Squeak-dev April 2002

squeak-dev@lists.squeakfoundation.org

195 participants
665 discussions

[ANN] AspectS (0.4.1)
by Robert Hirschfeld 30 Apr '02

30 Apr '02

AspectS 0.4.1 --- There is an updated version of AspectS (0.4.1). Besides several small fixes and enhancements, there is a new example that illustrates how to use AspectS to introduce different perspectives on the system, depending on different contexts. The example implements the AspectS tool support, previously done via subclassing of existing browser and support classes, now as an additional layer on top of the browser and support classes. The tools aspect is, amongst other characteristics, sensitive to projects as its context: To create a tools aspect instance:- aspect _ AsAspectToolsAspect new. To install an aspect:- aspect install. To make the tools aspect aware of a certain project:- aspect addProject: (Project named: 'Test'). aspect addProject: Project current. To make the tools aspect unaware of a certain project:- aspect removeProject: Project current. aspect removeProject: (Project named: 'Test'). To uninstall the aspect:- aspect uninstall. aspect _ nil. The new version (a change set and pre-configured image) is available at http://www.prakinf.tu-ilmenau.de/~hirsch/Projects/Squeak/AspectS/ Comments are welcome. Best, -Robert

3 3

Re: [Q][morphic] Collapsing menus. Was: how to hide submorphs
by Martin Drautzburg 30 Apr '02

30 Apr '02

Ned Konz <ned(a)bike-nomad.com> writes: > If you turn on clipping in the parent, the submorphs outside the > parent bounds shouldn't be a problem (because they won't be visible). > > Also if the parent overrides containsPoint: to return false when the > point is outside its bounds, you won't be able to interact with the > submorphs. Aaron J Reichow <reic0024(a)d.umn.edu> writes: > Send it #hide and #lock. That'll make it disappear and no longer recieve > events. I tried the code below, but #clipSubmorphs: true still lets me drag the owner morph from the invisible rectangle below. If I adjust fullBounds to bounds I can no longer grab, but I have trouble getting the hidden morph repainted after an #expand. Sending a self changed from an inspector does repaint it but the same message in the #expand method seems to have no effect. I may be completely on the wrong track ... how would *you* collapse a menu ? 'From Squeak3.2gamma of 15 January 2002 [latest update: #4843] on 30 April 2002 at 8:58:35 pm'! RectangleMorph subclass: #Collapser instanceVariableNames: 'child ' classVariableNames: '' poolDictionaries: '' category: 'dtz-tries'! !Collapser methodsFor: 'as yet unclassified' stamp: 'dtz 4/30/2002 20:56'! collapse self clipSubmorphs: true. fullBounds _ bounds ! ! !Collapser methodsFor: 'as yet unclassified' stamp: 'dtz 4/30/2002 20:58'! expand self clipSubmorphs: false. self layoutChanged. self changed. ! ! !Collapser methodsFor: 'as yet unclassified' stamp: 'dtz 4/30/2002 20:55'! grabableCollapse self clipSubmorphs: true. child lock; hide. ! ! !Collapser methodsFor: 'as yet unclassified' stamp: 'dtz 4/30/2002 20:57'! grabableExpand self clipSubmorphs: false. ! ! !Collapser methodsFor: 'as yet unclassified' stamp: 'dtz 4/30/2002 20:49'! initialize super initialize. self top: 50; clipSubmorphs: false. child _ RectangleMorph new top:100. self addMorph: child. ! ! "-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- "! Collapser class instanceVariableNames: ''! !Collapser class methodsFor: 'as yet unclassified' stamp: 'dtz 4/30/2002 20:45'! open self new initialize openInWorld.! !

2 1

Serialization formats (was Alan Kay in the news)
by Todd Blanchard 30 Apr '02

30 Apr '02

As a former WebObjects programmer and NextStep fan, I like the non-xml plist format. Its dead simple but very powerful. dictionary is { key1 = value1; key2 = value2; key3 = "Some text on\ntwo lines."; } list/array is ( one, two, "Complex string" ) or its a string. Any nesting of the above is possible - you can create oids if you like to do cross linking but the basic plist format is very powerful and you can do a parser in a few lines of code. I use it constantly from java :-( these days. -----Original Message----- From: Cees de Groot [mailto:cg@cdegroot.com] Sent: Tuesday, April 30, 2002 11:28 AM To: squeak-dev(a)lists.squeakfoundation.org Subject: Re: Alan Kay in the News Mark S. Miller <markm(a)caplet.com> said: >So don't be shy about proposing Squeak-based or Parcels-based serialization >formats. > Braindump: - SRP (State Replication Protocol) is a Smalltalk-platform-neutral protocol with some nice properties, and is expressly meant to be usable on platforms besides Smalltalk. Especially the data encoding is cool, but apart from that it also has good support for class evolution, etcetera. http://wiki.cs.uiuc.edu/CampSmalltalk/About+State+Replication+Protocol+( SRP) - Python has a neat serialization format that is based on some sort of mini-language which has textual and binary representations, support for custom formats and things like oid-replacements. Jim Fulton (Zope and formerly Smalltalk) thinks it is usable from other languages as well. The pickle.py module is less than 1000 lines of code. - I wouldn't suggest XML - all the code doing the XML parsing is probably too large for anything security-related. - RMI would be doable in Smalltalk, I think; it's well specified and the people behind RMI know what they were talking about so IMHO it's an OK wire protocol. The multiplexing makes it quite efficient from a resource-usage point of view. As you remark, Java serialization is a tad complex, so this might be a big job. http://java.sun.com/products/jdk/1.1/docs/guide/rmi/spec/rmi-protocol.do c.html http://java.sun.com/products/jdk/1.2/docs/guide/serialization/spec/seria lTOC.doc.html - The Parcel format is VisualWorks specific, although it probably could be ported (the problem is that the format has no formal specification AFAIK, so a 'port' would almost certainly mean having to read code from VisualWorks, which brings you in muddy legal waters). Even though I haven't followed the conversation on the E list on the necessity of a text format, I dissent. I think that if you want a human to talk to a running E instance through telnet you would better open a separate connection with a minimal command language (maybe one that would allow you to specify an object, step by step, and then you enter the "send" command). >7) The format must be compatible with our security requirements. In >particular, since E is only prepared to trust mobile E code, any need for >executable code in the serialization format must be E code. Parcels as is >fails this test of course. > In the context of heterogenous E implementations, the point of mobile code becomes tricky, of course :-) -- Cees de Groot http://www.cdegroot.com <cg(a)cdegroot.com> GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD 1986 F303 937F E098 9E8B

1 0

Re: Website look
by goran.hultgren＠bluefish.se 30 Apr '02

30 Apr '02

"Lex Spoon" <lex(a)cc.gatech.edu> wrote: > goran.hultgren(a)bluefish.se wrote: > > > Btw - perhaps we should change the wording "Squeak comes under an open > > source license" to perhaps "Squeak has a very liberal license" or > > something like that. I think the last word on OpenSource certification > > was that SqueakL is NOT OpenSource. > > It's not Debian compliant. But Debian has stricter requirements than > merely being open source. I don't know of any reason why Squeak > wouldn't be considered "open source", especially using lower case > letters. This time I wasn't referring to the problem with Debian but to the discussions about SqueakL that has been done on OSI mailinglist for certification. It was said quite clearly (IIRC) that SqueakL is NOT OpenSource. Just using small letters... Nah. That feels a tad like "pretending". I would write something else. regards, Göran

10 15

Re: Alan Kay in the News
by Mark S. Miller 30 Apr '02

30 Apr '02

Cees de Groot wrote: >The only thing that's not neat about E is that it sits on top of Java and has a >shitty development environment, so it's time to rob their ideas and transport >them to Squeak ;-). I just wanted to endorse this thought, and to make it clear that the E community, myself included, would be very eager to see progress in this direction, and can probably help. We're primarily interested in seeing the ideas succeed, rather than E per se. On the erights site, we regularly endorse "competitors" that seem to be getting things right. So please "rob" away! After seeing Lex Spoon's ( http://minnow.cc.gatech.edu/squeak/2074 & http://www.cc.gatech.edu/~lex/7431/term.ps ) and Rob Wither's ( http://minnow.cc.gatech.edu/squeak/2410 ) efforts in this regard, I extended the E history page ( http://www.erights.org/history/overview.html ) with my hope for these efforts. If it doesn't offend anyone, I've given this pair of efforts the unauthorized name "Squeak-E". On the e-lang list ( http://www.eros-os.org/mailman/listinfo/e-lang ) Rob Withers has been discussing with us where to go from here. Those of you interested in joining the discussion are most welcome. ---------------------------------------- Text by me above is hereby placed in the public domain Cheers, --MarkM

3 4

Re: PLEASE use MIME Atttachments! (was: Re: [FIX] Comments for Set...)
by Richard A. O'Keefe 30 Apr '02

30 Apr '02

Good point at the end, skip the rest if you aren't interested in the argument. goran.hultgren(a)bluefish.se wrote: > Looking at the code, I see that "mail to list" sends the message > to "squeak(a)cs.uiuc.edu". I note that the originating address for Eh. Where did you look? I just looked in my 3.2g image and I ended up in: ChangeSet>>buildMessageForMailOutWithUser: ...and as far back as 9/27/2001 (#4385) it seems to be correct. Mind though I might be looking in the wrong place. The address is in my image: squeak-dev(a)lists.squeakfoundation.org What update-level is your image? Squeak 3.0.1 of 2001.02.04, update 3552. Sure could! In that case I guess that Squeak fired it away alright but you didn't get to see the bounce(?) It was simpler and far more embarassing than that. See bottom. > Ear sick horse. Mail sent from Netscape worked fine. What does "Ear sick horse" mean? I am not native english speaking so I didn't get that. "Ear sick horse" = "Yes of course". > - they are not necessary, > - they are not supported by all mail user agents, > - they are notorious for insecurity. Just curious, how is uuencoding better than attachments then? It's not supported in all email clients AFAIK. The key point is that it doesn't _have_ to be. So you want us to change this? Obviously not. Why should I want to stop anybody doing anything that works for them? What I was hoping for was a more generous attitude. For example, for someone who understands Celeste, it should be a simple task to write a filter that decodes a uuencoded message and fires it out again as a Celeste attachment. Less than half the energy that has been spent in stepping on me could have given us the best of both worlds. > I am of course perfectly willing to use whatever method *works*. Well, you could have fooled me. But you say you read the bit where I said I had made repeated attempts to try the official way, with no success. How could such statements possibly fool you into thinking I was not willing to use a working method? OK, HERE's THE GOOD POINT. There are two ways of using mail services. I suppose you might call them the "institutional user" and "home user" ways. I've been using E-mail systems since 1979, always as an "institutional user". In this setting, which is the original RFC 821 (SMPT) setting, when you tell your mail user agent "send this message", it may send it immediately, or it may put it in a "spooling" area and send it later, but not _much_ later, typically at most an hour or two. As far as you are concerned, the message is _sent_ immediately. Maybe it's resting at a local "post office" (the spooling area), but it's already on it's way right now. In the "home user" setting, your mail is held on some other machine. You have to explicitly "pull down" your mail (like a trip to the post office to check a post box) and when you send mail, you are just adding it to a "mail bag" which will NOT be delivered to the "post office" (this time your ISP) until you send it. So "send a message" in this environment seems to mean "I have finished this letter, put it in the mail bag" and is a very different operation from "send this mail BAG to the post office". >From my point of view, what happened is that Celeste lied to me. Remember, I have Celeste on a Sparc and a Mac, but neither of them is my mail machine. On the rare occasions when I send mail from either of them using Netscape, it either didn't work (Sparc) or sent the message _immediately_ (Mac). So when I saw a great big button labelled "send message", I not unreasonably expected it to SEND THE MESSAGE. This was the lie. That button does NOT mean "send the message", it means "add the message to the .tosend category". The message will _never_ be sent if you don't then go to another window (which, don't forget, is _not_ the change sorter you started it, _nor_ the message editing window you've just finished with, but a Celeste control panel you may never have heard of before) and select ".tosend" (another thing you have never heard of, is not explained anywhere, and which you have no reason to suspect may be of interest) and then do the right thing, nothing will happen. If you _do_ discover that '.tosend' exists, open it up, select the message you just asked to be sent, and check the menu, you will find nothing there about sending the message. You will conclude, "Oh, this works like some other mailers, it builds up a queue and then sends it later AUTOMATICALLY." So you wait, and wait, and wait, and nothing happens. The way I finally figured this out is that someone got me to check whether mail sent to the STMP server using Telnet worked, and it did. So I wondered "has Celeste got the right server address", poked around a bit, tried 'Celeste smptServer' and found that it had never been initialised. So it was clear that Celeste had never tried to send the mail. (After my earlier unsuccessful attempts, I had deleted Celeste's files, assuming there was nothing interesting them. A mistake on my part.) I then tried _everything_ in the control panel, and eventually discovered 'send queued mail' in the menu on '.tosend' (NOT on the things that '.tosend' points to). Apparently Celeste users are happy with the way it works. I have not the slightest wish that they should be made to do anything different. However, in my environment, where the STMP server is on the same network as the machine running Squeak, there is absolutely no advantage in queueing the mail. (Except of course that you have one more chance to avoid sending a bad-mannered message.) There is, in contrast, a big advantage in sending mail right away. That is that Macintoshes have been known to crash. (As noted before, while I was trying to find out what was going on, Squeak (+ perhaps Acrobat Reader) crashed my Mac twice.) The sooner a message is sent, the less chance a crash will get in the way. It must be admitted that if you let the mouse hover over 'Send message' for long enough, eventually you will get a help balloon that says in small letters "add this to the queue of messages to be sent". But when you see "Send message", you don't suspect that you don't know what it means, so you just click away and DON'T wait for a help balloon. I propose the following change to Celeste. For all I know, it may already have been made in Squeak 3.2, but if not, I hope someone will do it. A "Mister Postman" window has two big buttons at the top, "send message" and "add attachment". Change this to three: "send message later", "send message now", and "add attachment". "send message now" should open up an SMPT socket and send the message right away.

6 6

Translating Squeak
by Diego Gomez Deck 30 Apr '02

30 Apr '02

Hello list! Translating the eToys vocabulary to spanish, I found (like many people do) that *too* many thinks get not tranlation with the current translation-support. We need a more complete solution to this problem. I propose to atack it in 2 fronts: 1) String 2) Date, Time and Number For 1) I created some methods in String (and a class variable named Languages and a few class methods to initialize and fill it) String>>translated ^ self translatedTo: World currentNaturalLanguage String>>translatedTo: languageSymbol | language | language := Languages at: languageSymbol ifAbsent:[Transcipt show:'*Language: ', languageSymbol, ' not present!'; cr. ^ self]. ^ language at: self ifAbsent:[Transcipt show:'*Sentence:', self, ' not present!'; cr. self] I took the Browser class and added #translated message to all hard-code strings. Then, watching in the Transcript for warnings about missing sentences, I translated the complete browser in just 1/2 hour. Example: Browser>>defaultBrowserTitle ^ 'System Browser' translated This mechanism could be not the best, but we need regionalized Squeaks to teach kids (or kids to teach us :)... I'd like to hear your comments... Cheers, Diego Gomez Deck

2 1

Are the translations used? (Re: [ENH] In Dutch)
by goran.hultgren＠bluefish.se 30 Apr '02

30 Apr '02

cg(a)cdegroot.com wrote: > > --==CelesteAttachment57917== > > Here's a first cut at Dutch EToy translation. Please give me feedback on > word usage. I wonder if anyone uses these translations... Hey, you EToy-guys! Have these translations been of any use? (I did one for Swedish) regards, Göran

3 3

Re: [ENH] DateAndTime
by Richard A. O'Keefe 30 Apr '02

30 Apr '02

"Lex Spoon" <lex(a)cc.gatech.edu> wrote: ANSI Squeak is here. Nobody seems to know about it. That's for a very simple reason. It isn't "here", it's "there". ("There" being http://minnow.cc.gatech.edu/squeak/2384 .) When those change sets came out, I thought it was a wonderful idea. I confidently expected that before long they would become part of the Squeak distribution. They were written for 2.9a; surely it is time they were adopted?

2 1

Squeak.org
by Dan Ingalls 30 Apr '02

30 Apr '02

Folks - My apologies for letting this drag on. I thought some reasonable support might be forthcoming from the ISP, but their attitude and responsiveness have been somewhere between reptilian and criminal. We're now on a fast track to move the domain to a different site. Thanks for your patience. - Dan

9 10

← Newer
1
2
3
4
5
6
7
...
67
Older →

Jump to page: