On Dec 9, 2011, at 24:06 , Alexandre Bergel wrote:
With the team at Bordeaux, we are working on an implementation to query histories (the tool is named VPraxis).
Which kind of query are you interested in? Is Hismo not enough?
We store facts, not models. It allows us to have the history (which method have been removed...) The use of prolog allows VPraxis to be very fast, as it can store hundreds versions.
I worked on a configuration importer. We can analyze many many versions now :-)
Cool ! How do you bypass the problem of memory in Pharo ?
Jannik
I implement a importer for Smalltalk. For now, I can load histories for a package, for a metacello configuration, from a repository.
Have a look at the Spirit project on ss3.gemstone.com/ss It contains a lot of scripts to load versions of Mondrian, Versionner, Spy, ...
A first part is build in Pharo (building snapshots from Monticello), and a second part in build with our java software and build the history file. We can export xml, sql or swipl, which allows us to build prolog queries.
Here (http://code.google.com/p/harmony/wiki/VPraxis) is a tutorial for the java VPraxis. For Smalltalk, I am working on it.
I generated the current Mondrian history in a swipl format, available here: http://dl.dropbox.com/u/7739334/MondrianHistorySwipl.zip An important point is that we do not into account the multiple branches (it is a current work :) ). So, when there are multiple branches, for now, we check were is the merge and we "ignore the multiple branches".
Now, for your question, we create a script: http://dl.dropbox.com/u/7739334/mondrianScript.txt Using swi-prolog (http://www.swi-prolog.org/) , you can call, in this script, the method moreN(L, 10).
For the package Mondrian-Layout, it returns: L = [id_MOAbstractGraphLayout, id_MOCircleLayout, id_MOSugiyamaLayout]. Which are the three classes that have changed more that 10 times.
For now, I am working on packaging all the things together. But if you have some other requests, do not hesitate to ask me :)
Looks cool!
Alexandre
On Nov 24, 2011, at 14:05 , Alexandre Bergel wrote:
Having a reification in Moose of 100 versions of Mondrian for example :-)
Just answering the question 'Which classes and methods of Mondrian have changed more than 10 times since the day Mondrian was born?' cannot be easily done without a lot of memory
Alexandre
On 24 Nov 2011, at 03:27, Francois Stephany wrote:
I'm wondering: how big is a dataset > 500MB ? I've no idea how big it is. Alex, what is your use case (in practice!) for more than 500MB?
On 23/11/11 18:25, Igor Stasenko wrote:
It is problematic, and requires different memory management than we currently have. I think if you need really big data sets, then use gemstone, which is developed to deal with that specifically.
-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
Jannik Laval
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
--- Jannik Laval