Squeak-dev April 2004

squeak-dev@lists.squeakfoundation.org

172 participants
488 discussions

[ENH] ChangeListTimeComparisionEnh-nk
by Ned Konz 29 Apr '04

29 Apr '04

"Change Set: ChangeListTimeComparisionEnh-nk Date: 29 April 2004 Author: Ned Konz Adds 'select newer methods' and 'select older methods' to the ChangeList menu. These compare the timestamps in the change records with those in the current versions of the corresponding methods. Also adds a Preference called #changeListSelectionsAdd that makes selections in the ChangeList additive. You can change this preference from the window menu, too. Also adds an 'invert selections' button to the optional button pane. "

1 0

[FIX] InspectElementFix re-posted ( [er][sm][et][approved] Looks fine to me. )
by Ned Konz 29 Apr '04

29 Apr '04

1 0

[ENH] FileListFileOpenFix-nk v2 (with ServicesInArchiveViewer-nk and BFAVFileListOpenFix-nk)
by Ned Konz 29 Apr '04

29 Apr '04

"Change Set: FileListFileOpenFix-nk (v2) Date: 29 April 2004 Author: Ned Konz Fixes the FileList so it doesn't open files unless you choose an action requiring a file to be opened. v2 makes it possible for the services to be used from the ArchiveViewer (using the companion ServicesInArchiveViewer-nk changset). "

1 0

[FIX] FileListFileOpenFix-nk
by Ned Konz 29 Apr '04

29 Apr '04

"Change Set: FileListFileOpenFix-nk Date: 15 February 2004 Author: Ned Konz Fixes the FileList so it doesn't open files unless you choose an action requiring a file to be opened. This allows Windows users to delete *.st and *.cs files, for instance. "

1 0

[FIX] InspectElementFix re-posted
by Ned Konz 29 Apr '04

29 Apr '04

"Change Set: InspectElementFixv2 Date: 9 March 2004 Author: Yoshiki Ohshima When printString contains return characters, the menus that assume the resulted string fits one line fails. This fixes one problem I encounterd. Also, the short-cut in the WriteStream>>nextPutAll: short-cuts the limit check in LimitedWriteStream. This change fix it. "

1 0

[Q][ETOYS] Constrained dragging?
by Lothar Schenk 29 Apr '04

29 Apr '04

How can I restrict the path of an eToy that's in the process of being dragged to an allowed region, or alternatively to avoid conflict with other eToys as obstacles? Curious, Lothar

4 5

[UPDATES] 27 for Squeak3.7beta
by dway＠mailcan.com 29 Apr '04

29 Apr '04

The latest batch of approved items for 3.7beta. Includes the class organizer fixes in KCP220, which is a significant change (and automatically closes any open browsers), but this fixes some important stuff. - Doug ------------------------------ 5879cursorfix 5880PackageInfoFromSM-avi -- Avi Bryant -- 6 April 2004 Update the PackageInfo package from SqueakMap. To avoid having to pull in Monticello for now, this first strips down the existing PackageInfo to a bare class (so as to not obsolete existing instances and references), and then loads the package as if for the first time from SqueakMap. In the future, we should update it 'properly' by just asking SM to pull in the next version." 5881KCP220ClassOrgnzFix -- Nathanael Schaerli -- 15 April 2004 This changeset cleans up the ClassOrganizer hierarchy and fixes some bugs in the SystemNotification framework. Furthermore, it fixes some other bugs that got appearent while I was working on this cleanup. In detail: - Cleans up the ClassOrganizer hierarchy. This was necessary because the old hierarchy was conceptually wrong and made it therefore unnecessarily hard to extend it. - Makes sure that changes to the class organizer always trigger the corresponding system change notifier event. (This was not the case so far). - All the system change notification triggers are in the kernel where they actually happen. In particular, I removed all the triggers in the class Browser. - Removes a few minor problems with nil values in the SystemChangeNotification framework. - Removes a bug when using drag and drop between classes without having message categories selected. IMPORTANT: - This changeset has been modified by hand in order to make it file in properly. - Close all the browser windows before filing this changeset in." "This script closes all the browser windows." PopUpMenu inform: 'This changeset will now close all your browser windows. This is necessary to avoid errors after filing it in.'. SystemWindow allInstancesDo: [:each | (each model isKindOf: Browser) ifTrue:[ each delete]]. 5882UpdatingMenuItemIcon-nk -- Ned Konz -- 13 April 2004 UpdatingMenuItems lose their icons, if any, when they change their contents. This causes the icons to be re-established after changes. 5883SARInstallerV28-nk -- Ned Konz -- 17 April 2004 This change set updates the 3.7b SARInstaller to v28, which is the latest version on SqueakMap. It also marks the SqueakMap package relase as being installed. 5884BorderFixes-nk -- Ned Konz -- 15 April 2004 This change set allows Players to change border widths to 0 and back up without losing the borders. I am proposing this instead of the 'Player borders' fix that Karl Ramberg posted on 9 April (ID 21791). This has the advantages of: - allowing for alternative implementations of borderWidth: and borderColor: - removing intimate knowledge of border style mechanisms from Player. v2: fix for borderColor: #raised, etc. 5885MonitorFix 5886ObjectsToolsBabel-dgd -- Diego Gomez Deck <DiegoGomezDeck(a)ConsultAr.com> -- 8 April 2004 The buttons for categories are now translatable (just like in Small-Land image). 5887MenuItemWidthFix-nk -- Ned Konz -- 14 April 2004 MenuItemMorphs with markers weren't getting laid out correctly. This is a fix for that problem. To see the problem, try this (from Stéphane Rollandin): | m | m _ MenuMorph new. m add: '<yes> some action' action: #yourself. m add: '<no> some other action' action: #yourself. m add: 'too short !' action: #yourself. m popUpInWorld 5888RemoveEmptyCatgyFix-nk -- Ned Konz -- 23 April 2004 Fixes a long-standing annoyance where the 'remove empty categories' menu choice would not work unless you had a category selected. 5889ParserRevertFix-gk -- Göran Krampe -- 3 March 2004 The issue was triggered by 'NS' (Nathanael?) in update 5700 were he changed 'self selectionAsStream' to 'self selection string' in #evaluateSelection. The result of that change is that instead of passing a ReadStream with a position into the full text, it passes only the selected code into the Parser. That works fine until the Compiler starts to try to givinng feedback using instvar requestor. Those methods, like for example when it tries to correct a selector (this one was caught by Frank's patch) or when it reports a compilation error (not caught by Frank's patch) then thinks that the position it has are absolute and can be used in messages to requestor. But they are now relative. Also, I note that there is still a sender of #selectionAsStream, namely when you choose 'debug it'. Given Frank's patch that case is now instead broken. So... for the time being, this changeset simply reverts this back to 'self selectionAsStream'. 5890condenseChangesFix-bg -- Boris Gaertner -- 23 April 2004 This change set fixes a bug that was reported on Apr 23, 2004 by Alexander Lazarevic. In condenseChanges, the LastQuitLogPosition was not told to SmalltalkImage." 5891WeekMissingStartDay-jf -- Julian Fitzell -- 23 April 2004 Week (class)>>starting:duration: was accessing StartDay directly rather than calling #startDay. This caused it to bypass the lazy initialization of StartDay. The bug can be noticed by doing: Date today asMonth weeks first dates first dayOfWeek It should return 1 (Sunday) but returns 7 (Saturday) in a new image without this fix." 5892monthAndYearInDateV2-avi -- Avi Bryant -- 21 February 2004 V2: I was wrong about #year, so it's been removed from the CS. #monthIndex, which was broken by this changeset, is now fixed. The Date class, though not part of the ANSI standard, is given ANSI semantics for #month and #year by the Chronology package, which has them return numeric values. This breaks backwards compatibility with Squeak code that expects them to return Month and Year objects respectively. This changeset restores these methods to their original behaviors (while keeping the ANSI standard behavior for the ANSI DateAndTime class)." 5893BitmapFillStyleOrgn-klc -- Ken Causey -- 27 January 2004 The origin was not being set in fromForm: which is I believe an error. It at least led a primitive failure when trying to use the resulting fillstyle. What is the form: class method for? It would seem to have much the same trouble since even more of the instvars will not be set." 5894TMEContextMenuFix-dgd -- Diego Gomez Deck <DiegoGomezDeck(a)ConsultAr.com> -- 8 April 2004 BUG: Take a TextMorph (from the Objects Catalog or from Supplies Flap), start to write on it and hit ESC while the focus is still on the morph. A DNU will raise. This changeset includes a fix, not so elegant... but it works. 5895IsKindOfForm-nk -- Ned Konz -- 18 April 2004 Changes callers of 'isKindOf: Form' and 'isKindOf: ColorForm' to use the existing queries isForm and isColorForm. Reorganizes the method protocols in Form. 5896norwQuote-ls -- Lex Spoon -- 10 April 2004 Fixes the quoting in the translation method for Norwegian. The Squeak compiler is happy with either version, but the Refactoring Browser needs the quotes to balance more strictly. This problem may point to a bug in Squeak's scanner of literal arrays." 5897ParagraphEditorExchange-cmm -- Chris Muller -- 9 April 2004 This appears to fix the exchange command in text editors." 5898FixImplIn-gk -- Göran Krampe -- 26 April 2004 Someone somewhere (?) fumbled the ball when letting 5577 into the stream. The method Debugger>>contextStackMenu:shifted: first got modified by Ned (autocreation of stubs) in 5511 and then was run over in 5577 (breakpoints). I am not sure how this slipped by - I have been trying to see in BFAV if this was due to an approved item not yet pushed out in the stream, but I am not sure. Might be interesting to figure out how this happened. This cs puts that 'implement in' choice back in the menu. Simple as that." 5899ZipExtractMakeDir-nk -- Ned Konz -- 22 April 2004 When you do an 'extract all' from the ArchiveViewer, subdirectories aren't created properly. This change set fixes that problem." 5900MoreEventSensorFixes-nk -- Ned Konz -- 13 April 2004 This is another modification to the changes made in updates 5784NoEventSensorProcess-ar and 5866MVCEventSensorFixes-nk This changeset separates the process that checks for the interrupt key being hit (by waiting on the interrupt semaphore) from the process that periodically gets more events if the UI is not doing so often enough. The advantage of doing this is that we can still remain responsive to interrupts whether the VM signals the semapore or if we detect the interrupt key via processEvent: These changes were suggested by Tim Rowledge. v2: better shutdown for images with prior versions. Sensor shutDown. Sensor interruptWatcherProcess ifNotNilDo: [ :p | p terminate ]. 5901TextDiffWithColors-nk -- Ned Konz -- 24 April 2004 If you use a font that can't do struck-out forms, you won't see the text diffs correctly. This change set makes the diffs appear as blue for old (also struck-out if possible) and red for new (which is the same as it was). v2 changes from blue only to blue/struck-out 5902TextMorphSpellingFix-gm -- German Morales -- 25 April 2004 just a simple spelling fix: rectangluar --> rectangular 5903LinkHighlightClipFix-nk -- Ned Konz -- 8 March 2004 Text links whose bounds extend past the visible bounds of their TextMorphs were not clipping their highlight boundaries when clicked. This CS fixes that problem. 5904multilistAdjust-ls -- Lex Spoon -- 17 April 2004 Tweaks SPS's fixes to make list widths auto-adjust, so that it works on multi-column lists as well. Without this change, I see a lot of debuggers during use of Celeste." 5905ZipCommentsCategories-nk -- Ned Konz -- 22 April 2004 Some comments and protocol reorganization for the Zip-related classes." <This automated message brought to you by the UpdateIncorporationTool.> <See the UpdateIncorporationTool package on SqueakMap for details.>

9 12

Re: FileDirectory>>fileExists: (was: Re: [BUG]Unable to load BFAV, various problems )
by Tim Rowledge 29 Apr '04

29 Apr '04

"Andreas Raab" <andreas.raab(a)gmx.de> wrote: [snip ] > Generally, I'm against this - it makes it a *lot* harder to write good > cross-platform code if you have to deal with the fact that the different > platforms have very different tradeoffs and implementations. [snip] Well, I have a bridge to sell you, young man :-) I agree with the nastiness of the current situation but disagree as to the desired state. I consider a single class with a tightly defined primitive interface to be tantamount to early binding when as we all know, late binding has many advantages. Ignoring for the moment Russell's point about the file naming issue (since file names should be a separate class to filestreams etc anyway) I would suggest that we would benefit from a single FileDirectory class BUT that said class needs a bridge to the actual machine primitives. Such a bridge class can be machine/OS/situation specific ie FileDirectory implements #fileNames by delegating to the currently selected bridge. A UnixFDBridge would probably open a dir stream, keep the pointer, use readdir etc and return the array of names. A RISCOSFDBridge would use quite different means to do the same but crucially the user code would only see the FileDirectory object. Another benefit would be that ServerDirectory could be replaced with a bridge class that handles the ftp stuff but again, without the user code having to know (Well, probably - maybe there are actually reasons to have ServerDirectory user accessible. The principle remains though). The important points are that the user class provides the user protocol and stores only the abstract state and uses a bridge class to handle al the details. The bridge class implements messages to support the user class and stores whatever state it needs to be able to drive the machine primitives. When snapshotting, the user class can drop the bridge object(s) since it must be able to recreate them at need. As another example, consider #directoryContentsFor: - which is a main interface to the platform prims. It is used to get full info about all the files in a dir, to get the names of all files and directories, to get files or directories separately, check for existence and so on. Building on such a prim is great for porting (barring the complication of implementing it!) but not very efficient in many cases. Using a bridge class would allow separate implementations for exists, filenames, stat-like information and so on when and if appropriate. It even allows for machine where it is best to (thinking of a complex way to do it) find out how many entries exist allocate an array to suit iterate, filling the array with names iterate, fetching stat info resort by date found in stat info. Hell, it would allow for a platform to use FFI to access the platform instead of more explicit primitives. Free the primitives from the dungeon! Late binding good, early binding baaaad! tim -- Tim Rowledge, tim(a)sumeru.stanford.edu, http://sumeru.stanford.edu/tim Useful random insult:- Low on thinking gas.

12 29

[FIX] ClassMethodDragFix-nk ( should be revived )
by Alexander＠Lazarevic.de 29 Apr '04

29 Apr '04

This got automatically closed after the first submission in 2002. Ned did resend this fix recently, but I think it has been overlooked residing still under the closed section of BFAV. Because this fixes a couple of annoying things I would like to see this revived and considered for approval.

1 0

[ANN]A plan for 3.8/4.0... (insert drumroll here)
by goran.krampe＠bluefish.se 29 Apr '04

29 Apr '04

Hi all! There has been something tasty cooking off-list in the Squeak community the last few days and it fell upon me to present it to the community at large. Which I do gladly. :) NOTE: The following is my perception, if there is anything wrong here - the appropriate people will surely slap me and correct it. :) ---- As most of you know we just entered 3.7beta and IMHO it looks pretty good. The 3.7 release will not be far away and at that time we typically open up 3.8alpha. So if there are any Cool Things going on that might affect the plan for the upcoming 3.8 and beyond, the timing was "just right" to bring it to the table... And a Cool Thing just happened to drop in! Dan Ingalls has been testing the waters for getting the support of the community in spearheading an effort for Squeak version 4.0 *including a full 64-bit port* to the IA64 ("Itanium"). Yup, you heard it right - with the help of Dan we are entering the land of terabyte images. :) And after exchanging a few emails we Guides of course decided to turn him down, we don't need no stinking 64 bi... eh, NOT! ;) The following is a short description of a battle plan for 3.8/4.0 that Dan brought forward and that we Guides support (well, I haven't heard anyone against): - Dan will be doing the 64-bit port as part of Alan's group at HP. That group also includes a bunch of Squeak heroes. - Since a 64-bit port inevitably brings us an image format change this will be version 4.0. - Of course this will also include most of the pending image format changes that have been talked about for 4.0. There are also (at least) two other important new things banging on the Squeak front door: - Yoshiki's m17n work - Anthony's work with full closures The idea here is a rather aggressive parallell schedule in which the current community process continues with 3.8alpha/beta/gamma "business as usual" and we bring in m17n etc, and at the same time the V4 team with Dan in the lead prepares a migration process that at the end of the release cycle can be executed fairly simply to produce the 4.0 version from the 3.8 version. This would then lead us to a simultaneous release of 3.8 (the last of version 3) and 4.0 (which should be identical in all but the image-format related areas) somewhere in june/july I think. Just as an icing on the cake Craig Latta is also getting anxious to get a part of the action and expressed his intent to try to get the Squat work ready for production in time with 4.0 which could mean that a 4.0k (kernel) might also be released at the same time! This was intended as a "short" announcement but turned lengthy. Dan will probably post shortly with more specifics, and I think Craig too etc. Anyone in the community that wants to participate more actively in these projects should of course not hesitate to contact Dan, Craig or any of us Guides. Phew. It looks like a busy spring, doesn't it? But FUN! Hmmm, Itanium... well, how hard can an Opteron port be after that, eh? regards, Göran PS. Note how I managed to write this posting without any cheezy jokes about "The Return of The King" or "The Return of The Jedi" or... eh, darn.

26 83

← Newer
1
...
5
6
7
8
9
10
11
...
49
Older →

Jump to page: