Squeak-dev August 1998

squeak-dev@lists.squeakfoundation.org

99 participants
188 discussions

Re: Polymorphism without protocol dilution
by Alan Kay 29 Aug '98

29 Aug '98

Maurice and Travis -- Dan Ingalls and I (seperately) were very taken with parts of APL in our youth, and. over the years, have been quite excited at the possibilities that polymorphism brings to this kind of generalization of collections. It is now time to take another pass at the algebraic roots of Smalltalk/Squeak. Suggestions please -- the best kind are not so much add-ons, but comprehensive schemes that are both powerful, readable, and help users think ... Cheers, Alan ------ At 7:36 AM -0000 8/26/98, Maurice Rabb wrote: >>Scary to have your own quotes, long forgotten, dredged up out of the dust. >>I feel >>like Bill (Gates or Clinton, take your pick). What continues to gnaw at >>brain is >>what you sum up so well with the statement >> > >I regret my inapropriate relationship with my Squeak archives. :-) >Seriously, I didn't mean to use your quote as a kidney punch. I just >though your comments were so apropos. The tough questions are definitely >cyclical. Unfortunately, the lousy ones (ala C++ vs Smalltalk) repeat as >well. > > >>For a while now, we've had a ?framework? in VisualWorks that I've thought >>about >>porting over to Squeak and releasing, but I've always feared the flame about >>diluting protocols. But tonight, I'm coming out. :) > >[munch] > >>blow my friends and teachers away. I would just whip up little expressions: >> >>#((1 2 3 4 5) + #(6 7 8 9 10) * 4) sum >> > >Hmmm, that looks pretty cool and compelling to me. > > >>Maybe the arithmetic protocol is one that dilutes better. > > >This reminds me of the "is Point a Magnitude or not" discussion. > >Your comments have given me the courage to add the arithmetic protocols I >really want to Rectangle. BTW, is Squeak arithmetic now using some sort of >hybrid between double dispatch and coercion? What is the advantage? It >used to use straight coercion, right? Travis are you still using your DD >changeSet? > >--Maurice > >--------------------------------------------------------------------------- > Maurice Rabb 773.281.6003 Stono Technologies, LLC Chicago, USA

1 0

Re: nil or #nil?
by David N. Smith 28 Aug '98

28 Aug '98

At 21:15 -0400 8/25/98, John Brant wrote: > >Some other anomalies with # are: > > ##Point value new ==> a Point > #(##Point) first ==> "#Point -> nil" > #(###Point) first value new ==> a Point > #(####Point) first ==> "nil -> nil" > ####Point ==> compiler error > > ## ==> "(Character value: 30) -> nil" > #(#) ==> compiler error > #(##) ==> compiler error John: This depends on your implementation. In IBM Smalltalk, only the second works and it produces an array holding an Atom. (An Atom is an object with a unique global value, like a symbol, but only has a name in the development environment. It is written like a symbol but with two pound signs. In runtime images it has no name. Atom's are used instead of Symbols for state variables.) The rest all produce a compile error, except for the first which produces a runtime error (Atom does not understand #value). I ran your examples in Squeak 2.1 and got a different result on the third one: #(###Point) first value new (new not understood) The ###Point seems to produce an Association (!) with nil as a key and the symbol #Point as a value. Sending new to the symbol produces the error. Maybe there are changes since 2.1 that I don't have? (An Association?!!?? Where does THAT come from??) The last two produce surprising results in that the error message appears at the front of the workspace and the whole workspace flashes the selection color momentarily. There is some odd bug somewhere. Dave _______________________________ David N. Smith IBM T J Watson Research Center Hawthorne, NY _______________________________ Any opinions or recommendations herein are those of the author and not of his employer.

1 0

RE: Definition of Squeak
by Stefan Matthias Aust 28 Aug '98

28 Aug '98

>A fine idea. Are you volunteering? Well, here you are :) The following rules use EBNF notations. Terminal symbols are enclosed in double quotes. [ ] means optional occurence, { } zero or more occurences and | is an alternative. These are the main rules for Squeak's variant of Smalltalk, as implemented in the class Parser. TODO: Define word, binary, string, number, scannerLiteral. start = method. method = pattern temporaries primitive statements. NOTE: Compiler asures that no other tokens follow. pattern = unaryPattern | binaryPattern | keywordPattern. unaryPattern = unarySelector. unarySelector = word. binaryPattern = binarySelector argument. binarySelector = (binary | "|"). NOTE: | isn't a binary and || isn't allowed at all; this might be an error. keywordPattern = keyword argument {keyword argument}. keyword = word ":". NOTE: Actually, this is already detected by the scanner. argument = word. NOTE: The argument must not already been declared. Write access is not allowed for arguments. temporaries = [ "|" {variable} "|" ]. NOTE: The variable must not already been declared. || (without whitespaces) is allowed because || isn't a binarySelector. variable = word. primitive = [ "<" primitiveDecl ">" ]. primitiveDecl = [ "primitive:" number ]. NOTE: you can omit the complete declaration. In this case, a primitive nil is returned. You can also use floats for the number. Both seems to be errors in the compiler statements = returnStmt | expressionStmt NOTE: if there's another statement after a return, the compiler notify us that it expects the block end, regardless whether we're in a block or not. returnStmt = "^" expression [ "." ]. expressionStmt = [ expression {"." statements} [ "." ] ]. NOTE: If no expression is available, method bodies are replaced with "self" and empty blocks with "nil". expression = (assignment | braceAssign | primaryExpr) messagePart assignment = variable assigner expression. NOTE: variable must be bound. assigner = "_" | ":=". NOTE: "_" is actually the left arrow. braceAssign = brace assigner expression. NOTE: I don't understand this. primaryExpr = varExpr | blockExpr | braceExpr | "(" expression ")" | literal. varExpr = variable NOTE: Variable must be bound or one of the following special variables: false, nil, self, super, thisContext, true and homeContext. blockExpr = "[" [ ":" argument {":" argument} "|" ] statements "]" NOTE: The arguments may shadow other declarations. Write access is not allowed for arguments. braceExpr = "{" [ expression {"." expression} ] "}". literal = string | number | ("-" number) | scannerLiteral. NOTE: ScanneLiterals include symbols, characters, literal arrays, etc. messagePart = [ messagePart3 {messagePart3} cascade ]. messagePart3 = keywordMsgPart | messagePart2. NOTE: This somewhat complex scheme is used to express the precedence rules of different message types. keywordMsgPart = keywordPart {keywordPart}. NOTE: All keywordParts are concatenated to one message send. keywordPart = keyword primaryExpr messagePart2 {messagePart2}. messagePart2 = binaryMsgPart | messagePart1. binaryMsgPart = binarySelector messagePart1 {messagePart1}. messagePart1 = unaryMessagePart. unaryMsgPart = unarySelector. cascade = {";" messagePart3}. NOTE: Compiler checks whether cascade is allowed. You cannot cascade super sends and sends to special selectors like ifTrue: which are replaced by the compiler. method = pattern temporaries primitive statements. NOTE: Compiler asures that no other tokens follow. pattern = unaryPattern | binaryPattern | keywordPattern. unaryPattern = unarySelector. unarySelector = word. binaryPattern = binarySelector argument. binarySelector = (binary | "|"). NOTE: | isn't a binary and || isn't allowed at all; this might be an error. keywordPattern = keyword argument {keyword argument}. keyword = word ":". NOTE: Actually, this is already detected by the scanner. argument = word. NOTE: The argument must not already been declared. Write access is not allowed for arguments. temporaries = [ "|" {variable} "|" ]. NOTE: The variable must not already been declared. || (without whitespaces) is allowed because || isn't a binarySelector. variable = word. primitive = [ "<" primitiveDecl ">" ]. primitiveDecl = [ "primitive:" number ]. NOTE: you can omit the complete declaration. In this case, a primitive nil is returned. You can also use floats for the number. Both seems to be errors in the compiler statements = returnStmt | expressionStmt NOTE: if there's another statement after a return, the compiler notify us that it expects the block end, regardless whether we're in a block or not. returnStmt = "^" expression [ "." ]. expressionStmt = [ expression {"." statements} [ "." ] ]. NOTE: If no expression is available, method bodies are replaced with "self" and empty blocks with "nil". expression = (assignment | braceAssign | primaryExpr) messagePart assignment = variable assigner expression. NOTE: variable must be bound. assigner = "_" | ":=". NOTE: "_" is actually the left arrow. braceAssign = brace assigner expression. NOTE: I don't understand this. primaryExpr = varExpr | blockExpr | braceExpr | "(" expression ")" | literal. varExpr = variable NOTE: Variable must be bound or one of the following special variables: false, nil, self, super, thisContext, true and homeContext. blockExpr = "[" [ ":" argument {":" argument} "|" ] statements "]" NOTE: The arguments may shadow other declarations. Write access is not allowed for arguments. braceExpr = "{" [ expression {"." expression} ] "}". literal = string | number | ("-" number) | scannerLiteral. NOTE: ScanneLiterals include symbols, characters, literal arrays, etc. messagePart = [ messagePart3 {messagePart3} cascade ]. messagePart3 = keywordMsgPart | messagePart2. NOTE: This somewhat complex scheme is used to express the precedence rules of different message types. keywordMsgPart = keywordPart {keywordPart}. NOTE: All keywordParts are concatenated to one message send. keywordPart = keyword primaryExpr messagePart2 {messagePart2}. messagePart2 = binaryMsgPart | messagePart1. binaryMsgPart = binarySelector messagePart1 {messagePart1}. messagePart1 = unaryMessagePart. unaryMsgPart = unarySelector. cascade = {";" messagePart3}. NOTE: Compiler checks whether cascade is allowed. You cannot cascade super sends and sends to special selectors like ifTrue: which are replaced by the compiler. bye -- Stefan Matthias Aust // Are you ready to discover the twilight zone?

2 1

Re: Dumb question on Morphic
by Tim Olson 27 Aug '98

27 Aug '98

>Apparently the process dies somewhere inside openWithTitle:; >maybe someone knows off the top of his head, where, what's >the matter, and what's the right way of doing in this situation? If you follow the chain of messages down from openWithTitle:, you find yourself at ControlManager>>scheduleActive: which says: scheduleActive: aController "Make aController be scheduled as the active controller. Presumably the active scheduling process asked to schedule this controller and that a new process associated this controller takes control. So this is the last act of the active scheduling process." <primitive: 19> "Simulation guard" self scheduleActiveNoTerminate: aController. Processor terminateActive So the last act of openWithTitle: is to terminate the current process. This is the standard problem with opening a new controller -- it must be the last thing that you do. If you want to continue to execute code after the open, you will have to use the ControlManager>>scheduleActiveNoTerminate: message, but that would involve duplicating a lot of the messages in the path of control from openWithTitle: -- tim

1 0

RE: Re: nil or #nil?
by Tim Olson 27 Aug '98

27 Aug '98

Bob Jarvis wrote: >How about > > Symbol newFrom: 1 printString > >or maybe > > Symbol internCharacter: $1 Or: 1 printString asSymbol [I had written in a previous message that you could use asSymbol to create symbols out of things which weren't identifiers, but I had forgotten that asSymbol only worked on Strings or Characters -- Bob's post reminded me that you had to perform "asSymbol" on the printString]. -- tim

1 0

Re: Re: nil or #nil?
by Tim Olson 27 Aug '98

27 Aug '98

> >In a message dated 8/25/98 11:35:36 PM, you wrote: > ><<I didn't know that "#1 == 1". > >I suspect that most folks fall into that camp. Currently in Squeak, # >means "literal quotation", and any literal which is quoted simply returns >that literal>> > >Whoa! Where have I been? > >I was not aware of this difference and I TRY to pay close attention. If you >can't use # to create a new Symbol, what do you use? Only identifiers and keywords (i.e. <letter>{<letter>|<digit>| $:}) can be quoted with # to create a Symbol. If you want to create a symbol from something that isn't an identifier, such as 1, you have to use "asSymbol". The key point here is that # means "Literal Quotation", not "as Literal Symbol". -- tim

1 0

Dumb question on Morphic
by Boris G. Chr. Shingarov 27 Aug '98

27 Aug '98

Fellow Squeakers, As I just started thinking on Cheese-over-Morphic (to return to a BiCheese image for those of you who follow this thread), I began to play with Morphic and got this question: I'm trying to programmatically open a world in the middle of some other actions, e.g. | myWorld | "action 1 - create and open a WorldMorph" myWorld := WorldMorph new initialize. myWorld openWithTitle: 'Hello there!'. "action 2 - do somthing else e.g. print a line" Transcript show: 'Umm?'. Apparently the process dies somewhere inside openWithTitle:; maybe someone knows off the top of his head, where, what's the matter, and what's the right way of doing in this situation? Boris

1 0

Re: Arithmetic (was: Polymorphism without protocol dilution)
by m3rabb＠stono.com 27 Aug '98

27 Aug '98

At 11:58 AM 8/26/98, Travis Griggs wrote: >I will give that double dispatching, at least for me, was a piece of the >Smalltalk enlightenment puzzle. Me too. >An #adaptToType: >type message is basically a polymorphic application of a type cast, >something that >we're all familiar with from more arcane systems. OTOH, the only people >that have >really delved into the core number stuff so far, have been people >comfortable with >100% messaging. I looked at the Squeak adaptTo stuff again, but I still find it more confusing than DD. Maybe my brain has become too adapted toward DD. > Ultimately, the whole thing would best be solved by adding >multi-methods to Squeak, but I'm not that good of a VM hacker. :) > As in multidispatch, ala CLOS? >> I was planning to use DD for the Rectangle arithmetic as well. When I am >> done how about we unify this with STP's bug fixes? > >Sounds great. I'll go ahead and move the numeric collection stuff in as >well. I also >intend to have a go at adding strings to numbers and vice versa (aka Perl). Great, I'll send you a private message when I am ready. --Maurice --------------------------------------------------------------------------- Maurice Rabb 773.281.6003 Stono Technologies, LLC Chicago, USA

1 0

Re: RE: nil or #nil?
by Mark4Flies＠aol.com 27 Aug '98

27 Aug '98

In a message dated 8/25/98 2:53:59 PM, you wrote: <<If I were to write something like #(nil true false) I'd expect an array equivalent to Array with: nil with: true with: false not Array with: #nil with: #true with: #false>> Me, too. It seems as though the special values should have precedence since they always exist and I should just use them (favor "Array with: nil with: true with: false" interpretation) but if I need literals (that happen to be the same string of characters) I have to create them, possibly going out of my way (accept "Array with: #nil with: #true with: #false").

1 0

Re: Unsubscribing
by johnson＠cs.uiuc.edu 27 Aug '98

27 Aug '98

I think the way to unsubscribe is to send to squeak-request(a)cs.uiuc.edu with "unsubscribe" as the SUBJECT. -Ralph

1 0

Jump to page: