Squeak-dev June 2010

squeak-dev@lists.squeakfoundation.org

96 participants
269 discussions

Re: WebClient-Monticello
by Chris Cunnington 23 Jun '10

23 Jun '10

I like this a lot and will be combing over it for a while. I did find in your instructions that: directory: (FileDirectory on: '/tmp/monticello') presented some problems for me. Putting a folder called "monticello" in the Squeak folder, ensuring anybody could write/read to it, and then using: directory: (FileDirectory default directoryNamed: 'monticello') worked as expected. I don't think MCWebServer could find the directory due to an incomplete path. I have a feeling that you expected the relative path to be assisted by the fuller path that gets put into the directory instvar during initialization. But I don't think the example instructions you've provided do that. They clobber whatever's in the directory instvar and leave it with only a relative path. That seems to be why I was getting errors. Thanks for sharing this. Chris

2 1

[Talk] Gilad Bracha: The Brave New World of Full Service Computing (HPI, 2010-06-17, 11:30-12:00)
by Robert Hirschfeld 23 Jun '10

23 Jun '10

Hi -- Gilad Bracha is going to give a talk on Newspeak and "The Brave New World of Full Service Computing" this Thursday at HPI's Symposium on Future Trends in Service-Oriented Computing 2010: http://kolleg.hpi.uni-potsdam.de/school/events/future_trends_in_soc_2010/ag… Best, Robert -- Robert Hirschfeld hirschfeld(a)acm.org www.hirschfeld.org

2 2

The Trunk: Tools-jcg.247.mcz
by commits＠source.squeak.org 23 Jun '10

23 Jun '10

Joshua Gargus uploaded a new version of Tools to project The Trunk: http://source.squeak.org/trunk/Tools-jcg.247.mcz ==================== Summary ==================== Name: Tools-jcg.247 Author: jcg Time: 23 June 2010, 12:13:50.59 pm UUID: a06587b9-7aaf-412f-a9a1-c49d97fb8110 Ancestors: Tools-cmm.246 Workspaces can now look up variable-bindings in a user-specified list of SharedPools. See the class-side methods #addLookupPool:, #removeLookupPool:, and #clearLookupPools. =============== Diff against Tools-cmm.246 =============== Item was added: + ----- Method: Workspace class>>removeLookupPool: (in category 'lookup pools') ----- + removeLookupPool: sharedPool + "Workspaces should no longer look up bindings in the specified SharedPool." + LookupPools ifNil: [^self]. + LookupPools remove: sharedPool ifAbsent: [].! Item was added: + ----- Method: Workspace>>lookupInPools: (in category 'binding') ----- + lookupInPools: aString + "Iterate through all of the lookup pools to find a match for the specified name." + LookupPools ifNil: [^nil]. + LookupPools do: [:pool | (pool bindingOf: aString) ifNotNil: [:assoc | ^assoc value]]. + ^nil + ! Item was added: + ----- Method: Workspace class>>clearLookupPools (in category 'lookup pools') ----- + clearLookupPools + LookupPools := nil. + ! Item was changed: ----- Method: Workspace>>bindingOf: (in category 'binding') ----- bindingOf: aString - mustDeclareVariables ifTrue: [^ nil]. "I want to have workspace that force the user to declare variables. Still subclasses may want to do something else" + bindings ifNil: [self initializeBindings]. + (bindings includesKey: aString) ifFalse: [ + | val | + val := self lookupInPools: aString. + val + ifNotNil: [ + "Caveat: since we're not inserting the pool's Association into + the dictionary, assigning a new value to the variable will only + affect the one Workspace (i.e. you can't assign to a variable in a + SharedPool)." + bindings at: aString put: val] + ifNil: [aString first isUppercase + ifTrue: [^nil] + ifFalse: [bindings at: aString put: nil]] + ]. - bindings isNil - ifTrue: [self initializeBindings]. - (bindings includesKey: aString) - ifFalse: [bindings at: aString put: nil]. ^ bindings associationAt: aString! Item was added: + ----- Method: Workspace class>>addLookupPool: (in category 'lookup pools') ----- + addLookupPool: sharedPool + "Add a SharedPool that bindings should be looked up in." + LookupPools ifNil: [LookupPools := IdentitySet new]. + LookupPools add: sharedPool! Item was changed: StringHolder subclass: #Workspace instanceVariableNames: 'bindings acceptDroppedMorphs acceptAction mustDeclareVariables shouldStyle' + classVariableNames: 'LookupPools ShouldStyle' - classVariableNames: 'ShouldStyle' poolDictionaries: '' category: 'Tools-Base'! !Workspace commentStamp: 'ls 10/14/2003 12:13' prior: 0! A Workspace is a text area plus a lot of support for executable code. It is a great place to execute top-level commands to compute something useful, and it is a great place to develop bits of a program before those bits get put into class methods. To open a new workspace, execute: Workspace open A workspace can have its own variables, called "workspace variables", to hold intermediate results. For example, if you type into a workspace "x := 5" and do-it, then later you could type in "y := x * 2" and y would become 10. Additionally, in Morphic, a workspace can gain access to morphs that are on the screen. If acceptDroppedMorphss is turned on, then whenever a morph is dropped on the workspace, a variable will be created which references that morph. This functionality is toggled with the window-wide menu of a workspace. The instance variables of this class are: bindings - holds the workspace variables for this workspace acceptDroppedMorphss - whether dropped morphs should create new variables!

1 0

The Trunk: Compiler-jcg.147.mcz
by commits＠source.squeak.org 23 Jun '10

23 Jun '10

Joshua Gargus uploaded a new version of Compiler to project The Trunk: http://source.squeak.org/trunk/Compiler-jcg.147.mcz ==================== Summary ==================== Name: Compiler-jcg.147 Author: jcg Time: 23 June 2010, 12:04:54.856 pm UUID: 714d775f-78af-45dc-affc-b33d59d68da5 Ancestors: Compiler-nice.146 Change Parser>>correctVariable:interval: to allow it to look up the proposed variable in the requestor even if it starts with an uppercase letter. =============== Diff against Compiler-nice.146 =============== Item was changed: ----- Method: Parser>>correctVariable:interval: (in category 'error correction') ----- correctVariable: proposedVariable interval: spot "Correct the proposedVariable to a known variable, or declare it as a new variable if such action is requested. We support declaring lowercase variables as temps or inst-vars, and uppercase variables as Globals or ClassVars, depending on whether the context is nil (class=UndefinedObject). Spot is the interval within the test stream of the variable. rr 3/4/2004 10:26 : adds the option to define a new class. " "Check if this is an i-var, that has been corrected already (ugly)" "Display the pop-up menu" + | binding userSelection action | - | tempIvar binding userSelection action | (encoder classEncoding instVarNames includes: proposedVariable) ifTrue: [^InstanceVariableNode new name: proposedVariable index: (encoder classEncoding allInstVarNames indexOf: proposedVariable)]. "If we can't ask the user for correction, make it undeclared" self interactive ifFalse: [^encoder undeclared: proposedVariable]. "First check to see if the requestor knows anything about the variable" + (binding := requestor bindingOf: proposedVariable) + ifNotNil: [^encoder global: binding name: proposedVariable]. - tempIvar := proposedVariable first isLowercase. - (tempIvar and: [(binding := requestor bindingOf: proposedVariable) notNil]) - ifTrue: [^encoder global: binding name: proposedVariable]. userSelection := requestor selectionInterval. requestor selectFrom: spot first to: spot last. requestor select. "Build the menu with alternatives" action := UndeclaredVariable signalFor: self name: proposedVariable inRange: spot. action ifNil: [^self fail]. "Execute the selected action" requestor deselect. requestor selectInvisiblyFrom: userSelection first to: userSelection last. ^action value!

1 0

intersect: when intsersects: is false
by Jim Rosenberg 23 Jun '10

23 Jun '10

[Using Squeak 3.8; apologies if more current versions "fix" this ... ] This expression yields false: ((0@0) corner: (50@50)) intersects: ((100@100) corner: (150@150)) So, what should this yield?: ((0@0) corner: (50@50)) intersect: ((100@100) corner: (150@150)) I was expecting perhaps nil. Instead I got (100@100) corner: (50@50). This is "a surprise". It's certainly not right mathematically ... --- Jim Rosenberg http://www.well.com/user/jer/ Internet: jr(a)amanue.com

8 16

problem building Unix Cog VM
by Chris Muller 23 Jun '10

23 Jun '10

I don't know why, but I can't seem to build the cog vm when I RE-unpack OpenSourceCog.tar.gz. The directory I unpacked it to earlier today still builds a VM just fine. I've renamed that one to .old for now, but I'm wanting to get this VM into our standard build process and that involves unpacking the tarball and building fresh via a script called by make. But when I unpack OpenSourceCog.tar.gz and build it exacly the same, it _appears_ that all builds fine, although it does seem perhaps a bit longer. When I then try to start an image, it says it can't find the X11 plugin at: /my/specified/prefix/lib/squeak/3.9-7//vm-display-X11 Notice the double-slash; that's where it says it's looking for it, which is obviously incorrect. I had specified my prefix without a / suffix, which I assume is correct. make install prefix=/my/specified/prefix I've been pulling my hair out because the one out of my cog.old directory builds a perfectly-working VM using these exact same steps, which I thought were merely unpacked from OpenSourceCog.tar.gz.. Any ideas?

1 0

Magma on Cog - initial test
by Chris Muller 23 Jun '10

23 Jun '10

Well, I still need to do more testing, but so far Cog looks to be Christmas in June! The Magma tests suite passed, and in 20 minutes (vs. 37), I'll add. I've also updated the Magma benchmark page with just an initial run of the local benchmarks. http://wiki.squeak.org/squeak/5606 The old numbers are under "Local" and the new numbers are under "Local with Cog" (open two browsers side-by-side to easily compare the #'s). In summary: - Read performance is increased roughly 4X across the board - Write performance with Cog, but without WriteBarrier is even faster than the old VM with WriteBarrier. - Bulk-load test seems to be about 2-3X faster. Maui and other morphic widgets are blazing fast. Dragging and animation seem to be much smoother. It's wonderful! - Chris

1 0

Daily Commit Log
by commits＠source.squeak.org 23 Jun '10

23 Jun '10

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2010-June/003480.html Name: System-ar.342 Ancestors: System-cmm.341 Add a test to identify whether the image is running under Cog or not. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-June/003481.html Name: Tests-ar.83 Ancestors: Tests-HenrikSperreJohansen.82 Guard TestObjectsAsMethods against crashes on Cog. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-June/003482.html Name: System-ar.343 Ancestors: System-ar.342 Fixes an ancient issue when launching Squeak with a script as argument. The code the converted the script name to a file url was just utterly and horribly broken and could not possibly work. With the fix, one can now again start Squeak with arguments of the form: squeak mysqueak.image file.st squeak mysqueak.image scripts/file.st squeak mysqueak.image ../../svn/scripts/file.st (and of course any kind of URL) ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-June/003483.html Name: System-tbn.344 Ancestors: System-ar.343 recategorize #isRunningCog under vm ============================================= http://lists.squeakfoundation.org/pipermail/packages/2010-June/003484.html Name: Multilingual-ul.121 Ancestors: Multilingual-nice.120 - use #ifNil: when possible - simplified UTF8TextConverter >> #nextFromStream:empty log message =============================================

1 0

Cog and Batteries?
by Casey Ransberger 23 Jun '10

23 Jun '10

I have a perhaps naive question: what's the expected impact of the JIT on battery life? Seems like the extra number crunching might have an impact. Regardless of whether or not people want us generating native code on mobile devices, does it generally make more sense to use the stack VM on mobile devices? FWIW, I think 10-15% better performance is still a f@(&!?g big deal on dog slow devices! Oh, this stuff just makes my whole year fantastic. Of course, the question may be a bit premature since we don't have a JIT for ARM yet:)

3 2

The Trunk: Multilingual-ul.121.mcz
by commits＠source.squeak.org 22 Jun '10

22 Jun '10

Levente Uzonyi uploaded a new version of Multilingual to project The Trunk: http://source.squeak.org/trunk/Multilingual-ul.121.mcz ==================== Summary ==================== Name: Multilingual-ul.121 Author: ul Time: 22 June 2010, 6:48:58.111 pm UUID: 1293f287-5ed7-354c-9df8-d36e42ef16c0 Ancestors: Multilingual-nice.120 - use #ifNil: when possible - simplified UTF8TextConverter >> #nextFromStream:empty log message =============== Diff against Multilingual-nice.120 =============== Item was changed: ----- Method: UTF8TextConverter>>nextFromStream: (in category 'conversion') ----- nextFromStream: aStream + | character value1 value2 unicode value3 value4 | + aStream isBinary ifTrue: [ ^aStream basicNext ]. + character := aStream basicNext ifNil: [ ^nil ]. + value1 := character asciiValue. + value1 <= 127 ifTrue: [ ^character ]. "1-byte character" - | character1 value1 character2 value2 unicode character3 value3 character4 value4 | - aStream isBinary ifTrue: [^ aStream basicNext]. - character1 := aStream basicNext. - character1 isNil ifTrue: [^ nil]. - value1 := character1 asciiValue. - value1 <= 127 ifTrue: [ - "1-byte character" - ^ character1 - ]. - "at least 2-byte character" + character := aStream basicNext ifNil: [ ^self errorMalformedInput ]. + value2 := character asciiValue. - character2 := aStream basicNext. - character2 = nil ifTrue: [^self errorMalformedInput]. - value2 := character2 asciiValue. - (value1 bitAnd: 16rE0) = 192 ifTrue: [ + ^ Unicode value: ((value1 bitAnd: 31) bitShift: 6) + (value2 bitAnd: 63) ]. - ^ Unicode value: ((value1 bitAnd: 31) bitShift: 6) + (value2 bitAnd: 63). - ]. - "at least 3-byte character" + character := aStream basicNext ifNil: [ ^self errorMalformedInput ]. + value3 := character asciiValue. - character3 := aStream basicNext. - character3 = nil ifTrue: [^self errorMalformedInput]. - value3 := character3 asciiValue. (value1 bitAnd: 16rF0) = 224 ifTrue: [ unicode := ((value1 bitAnd: 15) bitShift: 12) + ((value2 bitAnd: 63) bitShift: 6) + + (value3 bitAnd: 63) ]. - + (value3 bitAnd: 63). - ]. - (value1 bitAnd: 16rF8) = 240 ifTrue: [ "4-byte character" + character := aStream basicNext ifNil: [ ^self errorMalformedInput ]. + value4 := character asciiValue. - character4 := aStream basicNext. - character4 = nil ifTrue: [^self errorMalformedInput]. - value4 := character4 asciiValue. unicode := ((value1 bitAnd: 16r7) bitShift: 18) + ((value2 bitAnd: 63) bitShift: 12) + ((value3 bitAnd: 63) bitShift: 6) + + (value4 bitAnd: 63) ]. + unicode ifNil: [ ^self errorMalformedInput ]. + unicode > 16r10FFFD ifTrue: [ ^self errorMalformedInput ]. + unicode = 16rFEFF ifTrue: [ ^self nextFromStream: aStream ]. - (value4 bitAnd: 63). - ]. - - unicode isNil ifTrue: [^self errorMalformedInput]. - unicode > 16r10FFFD ifTrue: [^self errorMalformedInput]. - - unicode = 16rFEFF ifTrue: [^ self nextFromStream: aStream]. ^ Unicode value: unicode. ! Item was changed: ----- Method: EUCTextConverter>>nextFromStream: (in category 'conversion') ----- nextFromStream: aStream | character1 character2 offset value1 value2 nonUnicodeChar | aStream isBinary ifTrue: [^ aStream basicNext]. + character1 := aStream basicNext ifNil: [^ nil]. - character1 := aStream basicNext. - character1 isNil ifTrue: [^ nil]. character1 asciiValue <= 127 ifTrue: [^ character1]. + character2 := aStream basicNext ifNil: [^ nil]. - character2 := aStream basicNext. - character2 = nil ifTrue: [^ nil]. offset := 16rA1. value1 := character1 asciiValue - offset. value2 := character2 asciiValue - offset. (value1 < 0 or: [value1 > 93]) ifTrue: [^ nil]. (value2 < 0 or: [value2 > 93]) ifTrue: [^ nil]. nonUnicodeChar := Character leadingChar: self leadingChar code: value1 * 94 + value2. ^ Character leadingChar: self languageEnvironment leadingChar code: nonUnicodeChar asUnicode. ! Item was changed: ----- Method: MultiByteFileStream>>peekFor: (in category 'public') ----- peekFor: item | next state | "self atEnd ifTrue: [^ false]. -- SFStream will give nil" state := converter saveStateOf: self. + (next := self next) ifNil: [^ false]. - (next := self next) == nil ifTrue: [^ false]. item = next ifTrue: [^ true]. converter restoreStateOf: self with: state. ^ false. ! Item was changed: ----- Method: MultiByteBinaryOrTextStream>>peekFor: (in category 'public') ----- peekFor: item | next state | "self atEnd ifTrue: [^ false]. -- SFStream will give nil" state := converter saveStateOf: self. + (next := self next) ifNil: [^ false]. - (next := self next) == nil ifTrue: [^ false]. item = next ifTrue: [^ true]. converter restoreStateOf: self with: state. ^ false. !

1 0

← Newer
1
...
5
6
7
8
9
10
11
...
27
Older →

Jump to page: