stasenko. >> > > Thats the problem, no -DHAVE_LIBXRENDER and no -lXrender > (don't know exaclty how this gets enabled, some cmake rules?) > > Maybe I choosed the wrong words: > Change the config, so that xrender support gets enabled. > > squnixX11.c: > > static sqInt display_ioSetCursorARGB(sqInt cursorBitsIndex, sqInt extentX, > sqInt extentY, sqInt offsetX, sqInt offsetY) > { > #if defined(HAVE_LIBXRENDER) && (RENDER_MAJOR > 0 || RENDER_MINOR >= 5) > ... > return 1; > #else > return 0; > #endif > } > > > > Nicolai > -- Best regards, Igor Stasenko.

March 29, 2014

Re: [Pharo-dev] Roassal 3D empty windows

by Nicolai Hess

For me, it works now: I just have to change the default extent: GLViewportMorph>>initializeForNewSession display := NBGLDisplay extent: 800@800. "<- instead of self extent" session := NativeBoost uniqueSessionObject. form ifNotNil: [ form unhibernate ]. otherwise the display (and frame buffers ) are initialized with Morphs default extent (40@50), and after that, every change on the extent does not update the framebuffer dimension (although NBOpenGls FrameBuffer and RenderBuffer do recreate there buffers with the new extent, somehow the opengl buffers doesn't change the sizes). R3 IDE windows are still empty and I don't know what they are supposed to show. Are there any screenshots for the R3 IDE? Nicolai 2014-03-17 23:54 GMT+01:00 phil(a)highoctane.be <phil(a)highoctane.be>: > I loaded Roassal3D on my machine (Windows 8.1 Pro). > > Loaded fine. > > Then tried the tutorial basic cube. > > view := R3View new. > view add: R3CubeShape element. > view addInteraction: R3MKControl. > view open > > Shows a cube but all black. > Interaction works (on AZERTY keyboards, WSAD kinds of, well, you know...). > See screenshot. > > Graphics drivers up to date. > Card: NVidia Geforce GTX 770 ( > http://www.asus.com/Graphics_Cards/GTX770DC2OC2GD5/) > > Phil > > > > On Mon, Mar 17, 2014 at 11:35 PM, Nicolai Hess <nicolaihess(a)web.de> wrote: > >> As described in the "NBOpenGL on Windows" thread, I have a windows >> machine where NBOpenGL can not create frame buffer objects (driver issue >> maybe). >> >> But on another windows machine, there is no error but strange rendering >> results. >> >> On that machine, Roassals 3D IDE, still only shows empty windows. And >> other >> demos shows the same strange rendering (see screenshot, includes GL >> context driver info). >> >> >> (BTW, there is some class name missmatch in NBGLCurveRenderer in this >> NBOpenGL version. >> It referse to NBGLFragmentProgram although there is only a >> NBGLFragmentProgramARB class >> in the system). >> >> >> >> >> 2014-03-17 15:55 GMT+01:00 Ronie Salgado <roniesalg(a)gmail.com>: >> >>> Hi Nicolai, >>> >>> What are your system specs?, specially your graphic card. >>> Roassal3D requires a graphic card with at least OpenGL 2.0 support. Try >>> updating your graphic drivers >>> >>> Greetings, >>> Ronie >>> >>> >>> 2014-03-17 11:03 GMT-03:00 Alexandre Bergel <alexandre.bergel(a)me.com>: >>> >>> Oh... Have you tried to update Athens and NBOpenGL ? Maybe you are using >>>> an old version of them. >>>> Maybe an old version of OpenGL is installed on your Window? >>>> >>>> Ronie, any idea what's going on? >>>> >>>> Alexandre >>>> >>>> >>>> On Mar 17, 2014, at 10:55 AM, Nicolai Hess <nicolaihess(a)web.de> wrote: >>>> >>>> > Thank you Alexandre, >>>> > this was what I tried under windows, (I wasn't sure if this is still >>>> the right way to do it, I >>>> > read it the Roassal3dTutorial.pdf). >>>> > But it doesn't work. Either I am getting empty windows or (with an >>>> other >>>> > NBOpenGL version) >>>> > Frame buffer is incomplete: GL_FRAMEBUFFER_UNSUPPORTED') >>>> > >>>> > I tried this once with linux too, and it didn't worked. But I am not >>>> sure how long ago :) >>>> > I'll try it again and report back. >>>> > >>>> > (Another note, after downloading the configuration, this: >>>> > (Smalltalk at: #ConfigurationOfRoassal3d) loadDevelopment >>>> > takes awfully long on windows (~ 5 min). >>>> > I am sure it was much faster on linux, any idea how that can be?) >>>> > >>>> > >>>> > >>>> > >>>> > >>>> > >>>> > 2014-03-17 14:34 GMT+01:00 Alexandre Bergel <alexandre.bergel(a)me.com >>>> >: >>>> > Hi Nicolai, >>>> > >>>> > Roassal3d works well with the last Pharo 3.0. >>>> > You just need to execute the following incantation: >>>> > >>>> > -=-=-=-=-=-=-=-=-= >>>> > Gofer new smalltalkhubUser: 'ronsaldo' project: 'roassal3d'; package: >>>> 'ConfigurationOfRoassal3d'; load. (Smalltalk at: #ConfigurationOfRoassal3d) >>>> loadDevelopment >>>> > -=-=-=-=-=-=-=-=-= >>>> > >>>> > This does not work for you? >>>> > It works well on OSX and on Linux >>>> > >>>> > Alexandre >>>> > >>>> > >>>> > On Mar 15, 2014, at 10:52 AM, Nicolai Hess <nicolaihess(a)web.de> >>>> wrote: >>>> > >>>> > > Hi, >>>> > > >>>> > > what could be the reason, that any Roassal3d example from the >>>> > > Roassal3DTutorial.pdf only shows white empty windows? >>>> > > >>>> > > I loaded ConfigurationOfRoassal3d-RonieSalgado.10 and >>>> > > (Smalltalk at: >>>> > > #ConfigurationOfRoassal3d) loadDevelopment >>>> > > >>>> > > Any ideas? >>>> > > >>>> > > >>>> > > regards >>>> > > Nicolai >>>> > > >>>> > >>>> > -- >>>> > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>>> > Alexandre Bergel http://www.bergel.eu >>>> > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>>> > >>>> > >>>> > >>>> > >>>> >>>> -- >>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>>> Alexandre Bergel http://www.bergel.eu >>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>>> >>>> >>>> >>>> >>> >> >

March 29, 2014

Re: [Pharo-dev] Font problem identified

by Pharo4Stef

Thanks for you suggestion I will check how to implement a registration mechanism. I was wondering if I missed the obvious. Stef > > > i'd like to point out, that there is of course easy brute-force solution to this problem > (simply put dirty patch to force re-loading fonts).. > > however, i think we should wisely consider, how to deal with it in graceful manner by providing a font-registration mechanism for any potential 3-rd party package(s). > Like that, we will solve the problem for longer time period than just the single next day. > > -- > Best regards, > Igor Stasenko.

March 29, 2014

Re: [Pharo-dev] VistaCursors are killing my LinuxVM (beCursor primitive failing)

by Nicolai Hess

2014-03-29 18:08 GMT+01:00 Igor Stasenko <siguctua(a)gmail.com>: > > > > On 28 March 2014 22:18, Nicolai Hess <nicolaihess(a)web.de> wrote: > >> Yes we have this issue: >> >> 12960 <https://pharo.fogbugz.com/default.asp?12960> >> Primitive 101 (Cursor beCursor) fails on Ubuntu >> >> we just need someone who can edit the pharo vm build config. >> >> >> "We just need to enable xrender support in the vm build config" > i haven't seen anything related to 'xrender support' in configs, despite i > was one > of those who made those configs originally.. :) > can you elaborate, what exactly you talking about? > > > >> >> >> >> 2014-03-28 18:52 GMT+01:00 phil(a)highoctane.be <phil(a)highoctane.be>: >> >> I moved a windows image with VistaCursors to Linux and the beCursor >>> primitive failed, leading to a white screen opening. >>> >>> Maybe this is because the cursors are bigger than the 16x16 form. >>> >>> I saw discussions of this being fixed in Squeak. Are we having this >>> issue in Pharo? >>> >>> Phil >>> >> >> > > > -- > Best regards, > Igor Stasenko. > Thats the problem, no -DHAVE_LIBXRENDER and no -lXrender (don't know exaclty how this gets enabled, some cmake rules?) Maybe I choosed the wrong words: Change the config, so that xrender support gets enabled. squnixX11.c: static sqInt display_ioSetCursorARGB(sqInt cursorBitsIndex, sqInt extentX, sqInt extentY, sqInt offsetX, sqInt offsetY) { #if defined(HAVE_LIBXRENDER) && (RENDER_MAJOR > 0 || RENDER_MINOR >= 5) ... return 1; #else return 0; #endif } Nicolai

March 29, 2014

Re: [Pharo-dev] [Inspiration] Toward a better programming

by Igor Stasenko

Yes. It is harmful to throw away semantic in favor of high-level abstractions. As well as opposite.. You need to use (and master) both. IMO smalltalk environment, as such, is capable to demonstrate that is is possible to achieve this golden middle, where abstraction(s) meet low-level semantics, without competing each other. Hence, of course, my contribution to it - NativeBoost, because this is my vision. On 29 March 2014 19:02, phil(a)highoctane.be <phil(a)highoctane.be> wrote: > I was using Pharo today for a maths expo that my wife organized for her > school day. > > Feedback from the whiz kids: > > "That nice if one wants to control its environment. > > But I am learning C first because I want to understand how the machine > works and I got lost by using all kinds of languages and mastering none. > (He uses Python, C, C# and had a run on Pharo on his own)." > > So, Pharo + PharoVM + NB + Slang could be a good combo. > > Phil > Le 29 mars 2014 18:41, "Igor Stasenko" <siguctua(a)gmail.com> a Ã©crit : > > >> >> >> On 29 March 2014 18:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: >> >>> Hi, >>> >>> Sure he is repeating things. And sure he is reinventing half of what we >>> already have. >>> >>> However, he is likely to get more traction because he reinvents the UI, >>> not the core concepts. You could say it is not an essential contribution. >>> But, it is. >>> >>> >> The general misconception is that visual manipulation with objects can >> fully replace the coding. It won't and never will. Except from very simple >> cases (like computing the average :) >> You can try to visually express things like branching, conditional loops, >> common programming patterns, but i doubt you will find it convenient to >> program it in "visual" way rather than coding. >> >> >>> Doru >>> >>> >>> On Sat, Mar 29, 2014 at 6:28 PM, Igor Stasenko <siguctua(a)gmail.com>wrote: >>> >>>> Sure, guy just repeating almost same thesis as in the original Self >>>> demo dated back in '95, >>>> "direct manipulation with objects errr.. data".. >>>> except that now in crappy web-based environment.. and still missing the >>>> point.. >>>> because it is not about data, it is about objects. >>>> >>>> >>>> >>>> >>>> On 29 March 2014 18:20, Tudor Girba <tudor(a)tudorgirba.com> wrote: >>>> >>>>> Beautiful demo. This should be our game, yet others are playing it :(. >>>>> >>>>> Doru >>>>> >>>>> >>>>> On Sat, Mar 29, 2014 at 11:31 AM, Sven Van Caekenberghe <sven(a)stfx.eu>wrote: >>>>> >>>>>> >>>>>> On 29 Mar 2014, at 10:38, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: >>>>>> >>>>>> > This is a nice write down: >>>>>> > >>>>>> > >>>>>> http://www.chris-granger.com/2014/03/27/toward-a-better-programming/ >>>>>> > >>>>>> > with a nice demo of a prototype: >>>>>> > >>>>>> > https://www.youtube.com/watch?v=L6iUm_Cqx2s >>>>>> > >>>>>> > Luckily, the horrible C++ code computing standard deviation in the >>>>>> article can be written quite elegantly and directly in Pharo: >>>>>> > >>>>>> > | input | >>>>>> > input := #(2 4 4 4 5 5 7 9). >>>>>> > (((input - input average) raisedTo: 2) sum / (input size - 1)) sort. >>>>>> > >>>>>> > Sven >>>>>> >>>>>> Damn spelling correction ;-) >>>>>> >>>>>> | input | >>>>>> input := #(2 4 4 4 5 5 7 9). >>>>>> (((input - input average) raisedTo: 2) sum / (input size - 1)) sqrt. >>>>>> >>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> www.tudorgirba.com >>>>> >>>>> "Every thing has its own flow" >>>>> >>>> >>>> >>>> >>>> -- >>>> Best regards, >>>> Igor Stasenko. >>>> >>> >>> >>> >>> -- >>> www.tudorgirba.com >>> >>> "Every thing has its own flow" >>> >> >> >> >> -- >> Best regards, >> Igor Stasenko. >> > -- Best regards, Igor Stasenko.

March 29, 2014

Re: [Pharo-dev] [Inspiration] Toward a better programming

by phil＠highoctane.be

I was using Pharo today for a maths expo that my wife organized for her school day. Feedback from the whiz kids: "That nice if one wants to control its environment. But I am learning C first because I want to understand how the machine works and I got lost by using all kinds of languages and mastering none. (He uses Python, C, C# and had a run on Pharo on his own)." So, Pharo + PharoVM + NB + Slang could be a good combo. Phil Le 29 mars 2014 18:41, "Igor Stasenko" <siguctua(a)gmail.com> a Ã©crit : > > > > On 29 March 2014 18:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: > >> Hi, >> >> Sure he is repeating things. And sure he is reinventing half of what we >> already have. >> >> However, he is likely to get more traction because he reinvents the UI, >> not the core concepts. You could say it is not an essential contribution. >> But, it is. >> >> > The general misconception is that visual manipulation with objects can > fully replace the coding. It won't and never will. Except from very simple > cases (like computing the average :) > You can try to visually express things like branching, conditional loops, > common programming patterns, but i doubt you will find it convenient to > program it in "visual" way rather than coding. > > >> Doru >> >> >> On Sat, Mar 29, 2014 at 6:28 PM, Igor Stasenko <siguctua(a)gmail.com>wrote: >> >>> Sure, guy just repeating almost same thesis as in the original Self demo >>> dated back in '95, >>> "direct manipulation with objects errr.. data".. >>> except that now in crappy web-based environment.. and still missing the >>> point.. >>> because it is not about data, it is about objects. >>> >>> >>> >>> >>> On 29 March 2014 18:20, Tudor Girba <tudor(a)tudorgirba.com> wrote: >>> >>>> Beautiful demo. This should be our game, yet others are playing it :(. >>>> >>>> Doru >>>> >>>> >>>> On Sat, Mar 29, 2014 at 11:31 AM, Sven Van Caekenberghe <sven(a)stfx.eu>wrote: >>>> >>>>> >>>>> On 29 Mar 2014, at 10:38, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: >>>>> >>>>> > This is a nice write down: >>>>> > >>>>> > >>>>> http://www.chris-granger.com/2014/03/27/toward-a-better-programming/ >>>>> > >>>>> > with a nice demo of a prototype: >>>>> > >>>>> > https://www.youtube.com/watch?v=L6iUm_Cqx2s >>>>> > >>>>> > Luckily, the horrible C++ code computing standard deviation in the >>>>> article can be written quite elegantly and directly in Pharo: >>>>> > >>>>> > | input | >>>>> > input := #(2 4 4 4 5 5 7 9). >>>>> > (((input - input average) raisedTo: 2) sum / (input size - 1)) sort. >>>>> > >>>>> > Sven >>>>> >>>>> Damn spelling correction ;-) >>>>> >>>>> | input | >>>>> input := #(2 4 4 4 5 5 7 9). >>>>> (((input - input average) raisedTo: 2) sum / (input size - 1)) sqrt. >>>>> >>>>> >>>>> >>>> >>>> >>>> -- >>>> www.tudorgirba.com >>>> >>>> "Every thing has its own flow" >>>> >>> >>> >>> >>> -- >>> Best regards, >>> Igor Stasenko. >>> >> >> >> >> -- >> www.tudorgirba.com >> >> "Every thing has its own flow" >> > > > > -- > Best regards, > Igor Stasenko. >

March 29, 2014

Re: [Pharo-dev] [Inspiration] Toward a better programming

by Igor Stasenko

On 29 March 2014 18:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: > Hi, > > Sure he is repeating things. And sure he is reinventing half of what we > already have. > > However, he is likely to get more traction because he reinvents the UI, > not the core concepts. You could say it is not an essential contribution. > But, it is. > > The general misconception is that visual manipulation with objects can fully replace the coding. It won't and never will. Except from very simple cases (like computing the average :) You can try to visually express things like branching, conditional loops, common programming patterns, but i doubt you will find it convenient to program it in "visual" way rather than coding. > Doru > > > On Sat, Mar 29, 2014 at 6:28 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: > >> Sure, guy just repeating almost same thesis as in the original Self demo >> dated back in '95, >> "direct manipulation with objects errr.. data".. >> except that now in crappy web-based environment.. and still missing the >> point.. >> because it is not about data, it is about objects. >> >> >> >> >> On 29 March 2014 18:20, Tudor Girba <tudor(a)tudorgirba.com> wrote: >> >>> Beautiful demo. This should be our game, yet others are playing it :(. >>> >>> Doru >>> >>> >>> On Sat, Mar 29, 2014 at 11:31 AM, Sven Van Caekenberghe <sven(a)stfx.eu>wrote: >>> >>>> >>>> On 29 Mar 2014, at 10:38, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: >>>> >>>> > This is a nice write down: >>>> > >>>> > http://www.chris-granger.com/2014/03/27/toward-a-better-programming/ >>>> > >>>> > with a nice demo of a prototype: >>>> > >>>> > https://www.youtube.com/watch?v=L6iUm_Cqx2s >>>> > >>>> > Luckily, the horrible C++ code computing standard deviation in the >>>> article can be written quite elegantly and directly in Pharo: >>>> > >>>> > | input | >>>> > input := #(2 4 4 4 5 5 7 9). >>>> > (((input - input average) raisedTo: 2) sum / (input size - 1)) sort. >>>> > >>>> > Sven >>>> >>>> Damn spelling correction ;-) >>>> >>>> | input | >>>> input := #(2 4 4 4 5 5 7 9). >>>> (((input - input average) raisedTo: 2) sum / (input size - 1)) sqrt. >>>> >>>> >>>> >>> >>> >>> -- >>> www.tudorgirba.com >>> >>> "Every thing has its own flow" >>> >> >> >> >> -- >> Best regards, >> Igor Stasenko. >> > > > > -- > www.tudorgirba.com > > "Every thing has its own flow" > -- Best regards, Igor Stasenko.

March 29, 2014

Re: [Pharo-dev] [Inspiration] Toward a better programming

by Tudor Girba

Hi, Sure he is repeating things. And sure he is reinventing half of what we already have. However, he is likely to get more traction because he reinvents the UI, not the core concepts. You could say it is not an essential contribution. But, it is. Doru On Sat, Mar 29, 2014 at 6:28 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: > Sure, guy just repeating almost same thesis as in the original Self demo > dated back in '95, > "direct manipulation with objects errr.. data".. > except that now in crappy web-based environment.. and still missing the > point.. > because it is not about data, it is about objects. > > > > > On 29 March 2014 18:20, Tudor Girba <tudor(a)tudorgirba.com> wrote: > >> Beautiful demo. This should be our game, yet others are playing it :(. >> >> Doru >> >> >> On Sat, Mar 29, 2014 at 11:31 AM, Sven Van Caekenberghe <sven(a)stfx.eu>wrote: >> >>> >>> On 29 Mar 2014, at 10:38, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: >>> >>> > This is a nice write down: >>> > >>> > http://www.chris-granger.com/2014/03/27/toward-a-better-programming/ >>> > >>> > with a nice demo of a prototype: >>> > >>> > https://www.youtube.com/watch?v=L6iUm_Cqx2s >>> > >>> > Luckily, the horrible C++ code computing standard deviation in the >>> article can be written quite elegantly and directly in Pharo: >>> > >>> > | input | >>> > input := #(2 4 4 4 5 5 7 9). >>> > (((input - input average) raisedTo: 2) sum / (input size - 1)) sort. >>> > >>> > Sven >>> >>> Damn spelling correction ;-) >>> >>> | input | >>> input := #(2 4 4 4 5 5 7 9). >>> (((input - input average) raisedTo: 2) sum / (input size - 1)) sqrt. >>> >>> >>> >> >> >> -- >> www.tudorgirba.com >> >> "Every thing has its own flow" >> > > > > -- > Best regards, > Igor Stasenko. > -- www.tudorgirba.com "Every thing has its own flow"

March 29, 2014