I have just posted this new Monticello configuration to the "MagmaTester" project of squeaksource. I recommend using this for your development projects.
This rolls up several fixes and enhancements including, most recently, a major improvement in single-clause query performance. #size and #at: are now handled in a single, brief, server call! This is the way it was in Magma years ago when single-clause was all that was supported, before #where: came along.
Now, there are two subclasses of MagmaCollectionReader. MagmaCollectionFilteredReader and MagmaCollectionOptimizedReader. #where: will automatically determine whether your where: is a single clause and, if so, answer back the Optimized. The API is exactly the same, but Optimized will perform roughly 100X faster than Filtered with #size and #at:.
This, of course, creates a need for conversion if you have persistent readers in your repository. They will have to be converted to an instance of either subclass, but that conversion script is not yet written.
Enjoy, Chris
Hi there,
I was "googling" and "messing arround" Squeak for a couple of hours until I get how to load this release.
For the record: a fresh Monticello version is a prerequisite to make able Monticello to read and load .mcm files like r41Beta1 of this Magma release. So I've started from a fresh Squeak3.9-final-7067.image opened Monticello and went to MonticelloConfigurations, selected source.impara.de/mc repository, and loaded the lastest version of this packages in this order: 1. PackageInfo-Base 2. MonticelloConfigurations 3. Monticello
only then I've added 'http://www.squeaksource.com/MagmaTester' repository, opened it and selected r41Beta1, clicked the mcm file (this time withour error) and then the load completed succesfully.
cheers,
Sebastian Sastre
-----Mensaje original----- De: magma-bounces@lists.squeakfoundation.org [mailto:magma-bounces@lists.squeakfoundation.org] En nombre de Chris Muller Enviado el: Jueves, 28 de Febrero de 2008 01:04 Para: magma Asunto: r41Beta1
I have just posted this new Monticello configuration to the "MagmaTester" project of squeaksource. I recommend using this for your development projects.
This rolls up several fixes and enhancements including, most recently, a major improvement in single-clause query performance. #size and #at: are now handled in a single, brief, server call! This is the way it was in Magma years ago when single-clause was all that was supported, before #where: came along.
Now, there are two subclasses of MagmaCollectionReader. MagmaCollectionFilteredReader and MagmaCollectionOptimizedReader. #where: will automatically determine whether your where: is a single clause and, if so, answer back the Optimized. The API is exactly the same, but Optimized will perform roughly 100X faster than Filtered with #size and #at:.
This, of course, creates a need for conversion if you have persistent readers in your repository. They will have to be converted to an instance of either subclass, but that conversion script is not yet written.
Enjoy, Chris _______________________________________________ Magma mailing list Magma@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/magma
-----Mensaje original----- De: magma-bounces@lists.squeakfoundation.org [mailto:magma-bounces@lists.squeakfoundation.org] En nombre de Chris Muller Enviado el: Jueves, 28 de Febrero de 2008 01:04 Para: magma Asunto: r41Beta1
I have just posted this new Monticello configuration to the "MagmaTester" project of squeaksource. I recommend using this for your development projects.
This rolls up several fixes and enhancements including, most recently, a major improvement in single-clause query performance. #size and #at: are now handled in a single, brief, server call! This is the way it was in Magma years ago when single-clause was all that was supported, before #where: came along.
Now, there are two subclasses of MagmaCollectionReader. MagmaCollectionFilteredReader and MagmaCollectionOptimizedReader. #where: will automatically determine whether your where: is a single clause and, if so, answer back the Optimized. The API is exactly the same, but Optimized will perform roughly 100X faster than Filtered with #size and #at:.
This, of course, creates a need for conversion if you have persistent readers in your repository. They will have to be converted to an instance of either subclass, but that conversion script is not yet written.
Enjoy, Chris _______________________________________________ Magma mailing list Magma@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/magma
magma@lists.squeakfoundation.org