Squeak-dev March 1999

squeak-dev@lists.squeakfoundation.org

153 participants
338 discussions

HP Journada?
by mdr＠scn.org 23 Mar '99

23 Mar '99

Has anyone done a Squeak port for the HP Journada? How fast is it? The specs suggest it may be as fast as the Itsy, with keyboard and decent size display, and quick switch-on. Might make an ideal travelling machine. Mike p.s. Is there a set of benchmark results somewhere for various processors? I am specifically interested in relative performance of Alpha, PC, PowerPC, and palmtops.

2 1

Selector Changed
by William Barnett-Lewis 23 Mar '99

23 Mar '99

Hello all, Time for "Yet Another Newbie Question"... While (finally) loading all of the recent updates onto my PowerBook, I noticed many a "selector changed!" notifier box popping up. I looked at 'debug' on the first one and couldn't see anything that made sense to me so I simply proceeded from there and chose 'proceed' after that. The question I have is - was this the right thing to do? I would guess that part of the bug fixing process would need to change these selectors, but as documentation (esp. for newbies like me) is still lacking, I figured I better ask. TIA, William -- Live without fear; your Creator loves you as a mother. Go in peace to follow the good road and may God's blessing be with you always. St. Claire

1 0

Re: [Newbie Q.] MessageTally spyOn:
by Tim Olson 23 Mar '99

23 Mar '99

>What is this telling me? Is this an execution trace? What do the leading >numbers mean? The report is a tally of message sends and the percentage of overall time taken in each. This is done by setting a timer to interrupt the "spyed upon" process every so often and examine the execution context. > Are there any other neat tools like this in Squeak? Sure! For example, MessageTally also impelements "tallySends:", which simulates the block, counting the exact number of message sends. Try: MessageTally tallySends: [3.14159 printString] Maybe other folks will pitch in with their favorite "hidden tool" in Squeak. -- tim

1 0

RE: creating Complex
by Jarvis, Robert P. 23 Mar '99

23 Mar '99

Where can I pick up a Complex number class for Squeak? Was this in the updates since 2.3 (haven't gotten them - the firewall is my shepherd, I shall be safe no matter what)? Bob Jarvis The Timken Company > -----Original Message----- > From: Bert Freudenberg [SMTP:bert@isgnw.CS.Uni-Magdeburg.De] > Sent: Tuesday, March 23, 1999 10:07 AM > To: squeak(a)cs.uiuc.edu > Subject: Re: creating Complex > > On Tue, 23 Mar 1999, SLAWICK CHARLIER wrote: > > > I am a new squeak user (I use squeak to learn object oriented > > programming). > > To learn programming smalltalk, I actulally try to create a complex > number > > class. I've got a little problem: I want the programm to create a > complex > > when writing the command: > > 5+4i > > (for example); I've tried to add a method to the superclass 'number' but > I > > can't name it '+i' or '+ t1 i'... > > Is there a way for me to deal with this problem?... > > You can implement 'i' in Number returning a Complex with a real part of > zero. Then '5 + 4 i' would be 'Number + Complex' which should yield > another Complex by proper coercion (see Number's class comment). > > /bert > > -- > Bert Freudenberg Departement of > Simulation and > mailto:bert@isg.cs.uni-magdeburg.de Computer Graphics > http://isgwww.cs.uni-magdeburg.de/isg/bert.html Univ. of Magdeburg >

1 0

Re: creating Complex
by Bert Freudenberg 23 Mar '99

23 Mar '99

On Tue, 23 Mar 1999, SLAWICK CHARLIER wrote: > I am a new squeak user (I use squeak to learn object oriented > programming). > To learn programming smalltalk, I actulally try to create a complex number > class. I've got a little problem: I want the programm to create a complex > when writing the command: > 5+4i > (for example); I've tried to add a method to the superclass 'number' but I > can't name it '+i' or '+ t1 i'... > Is there a way for me to deal with this problem?... You can implement 'i' in Number returning a Complex with a real part of zero. Then '5 + 4 i' would be 'Number + Complex' which should yield another Complex by proper coercion (see Number's class comment). /bert -- Bert Freudenberg Departement of Simulation and mailto:bert@isg.cs.uni-magdeburg.de Computer Graphics http://isgwww.cs.uni-magdeburg.de/isg/bert.html Univ. of Magdeburg

1 0

Re: [Q] Why is JPG reading is so SLOW ?
by Tim Olson 23 Mar '99

23 Mar '99

>I'm newbie in squeak and just learned how to use file list. > And I tried to open JPEG file with open image in window(?) like > menu command, but it takes much time to display image. > That file is open less than 0.1sec in xv or gimp like program. > Why so slow ? And can I make it faster ? The JPEG Reader in Squeak is entirely written in Smalltalk for clarity, and, well, simply because it's nice to have a unified source language for everything, rather than having a lot of C code tacked on. JPEG decoding is a computationally expensive process, performing a lot of integer math on 8x8 macroblocks of the image, and Squeak isn't very fast on this type of code, because of the overhead involved in: * bytecode interpretation & dispatch * operand checking for SmallInteger arithmetic * overflow checking for SmallInteger arithmetic [Ian & Dan, how does Jitter-II perform on this?] However, Squeak also provides a mechanism to translate rudimentary Smalltalk code into a C primitive which can then be compiled into the VM. This was used by John Maloney for the sound generation primitives, for example. I believe that someone else on the list mentioned that they had converted the IDCT routine in the JPEG reader into a primitive, as well. The first step in making it faster is to know where the code is spending its time. Squeak supplies a great tool for this -- MessageTally. Try the following: MessageTally spyOn: [JPEGReadWriter formFromFileNamed: <your file here>] It will open up a Workspace window that looks something like: ----------------- - 195 tallies. **Tree** 100.0 JPEGReadWriter class(ImageReadWriter class)>>formFromFileNamed: 100.0 JPEGReadWriter class(ImageReadWriter class)>>formFromStream: 98.5 JPEGReadWriter>>nextImage 98.5 JPEGReadWriter>>nextImageDitheredToDepth: 34.9 JPEGReadWriter>>decodeMCU |34.4 JPEGReadWriter>>decodeBlockInto:component:dcTable:acTable: | 15.4 JPEGReadWriter>>decodeByteWithTable: | |8.2 JPEGReadWriter>>peekBits: | | |4.1 primitives | | |3.6 JPEGReadWriter>>fillBuffer | |4.1 JPEGReadWriter>>getBits: | | |2.6 SmallInteger(Integer)>><< | |3.1 primitives | 12.3 Array(SequenceableCollection)>>atAllPut: | 3.1 JPEGReadWriter>>getBits: | 2.6 JPEGReadWriter>>scaleAndSignExtend:inFieldWidth: 34.9 JPEGReadWriter>>idctMCU |34.9 JPEGReadWriter>>idctBlockInt:component: | 9.2 Interval(Collection)>>detect:ifNone: | |8.2 Interval>>do: | 5.1 JPEGReadWriter>>sampleRangeLimit: | 5.1 Interval>>do: | 5.1 SmallInteger(Number)>>to: | |3.6 Interval class>>from:to:by: | 4.1 SmallInteger(Number)>>negated | 3.6 SmallInteger>>bitShift: 27.7 JPEGReadWriter>>colorConvertMCU 27.7 JPEGReadWriter>>colorConvertIntYCbCrMCU 20.5 JPEGColorComponent>>nextSample 6.7 primitives **Leaves** 20.5 JPEGColorComponent>>nextSample 13.3 Interval>>do: 12.3 Array(SequenceableCollection)>>atAllPut: 7.2 SmallInteger(Number)>>negated 6.7 JPEGReadWriter>>colorConvertIntYCbCrMCU 6.7 SmallInteger(Integer)>><< 5.1 JPEGReadWriter>>sampleRangeLimit: 4.1 JPEGReadWriter>>peekBits: 3.1 JPEGReadWriter>>decodeByteWithTable: 2.6 SmallInteger(Integer)>>>> ------------- Here you can see that the JPEG reader spends it's time rougly equally in three main routines: colorConvertMCU idctBlockInt:component: decodeBlockInto:component:dcTable:acTable: The first two are prime candidates for conversion to primitives. -- tim

1 0

Re: multilingual Squeak
by ohshima＠is.titech.ac.jp 23 Mar '99

23 Mar '99

I'm sorry for my slow response. > > From: "Michael S. Klein" <mklein(a)alumni.caltech.edu> > > > > What about comparing two strings? > > > > If you got two strings each with one character that is the same code > > point in unicode, but the strings are in diferent encodings, are the > > strings equal, or not. > > > > In other words, for example (from the first Han character in Unicode) > > > > is U+4e00 = G(5027) ? > > is G(5027) = J(1676) ? > > > > U means Unicode 2.0 > > G means GB 2312-80 > > J means JIS X 0208-1990 They are 'not equal' to each other on the system which I think of. I think this is reasonable. > > For the Eurocentric amongst us (most of the list), > > > > is a Unicode $r the same as an ASCII $r ? > > is an English 'r' the same as a French 'r' ? The characters which can be mapped into 'A area' consitently may be mapped to the area. > > In defence of Unicode, it preserves round trip transcoding. > > It would be my standard of choice. Actually, "J -> U -> J" transcoding does not preserve the original text. The two 'reverse sollidus'es are mapped to the same code point and can't be reversed. OHSHIMA Yoshiki Dept. of Mathematical and Computing Sciences Tokyo Institute of Technology

3 2

RE: Floating point exceptions
by Jarvis, Robert P. 23 Mar '99

23 Mar '99

Given some of the oddities observed when the Float>>raisedTo: controversy arose awhile back, I don't think it's surprising to find that floating point calculations behave differently on different platforms. Irritating, aggravating, infuriating, and annoying certainly, but not surprising. :-} On NT I find that if I inspect the result of 1.7976931348623159e308 * 1.7976931348623159e308 (which is the same as MaxVal * MaxVal) I get Infinity. I think this means that the floating point code in the image somewhere is (correctly) determining that it's getting back an IEEE infinity, and is then creating a reference to the Infinity constant. Curiouser and curiouser... I suspect that the VM you're working with isn't trapping floating point errors (SIGFPE?), and thus the error is being kicked to the operating system with predictable results. Hunt around for a call to signal with SIGFPE as the signal name. It may be that it's not there, or that the signal handler isn't handling the errors correctly. Just a thought. <fifty-percent-joke> I think the obvious thing to do is to get rid of Floats. They cause more trouble than they're worth. </fifty-percent-joke> Bob Jarvis The Timken Company > -----Original Message----- > From: Tim Rowledge [SMTP:rowledge@interval.com] > Sent: Friday, March 19, 1999 5:17 PM > To: Squeak mailinglist > Subject: Floating point exceptions > > I'm having trouble with floating point class initialization. Hopefully > somebody > reading this is knowledgable enough to help me solve the problem. > > Float class>initialize creates a bunch of constants like Pi, MaxVal and so > on. > No problem there, but attempting to execute things like > Infinity _ MaxVal * MaxVal > causes a floating point exception and program exit. > > On the face of it, I'm not surprised; after all, we've just carefully > found the > maximum possible float and now we square it. Of course it overflows! > However, > Sq on W95 does not cause an exception, and presumably nor does Mac since > I've > heard no complaints from Mac users filing in the recent updates > (B3DExponentTable causes a similar exception). > > Is it really the case that the float libraries just ignore errors like > overflow? Is there some switch being set somewhere in the PC/MAc platform > code > that I haven't seen? Should Sq code assume that it is ok to square Float > MaxVal? Does ANSI/IEEE allow for this? Do unix VMs have similar trouble? > > tim > > -- > Software is mind work. Having the right frame of mind is essential. > Tim Rowledge: rowledge(a)interval.com (w) +1 (650) 842-6110 (w) > tim(a)sumeru.stanford.edu (h) <http://sumeru.stanford.edu/tim>

5 4

Squeak programming question
by kpgrant＠mindspring.com 23 Mar '99

23 Mar '99

Quick programming question from a Squeak/Smalltalk newbie: Is there any elegant method for handling occasions when you wish a method to return multiple objects? I'm writing a method that will return a set and a dictionary to the caller. I can't reasonably break up the method into two methods (one for each kind of data), and the contents of the set and dictionary are unrelated enough that it seems silly to make a whole new class of object containing nothing but pointers to each. The only thing that occurs to me is something like: ^((OrderedCollection new) add: myDictionary add: mySet) and then disassemble the thing on the receiving side. I'm not really crazy about this though. Re the discussion about turning this mailing list into a newsgroup. At the very least it would be nice if the mailing list were split up into sub-lists based on topic. There is a lot of traffic on the list that people like me, who aren't far enough along to even think about making improvements to the current environment, can't begin to follow. At the same time it seems kind of silly to post a "how do I do this in Squeak" kind of question to the whole list. A sub-list dedicated to programming questions would be nice. Thanks, Kevin

9 12

The mysteries of GC and WeakArrays....
by Peter Smet 23 Mar '99

23 Mar '99

I have set up a weakKeyDictionary (wkd) which appears to be working correctly. If I say: wkd at: Object new put: 'this' the GC reclaims the new Object, since it has no references. if I create a method as follows: | b | b := Bag new. wkd at: b put: 'this' and call this method the new Bag is not reclaimed, even though the temp var | b | has gone out of scope. However, if I now change the method to read: | b | b := Bag new. wkd at: b put: 'this'. b := nil. The b is reclaimed by the GC, and disappears from the wkd. Is this the correct behaviour? I know temp vars dont have to be set to nil to be reclaimed, so what is going on here?? If I don't set b to nil, is the b in my wkd ever going to be reclaimed? How do people program weakCollections such as dependents to get around this problem ? Another issue is blocks. Do blocks hold strong references to objects, or are all the references held by the enclosing scope? for eg: | a b | a := Set new. b := [a]. anotherProcedure: b if the block is passed to another procedure as a Parameter, does it now hold a strong reference to a, so that a cannot be GC'd? Thanks for any help Peter

1 0

← Newer
1
...
7
8
9
10
11
12
13
...
34
Older →

Jump to page: