Alvaro Fuentes
70c41dafd4
The main motivation for this change is to improve performance of BoM `explode`, which heavily rely on checking the lines to skip. `explode` is also called from product `compute_quantities_dict` which is called multiple times during upgrades. With this fix we improved the running time of an upgrade _test_ from 3h:30m to 2h:55m which gives a 16% performance improvement. The goal of `_skip_bom_line` is to check that for each attribute present on the line, at least one value associated to that attribute must be in the attribute values of the product(*). If none is found then we consider that we can skip the line. The previous implementation was inefficient. It grouped all values by attribute, then checked one by one if at least one value is on the product. In case one attribute does not have any value on the product it skipped the line. The implementation we propose here is to take the intersection of the product and line values, then check that their attributes are the same. The later can be done with a simple length check. In case they are different the line must be skipped. Note that this works because only one value is possible per attribute in a product. Both implementations are equivalent. The second is more efficient because does not branch and relies on (record)set operations. For example, let's consider a product with two attribute values `a` and `b`, and a line with multiple values `a`,`y` for attribute 1, and `z` for attribute 2. ``` Product Line +---+ +-----+ | a | <- same attribute -> | a,y | +---+ +-----+ | b | <- same attribute -> | z | +---+ + ----+ ``` This line must be skipped. The reason is that the value `b` is not among the list `[z]` of values for attribute 2 on the line. The new implementation would get the intersection of attribute values as `[a]` from there the comparison of the attributes will fail because `[a]` has only one attribute while the line has two. Let's consider a second case, where there is no value on the line for attribute 2. ``` Product Line +---+ +-----+ | a | <- same attribute -> | a,y | +---+ +-----+ | b | <- same attribute -> | | +---+ + ----+ ``` This line is not skipped because there is no value for attribute 2 on the line. Therefore the condition(*) per attribute is not violated for this product. The new implementation gets `[a]` as intersection of values, but now the attributes coincide: they are both attribute 1 for the intersection and the line. Finally, ``` Product Line +---+ +-----+ | a | <- same attribute -> | a,y | +---+ +-----+ | | <- same attribute -> | z,w | +---+ + ----+ ``` This line is skipped because none of `[z,w]` are in the product. The new implementation would get again `[a]` as intersection which does not match the attributes on the line. closes odoo/odoo#105019 X-original-commit: 47f4c3cc72ecb612ef4edd8c603e27b698187d8a Signed-off-by: William Henrotin (whe) <whe@odoo.com>
Odoo
Odoo is a suite of web based open source business apps.
The main Odoo Apps include an Open Source CRM, Website Builder, eCommerce, Warehouse Management, Project Management, Billing & Accounting, Point of Sale, Human Resources, Marketing, Manufacturing, ...
Odoo Apps can be used as stand-alone applications, but they also integrate seamlessly so you get a full-featured Open Source ERP when you install several Apps.
Getting started with Odoo
For a standard installation please follow the Setup instructions from the documentation.
To learn the software, we recommend the Odoo eLearning, or Scale-up, the business game. Developers can start with the developer tutorials