Squeak-dev September 2009

squeak-dev@lists.squeakfoundation.org

119 participants
370 discussions

The Trunk: SMBase-nice.88.mcz
by commits＠source.squeak.org 05 Sep '09

05 Sep '09

Nicolas Cellier uploaded a new version of SMBase to project The Trunk: http://source.squeak.org/trunk/SMBase-nice.88.mcz ==================== Summary ==================== Name: SMBase-nice.88 Author: nice Time: 5 September 2009, 10:22:37 am UUID: 92ba52f7-eb3b-4f70-aeae-603ac1d0d64c Ancestors: SMBase-edc.87 Avoid invoking super when self would do the job. ( super basicNew issue http://bugs.squeak.org/view.php?id=6977 ) =============== Diff against SMBase-edc.87 =============== Item was changed: ----- Method: SMObject class>>newIn: (in category 'instance creation') ----- newIn: aMap "Create a new object in a given map with an UUID to ensure unique identity." + ^(self basicNew) map: aMap id: UUID new! - ^(super basicNew) map: aMap id: UUID new! Item was changed: Object subclass: #SMUtilities instanceVariableNames: '' + classVariableNames: 'MailServer MasterServer' - classVariableNames: 'MasterServer MailServer' poolDictionaries: '' category: 'SMBase-UI'! !SMUtilities commentStamp: 'gk 11/13/2003 23:39' prior: 0! Various loose functions in SM.!

1 0

The Trunk: Collections-nice.126.mcz
by commits＠source.squeak.org 05 Sep '09

05 Sep '09

Nicolas Cellier uploaded a new version of Collections to project The Trunk: http://source.squeak.org/trunk/Collections-nice.126.mcz ==================== Summary ==================== Name: Collections-nice.126 Author: nice Time: 5 September 2009, 10:21:20 am UUID: 32bbec9c-28ef-4c94-bd03-b84942671351 Ancestors: Collections-ar.125 Avoid invoking super when self would do the job. ( super basicNew issue http://bugs.squeak.org/view.php?id=6977 ) =============== Diff against Collections-ar.125 =============== Item was changed: ----- Method: OrderedCollection class>>new: (in category 'instance creation') ----- new: anInteger + ^ self basicNew setCollection: (Array new: anInteger)! - ^ super basicNew setCollection: (Array new: anInteger)! Item was changed: ----- Method: OrderedCollection class>>new:withAll: (in category 'instance creation') ----- new: anInteger withAll: anObject + ^ self basicNew setContents: (Array new: anInteger withAll: anObject)! - ^ super basicNew setContents: (Array new: anInteger withAll: anObject)! Item was changed: ----- Method: Set class>>new: (in category 'instance creation') ----- new: nElements "Create a Set large enough to hold nElements without growing" + ^ self basicNew initialize: (self sizeFor: nElements)! - ^ super basicNew initialize: (self sizeFor: nElements)!

1 0

iPhone squeak VM goes to latest Pharo image
by John M McIntosh 05 Sep '09

05 Sep '09

On Friday after ESUG I migrated the iPhone VM source code to use the Closure VM logic, this is squeak.app V2.x for the iPHone. I also updated the January era Pharo image to the latest image. If you have a copy of the squeak VM that I build for you and would like to upgrade you need to download the Squeak.2.0.0.app.zip from ftp://ftp.smalltalkconsulting.com/experimental/iPhoneAdHoc/ I note if you have made your VM image writtable via the setting's panel we will launch the new VM with the old writable image versus the new one. If you want to launch the new image you must first delete Squeak from your iPhone, beware that would erase any changes, so use MC to preserve that work. After deleting Squeak.app 1.x from your phone please install the new 2.0.0 version. As always ensure you sync and backup your device before deleting applications. I'm heading out for dinner in London, if problems please email. -- = = = ======================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ========================================================================

1 0

Re: [squeak-dev] The Trunk: Kernel-ar.237.mcz
by Alexander Lazarević 05 Sep '09

05 Sep '09

Hi Andreas! I think it all stems from [1]. In short what I think happened was that it was reported that very long delays (>SmallInt) made the image hang, because a primitive would choke on not SmallInt parameters. I and others just wanted to prevent this by just not allowing very long delays and to have some LongDelay for very special use cases where very long delays would be required. I think this was harvested and made it into the image. There was also the idea of fixing the delay code to have no artificial limit on delays and a change for timeoutSemaphore:after: was proposed that juggled Delays to make long Delays possible. This also went into the image some time, BUT the code to not allow long delays was not removed. So I guess this code never had an effect. Now it was you, if my memory serves me well, that introduced handleTimerEvent et.al. which just naturally made long delays possible. This also went into the image, BUT also having the other fixes left in places. So I guess what was left was a mess of three disjunct fixes for the same problem. I will change the class comment of Delay that warns about too long delays and remove the Test that checks for long delays to be refused and now fails. Alex [1] http://bugs.squeak.org/view.php?id=854 On Fri, Sep 4, 2009 at 6:48 AM, <commits(a)source.squeak.org> wrote: > Andreas Raab uploaded a new version of Kernel to project The Trunk: > http://source.squeak.org/trunk/Kernel-ar.237.mcz > > ==================== Summary ==================== > > Name: Kernel-ar.237 > Author: ar > Time: 3 September 2009, 9:48:09 am > UUID: 0ce8d553-42c8-2d42-add2-fbadccdbfe28 > Ancestors: Kernel-tfel.236 > > http://bugs.squeak.org/view.php?id=7321 > > Change Set: DelayCleanup > Date: 23 March 2009 > Author: Andreas Raab > > Cleans up Delay by removing many of the no longer used methods (timerInterruptWatcher etc). It also provides some fixes for methods that got mysteriously broken even though their original versions were perfectly fine, including Delay class>>timeoutSemaphore:after:. > > =============== Diff against Kernel-tfel.236 =============== > > Item was changed: > + ----- Method: Delay class>>primSignal:atMilliseconds: (in category 'primitives') ----- > - ----- Method: Delay class>>primSignal:atMilliseconds: (in category 'testing') ----- > primSignal: aSemaphore atMilliseconds: aSmallInteger > "Signal the semaphore when the millisecond clock reaches the value of the second argument. Fail if the first argument is neither a Semaphore nor nil. Essential. See Object documentation whatIsAPrimitive." > - > - | guardianDelay | > <primitive: 136> > + ^self primitiveFailed! > - "VM code actually only fails if the time parameter is not a SmallInteger value" > - aSmallInteger isInteger ifFalse:["somebody messed up badly and we can't do much about it" > - aSemaphore ifNotNil: [ > - ActiveDelay := nil. > - aSemaphore signal. "Prevent an image crash"]. > - ^self primitiveError: 'primSignal:atMilliseconds: failed because of a non-Integer resumption time parameter. The Semaphore has been signalled as a best guess of the right thing to do']. > - > - "So now we feel fairly sure that the aSmallInteger resumption time is actually a large integer and we need to just wait some more. To make the system do that we need a fake Delay and a reasonable resumption time to feed to the VM. A decent value is SmallInteger maxVal since the VM handles correlating that sort of largish value and clock wrapping. > - First though we return the problem Delay to the queue" > - SuspendedDelays add: ActiveDelay. > - "Now we want a Delay set to fire and do nothing" > - guardianDelay := self guardianDelay. > - guardianDelay activate > - ! > > Item was changed: > ----- Method: Delay class>>runTimerEventLoop (in category 'timer process') ----- > runTimerEventLoop > "Run the timer event loop." > + [RunTimerEventLoop] whileTrue: [self handleTimerEvent]! > - [ > - [RunTimerEventLoop] whileTrue: [self handleTimerEvent] > - ] on: Error do:[:ex| > - "Clear out the process so it does't get killed" > - TimerEventLoop := nil. > - "Launch the old-style interrupt watcher" > - self startTimerInterruptWatcher. > - "And pass the exception on" > - ex pass. > - ].! > > Item was changed: > ----- Method: Delay class>>restoreResumptionTimes (in category 'snapshotting') ----- > restoreResumptionTimes > "Private!! Restore the resumption times of all scheduled Delays after a snapshot or clock roll-over. This method should be called only while the AccessProtect semaphore is held." > > | newBaseTime | > newBaseTime := Time millisecondClockValue. > SuspendedDelays do: [:d | d adjustResumptionTimeOldBase: 0 newBase: newBaseTime]. > ActiveDelay == nil ifFalse: [ > ActiveDelay adjustResumptionTimeOldBase: 0 newBase: newBaseTime. > + ]. > - ActiveDelay activate]. > ! > > Item was changed: > ----- Method: Delay class>>forMilliseconds: (in category 'instance creation') ----- > + forMilliseconds: anInteger > - forMilliseconds: aNumber > "Return a new Delay for the given number of milliseconds. Sending 'wait' to this Delay will cause the sender's process to be suspended for approximately that length of time." > > + anInteger < 0 ifTrue: [self error: 'delay times cannot be negative']. > + ^ self new > + setDelay: anInteger asInteger > + forSemaphore: Semaphore new > - ^ self new setDelay: aNumber forSemaphore: Semaphore new > ! > > Item was changed: > ----- Method: Delay class>>startUp (in category 'snapshotting') ----- > startUp > "Restart active delay, if any, when resuming a snapshot." > > + DelaySuspended ifFalse:[^self error: 'Trying to activate Delay twice']. > + DelaySuspended := false. > self restoreResumptionTimes. > - ActiveDelay == nil ifFalse: [ActiveDelay activate]. > AccessProtect signal. > ! > > Item was changed: > ----- Method: Delay class>>startTimerEventLoop (in category 'timer process') ----- > startTimerEventLoop > "Start the timer event loop" > "Delay startTimerEventLoop" > self stopTimerEventLoop. > - self stopTimerInterruptWatcher. > AccessProtect := Semaphore forMutualExclusion. > ActiveDelayStartTime := Time millisecondClockValue. > SuspendedDelays := > Heap withAll: (SuspendedDelays ifNil:[#()]) > sortBlock: [:d1 :d2 | d1 resumptionTime <= d2 resumptionTime]. > TimingSemaphore := Semaphore new. > RunTimerEventLoop := true. > TimerEventLoop := [self runTimerEventLoop] newProcess. > TimerEventLoop priority: Processor timingPriority. > TimerEventLoop resume. > TimingSemaphore signal. "get going" > ! > > Item was changed: > ----- Method: Delay class>>timeoutSemaphore:afterMSecs: (in category 'instance creation') ----- > timeoutSemaphore: aSemaphore afterMSecs: anInteger > "Create and schedule a Delay to signal the given semaphore when the given number of milliseconds has elapsed. Return the scheduled Delay. The timeout can be cancelled by sending 'unschedule' to this Delay." > "Details: This mechanism is used to provide a timeout when waiting for an external event, such as arrival of data over a network connection, to signal a semaphore. The timeout ensures that the semaphore will be signalled within a reasonable period of time even if the event fails to occur. Typically, the waiting process cancels the timeout request when awoken, then determines if the awaited event has actually occurred." > > + anInteger < 0 ifTrue: [self error: 'delay times cannot be negative']. > + ^ (self new setDelay: anInteger forSemaphore: aSemaphore) schedule > - ^ (self forMilliseconds: anInteger) schedule > ! > > Item was changed: > ----- Method: Delay class>>initialize (in category 'class initialization') ----- > initialize > + "Delay initialize." > - "Delay initialize" > self startTimerEventLoop.! > > Item was changed: > ----- Method: Delay class>>shutDown (in category 'snapshotting') ----- > shutDown > "Suspend the active delay, if any, before snapshotting. It will be reactived when the snapshot is resumed." > "Details: This prevents a timer interrupt from waking up the active delay in the midst snapshoting, since the active delay will be restarted when resuming the snapshot and we don't want to process the delay twice." > > AccessProtect wait. > self primSignal: nil atMilliseconds: 0. > self saveResumptionTimes. > + DelaySuspended := true.! > - ! > > Item was added: > + ----- Method: Delay>>printOn: (in category 'printing') ----- > + printOn: aStream > + super printOn: aStream. > + aStream nextPutAll: '('; print: delayDuration; nextPutAll: ' msecs'. > + beingWaitedOn ifTrue:[ > + aStream nextPutAll: '; '; print: resumptionTime - Time millisecondClockValue; nextPutAll: ' msecs remaining'. > + ]. > + aStream nextPutAll: ')'.! > > Item was changed: > ----- Method: Delay>>unschedule (in category 'private') ----- > unschedule > + AccessProtect critical:[ > + FinishedDelay := self. > + TimingSemaphore signal. > + ].! > - "Unschedule this Delay. Do nothing if it wasn't scheduled." > - > - | done | > - TimerEventLoop ifNotNil:[^self unscheduleEvent]. > - AccessProtect critical: [ > - done := false. > - [done] whileFalse: > - [SuspendedDelays remove: self ifAbsent: [done := true]]. > - ActiveDelay == self ifTrue: [ > - SuspendedDelays isEmpty > - ifTrue: [ > - ActiveDelay := nil. > - ActiveDelayStartTime := nil] > - ifFalse: [ > - SuspendedDelays removeFirst activate]]]. > - ! > > Item was changed: > ----- Method: Delay>>setDelay:forSemaphore: (in category 'private') ----- > + setDelay: millisecondCount forSemaphore: aSemaphore > - setDelay: milliseconds forSemaphore: aSemaphore > "Private!! Initialize this delay to signal the given semaphore after the given number of milliseconds." > > + delayDuration := millisecondCount. > - delayDuration := milliseconds asInteger. > - delayDuration < 0 ifTrue: [self error: 'delay times cannot be negative']. > - delayDuration > (SmallInteger maxVal // 2) > - ifTrue: [self error: 'delay times can''t be longer than about six days (', (SmallInteger maxVal // 2) printString , 'ms)']. > delaySemaphore := aSemaphore. > + beingWaitedOn := false. > + ! > - beingWaitedOn := false.! > > Item was changed: > ----- Method: Delay class>>forSeconds: (in category 'instance creation') ----- > forSeconds: aNumber > + "Return a new Delay for the given number of seconds. Sending 'wait' to this Delay will cause the sender's process to be suspended for approximately that length of time." > + > + aNumber < 0 ifTrue: [self error: 'delay times cannot be negative']. > + ^ self new > + setDelay: (aNumber * 1000) asInteger > + forSemaphore: Semaphore new > - ^ self forMilliseconds: aNumber * 1000 > ! > > Item was changed: > Object subclass: #Delay > instanceVariableNames: 'delayDuration resumptionTime delaySemaphore beingWaitedOn' > + classVariableNames: 'AccessProtect ActiveDelay ActiveDelayStartTime DelaySuspended FinishedDelay RunTimerEventLoop ScheduledDelay SuspendedDelays TimerEventLoop TimingSemaphore' > - classVariableNames: 'AccessProtect ActiveDelay ActiveDelayStartTime FinishedDelay RunTimerEventLoop ScheduledDelay SuspendedDelays TimerEventLoop TimingSemaphore' > poolDictionaries: '' > category: 'Kernel-Processes'! > > !Delay commentStamp: 'stephaneducasse 10/1/2005 21:07' prior: 0! > I am the main way that a process may pause for some amount of time. The simplest usage is like this: > > (Delay forSeconds: 5) wait. > > An instance of Delay responds to the message 'wait' by suspending the caller's process for a certain amount of time. The duration of the pause is specified when the Delay is created with the message forMilliseconds: or forSeconds:. A Delay can be used again when the current wait has finished. For example, a clock process might repeatedly wait on a one-second Delay. > > The maximum delay is (SmallInteger maxVal // 2) milliseconds, or about six days. A delay in progress when an image snapshot is saved is resumed when the snapshot is re-started. Delays work across millisecond clock roll-overs. > > > For a more complex example, see #testDelayOf:for:rect: . > > A word of advice: > This is THE highest priority code which is run in Squeak, in other words it is time-critical. The speed of this code is critical for accurate responses, it is critical for network services, it affects every last part of the system. > > In short: Don't fix it if it ain't broken!! This code isn't supposed to be beautiful, it's supposed to be fast!! The reason for duplicating code is to make it fast. The reason for not using ifNil:[]ifNotNil:[] is that the compiler may not inline those. Since the effect of changes are VERY hard to predict it is best to leave things as they are for now unless there is an actual need to change anything! > > Item was changed: > ----- Method: Delay>>schedule (in category 'private') ----- > schedule > + "Schedule this delay" > + resumptionTime := Time millisecondClockValue + delayDuration. > + AccessProtect critical:[ > + ScheduledDelay := self. > + TimingSemaphore signal. > + ].! > - "Private!! Schedule this Delay, but return immediately rather than waiting. The receiver's semaphore will be signalled when its delay duration has elapsed." > - > - beingWaitedOn ifTrue: [self error: 'This Delay has already been scheduled.']. > - > - TimerEventLoop ifNotNil:[^self scheduleEvent]. > - AccessProtect critical: [ > - beingWaitedOn := true. > - resumptionTime := Time millisecondClockValue + delayDuration. > - ActiveDelay == nil > - ifTrue: [self activate] > - ifFalse: [ > - resumptionTime < ActiveDelay resumptionTime > - ifTrue: [ > - SuspendedDelays add: ActiveDelay. > - self activate] > - ifFalse: [SuspendedDelays add: self]]]. > - ! > > Item was removed: > - ----- Method: Delay>>scheduleEvent (in category 'private') ----- > - scheduleEvent > - "Schedule this delay" > - resumptionTime := Time millisecondClockValue + delayDuration. > - AccessProtect critical:[ > - ScheduledDelay := self. > - TimingSemaphore signal. > - ].! > > Item was removed: > - ----- Method: Delay>>activate (in category 'private') ----- > - activate > - "Private!! Make the receiver the Delay to be awoken when the next timer interrupt occurs. This method should only be called from a block protected by the AccessProtect semaphore." > - TimerEventLoop ifNotNil:[^nil]. > - ActiveDelay := self. > - ActiveDelayStartTime := Time millisecondClockValue. > - ActiveDelayStartTime > resumptionTime ifTrue:[ > - ActiveDelay signalWaitingProcess. > - SuspendedDelays isEmpty ifTrue:[ > - ActiveDelay := nil. > - ActiveDelayStartTime := nil. > - ] ifFalse:[SuspendedDelays removeFirst activate]. > - ] ifFalse:[ > - TimingSemaphore initSignals. > - Delay primSignal: TimingSemaphore atMilliseconds: resumptionTime. > - ].! > > Item was removed: > - ----- Method: Delay>>unscheduleEvent (in category 'private') ----- > - unscheduleEvent > - AccessProtect critical:[ > - FinishedDelay := self. > - TimingSemaphore signal. > - ].! > > Item was removed: > - ----- Method: Delay class>>startTimerInterruptWatcher (in category 'timer process') ----- > - startTimerInterruptWatcher > - "Reset the class variables that keep track of active Delays and re-start the timer interrupt watcher process. Any currently scheduled delays are forgotten." > - "Delay startTimerInterruptWatcher" > - | p | > - self stopTimerEventLoop. > - self stopTimerInterruptWatcher. > - TimingSemaphore := Semaphore new. > - AccessProtect := Semaphore forMutualExclusion. > - SuspendedDelays := > - SortedCollection sortBlock: > - [:d1 :d2 | d1 resumptionTime <= d2 resumptionTime]. > - ActiveDelay := nil. > - p := [self timerInterruptWatcher] newProcess. > - p priority: Processor timingPriority. > - p resume. > - ! > > Item was removed: > - ----- Method: Delay class>>guardianDelay (in category 'instance creation') ----- > - guardianDelay > - "Make a Delay with a resumption time far in the future but still a SmallInteger so that it can be used as a guardian for the active delay queue. No process will be waiting on this and when triggered it will do nothing. What it allows is very long Delays where the resumption time is a large integer; should such a delay get activated it will fail the primitive and we creat one of these guardians to make sure the delay timer keeps going and triggers the resumption time recalculations in save/restoreResumptionTime" > - ^self new beGuardianDelay! > > Item was removed: > - ----- Method: Delay class>>stopTimerInterruptWatcher (in category 'timer process') ----- > - stopTimerInterruptWatcher > - "Reset the class variables that keep track of active Delays and re-start the timer interrupt watcher process. Any currently scheduled delays are forgotten." > - "Delay startTimerInterruptWatcher" > - self primSignal: nil atMilliseconds: 0. > - TimingSemaphore ifNotNil:[TimingSemaphore terminateProcess].! > > Item was removed: > - ----- Method: Delay class>>timerInterruptWatcher (in category 'timer process') ----- > - timerInterruptWatcher > - "This loop runs in its own process. It waits for a timer interrupt and > - wakes up the active delay. Note that timer interrupts are only enabled > - when there are active delays." > - | nowTime | > - [true] > - whileTrue: [TimingSemaphore wait. > - AccessProtect > - critical: [ActiveDelay == nil > - ifFalse: [ActiveDelay signalWaitingProcess. > - (nowTime := Time millisecondClockValue) < ActiveDelayStartTime > - ifTrue: ["clock wrapped so adjust the resumption > - times of all the suspended delays. No > - point adjusting the active delay since > - we've just triggered it" > - SuspendedDelays > - do: [:d | d adjustResumptionTimeOldBase: ActiveDelayStartTime newBase: nowTime]]]. > - SuspendedDelays isEmpty > - ifTrue: [ActiveDelay := nil. > - ActiveDelayStartTime := nil] > - ifFalse: [SuspendedDelays removeFirst activate]]]! > > Item was removed: > - ----- Method: Delay>>beGuardianDelay (in category 'private') ----- > - beGuardianDelay > - "see comment for class method guardianDelay" > - beingWaitedOn := false. > - resumptionTime := SmallInteger maxVal. > - delaySemaphore := Semaphore new! > > >

2 1

Graphical Application Where to Begin?
by JohnReed Maffeo 05 Sep '09

05 Sep '09

Hello, It has been awhile since I have written, but I am back trying to continue work on an application that has been on hiatus for awhile. My current goal is to write a graphical interface to display geo-referenced data in a scrollable/zoomable window. I have written code to read data from a file and use it to create an instance of GraphicSymbol. I can display the object on a form and display the form. Now I want to start building a simple map browser application, but I don't know where to start. From what I understand, there is a lot of Morphic that will/should be deprecated. What can I do to ensure that my design conforms to the appropriate Morphic roadmap? Any suggestions would be greatly appreciated. The vision is to create a virtual landscape from a combination of elevation data ie a matrix of elevation points and planar features ie various lines that identify roads, rivers, boundaries etc. Thanks, John-Reed Maffeo -- Powered By Outblaze

7 9

Cuis font sizes
by Andreas Raab 05 Sep '09

05 Sep '09

Hi Juan - This is mostly to Juan but others might be able to help as well. I noticed that the Cuis fonts disagree in their pixel sizes significantly with that we get from the plain DejaVu true type fonts in Squeak. A bit of math reveals that at the (assumed) screen resolution of 96dpi a 12pt font should result in 16px high font (12*96/72 see Eqn. 3 in [1]). The corresponding Bitmap DejaVu 12pt font, however, is 19px high which would mean a screen resolution of 114dpi (19*72/12). Is this choice intentional? Unfortunately, none of the pixel sizes for Bitmap DejaVu correspond exactly to a point size when assuming 96dpi. The reason this is causing me grief is that I am trying to set up a mechanism for allowing people to drop the DejaVu fonts into their images and have any glyphs not part of the cuis fonts rendered in DejaVu truetype instead. I can probably get the truetype rendering to spit out a precise pixel size to match it but I think it'd be good if we had clarity on how these pixel sizes came into being. Any insights appreciated! Cheers, - Andreas [1] http://www.emdpi.com/emsquare.html

9 18

Re: [squeak-dev] Clamato
by Avi Bryant 05 Sep '09

05 Sep '09

On Fri, Sep 4, 2009 at 11:53 AM, Dan Ingalls<Dan(a)squeakland.org> wrote: > I need to think about this some more and, more importantly, look at and write some code. We do a lot of point and rectangle arithmetic which needs returns, as in > > var loc = m.bounds().center().plus(delta) > <==> > loc = m bounds center addPt: delta > > How would you go about this? I'm not sure I see the problem... let me clear up a possible point of confusion. It's not that methods don't have a return value, it's that methods, like blocks, always return the value of their last statement (instead of having explicit returns with ^ ). Or am I missing what you're asking? > ...and while we're on the topic of point arithmetic, it drives me nuts that JS won't let you add infix methods. You can put them in the dictionary, but you can't call them. What I've thought of is to use St's aggregated infix for non-number users of '+', so 'a + b' translates to 'a + b' for numbers and, eg, 'a &+ b' translates to 'a plus b' for points, dates and the like. > > That at least gives us... > > loc = m bounds center &+ delta > > I've thought of using '.' for this case, because it is so much less intrusive, as... > > loc = m bounds center .+ delta FWIW: in clamato, a + b becomes a.$plus(b) It sounds like you're trying to map smalltalk selectors to existing, equivalent javascript method names, whereas I'm intentionally mangling them (foo:bar: becomes _foo_bar_) to avoid clashes with any methods that might already be there. This means you need to explicitly wrap any Javascript methods you might want to use in primitives. > That said, though, are you open to a possible collaboration to put a bit more attention on making it the best it can be without making it significantly more complicated. Absolutely. I would be delighted to have more collaboration on this (and honored, Dan, to have yours) - though of course I also reserve the right to ignore any and all suggestions :) > I'm still interested in my Lively April Fools prank, and it *could* have huge synergy with what you are doing if we agreed on the language. You might find that an entire Lively Kernel running in Clamato was a nice reservoir to draw on. Sure. I don't know too much about what Lively requires of its classes, but it can't be too hard to make this work. Though I should note that I personally am more interested in a DOM-based UI than in canvas or SVG (the DOM is, of course, much more limiting, but it's also more familiar to most users and that's important for adoption). > Have you integrated this development environment with a repository (haven't gone through your code yet)? We're very happy with our browser hookup that works with a local version of our repository by live WevDAV hookup. Yes, we do what's probably the same thing - every time you save a method, a new version of the file is saved over WebDAV. Avi

2 1

The Trunk: KernelTests-laza.92.mcz
by commits＠source.squeak.org 05 Sep '09

05 Sep '09

Alexander Lazarević uploaded a new version of KernelTests to project The Trunk: http://source.squeak.org/trunk/KernelTests-laza.92.mcz ==================== Summary ==================== Name: KernelTests-laza.92 Author: laza Time: 5 September 2009, 8:54:09 am UUID: 383b4c94-242c-9f49-99bc-1e999a57f2c4 Ancestors: KernelTests-ar.91 Removed the test that checks that too long delays are refused and added a test that checks that long delays are ok. =============== Diff against KernelTests-ar.91 =============== Item was changed: ----- Method: DelayTest>>testBounds (in category 'testing-limits') ----- testBounds "self run: #testBounds" self should: [Delay forMilliseconds: -1] raise: Error. + self shouldnt: [Delay forMilliseconds: SmallInteger maxVal + 1] raise: Error. - self should: [Delay forMilliseconds: SmallInteger maxVal // 2 + 1] raise: Error. "Not longer than a day" - self shouldnt: [Delay forMilliseconds: SmallInteger maxVal // 2] raise: Error. self shouldnt: [(Delay forMilliseconds: Float pi) wait] raise: Error. "Wait 3ms" !

1 0

The Trunk: Kernel-laza.241.mcz
by commits＠source.squeak.org 05 Sep '09

05 Sep '09

Alexander Lazarević uploaded a new version of Kernel to project The Trunk: http://source.squeak.org/trunk/Kernel-laza.241.mcz ==================== Summary ==================== Name: Kernel-laza.241 Author: laza Time: 5 September 2009, 8:51:42 am UUID: bb272481-19dd-ca4d-9f2d-283443900cf5 Ancestors: Kernel-ar.240 Removes the warning of too long delays from the Delay class comment =============== Diff against Kernel-ar.240 =============== Item was changed: Object subclass: #Delay instanceVariableNames: 'delayDuration resumptionTime delaySemaphore beingWaitedOn' classVariableNames: 'AccessProtect ActiveDelay ActiveDelayStartTime DelaySuspended FinishedDelay RunTimerEventLoop ScheduledDelay SuspendedDelays TimerEventLoop TimingSemaphore' poolDictionaries: '' category: 'Kernel-Processes'! + !Delay commentStamp: 'laza 9/5/2009 08:45' prior: 0! - !Delay commentStamp: 'stephaneducasse 10/1/2005 21:07' prior: 0! I am the main way that a process may pause for some amount of time. The simplest usage is like this: (Delay forSeconds: 5) wait. An instance of Delay responds to the message 'wait' by suspending the caller's process for a certain amount of time. The duration of the pause is specified when the Delay is created with the message forMilliseconds: or forSeconds:. A Delay can be used again when the current wait has finished. For example, a clock process might repeatedly wait on a one-second Delay. + A delay in progress when an image snapshot is saved is resumed when the snapshot is re-started. Delays work across millisecond clock roll-overs. - The maximum delay is (SmallInteger maxVal // 2) milliseconds, or about six days. A delay in progress when an image snapshot is saved is resumed when the snapshot is re-started. Delays work across millisecond clock roll-overs. For a more complex example, see #testDelayOf:for:rect: . A word of advice: This is THE highest priority code which is run in Squeak, in other words it is time-critical. The speed of this code is critical for accurate responses, it is critical for network services, it affects every last part of the system. In short: Don't fix it if it ain't broken!! This code isn't supposed to be beautiful, it's supposed to be fast!! The reason for duplicating code is to make it fast. The reason for not using ifNil:[]ifNotNil:[] is that the compiler may not inline those. Since the effect of changes are VERY hard to predict it is best to leave things as they are for now unless there is an actual need to change anything!

1 0

Somehow , the 'Network-Igor.Stasenko.36.mcz' commit to inbox was not broadcasted via mail
by Igor Stasenko 04 Sep '09

04 Sep '09

There is a small fix i made, you can read it in package snapshot comment. -- Best regards, Igor Stasenko AKA sig.

3 3

← Newer
1
...
27
28
29
30
31
32
33
...
37
Older →

Jump to page: