Squeak-dev May 2000

squeak-dev@lists.squeakfoundation.org

137 participants
680 discussions

Re: [BUG] DSA failure in LargePositiveInteger in #new: <primitive: 71>
by Bob Arning 30 May '00

30 May '00

On Tue, 30 May 2000 11:54:10 -0400 (EDT) Bob Arning <arning(a)charm.net> wrote: >Here is at least part of the answer. Inspect the following: > >zz _ LargePositiveInteger new: 1. >zz digitAt: 1 put: 1. >a _ zz = 1. >b _ zz printString. >c _ zz = 1. >{a. b. c} > >As far back as Squeak 2.5, I get (true '1' false ). Well, a slight correction - with the very latest Mac 2.8 VM (Squeak VM 2.8beta) and LargeIntegers it produces: (false '1' false ) With the previous VM (Squeak VM 2.8-NOT) it returns (as it does in 2.7 and 2.5) (true '1' false ) So the (or at least, one) difference is how #= is handled for "short" LargeIntegers with and without the plugin. Given this difference (plus who knows how many other places) and the class comment of LargePositiveInteger pointed out by Stephan, perhaps we should prohibit the creation of invalid LargeIntegers. I tried this : LargePositiveInteger class>>new: digits digits < 4 ifTrue: [self halt]. ^super new: digits and this trapped the DSA code much earlier. Looks like it will take some work there to be fully compliant with the new LargeIntegers. Cheers, Bob

1 0

Re: Squeak BOF and demos at ECOOP 2000
by Dan Ingalls 30 May '00

30 May '00

>from 12 to 16 june, the 14th ECOOP takes place in Cannes (France). On that >event, we would like to make noise concerning Squeak. We are thinking on >organizing a BOF-like meeting of all available squeakers and people that are >interested. The idea is first of all to meet with as many people as possible >(to learn to know each other better), to have some discussion and to see >some demoes from fancy stuff done with Squeak. BUT therefore I would like to >have an idea of how many people that are coming to ECOOP and are willing to >come to such event (and what demoes they could do, or what help they can >offer in some way). Please respond if you are coming, because if there are >not enough squeakers there, then I'll just drop the idea. All of Squeak Central (except for Ted who, with Cynthia, just had triplets!) will be at ECOOP. Most of us will arrive Wednesday and depart on Sunday. I don't get in until around 7pm on Wed (and that's assuming numerous connections all go well), but if there were a plan to gather at some convenient spot around 9, I would try to be there. Scott and John arrive earlier, so perhaps they, or some of our European Squeakers, could make a plan, once they are at the conference. We have not made any formal plans for a BOF, but it shouldn't be too hard to find a room and a time that will work for most of us. I think any needs for a larger forum should be pretty well met by Alan's talk Thursday evening. - Dan

1 0

Re: [BUG] ChangeSorter may forget changes
by Steve Wessels 30 May '00

30 May '00

I just checked and this bug does not happen in V2.7 - Steve Bob Arning wrote: > If you have a file containing > > 'From Squeak2.8alpha of 13 January 2000 [latest update: #2204] on 29 May 2000 at 7:51:26 pm'! > Object subclass: #Test1 > instanceVariableNames: 'var1 var2 var3 ' > classVariableNames: '' > poolDictionaries: '' > category: 'Tools-Browser'! > > and you > > 1. File it into a new change set. > 2. Delete var2 > 3. File it out > > then you get the expected output with only var1 and var3. However, if you > > 1. File it into a new change set. > 2. Make it the current change set > 3. Delete var2 > 4. File it out > > then var2 is still included in the fileout. > > Cheers, > Bob

1 0

POP/SMTP without user interaction?
by Stefan Matthias Aust 30 May '00

30 May '00

Hi! Has anybody already changed things like POPSocket and SMTPSocket so that they'll work in a GUI-less environment? At the moment, both classes will prompt the user in #waitForDataQueryingUserEvery: which is, well, not the best idea. I also dislike that this method will emit errors using #error: instead of generating more appropriate SocketExceptions. bye PS: The same question needs to be asked for FileStreams which also assume that there's always a user to answer questions... -- Stefan Matthias Aust Projektleiter/Softwareentwicklung Baltic Online Computer GmbH, Alter Markt 1-2, 24103 Kiel Fon: +49 (0) 431-54003-0 Fax: -99 http://www.baltic-online.de

2 1

Re: Copyright issues [was: Re: Who wants to refactor #deepCopy?]
by Andrew C. Greenberg 30 May '00

30 May '00

>How far can someone 'copy' an existing implementation of a commercial >Smalltalk? >- Totally? >- Just rename some vars? >- Change the class structure? >- Change some algorithms? The following provides broad summaries of U.S. law only. Your mileage may vary. Copyright protects expressions of ideas, and not ideas itself. To the extent a program implements a particular functionality, that functionality may be recreated without reference to the expression. If the expression is only one of a few limited ways in which the functionality may be implemented, it probably isn't protected by copyright at all (the so-called "merger doctrine" where the idea merges with the expression). Still further, you can only infringe if you are guilty of copying or deriving from the original work. If you are presented with a detailed specification, even if derived from an existing work, that embodies only the ideas of the work, and develop a program from the specification without reference to the original (a so-called "clean-room" implementation), there is no infringement. (Practice this only witht the guidance of competent counsel -- "clean rooming" is difficult to do right, and a disaster if done wrong). So, while there is great latitude to borrow concepts and ideas from an existing functional work, any of the preceding acts from the original post -- total copying or changes with just a few minor mechanical variations, or even derivations with sophisticated variations -- may well be infringement. In short, the devil is in the details. Don't count, however, on appropriating expression from a work, with or without changes, and then incorporating portions of it directly in your own. You are free to borrow the IDEAS, not the expressions. -- Andrew C. Greenberg acg(a)netwolves.com V.P. Eng., R&D, 813.885.2779 (office) Netwolves Corporation 813.885.2380 (facsimile) www.netwolves.com

2 1

Re: Squeak BOF and demos at ECOOP 2000
by Stephan Rudlof 30 May '00

30 May '00

I'm coming to ECOOP. BTW: It's my first conference of this kind and I've never visited a BOF-like meeting so far... Greetings, Stephan Roel Wuyts wrote: > > Hello folks, > from 12 to 16 june, the 14th ECOOP takes place in Cannes (France). On that > event, we would like to make noise concerning Squeak. We are thinking on > organizing a BOF-like meeting of all available squeakers and people that are > interested. The idea is first of all to meet with as many people as possible > (to learn to know each other better), to have some discussion and to see > some demoes from fancy stuff done with Squeak. BUT therefore I would like to > have an idea of how many people that are coming to ECOOP and are willing to > come to such event (and what demoes they could do, or what help they can > offer in some way). Please respond if you are coming, because if there are > not enough squeakers there, then I'll just drop the idea. > > Thx > > -- > Roel Wuyts Programming Technology Lab > rwuyts(a)vub.ac.be Vrije Universiteit Brussel > http://prog.vub.ac.be/~rwuyts > Webmaster of European Smalltalk User Group: www.esug.org -- Stephan Rudlof (sr(a)evolgo.de) "Genius doesn't work on an assembly line basis. You can't simply say, 'Today I will be brilliant.'" -- Kirk, "The Ultimate Computer", stardate 4731.3

1 0

EToy status update?
by Eric Scharff 30 May '00

30 May '00

This is somewhat a followup to the recent postings about Logo and HyperCard. I have a tremendous amount of personal interest in this style of programming. It seems that there are fragments of things all over the place, like the current EToy stuff (the scripting tiles, ...), Fabrik, the HyperCard-like features in BookMorphs, the Wonderland language, and probably others I know nothing about. I remember a conversation a while back about Squeak versus, say, Macromedia Director that also gets at similar issues. (Folks doing this stuff are probably aware of the AgentSheets / Visual AgenTalk work done by colleagues of mine here at Colorado. I choose both not to speak for them and also to maintain my neutrality, but it is an intriguing approach very much like the EToy stuff.) I'm not necessarily recommending merging all these things into one glorious unified thing and calling it the ideal end-user programming language or something. Squeak/Smalltalk itself provides a very good glue for various different approaches. One challenge in the current system is knowing whats there, and knowing the status of these various endeavors. What is the current state of all of these different things? In particular, what is the future of the EToy stuff? We've heard a lot (and even now see) the advances in Morphic, Image Segments, and other such things. However, I really like the EToy stuff (to the limited extent I understand it; basically exploration based on Demos I've seen the SqC folks do) and I would like to see that work continue. I find the connection between scripted objects that dynamically generate Smalltalk behavior particularly nice, because it leads to different levels of abstraction (script the objects, or change the script primitives, or re-code directly in Smalltalk depending on what you want to accomplish.) Any information about the current status and direction of the EToy stuff would be great (at least for me, he says selfishly... :) ) -Eric

2 1

Re: [BUG] DSA failure in LargePositiveInteger in #new: <primitive: 71>
by Bob Arning 30 May '00

30 May '00

On Tue, 30 May 2000 09:58:42 EDT JArchibald(a)aol.com wrote: >See if you can evaluate 'DigitalSignatureAlgorithm generateKeySet'; MVC or >Morphic. > Jerry, Here is at least part of the answer. Inspect the following: zz _ LargePositiveInteger new: 1. zz digitAt: 1 put: 1. a _ zz = 1. b _ zz printString. c _ zz = 1. {a. b. c} As far back as Squeak 2.5, I get (true '1' false ). The gotcha here is that LargeIntegers are assumed (in #printOn:base:, anyway) to have more than one digit. If not, then some destructive operations happen on the receiver (zz in this case). Where else this happens, I couldn't say. Why this seems to be different with/without the LargeIntegers plugin is, I suspect, that results of operations may be converted to SmallIntegers in one case and remain as LargeIntegers in another. These "short" LargeIntegers seem to be cropping up when DSA fails. Cheers, Bob

1 0

Re: Large array slows down Squeak
by Stefan Matthias Aust 30 May '00

30 May '00

Yoshiki Ohshima wrote: > Smalltalk garbageCollect. Thank you, you saved my day :-) bye -- Stefan Matthias Aust Projektleiter/Softwareentwicklung Baltic Online Computer GmbH, Alter Markt 1-2, 24103 Kiel Fon: +49 (0) 431-54003-0 Fax: -99 http://www.baltic-online.de

1 0

Large array slows down Squeak
by Stefan Matthias Aust 30 May '00

30 May '00

Hi, I just noticed that an Array new: 10000000 "ten millions" noticablely slows down Squeak (open it in an inspector or assign it to a global (or workspace) variable). Actually, it's now nearly impossible to work with Squeak. If I do a ByteArray new: 40000000 which should take the same amount of memory, no such effect occurs so I guess it's a garbage collector issue. Right? Is there a workaround? bye -- Stefan Matthias Aust Projektleiter/Softwareentwicklung Baltic Online Computer GmbH, Alter Markt 1-2, 24103 Kiel Fon: +49 (0) 431-54003-0 Fax: -99 http://www.baltic-online.de

2 1

← Newer
1
...
5
6
7
8
9
10
11
...
68
Older →

Jump to page: