Squeak-dev May 2002

squeak-dev@lists.squeakfoundation.org

160 participants
453 discussions

[hack/toy] Reified variables
by Vassili Bykov 02 May '02

02 May '02

As promised to a few people at Smalltalk Solutions, here is a hack I did a while ago to create an illusion of variables having some first-class status. See the file-out (known to work in 3.3a). The basic idea is that whenever there is a variable "foo" in scope, a keyword-like expression "foo:" (or "foo :") is allowed in the receiver position. It evaluates to an instance of (a subclass of) ReifiedVariable, which is an object that understands messages #value and #value: to get and set the value of the variable. This is pretty much all there is to it, plus a number of methods implemented in ReifiedVariable to do various variable-like things. For example, a := 3 actually works as (a:) = 3 where = is a message to the variable to assign the argument. (Sure, it clashes with comparison, but I figured that will do for a hack). Other things one can do are things like a: + 4 "increment the value of a by 4" a: decrement "decrement it by 1" p := a: "assign reified a to p (done by reified p!)" p + 10 "increment a by 10 through p" a: <> p: "swap a and p" See the example methods on the class side of ReifiedVariable. Of course, this is not nearly as cool as the "reifier" quote of Smalltalk-72, but FWIW... Cheers, --Vassili

1 0

Re: [RFI] Possible enhancements to work on at the Squeak-End
by goran.hultgren＠bluefish.se 02 May '02

02 May '02

Just wanted to remention a small detail I wrote about earlier: > 1. Add some form of filtering in the Hierarchy browser based on > categories/modules. When a newbie open up a hierarchy to find out which > class to use, say on "Morph" (cough) he/she drowns in morphs. And why? > Because lots of those classes are not in the Morphic categories - it > should be possible to filter out "application classes" like MinesTile > etc. I personally feel that this would make it much easier to understand Morphic for example which has extensive "reuse by inheritance" which "muddles the water". Anyway, just an idea. You guys might have better ideas! :-) regards, Göran

1 0

[Q] Regular Expresions
by Diego Gomez Deck 02 May '02

02 May '02

Hi... Where can I find a Regular Expression goodie for an actual version of Squeak? Any idea? TIA, Diego Gomez Deck

2 1

[Fix] ClassBuilder (Obsolete Classes) for 3.3a
by Henrik Gedenryd 02 May '02

02 May '02

> Thus, we have created a new version of the fix that incorporates both my and > his changes. This version works with Squeak 3.2gamma (4827), but it does not > work with Squeak 3.3 yet. As I said before, I'm going to look at that with > Henrik next week and try to make it compatible. > > Cheers, > Nathanael Here is a version that we have tested with 3.3a-4843 and 4827, early 3.3a images will not work (e.g. 4769). It should be fine for going into the update stream. Henrik & Nathanael

1 0

Re: Will the real sqUnixSocket.c step forward?
by goran.hultgren＠bluefish.se 02 May '02

02 May '02

"Lex Spoon" <lex(a)cc.gatech.edu> wrote: [SNIP] > Okay, Ian has patiently explained to me how the semaphore notification > stuff is supposed to work, and it turns out that the SourceForge code is > really a mess. Basically, the following line in notify(): > > int mask= pss->pendingEvents | eventMask; > > would cause all events to be signalled that the image is interested in, > whether the event has happened or not. Changing the "|" to the proper > "&" (duh!) causes many events *not* to be signalled. This led to a > cascade of changes to make sure that pendingEvents has all the flags it > is supposed to, and that select() gets called on all the events listed > in pendingEvents. This is all posted on SourceForge now. Super! I am rebuilding as we speak. > Goran's example code works properly, Scamper works, and with any luck > email sending works. :) Hopefully this is better on the whole, but > there may well be new problems. It's tricky stuff, and the code is less > than clear. > > The changes are posted on SourceForge. I haven't looked at Ian's > version, but there's a good chance that it has similar issues. Yes - the "|" bug was in his original code. He sent a patch to the list with the "|"->"&" change plus others. Did you check it? > Lex Spoon regards, Göran

1 0

Re: Drag and drop in 4843???
by Blaine Buxton 02 May '02

02 May '02

Hey, I also noticed this today and I also noticed that when you select a piece of code anywhere (workspace, transcript, browser pane) and select "Do it" from the menu that the highlighting disappears around the text. Not a big thing, but the drag and drop would be nice to have back....=) >From: stephane ducasse (home) <ducasse(a)iam.unibe.ch> >Reply-To: squeak-dev(a)lists.squeakfoundation.org >To: squeak dev <squeak-dev(a)lists.squeakfoundation.org> >Subject: Drag and drop in 4843??? >Date: Wed, 1 May 2002 21:33:24 +0200 > >Hi > >I just want to know if this is my system that is in a strange state. I >cannot drag and drop method in different categories. I'm in 4843 > >Stef > > ------------------------------------------------ Blaine Buxton http://www.mp3.com/blainebuxton "Think! It ain't illegal yet."-George Clinton _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp.

2 1

WinCE on HPC/Pro MIPS?
by Ned Konz 02 May '02

02 May '02

I got the WinCE VM for H/PC Pro on MIPS from http://www.is.titech.ac.jp/~ohshima/squeak/WinCE/SqueakVM-alpha6-PocketPC-M… and then installed it and a 6Mb stripped down 3.2g image on my CompactFlash card on my WinCE machine (NEC MobilePro 770). This identifies itself as: Microsoft Windows CE, Handheld PC Edition Version 3.0: Core system version 2.11 When I start the VM it asks for the name of the image file. I then select the image and it tells me that it can't find the image. It says: ---- Squeak! Could not open the Squeak image file '\Storage Card2\My Documents\Squeak\Squeak.image' There are several ways to open the Squeak image file. You can: 1. Double-click on the desired image file. 2. Drop the image file onto the Squeak application. Aborting... ---- Same thing (minus asking me for the file name) happens when I supply the full name of the image at the command line. And (contrary to the requestor text above) I can't start it by double-clicking or dropping. I have tried putting the image, changes, and sources file as well as the VM in both \Storage Card2\S\ and \Storage Card2\My Documents\Squeak\ Does anyone have any suggestions? I assume this is the right VM because it at least starts and asks for an image. I have had Squeak running on this machine before, but a different image and VM. I have 27Mb free space for program memory, and 5.3Mb free space on the RAM-based pseudo-drive. The CF card "\Storage Card2" has 30Mb free. I would very much like to get this going before spending 5 hours on a plane tomorrow. Thanks, -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE

4 8

Re: Sorting oddity - feature or ??
by Richard A. O'Keefe 02 May '02

02 May '02

Bob Arning <arning(a)charm.net> wrote: Is the following a known and accepted feature of our sorting algorithm? [((1 to: 10000) collect: [ :x | x]) asSortedCollection] timeToRun ==> 98 [((1 to: 10000) collect: [ :x | 3]) asSortedCollection] timeToRun ==> 20826 Array>>asSortedCollection results in a new empty SortedCollection is made (that) addAll: (the Array) is done This just dumps all the new elements in then sorts everything. The method used is one I'm not familiar with, but it's a variant of quicksort. Quicksort variants that don't use fat pivots are known to perform very badly when there are lots of equal elements. Quicksort was invented for a machine whose main storage capacity was 256 words (each 40 bits). Not 256 M, not 256 k, just two hundred and fifty six words of data. The inventor, Hoare, KNEW it was inferior, in the abstract, to a well-written merge sort, which to sort N items needs N/2 item workspace. He also KNEW that he didn't _have_ any spare memory, so quicksort was much faster than a method he couldn't use. Squeak doesn't run on those machines. Squeak can afford the memory to use merge sort. A sorting method with a guaranteed worst case that's better than the _average_ case of most quicksorts is not to be sniffed at. (Yes, I have read the papers and books which tell you that quicksort is blindingly fast. But Sedgewick's Acta Informatica paper assumed that the cost of #< was one quarter of the cost of a memory reference. That's not true in Squeak, and seldom true in most sorting applications.) Alternatively, if it's thought that Squeak _must_ use O(lg N) space sorting, perhaps the algorithm from Bentley & McIlroy "Engineering a Quick Sort", Software Practice and Experience about 8 years ago, could be adapted.

1 0

Drag and drop in 4843???
by stephane ducasse 02 May '02

02 May '02

Hi I just want to know if this is my system that is in a strange state. I cannot drag and drop method in different categories. I'm in 4843 Stef

2 1

Re: rounded corners (was: Re: straw-man 3.2 default preferences)
by Martin Drautzburg 02 May '02

02 May '02

Henrik Gedenryd <h.gedenryd(a)open.ac.uk> writes: > The argument for having rounding on at all in the first place should really > have nothing to do with speed. But a speed problem would add more weight > against them. Has anybody noticed that rounded corners on raised borders don't look right ? Especially the upper left corner looks like a mouse has eaten some pixels.

1 0

Jump to page: