Magma June 2007

magma@lists.squeakfoundation.org

10 participants
16 discussions

MagmaReadStrategies
by Keith Hodges 17 Jun '07

17 Jun '07

Dear Chris, I was using read strategies and found a bug in the depths generation process, here is a test which shows the problem. Problem A: Subclass defines depths on instVars, superclass defines a depth for all instVars - raises an error .. not indexable. This test case demonstrates the problem: (MagmaSession new) readStrategy: (MaReadStrategy new onAny: Collection readToDepth: 2; forVariableNamed: 'tally' onAny: Dictionary readToDepth: 3; yourself). Problem B: Looking at the code, if a class defines a depth for all instVars, no attempt is made to see if the superclasses have defined a greater depth. The code has an (a max: b) statement as if to intend this behaviour but in reality it looks to me like it is actually (a max: a). (MagmaSession new) readStrategy: (MaReadStrategy new onAny: Collection readToDepth: 3; onAny: Dictionary readToDepth: 2; yourself). Also Problem C: The ObjectExplorer behaviour is not completely in sync with Magma's behaviour. Ideally you want Magma to realize objects when you select them in explorer, not unexplicably due to explorers inner workings. And finally very useful Logging "D" ------ Solutions: In package 'magma seaside' I have included the fixes for A and B, in WAMagmaReadStrategyDefault. I have also added a couple of methods to MagmaMutatingProxy, and a MagmaProxyStandIn class to improve the way that ObjectExplorer behaves. Now you can really explore how magma thinks when it is using read strategies. Logging: The class WAMagmaReadStrategyWithQ2Logging, logs realizations verbosely to the Q2Log, a logging facility available in Gjallar. This is very useful. My fixes for A and B, have a couple of enhancements to the basic scheme. Negative depths are valid and subtracted from the inherited value. This can be useful so as to specify pulling a whole datastructure except this instVar or class type. ---- cheers Keith

3 6

OODB Design Question
by Michael Gorsuch 14 Jun '07

14 Jun '07

Hello, I'm looking to port one of my side projects to Seaside with a Magma back end, and need a little clarification on how Magma would handle one particular situation. This service is an online dream journal (http://www.dreamjournal.com), and has a very simple object model. Essentially, we have a User model, and a Dream model. User has an attribute called 'dreams' that holds all of that user's dream info. Dreams also need to be accessed by the public (if you mark the dream as so), therefore I was also holding all of my dreams in another collection (as well as with the User). This made it easy to handle sorting and what not for each view. Most importantly, the Dream object inside the User's 'dream' collection is the same exact Dream object stored in the master dream collection, allowing for changes to be made and reflected across both areas with ease. If a user edits a particular dream, it changes in both collections. So. Finally. The question: If I commit a User in a Magma repository, I know that all of his dreams will be saved with him. But, if I commit the master dream collection to the repository, and retrieve it, will it still contain the same exact objects that the User's have, or will they merely be copies? If they are copies, perhaps someone can suggest a more intelligent way to manage this dataset. Thanks! Michael -- Styled Bits 2.0: They're Mostly Broken, But I Try http://www.styledbits.com

2 1

MagmaCounter
by Balázs Kósi 12 Jun '07

12 Jun '07

Hi Chris, We would like to add objects from concurrent transactions to a MagmaCollection, which objects have unique ids, which ids are monotonically increasing in time. I've googled for this problem, and found a thread on this mailing list about generating unique ids. Particulary in this email ( http://lists.squeakfoundation.org/pipermail/magma/2005-August/000029.html ) you describe a MagmaCounter, which would be perfect for us, i think. Can you give us some pointers if you have an implementation, or help us getting started with the coding? Thanks, Balázs

2 1

Fwd: Query optimisation
by Chris Muller 09 Jun '07

09 Jun '07

Hi Florian, do you have Monticello Configurations loaded? You need to load that from impara's SqueakSource repository (add repository "http://source.imara.de/mc". THEN, you may load the r40Alpha1 *configuration* from the MagmaTester project. Hope this helps. On 6/8/07, Florian Minjat <florian.minjat(a)emn.fr> wrote: > I went on the MagmaTester repository on SqueakSource. > There are a lot of packages and I am quite confused ^^. > Should I load MagmaClientLoader and MagmaServerLoader or Magma server > and Magma client ? Do I still need to load Magma seaside ? > > By the way I just submitted 'Magma Client' and 'Magma Server' packages > on the Universe Browser to ease the installation of a new image. > > Florian > > Chris Muller wrote: > > Sure thing. I just a posted a new configuration to "MagmaTester" > > Project on SqueakSource under the name: > > > > r40Alpha1 > > > > This is the latest version of all packages. All tests pass, and > > legacy databases will automatically be upgraded to this version > > (you'll see a warning first, and you can't go back once you Proceed). > > > > I don't anticipate anything left in the final r40 that using this > > would create any issues migrating to the final version when it's > > finally done, so it should be ok to go ahead and use this. I can't > > guarantee that, of course. :) > > > > Enjoy. > > > > > > On 6/7/07, Florian Minjat <florian.minjat(a)emn.fr> wrote: > >> Chris Muller wrote: > >> > Today I noticed a performance bug; single-clause queries do not need > >> > to go through a (expensive) routine needed only for multi-clause > >> > queries. > >> > > >> > A simple guard in the code has now sped up single-clause queries by 100 > >> > times. > >> > > >> > A single-clause query is one which has no more than one "range" of > >> > keys specified, like: > >> > > >> > events where: > >> > [ : each | > >> > each date > >> > between: '1/1/2006' asDate > >> > and: '12/31/2006' asDate ] > >> > > >> > Like in older Magma (pre-queries) this is really fast again. > >> > >> Hi, > >> That's really cool because I only do single-cause query. > >> Is this fix available on the repository ? > >> > >> Florian > >> _______________________________________________ > >> Magma mailing list > >> Magma(a)lists.squeakfoundation.org > >> http://lists.squeakfoundation.org/mailman/listinfo/magma > >> > > > _______________________________________________ > Magma mailing list > Magma(a)lists.squeakfoundation.org > http://lists.squeakfoundation.org/mailman/listinfo/magma >

1 0

Query optimisation
by Chris Muller 08 Jun '07

08 Jun '07

Today I noticed a performance bug; single-clause queries do not need to go through a (expensive) routine needed only for multi-clause queries. A simple guard in the code has now sped up single-clause queries by 100 times. A single-clause query is one which has no more than one "range" of keys specified, like: events where: [ : each | each date between: '1/1/2006' asDate and: '12/31/2006' asDate ] Like in older Magma (pre-queries) this is really fast again.

2 3

Universe Magma Glorp
by Ramiro Diaz Trepat 05 Jun '07

05 Jun '07

Is there any reason why Magma and Glorp are not available in the Package Universe browser for installation? Thanks r.

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Magma June 2007