Migration of all python code, mostly withtout chagnes. And adapting
xml files to guidelines.
On calendar.py, some changes have been perform to make the migration works :
- `_compute` method has been split into `_compute_dates`, `_compute_attendee` and `_compute_display_time`.
- remove unused method `onchange_partner_ids`, `check_partners_email`
- make `compute_rule_string` use a cache record instead of a value
dict to avoid crashes
- remove `new_invitation_token` method, and make it a default value on attendee creation
- add a default value on `day` field, otherwise the "Please select a
proper day of the month" exception is raised during the onchange
A string attribute value can be translated,
and therefore the inherit broken if the term is indeed
translated.
This was the case for this term "Tracked link" in the Dutch
language, the "Tracked link" term of the inherited view has
been translated, and the inherit was therefore broken as
it couldn't find the given node in the base view due to the
fact this term was translated.
opw-683128
When the schema of a view change (e.g. drop of column), the "CREATE OR REPLACE"
will not be enough and dropping the view is required.
Fixes#12754Closes#12756
When making an advance payment in percent, the function onchange_advance_payment_method
set the Down Payment Product to False. Then each time an advanced payment in percent
was made, a new down payment product was created and set in sale.config.setting in the field
deposit_product_id_setting.
Now the function onchange_advance_payment_method let the Down Payment Product
even if the payment method is in percent.
opw:683167
During module update, we should only consider views of already
loaded modules. This filter is done by searching on xmlids of views.
We should also search for explicit view ids (current module is not yet
marked as loaded)
However, by doing this `OR` condition, the generated query is not
optimal and forbid postgresql to use the composite index
`ir_model_data_model_res_id_index` on `ir_model_data` table.
This is a performance issue for databases with a lot of ir_model_data rows.
Use 2 quicker search() to avoid this problem.
This only affect the module update. Normal view loading is unchanged.
Odoo S.A. invoices only 'non-share' (internal) users; the default
filter on the user list view does the same thing. It makes little
sense for the user counter to count all share users.
Now that the debug URL parameter can be equal to "assets", the session
debug variable has to be updated to not only be true or false. Now, its
value is false if the debug parameter is not specified in the url and
a the URL parameter string value otherwise (or 1 if the value is an
empty string).
Also modify all redirection which kept the debug mode to also keep
=assets. Use the $.param.querystring function.
* switch application flag between `hr_timesheet_sheet` and `hr_timesheet`
* renamed `hr_timesheet` application name from "time tracking" to "timesheets"
* move the icon from `hr_timesheet_sheet` to `hr_timesheet`
* removed unused icon from `hr_timesheet`
* renamed `hr_timesheet_sheet` application name from "timesheets" to "timesheets / attendances"
This makes the kanban view for pack_operation_product_ids more
consistent with the one for the move lines and also adds the UOM for the
pack operation to the kanban card.
Return key events don't get redispatched and even if they did, it
wouldn't work because 'fake' keypress events don't trigger native HTML
event handlers.
This should be safe because Return cannot start a valid barcode.
We hide the quants_grp if the stock.move is in the draft state. In this
case normally the value of the string_availability_info will be empty
anyway, because the _get_string_qty_information function skips moves in
the draft state.
This is especially useful on small displays, because on there the string
attribute of the field won't be shown, so you just end up with a
Reserved heading with nothing underneath it.
- Converted yml test cases to python unittest
- now, 'self.calendar' recordset from common.py will carry forward its context
so no need to pass context explicitely to the methods calling using that recordset.
- removed commented code
- used odoo.fields.Datetime to covert dates instead of python datetime library
90 percents of the file had to be migrated because of the methods
interval_remove_leaves and get_leave_intervals that were migrated in
the module resource
* Move files and reorganize directories according to guidelines
* Lint and clean code, remove dead code
* Migrate code to the new API
Applied to stock submodules: stock_account, stock_calendar,
stock_landed_costs.
The following modules were already in new API, but cleaned and reorganized:
stock_dropshipping, stock_picking_wave, sale_stock
Two fixes from saas-12 in yaml_import have been included in this branch
because originally developped when fixing issues related to the new API
migration.
If two addons path have a common part in the folder name (e.g. `/home/alice/dev`
and `/home/alice/devodoo`), the `get_module_from_path` method may match the
wrong folder.
A file `/home/alice/devodoo/bob/models.py` would wrongly match `/home/alice/dev`
path (due to the lack of separator) and the returned module would be `odoo`
(`"odoo/bob/models.py".split('/')[0]`).
In such scenario, the translations of files (code, static folder, report) would
not be included in the exported translation file.
Force the module path to ends with a folder separator to avoid wrong matching.
Closes#12757
Jérome Maes