Hi,
At ESUG's Camp Smalltalk we did an experiment with the loading the map via a changeset.
We did:
- make changeset (6679PachMC) that patches the MC Configurations to not load patches, but complete version (until the fixed squeaksource version is installed). - then we made a monticello configuration with the then current packages, put that into 6680Release-md changset.
If you load updates, the packages will be loaded according to the map defined. Very nice.
Two problems:
1) Afterwards,opening MCConfig Browser we get an empty configurarion. I would have thought that we want this to show the loaded config... 2) The update 6679 makes monticello believe that we have edited code, that will bring monticello to ask us to confirm that loading the package is ok.
I think we should release a 6680 3.9a image, this way people can test it. And we have something to replace the old 3.9a that is used a lot by people (the one that is older then 3.8...)
Marcus
Hi everyone. Just thought I'd let you know that I'm working on making the Traits stuff more friendly. This means among things that the packages shouldn't conflict with existing packages.
To this purpose, I intend to work against a 3.9 image, since that is all partitioned up. So an image that we are pretty sure will be the real basis for 3.9 would be useful.
I'm also in the process of writing a tool that's meant to make it easier to find and get rid of method overrides between packages.
Daniel
Marcus Denker wrote:
Hi,
At ESUG's Camp Smalltalk we did an experiment with the loading the map via a changeset.
We did:
- make changeset (6679PachMC) that patches the MC Configurations to not
load patches, but complete version (until the fixed squeaksource version is installed).
- then we made a monticello configuration with the then current
packages, put that into 6680Release-md changset.
If you load updates, the packages will be loaded according to the map defined. Very nice.
Two problems:
- Afterwards,opening MCConfig Browser we get an empty configurarion. I
would have thought that we want this to show the loaded config... 2) The update 6679 makes monticello believe that we have edited code, that will bring monticello to ask us to confirm that loading the package is ok.
I think we should release a 6680 3.9a image, this way people can test it. And we have something to replace the old 3.9a that is used a lot by people (the one that is older then 3.8...)
Marcus
Am 15.08.2005 um 16:13 schrieb Marcus Denker:
- Afterwards,opening MCConfig Browser we get an empty
configurarion. I would have thought that we want this to show the loaded config...
As I wrote previously: 'However, there is no such thing as THE configuration, so the config browser might not be the best place to put the option. Having a menu entry somewhere to browse the "system configuration" would make much more sense.'
Doug suggested having a method somewhere that answers a configuration which defines the packages belonging to the base system. You could get this, perform an update from image (to see the currently installed versions), and browse that.
- The update 6679 makes monticello believe that we have edited
code, that will bring monticello to ask us to confirm that loading the package is ok.
ChangeSets and MonticelloPackages do not work well together ... you could try to manually reset the "modified" flag of the working copy in the update.
- Bert -
- The update 6679 makes monticello believe that we have edited code,
that will bring monticello to ask us to confirm that loading the package is ok.
ChangeSets and MonticelloPackages do not work well together ... you could try to manually reset the "modified" flag of the working copy in the update.
Bad idea if you have modifications for real in your image. Those would get dumped without even asking you.
Cheers, - Andreas
Am 15.08.2005 um 17:44 schrieb Andreas Raab:
- The update 6679 makes monticello believe that we have edited
code, that will bring monticello to ask us to confirm that loading the package is ok.
ChangeSets and MonticelloPackages do not work well together ... you could try to manually reset the "modified" flag of the working copy in the update.
Bad idea if you have modifications for real in your image. Those would get dumped without even asking you.
Good point. You would have to check if there are no modifications before loading the change set.
- Bert -
On Mon, 15 Aug 2005 16:13:04 +0200, "Marcus Denker" denker@iam.unibe.ch said:
... I think we should release a 6680 3.9a image, this way people can test it. And we have something to replace the old 3.9a that is used a lot by people (the one that is older then 3.8...)
Probably a good idea. I will eventually get the source.squeakfoundation.org SS server working properly with the diffing, hopefully within the next week or two as I have time. We will still want this fairly soon, as the main slowdown with loading complete .mcz packages is not so much bandwidth, but the time it takes to load into the image.
But I agree that we should move forward, so let's go with your plan, and as soon as the SS server is fixed, we can change the Monticello package back again so that it loads .mcd files instead.
- Afterwards,opening MCConfig Browser we get an empty configurarion.
I would have thought that we want this to show the loaded config... 2) The update 6679 makes monticello believe that we have edited code, that will bring monticello to ask us to confirm that loading the package is ok.
With #2 you could try adding the test for changes in 6679 that Bert mentioned. With #1, also try Bert's other suggestion, but if that takes too much effort, we could fix that a bit later.
Then we could announce 3.9a-6680 as ready for people to try, with the caveat that we will be adding .mcd-loading capability relatively soon which should speed things up for people.
I notice that you've been doing a lot of harvesting of stuff into the 3.9a packages, which is great to see... people will want to try this stuff.
- Doug
packages@lists.squeakfoundation.org
-
Andreas Raab -
Bert Freudenberg -
Daniel Vainsencher -
Doug Way -
Marcus Denker