**Steps to Reproduce:**
-  Create a Server Action of type 'Webhook Notification'.
- Select a model containing a field that returns a `frozendict`-based structure (e.g. `account.move` → `needed_terms`).
- Add the field to the webhook fields.
- Open the webhook sample payload preview.

**Issue:**
- During sample payload generation:
   - The selected fields are read from a sample record.
   - A selected field returns a structure containing `frozendict` objects.
   - The payload is serialized using `json.dumps()`.
   - JSON serialization fails with:
```text
TypeError: keys must be str, int, float, bool or None, not frozendict
```
  - The webhook sample payload computation crashes and the preview cannot be displayed.

**Root Cause:**
- The webhook sample payload may contain `frozendict` objects returned by selected fields.
- The serializer used for payload generation does not handle such mapping-like objects, causing  `json.dumps()` to fail.

**Solution:**
- Use a serializer that converts mapping-like objects into JSON-compatible structures before serializing the webhook sample payload.

**OPW-6295777**

Description of the issue/feature this PR addresses:

Current behavior before PR:

Desired behavior after PR is merged:




---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

Forward-Port-Of: odoo/odoo#272324
Forward-Port-Of: odoo/odoo#271864

### Contains the following commits:

https://github.com/odoo/o-spreadsheet/commit/3529978d50 [REL] 19.2.19 [Task: 0](https://www.odoo.com/odoo/2328/tasks/0)
https://github.com/odoo/o-spreadsheet/commit/156b8921dc [FIX] config: bump node version in GH action [Task: 0](https://www.odoo.com/odoo/2328/tasks/0)
https://github.com/odoo/o-spreadsheet/commit/f3ae8587f9 [IMP] pivot: give full dimension to pivot normaliser [Task: 6023622](https://www.odoo.com/odoo/2328/tasks/6023622)

Co-authored-by: Florian Damhaut (flda) <flda@odoo.com>
Co-authored-by: Anthony Hendrickx (anhe) <anhe@odoo.com>
Co-authored-by: Alexis Lacroix (laa) <laa@odoo.com>
Co-authored-by: Lucas Lefèvre (lul) <lul@odoo.com>
Co-authored-by: Adrien Minne (adrm) <adrm@odoo.com>
Co-authored-by: Ronak Mukeshbhai Bharadiya (rmbh) <rmbh@odoo.com>
Co-authored-by: Dhrutik Patel (dhrp) <dhrp@odoo.com>
Co-authored-by: Rémi Rahir (rar) <rar@odoo.com>
Co-authored-by: Pierre Rousseau (pro) <pro@odoo.com>
Co-authored-by: Vincent Schippefilt (vsc) <vsc@odoo.com>
Co-authored-by: Marceline Thomas (matho) <matho@odoo.com>

#### Description of the issue this PR addresses:
- When an emoji shortcut ending with `/` (e.g. `:/` for 😕) is typed, the emoji plugin replaces the characters before the powerbox `on_input_handler` runs. Since `ev.data` still reflects the original typed `/`, the powerbox was incorrectly opening.
- Emoji shortcuts works only when it is used at the end of a text node, because the matching logic checked the whole remaining substring from the current position.
- Sometimes, pressing Backspace splits one text node into two, and then an emoji shortcut works at the end of the first text node even when the paragraph is visible as a single line.

#### Desired behavior after PR is merged:
- Check the DOM character at cursor position instead of `ev.data` to determine whether `/` is actually present before opening the powerbox.
- Emoji shortcuts now works when used with a preceding space anywhere in the paragraph.

Enterprise PR-https://github.com/odoo/enterprise/pull/118310
task-6243724

---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

- Description of the issue/feature this PR addresses:
- The mandatory product_id field required by the many2one_uom widget was omitted from the list view, causing the issue.
- The issue occurred when the customized 'Extend Order' button was clicked, opening the wizard with all sales order lines loaded into its one2many field.
- Error message: UncaughtPromiseError > OwlError

Uncaught Promise > An error occured in the owl lifecycle (see this Error's "cause" property)

Occured on apollohomecare-migration-v19-33341368.dev.odoo.com on 2026-06-26 10:18:38 GMT

OwlError: An error occured in the owl lifecycle (see this Error's "cause" property)
    Error: An error occured in the owl lifecycle (see this Error's "cause" property)

Caused by: Error: The widget 'Many2OneUomField' (field 'product_uom_id') needs a 'product.product' or 'product.template' field. 'product_id' is used but is related to 'undefined' model.

Description of the issue/feature this PR addresses:

Current behavior before PR:

Desired behavior after PR is merged:




---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

Forward-Port-Of: odoo/odoo#272417

Currently, an error occurs on an invoice when user selects a payment term and removes the currency.

Steps to replicate:
- Install account
- Turn on multiple currencies
- Open invoices.
- Create a new invoice 
- Add a line
- Add a customer
- Save
- Select payment term as `30% Now, Balance 60 Days`
- Remove the currency.

Error:
```
ValueError: Expected singleton: res.currency()
```

Cause:
- This error only occurs when the selected payment term contains at least two due term lines [1].

- When the selected payment term has atleast two lines the check [1] assigns `on_balance_line` as  false and the `else` block is evaluated where currency being an empty recordset (as the user removed it) causes the error from [line] when trying to perform `round()` on an empty res.currency recordset.

Solution:
- As the currency is a required field, user will not be able to save the record until a currency is assigned.

- Used journal's currency or company's currency as a fallback when computing payment terms if the current currency is empty.


[1]: https://github.com/odoo/odoo/blob/177002c0aca95de2de6458bb65bedf5c74541ac0/addons/account/models/account_payment_term.py#L229

[line]: https://github.com/odoo/odoo/blob/177002c0aca95de2de6458bb65bedf5c74541ac0/addons/account/models/account_payment_term.py#L240

sentry-7569922293

---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

Forward-Port-Of: odoo/odoo#272073

# How to reproduce
- Create a new Receipt
- Save
- Delete the new Receipt

# The issue
A traceback is shown :
`TypeError: Cannot read properties of undefined (reading 'label')`

# Cause
This is caused by the custom status bar for pickings `StockPickingLockedStatusBarField`. In its template, we replace the display of the current label : https://github.com/odoo/odoo/blob/490c355ae0bc77c1106e22b3afb2206583980324/addons/stock/static/src/fields/stock_picking_locked_statusbar_field.xml#L20-L23 https://github.com/odoo/odoo/blob/490c355ae0bc77c1106e22b3afb2206583980324/addons/stock/static/src/fields/stock_picking_locked_statusbar_field.xml#L4-L9

The issue is that the base implementation of the current label properly handles the case were no item is currently selected:
https://github.com/odoo/odoo/blob/7630f8fe2d5198b7a1ed538241795dc26a497fa0/addons/web/static/src/views/fields/statusbar/statusbar_field.js#L298-L300

But the picking implementation does not :
https://github.com/odoo/odoo/blob/490c355ae0bc77c1106e22b3afb2206583980324/addons/stock/static/src/fields/stock_picking_locked_statusbar_field.js#L12-L14

And it seems that the template is quickly rendered without any selected item before deletion.

opw-6345192

Forward-Port-Of: odoo/odoo#273030

**Current behavior before PR,**
Computing `main_user_id` of a partner involved filtering active users within a
loop. To avoid losing the prefetching, we manually forced all partner users to
be kept in the prefetch during each iteration. This caused performance issues as
the recordset grew large (e.g., during auto-subscription).

**Desired behavior after PR is merged,**
All partner users are prefetched and filtered once, removing the need for a
repetitive manual prefetch.

**Benchmark,**
The following observations were recorded when auto-subscribing users to a
discuss channel at different scales.
The _Before_ and _After_ results represent the max values from three consecutive
tests.

|  Records |   Before |      After |
| :--------| -------: | ---------: |
|    3k    |   ~2.3s  |    <90ms   |
|    5k    |   ~3.8s  |    <160ms   |
|    10k   |   ~7.6s  |    <300ms  |

part of task-6116079



---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

Forward-Port-Of: odoo/odoo#271549

Steps to reproduce
------------------
1. Install l10n_ke_edi_oscu_pos.
2. Select the Kenyan company.
3. Enable eTIMS on the PoS.
4. Register the products inside a combo, but not the combo itself.
5. Sell the combo in the PoS.

Observation
-----------
We see a warning that the combo must be registered to eTIMS, and the order can't be validated.

What's happening
----------------
In the PoS a combo adds a 0 price parent line for the combo product, but the combo is not a real item to send to eTIMS, only the products inside it are, and (as per step 4) the combo is not registered.

`checkEtimsFields` sees the combo as not registered, so it raises the warning in `showUnregisteredProductsWarning` and blocks the payment in `validateOrder`.

Fix
---
In the backend, we skip sending the parent combo line to eTIMS, and on the frontend, we make the combo parent line not need eTIMS registration, so the warning and the block don't apply to it.

opw-6253306

Forward-Port-Of: odoo/enterprise#122179
Forward-Port-Of: odoo/enterprise#119362

The following error is raised in the POS when paying with an Adyen terminal:
```
TypeError: Cannot read properties of undefined (reading 'uuid')
at Proxy.handleAdyenStatusResponse
```

Adyen delivers webhook notifications at-least-once, so the ADYEN_LATEST_RESPONSE event can fire several times for a single payment, running handleAdyenStatusResponse concurrently. After the await on get_latest_adyen_status, a previous (duplicate) notification may already have resolved the payment line, so getPendingPaymentLine no longer returns it and the subsequent line.uuid dereference crashes.

opw-6237987 patched the same root cause on a single line by adding an optional chaining operator in isPaymentSuccessful, which only moved the crash to the next dereference.

Fetch the pending line once at the start of handleAdyenStatusResponse and bail out when it is gone, so every dereference below is safe. The same guard is added to the remaining branches of _adyen_handle_response for consistency with the existing Reject branch.

opw-6237987

---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

Forward-Port-Of: odoo/odoo#271181
Forward-Port-Of: odoo/odoo#269720

Version:
--------
- 18.0+

Steps to reproduce:
-------------------
- Install `quality_mrp`
- Create a manufactured product with a BoM
- Create a Quality Point for the `Manufacturing` operation of that product
- Create and confirm multiple Manufacturing Orders
- Verify that each MO generates a quality check
- From the MO list view, select the MOs and merge them from the gear menu(merge)

Issue:
------
When Manufacturing Orders are merged, All MOs are cancelled but it keep their
quality checks in the 'To Do' state.

As a result:
- The quality checks remain linked to cancelled MOs
- The 'Quality Checks' smart button is still displayed on cancelled MOs

Expected behavior:
------------------
- Pending quality checks should be deleted when the MO is cancelled
- The 'Quality Checks' smart button should no longer be displayed

Cause:
------
A previous fix introduced logic to remove pending quality checks when a
Manufacturing Order is cancelled:
odoo-dev@db93bd2

This logic was implemented in `action_cancel()` by unlinking quality
checks associated with the cancelled MO:
https://github.com/odoo/enterprise/blob/20bc0eb5c2cec67eecd3b44450934e23370b48f2/quality_mrp/models/mrp_production.py#L94-L97

However, when MOs are merged, the merge flow does not call
`action_cancel()`. Instead, it directly invokes `_action_cancel()` on
the source Manufacturing Orders:

https://github.com/odoo/odoo/blob/aca0b7289c68fc7a75d47ab313f5f791ebf30f7d/addons/mrp/models/mrp_production.py#L2480


Since the quality check cleanup is implemented only in
`action_cancel()`, it is bypassed during the merge process.

As a result, the source MOs are cancelled but their pending quality
checks remain in place.


---

opw-6260735

Forward-Port-Of: odoo/enterprise#121717
Forward-Port-Of: odoo/enterprise#119525

### Steps to reproduce:

- Create a storable product P tracked by SN, fifo perpetual lot valuated using the MTO route and with a BOM: 1 x COMP
- Create, confirm and validate an MO for 10 unit of P using COMP's at 10$ (by setting its standard price).
- Create and confirm a Sales Order for 1 unit of P
> This generate an MO
- Validate this MO for a new serial say SN011 using a COMP at 20$.
- Validate the Delivery Order using SN011
- Create and post the customer invoice
#### > The invoice COGS uses Std Price rather than the 20$ lot's fifo value

### Cause of the issue:

The cogs value are generated based on the moves returned by the `_get_stock_moves` call of the acount.move.line:
https://github.com/odoo/odoo/blob/6a356cb0640ce20c71f20a8bd64ef99e76a82489/addons/stock_account/models/account_move_line.py#L66-L68 Currently, if an account.move.line is linked to a sale.order.line, this methods returns the entire pull of stock moves linked to the sol: https://github.com/odoo/odoo/blob/79c9e7de6764e7f8e47709b827df6eee81e72637/addons/sale_stock/models/account_move.py#L155-L156 However, these moves include both, the delivery move and the `move_finished_ids` of the MTO prodcution. This is problematic since the delivery move is considered as positive cogs qty and the MO is considered as incoming cogs qty leading to a sum of 0 cogs qty which in turns make the cogs price unit fall back to the product standard price instead of the fifo cost of the produced lot.: https://github.com/odoo/odoo/blob/6a356cb0640ce20c71f20a8bd64ef99e76a82489/addons/stock_account/models/stock_move.py#L257-L271

opw-6292048


---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

Forward-Port-Of: odoo/odoo#271018

### Steps to reproduce:

- In the settings enable: Multi-Steps Routes
-  Set your warehouse to manufacture in 2 steps (pick then manufacture).
- Create a final product (FP) with a BOM in flexible consumption:
  - 1 x COMP (lot tracked)
- Create and confirm an MO for 1 units of FP
- Set the quantity producing on the MO to 1
> The consumed qty was updated to 1 unit
- Set a lot on the pre-production pikcing and validate
#### > The lot is not transfered to the MO which you are not able to validate since the registered component is lot less

### Cause of the issue:

The issue is caused by https://github.com/odoo/odoo/commit/3223deb871ca4cb4ac0381e4321f2dbf79a60189 as the `qty_waiting` is based on the reservation state of the move origin of the move rather than its actual demand:
https://github.com/odoo/odoo/blob/00118002bd6eab2f4c34a32e993a9219fded06ac/addons/mrp/models/mrp_production.py#L1419-L1426
In particular, since the backorder of the pre-production picking was not reserved (since nothing was available in stock), it was not taken into account as it should have been.

Issue 2:

Steps to reproduce:

- In the settings Enable Multi-Steps Routes
- Unarchive MTO
- Create 3 products: 
  - Final Product: Tracked by SN with a BOM: 1 X Super Component
  - Super Component: Tracked by SN, MTO with a BOM: 1 X Component
  - Basic Component: Put 10 units in stock
- Create and confirm an MO for 3 units of Final Product
> This should create an MO for 3 units of Super Component
- Go to the Child MO > Cogs wheel > Split in 3 MO's
- Click "Generate serial" on each Child MO and validate the first one
- On the MO for Final Product > Cogs wheel > Split in 3 MO's
- On the first MO, click "Generate Serial"
> Error: Reserving a negative quantity is not allowed.

### Cause of the issue:

The `action_generate_serial` calls in turn the `set_qty_producing`:
https://github.com/odoo/odoo/blob/9fac1400fe5a8e665732c2ee3701c17c3495318f/addons/mrp/models/mrp_production.py#L1601
However, since the main MO was split the Super component demand is of 1 but each child MO provide an origin quantity of 1 so that the `new_qty` will be set to a negative one here:
https://github.com/odoo/odoo/blob/9fac1400fe5a8e665732c2ee3701c17c3495318f/addons/mrp/models/mrp_production.py#L1418-L1426
But, since the first child MO was validated, there is already a move line associated to the Super component move and the `_set_quantity_done` will therefore try to adapt the reservation to a negative quantity which leads to the error:
https://github.com/odoo/odoo/blob/9fac1400fe5a8e665732c2ee3701c17c3495318f/addons/stock/models/stock_move_line.py#L469-L470

opw-6128575
opw-6317083

---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

Forward-Port-Of: odoo/odoo#271123

The previous memory profiler was causing a lot of performance issues.
This is because tracemalloc tracks the allocations that happens at the
python interpreter level by attaching to the cpython allocators. This
first meant each allocation that happens through python has to go
through a callstack while holding the GIL and preventing the thread and
other threads from operating. This callstack does multiple things,
first is walking the allocation back from the current frame up until
the specified frame depth at the start of collection. The other is
updating the internal object that keeps track of the allocations and
what cause them up until now which degrades the performance even more
when the allocator keeps running for a long time. Increasing the frame
depth also means the partitioning becomes even more fragmented in the
internal object and leads to higher memory usage. This in turns means
lower performance as well. The issue becomes more evident when the
overhead of tracemalloc blocks any execution even turning it off
because the gil cannot be released until the full allocation execution
happens.

Currently this would happen on long enough requests or a high enough
depth.

Two PRs were made to try to address this issue.

1- https://github.com/odoo/odoo/pull/251950 : This PR tries the
solution of having a lower frame depth but matching the frames based on
a window of frames so that we can reconstruct an approximation of the
flamegraph, for example: matching window of 2 frames
1 - > 2 - > 3 - > 4
2 - > 3 - > 4 - > 5
would mean that we would match frames 2 and 3 in both stack traces and
append the first frame to the second callstack which would look like
1 - > 2 - > 3 - > 4 - > 5
Neverthless this was deemed to have too big of an assumption in the
building heuristic.

2- https://github.com/odoo/odoo/pull/253120: This PR was supposed to be
introducing memray as a profiler. Memray is the best tool for this
usecase. First because it attaches on the native system allocation
calls, and uses a file to append to on allocations. This solves both of
the issues that we had in the beginning but the issue with memray is
that it's an external tool that was deemed unnecessary to add.

The final solution is this PR:
The PR assumes a heuristic that in worker mode, a single worker handles
one thread which mean that the process memory can be fully attributed
to the request. The heuristic is also based that on a high enough
sampling rate, the delta can be fully attributed to the current frame.
This is a close enough approximation to know where to look but not what
is the actual memory usage by line.

Forward-Port-Of: odoo/odoo#270442
Forward-Port-Of: odoo/odoo#253604

A paid order can reach `sync_from_ui` more than once. In that case the order falls into the else branch of `sync_from_ui` and its payments are re-processed through `process_saved_payments`, which was not idempotent and led to two issues:

- The change/return cash payment is generated server-side in `_process_payment_lines` and has no uuid, so `_update_lines` cannot deduplicate it. Each extra sync therefore created an additional return payment. It is now removed before being recomputed, which also keeps it correct when the payments are edited after payment (new return amount, or no change at all).

- `_update_lines` replays the client commands as-is. On a second sync, a delete command (`[2, id]`) targets a payment that the first sync already removed, and `_create_pm_change_log` crashed with a MissingError while reading the deleted record. Update/delete/unlink commands referencing records that no longer exist are now skipped. Note that delete/unlink commands only carry 2 elements, so the check runs before the `len(line) < 3` guard.

Steps to reproduce:
- Pay an order, then re-sync it (or edit its payments and sync again). => the return payment was duplicated, or a MissingError was raised.

opw-6327912

---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr