Founder & CEO of Odoo, Fabien Pinckaers discusses the new Odoo 13 framework

New to Odoo?

 

The Odoo 13 framework is also called an In-Memory ORM, because it provides more considerable memory than before. When employed for operational measures, on an average, it runs 4.5 times faster when compared to earlier versions of Odoo.

Key features of Odoo 13 framework

Simplified Cache process

Pinckaers says that in the new framework, they have simplified the cache as the stored fields will now only need a single value. On the other hand, the non-stored fields’ computed value will depend on the keywords present in the context (eg. translatable and context).

He added that, in version 12, most fields did not need a cache so it contained only one global cache with an exception for fields that were text-dependent. It also had a new attribute for a multi-line inventory where the projects depend on “way roads”. However, the difficulty in this version is that when creating a field, users had to select the cache value and if the context of the field is changing, then the users had to again specify the new value of cache. This step is made simpler in version 13, as the user now needs to specify the value of the cache only once. “It seems simple but actually in the business code we're passing it to all the fields at the same time,” asserts Pinckaers. This simplified cache process will also reduce the alert memory access of the code.

In-memory updates

While specifying the various test field values, in the earlier versions, users had to update its validation value each time making it a time consuming process. To overcome this problem, the Odoo team has included all the data transactions in memory in the new version. Consequently,  in Odoo 13, when assigning the field value, the user can put it in the cache each time. Hence, when a field value needs to be read, it is taken from the cache itself.

To manage all the dependencies in Python, Pinckaers demonstrated how users should always: 

• Use the inverse field, instead of SQL query
Unlock access to the largest independent learning library in Tech for FREE!
Get unlimited access to 7500+ expert-authored eBooks and video courses covering every tech area you can think of.
Renews at €18.99/month. Cancel anytime
START FREE TRIAL
• Avoid using SELECT, as the implementation of the compute will read the same object
• When create(), set one2many to[]

Delaying the computing field for faster transactions

In order to delay a computing field in the line.product_quantity and the line.discount in the preceding Odoo versions, a user had to compute the dependency value for all the for line in order commands. Once the transaction was completed, the values were then recomputed and written in the code.

This process is also made easy in Odoo 13, as the user can now mark all the line commands to recompute and use the self.flush() command to compute the values after the transaction is completed. This makes all the computation transactions to be conducted in memory. According to Pinckaers, this support will help users with more than 100 customers as it will make the process much faster and simpler.

Optimize dependency tree to reduce Python and SQL computations

Pinckaers takes the ‘change order’ example to demonstrate how version 13 of Odoo has a clean dependency tree. This means that if the price list of the order is changed, the total cost of the order will also change indirectly, thus optimizing the dependency tree. He explains that this indirect change will happen due to the indirect dependency between the pricelist identity and the total cost list of the field in Odoo 13.

In the earlier versions, due to the recursive nature of the dependencies, each order of the line entailed the order ID of the field. This required the user to read sometimes even more than 100 lines of the list to get the order ID. In Odoo 13, this prolonged process is altered to get a more optimized dependency tree. This means that the user can now directly get the order ID from the dependent tree, without the Python and SQL computations. 

Improvements in browse optimization()

The major improvement instilled in Odoo 13 browse optimization() is the mechanism to avoid multiple format cache conversion. In the previous versions, users had to read and convert all the SQL queries to cache format followed by put in cache command. This meant that it required three commands just to read the data, making the process very tedious. With the latest version, the prefetch command will directly save all the similar data formats in the memory.

“But if the format is different, then we have to apply everything a color conversion method. As  Python is extremely slow,” Pinckaers says, “applying a dictionary that we see from outside the cache” makes the process faster because a C implementation can be used to directly convert the data in the cache format. You can watch the full video to see Pinckaer’s demonstration of code cleanup and Python optimization.

If you want to use Odoo to build enterprise applications and set up the functional requirements for your business, read our book ‘Working with Odoo 12 - Fourth Edition' written by Greg Moss to learn how to use the MRP module to create, process, and schedule the manufacturing and production order. This book will also guide you with in-depth knowledge about the business intelligence required in Odoo, its architecture and will also unveil how to customize Odoo to meet the specific needs of your business. 

Creating views in Odoo 12 – List, Form, Search [Tutorial]

How to set up Odoo as a system service [Tutorial]

Handle Odoo application data with ORM API [Tutorial]

Implement an effective CRM system in Odoo 11 [Tutorial]

“Everybody can benefit from adopting Odoo, whether you’re a small start-up or a giant tech company” – An interview with Odoo community hero, Yenthe Van Ginneken