Contributors mailing list archives


Re: Migration to version 9

Camptocamp SA, Yannick Payot
- 31/08/2015 08:35:03
My vote goes for the KISS option to set the module installable = False

However, me must check that it will be loadable on It
might complain that some directories are not modules and forbid to
load the list of ported modules. (to be checked)
Yannick Vaucher
Business Solutions Software Developer

Camptocamp SA
PSE A, CH-1015 Lausanne
Phone: +41 21 619 10 30
Office: +41 21 619 10 10

On Mon, Aug 31, 2015 at 10:22 AM,  <> wrote:
> On Mon, Aug 31, 2015 at 9:52 AM, Guewen Baconnier
> <> wrote:
>> my voice goes certainly to the "installable: False" solution.
> Me too, obviously.
>> KISS and supports well the different git workflows (forward ports
>> merges...). Also, a script as proposed by Stéphane alleviates the
>> issue raised by someone which is that they don't know which module is
>> migrated or not.
> I also argues that giving adequate visibility to unported modules is
> important.
> Regarding the one module / one repo, here is my take on this.
> I believe the perimeter of each repository must match what each PSC is
> ready/willing/has the resources to maintain.
> We must not confuse the repository structure with packaging, release and
> dependency management. For the latter we have decent tools in python and we
> must strive to use them (ie and all the tooling building on top of
> that).
> It should be perfectly possible to have one PSC managing one repository
> containing many modules and making independent releases of each modules with
> proper versioning and dependencies.
> So I suggest
> * keeping the current repo structure for know
> * reinforce the role of PSC's (which is a goal of the current board)
> * let the repository structure slowly evolve according to PSC's and
> functional perimeters
> * work (again?) on proper packaging and release management at the module
> level
> Best regards,
> -sbi
> _______________________________________________
> Mailing-List:
> Post to:
> Unsubscribe: