Odoo Changelog — 2026-05-29

Friday, May 29, 2026

2 changes

Branch: saas-18.2

1 change

Bug Fixes(1)

Resolved issues and error corrections

[PERF] l10n_se_sie: prevent OOM crashes on SIE 4 report

PR #118404

Enterprisesaas-18.2l10n_se_sie4HIGH

This update resolves a critical issue causing OOM crashes when generating the Swedish SIE 4 report with large datasets. By optimizing the database query and leveraging efficient data processing techniques, the report now executes much faster and with significantly reduced memory usage, ensuring reliable performance for users.

Original PR description

### Description of the issue/feature this PR addresses: Prevent Out of Memory (OOM) crashes and drastically improve execution speed when generating the Swedish SIE4 verification export for massive…

### Description of the issue/feature this PR addresses:
Prevent Out of Memory (OOM) crashes and drastically improve execution speed when generating the Swedish SIE4 verification export for massive datasets.

### Current behavior before PR:
When exporting a large volume of journal entries (e.g., 190,000+ account moves), the `_export_l10n_se_sie4_verification` method relies on iterating through heavy ORM recordsets and accessing relational child fields (move.line_ids) inside a loop. This triggers a severe N+1 query problem, maxing out server RAM and causing an OOM crash.

### Desired behavior after PR is merged:
The method now utilizes a hybrid data extraction approach:
- The ORM is used strictly to safely evaluate domains (multi-company rules, dates, states) and fetch a lightweight list of valid move_ids.
- A single SQL query with JOIN statements fetches all parent moves, child lines, and account codes in exactly one database query.
- itertools.groupby chunks the flat, lightweight dictionary results back into their respective journal entries.
The export now handles massive datasets in seconds with minimal memory overhead, while remaining perfectly secure.

### Benchmark:

For Memory:
| # Input Data | Before PR | After PR |
| -------- | -------- | -------- |
| ~190,000 moves | MemoryError | 407MB|
| ~200,000 moves | 1.8GB | 174.8 MB|

For Speed:
| # Input Data | Before PR | After PR |
| -------- | -------- | -------- |
| ~190,000 moves | MemoryError | 5.10s |
| ~200,000 moves | 1m29s| 5.3s|

### Reference:
opw-6067999

Forward-Port-Of: odoo/enterprise#117577
Forward-Port-Of: odoo/enterprise#113227

#performance#SIE4#Swedish Accounting

Branch: 18.0

1 change

New Features(1)

New functionality added to Odoo

[IMP] l10n_fr_pdp: authenticate before using pdp

PR #266981

Core18.0l10n_fr_pdpHIGH

This update enables French businesses to comply with mandatory electronic invoicing requirements by connecting to the Odoo-approved platform. It enforces authentication for PDP usage, adds support for French UBL invoicing formats, and incorporates tests for Flux 10 e-reporting, ensuring accurate and compliant data transmission.

Original PR description

Before this commit, we could use pdp without any authentication proof,
this commit will enfore the identity authentication before using pdp

task-5395365

#e-invoicing#france#peppol#compliance

### Description of the issue/feature this PR addresses: Prevent Out of Memory (OOM) crashes and drastically improve execution speed when generating the Swedish SIE4 verification export for massive datasets. ### Current behavior before PR: When exporting a large volume of journal entries (e.g., 190,000+ account moves), the `_export_l10n_se_sie4_verification` method relies on iterating through heavy ORM recordsets and accessing relational child fields (move.line_ids) inside a loop. This triggers a severe N+1 query problem, maxing out server RAM and causing an OOM crash. ### Desired behavior after PR is merged: The method now utilizes a hybrid data extraction approach: - The ORM is used strictly to safely evaluate domains (multi-company rules, dates, states) and fetch a lightweight list of valid move_ids. - A single SQL query with JOIN statements fetches all parent moves, child lines, and account codes in exactly one database query. - itertools.groupby chunks the flat, lightweight dictionary results back into their respective journal entries. The export now handles massive datasets in seconds with minimal memory overhead, while remaining perfectly secure. ### Benchmark: For Memory: | # Input Data | Before PR | After PR | | -------- | -------- | -------- | | ~190,000 moves | MemoryError | 407MB| | ~200,000 moves | 1.8GB | 174.8 MB| For Speed: | # Input Data | Before PR | After PR | | -------- | -------- | -------- | | ~190,000 moves | MemoryError | 5.10s | | ~200,000 moves | 1m29s| 5.3s| ### Reference: opw-6067999 Forward-Port-Of: odoo/enterprise#117577 Forward-Port-Of: odoo/enterprise#113227