Squeak-dev February 2002

squeak-dev@lists.squeakfoundation.org

202 participants
665 discussions

Re: Persistence & DTSTTCPW: ZODB clone?
by goran.hultgren＠bluefish.se 01 Feb '02

01 Feb '02

Have you guys ever worked with GemStone (or any similar OODB)? Anyway - Stephen Pair has been doing very interesting work in this direction and demonstrated parts of it at OOPSLA. One of these things is his "CATS" - a transaction service much like the one used in for example GemStone. Looked very neat and solid to me. I know he has been thinking about all the other parts too and a small bird has whispered in my ear that there might be stuff coming... As another sidenote, I have been playing with ImageSegments a bit and <plug> if you sign up for an account at SqueakDot </plug> you can actually see it in action right there - you can checkpoint/revert/analyze your account using ImageSegments and you will see some performance numbers right away. regards, Göran PS. <CSOTD> "Newbie tip: Bags are perfect for histograms and polls. It is what the SDPoll class uses at SqueakDot. Do a 'print it' of below to see the number of votes for 'Yes'." | votes | votes _ Bag new. votes add: 'Yes'; add: 'Yes'; add: 'No'. votes occurrencesOf: 'Yes' </CSOTD> DS

2 1

Re: AW: [VM] unstability
by Bob Arning 01 Feb '02

01 Feb '02

On Thu, 31 Jan 2002 18:22:34 -0800 Scott Wallace <scott.wallace(a)squeakland.org> wrote: >As you'll see, you are correct that it relates to the referencing >mechanism in scripting. The point of departure -- the method from >which, once entered in these "hang" conditions, one never returns -- >is Object.uniqueNameForReference. Scott, One problem is that Object>>uniqueNameForReference has a bug: aName _ Utilities keyLike: stem satisfying: [:jinaLake | nameSym _ jinaLake asSymbol. ((References definedNames includesKey: nameSym) not and: [(Module root moduleDefining: nameSym) notNil]) and: [(knownClassVars includes: nameSym) not]]. will return a name that *is* defined by a module. [(Module root moduleDefining: nameSym) notNil]) and: should be [(Module root moduleDefining: nameSym) isNil]) and: If you want to see this in action, before you make the change, try Morph new uniqueNameForReference Do this once and you get #Morph. It will fail the second time because the name #Morph now refers to the instance of Morph created in the first test. Needless to say, don't do this on an image you care about. Cheers, Bob

2 1

[BUG] class var refs broken
by Scott Wallace 01 Feb '02

01 Feb '02

The "class var refs" function in browsers, message lists, change sorters, etc., is broken in current 3.3a, with the error occurring in modules code. -- Scott

1 0

SqueakDot was unavailable a while...
by goran.hultgren＠bluefish.se 01 Feb '02

01 Feb '02

I had the excellent "CPU hogwatcher" (thanks Ned, saved me several times) on and it had kicked in for some reason... Now I have turned it off so... :-) We will see if it pops up again. regards, Göran

1 0

[VM] unstability
by Stephane Ducasse 01 Feb '02

01 Feb '02

Hi I'm using Squeak 3.2.1Beta6MT under mac Os 9.2 with the module image. And Squeak crashes every 10 minutes in average. Am I the only one? Stef

5 5

Re: Persistence & DTSTTCPW: ZODB clone?
by Avi Bryant 01 Feb '02

01 Feb '02

On Thu, 31 Jan 2002, Cees de Groot wrote: > The structure (now I'm going to tread dangerous territory here - this is from > the back of my head, I really should refresh my knowledge from the docs) is > basically: magic, transaction header, object, object, object, ..., transaction > header, object, object, object, ... etcetera. Objects have backpointers to > parent versions, transactions too - this mean you can 'timetravel' to older > transactions or older object versions (guess how trivial it is to build a Wiki > on top of that...). A few questions - Does anything without an oid get serialized in full to the transaction log, as part of its parent object? How does it recover the oids from the objects when writing to disk - does it keep a reverse map object->oid? Or does every business object have to have an oid field? Which brings me to - what can be used in business objects? In Squeak, will standard collections be usable, or will you have to use an ODBDictionary, ODBOrderedCollection, etc, that have oids, and can associate themselves with transactions? And how do most objects (that don't have an easy at:put: protocol) know they've been modified? Compiler hacks like kats uses? VM hacks? Or do your business objects have to send "self changed" all the time? I ask all of this because I've been thinking about similar designs myself without coming up with satisfactory answers. What I've been looking at instead, recently, is generalizing my hacked together little O/R mapping framework (sort of a GLORP-lite) so that it will work with simple in-memory tables as well as RDBs. Once you map an object structure down to simple relational tables (or arrays of dictionaries of strings, which is what a relational table looks like to my framework), transactions become very easy to serialize. Of course, the disadvantage is that you're constrained by the O/R model and have to have all of this metadata about relationships, etc. One upside is that you can freely mix your "object" (in-memory) DB and any relational DBs you need to use, in terms of having relationships between them, since they all use the same model. The other is that all of the DB-related stuff is totally external to your business objects, rather than forcing them to inherit from anything in particular (although you still have the "self changed" problem to some degree). I don't particularly like this approach, so if you manage to get a ZODB lookalike working, I'll be very pleased. I'm also happy to help out if you need it. Cheers, Avi

2 1

[Q][VM][Windows] USB graphic tablet
by Hannes Hirzel 01 Feb '02

01 Feb '02

Hi I'd like to buy a graphic tablet which can be plugged into the USB port on a Windows machine. Is this OK with the current Windows VM? Or are there any restrictions I have to pay attention to? Thanks for the answer in advance. Hannes Hirzel

1 0

Re: [ANN] sqdb now online
by Duane Maxwell 01 Feb '02

01 Feb '02

Actually, I created that entity in order to preserve whitespace - I did &cr;, &nl;, and &tab; - they-re listed on at the top of the documentation page. Some XML parsers strip out whitespace. What I should do is create a DTD that defines the entity and host it on the site so that validating parsers and IE won't choke on them. An alternative is to produce backslashed escapes for the whitespace - probably an easier but less XML-ish solution. Stephan Wessels was thinking about making the browser changes. -- Duane ----- Original Message ----- From: <Torsten.Bergmann(a)phaidros.com> To: <dmaxwell(a)san.rr.com> Sent: Thursday, January 31, 2002 11:15 PM Subject: RE: [ANN] sqdb now online > Hi Duane, > > If I try to click on the XML link in class comments of the > Squeak documentation database the browser reports (IE 5.0 on Win32): > > The XML page cannot be displayed. Cannot view XML input using XSL > style sheet. Please correct the error and then click the Refresh b > utton, or try again later. > ----------------------- > Reference to undefined entity 'cr'. Line 3, Position 347 > > Looks like the XML generator does not produce valid XML at the end > of a comment. "<class>...&cr;</class>" > > Thanks for the cool work. Is there a tool available to document directly > in Squeak browsers and up/download comments from the server ? > > Keep on talking small > Torsten > >

1 1

SqueakFFIPrims source code for mach-o carbon FFI gets posted to SF
by John M McIntosh 01 Feb '02

01 Feb '02

I posted the source code for the carbon mach-o OS-X SqueakFFIPrims to sourceforge. Early adopters are welcome to try and compile it. Beware of interesting ways to crash your VM... (Mind a bundle will go into a FTP directory soon) -- -- =========================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> 1-800-477-2659 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ===========================================================================

1 0

Re: Tail Call
by Anthony Hannan 01 Feb '02

01 Feb '02

In the block closure version, which uses stack frames for method contexts, I doubt the tailSend: bytecode would be any faster because, although your saving a return, you have to copy the receiver and args down. The regular send: just leaves the receiver and args where they are and starts the new context right from there. The biggest drawback of tailSend:, I think, is losing the calling context when looking in the debugger. It would just add to the confusion when trying to figure out a bug. The only benefit would be to save stack space. But (unless your in an infinite loop) it is rare that the stack uses too much space, usually it other data structures taking up too much space. And if your stack is taking up most of the space, usually you can rewrite your algorithm to use iteration instead of recursion without loosing much clarity. But I guess there are some algorithms where the tail recursive version is nicer. So I would vote for adding the tailSend: bytecode but requiring another language construct that invokes it, so you are aware of it (in the debugger). I would not automatically convert all "send: + returnTop" to "tailSend:" Cheers, Anthony

2 1

Jump to page: