Squeak-dev January 2004

squeak-dev@lists.squeakfoundation.org

156 participants
396 discussions

SqueakMap - analysis tools?
by Bill Schwab 14 May '04

14 May '04

Hello all, SqueakMap seems to be taking off nicely, and it is a nice way to grab goodies. However, I still come back to the problem of starting from scratch. Dolphin's commercial origins force the issue with each release, but it's also a nice sanity check on whether or not I know what is in my image. Are there any good tools for deciding what to load into a new image? I encourage and Dolphin-using Squeakers to take a look at Migrate from my Dolphin Goodies - especially the version packaged for D5. I would really like to see something like it for Squeak. The other issue, mentioned before, is that SM seems to favor downloading over locally cached copies. Ideally, it should offer to download updates, but be willing to load packages from the local drive. Has that been addressed? Bill Wilhelm K. Schwab, Ph.D. University of Florida Department of Anesthesiology PO Box 100254 Gainesville, FL 32610-0254 Email: bills(a)anest4.anest.ufl.edu Tel: (352) 846-1285 FAX: (352) 392-7029

3 2

HELP! with plugins
by Eric Scharff 11 May '04

11 May '04

(Apologies for nagging folks again, but I never received answers to any of my plugin questions. While creating plugins may not be nearly as exciting as discussing licensing issues, it is far more important to my current project. :) ) Here is a re-post of my latest plugin question: I have a trivial plugin: TestInterpreterPlugin subclass: #MyPlugin instanceVariableNames: '' classVariableNames: '' poolDictionaries: '' category: 'Eric-Test' With a "primitive" method I intend to compile to C: innerLoopX: x y: y limit: escape | | self primitive: 'primGuy' parameters: #(SmallInteger SmallInteger SmallInteger) receiver: #Oop. ^ (escape + x + y) asOop: SmallInteger (Note, simplified to do nothing useful, but illustrates the problem) If I execute: MyPlugin doPrimitive: 'innerLoopX:y:limit:' withArguments #(1 1 1) I would expect to get 3. Instead, I get a walkback: Error: attempt to index non-existent element in an ordered collection OrderedCollection(Object)>>error: OrderedCollection>>errorNoSuchElement OrderedCollection>>at: InterpreterProxy>>stackValue: InterpreterProxy>>stackIntegerValue: MyPlugin>>DoItIn: Compiler>>evaluate:in:to:notifying:ifFail: [] in TestTMethod>>simulatePrologInContext: Array(SequenceableCollection)>>keysAndValuesDo: TestTMethod>>simulatePrologInContext: MyPlugin(Object)>>primitive:parameters:receiver: MyPlugin>>innerLoopX:y:limit: MyPlugin class(TestInterpreterPlugin class)>>doPrimitive:withArguments: What am I doing wrong? -Eric

4 4

Enhanced & Extended Serial Port support.
by John M McIntosh 11 May '04

11 May '04

I've push a changeset and macintosh plugins onto the server, they should appear in a few days. The SerialExtendedMacOS9Plugin.sit goes in the mac 3.5 noncarbon plugin location The SerialExtendedUnixPlugin.bundle goes in the mac 3.5 carbon plugin location {yes the name implies unix, but os-x is unix...} The EnhancedSerialPorts-JMM changeset which is the classes and plugin interface logic you need to file in. I suspect usage will be limited but it enables you to do things with serial ports (and file handles, hint any file handle) that you can't do otherwise. I would think to compile this code on your favorite Linux box that would be a snap. Only one platform specific primitive or method exists serialPortFindNamesPlusKeysstorage which on os-x invokes the IOServiceGetMatchingServices() call to get path names to devices that support the serial protocols. This primitive call a common Linux box could be replaced by smalltalk code to iterate over /dev or read some sort of control file etc to figure out the path names to the serial devices. For mac os-9 this functionality was already part of the VM. I'm hoping that no platform specific C code will be required for unix boxes. Alas windows users require someone to do the dirty work if you want windows specific serial calls? Or will unix emulation somewhere/how etc work? Couldn't speak for parallel ports, never even occurred to me to investigate till now. I should point out that Dan Ingalls <Dan(a)SqueakLand.org> was kind enough to actually fund with real USA $ some of the work required to build this plugin, then willing to release it to the public. So enjoy, and thank Dan for motivating ($) me to write the code and for releasing it. PS I won't pass up an opportunity to point out if you've an obscure macintosh specific feature you would like and like Dan have a bit of cash to pay for it, then we should talk. -- ======================================================================== === John M. McIntosh <johnmci(a)smalltalkconsulting.com> 1-800-477-2659 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ======================================================================== ===

4 5

[BUG][3.6] Resignalling an exception
by Avi Bryant 11 May '04

11 May '04

Try the following line of code in both 3.5 and 3.6: [1/0] on: Error do: [:e | e signal] In 3.5, when the debugger comes up, the top context is the send to #/. In 3.6, it is the handler block, with the send to #/ several frames down. Which is correct? I personally much prefer the 3.5 behavior. Another symptom of what I think is the same bug: [1/0. self inform: 'inner'] on: Error do: [:e | e defaultAction. self inform: 'outer']. Try this code. When the notifier pops up, click Proceed. In 3.5, you will see 'inner'. In 3.6, you will see 'outer'. Again, I would prefer the 3.5 behavior here. Does anyone know what exactly has changed, and how easy it is to fix?

2 1

Re: [BUG][FIX]FileDirectory>directoryExists: a bit silly
by Bob Arning 29 Apr '04

29 Apr '04

On Sun, 3 Aug 2003 23:15:05 +0000 (GMT) tim(a)sumeru.stanford.edu wrote: >I'm fairly sure that >+++++++ >directoryExists: filenameOrPath > "Answer true if a directory of the given name exists. The given name >may be either a full path name or a local directory within this >directory." > "FileDirectory default directoryExists: FileDirectory default pathName" > > ^(FileDirectory default directoryNamed: filenameOrPath ) exists >++++++++++++++ >would actually be correct Umm... maybe you meant +++++++ directoryExists: filenameOrPath "Answer true if a directory of the given name exists. The given name may be either a full path name or a local directory within this directory." "FileDirectory default directoryExists: FileDirectory default pathName" ^(self directoryNamed: filenameOrPath ) exists ++++++++++++++ Cheers, Bob

3 2

[BUG]FileDirectory & ServerDirectory cannot agree on assureExistenceOfPAth: or assurePathExists:
by Tim Rowledge 29 Apr '04

29 Apr '04

In 3.7-5607 - FileDirectory>assureExistence uses #assureExistenceOfPath: ServerDirectory>assureExistence uses #assurePathExists: The two methods are very nearly identical in fact and appear identical in intent. Perhaps the caretakers might like to agree on just one method? tim -- Tim Rowledge, tim(a)sumeru.stanford.edu, http://sumeru.stanford.edu/tim Strange OpCodes: CPM: Change Programmer's Mind

2 1

[FIX][KCP][3.6a] IfUnhandledError-nk
by ned＠bike-nomad.com 29 Apr '04

29 Apr '04

from preamble: "Change Set: IfUnhandledError-nk Date: 27 June 2003 Author: Ned Konz This changeset: - edits all callers of BlockContext>>ifError: to use something else, usually #ifUnhandledError: - adds BlockContext>>ifUnhandledError: (see below) - marks BlockContext>>ifError: as deprecated - edits all callers of Semaphore>>critical:ifError: to use something else. - marks Semaphore>>critical:ifError: as deprecated - edits most callers of on: Error do: [ ... ] to use something else, mostly #ifUnhandledError: - restores the behavior of removing step messages from the WorldState step list when the step messages have an error (the 'start stepping again' debug menu choice will restart them) I wrote #ifUnhandledError:, which allows enclosing scopes to handle exceptions. This behaves exactly the same as #ifError:, except that: - if there is an outer exception handler, that handler gets to handle the exception instead, and - there is now a one-argument form as well as the 0 and 2 argument forms that gets passed the exception itself. "!

2 1

MySql driver broken in 3.6g
by ar 23 Apr '04

23 Apr '04

the MySql driver 1.10 appears to be broken in 3.6g, due to the changes in Socket and SimpleClientSocket. (SimpleClientSocket was renamed to OldSimpleClientSocket and Socket was renamed to OldSocket). Simply changing the MySql driver to use the old classes is not sufficient because then SocketStream breaks. There may be more to it but that's as far as I got. Is this a known problem or has it been fixed already?

5 6

[ENH] Accufonts changeset candidate for 3.7alpha
by Doug Way 15 Apr '04

15 Apr '04

With the help of a bit of handy code from Ned, I've converted the Accufonts.sar file on SqueakMap with all of its bundled .sf2 font files into a single changeset which could be added to the update stream. The entire changeset is just over 100K, not too bad, although a little too large to post to this list, so I put it here: http://www.mindspring.com/~dway/smalltalk/InstallAccuFonts.8.cs.gz This is the candidate for inclusion in the 3.7alpha update stream. Note that the main point of the changeset is to remove the original Apple fonts (NewYork) from the image, so that the font clause in Squeak-L becomes irrelevant, allowing the possibility of making progress with sublicensing (or even re-licensing) Squeak. The replacement fonts look very similar so there shouldn't be much that's noticably changed. As a bonus, several extra bitmap fonts are thrown in. This also removes the ComicPlain/Bold bitmap fonts, which are not free font and were removed so that potential business users of Squeak have a "clean" image to work from font-wise. On the other hand, apparently there is some legal precedent recently that bitmap fonts are not copyrightable, so this may or may not really be an issue. Anyway, "Accujen" is a reasonable looking sans-serif replacement for the Comic fonts. This changeset does not remove the ComicSansMS TrueType font (yet), which probably should be removed, because outline fonts definitely are copyrightable, and this is not a free font either as far as I know. However, I was going to wait and possibly remove this one as part of incorporating the BitStream Vera TrueType fonts, so we have a bunch of replacement TT fonts before getting rid of this one. The "Atlanta" bitmap font, which someone donated to Squeak awhile back, remains. The Accufonts come with their own (very simple) license, included in the changeset preamble. This seems fine, fonts are a very different entity from source code, and it is not uncommon for software to provide a separate license for included fonts. There are still a couple of smallish issues with these fonts, but I wanted to get these in soon so that people can try them out, and we can fix/tweak things based on feedback from that. The problems I know of are: The character widths for some punctuation characters could use some adjusting... e.g. the Accuny period character is quite wide and looks funny IMHO. Also, some fonts such as Accuny09 and Accuny10 have the same size of 12, which can cause problems for some tools, such as Boris Gaertner's useful FontEditor tool. One odd last thing... I notice that the "Smalltalk garbageCollect" at the end of the postscript does not remove the old fonts from "StrikeFont allInstances", but doing a manual garbageCollect after filing in the changeset does. Maybe the ChangeSet object is somehow holding onto the fonts up 'til then, or something. No biggie I guess, we'll make sure garbage is collected sometime before 3.7 goes final. ;-) - Doug

4 7

[FIX] AddAllFix-ac
by andrew＠beta4.com 07 Apr '04

07 Apr '04

from preamble: "Change Set: AddAllFix-ac Date: 6 August 2003 Author: Andrew Catton Fixed bug where: someCollection addAll: someCollection causes an infinite loop. - #addAll*: methods now use a copy of the parameter if it is the same as the receiver. - The old behavior was moved to #basicAddAll*: variants. - #basicAddAll: is used by SortedCollection>>copy to avoid an #addAll:->#copy->#addAll: infinite loop when adding a collection to itself. - Removed OrderedCollection>>addAll: as it didn't actually change the semantics inherited from Collection>>addAll: and would have required a further override on SortedCollection had it remained."!

4 5

← Newer
1
2
3
4
5
6
7
...
40
Older →

Jump to page: