If a user A deletes the res.users record of user B while B is connected,
the verification of the session token fails with a comparison of a boolean and
bytes values.
While the check should obviously fail, this patch gracefully inform the user B
its session has expired and redirect him to the login page.
Without the patch, the session is never invalidated in the user browser,
redirecting to a forbidden error page as long as the session has not been manually
cleared from the browser.
Fixes#25530Closes#25654Closes#25682
Cherry-Pick of 96f01c08f8
This reverts commit ef444da57a.
Changing a function signature is not supposed to happen in stable;
we have already received 3 opw's about broken customizations or
modules that extend the stock because code such as:
res = self.do_next_transfer()
if not res:
<bla bla>
stops without any warning.
As `pre-` migration scripts may use the registry, we must ensure that
all fields are set up before execution in order to have a consistent
registry.
This is required when loading a registry which contains modules to
install/upgrade without `-u` flag. In this case, the setup was only done
*after* module loading.
Backport following opw-1851612
[FIX] google_calendar: do not create an event with an invalid id
The id is useful to update existing events but sometimes we are getting some
ids that are not accepted by Google
Getting an error:
odoo.addons.google_account.models.google_service: Bad google request : {
"error": {
"errors": [
{
"domain": "global",
"reason": "invalid",
"message": "Invalid resource id value."
}
],
"code": 400,
"message": "Invalid resource id value."
}
}
Looks like existing events can have a _ in their id but new one, no longer.
It seems that these events are created by outlook calendar when synchronized
with Google Calendar.
- Store a token inside sessions to allow implicit session deactivation when needed.
backport of @da1f153d61d747d9357694382fe04f96c0ca886a @c8243e71c6da37547a19f61c58f25d5d03e13d38
In this commit, our hero backport c444b5a293 to 9.0
[FIX] google_account: fix google request exception management
Error thrown by google request is an urllib2.HTTPError that can be read
and loaded in JSON. However in some cases the result of the read may
be void or not JSON-ready. This was causing a crash in the error
management and hid the actual issue.
This commit tries to read and JSON-load the error but fall back on
simply displaying the raw error in case of issue when handling it.
opw-1851612
When returning None, the XML-RPC can trigger an error, making
the api unusable in certain cases.
So, we added return True and if the context is None we
use an empty dict, so if the context is returned in a dict,
it is not returning None either. Tests were adapted too.
Closes#22264
Steps to reproduce the bug:
- Create a recurring meeting, with a start date with time (not all day) (for example 09:00),
a duration (for example 5 hours) , each week for example on fridays, for 3 occurences.
-Save
Bug:
- The start_datetime ("Starting at") was increased with the duration of the meeting.
PS: The displayed start and stop in calendar view were computed in function "calendar_id2real_id"
with the virtual id.
opw:1858154
When removing Odoo Debian package, the directory /var/lib/odoo is also
removed. This directory could contain important data like filestore or
custom modules.
With this commit, this directory is preserved on removal and deleted
when the purge command is issued with a Debian package manager.
Fixes#22138
If we applied a link eg. on:
```
<span>hello <b>world</b></span>
```
The system actually gets the "label": hello worldworld because there is
3 nodes:
text node: hello
element node: `<b>world</b>`
text node: world
Also since "hello worldworld" is different than "hello world",
instead of just keeping existing nodes and adding the link, the system
would replace the selected range by:
`<span><a>hello worldworld</a></span>'
instead of:
`<span><a>hello </a><b><a>world</a></b></span>`
This commit ignores element nodes when creating a new link, since when
getting the label of the link from the selection, only the text nodes
insides the element nodes have any interest.
There was a second issue because if we had:
```
<i><a href="hello">world</a></i>!
```
and tried to put a link over "world!", the code would decide: "world" is
inside a link so we will just update that link.
Thus we would get:
```
<i><a href="hello">world!</a></i>!
```
instead of:
```
<i><a href="hello">world</a></i><a href="hello">!</a>
```
opw-1848351
closes#25187
Before this rev., if you went to Sales > Sales order, switched to
calendar view, edited the arch such that 'color="state"' is
replaced by 'color="invoice_status"' on the root node, it crashed.
The reason of the crash is that a possible value of the
'invoice_status' selection field is 'to invoice' (it contains a
whitespace), and the JQuery selector didn't wrap the value by
quotes.
OPW~1856305
- When receiving a S2S payment feedback from Ogone server, we call the method '_ogone_form_get_tx_from_data' which does a "pre-process" of the data to retrieve the payment.transaction linked to this payment.
It also checks the hash signature of the data to be sure it comes from Ogone.
Right after, we call "_ogone_s2s_validate" which make a HTTP call to ogone, to retrieve the data related to the transaction which were already sent by Ogone (maybe to be sure the data comes from Ogone?).
So instead of calling "_ogone_s2s_validate" we now call "_ogone_s2s_validate_tree" which processes the data from Ogone.
We are sure they come from Ogone, since they passed the hash signature when calling "_ogone_form_get_tx_from_data".
When updating the qty on a confirmed sale order line,
we want the system to add new procurement qty
for the cancelled procurements. That way you can
leave the cancelled procurements for what they are.
Fixes#10381 (wrong calculation on datetime)
if worked hours > 24h, calculation is wrong because "seconds" funtion does not take into account days
See: https://docs.python.org/2.4/lib/datetime-timedelta.html
total_seconds() should be used instead of seconds function
backport of 682ae1cc64
Before this commit, if you enable website_form_enable_metadata, that
will crash with a "KeyError: 'meta'"
This commit closes#24888
If you don't define an explicit address type, then the
record is not going to be accessible.
In the ORM:
>>> env['res.partner'].create({'name': 'Test', 'type': False})
res.partner(44356,)
>>> env['res.partner'].search([('type', '!=', 'private'), ('id', '=', 44356)])
res.partner()
>>> env['res.partner'].search([('type', '=', False), ('id', '=', 44356)])
res.partner(44356,)
due to the fact that NULL in the database is undefined and can't be
compared to a specific key.
- When merging partners with fields restricted to certain groups which the user doesn't belong, an access error is raised.
To avoid this, we only merge fields that are accessible by the user. We do so by using the method fields_get() that only returns the fields accessible by the current user.
Currently when connecting to a smtp server, if no response is returned
and no error is raised, the `SMTP()` instantiation method never ends.
This PR is inspired by 54a477d797 related to `fetchmail()`.
Possibly, a better way exists but it has to change some methods
signatures.
Closes#24877
opw-1851030
Before this commit:
* Module A defines a field X of model M
* Module B inherits from model M without touching field X
* Module C inherits from model M and extends field X by giving an INDEX
/ NOT NULL constraint.
* Module B and C depend from Module A, but not each other
If all three modules are installed and Module B is updated, the INDEX /
NOT NULL constraint could be dropped.
This happens because Module B can be loaded before Module C is loaded,
if that's the case, then after the upgrade of Module B, during the
schema checking, we verify that the field object we have and the field
on the DB are the same, since Module B doesn't introduce the index then
this check is false and we drop the index. When we get to loading Module
C, we do not do any schema checking because the module is not marked as
`to upgrade`, therefore the index is lost forever.
To solve this, we re-init the models that belong to the set of the intersection
between upgraded and modified models and loaded and modified models.
Fixes#24958
Steps to reproduce:
- Enable multi currencies
- Make sure that the current rate of a foreign currency (i.e. $) is not 1.0 (i.e. 0.5)
- Create a purchasable product with a cost (i.e. 100€)
- Create a vendor bill in a foreign currency (i.e. $)
- Add the product on the invoice
Bug:
The unit price was equal to 25$ instead of 50$
Technical reason:
The function "_onchange_product_id" was called twice (the second time by function
"_onchange_uom_id") and the price unit was converted twice in the currency of the vendor bill.
The idea of this fix is to only convert in the currency when the unit price is set
by the system.
Fixes#24751
opw:1849212
If the `web.base.url` contains a trailing `/`, the replacement of the
`/unsubscribe_from_list` link won't work since the string to replace
will be `my_url//unsubscribe_from_list` instead of
`my_url/unsubscribe_from_list`.
Fixes#24731
opw-1848572
In 11.0, this change e9454e79 solved the use case of:
- opening the registration of a ticket
- discard
=> the page must be reloaded to register a ticket
A new report is that since 9.0, if we try to register 0 ticket we would
also have to reload the page.
This commit backports e9454e79 and solves the 0 ticket registration.
opw-1851622
closes#24966
Commit https://github.com/odoo/odoo/commit/2eb344f23b3a9daa8e7c7ddaead145a8b05b39bf changed the dependancies of l10n_fr_certification which is not acceptable on stable. Instead, the method to check is now moved in account module (to avoid duplicated) and it is called by l10n_fr_certification and account_lock module.
Module account_lock has been introduced by:
https://github.com/odoo/odoo/commit/2eb344f23b3a9daa8e7c7ddaead145a8b05b39bf
A new constrains appears on the lock dates: their must not be set
after the last day of the previous month.
Then, it breaks the test on closed period that set the lock date 'yesterday'.
Have a XMLReceipt with the line:
<barcode encoding="CODE39">123456789</barcode>
Print the receipt.
Before this commit, jibbrish characters were printed and also kinda 'broke'
the spacing between commands
e.g. If you add an EAN13 barcode below the code39 it would have failed to print correctly too
After this commit, everything prints correctly
OPW 1849284
ref: https://reference.epson-biz.com/modules/ref_escpos/index.php?content_id=128closes#24965
When a t-field element was in an editable t-ignore environement,
modifying it was leaving the edit mode style attached to it. This
was because of:
1) When the t-field element was changed, it was marked dirty but
also its parent editable container. Fixing this, only solves
the case where only the t-field (and not one of its neighbors)
is changed but it was worth fixing anyway.
2) Before saving an element, the potential 'o_editable' and
summernote classes were not removed of its descendant and were
thus saved.
Bug found with task-38069, merged in stable as it might occur there
too.
product.product inheritS from product.template, and they both
define the 'standard_price' field, but implement it differently;
- product: the field is a company dependent one (so non stored)
- template: the field is a computed one based on tis variants
For the first case, since the field is not stored in database, when
doing SQL query, we have to get the value from the table ir_property.
That is what purchase report does, but instead of searching on resource
'product.product', it does it on 'product.template'. There are
obviously no entries in ir_property table for 'standard_price' field
on product template. As consequence, the "product value" (cost)
is always null in purchase reporting.
This commit fixes that by modifying SQL query to get the good
value from ir_property table.
Martin Trigaux