August 2013 - Pharo-dev - lists.pharo.org

Mariano & Behavior>>methodDict
by ClÃ©ment Bera Aug. 9, 2013

Aug. 9, 2013

Hello Mariano and others, In Pharo there is a method: Behavior>>methodDict methodDict == nil ifTrue: [^ MethodDictionary new ]. ^ methodDict It seems (according to the latest author) that it is Mariano who changed it from: Behavior>>methodDict methodDict == nil ifTrue: [self error: 'MethodDict is nil']. ^ methodDict A comment is present: "The method dictionary of a class can be nil when we want to use the #cannotInterpret: hook. Indeed when a class dictionary is nil, the VM sends the message cannotInterpret: to the receiver but starting the look up in the superclass of the class whose method dictionary was nil. Now the system relies that when the message methodDict is sent to a class a method dictionary is returned. In order to prevent the complaints of tools and IDE unaware of this feature, we fool them by providing an empty MethodDictionary. This will hopefully work in most cases, but the tools will loose the ability to modify the behavior of this behavior. The user of #cannotInterpret: should be aware of this." Now my problem is when I do: MyClass methodDict: nil. MyClass new name. It crashes the image in latest Pharo 3. Mariano, do you have an idea on how to fix it ? I ask you because you are the latest author and it seems tat the bug comes from this method. Thanks for any answer.

3 2

[update 3.0] #30328
by StÃ©phane Ducasse Aug. 8, 2013

Aug. 8, 2013

1 0

Changes in the repositories for the VM
by Esteban Lorenzano Aug. 8, 2013

Aug. 8, 2013

Hi, Ok, we finally decided to make some changes longtime postponed, to improve our way of building the VM. Squeak guys are still using the old mechanism, so this just applies to Pharo and the continuous integration under the Pharo umbrella... but please notice that ANY vm flavor can be built with this source repository. The first change was to put all sources together. No more MCZ files to keep VMMaker and plugins and platform sources as a GIT repository. We are using filetree to keep source files, so is just like a different kind of monticello repository. With this, now a commit is atomic and the platform sources will be in sync with VMMaker sources. Also, that simplifies a lot the ConfigurationOfCog file, since the commit number is the version number... now we just need the baseline (not version anymore). So now you can checkout a version of the repository and do: ConfigurationOfCog loadGit. and that's all :) (from our point of view, this simplification is an huge advantage over the older process). The second change is just a relocation. Now the VM source lives in GitHub :) The reason of this move is because GitHub has a lot more functionality than the old Gitorious, also most of the people using filetree are already using GitHub so is easier for them to fork, etc. Also, it is hyped so we wanted to be in that wave ;) The new repository can be found at: https://github.com/pharo-project/pharovm Cheers, Esteban

2 1

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 8, 2013

Aug. 8, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=win/… 7 regressions found. KernelTests.Classes.BehaviorTest.testBehaviorRespectsPolymorphismWithTraitBehavior KernelTests.Classes.MetaclassTest.testMetaclassAndTraitClassRespectsPolymorphismRules KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testNotEnvironment Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testOrEnvironment Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testRemoveSelectorByAndAndNot Tests.Release.ReleaseTest.testInstanceSideMethodsWithNilKeyInLastLiteral

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 8, 2013

Aug. 8, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=mac/… 7 regressions found. KernelTests.Classes.BehaviorTest.testBehaviorRespectsPolymorphismWithTraitBehavior KernelTests.Classes.MetaclassTest.testMetaclassAndTraitClassRespectsPolymorphismRules KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testNotEnvironment Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testOrEnvironment Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testRemoveSelectorByAndAndNot Tests.Release.ReleaseTest.testInstanceSideMethodsWithNilKeyInLastLiteral

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 8, 2013

Aug. 8, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=linu… 7 regressions found. KernelTests.Classes.BehaviorTest.testBehaviorRespectsPolymorphismWithTraitBehavior KernelTests.Classes.MetaclassTest.testMetaclassAndTraitClassRespectsPolymorphismRules KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testNotEnvironment Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testOrEnvironment Refactoring.Tests.Environment.RBBrowserEnvironmentTest.testRemoveSelectorByAndAndNot Tests.Release.ReleaseTest.testInstanceSideMethodsWithNilKeyInLastLiteral

1 0

[update 3.0] #30327
by Esteban Lorenzano Aug. 8, 2013

Aug. 8, 2013

30327 ----- 11144 Merge Latest NativeBoost with Pharo 3.0 https://pharo.fogbugz.com/f/cases/11144

1 0

Fwd: SmalltalkHub down
by Nicolas Cellier Aug. 8, 2013

Aug. 8, 2013

---------- Forwarded message ---------- From: Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com> Date: 2013/8/8 Subject: SmalltalkHub down To: Pharo Development <Pharo-project(a)lists.gforge.inria.fr> http://smalltalkhub.com/ -> 502 Proxy Error

1 0

Please test: New PharoVM built
by Esteban Lorenzano Aug. 7, 2013

Aug. 7, 2013

Hi, Guille built a new vm with an attept of fixing the key combinations problem. Can you test it? http://files.pharo.org/vm/pharo/ (and pick the latest for your platform) Thanks, Esteban

3 2

becomeCompactSimplyAt: and Integer class>>initialize
by Martin Dias Aug. 7, 2013

Aug. 7, 2013

Hi 1) I'm working on an issue related to compact classes index and I discovered that behaviours implement #becomeCompactSimplyAt:, which has this comment: "Make me compact, but don't update the instances. *For importing segments* ." This made me think that it's no longer needed in the image since we removed ImageSegment. In other hand, I don't know why sender might decide if the instances of the class must be updated or not... do you? 2) The only users in Pharo 3 are in: Integer class >> initialize "Integer initialize" "Ensure we have the right compact class index" "LPI has been a compact class forever - just ensure basic correctness" (LargePositiveInteger indexIfCompact = 5) ifFalse:[ (Smalltalk compactClassesArray at: 5) ifNil:[LargePositiveInteger becomeCompactSimplyAt: 5] ifNotNil:[self error: 'Unexpected compact class setup']]. (LargeNegativeInteger indexIfCompact = 4) ifFalse:[ (Smalltalk compactClassesArray at: 4) ifNil:[LargeNegativeInteger becomeCompactSimplyAt: 4] ifNotNil:[self error: 'Unexpected compact class setup']]. Now, looking for other users of #indexIfCompact, I see that here is the only place in the image where compact indexes are checked/fixed. 3) I wonder if we should remove both #becomeCompactSimplyAt: and Integer class>>initialize. Thank you in advance. MartÃn

1 0

NB and OS specific tests
by Torsten Bergmann Aug. 7, 2013

Aug. 7, 2013

Hi Igor, NB includes OS specific classes and examples that will/may require OS specific tests. IMHO it would make sense to change "NativeBoost-Tests" category into - "NativeBoost-Tests-Core" - "NativeBoost-Tests-Mac" - "NativeBoost-Tests-Unix" - "NativeBoost-Tests-Win32" while still keeping "NativeBoost-Tests" as one package. Later the package can be split easily (but only if necessary). So we have categories for the NB Core tests and OS specific tests similar to the other packages/categories. I would further provide common superclasses for OS specific tests, for instance TestCase subclass: #NBWin32SpecificTest instanceVariableNames: '' classVariableNames: '' poolDictionaries: '' category: 'NativeBoost-Tests-Win32' and similar NBUnixSpecificTest and NBMacSpecificTest also. By overriding their #isAbstract method we can make their hierarchy only be runnable on the specific OS. So the OS specific Jenkins slaves will show broken OS specific code, Unix tests will not run on Mac, etc. Therefore we can control that the tests only run on the OS platform and find out when things fail. What do you think? I can do it when there is another minute. Thx Torsten

3 2

NB and OS specific tests
by Torsten Bergmann Aug. 7, 2013

Aug. 7, 2013

Hi Stef, #nbBindingOf: is an extension on Behavior from the NativeBoost Core package. I tried and it works: - using Pharo2.0 Latest update: #20603 - adding http://smalltalkhub.com/#!/~Pharo/NativeBoost - loading latest version of Core, Examples, Pools, Mac, Unix, Win32, Tests All NB tests pass. Thx T.

2 1

[update 3.0] #30326
by StÃ©phane Ducasse Aug. 7, 2013

Aug. 7, 2013

2 1

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 7, 2013

Aug. 7, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=win/… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 7, 2013

Aug. 7, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=mac/… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 7, 2013

Aug. 7, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=linu… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[update 3.0] #30324
by Esteban Lorenzano Aug. 6, 2013

Aug. 6, 2013

4 3

moose build - pharo vm crash
by Tudor Girba Aug. 6, 2013

Aug. 6, 2013

Hi, It seems that the Moose build crashes the VM since yesterday evening due to a failing test. The image is based on Pharo 2.0 and is running a stable VM on Ubuntu. See some details here: https://ci.inria.fr/moose/job/Moose-latest-dev-4.8/804/console The script to reproduce the problem (on Ubuntu) is: #------------------- wget --quiet -O - http://get.pharo.org/20+vm | bash ./pharo Pharo.image save $JOB_NAME REPO=http://www.smalltalkhub.com/mc/Moose/Moose/main ./pharo $JOB_NAME.image config $REPO ConfigurationOfMoose --install=development ./pharo $JOB_NAME.image mooseimagesetup ./pharo $JOB_NAME.image moosetest --junit-xml-output mv ./pharo-vm/PharoV20.sources ./ zip $JOB_NAME.zip $JOB_NAME.image $JOB_NAME.changes PharoV20.sources #------------------- Cheers, Doru -- www.tudorgirba.com "Every thing has its own flow"

4 8

Flamel Update
by Gisela Decuzzi Aug. 6, 2013

Aug. 6, 2013

Hi, I'm working for the Gsoc project: better rw rule tools and my mentors suggested me to send the weekly progress to the mailing list because other people can be interested. Work done: - defining a basic UI that admit - write an example expression - write the matching / transforming pattern - restrict the scope where running the rule (now select classes or packages) - run a rule for matching / transforming This week progress: - Show / Navigate the result after matching: [image: ImÃ¡genes integradas 1] - blog post about defining the matching expression: http://pharorwrules.wordpress.com/2013/08/06/just-want-to-match/ - bugfixing - middle term evaluation - fixing repository and package mess (that I've done...) For next week: - correct problems when transforming - vitualize the result after a transformation (rehuse part of the search result for this) - add tests for transforming - blog post about rw engine (how to define a rule from the scratch and why I implemented my own rule that matchs and transforms) - (if time) define the matching pattern for the expression (is an expression, is a method? Should try with the best?), right now is matching as a method pattern and sometimes is confusing. If someone wants to see more planed task can have a look to the Trello board in: https://trello.com/b/XqfJGqeB/flamel If someone wants to give a try and give me some feedback the code is in smalltalkhub: MCSmalltalkhubRepository owner: 'gisela' project: 'Flamel' user: '' password: '' The blog for this project is: http://pharorwrules.wordpress.com Cheers, Gise.

2 4

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 6, 2013

Aug. 6, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=win/… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 6, 2013

Aug. 6, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=mac/… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 6, 2013

Aug. 6, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=linu… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[update 3.0] #30325
by Esteban Lorenzano Aug. 6, 2013

Aug. 6, 2013

30325 ----- 11294 Bugs in refactoring when using traits https://pharo.fogbugz.com/f/cases/11294 Diff information: http://smalltalkhub.com/mc/Pharo/Pharo30/main/Refactoring-Environment-Esteb… http://smalltalkhub.com/mc/Pharo/Pharo30/main/Refactoring-Core-EstebanLoren… http://smalltalkhub.com/mc/Pharo/Pharo30/main/NautilusRefactoring-EstebanLo… http://smalltalkhub.com/mc/Pharo/Pharo30/main/Nautilus-EstebanLorenzano.506…

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 6, 2013

Aug. 6, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=win/… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 6, 2013

Aug. 6, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=mac/… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Aug. 6, 2013

Aug. 6, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=linu… 1 regressions found. KernelTests.Numbers.LargePositiveIntegerTest.testMethodsOfTheClassShouldNotBeRepeatedInItsSuperclasses

1 0

[update30] #30323
by Esteban Lorenzano Aug. 6, 2013

Aug. 6, 2013

30323 ----- 11200 Repeated methods in superclasses, explicit requirement no implemented, repeated methods in traits https://pharo.fogbugz.com/f/cases/11200 (pass 1)

1 0

Re: [Pharo-dev] KeyDown and KeyUp without automatic key repeating.
by Ronie Salgado Aug. 5, 2013

Aug. 5, 2013

Thank you, Its not exactly what I need, but using it I can hack the result that I want using some dictionary holding the key down state. Greetings, Ronie Salgado 2013/8/5 Erwan Douaille <douailleerwan(a)gmail.com> > I made an example, i guess it's what you need. > Cheat with the keyUp > > ExempleKey new > > Mouse over the morph and try key left / right > > > 2013/8/5 Ronie Salgado <roniesalg(a)gmail.com> > >> Hello, >> I need to know how I can handle a KeyDown and a KeyUp morph event without >> automatic key repeating, to be able to setup an object that performs per >> frame animation. >> For the ones who have worked with SDL, what I need is something like >> SDL_EnableKeyRepeat(0, 0) >> >> Greetings, >> Ronie Salgado >> > > > > -- > Best regards, > > Douaille Erwan <douaille.erwan(a)gmail.com> >

1 0

renaming method propagation
by StÃ©phane Ducasse Aug. 5, 2013

Aug. 5, 2013

Hi guys I found a problem with method renaming. The scenario is the following one: - you have a message browser open with a list a methods (resulting from a sender/implementor). - you rename a method and you get an error why because the message browser does not replace the currently changed method. and the RGMethodDefinition that represented the renamed method defines the code like that MessageBrowser>>wrapItem: anItem | s | s :=String streamContents: [ :aStream | 3 to: (cachedHierarchy at: anItem) size do: [:i | aStream << ' ']. aStream << anItem methodClass name << ' ('. anItem isFromTrait ifTrue: [ aStream << anItem compiledMethod origin name; space ]. aStream << (anItem category ifNil: ['']) <<')']. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ {s. anItem selector}. RGMethodDefinition>>compiledMethod "Retrieves the compiled method of this definition if exists" | rClass | (rClass := self realClass) notNil ifTrue: [ (rClass includesSelector: self selector) ifTrue: [ ^rClass >> self selector ] ]. ^nil category ^ self protocol protocol self isActive ifTrue: [ ^ self compiledMethod category ]]. self isHistorical ifTrue: [ ^ self protocolAtPointer ifNil:[ self compiledMethod ifNil:[ protocol ] ifNotNil:[ :cm| cm category ] ] ]. ^ protocol So I patched the protocol method as follows protocol self isActive ifTrue: [ ^ self compiledMethod ifNil: [ ' '] ifNotNilDo: [ :cm | cm category ]]. "I hate this ifNil test, the problem is that when we rename a method, a method definition in a list may still refers to the old selector and the definition of compiled method is that case is not finding the associated method because it holds the old selector and not the one corresponding to the one defined in the class." self isHistorical ifTrue: [ ^ self protocolAtPointer ifNil:[ self compiledMethod ifNil:[ protocol ] ifNotNil:[ :cm| cm category ] ] ]. ^ protocol I imagine that the best solution would be that the tools update their contents but I wanted to know your point of view. Stef

2 2

KeyDown and KeyUp without automatic key repeating.
by Ronie Salgado Aug. 5, 2013

Aug. 5, 2013

Hello, I need to know how I can handle a KeyDown and a KeyUp morph event without automatic key repeating, to be able to setup an object that performs per frame animation. For the ones who have worked with SDL, what I need is something like SDL_EnableKeyRepeat(0, 0) Greetings, Ronie Salgado

1 0

WhatsUp from: 2013-08-05 until: 2013-08-18
by seaside＠rmod.lille.inria.fr Aug. 5, 2013

Aug. 5, 2013

Hi! We're sending this automatic email twice a month, to give the community an opportunity to easily know what's happening and to coordinate efforts. Just answer informally, and feel free to spawn discussions thereafter! ### Here's what I've been up to since the last WhatsUp: - $HEROIC_ACHIEVEMENTS_OR_DISMAL_FAILURES_OR_SIMPLE_BORING_NECESSARY_TASKS ### What's next, until 2013-08-18 (*): - $NEXT_STEPS_TOWARDS_WORLD_DOMINATION (*) we'll be expecting results by then ;)

3 2

Did canUnderstand: changed?
by StÃ©phane Ducasse Aug. 5, 2013

Aug. 5, 2013

Hi why canUnderstand: now handles shouldNotImplement? Stef

2 1

about on:do: -> onKey:do:
by StÃ©phane Ducasse Aug. 5, 2013

Aug. 5, 2013

Hi ben and others we are introducing onKey:do: for bindings so that we can control the complexity and the registration of announcements. So if you can I have a look it would be good. https://pharo.fogbugz.com/default.asp?11299 Stef

5 7

[update 3.0] #30322
by Esteban Lorenzano Aug. 5, 2013

Aug. 5, 2013

1 0

Smalltalk = strongly typed
by btc＠openinworld.com Aug. 5, 2013

Aug. 5, 2013

greetings all, I'm in the final weeks of writing up my Masters dissertation and seeking some scholarly references to Smalltalk being "Strongly Typed." I my review of Smalltalk I was surprised to find that [1] describes Smalltalk as Strongly Typed, since Smalltalk is sometimes denigrated as being untyped / weakly typed. From reviewing discussion forums this now makes sense, but I can only find one of scholarly reference that briefly mentions this [2]. The most enlightening is [3] which defines Type Strength as: "A strongly typed language prevents any operation on the wrong type of data. In weakly typed languages there are ways to escape this restriction: type conversions" meaning that getting a MNU is a form of Strong Typing since you can't make a Smalltalk object run a method that is not its own. The problem appears to be that Strong Typing has been synonymous with Static Typing for a long time, and Static Typing strongly ties types to variables, except in Dynamically Typed languages, I think types can be considered independently from variables, in which case the definition of [3] has some merit, hence Smalltalk is Strongly Typed. Sounds controversial, so I'm just hoping for some peer reviewed backup - but only you have something easily to hand. This is just a small thing I can just leave out if necessary. cheers -ben [1] http://www.squeak.org/Features/ [2] p15, http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.35.7507&rep=rep1&t… [3] http://www.cs.kuleuven.ac.be/publicaties/rapporten/cw/CW415.pdf

12 20

Aug. 4, 2013

addDependent: anObject self announcer on: MorphChanged do: [ :ann | anObject update: ann selector ]. self announcer on: MorphChangedWithArguments do: [ :ann | anObject update: ann selector with: ann arguments ]. ^ anObject is an example of the problem because addDependent: was holding also weakly its dependent and now we cannot use weak with blocks. Stef

1 2

Aug. 4, 2013

Hi guys we are looking at the memory leak problems with igor and here are the conclusions: we are renaming on:do: for shortcut into onKey:do: so that we can see a bit more (normally easy) Now the complex part. on:do: should not be used in weak announcement. because the block holds a context that prevents the garbage collection to reclaim the object. For blocks implicitly the receiver of the message containing the block is held strongly. Now the block cannot We are adding a message in the on:do: implementation and we will have a look at the complete image. on:do: should only be used for strong references. The problem is that for UI we need weak. So we are looking at users for now. Stef

1 0

Aug. 4, 2013

Hi I merged Jannik clean and I commited a new version of the configuration. Now it works on Pharo 20 and also on Pharo 30 And I will produce a new System-Sound for the kernel based on Pavel work and I will produce a new configuration. Stef

3 5

Aug. 3, 2013

Hi I discovered that UUIDGenerator and DigitalSignatureAlgorithm used in some encrypting algo are dependent of Sound. BaseSoundSystem>>:randomBitsFromSoundInput: bitCount "Answer a positive integer with the given number of random bits of 'noise' from a sound input source. Typically, one would use a microphone or line input as the sound source, although many sound cards have enough thermal noise that you get random low-order sample bits even with no microphone connected. Only the least signficant bit of the samples is used. Since not all sound cards support 16-bits of sample resolution, we use the lowest bit that changes." "(1 to: 10) collect: [:i | BaseSoundSystem new randomBitsFromSoundInput: 512]" | recorder buf mid samples bitMask randomBits bit | "collect some sound data" recorder := SoundRecorder new clearRecordedSound. recorder resumeRecording. (Delay forSeconds: 1) wait. recorder stopRecording. buf := recorder condensedSamples. "grab bitCount samples from the middle" mid := buf monoSampleCount // 2. samples := buf copyFrom: mid to: mid + bitCount - 1. "find the least significant bit that varies" bitMask := 1. [bitMask < 16r10000 and: [(samples collect: [:s | s bitAnd: bitMask]) asSet size < 2]] whileTrue: [bitMask := bitMask bitShift: 1]. bitMask = 16r10000 ifTrue: [^ self error: 'sound samples do not vary']. "pack the random bits into a positive integer" randomBits := 0. 1 to: samples size do: [:i | bit := ((samples at: i) bitAnd: bitMask) = 0 ifTrue: [0] ifFalse: [1]. randomBits := (randomBits bitShift: 1) + bit]. ^ randomBits Now I was thinking that when the sound package is not loaded we could have a random number generator. Stef

1 0

Aug. 3, 2013

When I modify a method and click on somewhere else I get a pop up and pressing discard does not work but pops up another. Now I'm trying to reproduce the problem to describe it clearly and I cannot anymore but I'm sure I did get this strange behavior. Stef

2 1

SimpleMIDIPort midiIsSupported returns the first time false and after all the time true :)
by StÃ©phane Ducasse Aug. 3, 2013

Aug. 3, 2013

Strange May be the logic in the primitive is strange. midiIsSupported "Answer true if this platform supports MIDI." ^ self primPortCount > 0 primPortCount "Answer the number of MIDI ports supported by this platform, or zero if this primitive is not implemented." <primitive: 'primitiveMIDIGetPortCount' module: 'MIDIPlugin'> ^ 0 Stef

1 0

Aug. 3, 2013

Because when I merge and I click on apply the change is not applied in the image (when I do not accept the button in the merge tool)? Stef

1 1

Still broken Date comparaison in 2.0?
by hilaire Aug. 3, 2013

Aug. 3, 2013

Instances value from stacking GrowlMorphs: TimeStamp now -> 4 August 2013 1:18:30 am vanishTime -> 4 August 2013 12:51:51.62 am TimeStamp now < vanishTime ->true Inspectors: <http://forum.world.st/file/n4702023/TimesStampInspectors.png> Hilaire -- View this message in context: http://forum.world.st/Still-broken-Date-comparaison-in-2-0-tp4702023.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

1 0

Aug. 3, 2013

Hi guys I wonder what we can do about AppRegistry I imagine that we can remove MailSender. I'm going over the code on the image as well as http://code.google.com/p/pharo/issues/detail?id=6996 and pavel changes. Any feedback is welcome. Stef

1 0

modified method temp references in closures: possible bug?
by Max Leske Aug. 3, 2013

Aug. 3, 2013

I just came across this (possible) bug and I'm able to reproduce it in Pharo 2.0 and 3.0 with the latest PharoVM (found in a 1.3 with NBCog): expected (by my feeble mind): 1 2 actual: 2 2 | methodScope | collection := OrderedCollection new. 1 to: 2 do: [ :index | methodScope := index. dict add: [ methodScope ] ]. collection do: [ :block | Transcript show: block value; cr ] I'm aware that using method temps in this way is not good programming style (usually) but I still think this behavior is curious if not at least very likely to cause a lot of weird errors. Should I open a bug report? Cheers, Max

4 6

Aug. 2, 2013

Display flash: (0@0 corner: 200@200) does not work on Pharo 30 while it was working on 2.0. It looks that this is VM related because on the linux of laurent it is working. I'm too tired to get how to extract the VM versionâ¦. I tried --version and others but it did not work. pharo -version bin unknown option: -version Usage: [<option>...] [<imageName> [<argument>...]] [<option>...] -- [<argument>...] Common <option>s: -help print this help message, then exit -memory <size>[mk] use fixed heap size (added to image size) -headless run in headless (no window) mode (default: false) Notes: <imageName> defaults to `Squeak.image'. If `-memory' is not specified then the heap will grow dynamically. <argument>s are ignored, but are processed by the Squeak image. The first <argument> normally names a Squeak `script' to execute. Precede <arguments> by `--' to use default image. bin pharo -version 'NBCoInterpreter NativeBoost-CogPlugin-EstebanLorenzano.18 uuid: a53445f9-c0c0-4015-97a3-be7db8d9ed6b Mar 13 2013 NBCogit NativeBoost-CogPlugin-EstebanLorenzano.18 uuid: a53445f9-c0c0-4015-97a3-be7db8d9ed6b Mar 13 2013 git://gitorious.org/cogvm/blessed.git Commit: 412abef33cbed05cf1d75329e451d71c0c6aa5a7 Date: 2013-03-13 17:48:50 +0100 By: Esteban Lorenzano <estebanlm(a)gmail.com> Jenkins build #14535 ' Stef too tired and exhausted to think

1 0

Aug. 2, 2013

Rectangle>>drawReverseFrame: rect Display border: rect width: 2 rule: Form reverse fillColor: Color gray and force user to create a dummy rectangle to invoke it. Stef

1 0

Aug. 2, 2013

convertFromCompoundText | readStream converter | readStream := self readStream. ^ self class new: self size streamContents: [ :writeStream| converter := CompoundTextConverter new. converter ifNil: [^ self]. [readStream atEnd] whileFalse: [ writeStream nextPut: (converter nextFromStream: readStream)]]. CompoundTextConverter new -> DNU convertFromSystemString | readStream converter | readStream := self readStream. ^ self class new: self size streamContents: [ :writeStream| converter := LanguageEnvironment defaultSystemConverter. converter ifNil: [^ self]. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ [readStream atEnd] whileFalse: [ writeStream nextPut: (converter nextFromStream: readStream)]]. systemConverterClass self subclassResponsibility. ^ Latin1TextConverter. What!!!! Stef

1 0

3.0 rolled back to 30321
by Esteban Lorenzano Aug. 2, 2013

Aug. 2, 2013

Hi, I needed to rollback to previous integration because there are some problems related to last integration (and Ben is in vacations and cannot deliver a "hot-fix" :) Esteban

2 1

Morph inspect weird behavior
by Erwan Douaille Aug. 2, 2013

Aug. 2, 2013

Hi, I found something else weird, When i inspect a Morph the bounds are resized. Do it : a := Morph newBounds: (Rectangle origin: (0@0) corner: (300@300) ). a color: Color red. a openInWorld Now click on the morph, open halo, inspect it and the bounds are resized. Is that normal ? (I guess not) Thanks :) -- Best regards, Douaille Erwan <douaille.erwan(a)gmail.com>

1 0