DROP CONSTRAINT (even with IF EXISTS is specified) acquires an ACCESS
EXCLUSIVE lock on the table, preventing e.g. inserts in an other
transaction, so ir_logging would systematically deadlock if configured
to the same database and a warning would be triggered during install
or update (if that ran ir.logging's init).
1. hand-roll the "IF EXISTS" bit, to avoid taking an ACCESS EXCLUSIVE
lock on the table if the problematic constraint does not exist and
thus doesn't need to be dropped (which by now should be the vast
majority of cases).
Replacing DROP CONSTRAINT with DISABLE TRIGGER does not fix the
issue as *that* acquires SHARE ROW EXCLUSIVE. While that's less
constraitning than ACCESS EXCLUSIVE, it still conflicts with an
insert's ROW_EXCLUSIVE.
2. add a timeout to the logging INSERT anyway, the deadlock is still
an issue if we're updating a database which does have the
problematic constraint, and we want to preclude the possible
eventual introduction of new deadlocks in the future.
closesodoo/odoo#34243
Signed-off-by: Xavier Morel (xmo) <xmo@odoo.com>
Stock.locations should be in no-update:
Limit the impact of recompute fields during the upgrade of stock module. (especially method _compute_product_availability)
Limit the impact on migration
Method _compute_product_availability is triggered during an upgrade of module and during migrations.
This may take a lot of time of big databases
closesodoo/odoo#34074
Signed-off-by: Nicolas Martinelli (nim) <nim@odoo.com>
In the Discuss app, using Chrome or IE11, try to send the same
attachment to different channels. The first attachment is correctly
detected but not the next ones.
The problem is due to a browser inconsistency, on firefox selecting the
same file in an `<input type=file>` triggers the `change` event. This is
not the case on Chrome/IE, selecting the same file doesn't triggers an
`onchange`.
opw-2006647
closesodoo/odoo#34076
Signed-off-by: Alexandre Kühn (aku) <aku@odoo.com>
Installing test_new_api without demo will fail since demo_user
does not exists. This is a problem if we want to test migration
with all modules installed. This simple fix solve the problem.
closesodoo/odoo#34065
Signed-off-by: Christophe Simonis <chs@odoo.com>
Before this task, when creating the related field from the studio, it is by
default stored. When we change the field name in 'ir.model.fields' using ORM's
write method, it also renames the column in the table associated with the
field's model.
This is the expected behavior. But when someone tries to change the 'name' of a
non stored field, the column is not available in the associated model, and so
trying to rename the column results into traceback. This issue was probably
there for a while but came into light with the task 1985710. More info can
be found on the chatter of the task.
This commit fixes the issue by adding an extra check before renaming the table
column. The query will be executed only when the field is stored and thus
exists in the table.
closesodoo/odoo#33944
Signed-off-by: Raphael Collet (rco) <rco@openerp.com>
The content of these fields can't be updated, there's no reason to
even show them. Just remove them from the default / auto-generated
view for client actions.
Task 1772242
closesodoo/odoo#29179
Signed-off-by: Xavier Morel (xmo) <xmo@odoo.com>
For each asset depreciation lines we verify that there are not linked
to a move. This verification was done for each line separately.
Moreover, during the creation of the journal entries, we were writing
on the 'line_ids'. In this case, accessing the move of a line
was invalidating the cache. This fix ensure that this operation is
done out of the loop and the cache is then not invalidated at each
step.
See opw-2000305
closesodoo/odoo#33908
Signed-off-by: Nicolas Martinelli (nim) <nim@odoo.com>
Since test_pylint was moved into the test_lint module in commit
b27077bb, this could break Odoo instances where a module has test_pylint in its dependency.
With this commit, the test_pylint module is coming back as an empty
module.
closesodoo/odoo#33871
Signed-off-by: Olivier Dony (odo) <odo@openerp.com>
Because Odoo still supports Android Kitkat and Internet Explorer, ES5 is
the highest ECMAScript allowed version.
This commit adds a new test that parses each javascript file found in
addons. The test fails if unsupported js code is found. As ES6 is
allowed during tests, the files under a path that contains 'static/test'
are just skipped.
This test uses the es-check tool
(https://www.npmjs.com/package/es-check).
If the es-check tool cannot be found, the test is simply skipped.
Also the test_pylint module is renamed to a more generic name test_lint,
that way, every linter test can find a shelter here.
closesodoo/odoo#33724
Signed-off-by: Christophe Monniez (moc) <moc@odoo.com>
In a web editor, add some text, select the text, add a link and
immediately discard the link modal. Traceback.
The traceback is a execute `setEnd` on `Range`, that error is thrown
because editable DIV doesn't save old range
Odoo wrap code does some text node split in order to attach the link
just on the selected text. If it correctly re-set the underlying
structure, it does not update the $(editable) structure used to restore
ranges.
After this commit, editable div will use newly created range for the element
for ex:
-> content: test
-> selected content "test"
when click on link dialog:
ec: test , eo: 4, sc: test, so: 0
selected content -> start 0, end 4 -> "test" like this
after clicking on discard:
ec: null text element, eo: 4, sc: null text element, so: 0
selected content -> ec and sc both are null, so it will throw an error
while setting cursor position
after applying fix when click on discard:
ec: test, eo: 4, sc: test, so: 0
selected content -> start 0, end 4 -> "test"
task - 1955863
closesodoo/odoo#33792
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
There was no way to inherit properly action_move_create method.
opw:1974069
closesodoo/odoo#33488
Signed-off-by: Nicolas Martinelli (nim) <nim@odoo.com>
The field had no triggers on the compute method.
As the field is not stored, it works fine in most cases where the value is
computed in a different transaction that the modification but when it is the
case (e.g. OCA/connector-prestashop#101), the values were incorrect
Add the correct triggers to recompute the value when needed
closesodoo/odoo#30024
Signed-off-by: Martin Trigaux (mat) <mat@odoo.com>
Backport of 610b813c87.
In this way the description of the delivery method can be used on a SO
line. If the description wasn't set, the name of the delivery method
will be used.
opw 2000169
closesodoo/odoo#33551
Signed-off-by: Nans Lefebvre (len) <len@odoo.com>
The compute method of a recursive field must be invoked on single
records even when several records are marked to recompute.
closesodoo/odoo#33529
Signed-off-by: Raphael Collet (rco) <rco@openerp.com>
When transaction is approved, if for any reason the `customerProfileId`
is missing from the response, the transaction is aborted.
It should succeed even if we cannot create a customer profile.
opw-1998505
closesodoo/odoo#33501
Signed-off-by: Nicolas Martinelli (nim) <nim@odoo.com>
Traceback generated when trying to pass a datetime object into a function tag
in xml.
<function name="action_name" model="model_name" eval="datetime.date.today"/>
Used to fail.
With this commit now one can pass time, datetime, timedelta, relativedelta,
version, ref, pytz in function tag in xml
Task-id: 1772614
Closesodoo/odoo#29212
Signed-off-by: Martin Trigaux (mat) <mat@odoo.com>
Co-authored-by: Dhaval Limbuwala <dli@odoo.com>
When multiple emails (`mail.mail`) are linked
to the same message (`mail.message`),
all the message notifications were being marked as sent
as soon as one of the mail was sent, and not all of them.
This is indeed possible to have multiple emails linked to a single
message, for instance when there are employees and customers in
the followers of a thread, and some must therefore have
the buttons meant for the employees ("Assign me") and some not.
In other words, it was possible notifications were marked
as sent while their associated email was not sent yet.
This revision also comes as an addition to
5d6656d739
as it was possible to receive a bounce coming from
the previous mail sent which wrote on the same notifications `email_status`
than the current message being sent before we could acquire the lock on
them (see commit message of the above mentioned revision),
as it was writing the `email_status` on more notifications that it
should have, including notifications of the previous email.
e.g
- Mail 1
- Message 1
- Recipient A, B, C
- Mail 2
- Message 1
- Recipient D, E
Let's say recipient A bounces
If the email notifying the bounce arrives during the processing of the "Mail 1",
it will be blocked until we release the lock on the `mail.notification` `email_status`
we acquire thanks to having written `exception` on `email_status` before sending the email.
But, as soon as it is unblocked (after the `cr.commit()`),
it will immediately re-attempt the delivery of the bounce email,
therefore writting "bounced" in `email_status` of recipient A,
while the transaction of the second email already began.
When the second email sending will try to write the `email_status`
of recipient A as well
(as it used to write the `email_status` on all the message notifications,
not just the one related to the current email),
it raised a concurrent update.
Therefore:
1. Interrupting the loop, as psycopg2 exceptions are raised and not passed,
2. Replay the sending of the second email a second time.
opw-1984293
closesodoo/odoo#33431
Signed-off-by: Denis Ledoux <beledouxdenis@users.noreply.github.com>
When website_customer is not installed, some endpoints are not handled
by the router, causing the following tests to fail:
- test_10_crawl_public
- test_20_crawl_demo
- test_30_crawl_admin
Those tests were failing due to /customers responding with 404.
Indeed, the route /customers is defined in website_customer, but had
some usage within website_crm_partner_assign, causing tests to fail when
the former module is not installed.
The new behavior can be explained as follows:
- module website_crm_partner_assign on its own will lose the hyperlink.
- module website_customer will now inherit the view definition in order
to wrap the targeted HTML text elements with an anchor tag.
Closes#33344
Signed-off-by: Christophe Simonis <chs@odoo.com>
Putting *args and **kwargs confuses the function `odoo.api.split_context`.
This causes that a call to `search` via XMLRPC gets the wrong argument: takes
the first argument of ``args`` as the context.
Example which ignores the `count=True`::
```
api.call_kw(self.env['product.product'], 'search', (
[], False, False, None, True, {'lang': 'en_US'}
), {})
```
would ignore the `count=True` and use the wrong argument for context.
Issue introduced in commit 15ea753a.
Since:
`def search(self, model=None, *args, **kwargs)`
is equivalent to:
`def search(self, args, offset=0, limit=None, order=None, count=False):`
because the call to the super will anyway call the parent method, we can
change the method arguments.
note: this is only for [10.0,saas-14] where 15ea753a is.
closes#33147
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
Co-authored-by: Nicolas Lempereur <nle-odoo@users.noreply.github.com>
Rounding each element of the sum decrease the
precision and prevent to reconcile large recordset.
closesodoo/odoo#33100
Signed-off-by: Laurent Smet <smetl@users.noreply.github.com>
With some escpos printers the cash drawer did still not open after
previous fixes. One of the problemetic printers is the Epson TM-m30.
This fix will check the status of the drawer and try to open it up to 5
times. The fix is successfully tested with the TM-m30.
closesodoo/odoo#33328
Signed-off-by: pimodoo <pimodoo@users.noreply.github.com>
This commit will correctly check for a custom payment.transaction create method
on 'self' instead of the acquirer's provider before calling it.
closesodoo/odoo#33033
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>
Following the fix in 4c35983d0, it appears that the excludes and
packages options are conflicting. py2exe seems smart enough to include
jinja2 package without the explicit include via the 'package' options.
closesodoo/odoo#32991
Signed-off-by: Christophe Monniez (moc) <moc@odoo.com>
Prefetching it triggers its translation, which quickly becomes most of
the overhead of /jobs despite the field not being displayed on that
page: an empty /jobs takes ~150ms, a /jobs with 99 jobs takes ~2000ms,
>70% of which are spent in html_translate and ultimately
translate.py:process (14600 calls, 60% of runtime, 40% own).
Removing prefetching lowers the runtime to ~700ms.
closesodoo/odoo#32985
Signed-off-by: Xavier Morel (xmo) <xmo@odoo.com>
Make a PO then an invoice with a product that will create an asset
cancel the invoice
Open the Assets Analysis report
Before this commit, the asset lines corresponding to the invoice
are shown
After this commit, they are not shown (only non archived assets are by default)
OPW 1974344
closesodoo/odoo#32955
Signed-off-by: Lucas Perais (lpe) <lpe@odoo.com>
Since 78ba90d5, when building the windows package, py2exe fails to
compiles jinja2 asyncsupport module. Even tough jinja detects the
environement, py2exe tries to compile the async part. The problem comes
from the fact that async keyword was introduced in python 3 and we use
py2exe with python 2.7.
With this commit the async part of jinja is execluded from py2exe.
closesodoo/odoo#32913
Signed-off-by: Olivier Dony (odo) <odo@openerp.com>
Christophe Simonis