Hi Petr,

> Is possible to make brief list "what needs to be done for 100% functional port for Pharo" and share it on some central place (github)? Maybe some terse todo overview about classes and specific porting issues.

In Udo's prior engagement on this list about porting Magma, I outlined
the list frameworks and test cases that would need to be ported in
dependency order.

> I personaly can very slowly rewrite, for example, something about file/dir access (use Pharo file system classes instead of Squeak file/stream classes), but hardly something about hasihing, indexing or something insisting deep knowledge about how Magma works internally. I am interested in Magma, because the need of "just simple, but transactional persistent object storage" annoys me frequently these days (GLORP or NoSQL databases are not the proper answer for persisting Smalltalk/Pharo native application domain models IMHO).
>
> Is there something like more detailed doc (but not functionally obsolete), which would help me understand how the Magma works inside?

This is an excellent, concise description.

   http://book.seaside.st/book/advanced/persistency/magma/workings

> A know only about this historic wiki place:
> http://wiki.squeak.org/squeak/2665

Inside that wiki is another PDF which describes the internals of
Magma's HashIndex files.

Best,
  Chris

_______________________________________________
Magma mailing list
Magma@lists.squeakfoundation.org
http://lists.squeakfoundation.org/mailman/listinfo/magma