Without this patch, when a product was added to cart, if this addon
was installed, it always landed in English in the SO.
This happened because the context, containing the current language,
was being aborted here (`context=None` instead of `context=context`).
This commit closes#14340
It seems Paypal does not always send the same responses on auto-return
even when PDT is off. Although not reproducible on a Paypal sandbox,
sometimes the system auto-return to /payment/paypal/dpn without any
meaningful POST data. This seems to only happen with new accounts
that use the 'Hermes' web application of Paypal.
The correct thing to do would be to add a new field on the paypal
payment provider for PDT token and make the PDT flow available
to users; but this is a stable branch and this fix is already
sufficiently delicate. This shall be done in master though.
From this revision on, users can then activate PDT on their paypal
account, set the PDT token as an ir.config_parameter value (WITH
GROUP RESTRICTION SET TO ADMIN/SETTINGS GROUP!!!) and the system
will process these requests correctly.
The revision
a1d6c2d631
has been ported with an oversight:
except_orm is no longer imported in this file,
besides, the new way to raise these warnings is to
use "UserError" instead.
opw-693779
Instead of hardcoding the IN & OUT endpoint
addresses for the ESC/POS printers, we now
attempt to auto-detect them.
This should increase compatibility with many
compatible thermal printers that are simply
using a different address, such as the
- STMicroelectronics POS58 Printer USB
- HU HAI SUNCSW Receipt Printer Co.,Ltd. Gprinter USB Printer
Fixes#12890
Before, it did a search to see if the package existed,
but the only thing it needs to do is see if the package
on the pack operation corresponds to that of the quant. (no need to check children also)
There is a test added:
A picking with 120 pieces incoming, 120 in pack 1, 80 in pack2
When we deliver those in a picking out, with product pack operations: (by taking them out of the pack)
120 from pack 1 and 80 from pack2,
we should only have 2 quants and links between moves in the end
And before, it generated 3 because it matched the wrong quants and made the wrong links.
opw 693760 closes#13836
This commit fixes the driver to respect the new device
registration mechanism via hw_proxy.rs232_devices.
It also refactors the code to more easily support multiple
scale protocols, and introduces support for the ADAM Equipment
AZExtra scale protocol. (Might be compatible with other ADAM
Equipment scales)
Support for th AZExtra scale is experimental at the moment,
especially given two annoying problems with this model:
- they do not support proper probing (stays mute until a
non-zero weight is measured), so they have to be probed
last and *assumed* to work
- the scale beeps when a read attempt is made and the
weight is not stable yet, or the weight has been already
read previously. This constant beeping during operation
is mitigated by extra delays between readings, but might
still prove to be a major issue for supporting this scale.
RS-232 drivers now need to register any device
they are handling in the hw_proxy.rs232_devices map.
This will prevent other drivers from probing them,
possibly messing up with the device.
Any update to rs232_devices must be done while holding
the hw_proxy.rs232_lock.
The hw_scale driver will be updated to use this mechanism
in the next commit, that will also handle a new RS-232
scale protocol.
- ngrok download URL has changed
- default kernel for raspbian image has changed and needs to match
- remove pre-existing git checkout to avoid conflicting
- wait a few seconds after setting up loop dev with kpartx to access image
contents - sometimes required to let the device appear in /dev
When an asset is set as Prorata Temporis, with a one month period, the
first depreciation amount is computed on the basis of the remaining days
in the purchase year. This doesn't make sense for a monthly
depreciation.
This commit makes the distinction between a monthly and a yearly
depreciation. In the case of a monthly depreciation, the remaining days
in the purchase month are taken into account.
opw-690034
Backport of 0a3899b72c
The fiscal year start date should be taken into account when calculating
the depreciation amount in case of prorata temporis.
opw-678072
Backport of e69d9b8b7d
The company currency is USD, the invoice currency is EUR.
- Create an invoice in EUR, set an invoice date
- Compute the taxes (click on "update" button)
- Change the exchange rate between EUR and USD
- Validate the invoice
At validation, the tax lines are not recomputed. Therefore, the tax
amounts, are still converted in the company currency using the old
rates. On the other hands, the other account move lines will have the
appropriate new rate.
Closes#14024
opw-692430
The company currency is USD, the invoice currency is EUR.
- Create an invoice in EUR, set an invoice date
- Compute the taxes (click on "update" button)
- Change the exchange rate between EUR and USD
- Validate the invoice
At validation, the tax amounts are not recomputed. Therefore, they are
still converted in the company currency using the old rate.
Closes#14024
opw-692430
When an asset is set as Prorata Temporis, with a one month period, the
first depreciation amount is computed on the basis of the remaining days
in the purchase year. This doesn't make sense for a monthly
depreciation.
This commit makes the distinction between a monthly and a yearly
depreciation. In the case of a monthly depreciation, the remaining days
in the purchase month are taken into account.
opw-690034
As of f814dd9908, debug
mode causes an extra X-Debug-Mode header to be sent
by the rpc() JS method.
This custom header was not whitelisted in the accepted
CORS headers, therefore any cross-origin call to a route with
`cors=True` would fail in debug mode, with a console error
along those lines:
"Request header field X-Debug-Mode is not allowed by
Access-Control-Allow-Headers in preflight response"
This would prevent loading the POS GUI in debug mode,
for example.
This commit is necessary in the 8.0 branch because
the POSBox is currently based on a 8.0 server and may
be accessed by a 9.0 POS or later, thus with the extra header.
Steps to reproduce :
1. Create a new customer
2. Create an invoice with an amount of 50.- for this new customer and validate it
3. Create a refund with an amount of 70.- for the same customer and validate it
4. Create a Customer Payment for this customer and validate it. The Invoice will be paid with 50.- of the refund. 20.- are still on residual in the refund.
5. Create a new invoice with an amount of 50.- for the same customer.
6. Create a new Customer Payment for the same Customer and validate it.
The Full Reconcile Boolean stays checked inspite of "Open Balance" of Refund amount is less than the Invoice amount.
After the fix:
Full Reconciliation Boolean doesn't get checked and Allocation amount on invoice journal item line should be same
as Open balance of the Refund Journal Item. So Invoice stays open with Balance amount.
opw:691577
Depending on the version of dateutil, rule._bynweekday can either be
a tuple or a set (see https://github.com/dateutil/dateutil/pull/54), which,
in the case of a set, breaks the access by index (see related issue:
https://github.com/dateutil/dateutil/issues/24).
By casting it into a list, we make sure that we can access [0] in both case.
Credits to jke ; closes opw-690761.
Before this patch, the debian package depends on `python-pybabel`.
According to the documentation, this is a dummy package for transition
from `python-pybabel` to `python-babel`[1].
This dummy package has thus been removed in debian stretch in favor of
`python-babel`, and the odoo package is thus not installable in debian
stretch.
To fix this, we depend directly on `python-babel`, which is available in
all debian releases[2].
Closes#13905
[1] https://packages.debian.org/jessie/python-pybabel
[2] https://packages.debian.org/jessie/python-babel
Underscore method _.each expects an array like object when a "length"
property is present.
This was an issue with a record having a numeric "length" field set to a
non-negative value. When changing a line the change would not appear on
blur.
This issue was fixed with 0e664c9e9 but introduced back with f0e331e00.
opw-691070
When returning an item by entering an orderline with a negative quantity
you got a warning message asking you to confirm a large payment
amount. Introduced by 5a3d9ce9ce.
Closes#13777
If you validate an order it will be sent to the backend. If you then
close the POS frontend somehow before clicking 'Next order' this order
will be reloaded from localStorage when reopening it. This is
problematic because it allows the user to then modify this order and
validate it again. Everything will appear fine but in the backend this
new, modified order will be ignored as it has the same pos_reference as
the one that was sent initially.
The deeper underlying issue is that there exists a brief timeframe where
the same order can exist twice in the db: both as 'order' and
'unpaid_order'. To ensure this does not happen this fix removes the
order from the unpaid orders after it is saved as a validated order.
Fixes#13538
This revert the fix made on 7bdd4de and replace it with a proper one that do create an account entry from stock_input to stock_output on a dropship move. That way, we avoid the manual change of account on supplier invoice. See the whole discussion on odoo#12687
Before this commit, the inventory by lot/pack/serial/... was only visible
if you switch quickly between res_config and adjustment inventory... once
the vaccumn clean the osv memory, the options was ignored.
Same issue with the _get_string_qty_information function.
Like in Invoices Analysis, the expenses must be expressed in the currency
of the company. In this way, it makes sens to sum them to compute the total.
opw:689760
The effect of this change is to trigger the recomputation of fields on larger
recordsets. This takes advantage of batch computations within compute methods.
Christophe Simonis