May 2012 - Pharo-dev - lists.pharo.org

[Pharo-project] Classes that implement unsafe #=
by Mariano Martinez Peck May 31, 2012

May 31, 2012

Hi guys. I found a problem while working with FuelPreview where we store objects of the graph in a Set. While doing a collect, there was an error because an implementation of #= was assuming that the argument was of a certain shape. Of course, that should answer false instead. So I wrote this test to see all existing wrong implementations: testAllClassesImplementSafeEqualsMethod "This tests that all classes in the system that implements #= do it in a way that they don't throw error when passing as an argument something different from expected. The correct behavior is that #= answers false." | wrongClasses | wrongClasses := IdentityDictionary new. ((SystemNavigation default allImplementorsOf: #=) collect: [:each | each methodClass]) do: [:each | | instance | "Some classes like CompiledMethod override basicNew to throw an error. In any case, the comparison will be false, so no problem" [instance := each basicNew] on: Error do: []. [instance = Object new] on: Error do: [: err | wrongClasses at: each put: err ]]. self assert: wrongClasses isEmpty The result is: {Magnitude. WideCharacterSet. KMKeymap. MCMockDefinition. ScaledDecimal. MCSnapshot} So, my I think we should change #= in those classes by adding: self == aKeymap ifTrue: [ ^ true ]. self class = aKeymap class ifFalse: [ ^ false ]. at the beginning of the #= what do you think? if agree, I open an issue. -- Mariano http://marianopeck.wordpress.com

4 9

[Pharo-project] moving Finder up to first level of World menu
by Ben Coman May 31, 2012

May 31, 2012

Has any consideration been given to moving World > Tools > Finder up to World > Finder. I find this is core to my work flow and I it is used much more often than any of System Browser, Workspace, Test Runner or Monticello Browser - and I wonder if anyone finds the same? cheers -ben

4 6

[Pharo-project] [ANN] NativeBoost 1.2 version released (#development)
by Igor Stasenko May 31, 2012

May 31, 2012

Hi, there i prepared new Metacello configuration of NativeBoost. Here the list of changes: Integer>>#| in AsmJit now - added a recursion detecting/handling via notifications - NBExternalHandle / NBExternalAddress is no longer subclasses of NBExternalStructure - more safer way to handle finalizing external objects survived from old session - avoid recursion when printing external struct fields - default alignment of external struct is machine word size - correct padding between struct fields (need to test) - ** fix ** conceptual bug when passing a struct by value. Now copying its contents on stack, instead of passing pointer - rewrote accessing struct fields, allowing nested structs - added #size_t type and NBSizeT marshaller - use stack pointer directly by getting it's address (via stackPointerAddress() available in Cog), so accessing ST stack is faster.. - added recursion guards in NBInterpreterProxy and in NBNativeCodeGen to prevent runaway train(s) - NBExternalType >> #typeSize answers the number of bytes which value of full type (including ptr arity) variable takes place in memory. NBExternalType >> valueSize is the number of bytes for base type (excluding pointers) takes in memory. NBExternalType >> stackSize taking into account stack argument alignment ------------------------- The main change is handling correctly passing structs by value. Before i was wrongly assuming that if function takes struct: int foo ( struct x a) you need to push the pointer to struct.. but in fact you need to copy struct contents on stack! This means that all function signatures which were written based on this wrong assumption should be changed to pass a pointer (if function takes a pointer to struct). So, don't be so haste to use it with Athens with new version of NB. It will definitely break (and i know where) because of this mistake. Fortunately, most of the libraries avoid passing structs by value, so most of the bindings should still keep working fine after update. But be warned about it! Thanks Luc for pointing out on this issue. Another improvement is because of more correct handling of structs, now NB allows nested structs. I blessed this version as #development. If no serious bugs will be found, after testing it with Athens and NBOpenGL (+ add your projects) i will change blessing to #release. -- Best regards, Igor Stasenko.

4 5

[Pharo-project] omni browser
by Milan Mimica May 31, 2012

May 31, 2012

I don't get it. What do I do to load a working OmniBrowser in latest Pharo 1.4? -- Milan Mimica http://sparklet.sf.net

3 5

[Pharo-project] Blog.
by Santiago Bragagnolo May 30, 2012

May 30, 2012

Hi all again, im writting to announce the opening of a new blog of mine. http://concretetypeinference.blogspot.com.ar/ Here ill write, organize my new knowledge and broadcast news about the concrete-types inference project. I was broadcasting just via twitter, but this way is better. (From now on, on twitter i'll just paste url with the related hashtags ) That's all. Keep tunned. Santiago.

3 2

[Pharo-project] Nautilus doesn't like Vistary theme
by Sean P. DeNigris May 30, 2012

May 30, 2012

When selecting a method, sometimes the code pane turns into the dreaded "big red X". Issue 5994: Nautilus doesn't like Vistary theme http://code.google.com/p/pharo/issues/detail?id=5994 -- View this message in context: http://forum.world.st/Nautilus-doesn-t-like-Vistary-theme-tp4632292.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

7 7

[Pharo-project] [update 2.0] #20101
by Marcus Denker May 30, 2012

May 30, 2012

20101 ----- Issue 5987: Remove SARInstaller http://code.google.com/p/pharo/issues/detail?id=5987 Issue 5979: Method names containing shortcut broken http://code.google.com/p/pharo/issues/detail?id=5979 Issue 5696: Failing tests: Traits http://code.google.com/p/pharo/issues/detail?id=5696 -- Marcus Denker -- http://marcusdenker.de

2 1

Re: [Pharo-project] [squeak-dev] GLMorphic
by Igor Stasenko May 30, 2012

May 30, 2012

On 30 May 2012 20:24, Colin Putney <colin(a)wiresong.com> wrote: > On Wed, May 30, 2012 at 11:01 AM, Tim Felgentreff <tim(a)nada1.de> wrote: > >> There's plenty of problems, and performance isn't great, but do people >> think better OpenGL integration is something that's still worth >> pursuing? (not necessarily based on GLMorphic) > > Yes! My dream GUI would be a Smalltalk widget library that draws into > Areithfa Ffenestri windows via OpenGL (or maybe Cairo). Native windows > would play nicely with the OS window manager, and Smalltalk widgets > would give us a consistent experience across platforms. Visualworks is > going in that direction with Cairo, and Igor's nascent Athens project > looks interesting, but we haven't really seen such a beast yet. > Yes, one of the goals of Athens is to incorporate it as default graphic framework for Morphic (and for rest of stuff of course). At some point, i will certainly start implementing OpenGL backend for Athens and will certainly use some of the work from GLMorphic. > Colin -- Best regards, Igor Stasenko.

1 0

[Pharo-project] Glamorous Inspector
by Camillo Bruni May 30, 2012

May 30, 2012

I'm trying to load the Glamorous Inspector in 2.0. Besides some difficulties with non-present versions in some Configs (RB and Co) I got the thing more or less running. However it doesn't find BorderedSubpaneDividerMorph, any chance where I might find this class? best cami

2 1

May 29, 2012

I'm trying to focus on getting all the tests green. So I will focus on that for a while. Any help is welcome :) Stef

2 3