Squeak-dev November 2013

squeak-dev@lists.squeakfoundation.org

43 participants
223 discussions

The Trunk: CommandLine-fbs.2.mcz
by commits＠source.squeak.org 02 Nov '13

02 Nov '13

Frank Shearar uploaded a new version of CommandLine to project The Trunk: http://source.squeak.org/trunk/CommandLine-fbs.2.mcz ==================== Summary ==================== Name: CommandLine-fbs.2 Author: fbs Time: 2 November 2013, 9:47:04.27 pm UUID: 414e59b8-4f4a-814d-9dac-b7b9886e92a0 Ancestors: CommandLine-fbs.1 SmalltalkImage current -> Smalltalk. =============== Diff against CommandLine-fbs.1 =============== Item was changed: ----- Method: CommandLineToolSet class>>debugError: (in category 'debugging') ----- debugError: anError "Print out a sensible stack trace and bail" | problemPlace s | self saveSnapshotOnError ifTrue: [Smalltalk saveAs: 'Debug-' , (Smalltalk imageName subStrings: '/') last]. problemPlace := anError signalerContext. s := FileStream stderr. (anError isKindOf: MessageNotUnderstood) ifTrue: [ s nextPutAll: anError messageText; cr; nextPutAll: problemPlace sender methodNode printString; cr]. (problemPlace stackOfSize: 20) do: [:ctx | s cr. ctx printOn: s]. s flush. + Smalltalk snapshot: false andQuit: true! - SmalltalkImage current snapshot: false andQuit: true!

1 0

backup trunk access
by Chris Muller 02 Nov '13

02 Nov '13

Just a reminder, the new-trunk repository is still running at box4.squeak.org:8888/trunk and is blazing fast compared to the regular Squeak trunk. It's kept up-to-date from trunk every hour, on the hour by a cron job. If you just want to access the web site or download individual packages, this works much better at least until source.squeak.org can be restarted. Unfortunately, we can't update our images from it because of the way our update process and MonticelloConfigurations are designed.

1 0

"Duplicate" MethodReferences in implementorsOf:
by Frank Shearar 02 Nov '13

02 Nov '13

Look for the implementors of #rename:. You'll see 17 Trait entries, each = to each other but not ==. I've traced this back to _something_ about Environment's declarations. In a Workspace, n := SystemNavigation default. a := OrderedCollection new. n allBehaviorsDo: [:cls | a add: cls]. {a size. a asSet size} "#(4384 4367)" "4384 - 4367 = 17" So we have 17 duplicates being reported. If you explore Smalltalk in a "full fat" trunk (specifically, one that still has 311Deprecated) and run "self allTraits" you'll get 17. Hmm. Explore "self allTraits" and you'll see the 17 Traits from 311Deprecated. So far, so good. Try cmd-b one of those Traits (say, TTransformationCompatibility and... you'll browse Trait. (You can get to TTransformationCompatibility by opening a new browser and searching for it.) So for some reason Smalltalk thinks that every one of these Trait names map to Trait. For example, cmd-b TTransformationCompatibility in a Workspace and you end up browsing Trait. I'm still investigating why this is broken, but if anyone knows, feel free to let me know :) frank

4 5

Re: [squeak-dev] The Trunk: CommandLine-fbs.1.mcz
by Nicolas Cellier 02 Nov '13

02 Nov '13

+1, please include in the update.mcm! I also like the Pharo alternative to Fuel out the error context... Lighter than full image save. Oh but we don't have Fuel yet :( Who wants to make it with a segment? 2013/10/31 <commits(a)source.squeak.org> > Frank Shearar uploaded a new version of CommandLine to project The Trunk: > http://source.squeak.org/trunk/CommandLine-fbs.1.mcz > > ==================== Summary ==================== > > Name: CommandLine-fbs.1 > Author: fbs > Time: 31 October 2013, 10:32:52.821 pm > UUID: dc871e85-bd16-d742-acbf-4d2b838113d1 > Ancestors: > > CommandLineToolSet does little more than dump errors to stderr. > > DummyUIManager comes from Pharo via Pavel Krivanek's minimising script. > > ==================== Snapshot ==================== > > SystemOrganization addCategory: #'CommandLine-Tools'! > SystemOrganization addCategory: #'CommandLine-UIManager'! > > UIManager subclass: #DummyUIManager > instanceVariableNames: '' > classVariableNames: '' > poolDictionaries: '' > category: 'CommandLine-UIManager'! > > !DummyUIManager commentStamp: 'fbs 10/31/2013 07:36' prior: 0! > I'm an alternative UIManager used to run an the image without GUI. I > redefine methods which require user input as these requests are irrelevant > in a headless environment. ! > > ----- Method: DummyUIManager>>checkForNewDisplaySize (in category > 'display') ----- > checkForNewDisplaySize > Display extent = DisplayScreen actualScreenSize ifTrue: [^ self]. > DisplayScreen startUp. > ! > > ----- Method: DummyUIManager>>chooseDirectory:from: (in category 'ui > requests') ----- > chooseDirectory: label from: dir > ^ nil! > > ----- Method: DummyUIManager>>chooseFrom:lines:title: (in category 'ui > requests') ----- > chooseFrom: aList lines: linesArray title: aString > ^ aList first! > > ----- Method: DummyUIManager>>chooseFrom:values:lines:title: (in category > 'ui requests') ----- > chooseFrom: labelList values: valueList lines: linesArray title: aString > ^ valueList first! > > ----- Method: DummyUIManager>>confirm: (in category 'ui requests') ----- > confirm: queryString > (ProvideAnswerNotification signal: queryString) > ifNotNil: [:answer|^answer]. > > self error: 'No user response possible'! > > ----- Method: DummyUIManager>>confirm:orCancel: (in category 'ui > requests') ----- > confirm: aString orCancel: cancelBlock > (ProvideAnswerNotification signal: aString) ifNotNil: [:answer | > ^answer == #cancel ifTrue: [cancelBlock value] ifFalse: [answer]]. > > self error: 'No user response possible'! > > ----- Method: DummyUIManager>>displayProgress:at:from:to:during: (in > category 'ui requests') ----- > displayProgress: titleString at: aPoint from: minVal to: maxVal during: > workBlock > ^ workBlock value: Association new! > > ----- Method: DummyUIManager>>edit:label:accept: (in category 'ui > requests') ----- > edit: aText label: labelString accept: anAction > ^ nil! > > ----- Method: DummyUIManager>>fontFromUser: (in category 'ui requests') > ----- > fontFromUser: priorFont > self error: 'No user response possible'! > > ----- Method: DummyUIManager>>inform: (in category 'ui requests') ----- > inform: aString > "Nothing to be done here"! > > ----- Method: DummyUIManager>>informUserDuring: (in category 'ui > requests') ----- > informUserDuring: aBlock > aBlock value: nil! > > ----- Method: DummyUIManager>>newDisplayDepthNoRestore: (in category > 'display') ----- > newDisplayDepthNoRestore: pixelSize > "Change depths. Check if there is enough space!! , di" > | area need | > pixelSize = Display depth ifTrue: [^ self "no change"]. > pixelSize abs < Display depth ifFalse: > ["Make sure there is enough space" > area := Display boundingBox area. "pixels" > > need := (area * (pixelSize abs - Display depth) // 8) > "new bytes needed" > + Smalltalk lowSpaceThreshold. > (Smalltalk garbageCollectMost <= need > and: [Smalltalk garbageCollect <= need]) > ifTrue: [self error: 'Insufficient free space']]. > Display setExtent: Display extent depth: pixelSize. > > DisplayScreen startUp! > > ----- Method: DummyUIManager>>request:initialAnswer: (in category 'ui > requests') ----- > request: queryString initialAnswer: defaultAnswer > (ProvideAnswerNotification signal: queryString) > ifNotNil: [:answer | > ^ answer == #default ifTrue: [defaultAnswer] ifFalse: > [answer]]. > > self error: 'No user response possible'! > > ----- Method: DummyUIManager>>requestPassword: (in category 'ui requests') > ----- > requestPassword: queryString > ^ self request: queryString initialAnswer: ''! > > ----- Method: DummyUIManager>>restoreDisplay (in category 'display') ----- > restoreDisplay! > > ----- Method: DummyUIManager>>restoreDisplayAfter: (in category 'display') > ----- > restoreDisplayAfter: aBlock > aBlock value. > Sensor waitButton.! > > StandardToolSet subclass: #CommandLineToolSet > instanceVariableNames: '' > classVariableNames: 'SaveSnapshotOnError' > poolDictionaries: '' > category: 'CommandLine-Tools'! > > ----- Method: CommandLineToolSet class>>debugError: (in category > 'debugging') ----- > debugError: anError > "Print out a sensible stack trace and bail" > | problemPlace s | > self saveSnapshotOnError > ifTrue: [Smalltalk saveAs: 'Debug-' , (Smalltalk imageName > subStrings: '/') last]. > problemPlace := anError signalerContext. > s := FileStream stderr. > (anError isKindOf: MessageNotUnderstood) ifTrue: [ > s > nextPutAll: anError messageText; cr; > nextPutAll: problemPlace sender methodNode > printString; cr]. > (problemPlace stackOfSize: 20) do: [:ctx | s cr. ctx printOn: s]. > s flush. > > SmalltalkImage current snapshot: false andQuit: true! > > ----- Method: CommandLineToolSet class>>debugSyntaxError: (in category > 'debugging') ----- > debugSyntaxError: anError > | s | > s := FileStream stderr. > s nextPutAll: '----- Syntax error -----'; cr. > s nextPutAll: anError errorCode; cr. > s nextPutAll: '----- Syntax error -----'; cr. > > self debugError: anError! > > ----- Method: CommandLineToolSet class>>saveSnapshotOnError (in category > 'preferences') ----- > saveSnapshotOnError > <preference: 'Save snapshot of image on failure' > category: 'debug' > description: 'If true, saves a snapshot of the failing image to > the current directory.' > type: #Boolean> > ^ SaveSnapshotOnError ifNil: [SaveSnapshotOnError := false].! > > ----- Method: CommandLineToolSet class>>saveSnapshotOnError: (in category > 'preferences') ----- > saveSnapshotOnError: aBoolean > SaveSnapshotOnError := aBoolean.! > > ----- Method: CommandLineToolSet class>>unload (in category 'class > initialization') ----- > unload > ToolSet unregister: self.! > > >

3 3

http://www.squeak.org/ is down
by Chris Muller 02 Nov '13

02 Nov '13

It says Bad Gateway.

1 0

Broken scoping in a Workspace
by Frank Shearar 02 Nov '13

02 Nov '13

a := OrderedCollection new. (Smalltalk globals instVarNamed: 'declarations') keysAndValuesDo: [:key :value | a add: value]. b := OrderedCollection new. Smalltalk allClassesAndTraitsDo: [:v | b add: value]. b Note that value should be scoped only to the first block. That second block should _not_ be able to see value. But it can, so (a) the code compiles and (b) produces #(nil nil nil ...). frank

2 2

Re: [squeak-dev] The Trunk: System-ul.613.mcz
by Frank Shearar 02 Nov '13

02 Nov '13

On 2 November 2013 12:54, <commits(a)source.squeak.org> wrote: > Levente Uzonyi uploaded a new version of System to project The Trunk: > http://source.squeak.org/trunk/System-ul.613.mcz > > ==================== Summary ==================== > > Name: System-ul.613 > Author: ul > Time: 2 November 2013, 1:47:09.671 pm > UUID: de29e18c-22f5-4299-823a-e6b207c3f552 > Ancestors: System-cmm.612 > > - fix: SystemNavigation >> #allBehaviorsDo: shouldn't enumerate Trait more than once. > > =============== Diff against System-cmm.612 =============== > > Item was changed: > ----- Method: SystemNavigation>>allBehaviorsDo: (in category 'query') ----- > allBehaviorsDo: aBlock > "Evaluate the argument, aBlock, for each kind of Behavior in the system > (that is, Object and its subclasses and Traits). > ar 7/15/1999: The code below will not enumerate any obsolete or anonymous > behaviors for which the following should be executed: > > Smalltalk allObjectsDo:[:obj| obj isBehavior ifTrue:[aBlock value: obj]]. > > but what follows is way faster than enumerating all objects." > > self environment allClassesAndTraitsDo: [ :class | > + aBlock value: class. > + class isTrait ifFalse: [ "class of a Trait is Trait, there are no MetaTraits" > + aBlock value: class class ] ]! > - aBlock > - value: class; > - value: class class ]! Thanks, Levente. I don't understand why that fixes the problem though. Er, unless it's that when you walk over the TFoos, you also walk (and walk and walk again) over Trait? frank

2 1

Re: [squeak-dev] The Trunk: Kernel-nice.818.mcz
by Frank Shearar 02 Nov '13

02 Nov '13

On 1 November 2013 23:10, <commits(a)source.squeak.org> wrote: > Nicolas Cellier uploaded a new version of Kernel to project The Trunk: > http://source.squeak.org/trunk/Kernel-nice.818.mcz > > ==================== Summary ==================== > > Name: Kernel-nice.818 > Author: nice > Time: 2 November 2013, 12:05:29.34 am > UUID: 3dae8edc-b873-491e-8b54-596cae38eb5f > Ancestors: Kernel-cmm.817 > > Avoid spurious warnings in Class>>rename: when an Undeclared isn't really referenced. > > This may help some SUnit tests to pass. > > =============== Diff against Kernel-cmm.817 =============== This is exactly the same as Trait >> #rename:. Is there a way to remove the duplication? What would be a good reason not to push this up to ClassDescription? (Maybe that this would imply that you can #rename: a Metaclass?) frank

3 5

The Trunk: System-ul.613.mcz
by commits＠source.squeak.org 02 Nov '13

02 Nov '13

Levente Uzonyi uploaded a new version of System to project The Trunk: http://source.squeak.org/trunk/System-ul.613.mcz ==================== Summary ==================== Name: System-ul.613 Author: ul Time: 2 November 2013, 1:47:09.671 pm UUID: de29e18c-22f5-4299-823a-e6b207c3f552 Ancestors: System-cmm.612 - fix: SystemNavigation >> #allBehaviorsDo: shouldn't enumerate Trait more than once. =============== Diff against System-cmm.612 =============== Item was changed: ----- Method: SystemNavigation>>allBehaviorsDo: (in category 'query') ----- allBehaviorsDo: aBlock "Evaluate the argument, aBlock, for each kind of Behavior in the system (that is, Object and its subclasses and Traits). ar 7/15/1999: The code below will not enumerate any obsolete or anonymous behaviors for which the following should be executed: Smalltalk allObjectsDo:[:obj| obj isBehavior ifTrue:[aBlock value: obj]]. but what follows is way faster than enumerating all objects." self environment allClassesAndTraitsDo: [ :class | + aBlock value: class. + class isTrait ifFalse: [ "class of a Trait is Trait, there are no MetaTraits" + aBlock value: class class ] ]! - aBlock - value: class; - value: class class ]!

1 0

Re: [squeak-dev] The Trunk: Traits-nice.300.mcz
by Frank Shearar 02 Nov '13

02 Nov '13

On 1 November 2013 23:31, <commits(a)source.squeak.org> wrote: > Nicolas Cellier uploaded a new version of Traits to project The Trunk: > http://source.squeak.org/trunk/Traits-nice.300.mcz > > ==================== Summary ==================== > > Name: Traits-nice.300 > Author: nice > Time: 2 November 2013, 12:30:54.426 am > UUID: fced9294-e4ef-4fb3-b2db-6e8da24a6e82 > Ancestors: Traits-nice.299 > > Avoid spurious warnings in Trait>>rename: when an Undeclared isn't really referenced. > > This may help some SUnit tests to pass. > > =============== Diff against Traits-nice.299 =============== > > Item was changed: > ----- Method: Trait>>rename: (in category 'initialize') ----- > rename: aString > "The new name of the receiver is the argument, aString." > > | newName | > (newName := aString asSymbol) ~= self name > ifFalse: [^ self]. > (self environment includesKey: newName) > ifTrue: [^ self error: newName , ' already exists']. > + ((self environment undeclared includesKey: newName) > + and: [(self environment undeclared unreferencedKeys includes: newName) not]) > - (self environment undeclared includesKey: newName) > ifTrue: [self inform: 'There are references to, ' , aString printString , ' > from Undeclared. Check them after this change.']. > self environment renameClass: self as: newName. > name := newName! There's a lot of text in this method. I think it would read more easily if we added an "undeclared := self environment undeclared" var. The + lines above would then be something like ((undeclared includesKey: newName) and: [(undeclared unreferencedKeys includes: newName) not]) ifTrue: [] frank

2 1

Jump to page: