Since Dan is tentatively scheduled to give a presentation on modules at the SqueakEnd/MMWorkshop, I thought it would be a good idea to have a project oriented around that. I've only been following modules off and on for the last 5 or 6 months and don't really know what issues are critical at the moment.
Can anyone comment on the current state of modularity and what they think might make for a good module oriented project at SqueakEnd?
- Stephen
Stephen Pair wrote:
Since Dan is tentatively scheduled to give a presentation on modules at the SqueakEnd/MMWorkshop, I thought it would be a good idea to have a project oriented around that. I've only been following modules off and on for the last 5 or 6 months and don't really know what issues are critical at the moment.
Can anyone comment on the current state of modularity and what they think might make for a good module oriented project at SqueakEnd?
I agree that a module-related project might be a good one for the SqueakEnd, since a bunch of core Squeak folks will be there.
I wonder if it might be appropriate to try to work on converting another chunk or two of Squeak to be cleanly unloadable? (I thought I'd heard that unloading works, and that there are a couple parts of the system that are unloadable right now, something like Balloon. Is this true? I haven't poked around with trying to unload anything yet.)
- Doug Way dway@riskmetrics.com
That sounds like a good idea. I think it'll be a good way for a few of us to learn the module system. So the question is, what is a good chunk of squeak that needs to be made un-loadable? You mentioned Balloon...any others?
- Stephen
-----Original Message----- From: squeak-dev-admin@lists.squeakfoundation.org [mailto:squeak-dev-admin@lists.squeakfoundation.org] On Behalf Of Doug Way Sent: Saturday, March 30, 2002 12:37 AM To: squeak-dev@lists.squeakfoundation.org Subject: Re: [SqueakEnd] and modules
Stephen Pair wrote:
Since Dan is tentatively scheduled to give a presentation
on modules
at the SqueakEnd/MMWorkshop, I thought it would be a good
idea to have
a project oriented around that. I've only been following
modules off
and on for the last 5 or 6 months and don't really know what issues are critical at the moment.
Can anyone comment on the current state of modularity and what they think might make for a good module oriented project at SqueakEnd?
I agree that a module-related project might be a good one for the SqueakEnd, since a bunch of core Squeak folks will be there.
I wonder if it might be appropriate to try to work on converting another chunk or two of Squeak to be cleanly unloadable? (I thought I'd heard that unloading works, and that there are a couple parts of the system that are unloadable right now, something like Balloon. Is this true? I haven't poked around with trying to unload anything yet.)
- Doug Way dway@riskmetrics.com
Hi
henrik sent me the following information this may be useful for you. I did not had the time to look at that yet. But I will.
ducasse wrote:
Do you have some scripts? For me I was planning to take the discards methods and see if I can fileout them and filein them after.
For image decomposition, the analysis tools are in Module, categories "code analysis" and "system conversion". The most important ones are listed in the ModuleExplorer code pane. You should be able to understand it from the code.
Then look at the ModuleRefactorer class and the FromVersion... classes to see how to code a class that will refactor the system. Feel free to extend this as it was only written to cover what I needed to have so far.
Let me know if you need more help.
Henrik
Stephen Pair wrote:
That sounds like a good idea. I think it'll be a good way for a few of us to learn the module system. So the question is, what is a good chunk of squeak that needs to be made un-loadable? You mentioned Balloon...any others?
- Stephen
Sorry Stephen, I have been intending to write something up for you--not enough hours in the day--but until then there's a page on the Swiki called
"A guide for refactoring the image" (partial, but with info on what can be outloaded right now)
and another called
"Modules: What to do next"
plus a few more pages that should be helpful. The latter page holds to-do items that are up for picking. This should obviously be complemented with a "done items" page listing what works. But basically the test cases should be the benchmark of what should work, anything else is a bonus ;-)
The new class comments were also posted with you in mind.
Our week in Bern (29/4-3/5) comes just before yours (3-5/5). I will definitely take time then to write down better instructions for refactoring, and then we'll also have a chance to "pass on" whatever results or conclusions we reach.
Henrik
Here is what henrik sent me. It seems that my first attempt to send it into the list failed
ducasse wrote:
Do you have some scripts? For me I was planning to take the discards methods and see if I can fileout them and filein them after.
Sorry you should have had this a long time ago.
For image decomposition, the analysis tools are in Module, categories "code analysis" and "system conversion". The most important ones are listed in the ModuleExplorer code pane. You should be able to understand it from the code.
Then look at the ModuleRefactorer class and the FromVersion... classes to see how to code a class that will refactor the system. Feel free to extend this as it was only written to cover what I needed to have so far.
Let me know if you need more help.
Henrik
squeak-dev@lists.squeakfoundation.org