Closed sbidoul closed 5 years ago
Merging #1 into master will decrease coverage by
1.01%
. The diff coverage isn/a
.
@@ Coverage Diff @@
## master #1 +/- ##
==========================================
- Coverage 96.96% 95.95% -1.02%
==========================================
Files 1 1
Lines 198 198
Branches 31 30 -1
==========================================
- Hits 192 190 -2
- Misses 3 4 +1
- Partials 3 4 +1
Impacted Files | Coverage Δ | |
---|---|---|
src/initializer.py | 95.95% <0%> (-1.02%) |
:arrow_down: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 50cd532...de4788e. Read the comment docs.
Sure. merged.
Although I do not understand the goal of this fork
This is an attempt of unbundling contrib. While contrib
might work for some people, it doesn't work for all, but some components of it might indeed.
It tries to solve for example those problems of scope. By decoupling, all those problems become non-existent all of a sudden.
Would you be willing to host click-odoo-initializer
under acsone
's github org? So that it becomes a concerted move? - There is close to nothing I have added so far ... ... to me that would feel like the right thing to do (and delete this xoe-labs repo).
Moving click-odoo-initdb to another project would create compatibility issues for people expecting to have it when installing click-odoo-contrib. So, rather no.
@sbidoul Moving it out and developing it further outside while keeping contrib
as is? Combined with a classical deprecation notice that doesn't seem too crazy to me... (And honestly, there is no intent on my side of forking this work at all, if it can be avoided)
Still no, sorry. Too much trouble for no benefit that I'd understand.
Btw, I plan to enhance other components of contrib, gradually, and I'm sure they would be all "out of scope" to be accepted upstream. I feel: this is an artificial constraint placed upon collaboration by the mere fact of bundling otherwise not related (within the click-odoo scope) components together... This dynamic is virtually forcing into involuntary forking. I don't think that's something that could be avoided without unbundling.
Let's advise if/when that happens.
makepot
is a candidate to be enhanced into a full web late integration capable of doing git-commits, as an example. But I can agree, to demonstrate the problem in a real case (if/when that happens :wink:)... I hope, though, I may ask you to do me a favor and keep those "buts" somewhat in the back of your mind going forward.
Although I do not understand the goal of this fork, allow me to suggest a couple of clarifications.