The balance start should only be set with the ending balance of the
previous session when journal type is 'cash'. When the journal type
is 'bank', it must be 0.
Fine tuning of this commit: 46d0b461cd
opw:1981984
closesodoo/odoo#34735
Signed-off-by: Simon Goffin (sig) <sig@openerp.com>
Same reason as: 4ba4d61903
Some messages from payment are displayed on the frontend (eg. when the
transaction is being done on the ecommerce), but only translations of
module matching a given domain (computed with the method of ir.http
model: _get_translation_frontend_modules_domain) are made available on
the frontend.
With this changeset, payment is added in the domain (which can be done
since payment depends on account, account on portal, and portal in
http_routing that declares the method).
opw-2033528
closes#34712
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
This commit fixes multiple issues regarding the /shop/address and /my/account
forms where both are displaying `vat` and `company_name` fields.
1. For a company's child partner, the `company_name` was empty instead of
showing its company name (the name of its parent_id which is the commercial
entity).
2. It was possible for a child partner to edit its `vat` (which is a field
synced from parent to children, but only parent should be able to edit it as
it is the commercial entity, see 4a1bc45934 and 98b1fffd4e).
3. Company name now share the exact same condition than the vat field in order
to be editable (as it should have been). It was possible to edit vat but not
company name on /my/account.
Steps to reproduce the bug:
- Create a company contact CO with a VAT number
- Create a contact C in CO with portal access
- Log with C and go to the shop
- In the checkout, try to edit the address details
opw-2007368
closesodoo/odoo#34564
Signed-off-by: Jérémy Kersten (jke) <jke@openerp.com>
When an email is sent and is related to a record, there is a header with a
button link to access the record.
Before this commit, that button href was always linking to the base_url URL.
That was an issue on multi-website, since we need to use the website url and
not the base_url if the record is related to a website.
That's the case for a SO eg, after checkout you receive a confirmation email.
In a multi-website environment, we need to access the record website.
To do so, a previous imp/fix was done with c1be1993f9.
But this occurence still needed to be replaced.
Fixes#34309closesodoo/odoo#34397
Signed-off-by: Jérémy Kersten (jke) <jke@openerp.com>
Future runbot improvement may share sources between build,
meaning that sources will be readonly to avoid any interaction.
_touch() was supposed to handle ro filesystem but
f7130556 introduced a new test that was not working
in this case.
Instead of trying to touch file and pseudomocking the result
if the filesystem is readonly, this commit add a real patch
on getmtime.
closesodoo/odoo#34710
Signed-off-by: Martin Trigaux (mat) <mat@odoo.com>
Follow up to:
- commit 4cdcab43ee
- commit cb4afb263b
- commit fb1839fa17
In order to be able to edit a done raw move after the production
and having a correct cost report with consistent values. The
simple rule that a done move should have its quantity equals to
its initial demand. Since report is based on initial demand it
should follow this rule and it's also a good practice because a
lot of computation are based on it (easier than quantity done
because it's already set on the correct quant uom).
opw-2033259
closesodoo/odoo#34695
Signed-off-by: Arnold Moyaux <amoyaux@users.noreply.github.com>
Steps to reproduce the bug:
- Create an invoice I
- Set the invoice date in the past and add payment terms
- Validate I
- Create refund from I
Bug:
The invoice date is set as today but the due date was set in the past.
opw:2033438
closesodoo/odoo#34692
Signed-off-by: Simon Goffin (sig) <sig@openerp.com>
If a test is failing with assertRecordValues, the comparison and the error
logging were not computed the same way.
The _compare_candidate was accessing the record via
record[field_name]
which uses the cache
While the _format_message was doing
records.read(...)
which queries directly the database.
If the test was failing (correctly) because of a wrong cache invalidation
(e.g. update the value directly in SQL without invalidating the cache),
the error message would look like:
AssertionError: Record doesn't match expected values at index 0.
Current values:
{'id': 42, 'name': 'Foo'}
Expected values:
{'name': 'Foo'}
closesodoo/odoo#34589
Signed-off-by: Martin Trigaux (mat) <mat@odoo.com>
Go to the database manager, configure a password either via the
interface either via the `admin_passwd` `.odoorc` config file. Click on
the backup menu, let the `password` field empty and submit the form. The
modal is closed without any warning and no query is sent.
The problem is that even if the field is marked as `required`, there is
a event listener that catch the `onsubmit` event and close the modal
even if it is not valid.
opw-2031461
closesodoo/odoo#34669
Signed-off-by: Julien Castiaux <Julien00859@users.noreply.github.com>
Setup a multi-company account and multiple website (one for each
company), create a mail campaign from a different company than the
superuser, send the emails and unsubscribe, the company logo is the logo
of the company the superuser is in instead of the logo of the company
sending the email.
opw-2026528
closesodoo/odoo#34701
Signed-off-by: Nans Lefebvre (len) <len@odoo.com>
The patch fa492d87f4 has been backported
from Odoo 12.0, that runs on Python 3.
The string '\n\n({} {}, {} {})' to be formatted is a byte-string in
Python 2, while the return value of _() is always a unicode-string.
As format() is (too?) nice, it attempts to convert the unicode-strings
into ascii in order to inject them in the format pattern.
With some languages that are written in ascii, this works -- by chance.
When you use non-ascii languages like Japanese, it fails.
We then fix that issue by using unicode-strings in the formatting
pattern.
#OneCharacterPatch B-)
opw-2032016
-----------------------------
For full technical understanding:
Python 2.7.16 (default, Mar 11 2019, 18:59:25)
[GCC 8.2.1 20181127] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> '{}'.format('test')
'test'
>>> '{}'.format(u'test')
'test'
>>> '{}'.format(u'エ')
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
UnicodeEncodeError: 'ascii' codec can't encode character u'\u30a8' in position 0: ordinal not in range(128)
>>> u'{}'.format(u'エ')
u'\u30a8'
closesodoo/odoo#34698
Signed-off-by: Richard Mathot (rim) <rim@openerp.com>
Steps to reproduce the bug:
- Create two companies C1 and C2 where C2 is the child of C1
- Create two purchase taxes T1 and T2 where T1 is in C1 and C2 is in T2
- Create a prodcuct P with T1 and T2 as supplier taxes
- Be in C1 as current company
- Create a RFQ and add P
Bug:
T1 and T2 were set on the order line of P instead of T1
PS: This fix is insired from product_id_change in model sale.order.line
opw:2032113
closesodoo/odoo#34658
Signed-off-by: Simon Goffin (sig) <sig@openerp.com>
Have a user U that is Project User, Timesheets User, Employeed Officer.
Give him access to a private task T (by adding U as a follower of T)
on private project P.
Because of Employeed Officer rights, view_task_form2_inherited is rendered.
This shows field is_project_map_empty, which computation depends on project_id.
If U does not have access to P, then this triggers an access error, so the Task
cannot be displayed.
It is in general a legitimate configuration to allow U to interact with T
without being given full access to P, since it works in all cases but the rights
described above.
Since the computation of is_project_map_empty is the only blocking point,
we put it in sudo.
opw 2031124
closesodoo/odoo#34676
Signed-off-by: Nans Lefebvre (len) <len@odoo.com>
In multicompany mode,
/!\ Adress book sharing should be deactivated
User A in company A creates a vendor bill without a partner
User B in company observes the list. The new bill is present
User A changes companies and goes to company B
User B doesn't have access to company B
User B reloads the list
Before this commit, there was an access right error on User B side because the partner associated
with User A changed company, and is now unreadable from User B perspective
After this commit, there is no crash and we have the string: Created By User A in the list
in place of the vendor display name
OPW 2028451
closesodoo/odoo#34668
Signed-off-by: Lucas Perais (lpe) <lpe@odoo.com>
Before this patch, this method was altering the signature of its parent method, moving the position of `failure_type` to 1 and making the other 2 arguments kw-only.
It seems this accidentally didn't break anything because all calls happened to be done in kwarg mode. However, it's very possible that a downstream module that is not based on `mass_mailing` and makes positional calls gets broken when `mass_mailing` is installed.
The fix is to respect original method signature.
closesodoo/odoo#34648
Signed-off-by: Xavier Dollé (xdo) <xdo@odoo.com>
The callback will usually check the transaction's state during
its execution, hence it should be executed after the state change
closesodoo/odoo#34666
Signed-off-by: Damien Bouvy (dbo) <dbo@odoo.com>
Move inside a node so the term can be exported (t-esc value are not
translatable)
opw-2031550
closesodoo/odoo#34663
Signed-off-by: Martin Trigaux (mat) <mat@odoo.com>
Before this commit, when a transaction has as status 'authorized', the
transactions' confirm page gives an error : 'Oops! There was a problem
with your payment.'
Now, the page gives the same message as when the transaction has the
status 'done' : 'Your payment was successful! It may take some time to
be validated on our end.'
opw-1984325
closesodoo/odoo#34652
Signed-off-by: Damien Bouvy (dbo) <dbo@odoo.com>
[FIX] mrp: Manufacturing cost analysis report
Steps to reproduce the bug:
- Create a finished product P with a manufacturing BOM B
- Set two components on B C1 and C2
- Create a MO with P and process it(plan and produce it)
- Unlock the MO and set the consumed qty of C1 to 0
- Lock and click on Mark as Done
- Click on Cost Analysis
Bug:
C1 was displayed in the report with a qty = 1.0
Fine tuning of this commit: cb4afb263b
opw:2010912
closesodoo/odoo#34642
Signed-off-by: Simon Goffin (sig) <sig@openerp.com>
Before this commit, on a device with a "pen" (e.g. Surface Pro, ...),
it was not possible to sign a document on some browsers recently
updated (e.g. Chrome, Firefox, ...) only Internet Explorer was working.
Instead of drawing continuous lines only few dots appeared.
For Chrome and Firefox we have to catch events with "addEventListener".
For Edge we add touch-action property to convert touch events into click.
Steps to reproduce:
1) Go to sales app
2) Then go to quotations (menu Sales)
3) Create a new quotation
4) Click on preview
5) Click on the "Accept" button at the end of the preview
6) Use the "pen" to draw the signature (BUG)
We use a custom version of the jSignature lib (c.f. odoo/odoo@eddcb46),
so we have to patch the file in the Odoo repo with a mix of some
pull request found in the official repo and forks.
Link:
https://github.com/brinley/jSignature/pull/109https://github.com/brinley/jSignature/pull/159https://github.com/willowsystems/jSignature/pull/96
opw-2029684
Set an automatic reconciliation model with a tax
Create a bank statement on which one line will be caught by that model
Click on reconcile to get to the reconciliation widget
Observe the account move lines creates
Before this commit, the move line representing the tax amount
(and in the account for taxes) did not have a tax_line_id, which is
a reference to the tax that made the line exist
This messed up tax reports
After this commit, the move line of the tax has a reference
to the tax it originated from
The tax reports are correct
OPW 2006826
closesodoo/odoo#34627
Signed-off-by: Lucas Perais (lpe) <lpe@odoo.com>
On a small display, go to accounting > journal entries > create
In the (now kanban) list of account move lines, click ADD
Then, try to choose an account
Before this commit:
The list of accounts was empty.
This was because the form view spawned was the automatic generic one
The generated view did not have thr necessary structure to get fields' value
from the parent form view.
In this case, the domain of the name_search for account.account was wrong
After this commit:
The flow works on mobile as on desktop. The form view's arch is a 1 to 1
copy of the list view in terms of fields and their definition
OPW 2030837
closesodoo/odoo#34601
Signed-off-by: Lucas Perais (lpe) <lpe@odoo.com>
Backport somewhat improved version of this fix (we don't need to check
for the existing extension since we're not doing anything if there is
an extension at all) merged into later branches, to avoid forward-port
conflicts.
closesodoo/odoo#34509
Signed-off-by: Xavier Morel (xmo) <xmo@odoo.com>
Before 8557bcf the URLs were never rewritten for multilang, which was an
error and could duplicated the number of requests on frontend.
Since 8557bcf the URLs are rewritten. In Odoo the werkzeug library is
used to parse the URLs (this is similar to python parsing URL but works
in python2 and python3 the same), and in this particular instance it
chooses that tel:800800 is website http://tel:800800 and not a telephone
number.
The problem is when the url uses the uri 'tel:' and only numbers as a
phone number, if the user uses the global format for telephone number
(with the +) or using a separator for the numbers (e.g. '-' , '/'), in
these cases there are no problems parsing the url (phone number RFC:
https://tools.ietf.org/html/rfc3966).
This commit changes the uri 'tel:' to 'tel://' this one is correctly
parsed by the werkzeug library.
opw-2029844
closes#34306closesodoo/odoo#34586
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
Have a product with a fixed amount tax.
Make a "return" of that product in the pos (i.e. negative quantity)
Before this commit:
- The amount of the return was `tax_amount + product_amount` instead of `-1 * positive_total_amount`
- The amount differed from what can be observed in sales
This was because the sign of the quantity was applied twice
After this commit,:
- the amounts of positive and negative amounts are symmetrical
- they match the behavior in sale
The logic is very similar to bb72dea98d
OPW 2026278
closesodoo/odoo#34548
Signed-off-by: Lucas Perais (lpe) <lpe@odoo.com>
Incorporate Deivis Laya (deivislaya) as Vauxoo's contributor
I confirm I have signed the CLA and read the PR guidelines at
www.odoo.com/submit-pr
closesodoo/odoo#34569
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
smtplib.SMTP API calls such as .sendmail() or .login() make sure on
their first line that this command is sent, by calling
.ehlo_or_helo_if_needed().
However, test_smtp_connection() does not use SMTP.sendmail() to simulate
sending email, or the email would really be sent.
Instead it uses the low-level API SMTP.mail() which does not make sure
that HELO was sent.
The connection test could be fixed by ensuring that this command was
sent to the server in the test method. However, this could lead to
inconsistent cases where the connection test passes whereas another
usage in the code fails because the command was not sent.
For example, someone could use the low-level API for some reason.
EHLO could have already been sent because if authentication is enabled
smtplib.SMTP.login() calls ehlo_or_helo_if_needed(). Therefore it is
correct to call it ourselves at the end of our connect() method.
STARTTLS sends a first EHLO, negociates the encryption, and leaves the
state without the second EHLO, which should still be sent to the server,
as stated by RFC 3207, in section "4.2 Result of the STARTTLS Command".
https://www.ietf.org/rfc/rfc3207.txtclosesodoo/odoo#34550
Signed-off-by: Julien Legros (jle) <jle@odoo.com>
Fix/improve commit d870749d79
that was removing line feed (\n).
In this one, we also remove carriage return (\r)
to address all cases.
closesodoo/odoo#34536
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
In sale_expense, there is a rules allowing employee to see all confirmed sales
order. This rule breaks the one from sale "see own document" for salesman user.
The initial goal of this rule is explained in Task-29303: we wanted to ease the
reinvocie flow. For an employee, is it diffuclt to know on which analytic account
the expense should be reinvoice (no knowledge of that, no "analytic rights", ...).
Anyway, this rule brings more problem than expected and prevent people to work.
So, we decided to functionnaly revert the feature by desactivate the rule and hide
the sales order field on expense for user that are not salesperson. Salesperson can
set the SO, and the onchange will set the analytic account. Futher work will be
done in master (for 13.0) in order to solve that matter.
To apply the fix, the rule should be desactivated, and the module sale_expense can
be updated.
opw-2027005
Coming from Task-29303
closesodoo/odoo#34514
Signed-off-by: Jérome Maes (jem) <jem@openerp.com>
Valerian Dorcy