Squeak-dev June 2009

squeak-dev@lists.squeakfoundation.org

80 participants
92 discussions

Faster Sets Project
by Ralph Boland 24 Jun '09

24 Jun '09

I finally released my first project to the www.squeaksource.com repository. Its called FasterSets. It makes sets faster by not doing compares during grow operations which makes adding to sets use about 14% fewer compares on average during an add: operation assuming no deletions or preallocation of space before adding elements. Code for measuring performance is included. See Class FastSetsComment. Comments welcome. I consider the project complete but am willing to make changes as required or give permissions for others to do so or to take over the project. My hope is that this code will be incorporated into Squeak someday. I have not signed any release forms to allow this but am willing to do so. The project runs in Squeak 3.10.2 but can be easily modified to work in previous versions of Squeak. Regards Ralph Boland

4 3

Problems building squeak VM (Fedora core 11)
by Casimiro de Almeida Barreto 24 Jun '09

24 Jun '09

Hello, I've had problems building VM. Apparently the file $SOURCE_ROOT/platforms/unix/src/vm/debug.c and debug.h were adjusted to handle a Mac OS X problem regarding to dprintf. In file debug.c it is necessary to include stdlib.h to handle abort(). But then, dprintf is polymorphic and is not properly handled: gcc -g -O2 -fomit-frame-pointer -DLSB_FIRST=1 -DHAVE_CONFIG_H -DSQUEAK_BUILTIN_PLUGIN -I/home/casimiro/Softwares/squeak/squeak-svn-sources/bld -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/Cross/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/src/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/Cross/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/src/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/Cross/plugins/FilePlugin -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/plugins/B3DAcceleratorPlugin -I/home/casimiro/Softwares/squeak/squeak-svn-sources/bld -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/Cross/vm -I/home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/src/vm -c -o debug.o /home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/vm/debug.c In file included from /home/casimiro/Softwares/squeak/squeak-svn-sources/platforms/unix/vm/debug.c:3: /usr/include/stdio.h:399:66: error: macro "dprintf" passed 3 arguments, but takes just 1 make[1]: ** [debug.o] Erro 1 make: ** [vm/vm.a] Erro 2 Problem code at debug.h is: #if (DEBUG) /* the thing to use here is a variadic macro, but Apple's gcc barfs on ** them when running in precomp mode. did they _really_ have to break ** the preprocessor just to implement precomp? good _grief_. */ extern void __sq_dprintf(const char *fmt, ...); # define dprintf(ARGS) __sq_dprintf ARGS #else # define dprintf(ARGS) ((void)0) #endif

1 0

Faster Sets Project: Response to comments.
by Ralph Boland 23 Jun '09

23 Jun '09

First, as suggested I have changed the license to MIT. The follow suggestion was made: if you would like your new implementation to replace the old one in Squeak, you might want to write unit tests. These tests ensure that there is no problem. FasterSets includes only one new class used for measuring the performance improvement of the other changes. All of the remaining improvements are changes to class Set and its many subclasses. Allow me to consider 4 subsets of the classes Set and all of its subclasses. 1) (Set WeakIdentityKeyDictionary Dictionary) These classes have unit tests. I have run them and they all pass with FasterSets. 2) ComparesCountingSet This is the only new subclass of Set that I have added. It only measures performance and will disappear once the FasterSets project is incorporated into Squeak. Is is exercised sufficiently in doing its performance measurements that I do not feel that it needs a unit test. 3) (WeakKeyDictionary IdentityDictionary WeakSet KeyedSet PluggableDictionary WeakKeyToCollectionDictionary WeakValueDictionary PluggableSet MethodDictionary LiteralDictionary IdentitySet SystemDictionary KeyedIdentitySet) These are the subclasses of Set that have no unit tests. If unit tests for these classes are added to Squeak I will run FasterSets code against them and fix any FasterSet bugs I find. I have no desire to write them though. I don't know what most of them do and feel these tests are best written by someone who does. Even if all these test existed and passed with FasterSets it wouldn't establish that FasterSets has not introduced bugs. I have tested them in a rather poor way. I have been running Squeak using FasterSets since Squeak 3.6 and, as it happens, my application uses Set and a number of its subclasses A LOT! So some of these subclasses should be considered reasonably well tested (Set, Dictionary,IdentitySet). The others have been tested to the extent that they have been exercised by my using Squeak. However most of the work I do is algorithmic (finite state machine construction mostly) and so I'm sure large sections of Squeak are not exercised or not exercised very well by me. So there is real potential that anyone who loads FasterSets into their image will discover bugs and I recommend caution to anyone who loads FasterSets. Fixing bugs though should be easy. I will fix any reported to me or give permission to anyone who wants to to make and install the fix themselves. This may all sound very alarming but keep in mind that FasterSets actually contains very little code and not very complicated code at that. Caution is warranted because the classes modified are all low level and some are extensively used. The real problem though is, I suspect, some are not extensively used. 4) (WeakKeyToCollectionDictionary WeakValueDictionary PluggableSet LiteralDictionary SystemDictionary KeyedIdentitySet) These subclasses of Set have not been modified in FasterSets; in other words most have. Even the unmodified ones have superclasses that have been modified so all need to be tested. In fact the only reason for classes not being modified is because they inherit the necessary added behavior. Thus in my opinion the only classes that have been adequately tested are Set, Dictionary, and IdentityDictionary. If someone wants to claim that the five tests run by WeakIdentityKeyDictionaryTest are sufficient to claim WeakIdentityKeyDictionary is adequately tested I can add it to the list?. If anyone can send me a list of classes adequately tested just by running Squeak I will add them to the list. If I ever get to the point that all subclasses of Set are claimed to be adequately tested by somebody I will make a posting state this and who claims the tests are adequate. If anybody wishes to write any of the missing unit tests and add them to FasterSets until such time as they can be added to Squeak such efforts are most welcome. A final comment. I am most interested in getting FasterSets incorporated into Squeak and that means having FasterSets being properly tested. However I don't consider myself to be the best person to do this work though I will help if I can. The best persons to do so are those who can load FasterSets into their image and not be too upset if they get blown out of the water! They also must have images they can claim exercises one or most of the Set subclasses adequately. Regards, Ralph Boland On Sat, Jun 20, 2009 at 3:21 PM, Ralph Boland<rpboland(a)gmail.com> wrote: > I finally released my first project to the www.squeaksource.com repository. > Its called FasterSets. It makes sets faster by not doing compares during > grow operations which makes adding to sets use about 14% fewer compares > on average during an add: operation assuming no deletions or preallocation > of space before adding elements. Code for measuring performance is included. > See Class FastSetsComment. > > Comments welcome. > > I consider the project complete but am willing to make changes as required > or give permissions for others to do so or to take over the project. > > My hope is that this code will be incorporated into Squeak someday. > I have not signed any release forms to allow this but am willing to do > so. > > The project runs in Squeak 3.10.2 but can be easily modified to work > in previous > versions of Squeak. > > Regards > > Ralph Boland > >

1 0

Cuis: can't change the method category
by Igor Stasenko 23 Jun '09

23 Jun '09

I have a problem. I can't change the method's category in browser by dragging & dropping it in list(s). -- Best regards, Igor Stasenko AKA sig.

4 5

(no subject)
by Ken G. Brown 22 Jun '09

22 Jun '09

Undo, undo! Sighhh... Please forgive me, this was sent totally by accident from my iPhone while reading emails. Not exactly sure why, clicking around too fast I guess. Ken At 5:00 AM -0700 6/22/09, squeak-dev-request(a)lists.squeakfoundation.org apparently wrote: >Date: Sun, 21 Jun 2009 09:36:57 -0600 >From: "Ken G. Brown" <kbrown(a)mac.com> >Subject: [squeak-dev] (no subject) >To: "squeak-dev(a)lists.squeakfoundation.org" > <squeak-dev(a)lists.squeakfoundation.org> >Message-ID: <842A5FD2-4DD8-48F1-BF66-D9CD871EDF93(a)mac.com> >Content-Type: text/plain; charset=us-ascii; format=flowed > > > >Ken

1 0

[Chronology] Is this a bug or a feature?
by David T. Lewis 21 Jun '09

21 Jun '09

DateAndTime fromSeconds: 0.5 ==> walkback Prior to Mantis 4669 (which is now included in Squeak 3.10.2) the argument to #fromSeconds: could be a non-integer value, but now it is required to be an integer. Is this a bug or a feature? I have relied on the original behavior in my TimeZoneDatabase package, because in this context I was expecting "seconds" to represent elapsed time since the epoch. However, it is also quite common to treat "seconds" as integer numbers in other contexts, so I don't know if my expectation was appropriate. I took a look at a Squeak 3.6 image (prior to the addition of class DateAndTime), and we have this: Time dateAndTimeFromSeconds: 0.5 ==> #(1 January 1901 12:00:00.5 am) Time fromSeconds: 0.5 ==> 12:00:00.5 am But in later Squeaks with Chronology, we have: Time fromSeconds: 0.5 ==> 12:00:00 am So I am inclined to think that the original Squeaky behavior implies that #fromSeconds: should accept non-integer numbers, in which case the walkback in DateAndTime class>>fromSeconds: would be a bug. Opinions? Dave

1 0

(no subject)
by Ken G. Brown 21 Jun '09

21 Jun '09

Ken

1 0

displaying attributed characters on workspaces
by Ralph Boland 20 Jun '09

20 Jun '09

I have some code that draws trees using '|' and '-' characters to link nodes in the tree. For example: 8 8---| | 7 8---| | | 6 | 6---| | 5 8---| | 4 | 4---| | | 3 4---| | 2 2---| 1 I have a number of problems when I display the tree in a workspace (or debugger for that matter). 1) I need a fixed font so that symbols are properly aligned. This problem can be seen in the above figure but the effect is worse in Squeak. 2) I also need to add some additional information to my trees. The addition information is text ('|' and '-' characters linking nodes in the tree) and needs to be in color (Red) in order avoid making the figure too confusing. Thus I need to display attributed characters. Workspaces can handle this but I need for the 'printOn:' method to handle them as well. 3) I would like to add my own special characters to use instead of '|' and '-' so that there are no gaps between these special characters when used in my tree drawings. I know how the code works to draw the return and assignment arrows but I do not know if I need to use special fonts so that there are no forced gaps between characters and lines of text and if so which ones. Is it possible for me to do these things and have my tree printed accordingly when my tree is displayed in a workspace (or debugger) resulting from a 'printOn: aStream' message? I know that the stream I get when the printOn: is invoked is a LimitedWriteStream. but not much more. If it is not possible to do these things then I will settle for displaying it in a Text Pane. I still need to know which font to use though. The code is part of my implementation of the Spine Tree Decomposition data structure which I will release to www.squeaksource.com when it is completed. Any help most appreciated. Ralph Boland

2 1

What can be an actionSelector
by Gary Dunn 20 Jun '09

20 Jun '09

Thanks for all the help getting me started with Squeak. I decided to play around with the morphic development style rather than working directly in the Class Browser. My goal is to develop techniques for developing apps for the Open Slate project. I have always wanted a HyperCard-like application to use for this purpose and am delighted to discover that a StackMorph does just that. Having created a minimal stack, I set out to learn how to provide navigation controls. I placed a buttom morph on the background used later in the stack with a target of the name of the StackMorph and an actionSelector of goToFirstCardOfStack. This works. But when I put a button on the first page, which has a different background, and make its actionSelector goToNextCardInStack, nothing happens. If I make it flash, the default, the book flashes, so the target is correct. Any idea why this is not working? Is a background a stack? I thought all the pages in a StackMorph were one stack. The book's navigation controls work as expected, and the menu uses goToNextCardInStack. I have another unrelated issue. I have Squeak installed on two computers. Both use the same version, 3.9, and both are running FreeBSD and Gnome but slightly different versions. On one, morphs dropped onto a stack page disappear when anything is right-clicked (yellow). On the other one they don't. I can get them to stick by adding them to the background, but I want them to be unique to that page. What am I doing wrong? -- Gary Dunn, Honolulu osp(a)aloha.com http://openslate.net/ http://e9erust.blogspot.com/ Sent from Slate001

2 2

Ideas about shrink more MinimalMorphic
by Edgar J. De Cleene 19 Jun '09

19 Jun '09

Working side to side with Cuis, SqueakLightII and MinimalMorphic I found: SystemOrganization classifyAll: #(#PluggableTabButtonMorph #DropDownChoiceMorph #LedTimerMorph #TickIndicatorMorph #SimpleSwitchMorph #DumberMenuMorph #PluggableMultiColumnListMorphByItem #FloatingBookControlsMorph #BasicButton #PianoRollScoreMorph #ListComponent #EToyProjectDetailsMorph #StringButtonMorph #CircleMorph #ScaleMorph #FunctionComponent #PrintComponent #SquishedNameMorph #FlexMorph #ClickExerciser #TransitionMorph) under: 'Morphic-NotInCuis'. (MCPackage named: 'Morphic-NotInCuis') unload. So what¹s the criteria? Morph allSubclasses in all. List of classes with allCallsOn size = 0 and allSubclasses size = 0 And manual select some sure at some time you need. Anyway, both SqueakLightII and MinimalMorphic could load classes from repository if need to http://www.morphle.com/~edgar/LoadingSomeUnknow.mov You know 3.10 don¹t have .mp3 and .mpg play right ? And how you have this again ? The .mov shows the pre-alpha of this, how drag and drop a .morph coming from any 3.8, 3.9 or 3.10 image and asking to reload ³missed² classes from server. No need of know nothing about Monticello or repositories... Edgar

1 0

← Newer
1
2
3
4
5
6
7
8
9
10
Older →