[squeak-dev] Smaller mcd
nicolas.cellier.aka.nice at gmail.com
Fri Mar 6 16:59:50 UTC 2020
Le ven. 6 mars 2020 à 01:49, Bert Freudenberg <bert at freudenbergs.de> a
> On Thu, Mar 5, 2020 at 4:28 PM David T. Lewis <lewis at mail.msen.com> wrote:
>> On Thu, Mar 05, 2020 at 10:34:52PM +0100, Nicolas Cellier wrote:
>> > Hi all,
>> > While reviewing some contributions, I saw this: Monticello-mva.667
>> > Browse the whole thread, it's interesting.
>> > How can we finish the job and make it happen?
>> Thank you for reviving this thread. It is a very worthwhile proposal
>> and it is sad to think that we let it drop for so long. Milan Vavra
>> clearly put a lot of thought into this, and in the earlier discussion,
>> Bert gave the concept a nod of approval too.
>> If I understand the discussion correctly, there are a couple of potential
>> issues that may need to be addressed concerning possible impact of
>> source.squeak.org and squeaksource.com and ss3, but I'm sure we can
>> follow up on those issues. Aside from that, there seems to be no reason
>> that this cannot be tested, reviewed, and moved into trunk.
>> I am saying this without haviing actually tested it myself, but it
>> really does look like a good idea, and we should find a way to make
>> it happen.
> Agreed, this looks worthwhile doing.
> The tricky part is the diffs generated by the server. We would need some
> extended naming scheme to indicate that the MC client is able to handle the
> pruned histories.
> - Bert -
> As I understand it, there are two issues:
1) the server must serve both old image without short history support, and
therefore, some discrimination has to happen in the query
code for generating short and long mcd must coexist
2) old and new images might share a common package-cache, and old image
shall not use mcd with shorten history
The thread contains some proposals for preventing such accidental
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Squeak-dev