- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Compiler Explorer (for the Compiler guys) #offtopic

by ClÃ©ment Bera

On Wed, Oct 18, 2017 at 3:27 PM, Jan Vrany <jan.vrany(a)fit.cvut.cz> wrote: > Hi, > > > I was thinking of changing that, adding a primitive which answers a > > CogMethod as a bytearray, but you still need to use external libs > > such as gdb/lldb to disassemble your binary > > I use udis86 [1] for that purpose. Tiny, clean & dead easy to use. > x86 only, though. > > HTH. Jan > > [1]: http://udis86.sourceforge.net/ Tiny that's good. I normally use the facilities provided by the processor simulators (but that's not tiny). Not sure the production environment should have those features anyway > > > > (which is possible but not working out of the box everywhere and it > > does not make sense to include that as part of the VM), and you still > > need something extracted from VMMaker to make sense out of the > > CogMethod header and CogMethod map, which has a lot of dependencies, > > so likely the whole VMMaker itself. And I'm not talking about all the > > different CogMethod flavors. > > > > With VMMaker loaded you can do it with in-image compilation but you > > will have the machine code with only unlinked sends. With linked send > > it's more difficult as you need to really simulate the whole runtime. > > > > Overall there's no amazing solution. In the example shown it's easier > > as in C++ the code is not modified while it runs and you've already > > have gdb/lldb APIs to show what they show. > > > > > > > > > > > > > > > > > > > > > > > > > Cheers, > > > > Doru > > > > > > > > > > > > > On Oct 17, 2017, at 10:03 PM, Esteban A. Maringolo <emaringolo@ > > > > > gmail.com> wrote: > > > > > > > > > > I'm sure we could do this with ease, because we already can see > > > > > the bytecodes. > > > > > > > > > > https://godbolt.org/ > > > > > > > > > > Regards! > > > > > > > > > > Esteban A. Maringolo > > > > > > > > > > > > > -- > > > > www.tudorgirba.com > > > > www.feenk.com > > > > > > > > "The coherence of a trip is given by the clearness of the goal." > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > Mariano > > > http://marianopeck.wordpress.com > > > > > > > > > > > -- > > ClÃ©ment BÃ©ra > > Pharo consortium engineer > > https://clementbera.wordpress.com/ > > BÃ¢timent B 40, avenue Halley 59650 Villeneuve d'Ascq > > -- ClÃ©ment BÃ©ra Pharo consortium engineer https://clementbera.wordpress.com/ BÃ¢timent B 40, avenue Halley 59650 Villeneuve d'Ascq

Oct. 18, 2017

Re: [Pharo-dev] StringMorph removing Text styling

by Peter UhnÃ¡k

> So maybe renaming (subclassing?) StringMorph to LabelMorph to make it more > clear? > Never mind. This apparently exists and I am blind... Peter

Oct. 18, 2017

Re: [Pharo-dev] StringMorph removing Text styling

by Peter UhnÃ¡k

There's a cursor issue though... when you have TextMorphs in e.g. list (FastTable), then the cursor changes to text selection, that's why I wanted to use StringMorph. So maybe renaming (subclassing?) StringMorph to LabelMorph to make it more clear? Peter On Wed, Oct 18, 2017 at 1:30 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > Personally I would expect that the second version should be fixed to > discard formatting. We have the TextMorph for such cases. > > -- Pavel > > 2017-10-18 11:40 GMT+02:00 Peter UhnÃ¡k <i.uhnak(a)gmail.com>: > >> Hi, >> >> I am not sure if this is a bug or a feature, but converting Text to >> StringMorph clears some of the formating. >> >> Compare: >> >> |text| >> text := 'hello' asText makeAllColor: Color red. >> >> text asStringMorph openInWindow. "no color" >> (StringMorph contents: text) openInWindow. "with color" >> >> Bug? Feature? >> >> Thanks, >> Peter >> > >

Oct. 18, 2017

Re: [Pharo-dev] Compiler Explorer (for the Compiler guys) #offtopic

by Jan Vrany

Hi, > I was thinking of changing that, adding a primitive which answers a > CogMethod as a bytearray, but you still need to use external libs > such as gdb/lldb to disassemble your binary I use udis86 [1] for that purpose. Tiny, clean & dead easy to use.Â x86 only, though. HTH. Jan [1]: http://udis86.sourceforge.net/ > (which is possible but not working out of the box everywhere and it > does not make sense to include that as part of the VM), and you still > need something extracted from VMMaker to make sense out of the > CogMethod header and CogMethod map, which has a lot of dependencies, > so likely the whole VMMaker itself. And I'm not talking about all the > different CogMethod flavors. > > With VMMaker loaded you can do it with in-image compilation but you > will have the machine code with only unlinked sends. With linked send > it's more difficult as you need to really simulate the whole runtime. > > Overall there's no amazing solution. In the example shown it's easier > as in C++ the code is not modified while it runs and you've already > have gdb/lldb APIs to show what they show. > > Â > > > > > > Â > > > > > > > > > Cheers, > > > Doru > > > > > > > > > > On Oct 17, 2017, at 10:03 PM, Esteban A. Maringolo <emaringolo@ > > > > gmail.com> wrote: > > > > > > > > I'm sure we could do this with ease, because we already can see > > > > the bytecodes. > > > > > > > > https://godbolt.org/ > > > > > > > > Regards! > > > > > > > > Esteban A. Maringolo > > > > > > > > > > -- > > > www.tudorgirba.com > > > www.feenk.com > > > > > > "The coherence of a trip is given by the clearness of the goal." > > > > > > > > > > > > > > > > > > > > > > > > > > --Â > > Mariano > > http://marianopeck.wordpress.com > > > > > > --Â > ClÃ©ment BÃ©ra > Pharo consortium engineer > https://clementbera.wordpress.com/ > BÃ¢timent B 40, avenue Halley 59650Â Villeneuve d'Ascq

Oct. 18, 2017

Pharo7 DebugSession fails because of Roassal2Spec?

by Alistair Grant

Hi All, Attempting to load DataFrame with: Metacello new baseline: 'DataFrame'; repository: 'github://PolyMathOrg/DataFrame'; load. results in what looks like a stack trace being displayed on the window background. See the attached image. The system is unresponsive until interrupted (Ctrl-. on linux). The stack in the image appears to match the stack included below from MorphicUIManager>>onPrimitiveError: onwards. I'm not at all familiar with this code, but it looks a bit like the load has failed, and then the attempt to trap it and open a debugger has failed. One of the failures, at least, appears to be because the Roassal2Spec package overrides the inherited definition of #announcer in NewValueHolder to remove the lazy initialisation. However the NewValueHolder>>announcer extension has been there for a couple of years, so I presume it is some other change that has triggered the problem. Can someone, Alexandre? :-), comment on whether NewValueHolder>>announcer should be removed? Thanks, Alistair [ delaySemaphore wait ] in Delay>>wait in Block: [ delaySemaphore wait ] BlockClosure>>ifCurtailed: Delay>>wait [ | evt type | (Delay forMilliseconds: 20) wait. Sensor nextEvent ifNotNil: [ :evtBuf | type := evtBuf first. type = EventTypeKeyboard ifTrue: [ evt := World currentHand generateKeyboardEvent: evtBuf. evt isKeystroke ifTrue: [ | c | c := evt keyCharacter. (c asciiValue < 128 or: [ c isSeparator or: [ c isSpecial or: [ c = Character backspace ] ] ]) ifTrue: [ Sensor flushKeyboard. ^ evt keyCharacter ] ] ] ] ] in Transcripter class>>nextChar in Block: [ | evt type |... BlockClosure>>repeat Transcripter class>>nextChar Transcripter class>>askForEmergencyEvaluatorOrExitWithText: MorphicUIManager>>onPrimitiveError: DebugSession(Object)>>primitiveError: DebugSession>>signalDebuggerError: [ :ex | debugSession signalDebuggerError: ex ] in [ "schedule debugger in deferred UI message to address redraw problems after opening a debugger e.g. from the testrunner." [ Smalltalk tools debugger openOn: debugSession withFullView: bool andNotification: notificationString ] on: Error do: [ :ex | debugSession signalDebuggerError: ex ] ] in MorphicUIManager>>debugProcess:context:label:fullView:notification: in Block: [ :ex | debugSession signalDebuggerError: ex ] BlockClosure>>cull: Context>>evaluateSignal: Context>>handleSignal: MessageNotUnderstood(Exception)>>signal UndefinedObject(Object)>>doesNotUnderstand: #announce: DictionaryValueHolder(NewValueHolder)>>valueChanged: [ | oldValue | oldValue := value. value := anObject. self valueChanged: oldValue ] in DictionaryValueHolder(NewValueHolder)>>value: in Block: [ | oldValue |... BlockClosure>>ensure: DictionaryValueHolder(NewValueHolder)>>handleCircularReferencesWhile: DictionaryValueHolder(NewValueHolder)>>value: DictionaryValueHolder>>initialize DictionaryValueHolder class(Behavior)>>new DictionaryValueHolder class(NewValueHolder class)>>value: OrderedIdentityDictionary(OrderedDictionary)>>asValueHolder GTSpecPreDebugWindow(DynamicComposablePresenter)>>initialize GTSpecPreDebugWindow(SpecPreDebugWindow)>>initialize GTSpecPreDebugWindow class(Behavior)>>new GTGenericStackDebugger>>openWithNotification: GTGenericStackDebugger class(GTMoldableDebugger class)>>openOn:withFullView:andNotification: --

Oct. 18, 2017

[Pharo 7.0-dev] Build #209: 17754-CheckBoxModel--RadioButtonModel-ignore-disable

by ci-pharo-ci-jenkins2＠inria.fr

There is a new Pharo build available! The status of the build #209 was: SUCCESS. The Pull Request #372 was integrated: "17754-CheckBoxModel--RadioButtonModel-ignore-disable" Pull request url: https://github.com/pharo-project/pharo/pull/372 Issue Url: https://pharo.fogbugz.com/f/cases/17754 Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…

Oct. 18, 2017

Re: [Pharo-dev] Compiler Explorer (for the Compiler guys) #offtopic

by ClÃ©ment Bera

On Wed, Oct 18, 2017 at 1:42 PM, Mariano Martinez Peck < marianopeck(a)gmail.com> wrote: > > > On Wed, Oct 18, 2017 at 3:36 AM, Tudor Girba <tudor(a)tudorgirba.com> wrote: > >> Hi, >> >> This is already available since some time in the GTInspector. You can >> just inspect a compiled method. You will get both the bytecode and the AST >> (with source tracking, too). In fact, this was one of the original reasons >> why GTInspector exists: I wanted to figure out how the AST is structured so >> that I can write queries against it :). >> > > > Sure, we just miss machine code :) > Yeah the machine code is what matters. > But...AFAIR to get the machine code of a method it needs VMMaker to be > loaded.... > I was thinking of changing that, adding a primitive which answers a CogMethod as a bytearray, but you still need to use external libs such as gdb/lldb to disassemble your binary (which is possible but not working out of the box everywhere and it does not make sense to include that as part of the VM), and you still need something extracted from VMMaker to make sense out of the CogMethod header and CogMethod map, which has a lot of dependencies, so likely the whole VMMaker itself. And I'm not talking about all the different CogMethod flavors. With VMMaker loaded you can do it with in-image compilation but you will have the machine code with only unlinked sends. With linked send it's more difficult as you need to really simulate the whole runtime. Overall there's no amazing solution. In the example shown it's easier as in C++ the code is not modified while it runs and you've already have gdb/lldb APIs to show what they show. > > > > > >> >> >> Cheers, >> Doru >> >> >> On Oct 17, 2017, at 10:03 PM, Esteban A. Maringolo <emaringolo(a)gmail.com> >> wrote: >> >> I'm sure we could do this with ease, because we already can see the >> bytecodes. >> >> https://godbolt.org/ >> >> Regards! >> >> Esteban A. Maringolo >> >> >> -- >> www.tudorgirba.com >> www.feenk.com >> >> "The coherence of a trip is given by the clearness of the goal." >> >> >> >> >> >> > > > -- > Mariano > http://marianopeck.wordpress.com > -- ClÃ©ment BÃ©ra Pharo consortium engineer https://clementbera.wordpress.com/ BÃ¢timent B 40, avenue Halley 59650 Villeneuve d'Ascq

Oct. 18, 2017

Re: [Pharo-dev] Compiler Explorer (for the Compiler guys) #offtopic

by Esteban A. Maringolo

Dolphin's debugger was what came to my mind, it had the "dissasembly" way of debugging VM bytecode. It doesn't provide was native code ASM, and as I understand GTInspector neither. Regards! Esteban A. Maringolo 2017-10-18 6:55 GMT-03:00 Dimitris Chloupis <kilon.alios(a)gmail.com>: > Yeap GTInspector was the first thing that came to my mind , great tool :) > > On Wed, Oct 18, 2017 at 9:37 AM Tudor Girba <tudor(a)tudorgirba.com> wrote: > >> Hi, >> >> This is already available since some time in the GTInspector. You can >> just inspect a compiled method. You will get both the bytecode and the AST >> (with source tracking, too). In fact, this was one of the original reasons >> why GTInspector exists: I wanted to figure out how the AST is structured so >> that I can write queries against it :). >> >> >> Cheers, >> Doru >> >> >> On Oct 17, 2017, at 10:03 PM, Esteban A. Maringolo <emaringolo(a)gmail.com> >> wrote: >> >> I'm sure we could do this with ease, because we already can see the >> bytecodes. >> >> https://godbolt.org/ >> >> Regards! >> >> Esteban A. Maringolo >> >> >> -- >> www.tudorgirba.com >> www.feenk.com >> >> "The coherence of a trip is given by the clearness of the goal." >> >> >> >> >> >>

Oct. 18, 2017

Re: [Pharo-dev] Compiler Explorer (for the Compiler guys) #offtopic

by Mariano Martinez Peck

On Wed, Oct 18, 2017 at 3:36 AM, Tudor Girba <tudor(a)tudorgirba.com> wrote: > Hi, > > This is already available since some time in the GTInspector. You can just > inspect a compiled method. You will get both the bytecode and the AST (with > source tracking, too). In fact, this was one of the original reasons why > GTInspector exists: I wanted to figure out how the AST is structured so > that I can write queries against it :). > Sure, we just miss machine code :) But...AFAIR to get the machine code of a method it needs VMMaker to be loaded.... > > > Cheers, > Doru > > > On Oct 17, 2017, at 10:03 PM, Esteban A. Maringolo <emaringolo(a)gmail.com> > wrote: > > I'm sure we could do this with ease, because we already can see the > bytecodes. > > https://godbolt.org/ > > Regards! > > Esteban A. Maringolo > > > -- > www.tudorgirba.com > www.feenk.com > > "The coherence of a trip is given by the clearness of the goal." > > > > > > -- Mariano http://marianopeck.wordpress.com

Oct. 18, 2017

Re: [Pharo-dev] StringMorph removing Text styling

by Pavel Krivanek

Personally I would expect that the second version should be fixed to discard formatting. We have the TextMorph for such cases. -- Pavel 2017-10-18 11:40 GMT+02:00 Peter UhnÃ¡k <i.uhnak(a)gmail.com>: > Hi, > > I am not sure if this is a bug or a feature, but converting Text to > StringMorph clears some of the formating. > > Compare: > > |text| > text := 'hello' asText makeAllColor: Color red. > > text asStringMorph openInWindow. "no color" > (StringMorph contents: text) openInWindow. "with color" > > Bug? Feature? > > Thanks, > Peter >

Oct. 18, 2017