When evaluating js tests using the browser_js method, they are
considered a success when 'ok' is found in the console log and a failure
if 'error' is found instead.
This is not very robust and recently failed with this commit : 7410de111f
It adds a filter named 'Error' in a view, when the 'click_all' test
clicks on this filter, a message with the name of the filter is written
in the console, wrongly leading to a test failure.
With this commit, the browser_js method expects a log message in the
browser console with the explicit text "test successful".
On the other hand, any error in the console log will lead to a test
failure but a test can be forced to fail with the explicit error message
in the browser console "test failed".
As a bonus, the python logger should now also log browser js trace messages.
closesodoo/odoo#31158
Purpose
=======
The "publish/unpublish" widget is always shown on slide.channel/slide.slide for
a website_publisher user even if he's not allowed to manage the publishing of
the channel (see slide.channel.can_publish field).
This commit will hide the widget on both the slide.channel and the slide.slide frontend
pages if the user doesn't have the right to publish/unpublish. A generic solution
implemented directly in the publish mixin has been chosen to ease understanding
and allow easy reuse in other applications.
Commit linked to task ID 1902304 (main elearning task).
closesodoo/odoo#31187
Purpose
=======
The "publish/unpublish" widget is always shown on the frontend as long as the user
is a website_publisher and the object has a "website_published" field.
In some modules (namely website_slides) we need to handle a more refined level of access.
This commit adds a "can_publish" computed field that can be overridden to add some additional
access rules to the object and hide the widget if the user doesn't have the rights to
publish/unpublish.
When creating a pos session we try to link the session to
the first cash journal we find.
The problem is that we cannot link a journal comming from another
company this is strictly forbidden by a constraint.
This is why we should make sure that we search only for journals in the
same company as the config.
Task 1913733
Purpose of the merge is to continue the work done from PR #30988.
This completes the website profile page, gamifies the under development eLearning [1][2]
and continues integration between eLearning and the forum.
This merge contains notably:
- Complete profile page with course progression detail and eLearning certifications
- Apply Karma gains on various action on the eLearning
- Allow to set the communication type on eLearning course at creation (frontend)
- Redesign website users page
See sub commits for more details.
Merge linked to task ID 1922159 (elearning gamification / website_profile)
and task ID 1940516 (forum comments) PR #31097.
[1] see task ID 1902304 (main eLearning task) PR #29876.
[2] see task ID 1936153 (new homepage for slides) PR #30770.
closesodoo/odoo#31109
As the eLearning (website_slides) is also using website_profile users,
the All Users page is moved from website_forum to website_profile, to be
useable in both forum and eLearning.
This page displays also the number of karma point, badges and certificates
of each user.
The users are sorted by their karma points and the first 3 users are
promoted on a podium.
Use span instead of b tag in profile page
Task ID : 1922159
PR #31109
For users viewing their own profile page on the website,
a new section 'Certifications' is added to the about tab.
Thoses certifications are the survey that activated certificate.
Task ID : 1922159
PR #31109
Since 556b6454dab07cdf716d404ec0db9aa51a09b165, channel can be linked to a dedicated forum.
But the link could only be done in backend, in the slide.channel configuration
For now on, website_publishers can create a new dedicated forum when they create
a new slide.channel.
Link buttons have been added in both forum and slide.channel headers to navigate easily
between them.
Fix the relation between course and forum in a model point of view.
Task ID : 1922159
PR #31109
To gamify the new elearning, karma can be gained on various actions
- rank a slide_channel (course)
- vote for a slide
- finish a course
This commit also fix the 'completed' computation on slide.channel.partner
Task ID : 1922159
PR #31109
Two new sections are added in the profile page : Completed courses and Followed courses
Thoses sections replace the former 'Courses' section.
Use span instead of b tag in profile page
Task ID : 1922159
PR #31109
Previously we'd `append` the extensions and odoo parent directories to
sys.path, however this means if an `odoo` is already installed in a
sys.path (e.g. global installation, or setup.py develop in a venv,
...) *that* will be used as the source of Python code (for autodoc)
instead of the Python source being synchronised with the rst, which
can lead to very odd behaviour.
Prepend the doc's stuff to the PYTHONPATH instead, to ensure they get
priority and the code matching the rst gets loaded.
Make UnknownNS more capable (of giving no fucks) so the documentation
builder blows up less on weird-ass code at module toplevels.
UnknownNS should probably be split out from NSDoc though, that requires
odd workarounds in `__getitem__`.
The 'scrap button' should be visible only if:
- The picking is a done receipt
- The picking is a not done delivery
- The picking is internal
TaskID: 1930221
closesodoo/odoo#30466
_reverse_move presents a critical performance issue when
unreconcile moves with a lot of account_move_line.
The issue is that `account.move` and `account.move.line` models are triggering computed fields/methods for each new
`account.move.line` added, computing all lines again and again.
Making a factorial performance issue.
For each line the function try create cash basis entries for tax/base lines and this is combined with all the
fields compute that exists on the model so this introduce a performance problem
The issue is that `account.move` and `account.move.line` models are triggering
computed fields/methods for each new `account.move.line` created isolated,
computing all lines again and again.
e.g.
_compute_matched_percentage at
https://github.com/odoo/odoo/blob/5e8289cfeaf52a1eac64bcc975b186f0b5f9594b/addons/account/models/account_move.py#L44-L62
Then for each line modified/created this compute is executed for all lines (not just the last one).
Now by using self.env.norecompute() and self.recompute() the problem was reduced.
Unreconciling an 'account.move' with 1576 lines with a payment takes almost 4
minutes when before this changes, the conciliation process seemed not to never
end.
Fix https://github.com/odoo/odoo/issues/30934closesodoo/odoo#31014
create_tax_cash_basis_entry presents a critical performance issue when
reconciling moves with a lot of account_move_line.
The issue is that `account.move` and `account.move.line` models are triggering computed fields/methods for each new
`account.move.line` added, computing all lines again and again.
Making a factorial performance issue.
For each line the function try create cash basis entries for tax/base lines and this is combined with all the
fields compute that exists on the model so this introduce a performance problem
The issue is that `account.move` and `account.move.line` models are triggering
computed fields/methods for each new `account.move.line` created isolated,
computing all lines again and again.
e.g.
_compute_matched_percentage at
https://github.com/odoo/odoo/blob/5e8289cfeaf52a1eac64bcc975b186f0b5f9594b/addons/account/models/account_move.py#L44-L62
Then for each line created this compute is executed for all lines (not just the last created one).
Now by using self.env.norecompute() and self.recompute() the problem was reduced.
Reconciling an 'account.move' with 1576 lines with a payment takes almost 4
minutes when before this changes, the conciliation process seemed not to never
end.
Fix https://github.com/odoo/odoo/issues/30934
In some exceptional circumstances, the computation of the remaining
leaves might lead to numerical errors, e.g. a very small but non-zero
number (e.g. 10^-17)
This value will be displayed in the name of the leave type because of
the `%g` used in the string format. To avoid breaking the translation,
we round the value.
opw-1935232
closesodoo/odoo#31245
When defining an ABA routing number with leading zeros (which is
possible), the value will always be displayed without them.
This is because the field is an integer. Since the ABA routing transit
number must be a nine-digit code, we automatically fill it with leading
zeros.
opw-1940138
closesodoo/odoo#31223
- Enable multi-currency. Company currency = USD, another currency = SGD
- Update the exchange rate
- Create a purchase journal. Assign SGD to this journal
- Create a purchase tax of 7.00 %
- Enable tax rounding globally
- Create a purchase receipt with the following values
62.04 @ 7% tax
7.29 @ 7% tax
An error occurs: 'Cannot create unbalanced journal entry.'
Each line of the receipt creates its own tax line, which is therefore
rounded => the total of the receipt is computed with a 'round globally'
configuration, while the taxes are computed with a 'round per line'
configuration.
This is because `apply_taxes=True` is set in the context. This creates
the tax line by forcing the rounding. Therefore, we call the method
`_apply_taxes` explicitly in this specific case, and manually round the
taxes once they are merged.
opw-1938623
closesodoo/odoo#31194
We extract the tax creation from the AML creation in an external method,
in order to be able to call it separately.
This is necessary for the subsequent commit.
opw-1938623
Before this commit, when new messages were posted in a new
livechat, the operator received them after a delay. This
may take up to a minute before receiving the notification.
This behaviour was causing a bigger issue: those notifications
may never be received by the operator. This case happened when
the operator was receiving more recent notifications while
notifications from new livechats were pending.
The delay was caused by the operator not yet subscribed to
notifications on the newly created livechat. It is only on
the next poll of longpolling that the subscription occurred,
then followed by dispatching the notifications to the operator.
This commit fixes the issue by dispatching new messages of new
livechats directly to the operator. This is done by enforcing
the operator to make another polling when a new livechat is created,
so that he is always subscribed to those livechats. As a result,
new messages are received instantaneously.
opw-1933951
closesodoo/odoo#31191
This commit has been backported in 11.0 [1] and then reverted [2],
because it introduced a regression in web_studio, discovered at opw-1947380
As no forward-port has been done between the backport and the revert,
the forward-port of the revert also contains the patch itself, resulting
in a noop.
Actually revert 06d73eab62 in branch `12.0`.
[1] d855e3911f
[2] 1748b46718closesodoo/odoo#31637
Signed-off-by: Christophe Simonis <chs@odoo.com>
Before this commit, when printing an invoice, the taxes were grouped
by their tax group (NOT by their parent tax)
Summing the base of each tax in the process.
This was largely insufficient, as tax groups can have taxes
with different rates or computation method within them
The summing of the base of each tax for reach tax group was in this case plain wrong
This commit aims at grouping by
group > computation method > amount
Which makes more sense
OPW 1947202
OPW 1945353
closesodoo/odoo#31617
Signed-off-by: Lucas Perais (lpe) <lpe@odoo.com>
When duplicating a project, every task linked to the project is
duplicated too but the link between sub-tasks and their parent
task is not updated thus they are left linked to the task of
the original project.
opw-1937371
closesodoo/odoo#31154
When duplicating a journal, the alias is kept. Therefore, when the user
tries to change it, it updates the value on both records.
opw-1941717
closesodoo/odoo#31242
If the stage breadcrumb on the portal view is too long, it could get on
two lines but since they are righ-aligned in inversed order, the second
line (or more) will be the first stages.
With this fix, the breadcrumbs are left-aligned so on two line the
second line are the last breadcrumbs.
Also there was possibly a different order than backend since the order
is on "sequence, name, id" in backend but was just "sequence" here.
opw-1934497
closes#31237
Before this patch, cancelling a PO of 2000 lines would take a **very,
very** long time to process... so long that I terminated it before it
finished... yikes.
This was due to the call to `_do_unreserve` on every single move when
calling `_action_cancel`, this meant that n² calls to unlink where being
made (at least), which also meant n² calls to
`stock.move._recompute_state` were being made... that's a lot of
recomputes... = bad performance
The patch simply calls unlink once, which means `_recompute_state` is
called only once, saving a lot of time.
Cancel time for a PO containing 2000 lines:
- Before patch: 1h30min+ (process terminated before finishing)
- After patch: 7min
Other modules making use of stock and `_action_cancel` should also see
benefits from this patch
closesodoo/odoo#31787
Signed-off-by: Arnold Moyaux <amoyaux@users.noreply.github.com>
For every line in a RFQ, we create a stock.move upon confirmation of the
RFQ. This create, in turn, triggers a lot of recomputes that are not
needed for the creation of the next moves and thus should only be
triggered once.
This commit disables the recompute during the step that creates the
moves and then manually triggers the recompute **once**, after all the
moves have been created.
Confirmation time for a PO with 2000 lines:
- Before patch: ~1h
- After patch: ~9min
This patch is similar to the one introduced at [1], however the
previously mentioned patch uses batch create which is not available in
this version, so we manually emulate batch create's behavior.
[1] https://github.com/odoo/odoo/pull/31471
The reference fields is supposed to contain the reference number in the vendor's
system. E.g. if the vendor is using Odoo, it could be SO0744.
Therefore this reference is not supposed to change once written by the user.
However there is the _onchange_origin that uses the origin purchase order lines
to populate the reference fields on invoice_line_ids changes.
We remove entirely this behaviour for reference,
essentially reverting 052a5702c5.
Instead we put a default for origin and reference in action_view_invoice
so that these values are initially used at create.
opw 1920747
opw 1944840
closesodoo/odoo#31757
Signed-off-by: Nans Lefebvre (len) <len@odoo.com>
Let's assume a form view with a clickable statusbar, in readonly.
When the user clicks on a status, the change is saved directly.
When the save fails (e.g. access rights), the statusbar widget
is in an inconsistent state and must be reloaded.
This rev. reloads the whole record and re-render the form when
this happens.
Related to Issue: #1915391closes#29519
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Co-authored-by: Mohammed Shekha <msh@openerp.com>
As the message was set globally in the file, a _lt was needed for a lazy
translation
Fixesodoo/odoo#26288closesodoo/odoo#31761
Signed-off-by: Martin Trigaux (mat) <mat@odoo.com>
Search for orders that are not abandoned
Before this commit, the same results for abandoned and not abandoned were returned
This was because the not abandoned domain was not negated with the '!' operator that
expression.distribute_not() relies on
After this commit, the abandoned domain is negated
OPW 1942311
closesodoo/odoo#31739
Signed-off-by: Lucas Perais (lpe) <lpe@odoo.com>
Before this commit, users wasn't able to input values in some
input elements like password, email and phone in mobile.
Steps to reproduce:
- Go to user
- Select Preferences
- Select Change password
- Try to input something
=> The focus was lost and the keyboard disappeared
In fact, a hidden input always have the focus when no other fields
need it. This input is used to intercept a value from a barcode
scanner.
So now, we added some input types to avoid to set the focus on the
barcode input.
It's quite difficult to test this feature because we aren't allow
to simultate a keyboard to write in a input field.
See https://developer.mozilla.org/en-US/docs/Web/API/KeyboardEvent
Plus, we don't have the focus in mobile except if we have the cursor
inside the window.
So, we can't test focused elements or input values.
opw-1936712
opw-1948266
closesodoo/odoo#31566
Signed-off-by: Pierre Paridans <pparidans@users.noreply.github.com>
On /shop, we access the image field to generate the image widget, but we don't
actually need the image content, we just need its url. When `bin_size` is added
we don't compute all the resize for nothing.
opw-1923999
task-1922055
PR: #30274
Signed-off-by: Jérémy Kersten (jke) <jke@openerp.com>
The `_get_combination_info` override in website_sale_stock is only necessary
when we want to get the stock, which is on the product page RPC.
Before this commit, the stock would be computed all the time, which slowed down
the requests for no reason.
opw-1923999
task-1922055
PR: #30274
In order to increase performance, it was decided to avoid computing the first
combination on the /shop page, because it has to be done for many products in a
loop. Instead, we only display the template data, which is faster to get.
This implies that when using only one variant, we need to always write on the
template image instead of the variant image. This is needed because when there
is only one variant, the user doesn't know there is a difference between
template and variant, he expects both images to be the same.
To achieve this, we also had to change how the quick buttons work on the /shop
page ("add to cart", "add to wishlist" and "add to compare"). Indeed it was
decided that their behavior should be the same, no matter how many combinations
existed or not on the template. Therefore, the standard behavior of those
buttons is to always do their action on the first possible variant, that is
cached so we can use this logic without ever needing to read anything else than
the template data (and the cache).
opw-1923999
task-1922055
PR: #30274
Christophe Monniez