Moose-dev

moose-dev@list.inf.unibe.ch

6621 discussions

Start a nNew thread

moose.unibe.ch-bug
by Toon Verwaest 23 Nov '07

23 Nov '07

The searchbox doesn't appear to work in firefox cheers

1 0

Changed location of MSE4Java
by Adrian Kuhn 23 Nov '07

23 Nov '07

https://www.iam.unibe.ch/scg/svn_repos/Sources/Famix/MSE4Java/ https://www.iam.unibe.ch/scg/svn_repos/Sources/Famix/MSE4Java/famix.jar Please update your links. cheers, AA

1 0

A bug to smash in moose
by Sandro De Zanet 23 Nov '07

23 Nov '07

Hello everyone I generated a mse file (see attachment) which should generate 14 methods in the class ' Superclass', but here are 17 methods shown in the moose browser (some methods are duplicated). The file seems ok, so I think there is a bug in moose, reading the file. Cheers Sandro (Moose.Model (name 'frittelitest') (entity (FAMIX.Method (id: 1) (name 'foo') (belongsTo (idref: 14)) (signature 'foo')) (FAMIX.Method (id: 2) (name 'put') (belongsTo (idref: 14)) (signature 'put')) (FAMIX.Method (id: 3) (name 'size') (belongsTo (idref: 14)) (signature 'size')) (FAMIX.Method (id: 4) (name 'main') (belongsTo (idref: 15)) (signature 'main')) (FAMIX.Method (id: 5) (name 'isEmpty') (belongsTo (idref: 14)) (signature 'isEmpty')) (FAMIX.Method (id: 6) (name 'foo') (belongsTo (idref: 14)) (signature 'foo')) (FAMIX.Method (id: 7) (name 'containsValue') (belongsTo (idref: 14)) (signature 'containsValue')) (FAMIX.Method (id: 8) (name 'entrySet') (belongsTo (idref: 14)) (signature 'entrySet')) (FAMIX.Method (id: 9) (name 'putAll') (belongsTo (idref: 14)) (signature 'putAll')) (FAMIX.Method (id: 10) (name 'remove') (belongsTo (idref: 14)) (signature 'remove')) (FAMIX.Method (id: 11) (name 'containsKey') (belongsTo (idref: 14)) (signature 'containsKey')) (FAMIX.Method (id: 12) (name 'get') (belongsTo (idref: 14)) (signature 'get')) (FAMIX.Namespace (id: 13) (name 'frittelitest') (class (idref: 14) (idref: 16) (idref: 15))) (FAMIX.Class (id: 14) (name 'Superclass') (NOA 0) (NOM 14) (method (idref: 1) (idref: 6) (idref: 19) (idref: 11) (idref: 7) (idref: 8) (idref: 12) (idref: 5) (idref: 17) (idref: 2) (idref: 9) (idref: 10) (idref: 3) (idref: 20))) (FAMIX.Class (id: 16) (name 'Subclass') (NOA 0) (NOM 1) (method (idref: 18))) (FAMIX.Class (id: 15) (name 'Main') (NOA 0) (NOM 1) (method (idref: 4))) (FAMIX.Method (id: 17) (name 'keySet') (belongsTo (idref: 14)) (signature 'keySet')) (FAMIX.Method (id: 18) (name 'foo') (belongsTo (idref: 16)) (signature 'foo')) (FAMIX.Method (id: 19) (name 'clear') (belongsTo (idref: 14)) (signature 'clear')) (FAMIX.Method (id: 20) (name 'values') (belongsTo (idref: 14)) (signature 'values'))))

3 4

collective behavior
by Tudor Girba 22 Nov '07

22 Nov '07

Hmm, I was about to write that the goal of collective behavior is to model the behavior of the groups (like you demoed at ESUG), and I wanted to ask how do you store this behavior in Store. Now I see that I misunderstood completely. From the arguments below, I believe it is not really worth the trouble. Doru >> B1) the reasons for collective behavior >> > > To save memory. > > Each moose element has a back pointer to its model. This is 4 byte > per element, ie 4 MB per 1 Mio elements. Which are realistic maybe > even low numbers when analysing histories or traces. With collective > behavior the memory footprint is (4 + F) * M * T, where F is the size > of an anonymous class, M is the number of models and T is the number > of element types used per model. Given that F is about 80 byte > (including object headers and metaclasses etc pp) and T maybe about > 25, that is 2 KB = 0.002 MB per Model. > > >> B2) the solutions for collective behavior >> >> I propose to talk about each of these two points individually, >> because they are not the same to me. > -- www.tudorgirba.com www.tudorgirba.com/blog "Some battles are better lost than fought."

1 0

[ANN] moose.unibe.ch
by Tudor Girba 22 Nov '07

22 Nov '07

Hi, A new webpage based on Pier is up and running at: http:// moose.unibe.ch . Feedback is welcome :). Here is an overview: -- the front page is supposed to offer all sorts of entry points into the contents. -- publications are read from a bib file that contains the entries from scg.bib that are marked as moose-pub. For each publication a page is created whose title is the title of the paper, the url is the key of the entry and the tags of the page are the keywords of the entry. This is useful for the publications cloud that you can see on the front page. I spent quite some time to add missing moose-pub keywords and to add several other keywords. Now that the keywords are actually used, perhaps we will have an incentive to add some more for our publications :) -- on the front page there is a place for a highlight that is randomly shown. At the moment there are just 2 movies with Mondrian. It would be cool to have more such videos with your nice tools :). To add your highlight, you have to add a page below randomhighlight similar to: http://moose.unibe.ch/randomhighlight/videomondrianbasics -- on the tools page, each tool has a page allocated. You can add tags in each tool page, and these keywords are taken into account in the tools cloud. For example, you can specify whether the tool is in- visualworks or in-java. -- there is a news feed. At the moment it is broken due to some Apache settings or something :), but it should be available at: http://moose.unibe.ch/news?view=PBEntriesRssView -- the site contains the content from the Smallwiki page. I also took the liberty to copy some content from personal sites (like codecity or softwarenaut). I hope that is Ok. -- you can follow the changes via RSS: http://moose.unibe.ch/?view=PRChangesRssView -- the webpage is persisted in the image, and the image is scheduled to be saved under a different name 1 hour after a change. You need a user to change the page, please ask me for one. There are two types of users: admin and editor. The admin can do everything, while the editor can just edit :). You will probably get an editor user. Further details about how to edit the site can be found at: http://moose.unibe.ch/varia/howtoedit This is just version 1. There are several things that are still in plan: - make publications updated periodically from external urls. Currently it is updated manually. - make the tool page a special component that has explicit links to publications and developers. The tool should also have an explicit license field. - port the Store queries from VW to Squeak to create activity reports. - add last update information on each page. - work on the css to make it use a better alignment Cheers, Doru p.s. I would like to thank Lukas Renggli and Adrian Lienhard for their help and patience :). -- www.tudorgirba.com www.tudorgirba.com/blog "We cannot reach the flow of things unless we let go."

6 12

Fwd: moose.unibe.ch
by Tudor Girba 22 Nov '07

22 Nov '07

> From: Tudor Girba <girba(a)iam.unibe.ch> > Date: November 22, 2007 11:51:01 AM GMT+01:00 > To: Toon Verwaest <toon.verwaest(a)gmail.com> > Cc: Moose-dev Moose Dev <moose-dev(a)iam.unibe.ch> > Subject: Re: moose.unibe.ch > > Indeed, this is the main rule of web design. > > But, in this case, the problem is in Pier. I will fix it soon. > > Doru > > > > > On Nov 22, 2007, at 11:21 AM, Toon Verwaest wrote: > >> This is how your moosepage looks on firefox. Since 44.68% of your >> visitors come from firefox (+3% mozilla). I would strongly advice >> you to have a look at designing the site so its compatible with >> non-safari browsers. On internet explorer it already looks fairly >> ok but not great either. And actually, safari represents only 15% >> of the users, so only 15% of the users see the page correctly. >> Isn't this one of the major web design rules? >> >> cheers >> Toon > > > > -- > www.tudorgirba.com > www.tudorgirba.com/blog > > "Obvious things are difficult to teach." > > -- www.tudorgirba.com www.tudorgirba.com/blog "One cannot do more than one can do."

1 1

Announcement, #moose-dev channel on freenode
by Adrian Kuhn 21 Nov '07

21 Nov '07

Toon and me created a moose-dev channel on freenode to hang around, please feel free to join us irc://chat.freenode.net:8000/moose-dev Is this a correct IRC link? ... we'll see :) cheers, AA

2 2

Problem with Hapax when loading Moose Config
by Matthias Junker 21 Nov '07

21 Nov '07

When i try to load moose config 1.23 in a fresh image i get the following error: Warning: Package "HapaxNamespace" cannot be loaded. The following classes cannot be loaded because their environment is not in the image nor in the package being loaded: LICENSE (environment path: Root.Smalltalk.Hapax) When i hit ok anyway, i get a lot of unloadable class definitions Warning: Package "Clustering" cannot be loaded. The following classes cannot be loaded because their environment is not in the image nor in the package being loaded: CorrelationSquare (environment path: Root.Smalltalk.Hapax) ClusterTest (environment path: Root.Smalltalk.Hapax) CorrelationVector (environment path: Root.Smalltalk.Hapax) DendrogramLeaf (environment path: Root.Smalltalk.Hapax) ClusteringData (environment path: Root.Smalltalk.Hapax) TimeWarpingEngine (environment path: Root.Smalltalk.Hapax) ValueItem (environment path: Root.Smalltalk.Hapax) CollectionItem (environment path: Root.Smalltalk.Hapax) DistanceSquare (environment path: Root.Smalltalk.Hapax) ClusterEngine (environment path: Root.Smalltalk.Hapax) SeriationEngine (environment path: Root.Smalltalk.Hapax) Dendrogram (environment path: Root.Smalltalk.Hapax) DendrogramTest (environment path: Root.Smalltalk.Hapax) ClusteringVector (environment path: Root.Smalltalk.Hapax) DendrogramNode (environment path: Root.Smalltalk.Hapax) VectorItem (environment path: Root.Smalltalk.Hapax) FeatureCollection (environment path: Root.Smalltalk.Hapax) etc... Cheers, Matt

2 2

Famix30beta-14.pdf
by Adrian Kuhn 19 Nov '07

19 Nov '07

Please find attached the latest PDF of FAMIX.Core beta more informations will follow cheers, AA & Hani

4 5

problems with loading moose development
by Richard Wettel 19 Nov '07

19 Nov '07

Hi, I am trying to load my prerequisites (MooseDevelopment) and first I get this: "Warning: Package "Meta-FM3-Model" cannot be loaded. The following classes cannot be loaded because their environment is not in the image nor in the package being loaded: MultivalueLink (environment path: Root.Smalltalk.Meta)" Followed by a bunch of unloadable definitions in Meta. After loading everything I could, I try to initialize the meta descriptions and I get: "Unhandled exception: Message not understood: #on:opposite: UndefinedObject(Object)>>doesNotUnderstand: EMOF.Class>>initialize ..." Could you guys fix it? Thanks, Ricky

2 1

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

Moose-dev