TO ALL:
Just we meet with Matthew and some words with Craig. I also see Steph and Andreas sending feedback here or in release team list. It's important we reach some agree for next 3.11. We only exchange ideas , trying to see the best road to follow. I repeat my vision. One day I wish to have a very small and smart thing. I open workspace and say "Pier" and all loads and I in bussiness. Maria wish play "DressTheDoll" with Etoys, then she type DressTheDoll (and the thing only loads the needed to she). Some talented music (many Squeakers ) choose the last Stephane Rollandin have and voila, he is doing jazz, tango or Nepal music (and not load Etoys or Web) Sure many more examples. Now we have 3.10 , some just discover could load several things automatic, some don't know 3.10 lack some packages living outside the base image. 3.11 could cut more packages and load again if needed. The goal is going closer to MinimalMorphic. Then to Kernel. Then to Spoon. I apologize if Craig or Pavel think different, sure they could explain better. The point is if we should start from 3.8.2 (without Traits and Monticello) or we should continue from 3.10 is now. The tentative image I put in ftp is only a first prototype idea and was 3.10. But some believe could have sense do some similar from 3.8.2 Kind of 3.10 without Monticello and Traits. So we listen all ideas, wait all help. ======== Next meeting is scheduled tomorrow 2 February 2008 on IRC 8:00 Phoenix 16:00 Rosario 13:00 Europa ====== |
On Feb 2, 2008, at 00:05 , Edgar J. De Cleene wrote: [...] > > But some believe could have sense do some similar from 3.8.2 > Kind of 3.10 without Monticello and Traits. ...and without more than 350 fixes and enhancements that went into 3.9. Who would be going to redo this work? Just curious Adrian |
On 2/1/08, Adrian Lienhard <[hidden email]> wrote:
> ...and without more than 350 fixes and enhancements that went into 3.9. > > Who would be going to redo this work? Tangentially - what's the easiest way to see a list of these 350 enhancements? Avi |
you know that this is mc so the changesets are not available in
changesorter - but you can browse the window to get an idea - read the [ANN] in the mailing-list - check the closed item on mantis - and we could (not me) write a MC script that extract all the comments in the packages since everything was recorded there even the entry of the mantis archive and also perform a diff for the - this is basically the same than the way Sophie, seaside are maintained. Of course, using package was certainly a mistake but other communities succeeded and they do not have cs. Stef On Feb 2, 2008, at 1:14 AM, Avi Bryant wrote: > On 2/1/08, Adrian Lienhard <[hidden email]> wrote: > >> ...and without more than 350 fixes and enhancements that went into >> 3.9. >> >> Who would be going to redo this work? > > Tangentially - what's the easiest way to see a list of these 350 > enhancements? > > Avi > > |
In reply to this post by Edgar J. De Cleene
BTW I'm a bit amazed that we do not draw any conclusion of the failure
in social engineering and the rest in 3.10. Someone coined it a long agony. And I thought he has a point. This is strange since 3.9 and 3.10 used the same process and tools. So of course I would love to have the best of both world changeset and packages. Stef On Feb 2, 2008, at 12:05 AM, Edgar J. De Cleene wrote: > TO ALL: > > Just we meet with Matthew and some words with Craig. > I also see Steph and Andreas sending feedback here or in release > team list. > > It's important we reach some agree for next 3.11. > We only exchange ideas , trying to see the best road to follow. > > I repeat my vision. > > One day I wish to have a very small and smart thing. > I open workspace and say "Pier" and all loads and I in bussiness. > > Maria wish play "DressTheDoll" with Etoys, then she type > DressTheDoll (and > the thing only loads the needed to she). > > Some talented music (many Squeakers ) choose the last Stephane > Rollandin > have and voila, he is doing jazz, tango or Nepal music (and not load > Etoys > or Web) > > Sure many more examples. > > Now we have 3.10 , some just discover could load several things > automatic, > some don't know 3.10 lack some packages living outside the base image. > > 3.11 could cut more packages and load again if needed. > > The goal is going closer to MinimalMorphic. > Then to Kernel. > Then to Spoon. > > I apologize if Craig or Pavel think different, sure they could explain > better. > > The point is if we should start from 3.8.2 (without Traits and > Monticello) > or we should continue from 3.10 is now. > > The tentative image I put in ftp is only a first prototype idea and > was > 3.10. > > But some believe could have sense do some similar from 3.8.2 > Kind of 3.10 without Monticello and Traits. > > So we listen all ideas, wait all help. > > ======== > Next meeting is scheduled tomorrow 2 February 2008 on IRC > > 8:00 Phoenix 16:00 Rosario 13:00 Europa > > ====== > > > > > |
In reply to this post by Adrian Lienhard
El 2/1/08 8:59 PM, "Adrian Lienhard" <[hidden email]> escribió: > ...and without more than 350 fixes and enhancements that went into 3.9. > > Who would be going to redo this work? > > Just curious > Adrian No, with all the Squeak fixes. 350 or 3500. And all relevant fixes from Sophie,SmallLand, OlPC,Croquet ( if exists and not today in 3.10) Who you think is working now ? :=) Edgar |
In reply to this post by Edgar J. De Cleene
> ======== > Next meeting is scheduled tomorrow 2 February 2008 on IRC > > 8:00 Phoenix 16:00 Rosario 13:00 Europa > > ====== As I said , yesterday I broke my watch Should read 8:00 Phoenix 16:00 Europa 13:00 Argentina By the way , exist some Dick Tracy watch today ? Edgar I on IRC and Skype as edgardec, off course I read mails to here until this was solved. Then discuss should continue on release team list. Edgar |
In reply to this post by Edgar J. De Cleene
2008/2/1, Edgar J. De Cleene <[hidden email]>:
> > > The point is if we should start from 3.8.2 (without Traits and Monticello) > or we should continue from 3.10 is now. My vote is without Traits!!! (Traits: big example of a not useful thing, IMHO), Cheers. |
In reply to this post by Avi Bryant-2
On Feb 2, 2008, at 01:14 , Avi Bryant wrote: > On 2/1/08, Adrian Lienhard <[hidden email]> wrote: > >> ...and without more than 350 fixes and enhancements that went into >> 3.9. >> >> Who would be going to redo this work? > > Tangentially - what's the easiest way to see a list of these 350 > enhancements? In Mantis I filtered for: Fixed in Version = 3.9 and: [ status = closed ] Adrian |
In reply to this post by Edgar J. De Cleene
On Fri, Feb 01, 2008 at 08:05:25PM -0300, Edgar J. De Cleene wrote:
> TO ALL: > > Just we meet with Matthew and some words with Craig. > I also see Steph and Andreas sending feedback here or in release team > list. Edgar and I met for 30 minutes this morning and discussed what a 3.11 based on 3.8.2 would look like, how to get there, and how one could have the best of both changesets and packages for the release process: - Porting the 3.9 and 3.10 fixes back to changesets would be the best way to ensure the sanity of the release team. Edgar says the tool World > Open > PackageList could keep the image modular without MC loaded. I'm not sure how this works, but sounds interesting - If we did that, how would we generate these change-sets? Edgar says an experimental branch using DeltaStreams in parallel with change sets may be useful, but we didn't discuss it much. My comments about the meeting: - A deltaStream-based update stream would be a good experiment, but DeltaStreams doesn't yet have a script to generate diff's from MCZ packages. However, I have made one before (as part of my SqueakSource commit emails project) that we could use to partially automate this conversion. - DeltaStreams would be usable for this task, but not much so. But this would only be an experiment. I don't see DeltaStreams making the task of release easier for some time yet, especially as the file-out format is still not-forward-compatible binary, and no "subscription" mechanism exists yet in DS, due to it's alpha status. However, the core mechanism of change-model and SystemEditor is pretty stable, and is due for a stress test; this would be a good stress test - I don't want to give you the impression that we are going to abandon the 3.10 release. Keith and I would like to maintain it via LPF just as much as Edgar wants to make 3.11 out of 3.8.2. I would like there to be two 3.11 releases: one with traits and the other without. This just didn't arise in the meeting > > It's important we reach some agree for next 3.11. > We only exchange ideas , trying to see the best road to follow. > > I repeat my vision. > > One day I wish to have a very small and smart thing. > I open workspace and say "Pier" and all loads and I in bussiness. > > Maria wish play "DressTheDoll" with Etoys, then she type DressTheDoll (and > the thing only loads the needed to she). > > Some talented music (many Squeakers ) choose the last Stephane Rollandin > have and voila, he is doing jazz, tango or Nepal music (and not load Etoys > or Web) > > Sure many more examples. > > Now we have 3.10 , some just discover could load several things automatic, > some don't know 3.10 lack some packages living outside the base image. > > 3.11 could cut more packages and load again if needed. > > The goal is going closer to MinimalMorphic. > Then to Kernel. > Then to Spoon. > > I apologize if Craig or Pavel think different, sure they could explain > better. > > The point is if we should start from 3.8.2 (without Traits and Monticello) > or we should continue from 3.10 is now. > > The tentative image I put in ftp is only a first prototype idea and was > 3.10. > > But some believe could have sense do some similar from 3.8.2 > Kind of 3.10 without Monticello and Traits. > > So we listen all ideas, wait all help. > > ======== > Next meeting is scheduled tomorrow 2 February 2008 on IRC > > 8:00 Phoenix 16:00 Rosario 13:00 Europa > > ====== > > __._,_.___ > Mensajes con este tema (1) Responder (mediante la Web) | Crear un tema > nuevo > Mensajes | Archivos | Fotos | Marcadores | Base de datos | Sondeos | > Miembros | Agenda > correo electronico a: [hidden email] > > correo electronico a: [hidden email] > > Yahoo! Grupos > Modificar la configuracion mediante la Web (ID de Yahoo! obligatoria) > Modificar la configuracion mediante el correo: Cambiar a resumen diario | > Cambiar el formato a Tradicional > Visita tu grupo | Condiciones de uso de Yahoo! Grupos | Cancelar > suscripcion > Actividad reciente > * 1 > Miembros nuevos > Visita tu grupo > Y! Respuestas > > Pregunt?/a> > > Gente real te > > responde > > Yahoo! Messenger > > Mensajero perfecto > > Llam?de PC a PC > > sin costo alguno > > Barra Yahoo! > > Todo a un toque > > Acceso rapido a > > servicios Yahoo! > > . > __,_._,___ -- Matthew Fulmer -- http://mtfulmer.wordpress.com/ Help improve Squeak Documentation: http://wiki.squeak.org/squeak/808 |
In reply to this post by Avi Bryant-2
Hi avi
I was not sure that I was clear when I mentioned the window... Here are some of the changes that have been done in 3.9 7067: ----- - browser speedup - some green tests 7066: ----- - class rename fix 7065: ----- - eval fix - font fix 7064: ----- - fix MC repository - fix color of SM 7063: ----- - Squeakmap purge - Pragma fixes - ToggleBreak fix - HaloDup 7061: ----- - required selector traits - fix cs 7058: ----- - cleaning services + window color registry 7057: ----- - clean cs 7056 ----- - Change version info to gamma - fixes traits - Write new source file. 7053 ----- - unload FFI (but not discardFFI 7052 ----- - Fix Changeset Unamed1 http://bugs.impara.de/view.php?id=4391 - 0004349: some class comments are outdated (nicolas cellier) - 0004328: typo in Balloon text for Fridge, Listener and Text chat (espin) - 0004350: In 7048 Too persistant halos pop to front whenever an object is dropped.(wiz) - 0004392: Bad example in comment for SmalltalkImage>>vmVersion (lewis) - 0004391: Changes disappear from changes set (lewis) - Fix for SqueakMap 7051 ----- - prepare release 7050 ----- - 0003790: Text cursor too bright - cleanup methodClass/selector - fix for parserClass - call cleanup code 7049 ----- - updated SUnit - fixed overrides Services - make sure all methods are restarted - recompile some methods / tests for properties (Klaus) - fix for methodFinder to work with RB AST (Math) 7048 ----- -> Load: TrueType-ar.4 -> load 0003766: Hundreds of methods have source containing LF characters -> add services / call to service setup -> Unload packages. call some fix code -> disable multiple text undo 7047 ---- 0004312: Refactoring for behavior>>#new 0004307: DependentsArray>>copyWith: 0003133: [BUG][FIX] Float>>#asIEEE32BitWord and #fromIEEE32Bit: 0003309: In 7008 Requesting setTarget for menus can get a UndefinedObject DNU>isHandMorph 004249: SmLoader>>#categorySpecificOptions has a . after the variable definition 0004298: OrderedCollection>>#removeLast: 2; size is negative 0004314: CleanUpMacOptionKeyPressed-md.cs 0003765: ChangeLists and VersionBrowsers don't respect standardCodeFont preference. 0003777: In 7033 In ImageBrowser (omnibrowser) copyListToClipboard menu item causes a DNU. 0004273: small fix for prettyprinting (for 3.9) - merge in changes of Monticello-avi.276 and Monticello-avi.277 from the SqS MC repository: - Don't allow the active repo to be switched mid commit. - Minor optimization to dependency commits. - Traits test fixes: expected sentinfo test failures 7046 ----- smbase fixed traits undeclared fixed http://source.squeakfoundation.org/inbox/OmniBrowser.39-cwp.1.mcz http://source.squeakfoundation.org/inbox/OB-Standard.39-cwp.3.mcz 7045 ------ ToolBuilder colin fixes http://bugs.impara.de/view.php?id=3588 was bogus Name: Morphic-md.103 Author: md Time: 14 July 2006, 9:32:32 pm UUID: e9a3551d-a01a-4129-959a-87afdea47a97 Ancestors: Morphic-md.102 0003753: Loading a .morph file from FileList causes error: Invalid utf8 input detected ================================= Name: MorphicExtras-md.31 Author: md Time: 14 July 2006, 9:35:39 pm UUID: 28f52c31-f38d-4cd2-b659-2fa4d9eb0f8d Ancestors: MorphicExtras-md.30 0003753: Loading a .morph file from FileList causes error: Invalid utf8 input detected ==================================== Name: KernelTests-md.43 Author: md Time: 14 July 2006, 9:31:17 pm UUID: 3b672bd1-ae61-4344-b4e9-e5400a269725 Ancestors: KernelTests-sd.42 0003512: Number>>readFrom: does not handle error cases correctly =================================== Name: Kernel-md.141 Author: md Time: 14 July 2006, 9:29:25 pm UUID: 00e649d3-606d-4d42-b9e5-548e765bf1fd Ancestors: Kernel-md.140 0002151: Max number of literals checked in MethodNode instead of CompiledMethod 0003512: Number>>readFrom: does not handle error cases correctly ================================ Name: CollectionsTests-md.32 Author: md Time: 15 July 2006, 6:12:47 pm UUID: b44f4f02-6ab0-40cf-8717-3604be42fdf6 Ancestors: CollectionsTests-md.31 test from 0003574: String>>findFirstInString:inSet:startingAt: does not work if character of interest is last character in String, also WideString 0004265: SUnit test WeakIdentityKeyDictionaryTest>>testNoNils fails =================================== Name: Files-md.18 Author: md Time: 15 July 2006, 6:13:59 pm UUID: 50358b69-2700-4d87-a768-b80af0d4bc68 Ancestors: Files-md.17 0004266: AsyncFile class variable Error collides with global class name ====================================== Name: Kernel-md.142 Author: md Time: 15 July 2006, 6:16:17 pm UUID: bc06ef54-5012-4b93-805f-f0ea2a234cd7 Ancestors: Kernel-md.141 0002120: Remove Object>>externalCallFailed from Kernel package 0004138: ScaledDecimals are parsed incorrectly in 3.9b-7036 0003609: Integer >> crossSumBase: aBase\ 0003820: Code or comment in Date>>fromString: incorrect ======================= Name: KernelTests-md.44 Author: md Time: 15 July 2006, 6:17:58 pm UUID: 7fa2d72a-b000-4964-9607-7e92f885ba35 Ancestors: KernelTests-md.43 0004138: ScaledDecimals are parsed incorrectly in 3.9b-7036 0003609: Integer >> crossSumBase: aBase\ 0004260: SUnit test ClassBuilderChangeClassTypeTest>>testClassCreationAndChange fails =========================== Name: Morphic-md.104 Author: md Time: 15 July 2006, 6:19:45 pm UUID: b6fe7c5a-13b0-447f-af8b-a2bce9939684 Ancestors: Morphic-md.103 0003526: DNU when detaching a submenu ============================== Name: Speech-md.9 Author: md Time: 15 July 2006, 6:21:12 pm UUID: d8ae2a9c-d222-4eac-b069-6d31a5c457f9 Ancestors: Speech-md.8 0003506: some _ wrongly converted to _ Name: FFI-md.12 Author: md Time: 15 July 2006, 6:27:16 pm UUID: b53777ee-054a-4055-9730-7b9b92874d8d Ancestors: FFI-ar.11 add Object>>#externalCallFailed Name: System-md.90 Author: md Time: 15 July 2006, 6:30:58 pm UUID: e1372493-ff32-47ca-b091-4685e893552a Ancestors: System-md.89 0003790: Text cursor too bright First part 7044 ------ Name: Compiler-md.51 Author: md Time: 14 July 2006, 4:39:56 pm UUID: be84aa2e-c1fc-4783-886b-42ba38142007 Ancestors: Compiler-lr.50 removed broken tests =================================== Name: Kernel-md.140 Author: md Time: 14 July 2006, 4:41:50 pm UUID: f3d2af43-b5d8-4404-b1a8-fa2c562dee3b Ancestors: Kernel-md.139 0003993: Number does not read '1.0e-10 .txt' from file ==================================== Name: Network-md.32 Author: md Time: 14 July 2006, 4:55 pm UUID: 44e40f51-06fe-4d40-8b7a-33ca8d0157b1 Ancestors: Network-md.31 0003762: (Fast)SocketStream's class comment has not been updated since image inclusion Name: Morphic-md.102 Author: md Time: 14 July 2006, 4:52:08 pm UUID: 1ada4657-9c73-47ec-b9c3-0f68d6c1a8d7 Ancestors: Morphic-md.101, Morphic-md.100 merge with 100 Name: Morphic-md.101 Author: md Time: 14 July 2006, 4:44:01 pm UUID: 55361be6-6ac2-4005-a77c-9b56ee4f02b0 Ancestors: Morphic-sd.99 0003747: TransformMorph crashes when picked up by handMorph after scaleFactor: is called 0003533: [Enh] Targetsighting menu items want to use a target sight as a cursor. ==================================== Name: NetworkTests-md.9 Author: md Time: 14 July 2006, 4:56:20 pm UUID: 8ce5bafc-7d80-4ea8-9730-beeced99ff08 Ancestors: NetworkTests-gk.8 0003589: [BUG][FIX] UUIDPrimitivesTest>>testCreationRandom ==================================== Name: Tools-md.69 Author: md Time: 14 July 2006, 4:58:37 pm UUID: b40e0d66-5103-449d-946d-7cc8bdb4c6c6 Ancestors: Tools-md.68 Fix for FilecontentsBrowser>>browseStream Name: Morphic-md.100 Author: md Time: 14 July 2006, 2:52:06 pm UUID: da4fe617-8ded-40e8-810e-6d60c3a8e551 Ancestors: Morphic-sd.99 0003530: SimpleSwitchMorph does not actually implement switch functionality 0003690: CircleMorph has inimplemented message sends ======================================== Name: MorphicExtras-md.30 Author: md Time: 14 July 2006, 2:57:01 pm UUID: fdd65f8d-6bae-41aa-acaa-505086574ca4 Ancestors: MorphicExtras-sd.29 0003530: SimpleSwitchMorph does not actually implement switch functionality ========================================= Name: MorphicTests-md.6 Author: md Time: 14 July 2006, 3:02:09 pm UUID: 12154457-e61a-4c5c-bc48-828ee1a9d733 Ancestors: MorphicTests-md.5 0003530: SimpleSwitchMorph does not actually implement switch functionality 0003690: CircleMorph has inimplemented message sends 0003699: [BUG][FIX]ClockMorph toggleRoman broken ========================================= Name: Network-md.31 Author: md Time: 14 July 2006, 3:04:24 pm UUID: 427298a6-340d-4916-ad92-1b0303a4dbff Ancestors: Network-sd.30 0003264: (Fast)SocketStream SocketStream>>print: missing ============================================= Name: System-md.89 Author: md Time: 14 July 2006, 3:08:03 pm UUID: 4e78aa2c-2640-44bd-9c19-814b516c15d7 Ancestors: System-md.88 0003647: Smalltalk>>version should not be deprecated - Commenrt for MessageSend 7043 ------ Adds a new preference for the caret-with and changes the default color of the caret http://bugs.impara.de/view.php?id=3790 Name: Graphics-ar.39 Author: ar Time: 13 July 2006, 11:01:44 am UUID: 8493073d-14aa-7643-9a13-574d84f7b2c3 Ancestors: Graphics-ar.38 - include fix from http://bugs.impara.de/view.php?id=3570 + test http://squeaksource.com/ToolBuilder/ToolBuilder-Kernel-ar.16.mcz http://squeaksource.com/ToolBuilder/ToolBuilder-Morphic-ar.19.mcz http://squeaksource.com/ToolBuilder/ToolBuilder-SUnit-ar.11.mcz http://squeaksource.com/ToolBuilder/ToolBuilder-MVC-dtl.12.mcz The main change is that some support for menus has been added. 7042 ------ Smacc preferenceBrowser underscore Name: Kernel-md.135 Author: md Time: 12 July 2006, 10:25:06 pm UUID: 61de8206-a937-4ad9-a401-c82f20946fa0 Ancestors: Kernel-sd.134 fix CompiledMethod>>#= to ignore the last two literals (we may want to extend it to take pragmas into account for equality later...) Name: Files-md.17 Author: md Time: 12 July 2006, 11:10:35 pm UUID: e4f7668b-a5ed-4a46-b9ce-a100fc0985fd Ancestors: Files-md.16 0003723: [FIX] DosFileDirectoryTests>>testFileDirectoryNonExistence Description The change set included with this bug report fixes the test DosFileDirectoryTests>>testFileDirectoryNonExistence Name: MorphicTests-md.5 Author: md Time: 12 July 2006, 11:09:16 pm UUID: f2394cd3-e068-4a37-acfd-0258e0937a9a Ancestors: MorphicTests-stephaneducasse.4 0003566: TileMoprhTest>>testSoundMorph fails; fix attached Description in 3.9a 7029 (and 3.8 6665) TileMoprhTest>>testSoundMorph fails because it expects 'silence' to be one up from 'croak' but the menu provides 'horn'. This just fixes the test. 0003571: EventManager commen Description In 3.9a 7029 (and earlier) EventManagerTest produces one failure because of missing class comments. This cs adds reasonable class comments for EventManager and two related classes as part of the effort to get all tests to run green. 0003729: [FIX] FontTest>>testResetAfterEmphasized Name: Exceptions-sd.8 Author: sd Time: 13 July 2006, 3:15:11 pm UUID: 169d4c2b-c7f8-44bd-8621-ffcf2feb1c97 Ancestors: Exceptions-md.7 0000530: ExceptionTests>>#testTimeout works only some of the time Description When I run this test several times with the SUnit Test Runner it sometimes works and sometimes fails 0003648: explicit ^self in #yourself Description ... when it's important to return self, write it. (from the best practice patterns). yourself would be a nice example to show this in a lecture. 0003628: refactoring for hasInstVarRef Description hasInstVarRef was only defined on the Context, this one moved the implementation from MethodContext to compiledMethod (and just forwards from MethodContext). Additionally, the changeset has a small refactoring to use BlockContext>>#endPC in BlockContext>>#hasInstVarRef 0003529: BlockContextTest tests fail in 7025; fix attached Description In 3.9a 7025 four BlockContextTest tests failed because Jerome Peace's valueSupplyingAnswers: functionality was regressed out of PopUpMenu when icons where added to the menu. I suspect that the reason this wasn't caught is that the change was made to PopUp but the relevant s- unit tests were in BlockContextTest, so I added an test few method for the so called ST80-Menus. There was also one trivial error in a single BlockContextTest which I fixed. Without this fix running the test requires user responses on the four failing tests and several other. 7041 ----- 0004166: Fix for rejected Graphics extensions Description The following extensions to the graphics packages have been proposed and rejected: Color class>>saturatedRandom Form>>dominantColorWithoutTransparent The attached CS provides code such that Kedama works correctly without the rejected extensions. 'From Squeak3.9alpha of 4 July 2005 [latest update: #7035] on 8 July 2006 at 8:35:26 pm'! Change Set: KedamaFix Date: 8 July 2006 Author: Andreas Raab Change set for rejected extensions Color class>>saturatedRandomColor and Form>>dominantColorWithoutTransparent 0004168: Fix for Color's colorPalette: removal Description Color>>colorPaletteForDepth: has been removed to avoid the dependency between Color and NaturalLanguageTranslator. The attached change set provides an implementation in ColorPickerMorph and fixes the existing references. 0004169: StrikeFont>>decodedFromRemoteCanvas: does not belong to Graphics Description StrikeFont>>decodedFromRemoteCanvas: was included in the graphics package. The attached change set puts it under Nebraska, where it belongs. 0004170: Fix for rejected SmallLand extensions of Graphics Description The following proposed extensions of the Graphics package have been rejected: - Color>>iconOrThumbnailOfSize: - Form>>iconOrThumbnailOfSize: - Form>>scaledIntoFormOfSize: The attached change sets provides them as *Morphic extensions. 0003821: addMonths: does not handle leap years correctly 7040 ------ http://bugs.impara.de/view.php?id=3737 0003737: [BUG][FIX] condenseChanges broken with Traits 7039 Name: Compiler-lr.50 Author: lr Time: 9 July 2006, 7:56:43 pm UUID: 47ef9423-0f74-11db-b64a-000a9573eae2 Ancestors: Compiler-lr.49 - fixed more pragma bugs (throw an error for <1> and <foo bar>) Name: Collections-pmm.68 Author: pmm Time: 9 July 2006, 11:53:56 am UUID: 20686bef-7acd-4afc-a610-a4485c82ffb1 Ancestors: Collections-sd.67 changes with respect to: http://bugs.impara.de/view.php?id=1876 - changed Set class >> #new: according to comments of ar, this fixes the hiccup at around 4k elements - removed the progress bar from Set class >> #rehash all sets. This allows to load the changesets and makes in possibly faster (old version waits 2ms for each set when querying the mouse position) Name: Balloon-ar.13 Author: ar Time: 5 April 2006, 12:23:02 am UUID: bdaaa697-27c2-4049-a7e8-13e44a41f929 Ancestors: Balloon-ar.11 - replace underscore assignment with colon-equals Name: Compression-ar.8 Author: ar Time: 5 April 2006, 12:25:07 am UUID: 3c075d58-b256-704c-9870-6998a0c8e9ce Ancestors: Compression-ar.6 - replace underscore assignment with colon-equals Name: FFI-ar.11 Author: ar Time: 8 July 2006, 9:25:14 pm UUID: e964ce57-a9a6-2a4e-9572-685f6b711fba Ancestors: FFI-ar.10 - include non-Mantis fixes Name: Flash-ar.5 Author: ar Time: 8 July 2006, 9:21:53 pm UUID: d7099247-3e7a-d742-921d-8dd73cf0fa07 Ancestors: Flash-ar.4 - include improved image supports Name: Graphics-ar.38 Author: ar Time: 8 July 2006, 8:51:07 pm UUID: 38143d92-893f-ec4f-9882-0e7d947ca710 Ancestors: Graphics-ar.37 - include fix for BitBlt rounding Name: GraphicsTests-ar.9 Author: ar Time: 8 July 2006, 7:08:18 pm UUID: ce2beab2-9473-4d49-8302-b959486e7bf8 Ancestors: GraphicsTests-ar.8 - Integrate http://bugs.impara.de/view.php?id=3804 7038 0003784: String>>unescapePercentsWithTextEncoding: is subtle broken http://bugs.impara.de/view.php?id=3784 0003568: [FIX] Fraction>>asFloat does not answer nearest floating point number 0003564: LargeInteger>>asFloat does not honour IEEE rounding mode 0004078: Misuse of #inject:into: in DependentsArray>>size 0003339: [Fix] Simple button morphs don't reset their color properly when balked. 0003531: [Fix] In 7027 Scriptable button now turns wrong color when balked. 7037 " Name: Kernel-pmm.130 Author: pmm Time: 6 July 2006, 8:59:24 pm UUID: ca825d54-f890-4bcb-be38-4a718f29459b Ancestors: Kernel-al.129 - fix browsing with alt+b for globals like Smalltalk - push #isTrait to Object like #isBehavior Name: Kernel-al.129 Author: al Time: 3 July 2006, 3:58:12 pm UUID: ea1c48cf-2455-4fa5-ad13-75616626321b Ancestors: Kernel-sd.128 - small fix of the ClassBuilder (when redefining a class, its class side traits were lost) Name: Collections-lr.64 Author: lr Time: 7 July 2006, 11:33:53 am UUID: b45346c1-0d9b-11db-a7b8-000a9573eae2 Ancestors: Collections-sd.63 - revert string comparison to the 3.7 behavior Name: CollectionsTests-lr.28 Author: lr Time: 7 July 2006, 11:34:31 am UUID: cac623bf-0d9b-11db-a7b8-000a9573eae2 Ancestors: CollectionsTests-sd.27 - revert string comparison to the 3.7 behavior Name: OB-Standard-al.112 Author: al Time: 7 July 2006, 11:14:56 am UUID: fd3a8638-1e50-45b5-ad4a-f5224054a331 Ancestors: OB-Standard-pmm.111 - manually merged in Alex' enhancement: Enhancement: When you find a class by pressing Alt-f for example, you may have a very long list. Even if you type the exact name of the class. This changeset place as the first choice the class you entered. For instance, try to search for Object. You have a very long list, this change makes Object the first entry of the menu " 7036 - Pragmas fixes - traits fixes, Pakcageinfo fixes 7035 - goran squeakmap fixes 7034 some MC traits fixes New 7033.... to get back the history of the Monticello package!!! Arghhhh Stef Name: System-gk.45 Author: gk Time: 21 October 2005, 11:58:31 am UUID: f8210799-b1de-5f46-b675-8c7b83550503 Ancestors: System-gk.44 Two good refactorings based on work from Bernard Pieber for Mantis #861 in the Network package. These depend on changes in the Network package: - Project class>>fromUrl: - ProjectLoading class>>bestAccessToFileName:andDirectory: This also removes the only sender of Project class>>serverFileFromURL: and we could nuke it, but we leave that decision for the System stewards. :)" "Name: Network-sd.30 Author: sd Time: 29 May 2006, 9:29:37 pm UUID: 6da57032-4ee0-412e-830a-e6169251929e Ancestors: Network-KLC.27, Network-tb.28, Network-tb.29 merge Network-tb-29, tb-28, KLC27 with netweork-md.27 Name: Network-tb.29 Author: tb Time: 24 May 2006, 10:23:58 pm UUID: dfdd5a1c-4494-41da-afff-925a1597b1c3 Ancestors: Network-md.27 Fixes to address http://bugs.impara.de/view.php?id=2107 FileDirectory class>>retrieveMIMEDocument: class non-existent MIMEType class>forExtension: Name: Network-tb.28 Author: tb Time: 24 May 2006, 9:10:03 pm UUID: 75f7e92a-6087-4223-a373-9d6516cd35e1 Ancestors: Network-md.27 Fix for http://bugs.impara.de/view.php?id=2454 - correctly set accent- type in HttpUrl retrieveContents so fetches of .css files succeed Name: Network-KLC.27 Author: KLC Time: 3 April 2006, 11:18 am UUID: b2510551-ffee-4449-94c5-48d8c31bef77 Ancestors: Network-md.26, Network-gk.18, Network-KLC.21, Network-gk.23 This is a merge of: Network-gk.18.mcz Network-KLC.21.mcz Network-gk.23.mcz Combined notes: Network-gk.18: Changes: - Adds #fileName to FileUrl and HierarchicalUrl (with HttpUrl as subclass), ok - I think that is fair even if the last part of the path not necessarily denotes a file. - Adds Url class>>directoryUrl to FileUrl and HierarchicalUrl (with HttpUrl as subclass) which simply returns a new URL with the last element in the path removed. - Adds #absoluteFromFileNameOrUrlString: which then is used from two different places (refactoring). - A fix so that copying a URL does not share the path collection with the original. Not sure exactly if this has caused an actual bug yet, but I agree it is a good idea to not share. - Fix bug in FileUrl>>pathForDirectory which used an explicit pathname delimiter in one place. Duh. - Removed class MswUrl, sorry but this should not be in standard Network package IMHO. There is no use of it in the current image. Non behavior changes: - Refactoring adding Url class>>urlClassForScheme:. - Improved naming in ServerDirectory class>>serverForURL:, no actual code change. - Refines and adds lots of comments here and there. Differences from proposed changes from Bernard Pieber in Mantis #861: - Did not include the proposed HierarchicalPath>>pathForDirectory because it doesn t return a path (like the same method in FileUrl does), instead it returns a URL turned into a String. I changed the few senders that Bernard Pieber (Mantis #861) introduced to directoryUrl asString instead. - Did not add isSuperSwikiUrl, it does not belong there. It is a hack from the start. Network-KLC.19: This is an adaptation of the updated provided by Patrick Mauritz at Mantis ID #0411: Change Set: Enh-ConnectionQueue Date: 16 October 2004 Author: Patrick Mauritz Allow classes to subscribe to the ConnectionQueue via addDependent: to get notifications when new connections are opened. Receivers need to implement update:aSocket for this to work. Thanks to Ken Causey for guidance on IRC. I adapted Patrick s changes to ConnectionQueue>>listenLoop the result being that a check for a ConnectedTimedOut exception is added and a call to self changed on successful connection. Otherwise this should be the same as previously in the image. Network-KLC.20: As reported in Mantis ID #2106 this method included a call to the now non-existent Notification class>deprecated method. I ve replaced that call. With what I think is the appropriate mechanism. The next question is at what point do deprecated methods get removed entirely? Network-KLC.21: This removes 3 methods that had no implementation other than calls to self halt. None of the 3 methods were being used and it didn t appear that they would likely ever be used. Network-gk.23: Fix for Mantis #2119, the simplest one described. Fix for Mantis #1585, moved 3 test methods from OldSocket class to Socket and updated the code so that it works. Doing the above I also fixed an issue in Socket>>waitForDisconnectionFor:, it was wrongly waiting on semaphore instead of self readSemaphore. I also removed an unnecessary call to dataAvailable, unneeded temp var #extraBytes etc. The problem was that loopbackTest randomly hung in this method. Optimized Socket>>discardReceivedData, unneeded call to dataAvailable removed. Better class comments in ProtocolClient, Socket and SocketStream. Improved comment in Socket>>closeAndDestroy:' Name: Monticello-al.303 (to get back ancestry!!!!!!!) Author: al Time: 9 May 2006, 4:54:21 pm UUID: bc893983-a6cd-430c-bf5d-a4d60fdb04b3 Ancestors: Monticello-al.295 - fix loading traits with empty trait composition Fixing traits. Fixing canUnderstand + ivtraitshifting --------------------------------- 0003520: Chronology Tests for hash failed in 7025; fixes included (by Tom Koenig) --------------------------------- 0003518: SkipListTests fails/errors in 7025;SkipListTestsFixes (by Tom Koenig) --------------------------------- 0003373: Float asFraction does not handle gradual underflow --------------------------------- 0003339: [Fix] Simple button morphs don't reset their color properly when balked. --------------------------------- ChangeLog ========== Change Set: LongDelayFixes Date: 4 October 2005 Author: [hidden email] See Mantis bug 1840 and 854. This changeset allows a too-long Delay to fail the relevant primitive and schedules a fake delay to keep the timer system going. I claim this is preferable to disallowing long delays. Also slightly improves handling of rollover and the treatment of multiple delays with the same resumption time" --------------------------------- 3515 - Parser fails in #defineClass: becasue Metaclass doest not understand #category (DF) --------------------------------- 003513: [ENH] ToolBuilderBldEnh-KR (korakurider) ToolBuilder>>(class)build: enable default builder to build objects without opening. It's useful for building authoring prototypes for ToolBuilder-based tools in parts bin. --------------------------------- 0003510: In 3.9a requesting a language change leads to infinite generation of testRunners (korakurider) --------------------------------- 0003504: Float asInteger conversion is inexact... (nicolas cellier) CHANGELOG: -------------------- Compiler: - Small refactoring of MethodNode>>generate and related methods (needed by MethodWrappers) (Diego Fernandez (and Andrian Lienhard (netstyle.ch)) ) - == <integer literal> to = <integer literal> in response to Dan Ingalls' note (fbs) OmniBrowser: (Lukas Renggli of netstyle.ch) - fixes an bug when the browser is opened (Ctrl+B) from non OmniBrowser windows (Debugger, Workspace, ...) - drag and drop methods and categories onto classes -------------- SUnit: 0003489: ClassTestCase >> selectorsTested misses 'special' methods. The method ClassTestCase >> selectorsTested computes the sent messages in a test incorrectly. It looks in the literals array, and so misses the 32 'special' message selectors. The correct code is both shorter and simpler. (Andrew Black) --------------- Morphic: - 0003416: Faster Moprhic invalidation Morph>>invalidRect:from: was testing for wonderland textures (which where stored in a property...). (Thanks to Andrew Black for reporting this). - Better looking menu title bar (from Thorsten Bergman, if I remember correctly) - some method were lost in 7024 (from etoy-customevents-*). They are back now, fixes lots of bugreports on Mantis. --------------- Change Set: ExternalSettingsReorg-tpr Date: 5 April 2006 Author: [hidden email] Reorganise the categories in ExternalSettings to get rid of an explcit '--all--' that can make a real mess of filing out the code -------------- Change Set: KedamaMissing-KR Date: 14 April 2006 Author: Korakurider add missing method for Kedama ---------------- Change Set: KedamaWODepreciated-KR Date: 14 April 2006 Author: Korakurider refactor kedama not to call depreciated methods. note: As showDeprecationWarnings preferece is set to false in Squeakland image, original kedama code run without no problem. ------------------ renamed BlockContext>>endpc to be endOC and fix the one sender (md) ------------------ Change Set: MacRomanFix Date: 5 April 2006 Author: Andreas Raab Raise an error if somebody tries to write wide characters. ---------------------- 0003458: PositionableStream>>copyPreamble:from:at: broken (Andreas Raab) --------- 0003415: Changeset is not updated when renaming a class (Noury Bouraqadi) After renaming a class, the changeset is not updated to refer to the new name. Therefore, one virtually lose recorded changes. The fix also cleans a bit SystemDictionary. It removes method SystemDictionary>>renameClass:from: and uses instead SystemDictionary>>renameClass:as: used by traits ----------- 0003492: Complete the job of changing the way that arrays are printed (Andrew Black) The new array printing logic introduced in 3.9 had some holes, and needed some cleanup. The attached changeset provides it. Tests are included. Specifically, the changeset - marks Fractions as selfEveluating and Booleans as literals - puts Array>>printOn: back in the printing ctaegory, where someone might expect to find it - removes duplicated logic from Collection>>printElementsOn: - adds tests for all of the above. -------------- faster asUTC from Avi Bryant -------------- FixInvisible from Bert Freudenberg -------------- smallland fixes Tests for Heap Tests for SystemChangeNotifications Fix of roel for double event notification when class recategorisation 0003440: UTF8TextConverter>>nextFromStream: silently truncates input 0003375: mcz files saved in 3.9 don't load in 3.8 0003376: mpeg plugin has changes to image to support read from buffer versus file 0000521: [BUG] Button properties Morph need to be fixed or removed. 0002116: Halo rotation of a true type font throws up a debug window. 0001608: [BUG][FIX] Morphic: File Selection Dialog 0003378: Add FloatMathPluginTests class to Tests-VM package/category 0003377: mpeg plugin has changes to image to support dropping bits into an offset into a form so you can write to say Display OmniBrowser: - format non-local methods (e.g from traits) in italic (does also work in images that do not have support for traits) 0001604: [FIX] [BUG] smoothOrSegmentedPhrase-wiz 0001200: Outdent/Indent issue if code contains WideStrings SUnit: Merged in SUnit extensions by Mercap Software. CHANGELOG ============ - 0003215: saving and reloading a string morph with a true type font breaks the font - 0003354: Can't select some colors in PaintBoxColorPicker - 0003355: [BUG] Fresh 3.9a image (build 7020) has text morphs in objects window wrong size - added CollectionsTests from stef - OmniBrowser (fixed by Lukas) - makes it possible to drag meta-classes to new categories - add missing OBClassActor to OBMetaclassNode - fix as published in http://bugs.impara.de/view.php?id=3269 - fix a debugger when selecting omnibrowser as default browser - made the buttons in omnibrowser look like they do in 3.9 - fixes an infinite recursion occuring when a drag&drop operation is started - fix for the SlowRedraw bug (Adrian) - valueWithExit for BlockContex / BlockClosure - added Symbol>>#value: ------- Change Set: SuperSwikiFixKR Date: 20 February 2006 Author: Korakurider A one-method fix to the recent SuperSwikiStr update. ------- Change Set: TTCFont-fix-ascent-jrp Date: 27 March 2006 Author: John Pierce Add ascent instance variable back to the class so method ascent will work again. CHANGELOG ------------------- 7020 reload code added via updatestream from repository 7019 DNUFixScrollBar-md Hack (for now) to not get DNU on pre-debugWindow 7018 Change Set: NebraskaFontSet Date: 17 December 2005 Author: Yoshiki Ohshima Support StrikeFontSet in better way. 7017 Change Set: ja-forms Date: 29 July 2005 Author: Andreas Raab Install the japanese fonts. 7016 Change Set: ProportionalSplitter-honor-boundaries-jrp Date: 21 March 2006 Author: John Pierce Keep splitter from moving outside the natural window boundaries ----------- Change Set: projectViewMenuFix-yo.cs Date: 2 August 2005 Author: Kazuhiro Abe and Yoshiki Ohshima The project name of menu label was incorrect. ------------ Change Set: testTileFix-kfr Date: 24 January 2006 Author: Karl Ramberg When clicking the button to add a new TestTile the ScriptEditor is unresponsive to place the TestTile because the tile never signaled a mouseEnter to the ScriptEditor. This is a fix for that. ------------- Change Set: retainRotationCenter-sw Date: 25 January 2006 Author: Scott Wallace When repainting a sketch, retain any existing rotation-center that may have been specified. -------------- Change Set: undoFixes-sw Date: 2 February 2006 Author: Scott Wallace Implements uindo for 'dismiss via halo', which works even if trash is not being preserved. Make the Undo button in the NavBar dynamically show what action it *would* undo. Makes that button work propery in a redo situation (formerly it had malfunctioned when in redo mode.) Elaborates the wording of the 'undo' item to include identification of the target, for resize and rotate now, as it always has done for move. Fixes some bugs encountered along the way. Fixes the bug that could break undo of a rotation, if an addition or removal of a flex shell was involved. ------------------ Change Set: SuperSwikiString-KR Date: 1 February 2006 Author: Korakurider Current m18n implementation assumes shift_jis encoding is used in wire protocol between SuperSwiki server and client. Then some existing projects encoded in other than shift_jis cause walkback. This patch introduces 'encodingName' for ServerDirectory entry (even in knownServer entry) so that appropriate encoding can be specified for each server. If not specified, encoding is determined based on current locale. ------------------ Change Set: rotCenterHaloHelp-sw Date: 16 February 2006 Author: Scott Wallace Supplements the balloon help for the rotation-center handle with the additional informaiton that the user should hold down the shift key to move it. ------------------- Name: Balloon-mir.11 Author: mir Time: 22 March 2006, 2:14:18 pm (backported from Jan Fietz' changes) FlashFileReader support for more image tags: processDefineBitsJPEG3 (JPEG + alpha channel) processDefineLossless2 --------------------- Name: Flash-mir.2 Author: mir Time: 22 March 2006, 3:04 pm merge with iSqueak flash enh. (was ballon) ----------------------- Name: SUnitGUI-lr.6 Author: lr Time: 22 March 2006, 7:49:27 pm - fixed a couple of issues related to the change-notifications - no more debugger if the system contains classes that are not categorized yet (can happen when loading mcz files) - register for some different change events that were missed - unregister for changes when window closes ------------------- CHANGELOG ------------------- Image is now in 32bit graphics mode by default (faster). --------------- 0003328: Ctrl-E does not look for strings -------------- Change Set: 0479iso3Countries.cs Date: 1 September 2005 Author: John Mcintosh, Michael Rueger Add ISO3 country codes to the iso definitions ------------- Change Set: phraseDupFix-sw Date: 23 October 2005 Author: Scott Wallace Fixes a bug that broke the duplication of phrase tiles that don't reside in viewers ------------- Change Set: suppliesTwoLiner Date: 7 October 2005 Author: Scott Wallace, Michael Rueger changes the supplies flap to show two lines of items by default -------------- 0498MacUnicodeIInterp.cs --------------- Change Set: FontRegistryFix Date: 9 October 2005 Author: Yoshiki Ohshima registry wasn't properly used for TTCFont. ---------------- Change Set: percentEncodingFix Date: 22 September 2005 Author: Korakurider + When Squeak is launched by openning .pr file, passed filename is systemString (shift-jis on windows for instance). That has to be converted to Squeak-native string. + reimplementation of decoding percent-encoded-string has to be modified to support utf8 based encoding. ------------------ 0502nebraskaBuffer-yo.cs ------------------- Change Set: NebraskaTextAndImageFix Date: 6 October 2005 Author: Yoshiki Ohshima * DropShadow + Text + Nebraska wasn't working right. * Form depth = 32 wasn't working right. It isn't right or optimal but better in many ways. ------------------ Change Set: revertStamp-tak Date: 28 November 2005 Author: Takashi Yamamiya Revert stampOnTransformedOwner.cs (it is included in 6580changesFromSql38ToJa) to fix a problem of stamp of rotated object. -------------------- Change Set: nebraskaBuffer2-yo Date: 10 October 2005 Author: Yoshiki Oshima Republished as Squeakland update 0506 to make certain that the code will be present in images that had already loaded SqueakAlpha update 0502testPercentEncoding before it was rescinded. -------------------- Change Set: FixKedamaBundle Date: 23 October 2005 Author: Yoshiki Ohshima KedamaBundle used to create an extra patch. --------------------- Change Set: BOMLangEditor Date: 29 November 2005 Author: Yoshiki Ohshima Fix BOM problem in Language Editor. -------------------- Change Set: ProjectManifest Date: 29 November 2005 Author: Yoshiki Ohshima Put a 'manifest' file that stores some additional infomation about the project. ---------------------- Change Set: macosxJaMarks Date: 22 November 2005 Author: Tetsuya Hayashi Fixes Japanese input method related problem. It dealsd with special case involved in unicode conversion. ----------------------- Change Set: setupLanguageSpec-tak Date: 28 November 2005 Author: Takashi Yamamiya - Add language specific settings. - Remove setup scripts for old Japanese image. --------------------- Change Set: variableCategory-ka Date: 29 November 2005 Author: Kazuhiro Abe Patch for 'variable category doesn't apper when it's needed'. It occurs because missing 'translated' message after 'ScriptingSystem nameForInstanceVariablesCategory' in StandardViewer>>likelyCategoryToShow. ------------------------ Change Set: Use32BitPatch Date: 30 November 2005 Author: Yoshiki Ohshima ------------------------- Change Set: TextAlignEmphMenu-KR Date: 5 December 2005 Author: Korakurider fix problem that current alignment is shown wrongly on emphasis/ alignment Menu for TextMorph ------------------------ Change Set: colorPickerFix-KR Date: 14 December 2005 Author: Korakurider Get the alpha setting on a morph's property sheet's fill-color colorpicker set properly. ------------------------------ Change Set: shouldntBeRounded-ka Date: 4 December 2005 Author: Kazuhiro Abe Some morph, such as an Ellipse or a Polygon, shouldn't have rounded corners. They drew fake corners. ------------------------------------ Change Set: FixAssignmentSuffix Date: 15 December 2005 Author: Takashi Yamamiya Fixed duplicated assignment arrows in old project. To load old project which is made in 3.2 era, assignment arrows (_) are duplicated. The reason is sometimes symbols and strings are confused in #translatedWordingFor:. ---------------------------- Change Set: fixCharConv128-255 Date: 5 November 2005 Author: Takashi Yamamiya Some characters could not be converted from Squeak to OS encoding. To reproduce: - Open workspace. - Input MULTIPLICATION SIGN character with your IME. - select and copy a few times for copying to the system paste buffer. - You would see an error. Reason: A charactor which value is 128-255 should be converted. e.g. 00D7 MULTIPLICATION SIGN 00F7 DIVISION SIGN see http://www.unicode.org/charts/PDF/U0080.pdf ------------------------- Change Set: fixManifest Date: 16 December 2005 Author: Korakurider fix and enhance the facility to save manifest of project: + Original implementation use constant filename 'MANIFEST'. That behavior causes conflict of filename on server(Swiki etc). This patch use basename of project for file name for manifest + save latest update number of SmalltalkImage in manifest (enhancement) ---------------------------- 0528FixAPartOfBobConv.cs CHANGELOG =========== updated to latest OmniBrowser ------- Change Set: AddLineToTTCFont-yo Date: 1 August 2005 Author: Yoshiki Ohshima call #addLined on demand. It should be handled more statically, but I have a feeling that writing out TTCFont with LinedTTC in their derivative may not be handled very cleverly. -------------- Change Set: TTCRegistry-yo Date: 1 August 2005 Author: Yoshiki Ohshima A class variable (Registry) was missing from the class definition of TTCFont. This change set re-instate that, and register everything in postscript. -------------- Change Set: ttcNewSizeFix Date: 3 August 2005 Author: Yoshiki Ohshima Fallback font handling for new sized font has a bug. The main problem is that FixedFaceFont doesn't have the corresponding TextStyle. I could add it, but so far I gave nil to some implicit semantics in the methods in this change set. ------------------ Change Set: searchPaneFix-sw Date: 3 August 2005 Author: Scott Wallace Make viewer search panes clip their submorphs, thus avoiding an ugly green artifact that had crept in. (Thanks to Andreas for supplying the fix.) ------------------- Change Set: AddTranslationNov30 Date: 30 November 2005 Author: Yoshiki Ohshima Adds #translated to the ? halo of buttons. ------------------- Change Set: fixFileList2-bf Date: 5 December 2005 Author: Bert Freudenberg Fix image opening in FileList2 -------------------- 0446ProjLoadFixes.cs --------------------- Change Set: joystickEnh-mir-bf Date: 5 September 2005 Author: Michael Rueger, Bert Freudenberg Make joystick buttons available in Etoys. --------------------- Change Set: URLEncodingAttempt Date: 29 July 2005 Author: Yoshiki Ohshima An attempt to support more characters in encodeForHTTP. However, this introduces another backward compatibility problem. The default behavior of #encodeForHTTP uses UTF-8 as the byte sequence interpretation, but it was Latin-1 for older implementation. To get the backward compatible behavior, the senders of #encodeForHTTP should use an expression: aString encodeForHTTPWithTextEncoding: 'latin1'. instead. This changeset suppress the trivial error, but may cause the problem when somebody tries to load a project from older image with latin-1 characters in its name. ------------------ 0448loadNihongo4Projects.cs ------------------ Change Set: worldPreferences-sw Date: 3 August 2005 Author: Scott Wallace Don't show the Preferences category in the world's viewer in eToyFriendly mode. When *not* in eToyFriendly mode, do show the category, to which othe alternatives have now been added. ------------ Change Set: vectorError-sw Date: 3 August 2005 Author: Scott Wallace Circumvents error condition when clicking on the menu icon of some of the vector-vocabulary phrases. ------------- Change Set: EditableWatcher Date: 2 August 2005 Author: Takashi Yamamiya This fix allows you to edit variables with simple watchers ------------- Change Set: playfieldPrefBug-sw Date: 12 August 2005 Author: Scott Wallace Fixes the bug that the Preferences category could show up in the viewer of a non-world. ----------- Change Set: KedamaFixesAug Date: 15 August 2005 Author: Yoshiki Ohshima some fixes for a few known bugs. ------------ Change Set: KedamaFixesAug2 Date: 17 August 2005 Author: Yoshiki Ohshima suppress the dup halo for turtle exemplars, and make the forms in patches behaves better. ------------ Change Set: symbolTranslationFix Date: 18 August 2005 Author: Yoshiki Ohshima Fix a few more problems with language transition. ------------- Change Set: KedamaFixesAug3 Date: 19 August 2005 Author: Yoshiki Ohshima The backup code for a primitive was wrong. It shows wrong boundary behavior when the plugin is not present. -------------- Change Set: fixObsDefault Date: 1 September 2005 Author: Michael Rueger Removes obsolete default messages and references still using them. ------------ CHANGELOG -------------------- 7012 0003319: In 7011 asking a morph for a viewer gets a DNU due to missing methods. ------------------- Refactoring of Class|Trait>>category lookup. Improves category lookup by a factor of 10 and for example loading of a Monticello package by a factor of 2. (Adrian Lienhard) 7013 Change Set: nihongo7ModalWindows Date: 22 July 2005 Author: Takashi Yamamiya and Kazuhiro Abe Make some UI elements modal so that the user don't see multiple project saving dialog, etc. ------------- Change Set: nihongo7BitBlt Date: 22 July 2005 Author: Yoshiki Ohshima Prevents simple crash with errorneous numbers in geometory. ----------- Change Set: nihongo7WatchersAndReadouts Date: 22 July 2005 Author: Takashi Yamamiya and Yoshiki Ohshima The proposed improvement for watchers and readouts. The areas are: * Simple watcher bahves better with Alt-Click. * Allows type-in of non-ascii number strings. * Should handle ok with translated symbols in symbol tiles. ---------------- Change Set: nihongo7SimpleButtons Date: 22 July 2005 Author: Takashi Yamamiya Hide more buttons in the nav-bar. Controlled by a preference called showAdvancedNavigatorButtons. ---------------- Change Set: ExtNamesStay-tk Date: 27 July 2005 Author: Ted Kaehler When an object is being brought in from a saved project, if its name is already in use, make a new unique internal name (as before). Set the external name in the costume to its old name. Internal name is the key for it in References. ---------------- Change Set: SketchFix Date: 28 July 2005 Author: Andreas Raab Restore two missing methods from SketchMorph. --------------- Change Set: undo-kedama Date: 17 March 2006 Author: Michael Rueger Undoes two squeakland changes that broke etoys ----------- Change Set: polygonFix-tak Date: 10 February 2006 Author: Takashi Yamamiya Fixes a bug in generic handling of cached morph state, which manifested itself in a variety of ways involving PolygonMorphs, such as the fact that a drop of a Star always required \ a second click. Reverts Morph>>releaseCachedState so that it no longer resets formerOwner, formerPosition, and undoGrabCommand, thus making #releaseCachedState once again safe to call while drag is underway, The resetting of formerOwner, etc. is moved to #prepareToBeSaved ---------- 0003295: In 7006 MorphicMenuItem>>keystroke: code versions reveal a conflict. CHANGELOG =========== 7009 Change Set: kedama2.7-yo Date: 9 May 2005 Author: Yoshiki Ohshima ------------------------------------------- Change Set: flexFix-sw Date: 30 March 2005 Author: Scott Wallace Fix for Mantis bug 993: 'rotating morph changes z-order.' This is a bug that dates to the original implementaiton of 'flexing' (generic Morph rotation) in early 1998. After this fix, when a flex shell is added, we now remember the z-position of the original morph and ascribe that same z-position to the TransformationMorph that replaces it. This update also fixes the dual of that bug, which is that when a TransformationMorph decides it's no longer needed, and gets itself replaced by the morph it formerly had been governing, the unrotated morph had not been getting inserted in the z- ordering in the same place. --------------------------------------------- Change Set: noScritch-sw Date: 6 April 2005 Author: Scott Wallace The Scritch sound is supressed ---------------------------------------------- Change Set: dupFixes-sw Date: 19 April 2005 Author: Scott Wallace When the user chooses 'duplicate' from the halo-actions submenu of the halo menu of a rotated object, the result had formerly been an unrotated object that was not known to the allExtantPlayers list of the project's presenter, so would not come up ticking, etc. This update fixes those glitches. Also: When you dismiss a morph via its halo, its player gets removed from the project's allExtantPlayers list -- an optimization. ----------------------------- Change Set: selectionSibFix-sw Date: 19 April 2005 Author: Scott Wallace Fix for Mantis bug 981: Making a sibling of SelectionMorph causes error. Version 4 of this fileout also fixes bugs involving use of the 'duplicate' feature from the 'halo actions' submenu of a SelectionMorph -- such duplicates were flawed in a couple of way. ------------------------------- Change Set: tickingPainting-sw Date: 30 April 2005 Author: Scott Wallace Two responses to the concern about ticking-while-painting: (1) Adds the #keepTickingWhilePainting preference; when true, scripts will continue to tick while the user is in a painting session. (2) If this new preference is *false*, scripts that had been ticking when painting was initiated now resume ticking once painting is done. ------------ Change Set: acceptPlayerTile-sw Date: 1 May 2005 Author: Scott Wallace Implements a short-cut feature requested by Alan (or rather 'remembered' by him ;-) as follows: a player-valued tile can now be dropped onto a graphic- valued pad, whereup it expands into appropriate retrieval tiles for obtaining a graphic from the player. --------------- Change Set: unCustoming-sw Date: 17 June 2005 Author: Scott Wallace When the allowEtoyUserCustomEvents preference is false, refrain from showing the triggeringObject item in the Viewer and from offering the CustomEvents type in type-list popups. Also now makes the allowEtoyUserCustomEvents preference behave like a normal preference, i.e. independently of eToyFriendly, but sets it appropriately whenever eToyFriendly is changed -- after which it *can* subsequently be changed manually if desired, thus making it possible to set things up to operate with eToyFriendly on yet offering those CustomEvents. ------------------------ Change Set: vectorAccess-sw Date: 9 July 2005 Author: Scott Wallace Gives easier access to the command for turning on the vector vocabulary. Makes it possible to set an image up such that the vector vocbulary is always in use by default. The primary access to this control is now via the World's Viewer, which now gets a 'preferences' category, into which is also put a #dropProducesWatcher control. 7010 Change Set: soundUpdate Date: 6 April 2005 Author: Michael Rueger Loads an updates set of sounds 7011 0003312: Can't delete projects (ImageSegment>>rootsIncludingPlayers still refers to Environment) ------------------------- Change Set: watcherTypeChgFix-sw Date: 25 July 2005 Author: Scott Wallace Makes the new watchers launched to replace existing ones upon a type- change to a variable conform to the new protocol for creating watchers; also fixes the bug that simple watchers were being replaced by fancy ones after a type change. --------------------------- Change Set: StringEqual2 Date: 15 December 2005 Author: Yoshiki Ohshima 'abc' = 'abc' asWideString wasn't returning true after the String refactoring. Supposedly, primitiveCompareString should fail when the argument or receiver isn't a byte object, but currently the primitive just traverses the internal of the word object. The right fix would be to fix the primitive, but so far the Squeak side code should cope with it. ------------------------------ Change Set: tileFix-sw Date: 18 July 2005 Author: Scott Wallace The tear-off-a-tile halo handle could sometimes tear off a bogus tile, if a non-Sketch object was rotated before a player was allocated to it. This update fixes that. Thanks to Dave Briccetti for pointing out this bug. -------- Change Set: FixMorphDelete Date: 28 July 2005 Author: Andreas Raab Always #dismissViaHalo even if #preserveTrash is true. -------- Change Set: ProjectViewDismiss Date: 1 August 2005 Author: Yoshiki Ohshima Match the wording when a ProjectViewMorph is dismissed via halo. ------- Change Set: turnTowardFix-sw Date: 29 July 2005 Author: Scott Wallace As per suggestion from Randy Heiland, Squeakland mailing list 7/29/05, makes 'turn toward' always carry out the turn (unless the positions of the two objects coincide), whether or not the objects overlap. -------- 0461MophDissmissal.cs -------- Change Set: LocaleRefactoring Date: 8 August 2005 Author: Takashi Yamamiya ------- Change Set: localeEnabler Date: 31 August 2005 Author: Michael Rueger Enables the use of the language setting through the locale plugin. If no plugin is found, the current locale is used. The preference useLocale turns the automatic settings on or off. ----------- Change Set: smWorkaround-sw Date: 16 March 2006 Author: Scott Wallace, Michael Rueger Adapted for 3.9. A change to allow SqueakMap to load packages successfully into images whose version numbers are less than 5302, which is the Squeak.org update level at which the convention about where the system changes were stored changed. Goran's version of this code (7/13/04) assumed that any version number less than 5302 indicated an ancient Squeak, whereas in practice nowadays it is much more likely to indicate a recent Squeak that has a different formal version name, such as 'Tweak1.2' or Squeakland3.8', and which has its own sequence of update numbers. This workaround was prompted a need to load Monticello into the Squeakland image. Mantis-3061-SnapPop, fix by Jerome Peace (wiz) Mantis-3075-MixedCurves, fix by Jerome Peace (wiz) Mantis-2589-ButtonTargets, fix by Jerome Peace (wiz) Mantis-2709-CurveBoundsTRFix, fix by Jerome Peace (wiz) Mantis-2496-Halos-Menus, fix by Jerome Peace (wiz) 0003294: KeyedSet>>add: cand send #errorKeyAlreadyExists: which is not implemented 0003301: FileList has two install buttons 0003303: Browser --> open does allways open Browser (not SystemBrowser default) ChangeLog: ----------------- 0003249: Mac VM pops up debugger on every launch (InternetConfiguration broken?) 0003254: faster #binding 0003258: [ENH] Entry in Fonts menu for easily getting all big fonts 0003150: In 7002 Asking for priorVersions causes browsing some primitive methods to get a DNU 0003268: NumberTest>>testPrintShowingDecimalPlaces: broken CHANGELOG -------------------- - fixed #binding - 0003249: Mac VM pops up debugger on every launch (InternetConfiguration broken?) - 0003246: In 7005 Getting objects from Object morph not behaving as expected. - 0003245: 7005: ProcessBrowser DNU CHANGELOG -------------------- 0001613: [ENH] truncatedChanges-ls Date: 4 September 2003 Author: Lex Spoon Handle truncated changes files more gracefully. The problem actually can happen, and has been observed at Georgia Tech, e.g. if someone downloads a file but does not complete the download. The current behavior is that pink debuggers pop up; this is extremely confusing for new users when it happens in places like the Browser. The solution in this changeset is to fall back on the decompiler. A more thorough solution should probably trigger a systematic invalidation of all source pointers past the end of the changes file, because as time goes on the changes file will eventually grow large enough to cover the lost code and the routine will start returning random method code for the invalid source pointers. ---------------- 0001354: MIDI port name is garbled in selection dialog Change Set: AddTransMIDIPort Date: 31 May 2005 Author: Tooru Nosse fix for query of MIDI portName with SJIS name enable translation for midi port selection ---------------- 0002341: MorphicEventDispatcher minor refactoring 0001284: Smalltalk condenseSources leads to UndefinedObject(Object)>>error: 0003180: Selector requester does not strip blanks tests from 0001798: #critical:ifError: on Semaphore does not always signal the semaphore test from 0003133: [BUG][FIX] Float>>#asIEEE32BitWord and #fromIEEE32Bit: ---------------- Change Set: dupAllCtrlAltKeysPref-dew Date: 13 December 2004 Author: Doug Way Adds a new preference duplicateAllCtrlAndAltkeys to duplicate all Ctrl and Alt modifier key commands, which causes all Ctrl key commands to behave the same as Alt-key commands. The original Ctrl (a.k.a. 'shift') key commands are still available via Ctrl-Shift and Alt-Shift. This differs from the existing ctrlAlt prefs which only duplicate/ swap the eight editing keys. Also includes code to prevent these other two conflicting prefs from being activated at the same time. The three related prefs (swapCtrlAndAltKeys, duplicateCtrlAndAltKeys, duplicateAllCtrlAndAltKeys) should really be made into a radio button preferences set sometime in the future (like the Halo look preferences), since only one can be turned on at a time. This would eliminate the need for the #changed methods. See 'modifier' discussion on squeak-dev from 12/3/2004 for more details on why this was provided. ----------- - postCopy for MethodProperties - Preferences panel size - remove Preference #ansiAssignmentOperatorWhenPrettyPrinting - remove Preference #resizeOnAllSides, #twoSidedPoohTextures - set selector when evaluating a #Doit. Fixes MessageTally tallySends: - cleanups in Parser: remove old Properties code - cleanups in Encoder: use #binding, not #associationFor:. - remove the last new initialize cases - Fix ClassTestCase to not turn off Deprecation - Fix Tests for ObjectsAsMethods - Refactor/deprecate old environment methods in SystemDictionary Graphics: ----------- - 0002449: [Enh] Adds a target-sighting cursor to the menagerie of cursors. - 0002450: [Fix] Form fromUser can create a form w/ zero area if user specifies a corner to upper left of origin. - bullet-proof ImageReadWriter class>>understandsImageFormat: - 0002684: [Fix] In 6719 rotateBy: #right or #left returns wrong aspect ratio. - 0003060: Graphics-md.27: additions from Connectors / merge with image 3.9a7004 ------------- fasterMorpic: optimized extension/properties code for number of sends ------- Change Set: AutoDeselectToolFixes-wiz Author: (wiz) Jerome Peace For doubleClick to work in lists autoDeselect must be explicitly disabled. There are four places where This needs to be done on the tool end of thing. ------- 0003108: [quickFix] In 7002 selecting connector catagory or alpha p catagory gets a debug box. 0002500: In6713 Sometimes a Legitimate request generates DNU PseudoClass>>isTraits 0002991: [Fix] Workspaces initial extent is too large for most uses. 0002151: Max number of literals checked in MethodNode instead of CompiledMethod 0003128: [Fix] In 7002 Using browser buttons to ask for prettyprinting gets DNU Compiler>>asText 0001048: [ENH] SystemNavigation>>allImplementorsOf:localTo: 0002779: 3.9a-6721 theme... button on Services Browser gives 'key not found' walkback 3.9a7003 ------------- Change Set: decorateFix-bf Author: Bert Freudenberg Bullet-proof browser against removed class --------- 0001913: Make all implementors of #nextPut: to return the argument (instead of self) Changed some implementors of #nextPut: to return the argument (and not return self). This to be consistent with primitiveNextPut and all other implementors of #nextPut: kwl: As dicusses with Ken and Craig on #squeak IRC. -------- 0001733: [ENH][FIX] String-upToDep-huma Really deprecates SequenceableCollection>>#upTo: to #copyUpTo: -------- PlusTools do not register in Filelist when not active ------- 0003049: fast window resizing needed 0001596: [BUG][FIX] lurking signals in EventSensor ------- - PolygonMorph.st from Connectors - cleanups for preference removal ------ removed preferences #allowCelesteTell #useFileList2. #enableInternetConfig #browserNagIfNoClassComment #alternativeWindowLook #alternativeScrollbarLook #inboardScrollbars #fastSplitterResize ChangeLog ---------------- - updated SUnit packge (refactorings and a small bugfix) - SequenceableCollection.st from Connectors - BlockContext.st from Connectors - MessageSend.st from Connectors - TwoWayScrollPane.st from Connectors - Object.st from Connectors - CircleMorph.st from Connectors - TextMorph.st from Connectors - TileMorph.st from Connectors - SketchMorph.st from Connectors - StringMorphEditor.st from Connectors - EllipseMorph.st from Connectors - TTSampleStringMorph.st from Connectors - TTSampleFontMorph.st from Connectors - HandMorph.st from Connectors - SketchMorph.mir.1.cs from SqueakLand - TextMorph.mir.1.cs from SqueakLand ---------------- Change Set: PersMenuFix-wiz Author: (wiz) Jerome Peace The Yellow button personal menu item 'about this system' throws up a DNU. The message is being sent to SmallTalk (the dictionary) instead of SmallTalkImage current. As I was here anyway I also took the time to remove the only reference to the isFlagship preference and the hardcoded preference method itself as per discussions with sw in mantis #2690. ---------------- Change Set: watcherTypeChange-sw Date: 24 March 2005 Author: Scott Wallace If the user changes the type of a Variable, watchers looking at that variable are now fixed up. Any existing Watcher for the variable gets replaced by a labeled watcher appropriate for the new type. This is a fix for Squeakland Mantis bug #1001. ---------------- - remove Preferences: #selectionsMayShrink #warningForMacOSFileNameLength #celesteHasStatusPane. #celesteShowsAttachmentsFlag #autoAccessors #classicNewMorphMenu #morphicProgressStyle #showLinesInHierarchyViews #testRunnerShowAbstractClasses ---------------- - fix BlockContext>>#decompile to not reference Decompiler - add CompiledMethod#decompileWithTemps (logic was in Tools... sigh!) - ContextPart>>sourceCode simplified. - 0002922: Debugger shows only pretty printed code - #decompileString for BlockNode and BlockContext - Class class>>#rootsOfTheWorld (from RB/VW) - More changes from NewCompiler package - refactorings and cleanups related to decompiling in the Tools - complete OB - EToys-jmv.7.mcz, MorphicExtras-jmv.12.mcz and Morphic-jmv.70.mcz to the inbox. This version includes: - Mantis-0504-CurvierMorph, fix by Jerome Peace (wiz) - Resizing SystemWindows on any corner, fix by Juan Vuletich (jmv) and Robin (rl) - Removal CDScreenShotMorph, indicated by Jerome Peace (wiz) Reviewed by Juan Vuletich (jmv) CHANGELOG =========== 3.9a6723 ------------- - load Lukas' MethodPropertiesPragmas changeset 3.9a6724 ------------- - 0002498: In Sq6713 Strange behaviours when asking for class ref. - re-load 6723 via MC 3.9a6725 -------------- - 0002345: Cannot open Workspace from world menu in MVC - fixes to MC tests by Adrian - CompiledMethod>>decompileTree deprecated - String>>numArgs: for building selectors easily - cleaned up unused old property code in parser - first cut at a fast #who (see Changelog Kernel-md.92) - cleanups in CompiledMethod (see Changelog Kernel-md.92) - some first refactorings to not call #who Name: Kernel-md.92 Author: md Time: 16 February 2006, 2:18:22 pm UUID: 938634c4-fd4e-49ca-8921-4be99275f270 Ancestors: Kernel-md.91 - remove MethodPropertie>>#propDict (leftover) - Pragma>>#selector fixed to no call #who - First cut at "fast #who" - CompiledMethod>>#methodClass now gets class from classbinging in the last literal - cm>>#selector searches this class - searchForClass/seachForSelector search like old #who - who fixed to use methodClass/selector. does return #(unkown unkown) when not installed. deprecated. - isInstalled: tests if method is installed. - methodRef fixed to not use #who - #decompileTree deprecated - #decompile now sets class / selector - #decompileString changed to call #decompile - Behavior>>#binding returns the association form systemdict - Behavior>>#basicAddSelector:withMethod sets class in method on installation. 3.9a6726 -------------- Change Set: BehaviorHashEnh v1.2 Date: 22 June 2004, 16.02.2006 Author: Stephan Rudlof, md md: added a line to the poscript to uncompactify the MethodProperties class. We want to add an instVar for the selector. Improves the default Object>>hash for Behaviors by installing Behavior>>hash. String>>hash has been changed a little to avoid infinite recursion (without changing its semantics). All is done in the postscript. 3.9a6727 -------------- -Put back in the simplified Decompiler>>#decompileBlock: - deprecated #who - CompiledMethod>>#methodNode now sets Class and Selector for non- installed methods to Object/ CompiledMethod>>#defaultSelector - refactored ContextPart to not call #who - refactored ProcessBrowserPlus to not use #who - refactored ProcessBrowser to not use #who - simplified CompiledMethod>>#defaultSelector - String>>#hash now uses identityHasch as initial hash (needed for Behavior>>#hash) - moved numArgs: to Symbol, faster - add Behavior hash - simplify CompiledMethod: #defaultSelector, #equivalentTo:, #methodNode - add CompiledMethod>>#selector: - add iVarselector to MethodProperties, accessors - make MethodProperties compact again." 3.9a6728 -------------- deprecate behavior>>#selectorAtMethod:setClass: - deprecate ContextPart>>#mclass (use methodClass, like in CompiledMethod) - deprecate ContextPart>>#methodSelector (use #selector, like in CompiledMethod) - deprecate MethodContext>>#answer: (comment indicates 'will soon be removed' - refactor to not call deprecated methods. - Fast #who part 2: Selector from compiledMethod. - Forward from ContextPart to method (selector, decompile.) no need to search anymore - installed methods now get selector/class set. 3.9a6729 -------------- - fixed override of Services to not revert Traits browser change - 0002868: Copying the text of a list morph via the halo menu does return empty - fix cm>>selector, cm>>methodclass to test for new method layout, if not --> return nil Traits: Change log: - fixed marker methods that were broken because of new default literals in CompiledMethod - removed obsolete requries algorithm tests - prefixed all requires performance tests to exclude them from normal test runs - fixed issue [0002814] (targetTraits is shadowed in Change Set: TestRunnerEnh Date: 18 February 2006 Author: Alexandre Bergel With this fix, the TestRunner now update its list of classes and categories accordingly to system changes. It uses the SystemChangeNotifier. Name: CollectionsTests-fbs.16 Author: fbs Time: 14 February 2006, 5:44:42 pm UUID: bfeee468-b072-1048-9409-ed02f938ff47 Ancestors: CollectionsTests-md.14 This version converts == <integer literal> to = <integer literal> in response to Dan Ingalls' ChangeLog: ----------------- - remove Deprecated Methods from 3.8 - moved all new deprecated method to *39Deprecated - 0001824: Add methodNode>>generate, fix senders to use it - 0002513: arcTan: returns angle in strange intervale - 0002118: Integer class >> #primesUpTo: - 0001109: ScrollPane code/comment mismatch - 0001734: [ENH] remove deprecated Chronology methods in 3.9 [cd][su] [sm] - 0002570: [Fix] When Color pret†y printing it is hard to read the light tan literals against the white background. - start of a ReleaseTest (testing for Undeclareds etc...) - BorderedMoph.st from Connectors - Change Set: versionCats-bf Author: Bert Freudenberg Show method categories in version listing. Particularily useful when checking overrides, where the only thing changed is the categorization. - Change Set: systemSupportFixes-bf Author: Bert Freudenberg Fixes to SmalltalkImage - rename readDocumentFile to recordStartupStamp, which is what it does nowadays - changesName is full path now, like imageName - derive full name for new image or changes from image path primitive, rather than default directory (in all regular cases they are identical) - Change Set: AnnotationMorphFix-wiz Author: (wiz) Jerome Peace Made some style changes to the annotation preference widget. Two things. put the extra button in a better place. Made a more harmonious aspect ratio for the panels. 6720 - contains a fix related to tools in context of traits - fixing underscores (_ -> :=) in '38Deprecated' 'Tests' 'SUnit' 'SUnitGUI' '39Deprecated' 'CollectionsTests' 'Compression' 'Files' 'FlexibleVocabularies' 'Monticello' 'MonticelloConfigurations' 'Movies' 'Nebraska' 'PackageInfo' 'PreferenceBrowser' 'Protocols' 'ToolBuilder-Kernel' 'StarSqueak' 'Sound' 'VersionNumber' 'Tools' 'ToolBuilder-SUnit' 'ToolBuilder-MVC' 'Services-Base' 'SmaCC' 'SMLoader' 'SMBase' 'ToolBuilder-Morphic' 'Speech' 'ReleaseBuilder' 6719 - contains TextAnchor fixes of lukas - fix underscores in some packages (more to come) this is why this is slow to load - French mac os x emergency halt - Extracted ReleaseBuilder from SystemSupport 6718 includes call 'TheWorldMenu removeObsolete' to cleanout obsolete ref. ----- 0002542: Morphic testing failure, Morph>>overlapsShadowFormbounds ----- revert Decompiler>>#decompileBlock: ----- ServiceAction>>perform:orSendTo: for romain ----- Adding andreas packages the latest version of - http://source.squeakfoundation.org/Balloon/Balloon-ar.9.mcz - http://source.squeakfoundation.org/Balloon/Flash-ar.1.mcz - http://source.squeakfoundation.org/Balloon/TrueType-ar.1.mcz - http://source.squeakfoundation.org/Balloon/Morphic-Balloon-ar.1.mcz - http://source.squeakfoundation.org/Balloon/Morphic-TrueType-ar.1.mcz - http://source.squeakfoundation.org/Graphics/Graphics-ar.25.mcz ChangeLog: ---------------- Some Changes form the ClosureCompiler ---- Name: SUnitGUI-lr.4 - changed the label from 'Essential Test Runner' to 'Test Runner', since there is no other anymore this makes no sense to distinguish - renamed the button 'Run All' to 'Run Selected' to avoid confusion ----- Change Set: fileinUndeclared-bf Date: 12 January 2006 Author: Bert Freudenberg When filing in, do not ask whether it's okay to moveclass var to Undeclared. Just move it. ----- Change Set: DebuggerMVCSchedulingFix-dtl Date: 12 December 2005 Author: David T. Lewis Mantis bug 0002359: Debugger in MVC opens in next available Morphic world ----- Name: CollectionsTests-zz.11 Author: zz Add more tests to several Collection classes : Array, Association, Dictionary, Heap, Interval, LinkedList, OrderedCollection, SequenceableCollection, SortedCollection All tests are green ;-) ----- added haltOnce (with halt on count, inspect at count, and inspect until count) ----- removed emptyCheck from OrderedCollection removeLast, removeFirst Here's the changelog: Change Set: isSourceFileSuffix-M1407 Date: 15 September 2005 Author: [hidden email] A trivial change to FileStream class>isSourceFileSuffix: to remove the spurious seeming inclusion of '*' as a legitimate source file suffix. Seems to break nothing. ---- Change Set: selectorsWithArgs-md Date: 2 January 2006 Author: Marcus Denker Make Behavior>>#selectorsWithArgs: use symbol>>#numArgs. Much simpler: one line. And it fixes a bug. ---- Change Set: removeUndefTest Date: 14 January 2006 Author: Marcus Denker removes UndefinedObject>>#test, which was a test method for alternate syntax ----- 0002373: ParagraphEditor >> #makeCapitalized: comment is not correct ----- Change Set: SysWindowClean01-wiz Date: 18 January 2006 Author: (wiz) Jerome Peace wiz 1/18/2006 15:34 Ok. I think that does it. Removed unused preference and SystemWindow methods that use it Mantis # 2532. ------ Change Set: decompileBlock-md Date: 24 November 2005 Author: Marcus Denker A simplification for BlockContext>>decompileBlock: The Method got the tempNames by compiling the source of the method. This is already implemented in methodNode tempNames... thus not needed here. ------ remove emptycheck for Heap>>fist and SequencableCollection #first, #last, #middle ------ 0002514: [FIX] condenseSources (was: Re: [Q] Removing changes file content. ------ Delay startTimerInterruptWatcher to fix 0002379: BlockContext>>decompile broken, ProcessBrowser does not work ------ 0002377: Cannot rename a project from its window menu 0002497: MNU when changing title on Morphic Project 0002415: Error after one presses ESC key two times. 0002145: KlattFrame: ZeroDivide exception when the value of ro, rk or ra is 0 ----- Change Set: FileListCodeDNUfix-efc Author: Eddie Cottongim Apparently Object got (errantly?) registered with FileList as a file reader. This was throwing some DNUs. This simple doit unregisters Object and appears to solve the problem. ----- Change Set: RemoveAlternateSynPref ----- Change Set: PlusToolsAlternateSyntaxFix Date: 14 January 2006 Author: Marcus Denker The horrible alternateSyntax hack has been removed in 3.9. This fixes three methods in the plustools to not use definitionST80: but definitionST80 ------ - move Behavior>>#scopeHas:ifTrue: to *39Deprecated, fix the one sender - JMMRemoveExtraIndexCheck: faster second...8th - fix open file directly - WindowColorRegistry (Hernán Tylim) - Fix to open real Workspace the new update contains - debugger fixes (asked by diego) and done by marcus - Services integration (I asked romain to send me the RB engine services so soon we will be able to have access to all the RB operations from all the browsers). I got some merge problems with MC but it seems to be resolved. - A nice cool and fast TestRunner Thanks lukas this is great (integrated by marcus) Marcus removed the old one from the Tools package. new update + new file on the server: - adrian fixes for traits. now if you use the standardtoolset you will not see the traits if you do not need them. - Tim VMMaker fixes and cleanups VMM38-64bit-imageUpdates.1.cs - VMM38-gc-instrument-image.1.cs - usefull cleanups RemoveLeftoverVMMbits-38b4.1.cs - cleans out some leftover vmm methods that shouldn't be in the image WideStringPrimKill.1.cs - performance cleanup of left over string rework. LocalePluginAddins.1.cs - change methods to make use of LocalePlugin if it exists. Mantis-1092-MorphDropFix, fix by Edgar De Cleene (edc) and Scott Wallace (sw) Mantis-0503-TargetSighting, fix by Jerome Peace (wiz) Mantis-1015-SnapView, fix by Jerome Peace (wiz) Mantis-1771-ClickExerciser, fix by Jerome Peace (wiz) Mantis-1625-Thumbnails, fix by Jerome Peace (wiz) Mantis-1484-TrashHalo, fix by Jerome Peace (wiz) Mantis-1454-ArrowPrototypeFix, fix by Jerome Peace (wiz) Mantis-1347-ListDoubleClick, fix by Jerome Peace (wiz) Reviewed by Juan Vuletich (jmv) integer printOn:padded 2229 character fixes 2254 date fixes 2117 compiler fix of lukas evaluate:in:to:notifying: http://bugs.impara.de/view.php?id=1041 LowSpaceAndInterruptHandler-3- dtl OB - http://bugs.impara.de/view.php?id=2229 integer printOn - http://bugs.impara.de/view.php?id=2254 character fixes for Smacc on MC - http://bugs.impara.de/view.php?id=2117 date fixes - compiler fix of lukas for evaluating - http://bugs.impara.de/view.php?id=1041 Yes david finally :) Low watcher Traits have been pushed in the update stream. - compiler fixes so that we can load code with MC - flaps fixed - browser fixes - underscore fixes of bert - stroke event fixes of bert ToolPlus changes of andreas. - Basically some fixes by bert of MC - Some changes in the classBuilder necessary for traits and some bugfixes. (script 1) morphic splitter (script 2) deprecated + CurrentProjectRefactoring + collection tests (script 3) explore label + print fixes + Network fixes+ starMorph + SMloadder + double initialize (script4) Removed dictionary fixes + Delay class comment + TTCFont speedup + menuMorph enh + handMorph cursor enh Tests: ----------------------- -> fix #testAsMutator in Collection -> fixed some tests in Kernel -> fix undeclared vars in ClassRenameFixTest Kernel --------- -> removed TempNameCache from CompiledMethod Monticello -------------- fix two Undeclared: - MCGOODSRepository referenced KKDatabase directly - MCVersionTest: add instVar "visited", fix a method to use it. SmaCC ---------- - add isRBToken ST-80 ------- - remove the warning in #methodSourceContainingIt (This is quite fast now... under a minute) ->0001759: GradientCache ->Change Set: RemoveUnusedTempNameCache-bvs ->0001757: [ENH] New example method for HtmlFileStream class ->Change Set: CategorizerFix ->Subject: [BUG][FIX] Class>>rename fix 6/2004 (needs approval) ->0001732: [Fix] Event handler warning still incorrect in 3.8. Time it was corrected. ->Change Set: NumSelectionsInViewFix-nk ->0001688: [BUG] [ FIX] noticeMouseOver-wiz (sm) ->delete GeeMailMorph class>>initialize (DNU) ->Change Set: ProjectViewYellowMenu-fc ->Change Set: SystemWindowLabel-fc ->Change Set: QuickieDropShadowFix ->Change Set: SqueakMapSystemVersionFix ->0001663: [FIX] TextEmphasisMagicNumberCleanup-fbs ->A simple Stack class ->improved Magnitude comment ->Move http related methods from MC into the network package ->0001761: Exeption Hangling in SmaCCParser use undefined method ->001766: [Fix] Preferences Help not working Squeak 3.8 ->Change Set: UpdateProgress ->fix to make toolbuilder apps look better with lookenh ->Deprecated methods now in dedicated package The LookEnhancements are now merged into 3.9a: Highlights of this release: -> Hernan Tylim's great Preference Browser -> Monticello now loads patches when updating from source.squeakfoundation.org More complete list of changes: ------------------------------------------ -> register conflictchecker in FileList -> integrated method from ToolBuilder-Spec -> Change Set: AlphaNumUID -> Change Set: MakeArrows-sps -> Bugfix: use isDoit for testing doit selector -> comment for #openMorphicProject -> remove class SyntaxTestMethods. -> remove class for the proxyeditor -> categorize all uncategorized methods of package system Change Set: DecompileTree Date: 30 January 2005 Author: Dean Tribble, Alexandre Bergel Allow a single method to be properly decompiled. Change Set: PreferredKeyboardPosition-fbs Date: 7 January 2005 Author: Frank Shearar Update 5994 included a bunch of methods with badly spelled names. This changeset corrects those spellings Change Set: LayoutFrameReorg-fbs Date: 17 January 2005 Author: Frank Shearar This changeset simply moves a bunch of unclassified methods of LayoutFrame class into the instance creation protocol. 0001516: [ENH] FlapTab performance improvement Description Michael Latta <[hidden email]>: "When working with some performance tests I noticed that FlapTab was consuming too much time repositioning itself when morphs in the world view resized. The attached change list cuts the time for FlapTab>>ownerChanged by about 33%. With 9 tabs on my image it amounted to a 3% overall improvement. The changes involve testing for position changes before making them so it does not reposition to the same location, and using symbols rather than strings, and == rather than collection>>includes:. Change Set: FTPMultiLineResponseFix-nk Date: 19 January 2005 Author: Ned Konz The FTP protocol (RFC959), allows for multiple lines responses to commands. If a response is multi-line, the fourth character of the first line is a $- and the last line repeats the numeric code but the code is followed by a space. This CS fixes our inability to deal with such responses. Change Set: HTTPProxyPrefs-hpt Date: 10 December 2004 Author: Hernan Tylim This changeset is part of the Preferences Refactorings. The idea is to start moving all the user preferences and settings that are currently dispersed on the image to the Preferences class to be Preference instances. For more info look into the PrefViews-hpt preamble. The changes present on this .cs are simple refactorings on HTTPSocket to make it use getter and setter to access to the http proxy info, which will be stored on Preferences instead of class variables. Change Set: 6681PhonemeRecognizeMorph-dd Date: 28 August 2005 Author: Dominique Dutoit Remove the hard-coded bounds and tell the morph to resize itself if its content doesn't fit. Change Set: WindowColorPrefs-hpt Date: 8 December 2004 Author: Hernan Tylim This changeset is part of the big Preferences refactoring that started on 3.8 with PrefViews-hpt, and PrefViewsRegistry-hpt. On these previous changesets we prepared Squeak to use Preference instances to store any type of preferences, not only boolean. This changeset will move where squeak stores its window colors preference to regular Preference instances, so they will be editable from the new PreferenceBrowser Change Set: TimeProfileFullMethodNames-efc Date: 27 August 2005 Author: Eddie Cottongim TimeProfileBrowser was truncating method names making it a real pain to use on long selectors. This turned out to be a bug; MessageTally creates new instances of itself to do its work but fails to pass along the length preferences, and uses a short default length. This changeset makes it pass the length preference down. Highlitghts in this release: -> ConflictsChecker integrated (just a handfull of methods, no classes and important to have around) -> FastSocketStream replaces SocketStream -> MultiLevel undo/redo patch integrated -> Preference for auto-indenting -> alternate "SQ00" syntax support removed. -> Many small fixes and enhancements -> first steps to add URI and ToolBuilder (both not yet complete) A more extensive list can be found below. On the TODO List -> LookEnhancements -> patches from ToolBuilder -> MethodAnnotations -> ..... Marcus New Packages: ToolBuilder* Collections ---------------- Change Set: WideStringPrimKill Change Set: ConflictChecker recategorize Array>>isLiteral under testing 001382: 3.8 performance regression - ByteSymbol>>#hash again Compiler ------------- - remove support for #SQ00 alternate Syntax - removed dialect support from DialectStream - renamed DealectStream to ColoredCodeStream Exceptions --------------- -remove support for #SQ00 alternate Syntax Kernel --------- - remove support for #SQ00 alternate Syntax - Moved RuleDate, RuleIndexDate and RuleSelectionCodeDate in package on SM - Change Set: fasterDayOfYear-brp: - Change Set: JonClassCommentBlank Morphic ----------- - 0001673: [ENH] Unclassified method moved to proper category. - 0001674: [ENH] More useful comment for `Morph roundedCorners' method - 0000589: Changing text style in text view fails with TextMorphForEditView dNU: selectFrom:to: - removed dialect support from DialectStream - renamed DealectStream to ColoredCodeStream - Change Set: MultiUndo-v2-sps: MultiLevel undo/Redo for TextMorph - 0001685: [ENH] Added example (demo) method to the TransformationMorph class. - 001720: [BUG][FIX] TickIndicatorMorph-stepTime-huma - remove recognizeCharactersWhileMouseIn: - Change Set: PreferredKeyboardPosition-fbs Network ----------- -Change Set: UUIDGeneratorSeedFix-nk -Change Set: FastSocketStream-gk: replaced SocketStream by FastSocketStream -add Impara's URI package (not yet functional, more changes needed) SmaCC ----------- - add realValue to SmaCCToken ST80 ------- - remove support for #SQ00 alternate Syntax - Change Set: MultiUndo-v2-sps - Change Set: AutoIndentPreference-mikki System ---------- - remove support for #SQ00 alternate Syntax - really delete the Character Recognizer - Change Set: AutoIndentPreference-mikki Tools ------- - remove support for #SQ00 alternate Syntax - Change Set: ConflictChecker From ESUG: - Multi-pattern selection added for the StandardFileMenu - Minor refactoring of FileList2, and a new class called FileChooser for selecting a file to open. - Change Set: ObjectExplorerWrapperEnh (display classnames when error occurs) Baloon: Change Set: BitMapFillFix01-wiz 0001018: Grab new morph from screen can produce a null form. Collection: -> 0001382: 3.8 performance regression - ByteSymbol>>#hash -> add contains: to Collection (for compatibilityu with visualworks) -> changes LookupKey class>>#key: to use basicNew -> adds Collection>>#do:displayingProgress (thanks to Phillippe Marschal) -> moves some methods of SequencableCollection from other in-image packages to package Collection: - do: aBlock displayingProgress: aString - groupsOf: n atATimeCollect: aBlock - groupsOf: n atATimeDo: aBlock -> String>>#string -> fixes OrderedCollection and Set to not call initialize -> 0001532: Remove some VMMaker methods to package -> Change Set: revertAsExplorerString-dew -> Change Set: DelCompareTiming -> Change Set: Misc018-HeapIndexCallback -> Change Set: ByteArray-doubleAt -> Change Set: TextLinkValidateFix-nk -> Change Set: CollectionPrintingEnh -> Change Set: CollectionPrintingEnhFix -> add isArray -> fix for: 0001553: [FIX][BUG] Set>>atRandom: is not random -> Change Set: raok-elementPrinting -> introduce self evaluation for printing -> adds Character>>sameAs: -> adds the expandMacros* methods from the RB to String -> Change Set: WeakKeyDictionaryNilsFix -> Move SparseLargeTable from Multilingual into Collections -> add isDictionary Compiler -> moved primitiveNode into another category -> Change Set: defaultClassCategory-hboon -> adds format: textOrStream in: aClass notifying: aRequestor contentsSymbol: aSymbol -> adds format: textOrStream in: aClass notifying: aRequestor decorated: aBoolean (both from RB) -> add Compiler class>>recompileAll -> Change Set: 6603CleaningUtilities -> make Decompiler>>decompileBlock use BlockCtx>>endpc Compression: -> Change Set: HexFixes Exceptions -> moves an sunit method into sunit package FFI -> Change Set: HexFixes Files -> Change Set: mimeTypesFor-brp -> fix for 0001508: misleading message if .sources not found -> Change Set: FileDirectoryFix -> Change Set: UrlToText-fbs -> FileNameDo Graphics -> Change Set: ColorfromUser-fc -> Change Set: ColorTimeArray-fbs -> Change Set: BasicNew-md -> Change Set: FasterDistance-spfa -> a patch to the PNM support -> introduce self evaluation for printing -> Change Set: TextStyleWriteSF2Fixup -> Change Set: HexFixes Kernel -> 0001287: [ENH] Integer primality testing -> Change Set: CategorizerFix -> Change Set: TimespanEqualityFix -> an additional translated string from esther -> Change Set: EasterDate -> make Object>>-> to call Association basicNew -> 0001532: Remove some VMMaker methods to package -> Change Set: IntegerByteEncodeFix-850-v1-laza -> Change Set: KCP-0112-FixCanUnderstand -> Change Set: deprecate-isKindOThenDo -> Change Set: IntegerReadFromBaseFix-369-v2-laza -> fix for 0001423: [BUG] BlockContext>>valueWithArguments issues misleading error when sent non-Arrays -> 0001033: Inconsistency in naming of the formal parameters in certain methods of the True and False classes. -> add Object>>isArray -> introduce self evaluation for printing -> delete RenamedClassSourceFileRead -> adds Behavior>>formatterClass and Behavior>>prettyPrinterclass -> query methods for Behavior (which*) -> change ClassDescription>>reformatMethodAt: to use prettyPringerClass -> added to CM: allLiterals, getSource, literalsDo:, messagesDo:, methodClass, refersToLiteral, sendsSelector: -> changes CM>>methodNodeFormatted... to use prettyPrinterClass -> Added to Object: error, needwork -> Change Set: HexFixes -> Change Set: ObjectExplorerIntegerBases-v2-laza -> Change Set: 6603CleaningUtilities -> comment for CompiledMethod>>methodClass -> add BlockContext>>endpc: the end of the block bytecodes -> add isDictionary -> Change Set: SemaphoreWaitAndWaitTimeoutFix-jf Morphic -> Change Set: EsthersTranslations -> Change Set: WebpageMorphFix -> Change Set: MorphicEvtPos-wiz -> Change Set: fasterHScrollbar -> removed component1 -> moved groupsOf: methods to collection package -> removed the broken WordNet / Translation entries in the menus -> removal of docPane -> Change Set: KCPrmCharRecog -> Change Set: TextLinkByKbd -> removes a call to Object>>#isKindOf:thenDo: -> Change Set: DisplTextSol3 -> removed call to depr. method in NewParagraph -> Change Set: MetaMenu-fc -> fix for eventRecoder to work again -> Change Set: ListText-bf -> Change Set: WatcherDoubleClickFix-nk -> Change Set: revealPlayer -> Change Set: StarMorph-wiz -> Change Set: MorphInAScrollPane-fbs -> removed PDA -> Change Set: 6603CleaningUtilities -> 0001626: New comments to two methods -> 0001634: [ENH] Improved comment and new example methods for ScrollPane. -> move WebPageMorph to Scamper -> Change Set: MorphicPerformance1-efc -> 0000807: Setting a dotted lines gradient fill property to true in view invokes emergency debugger -> Change Set: MorphicPerformance1-fix1-efc Multilingual -> Change Set: HexFixes -> Change Set: MorphicPerformance1-efc -> Move SparseLargeTable to Collection package Nebraska: -> Change Set: objectSocketFixes Network -> 0000597: Network-TelNet WordNet broken, should be removed -> Change Set: icsMimeTypeAddition -> Change Set: MimeTypeFix -> Change Set: HTTPServerDirectoryFix -> Change Set: MailComposition-fc -> Change Set: FTPSymbolicLinkHack -> Change Set: UrlToText-fbs -> Change Set: SocketStreamPrintString-dew Speech -> Change Set: PhonemeRecognizerMorphFixEnh-dd ST80 -> Change Set: EsthersTranslations -> cleanups for WordNet removal -> Change Set: KCPrmCharRecog -> Change Set: defaultClassCategory-hboon -> fixes two methods to use aClass prettyPrinterClass -> fix for CodeHolder>>#addOptionalAnnotationsTo:... from Ned (was part of RB overrides) -> Change Set: WorkspaceBindingPresence System -> Add a synonym for #unloadModule: t -> Change Set: EsthersTranslations -> Change Set: LocalePluginAddins -> Change Set: searchFix-bf -> Change Set: MisspelledWarning -> removed component1 reference -> 000915: [BUG] [FIX] Preferences fontConfigurationMenu don;t open and is incomplete -> Change Set: VMM38-gc-instrument-image -> Change Set: VMM38-64bit-imageUpdates -> cleanups for WordNet removal -> 0001410: [BUG][FIX] "HTTPClient examplePostArgs" is broken -> deprecated Utilities>>isObject:memberOfOneOf: -> edit discardOddsAndEnds because of removals -> Change Set: PrintShowingDecimalPlacesFix-dtl -> a fix to make DigitalSignatures work again -> Change Set: SystemChangeNotificationFix -> Change Set: UrlToText-fbs -> changed testFormatter to use prettyPrinterClass method -> removes SystemDictionary discard3d, discardSoundAndSpeech, discardWonderland -> Change Set: CaseAwareDropHandler-spfa -> Change Set: 6603CleaningUtilities Tools -> Change Set: EsthersTranslations -> Change Set: MethodReferenceEnh -> Change Set: ProcessBrowserRegisterProcess-dtl -> Change Set: MessageTallyIndent-dew -> Change Set: FileList2 -> Change Set: KCPFixBrowserOrganization -> Change Set: FixDebuggerOpening -> some changes from RB overrides -> Change Set: classCommentStamps-dew On Feb 2, 2008, at 1:14 AM, Avi Bryant wrote: > On 2/1/08, Adrian Lienhard <[hidden email]> wrote: > >> ...and without more than 350 fixes and enhancements that went into >> 3.9. >> >> Who would be going to redo this work? > > Tangentially - what's the easiest way to see a list of these 350 > enhancements? > > Avi > > |
In reply to this post by garduino
Sure speech, nebraska and a lot of broken code is a useful thing....
On Feb 2, 2008, at 1:42 PM, Germán Arduino wrote: > 2008/2/1, Edgar J. De Cleene <[hidden email]>: >> >> >> The point is if we should start from 3.8.2 (without Traits and >> Monticello) >> or we should continue from 3.10 is now. > > My vote is without Traits!!! (Traits: big example of a not useful > thing, IMHO), > > Cheers. > > |
In reply to this post by Tapple Gao
> traits and the other without. This just didn't arise in the > meeting > > I think that this is missing a point. I may be reading what Andreas said wrongly, and I am sure he will be quick to correct me if I am wrong. I do not think that the issue is with the concept of traits or the fact that traits work. It is the application of traits to the class/metaclass heirarchy which renders elements of it difficult to understand. I would like to see the concept of traits retained, and if it simplifies things, extended to include stateful traits. I would like for the concept of traits which enables the behaviour defined by a class to be an assembly of components to allow one of those components to be a collection of Package extension-methods. So that extensions have a true identity, as opposed to being one method with a '*' in its method category. best regards Keith |
In reply to this post by stephane ducasse
El 2/2/08 12:03 PM, "stephane ducasse" <[hidden email]> escribió: > Sure speech, nebraska and a lot of broken code is a useful thing.... 3.10 don't have speech and I don't was let download nebraska By the way, I send code for fix some the nebraska to Yoshiki a long time ago Edgar |
In reply to this post by Edgar J. De Cleene
Hi egdar
to make sure that I understand that correctly: 3.10 harvested a couple of fixes (compared with the 685 mantis items marcus closed) and you are telling that 3.11 will in addition merge everything and reconcile all the forks. This sounds optimistic and not really aligned with the results of 3.10. Don't you think? Now a couple of weeks ago someone sent me this mail "Just out of curiosity... I wanted to write a mail, along the lines of... I looked into the bugs fixed in 3.10... I'm a bit confused, though, because there does not seem to be much progress in 3.10 with respect to bug fixing. Is Mantis up to date? Is Mantis showing what I think it should (I used the date + status filters)? Below is a comparison of the number of reports for each possible state in 3.9 (March 1, 2005 - Sept 30, 2006) and 3.10 (Oct 1, 2006 - today): 3.9 3.10 status ---------------------------- 276 254 new 19 8 feedback 4 1 acknowledged 2 3 confirmed 70 111 assigned 47 10 resolved 685 25 closed ...but I did not send it (yet)" And you know the saddest part is that is not that 3.10 harvested so few fixes because this can happen. The saddest part is that this person did not send his email because it means that people get bored or at the end that we do not succeed to bring their enthousiasm into the community. With this kind of attitude will may lose on the long term. I like squeak because people are sharing and are enthousiast about it. Of course managing the image with MC does not really work for streaming changes. I really hope that MC2 will solve that. Colin told me that it will: MC2 will really load only the difference and will not have to rescan everything. Now MC2 is not ready and may be not ready. So I will do 3.9.1 because some people asked me and after I will really take the time to think about where I put my energy. May do a call for building another fork of something different from Squeak as it is now or simply do not care. May be other dynamic languages are more fun at the end. Stef |
In reply to this post by stephane ducasse
I was who talked about Traits (Not useful to me and several other persons).
Is only an opinion, my opinion, but not was Edgar who commented so. Cheers. 2008/2/2, stephane ducasse <[hidden email]>: > Sure speech, nebraska and a lot of broken code is a useful thing.... > > On Feb 2, 2008, at 1:42 PM, Germán Arduino wrote: > > > 2008/2/1, Edgar J. De Cleene <[hidden email]>: > >> > >> > >> The point is if we should start from 3.8.2 (without Traits and > >> Monticello) > >> or we should continue from 3.10 is now. > > > > My vote is without Traits!!! (Traits: big example of a not useful > > thing, IMHO), > > > > Cheers. > > > > > > > |
In reply to this post by Tapple Gao
Here are the topics we discussed in our first skype call, from 14:00 to
15:00 utc. Edgar, Colin, Keith, Esteban, and Matthew participated - MC2 supports both PackageInfo-type files, and arbitrary slices. - Damien's mc2 work was in the direction of making MC2 emulate MC1, and colin didn't use that - Projects are something - The biggest issue with traits seems to be that the implementation is too complicated to unerstand, and for the tool developers to write tools for. - Talk in spanish about traits - Could traits be made loadable/unloadable? Is someone working on a simpler implementation? Andreas may be. - Will Stateful Traits be simpler to incorperate? -- Matthew Fulmer -- http://mtfulmer.wordpress.com/ Help improve Squeak Documentation: http://wiki.squeak.org/squeak/808 |
In reply to this post by stephane ducasse
stephane ducasse schrieb:
> Hi egdar > > to make sure that I understand that correctly: 3.10 harvested a couple > of fixes (compared with the 685 mantis > items marcus closed) and you are telling that 3.11 will in addition > merge everything and reconcile all the forks. > This sounds optimistic and not really aligned with the results of > 3.10. Don't you think? > > Now a couple of weeks ago someone sent me this mail > "Just out of curiosity... > I wanted to write a mail, along the lines of... > I looked into the bugs fixed in 3.10... I'm a bit confused, though, > because there does not seem to be much progress in 3.10 with respect > to bug fixing. Is Mantis up to date? Is Mantis showing what I think it > should (I used the date + status filters)? > > Below is a comparison of the number of reports for each possible state > in 3.9 (March 1, 2005 - Sept 30, 2006) and 3.10 (Oct 1, 2006 - today): > 3.9 3.10 status > ---------------------------- > 276 254 new > 19 8 feedback > 4 1 acknowledged > 2 3 confirmed > 70 111 assigned > 47 10 resolved > 685 25 closed > potential to annoy many. Redoing most of the patches of 3.9 and 3.10 seems to be a huge amount of work. So there need to be a consensus that the result is worth the efforts. Are Traits the only show stopper for a unification of Squeak distributions? If not, what else has to be done? And last but not least: isn't the unification a contradiction to the minimal image target? Regards Andreas |
In reply to this post by stephane ducasse
First of all, I would say, thath I appreciate your work..very much. I
admire you, like I admire all the "Squeak's Masters". This message, is for all the community. Probably, I'm not the right person to speak, because, I'm only a newbie, learning this fantastic world, and I'm more an observer than a contributor (sad to me). I'm concerned because in all the time I'm in this list, always thath the team want to improve, or do some thing to the project, the people thath disagree, create a fork, or don't want to help. The forks are not good for any project. In spanish maillist (as all can say, my english is terrible, and there I can express me better), I said thath IMHO, the best thath Squeak can do, is to go to a minimal Squeak, and all the things has now, could be loaded. and I think all we need to do is help to this, because this, is for all. Only can express my sadness, because I'm attending how this community is destroyed. Why all don't help? Why all only want his own purposses?.This is what we want to show to the newest members? a selfish community? The best thing we can do, if I don't like something, I will see to other side? We have a board, true? This was elected by the Squeak community, and IMHO, the opinions of the board, could be voted by the community, and the community, will get a Squeak for all. The example of Traits. Some wants Traits integrated on the image. Other Don't want it. Ok, a poll for this, and the community may respect the results. Only is an opinion, and, as the mail you received, I didn't have the value courage to send it. This mail go with good intentions, but I don't know how express me better, sorry. El 02/02/2008, a las 16:36, stephane ducasse escribió: > Hi egdar > > to make sure that I understand that correctly: 3.10 harvested a > couple of fixes (compared with the 685 mantis > items marcus closed) and you are telling that 3.11 will in addition > merge everything and reconcile all the forks. > This sounds optimistic and not really aligned with the results of > 3.10. Don't you think? > > Now a couple of weeks ago someone sent me this mail > "Just out of curiosity... > I wanted to write a mail, along the lines of... > I looked into the bugs fixed in 3.10... I'm a bit confused, though, > because there does not seem to be much progress in 3.10 with > respect to bug fixing. Is Mantis up to date? Is Mantis showing what > I think it should (I used the date + status filters)? > > Below is a comparison of the number of reports for each possible > state in 3.9 (March 1, 2005 - Sept 30, 2006) and 3.10 (Oct 1, 2006 > - today): > 3.9 3.10 status > ---------------------------- > 276 254 new > 19 8 feedback > 4 1 acknowledged > 2 3 confirmed > 70 111 assigned > 47 10 resolved > 685 25 closed > > ...but I did not send it (yet)" > > And you know the saddest part is that is not that 3.10 harvested so > few fixes because this can happen. The saddest part is > that this person did not send his email because it means that > people get bored or at the end that we do not succeed to bring > their enthousiasm into the community. With this kind of attitude > will may lose on the long term. I like squeak because people are > sharing and are enthousiast about it. > Of course managing the image with MC does not really work for > streaming changes. I really hope that MC2 will solve that. Colin > told me that it will: MC2 will really load only the difference and > will not have to rescan everything. Now MC2 is not ready and may be > not ready. > > So I will do 3.9.1 because some people asked me and after I will > really take the time to think about where I put > my energy. May do a call for building another fork of something > different from Squeak as it is now or simply do not care. > May be other dynamic languages are more fun at the end. > > Stef > |
In reply to this post by Andreas Wacknitz
>> I don't know what the outcome of this discussion will be. It has
>> the potential to annoy many. > Redoing most of the patches of 3.9 and 3.10 seems to be a huge > amount of work. > So there need to be a consensus that the result is worth the efforts. > Are Traits the only show stopper for a unification of Squeak > distributions? are you joking? Do you know what is to merge two pacjkage of graphics or other that diverged. Check all the changes made in OLPC or Sophie. I'm in both mailing-lists. And I already said that I was sad that we cannot harvest those changes because this is a huge tasks and it has ***nothing*** to do with traits. > If not, what else has to be done? > And last but not least: isn't the unification a contradiction to the > minimal image target? Not at all, if people want to remove traits. I personnally do not care. If the community does not want that this is ok for me. This will be just an error but this will not be my failure because we would have try. You know PHP people implemented stateful traits and apparently some want to push that in the mainstream. The Ruby guy said that he would have preferred traits over his mixin. But again this is ok for me/us if traits get removed. Traits are not the problems because there are small and do not change the run-time. In 3.9 we integrated a lot of fixes and pragmas and other stuff like that. Now if the community wants to lose that then this is stupid and I will simply do something else. There are plenty of dynamic language today with cool communities. We are not the only one. |
Free forum by Nabble | Edit this page |