Squeak-dev September 2003

squeak-dev@lists.squeakfoundation.org

163 participants
651 discussions

veryDeepCopy, ImageSegments, and actionMaps
by Ned Konz 17 Aug '23

17 Aug '23

Any Object can have dependents whether or not it has an instance variable to keep track of them (these are kept in the DependentsFields class variable of Object). The only class that keeps its own dependents is Model. Any Object can have an actionMap. These are supplied by EventManager class by default, and are stored in the ActionMaps class variable of EventManager. The only class that keeps its own action maps is Morph. When we veryDeepCopy an object we also copy the global dependents (see DeepCopier>>fixDependents). However, we don't copy the actionMap in those cases where it's held by the EventManager class. When we file out an object that uses DependentsFields out into an ImageSegment, we also file out the dependents (see ImageSegment>>dependentsSave:) if it's in a project. We do this by adding a project parameter called #GlobalDependentsInProject. However, we don't file out the actionMap in those cases where it's held by the EventManager class. My question is this: should we fix these behaviors (copying and serialization) to include these "global action maps" just like we already include the "global dependents"? Or should we discourage the use of the "global action maps"? -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE

3 2

Trying to build SerialPlugin for IrDA under MPW (MacOS 8.6)
by Hans-Martin Mosner 24 May '12

24 May '12

Hi, I recently got a serial IrDA adapter and wanted to play around with it under Squeak. This adapter has a baudrate switching mechanism which depends on toggling the DTR and RTS lines in the correct sequence. To support this, I wanted to build a new SerialPlugin with an additional primitive. However, the plugin as built with MPW does not seem to be accepted by the VM, which still uses the internal plugin in all cases. I've tried placing the plugin in a 'Plugins' subfolder of the VM's folder, and then directly into the VM folder, without any success. I have considered placing debug outputs in appropriate places in the VM, but building a complete VM with MPW is not easy - I would have to create an MPW makefile for the whole shebang, which is more work than I can invest at the moment. Does any of you Mac aficionados have an idea where the problem could be? I have appended the plugin and corresponding source files in a ZIP file (it's not big) so if you want to try it out you could do it right now :-) BTW, would it make sense to incorporate support for switching serial port parameters of an open port into the standard plugin? Closing and opening a port just to change the baud rate seems a little overkill... Cheers, Hans-Martin

6 14

RE: Progrmaming in Bytecode?
by Swan, Dean 24 Nov '06

24 Nov '06

>> It's not uncommon to find x86 code padded with a lot of NOPs >> that weren't in the original source. > >Cowardly assembler! ;-) > >> [NOPs...] Sometimes they're there to pad the execution pipeline, and >> sometimes it's just because the code generator allocates a whole large >> enough for the largest form of an instruction to simplify the branch >> target calculations. > >We must be talking about M$oft assemblers here. ;^p) Yes. This has been pretty characteristic of MS assemblers as far back as I can remember (i.e. about MASM 2.0 - I remember being quite excited when MASM 4.0 came out - '186 and '286 code!). >The real problem is that the solution in general might require N >iterations for a piece of code with N branches -- if promoting one of them >from short to long causes the next (or, much more problematically, a >preceding) one to become long too. You're such an optimist. O(n) is close to best case for this problem. >This is a weak form of what I suppose one might call an "oscillating >constraints" problem. (Hands up anyone who has ever seen [La]TeX get Hence, stochastic optimization. Even MS uses stochastic optimizers now, even in vb. (didn't want to say that too loud here. ;) ) >The usual cause for "padding" in (capable) assemblers (other than for >prefetch alignment, as you point out) is due to dynamic linking where the >size of call sites isn't known until link time. (On modern systems the Not only prefetch alignment. I have heard, although I don't know the details that sometimes it's desirable to force a dead cycle in the execution pipeline. I can only guess that this would be to influence the scheduling of the execution units for complex instructions, but I consider myself to be only a "casual observer" with respect to this kind of thing. >In Smalltalk you've also got a bigger problem: for very long methods the >range of even unconditional branches might not be enough, at which point >I know this has nothing whatsoever to do with Smalltalk...) Well, to kind of tie this back into something relevant to Squeak - I know a few people have talked about "new" VM instruction sets. It seems that it might help to think about a VM where all instructions are the same size. This could simplify code generation (i.e. compiling methods), and possibly improve performance and *reduce* code size (at least in the case of long methods). The anticipated performance improvement would come from forcing aligned fetches more often and potentially simpler instruction decoding. It's really nice to work with an architecture where all instructions take up one word of memory, take one cyle to execute, and sometimes can cause five or six different useful things to occur during that one cycle. (ok, so I write a lot of DSP code...) The single cycle thing isn't necessarily relevant for Smalltalk, but those "extended" bytecodes are really distasteful to me. -Dean

10 15

[FIX] FreeTranslationFixes-dgd
by diegogomezdeck＠consultar.com 26 Sep '04

26 Sep '04

'From Squeak3.5 of ''11 April 2003'' [latest update: #5180] on 28 August 2003 at 9:29:13 am'! "Change Set: FreeTranslationFixes-dgd Date: 28 August 2003 Author: Diego Gomez Deck <DiegoGomezDeck(a)ConsultAr.com> Two really simple fixes. - the port isn't 5081 anymore - the argument template isn't need To try take a look to the items 'translate it' and 'choose language' in the shiftedYellowButtonMenu (shift right click on windows). "

3 3

[ENH] InspectorsAndExplorersWithColor
by diegogomezdeck＠consultar.com 25 Sep '04

25 Sep '04

Change Set: InspectorsAndExplorersWithColor-dgd Date: 28 June 2003 Author: Diego Gomez Deck The inspectors and the explorers now will use the color of the inpected/explored object (using the idiom of #defaultBackgroundColor). Two methods are added. Inspector>>defaultBackgroundColor and ObjectExplorer>>defaultBackgroundColor and the implementation is a simply delegation to the inpected/explored object.

2 1

[ENH]Comma operator to Etoy strings.
by Karl Ramberg 23 Sep '04

23 Sep '04

"Change Set: ExtendableText Date: 22 September 2003 Author: Karl Ramberg Adds #, (comma) as an operator in Etoys strings so strings can grow."!

6 10

[ENH] Undeclared cleanup
by Michael Rueger 23 Aug '04

23 Aug '04

Hi all, this tiny goodie removes all entries without references from Undeclared. Maybe this should be called after every file in. Michael 'From Squeak2.8alpha of 19 January 2000 [latest update: #2052] on 5 May 2000 at 1:27:13 pm'! !Utilities class methodsFor: 'miscellaneous' stamp: 'mir 5/5/2000 13:26'! cleanupUndeclared | toBeRemoved refs | "Utilities cleanupUndeclared" toBeRemoved _ Set new. Undeclared associationsDo: [:assoc | refs _ Smalltalk allCallsOn: assoc. refs isEmpty ifTrue: [toBeRemoved add: assoc key]]. toBeRemoved do: [:each | Undeclared removeKey: each]! !

4 19

[BUG][FIX]SelectionMorph organizeInto[Row/Column]
by tblanchard＠mac.com 22 Aug '04

22 Aug '04

This fixes the very annoying organizeIntoRow and organizeIntoColumn features in SelectionMorph. Previously it ordered the items by their front to back order - which resulted in an inversion of my list. These changes first sort the selected items top to bottom (column) or left to right(row) so you get something a little closer to what you likely envisioned when you executed the command. Todd Blanchard

3 2

[BUG][FIX] Morphic: File Selection Dialog [was MVC: File Selectio n Dialog]
by Magistrello Alejandro (SFA) 04 Aug '04

04 Aug '04

>De: Boris Gaertner [SMTP:Boris.Gaertner@gmx.net] >Asunto: Re: [BUG][FIX] MVC: File Selection Dialog (shows another bug) > >I tried to execute > (PluggableFileList oldFileMenu: (FileDirectory on: 'c:\windows\fonts')) > startUpWithCaption: 'Select a File, please:' >in a Mrophic project and receive a strange effect, that I cannot quickly >fix: >I put that statement into a workspace that I place more or less centered >on the screen. When I execute the statement, the selection dialog is >displayed above the workspace. This looks good, but as soon as a >call up one of the menues of the dialog, the selected text of the workspace >comes into the foreground. >I think that in PluggableFileList>>open one should have something like > Smalltalk isMorphic > ifTrue: [<...>] > ifFalse: [<...>] > >I tried to do that, the result can be found in MorphicFSD.1.cs <<MorphicFSD.1.cs>> >In SystemWindow>>update: I added a reaction to the >message #close. >In PluggableFileList>>openAsMorphLasbel:inWorld: >I added an event loop. The condition of the loop uses the >fact that the model of a SystemWindow is niled when the >window is deleted. >I am not very experienced user of the Morphic interface, >so comments are highly wellcome. > > >Greetings, >Boris > if you use the closeButton you'll get a walkback. btw, this is pretty similar to FileList2 modalFileSelector regards, Alejandro

2 1

[ENH] ExceptionDialogs-avi
by avi＠beta4.com 28 Jul '04

28 Jul '04

from preamble: "Change Set: ExceptionDialogs-avi Date: 31 August 2003 Author: Avi Bryant Adds Notifications for #inform:, #confirm:, and #confirm:orCancel:. By default these still use PopUpMenu, but they could be caught and dealt with more appropiately in different UI contexts."!

2 1

Jump to page: