[16.0][MIG] purchase_order_qty_change_no_recompute: Migration to 16.0

[victoralmau]

Superseed #1886

Changes done:

• Make it work correctly taking into account that the price_unit, name and date_planned fields use the same compute method (and therefore the same depends).

@Tecnativa TT46595

[victoralmau]

After changing the approach of purchase_blanket_order and purchase_only_by_packaging to make them all compatible, I am now running into purchase_triple_discount issues.

Problem: Previously (v15) there was a context that allowed (in the tests) to omit onchange the relative to purchase_order_qty_change_no_recompute, so it "seemed" to have the standard operation and therefore the tests worked correctly.

Now: Using monkeypatching of get_depends() we can not perform test.

How could we add a context (or similar) in the purchase_order_qty_change_no_recompute tests so that in that case the test DOES change the behavior but in the rest it doesn't? (I thought about creating an ir.config.parameter with a value to re-set it when running the module tests, but I think that wouldn't work either). If we got to do it (in one way or another) the rest of changes in other modules will NOT be necessary.

Ping @pedrobaeza

[pedrobaeza]

In this case, there's no other solution except put it as rebel module. I have push an update branch with the following changes:

• No dependency on purchase_discount. This was added on the previous migration work and it's totally incorrect, and at first sight, unneeded. If something needs to be done, it should be done on the module itself or in a new glue module.
• Improve documentation and test.
• Proper fields in the monkeypatching.
• Remove other modules commits, as it's not needed.

[pedrobaeza]

/ocabot migration purchase_order_qty_change_no_recompute

[chienandalu]

Did you study the possibility to hook into the model _setup_fields method and use a similar estrategy to the one in _pop_field?

[pedrobaeza]

I already did. It doesn't serve, as the depends is something dynamic that can be expanded through other modules.

[chienandalu]

I see, the main issue here is that this technique can be use only once? Am I wrong?

[pedrobaeza]

Yes I'm afraid

[pedrobaeza]

Although maybe subsequent monkey patches will work over the previous replaced method...

[pedrobaeza]

/ocabot merge nobump

[OCA-git-bot]

Hey, thanks for contributing! Proceeding to merge this for you.
Prepared branch 16.0-ocabot-merge-pr-2193-by-pedrobaeza-bump-nobump, awaiting test results.

[OCA-git-bot]

Congratulations, your PR was merged at 6f0b690. Thanks a lot for contributing to OCA. ❤️