Before this patch, it was not possible to manually process a pack (i.e.
checking the "is_done" box), because the field was always reset
*before* saving. It was still possible to process it by hitting
"validate", but it was not possible to put a pack in a pack.
Manually process a pack means hitting the checkbox "is_done" on the pack
operation. This "is_done" field is a boolean computed field with a reverse
function, and this reverse function set the "qty_done" on the pack
operation to 1.0.
The compute field doesn't have any depends declared, so the value was
invalidated after any changes in the form view. The computed method was
used to check the checkbox is the "qty_done" is set to 1.0. It was
probably used to check the checkbox when opening later a picking for
which the pack was already processed.
The issue is that the "qty_done" was never updated, because the inverse
method is only called when saving, and the compute method already reset
the "is_done" field to False, as the "qty_done" was not updated.
To fix this issue, we do not use an inverse method but an onchange.
Indeed, this was the expected behavior: hitting the checkbox should
change the "qty_done".
We already have a test covering this use case, committed in v9 [1]. It
didn't discovered this regression because the test was using directly
the "qty_done" field on the pack operation, not the "is_done" field used
in the view. We change the test to use the "is_done" field and manually
trigger the onchanges.
This is due to a bad new API migration [2].
[1] rev e248420348
[2] rev d8548e0771
This record rule added in the below revision:
1950fc4216
has been removed for an unknown reason during the account refactoring
c04065abd8
Without this rule, a user can see invoices from other companies
through the invoices analysis.
opw-705718
The search on qty_available was refactored twice:
- b9853db8 to split the method
- a989680b to remove the eval operator
At b9853db8 the order of evaluation introduced a bug so that the domain
('qty_available', '=' 10.0)
was evaluated
if product[qty_available] = 10.0:
which crashed (`=` instead of `==`).
Before b9853db8, this code did not crash as the operator was converted via
if operator == '=':
operator = '=='
which is no longer executed after the split
a989680b replaced the eval by a dictionnary that replaced the eval error by a
KeyError when searching with `=` operator.
This commit changes several things:
- replace the `operator in ('==', '>=', '<=')` as this was evaluated before the
operator replacement code mentionned above
- remove the replacement code as no longer justified since a989680b
- replace the OPERATOR key `==` to always match given key
This is related to revision
52db42d3df
The total sent to the `compute` method is always
within the invoice company currency, not
in the invoice currency.
`total` comes from the first tuple result
of the method `compute_invoice_totals`,
which is always in the invoice company currency.
if the company currency is different than
the invoice currency,
the total is the sum of the `line['price']`
converted to the company currency, as stated
by
`line['price'] = currency.compute(line['price'], company_currency)`
and
`total += line['price']`
In the above revision, the possibility to pass
its own currency in the context is added. This is in case
the payment term used comes from another company
than the one of the invoice (see the commit message of the revision).
The mistake was to use the invoice currency, while it should
use the invoice company currency, as the `total` amount is
within the invoice company currency, not the invoice currency.
opw-705604
A byproduct is a produced when producing another product (the targetted
product).
A move can be linked to another move. eg: we could have a manufacturing
order of 3 units linked to a move of a delivery order of these 3 units.
If we produce 2 units, the manufacturing order is split in 2 units and
1 units, and the delivery order is split similarily because of the link
1. [T manufacturing move split] -> [T delivery move split]
(T is the targetted product, B is the byproduct)
But in 8c307d7b1 a move_dest_id of a byproduct was linked to the targetted
product delivery move, thus in some situation the move of the delivery
order of the byproduct would erroneously be split 2 times instead of one.
1. [B manufacturing move split] -> [T delivery move split]
2. [T manufacturing move split] -> [T delivery move split]
This could also be the source of other issue, and since the byproduct
and targetted product are different, the link should anyway not be done.
opw-697151
note: this change is already in 10.0 (it is inside mrp refactoring 2ddc35a53)
The field 'Type' should always be readonly since:
- it must not be changed for standard models
- its value must be 'Custom Object' for a custom model
opw-706130
When editing translations, you were not able to copy/paste texts before
this commit. Indeed, in some contexts, in some browsers, copy/pasting
text was creating a new paragraph inside the translation, preventing it
to be saved.
This commit simply adds code to remove these paragraphs each time text
is changed.
A more elegant solution will be found for master with website and
editor improvements.
When the quantity on hand is updated from the wizard, it may result in
completely inconsistent results.
This happens for example in the following case:
- 10 Units in Stock/WH
- 18 Units in Stock/WH/Shelf 1
The "New Quantity on Hand" suggests a theoretical quantity by taking
into account the location and its sub-locations. It the example, that
would mean 28 Units in location 'Stock/WH'.
However, the `_get_quants` method of the `stock.inventory.line` model
doesn't take into account the children locations. Therefore, the
theoretical quantity would be 10 Units in location 'Stock/WH'.
This inconsistency confuses the user, and the new quantity added might
introduce unexpected results.
opw-703886
Before this commit, when saving the pivot state in a favorite filter,
then restoring it, the row groupbys were ignored. This commit makes
sure that we take them into account if necessary.
Before this commit, when changing measures from the favorites menu (by
selecting a filter with some pivot_measures), the measures in the
dropdown menus were not correctly updated.
This commits make sure that we update the active measures in the menu after
each changes coming from a do_search.
In some instances we could want to access the barcode without being
connected.
For example if we print the report in an scheduled action, the report
would be rendered by wkhtmltopdf as public user making the barcode
blank (we have a redirection to a html login page instead of the image).
A solution requiring view updates in stable version was implemented with:
https://github.com/odoo/odoo/commit/14282f165 in 8.0 ( further explained
in https://github.com/odoo/odoo/issues/10621 ).
This commit fix this another way in 9.0 by making the barcode route
public since this doesn't bring any harm, solve this issue and allow the
use of the barcode route for public user (which might be wanted).
opw-694470
1. Activate the variant option on sale settings
2. Create a product and set invoicing on delivered qty and make it a
stockable product
3. Create 2 attributes, like color red and color blue
4. Odoo creates 2 variants
5. Make a BoM on the product with color red (So on Bom select the
product template and the product variant)
6. Set the Bom as "Ship this product as a set of components (kit)"
7. Now create a sale order with sale order line for the product variant
BLUE
8. Delivery order is created, deliver the whole delivery order
9. Go back to sale order and see that delivered qty is not updated and
thus we can not invoice the lines
opw-706423
The 'Start here to discover Odoo' is a data to help users to use Odoo.
Once they have understood, user often delete it but it comes back after an
update. The only way was to archive a record.
Using forcecreate is another way to avoid recreating it after an update.
Fixes#15262
Description of the issue/feature this PR addresses:
This commit adds the recompute conditional clause in the ORM unlink method.
This behaviour is already present in branch 10.0 as it can be seen in line 3431.
With this, the possibility of disabling the automatic recalculation of computed fields is given back to the developer of custom modules in performance critical scenarios. By using the Environment class "norecompute" context manager, the developer can now unlink records in a very fast manner and afterwards, execute the computed fields manually if needed (via an explicit method call or a direct sql update).
Current behavior before PR:
Having this option in the ORM will allow to speed up the system if needed when massive collateral changes must be made. One of this scenarios has to do with the modification of one2many fields and the write_uid and write_date computed fields. In the account module for example, when unreconciling account move lines, the deletion of the partial reconcile records involved in the reconciliation triggers a one by one update of the related account move lines and account moves. If the involved documents are big, such an account move (Journal Entry) with thousands of lines, if one of its lines is reconciled and afterwards unreconciled, the process can take several minutes due to expensive one by one recomputations such as the ones involved in the modification of the write_uid and the write_date in the models.py recompute() method. This modifications can however be enforced later with a couple of SQL update statements, instead of thousands of them executed by a for loop.
Desired behavior after PR is merged:
Developers of custom modules will now have the possibility of disabling the automatic calculation of computed fields, as it can be done in the create method (models.py "_create()" line 4370). This is vital for performance reasons, specially when handling big documents such as thousands of lines account moves or sales orders.
This is basically a backport of revision f259e74f8a
Complements the patch in 4715d18e12
in order to properly bootstrap a writeable data_dir when it is
(partially) nonexistant.
Depending on the startup parameters the data_dir might otherwise
have ended up read-only, preventing the creation of its necessary
components (session store, file store). Only the `addons` directory
of the data_dir needs to be read-only by default.
When validating a new order in between the create_from_ui RPC call and
the corresponding callback this new order was lost. There were two
reasons for this.
First of all, the Mutex supposed to prevent multiple create_from_ui RPC
calls from happening at the same time did not return a Deferred. This is
problematic because this function gets passed to jQuery.when() which
will consider functions that do not return a Deferred to be resolved
upon return.
Secondly, the resolved callback of the create_from_ui RPC call naively
deleted all paid orders, assuming they were all successfully sent to the
backend. This is not the case when an order is added to the orders array
in between the RPC call and the callback.
Closes#15190
Complements the patch in 15583a4813
in order to properly bootstrap a writeable data_dir when it is
(partially) nonexistant.
Depending on the startup parameters the data_dir might otherwise
have ended up read-only, preventing the creation of its necessary
components (session store, file store). Only the `addons` directory
of the data_dir needs to be read-only by default.
Modern browser implement the autofill specs as described at
https://html.spec.whatwg.org/multipage/forms.html#autofill
This means that there is now a way to tell the browser when a password
field is supposed to be filled with the user's own password (e.g. on a
login field), and when it's supposed to be used to input a new secret
value (e.g. when changing someone else's password).
This can be distinguished by setting the autocomplete attribute
to `current-password` or `new-password`.
In addition, many browsers have decided to entirely disregard the
autocomplete='off' attribute value, because it is felt that using
a secure password manager (as provided by those browsers) is a net
gain for security, allowing users to pick stron passwords.
See also https://developer.mozilla.org/en-US/docs/Web/Security/Securing_your_site/Turning_off_form_autocompletion
Based on this, the framework should generally consider that
password fields in the backend are meant to be used for *setting*
new values (autocomplete="new-password", i.e. no autocompletion!),
whereas the password field on the login page can indeed be autofilled
with the user's current password (autocomplete="current-password").
In addition to making the above changes, this patch also
lets developer specify their own `autocomplete` attribute on
text input fields.
Rev 0ff26cf7cd disabled this option
because it was causing JS errors, which is not the case anymore.
We now get the expected result:
- if product has no variants, open quick add modal dialog
- if the product has variants, open the product page to let
the user choose a variant
Also fix the bootstrap modal bug that happens when the quick add
modal is opened on a product that is not published (modal appeared
partially transparent and unusable)
Some printers (e.g. matrix/impact printers) may have a hard time
keeping up with the text output, and may trigger timeout errors
because of this, even though they would otherwise produce a correct
result.
Increasing the default timeout to 5s (from the default 1s) should
take care of most slow printers out there.
As discussed on issue #15225, it should be possible for system administrators
to disable the 1-click installation system.
The plan is to disable the feature by default, but make it relatively easy
to turn on when it is explicitly desired.
1. At the moment we cannot guarantee that all Apps published on the Odoo Apps
Store are safe. And it is a security risk to let end-users deploy Python
code on their Odoo servers without requiring any review/deployment by a
competent system administrator.
We will work on improving the validation process of the Store, but this
will require time, and won't probably be a 100% safe process in any case.
2. The one-click install feature is however really useful to help
non-technical users install Apps, as long as the feature has been
explicitly allowed by the system administrator. This is a common feature
in other software suites as well. So we'd like to keep it as an opt-in
feature.
3. Administrators of multi-tenant servers, cloud hosting services, etc.
understandably expect to be able to turn off the feature for
security/control reasons.
4. By turning off the feature by default, but still exposing it in the UI,
we keep it *discoverable* for users. The error message should be
helpful to direct users to their sysadmins.
5. By using the permissions of the download folder as a flag for turning
off the feature, we avoid introducing an extra server parameter.
The folder is still created (read-only) by default, for the sole purpose
of making it easier to locate.
Fixes#15225
As discussed on issue #15225, it should be possible for system administrators
to disable the 1-click installation system.
The plan is to disable the feature by default, but make it relatively easy
to turn on when it is explicitly desired.
1. At the moment we cannot guarantee that all Apps published on the Odoo Apps
Store are safe. And it is a security risk to let end-users deploy Python
code on their Odoo servers without requiring any review/deployment by a
competent system administrator.
We will work on improving the validation process of the Store, but this
will require time, and won't probably be a 100% safe process in any case.
2. The one-click install feature is however really useful to help
non-technical users install Apps, as long as the feature has been
explicitly allowed by the system administrator. This is a common feature
in other software suites as well. So we'd like to keep it as an opt-in
feature.
3. Administrators of multi-tenant servers, cloud hosting services, etc.
understandably expect to be able to turn off the feature for
security/control reasons.
4. By turning off the feature by default, but still exposing it in the UI,
we keep it *discoverable* for users. The error message should be
helpful to direct users to their sysadmins.
5. By using the permissions of the download folder as a flag for turning
off the feature, we avoid introducing an extra server parameter.
The folder is still created (read-only) by default, for the sole purpose
of making it easier to locate.
Fixes#15225
In some instances we could get a traceback in the form view of an
account.invoice.line object. This is because the `parent` field can be
undefined depending of the way the form view was opened.
For example there can be a account.invoice.line list in a
sale.order.line form and this lead to a product_id with a traceback
capable context.
There are some problems in the way the picking compute its state
according to its move lives if the delivery strategy is partial.
When a picking was in a partially available state and then you would
force availability, the state would stay in partial availability.
Therefore, we changed the way the state is calculated more like it was done before.
This issue stems from the new API migration.
The bank account didn't appear when selecting SEPA credit transfer in ACCOUNTING>PURCHASE>PAYMENTS
to refund an employee. The bank account is taken from partner set with the field home_address_id.
So the account holder have to be set on the bank_account_id of the employee with the home_address_id.
opw:702881
The reverse field of a one2many could be originating from an
inherits'd field, this was solved in some instance with f5e5bbda.
The issue could still happen in some instances when doing a comparison
of:
- the one2many field to a False value,
- the one2many with a negative operator and an empty set to negate,
With this change, the ORM is used in such a situation.
closes#15234
opw-704962
Simon Lejeune