Squeak-dev April 2005

squeak-dev@lists.squeakfoundation.org

158 participants
287 discussions

[Q] Morphs seems do not remember formerPosition
by Lic. Edgar J. De Cleene 27 Apr '05

27 Apr '05

When Squeak changes from 3.x to next 3.x , I check my old projects still works . I found following is failing as miMorph formerPosition returns nil acceptDroppingMorph: miMorph event: evt | centro | "When pieces are dropped into board, here you check if the center of correct place is inside the piece and if true stick it in place" miMorph class = PiezaSelectionMorph ifTrue: [miMorph submorphsDo: [ :m | self acceptDroppingMorph: m event: evt ]. miMorph delete. ] ifFalse: [ centro := centrosXY at: miMorph numero. (miMorph containsPoint: centro) ifTrue: [SampledSound playSoundNamed: 'clink'. miMorph center: centro. miMorph borderColor: Color red. miMorph sticky: true. self addMorphFront: miMorph. self fin] ifFalse: [miMorph position: miMorph formerPosition. miMorph openInWorld]] In Squeak3.8g-6548.image is workning and in newer not, so I wish a clue what update could causing this and if is really a bug what I should report to Mantis Thanks in advance. Edgar PD. The .sar is in http://minnow.cc.gatech.edu/squeak/3938 "Rompecabezas"

1 0

Re: Sun Fortress language spec
by Alexandre Bergel 27 Apr '05

27 Apr '05

Hello, Fortress is a new language from Sun. Thanks Marcus for having pointed out it. A spec is available on http://research.sun.com/projects/plrg/fortress0618.pdf However, here is a short summary of the points for which I am interested in. In Fortress there is no class. Just objects and traits. The language is statically typed, however it performs some inference if some type notation are missing. In Fortress, not everything is an object (e.g., top-level function can be defined). Function or methods can return multiple values, and not just one as in Smalltalk. Contracts (pre and post conditions) are also supported. The language has many other characteristics, but I will focus now more on traits. I was curious to see what is the thing they call "trait". In Fortress, a trait is a named group of methods (abstract or not). A object is defined by using a set of traits and declaring variables. A trait that can inherit from other traits. Yes, multi-inheritance of traits. A trait is stateless. In case of conflict (two methods or more having the same name are inherited), conflicting methods have to be compatible in order to satisfy multiple dispatch. It does not seem that a particular method can be excluded. A trait can exclude a trait. If a trait T exclused a trait S, the two traits are mutually exclusive. No object can have them both, no third trait can extend them both, and neither may extend the other. I know that I am not so accurtate on trait exclusion, but the doc I found is no accurate more. Something else (but I do not know in what it is useful for): if a trait definition T can include a 'bounds' clause, the trait must not be extended (i.e., inherited) with immediate subtraits other than those that appear in its 'bounds' clause. Traits can also be parameterized with some types. It seems that one can write some very complicated trait declarations with that. Ok, so traits does not have field but just methods. Fields are declared when creating an object, and for each field, an accessor and a mutator are automatically defined. Cheers, Alexandre On Wed, Apr 27, 2005 at 10:35:17AM +0200, Marcus Denker wrote: > Hi, > > A first language spec is out > > http://research.sun.com/projects/plrg/fortress0618.pdf > > (this is a research language with traits done at Sun ). > > Marcus -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.iam.unibe.ch/~bergel ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

1 0

Unmatching methods/calls.
by François THIMON 27 Apr '05

27 Apr '05

Hi. Something really annoying in squeak is you often find invokations of methods that don't exist. Just to give some examples, I found : WriteStream on:(String new:128) in XMLTokenizer class but though, no on: method is present amongst the static methods of WriteStream. I also found some examples in socket classes that I needed to correct by myself to get them to work. Why isn't there any method available in XMLParser? They do but call subclassResponsibility! Does someone know whether this will be set back in order one day? (I'm currently using the latest version,ie the 3.7-5989 full). Thanks in advance.

8 12

Hi all / Available and keen on joining a Squeak´s open Project
by Felix Madrid 26 Apr '05

26 Apr '05

Hi all, I am an Argentinian Smalltalker and I would like to participate in a Squeak Project´s where colaboration is needed. I entered http://minnow.cc.gatech.edu/squeak/444 looking for code requests but I didn´t find any created and open project to get involved with. Thanks! Félix

2 1

puzzle
by Stéphane Rollandin 26 Apr '05

26 Apr '05

hello list, here is a little puzzle: in my working image I keep having a very specific trouble with StrikeFonts, where my default font (which one it is does not seem to matter, only the fact that it is the default one) gets at time (rarely) its emphasis changed from 0 to some other value (4, 5...) nothing is visible until I save a morph as a *.morph file then read it back: at this point all text in the morph indeed show the emphasis, being bold or underlined or both. so I run the following snippet: (StrikeFont allInstances select: [:f | f name last isDigit and: [f emphasis isZero not]] ) do: [:f | f emphasis: 0] and everything is fine again... until the next time the emphasis gets changed. so here is my question: where should I have a look for the origin of this trouble ? I'm puzzled... and did someone else get a similar behavior ? thanks, Stef

1 0

OmniBase and Commits
by Daniel Salama 26 Apr '05

26 Apr '05

Hi, I'm trying to test uploading several thousand objects into an OmniBase db and I'm trying to figure out how to commit every 100th object. Can anyone shed some light into this? Every time I try something, it must be surrounded by a transaction, so I can't figure out how to do it. I have stored a LinkedList to a root element. I have something similar to this: txn := db newTransaction. 1 to: 1000 do: [:i| base := txn root at: 'linkedlist'. base add: anObject. (i \\ 100) = 0 ifTrue: [ txn commit. txn := db newTransaction]] But this doesn't seem to work. Please help. Thanks, Daniel

3 2

Monticello Package Browser
by Alexandre Bergel 26 Apr '05

26 Apr '05

Hello, I have been working on a new package browser comming with a clean Package class. I was quite busy the previous weeks, however I would like to push it further. It slowly become more and more usuable. I am currently working on showing information related to class extension. Have a look at http://www.iam.unibe.ch/~scg/Resources/SqueakPackages/packagebrowser.jpg The left most pane shows the list of packages and categories that compose them. The class pane shows classes and class that are extended (*Class). Clicking on a package and an extended class shows (as on the picture) shows the list of methods that belongs to the selected package. You have to read the screenshot as: The class OBMethodNode defined in the package OB-Standard is extended by the package 'Package' with two methods extensions: #isDescendantOfPackage: and #isDescendantOfPackageSubcategory: What I plan to do in the very short term: - Allowing classes to belong in a package without caring about subcategory. - Have some better menu - Adding a star in front of a method if it is a method extension. For mid-range term: - To make the browser fast. Now it is slower to the normal squeak browser. - To add two buttons Package and Hierarchy (a la Visualworks) - Work on a binary support for Monticello The website http://www.iam.unibe.ch/~scg/Resources/SqueakPackages contains a runnable image. Thanks Avi, Colin and Stef for your feedbacks. Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.iam.unibe.ch/~bergel ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

1 0

Harvesting?
by Andreas Raab 26 Apr '05

26 Apr '05

Hi Folks - Stef just wrote a message where he complained that there's no harvesting going on. I kinda agree and for me it triggered the interesting question what nowadays process is. With the guides being fired and all I wonder who is in charge to actually accept contributions? I think the pages on minnow etc. could use some updating in this regard but some general guidelines about what todays processes look like would be nice, too. Just curious, - Andreas

4 5

nearly invisible scrollbars [release usability issue]
by Ross Boylan 26 Apr '05

26 Apr '05

I've just been playing with a 3.8gamma image (got from the Debian apt archive, then pulled in latest updates). The scrollbars are almost invisible in several settings: it's white on white. This is with the default theme (brightsqueak). For example, a system window on a big pane or the package loader have this problem. It might be good to tweak this a bit before final release. Running under XFree86, if that matters.

1 0

Harvesting Party Friday April 29th 21:00 GMT
by Ken Causey 26 Apr '05

26 Apr '05

Just a reminder that the next Harvesting Party will take place between 21:00 (9 PM) GMT Friday April 29th and 01:00 (1 AM) GMT Saturday April 30th, 2005. As usual we will meet in #squeak on irc.freenode.net. More information about the #squeak IRC channel can be found at http://people.squeakfoundation.org/article/7.html Information about the Harvesting Party idea can be found at http://people.squeakfoundation.org/article/29.html If you have the time please attend and work toward the future of Squeak. The next seven parties are scheduled for 01:00 (01:00 AM) GMT Wednesday May 04, 2005 05:00 (05:00 AM) GMT Sunday May 08, 2005 09:00 (09:00 AM) GMT Thursday May 12, 2005 13:00 (01:00 PM) GMT Monday May 16, 2005 17:00 (05:00 PM) GMT Friday May 20, 2005 21:00 (09:00 PM) GMT Tuesday May 24, 2005 01:00 (01:00 AM) GMT Sunday May 29, 2005

1 0

← Newer
1
2
3
4
5
6
7
8
9
...
29
Older →

Jump to page: