- Pharo-dev - lists.pharo.org

[Pharo-project] [ANN] 10093
by StÃ©phane Ducasse Oct. 12, 2008

Oct. 12, 2008

- Network fix by Yann Monclair Modified the deprecation comment on Url>>#toText. It was recommending using Url>>#asText instead, but that would mean the method would return a Text instead of a String as it currently does. I replaced the deprecation comment to recommend replacing #toText with #asString. All NetworkTests run green - Fix by A. Bergel. The method PasteUpMorph>>findAPreferencesPanel: is now an extension of MorphicExtras. Before it was part of Mophic- Worlds. The reason to move findAPreferencesPanel: to MorphicExtras is the method references PreferencesPanel, which is defined in MorphicExtras. PasteUpMorph calls findAPreferencesPanel: within PasteUpMorph>>defaultDesktopCommandKeyTriplets. The cycle is therefore not completely removed. However, defaultDesktopCommandKeyTriplets should be turned into a registration mechanism to completely remove the cycle. - allUnSentMessages -> allUnsentMessages, removeAllUnSentMessages -> removeAllUnsentMessages, allUnSentMessagesIn: -> allUnsentMessagesIn: Stopping for this evening :) Stef

1 0

[Pharo-project] please do not use PopUpMenu but
by StÃ©phane Ducasse Oct. 12, 2008

Oct. 12, 2008

but UIManager Tx

1 0

[Pharo-project] Pharo Harvesting Infrastructure
by Matthew Fulmer Oct. 12, 2008

Oct. 12, 2008

Hi. I found a bug in Pharo, and I started wondering how to determine where it came from. I checked a few methods in the debugger to see if they had recently been modified, but that was no help. I checked the change sorter to see if it had any idea what had been recently installed, but it was choked by Monticello updates. Then I opened a new Pharo Image in order to try each update one by one (whatever Monticello you guys use to load your changes is *really* slow. I gotta fix you guys up with MC1.6). But before I did that, I noticed on the right of the screen is something that looks like a change log. And at the bottom, it said: (full list of changes: print "ScriptLoader new logContents") So I tried it in my 10091 image, and sure enough, it printed up a changelog. So I went to see what this mysterious class ScriptLoader is. I found a whole lot of methods. Some looked auto-generated. Some seemed to be convenience methods for loading MC packages from Pharo and PharoInbox. It looks kind of like Installer, which I am more familiar with. I'm rather curious now about how you manage Pharo. I took a look at http://code.google.com/p/pharo/wiki/HowToContribute and wondered what happens after that. I now imagine it involves ScriptBuilder building the update, and then something uploading it. What is the mechanism? How is Pharo tested? is it regularly tested by someone using TestRunner, or something fancier? What are all the auto-generated methods in ScriptLoader for? How do you typically trace a bug from image to problem package to update to issue? I'm interested because keith and I are building a build server that is able to build images, build packages, load packages, run tests, build VM's and such, all automated and with an easy interface. So far, we have the easy interface (Installer), but are still working on making it do more. So I'm interested in learning how it works here, so that we can at least end up with something as good as this -- Matthew Fulmer -- http://mtfulmer.wordpress.com/

3 6

[Pharo-project] I'm doing a small release to clean a bit the inbox
by StÃ©phane Ducasse Oct. 12, 2008

Oct. 12, 2008

1 0

[Pharo-project] Pharo image size: 10.9MB
by Adrian Lienhard Oct. 12, 2008

Oct. 12, 2008

I've been looking into how the current 13.4MB a Pharo image uses are spent... I expected that at least half of it is used by classes (compiled methods). But this is not the case. Actually, only about 1/4th are classes. As it turned out, almost the same amount of memory is used by forms (actually bitmaps)! If you want to know which are the largest 10 forms try this: (Form allInstances asSortedCollection: [ :e1 :e2 | e1 bits size >= e2 bits size ]) from: 1 to: 10 do: [ :each | each asMorph openInWorld ] ...among which for example are 4 different color pickers. In total there are around 450 forms, most from EToys. I've manually traced down where the larger forms come from and have cleaned up those that were possible (many are only referenced by caches). The number of forms is at 320 now. In the process I produced the script below, which among other stuff removes all but the default font. Also, it actually removes the flaps from the system instead of only hiding them. Here a history of image sizes: Squeak 3.9: 15.6 MB Pharo 10091: 13.4MB (before cleanup) Pharo 10091: 10.9MB (after cleanup) Main conclusion: the system needs cleanup not only of the code (where the mess is more directly visible), but also of the object memory. Furthermore, I expect removing EToys will have an nice impact on the image size. Also removing Sound will release some memory because of sound samples. If anybody has suggestions to further reduce the image size (<10MB would be a nice target for the first release...), please let me know. Cheers, Adrian Preferences setDefaultFonts: #( (setSystemFontTo: 'Accuny' 10) (setListFontTo: 'Accuny' 10) (setMenuFontTo: 'Accuny' 10) (setFlapsFontTo: 'Accuny' 12) (setEToysFontTo: 'Accuny' 10) (setPaintBoxButtonFontTo: 'Accuny' 10) (setWindowTitleFontTo: 'Accuny' 15) (setBalloonHelpFontTo: 'Accuny' 10) (setCodeFontTo: 'Accuny' 10) (setButtonFontTo: 'Accuny' 8)). TextConstants associations do: [ :association | ((association value isKindOf: TextStyle) and: [ association key ~= #Accuny ] and: [ (TextStyle defaultFamilyNames includes: association key) not ]) ifTrue: [ TextConstants removeKey: association key ]]. TextConstants at: #DefaultFixedTextStyle put: (TextConstants at: #Accuny). (Flaps classPool at: #SharedFlapTabs) removeAllSuchThat: [ :each | (#(Tools Pharo) includes: each labelString) not ]. NaturalLanguageFormTranslator classPool at: #CachedTranslations put: nil. NaturalLanguageTranslator resetCaches. PartsBin clearThumbnailCache. PaintBoxMorph initialize. PaintBoxMorph classPool at: #ColorChart put: nil; at: #Prototype put: nil. Utilities classPool at: #ScrapsBook put: nil. Project allInstancesDo: [ :each | each setThumbnail: nil ]. InternalThreadNavigationMorph clearThumbnailCache. ScriptingSystem stripGraphicsForExternalRelease. ProjectHistory currentHistory initialize. Utilities zapUpdateDownloader. Smalltalk forgetDoIts. Behavior flushObsoleteSubclasses. CommandHistory allInstancesDo: #initialize. ScriptLoader new cleaningCS. RequiredSelectors initialize. ProvidedSelectors initialize. LocalSends initialize. SendCaches initializeAllInstances. Utilities cleanseOtherworldlySteppers. Smalltalk organization removeEmptyCategories. SystemBrowser removeObsolete. TheWorldMenu removeObsolete. AppRegistry removeObsolete. FileServices removeObsolete. SystemNavigation default obsoleteBehaviors. MCFileBasedRepository flushAllCaches. MCDefinition clearInstances. Undeclared removeUnreferencedKeys. Smalltalk flushClassNameCache. 3 timesRepeat: [ Smalltalk garbageCollect. Symbol compactSymbolTable ]. Set rehashAllSets. Smalltalk garbageCollect. ___________________ http://www.adrian-lienhard.ch/

4 10

[Pharo-project] SLICE-Mantis-KernelNumber-Fix-Mining-nice.1.mcz
by nicolas cellier Oct. 11, 2008

Oct. 11, 2008

I push here some more or less important fixes. These are limited to one or two method per fix and should not take very long to review. Cheers Nicolas This gathers several Tests and Fixes from Mantis for Kernel-Number ------------------------------------------------------------ http://bugs.squeak.org/view.php?id=6781 0.0 raisedTo: -1.0 unexpected result The patch correct Number >> #raisedTo: and uses [^self class one] so that Float >> #raisedTo; can be removed (99% identical code). ------------------------------------------------------------ http://bugs.squeak.org/view.php?id=6729 -Infinity closeTo: Infinity self deny: (Float infinity negated closeTo: Float infinity) ------------------------------------------------------------ http://bugs.squeak.org/view.php?id=6695 Float storeStringBase: does not store the radix self assert: (20.0 storeStringBase: 16) = '16r14.0' compare to this: (20 storeStringBase: 16) Note: in base 10, radix is not prepended (same as Integer). ------------------------------------------------------------ http://bugs.squeak.org/view.php?id=6697 Number>>storeOn:base: default implementation is wrong Default implementation does omit the radix prefix when it should not! Change it to subclassResponsibility, since every subclass will have a different one... ------------------------------------------------------------ http://bugs.squeak.org/view.php?id=6990 0.0 significandAsInteger is too big ------------------------------------------------------------ http://bugs.squeak.org/view.php?id=7114 SmallInteger minVal highBitOfMagnitude Try it, it should not raise an Error... ------------------------------------------------------------ http://bugs.squeak.org/view.php?id=5640 additional tests and fixes for Number>>printShowingDecimalPlaces: http://bugs.squeak.org/view.php?id=7028 printShowingDecimalPlaces: is broken in 3.10 Try (-0.9 printShowingDecimalPlaces: 1) for example... Where is the minus sign? Fix included in 3.10 is incorrect and not complete ------------------------------------------------------------

1 0

[Pharo-project] About trait mop
by StÃ©phane Ducasse Oct. 11, 2008

Oct. 11, 2008

Hi I have been working on a new API to access class, traits, selector for traits I wrote some tests in the PharoTaskForces/Traits-MOPTests package Now in essence we have - The selector of a compiled method should be its name. An aliased method should have the name of its alias name. - The class of a compiled method is the class that contains it. A compiled method cannot be shared. - The origin of a compiledMethod is its defining class or trait. - The reachHome of a compiled method is class or traits that you changed (and not override) when you want to change a method in the flat view. T1 c c1 T2 c c3 uses: T1 Class A foo uses: T2 testSelectorT11T2 "self debug: #testSelectorT11T2" "The selector of a compiled method should be its name. An aliased method should have the name of its alias name." self assert: (AUsingTrait2>>#foo) selector = #foo. self assert: (Trait1>>#c) selector = #c. self assert: (Trait2>>#c) selector = #c. self assert: (AUsingTrait2>>#c) selector = #c. self assert: (AUsingTrait2>>#c3) selector = #c2. self assert: (BUsingTrait1>>#c) selector = #c. testClassT11T2 "The class of a compiled method is the class that contains it. A compiled method cannot be shared." self assert: (AUsingTrait2>>#foo) methodClass = AUsingTrait2. self assert: (AUsingTrait2>>#c) methodClass = AUsingTrait2. self assert: (AUsingTrait1>>#c3) methodClass = AUsingTrait2. self assert: (AUsingTrait2>>#c1) methodClass = AUsingTrait2. self assert: (BUsingTrait1>>#c) methodClass = BUsingTrait1. self should: (Trait1>>#c1) methodClass = nil. "T1 or nil" self should: (Trait2>>#c2) methodClass = nil. "T2 or nil" self should: (Trait2>>#c1) methodClass = nil. "T2 or nil or T1" testReachHome "self debug: #testReachHome" " The reachHome is the method that you changed (and not override) when you want to change a method in the flat view." self assert: (AUsingTrait2>>#foo) reachHome = AUsingTrait2. self assert: (AUsingTrait2>>#c3) reachHome = Trait2. self assert: (AUsingTrait2>>#c1) reachHome = Trait1. self assert: (AUsingTrait2>>#c) reachHome = Trait2. self assert: (Trait11>>#c) reachHome = Trait11. self assert: (Trait2>>#c) reachHome = Trait2. self assert: (Trait2>>#c1) reachHome = Trait11. self assert: (AUsingTrait1>>#c2) reachHome = AUsingTrait1. testOrigin "self debug: #testClass" "The origin of a compiledMethod is its defining class or trait." self assert: (AUsingTrait2>>#foo) origin = AUsingTrait2. self assert: (AUsingTrait2>>#c3) origin = Trait2. self assert: (AUsingTrait2>>#c1) origin = Trait11. self assert: (AUsingTrait2>>#c) origin = Trait11. self assert: (BUsingTrait1>>#c) origin = Trait1. self assert: (Trait2>>#c3) origin = Trait2. self assert: (Trait11>>#c1) origin = Trait11. self assert: (Trait2>>#c1) origin = Trait11. "since it introduced the alias it is its origin" self assert: (AUsingTrait1>>#c2) origin = AUsingTrait1. testTraitT11T2 "it seems to me that we do not need trait: but we need origin and reachHome" "(A>>foo) trait -> nil (A>>c3) trait -> T2 (A>>c) trait -> T2 (A>>c1) trait -> T2 ***** since it is flat (T1>>c1) trait -> ???? T1 or nil (T2>>c1) trait (T2>>c2) trait -> ???? T2 or nil" self assert: (AUsingTrait2>>#foo) trait isNil. self assert: (AUsingTrait2>>#c3) trait = Trait2

4 7

Oct. 11, 2008

Apparently #2 value is a number? but the parser parses it. Stef

3 9

[Pharo-project] Monticello 1.6 or 2.0?
by Janko MivÅ¡ek Oct. 10, 2008

Oct. 10, 2008

Hello, We have now three Monticellos: 1, 1.6 and 2.0. What is the Pharo plans here, to go to 1.6 then to 2.0? How to migrate old repositories like SqueakSource? When to migrate? How to persuade all Squeakers to upgrade MC in their images? That questions which are raising in my mind for a while... Janko -- Janko MivÅ¡ek AIDA/Web Smalltalk Web Application Server http://www.aidaweb.si

2 1

[Pharo-project] Installer refactor
by Matthew Fulmer Oct. 10, 2008

Oct. 10, 2008

I spent most of the day yesterday refactoring Installer. I split it up into 14 classes, so now it should be a bit easier to understand. I'll fix badly categorized methods and add some comments. -- Matthew Fulmer -- http://mtfulmer.wordpress.com/

3 7