Before this commit, some editor options could break the autocompletion
of form fields. This commit ensures that auto-fillable fields remain
auto-fillable when using the editor. In addition, this commit improves a
test to ensure that this bug does not reappear.
Steps to reproduce the fixed issue:
- Go to the "contact us" page or drop a form in a page
- Edit an auto-fillable field of the form ("Your Name" for example)
- Change the label position of the field
- Save
-> The field is no longer auto-fillable.
task-2715201
closesodoo/odoo#90103
X-original-commit: 455e0cc58c7fa1e02ebaa19dd2f99d829ca6e302
Signed-off-by: Quentin Smetz (qsm) <qsm@odoo.com>
This commit allows to give priority to auto-fill values and to values
coming from a data-for when in addition to one of these values this same
field has a default value. The form's test are also improved so that it
checks for this desired behavior.
task-2715201
X-original-commit: 7b23d3aacd22f87cb0c22ecd0478eba4a17b4bf3
Part-of: odoo/odoo#90103
Some form fields are dynamically filled. As the inputs that hold these
values do not always have a value property, their HTML value attribute
can be modified. Before this commit, the form option had to remove these
values before saving so that these values wouldn't become default values
(see [1]). Sadly this system is problematic and can remove default
values set by the user. This commit ensures that only values from
auto-fillable fields and from data-for fields will be removed and from
now, the suppression of these values is done at a more appropriate time
than what is done in [1]. Test is added with this commit to verify that
the default values are not erased on later form edit.
Step to reproduce the fixed issue:
- Create a form, add a (required hidden) field with a default value
- Save, then enter edit mode again
- Add a new field and save again
The default values of the form are gone. Enter edit mode again to see
the hidden input, you will see there is no value.
[1]: https://github.com/odoo/odoo/commit/b637a5e32f767b62736241042f88fa0cecf9f10b
task-2715201
X-original-commit: 043e1fdf923d2037dd8da128ab99388f0c92e544
Part-of: odoo/odoo#90103
File uploader might be deleted because it is linked to views, but the process of
upload should still happen for the related thread.
closesodoo/odoo#90102
X-original-commit: aa57c1804ecd9c88188e33c098b805f65f103a33
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
Translations of type model_terms are similar to model and the access
rights can be checked on the referenced record instead of
ir.translation
closesodoo/odoo#90101
X-original-commit: 992a919d09c3ba224a64241f4a6c16c1a086eb4b
Signed-off-by: Martin Trigaux (mat) <mat@odoo.com>
- Filter in attributes rather than filter out.
The goal is to not gather attributes which are
costly in term of performance if they are not requested
in the first place.
e.g., with all modules installed:
- Before rev:
```py
In [1]: %time for _i in range(1000): self.env["res.partner"].fields_get(attributes=['readonly', 'required', 'states', 'invisible']);self.invalidate_cache()
CPU times: user 1.99 s, sys: 9.83 ms, total: 2 s
Wall time: 2.03 s
```
- After rev:
```py
In [2]: %time for _i in range(1000): self.env["res.partner"].fields_get(attributes=['readonly', 'required', 'states', 'invisible']);self.invalidate_cache()
CPU times: user 345 ms, sys: 0 ns, total: 345 ms
Wall time: 345 ms
```
- Use the `_description_` mechanism for the attributes `name` and `type`,
so its no longer needed to treat them as exception in
`field_get` and `get_description` respectively,
and make the code shorter and cleaner.
- Move out from `fields_get` the block
```py
has_access = functools.partial(self.check_access_rights, raise_exception=False)
readonly = not (has_access('write') or has_access('create'))
...
if readonly:
description['readonly'] = True
description['states'] = {}
```
because:
- It meant you had a different behavior using `fields_get` or `get_description`
for the keys `readonly` and `states`, meaning a field could be marked as `readonly`
by `fields_get` but not by `get_description`, which is confusing.
- It is actually used in only one place, the post-process of back-end views,
which mark the fields readonly for the web client if you do not have
the create or write access to their model.
closesodoo/odoo#87273
Signed-off-by: Denis Ledoux (dle) <dle@odoo.com>
* web_editor
Prior to this commit, removing a snippet while its widget was still
active could lead to slow removal as editors would be created for its
inner content.
Stopping a widget before removing it is good practice as it will clear
the DOM from any content it may leave behind and will prevent snippet
editor to be created for its dynamic inner content.
As an example, the `Products` snippet had SnippetEditors being created
for its product cards, which could lead to slow removal especially
in later versions of Odoo. E.g. in 14.0, because ImageTools were
introduced.
Closes#85464
task-2781418
closesodoo/odoo#90100
X-original-commit: ca1359e0f40b94b93eb8cfd6834a05796f64c675
Signed-off-by: Quentin Smetz (qsm) <qsm@odoo.com>
Steps to reproduce the bug:
- Create a MO with work orders
- Confirm the MO
- Under the tab Work Orders > click on the button start multiple times quickly
Problem:
Several `mrp.workcentre.productivity` are created for the work order,
whereas normally the workorder can have only one timer started (without date_end).
Solution:
- If the workorder already has a timer running, return true to stop the process
- Add a constraint to block the user when he tries for example to create with an rpc call several `mrp.workcenter.productivity`
without an end date for the same workoder
opw-2802436
closesodoo/odoo#90094
X-original-commit: dd22195a75ec437e42424ee95ca9a41feeb48b0b
Signed-off-by: William Henrotin (whe) <whe@odoo.com>
Signed-off-by: Djamel Touati (otd) <otd@odoo.com>
In a multi-company environment, creating a product template will always
try to assign it's route to `route_warehouse0_buy` but this throws an
error if the current company is not the one of the route
Steps to reproduce:
1. Install Inventory and Purchase
2. Go to Settings > Inventory > Warehouse and enable Multi-Step Routes
3. Open Inventory and go to Configuration > Warehouse Management >
Routes
4. Open the 'Buy' route, assign it to one of the company for example 'My
Company (San Fransisco)' and save
5. Go to Products and select an other company for example 'My Company
(Chicago)'
6. Click on `CREATE`
7. An error is thrown, preventing the user from creating new products
Problem:
Function `_get_buy_route` always fetched the same route but this route's
company might have been changed so an error is thrown when trying to
access it from the other company
Solution:
Check if we have access to `route_warehouse0_buy` before using it.
opw-2805154
closesodoo/odoo#90093
X-original-commit: 1a733ae0bfd9f1cb1e90a2cab80af290560c6f12
Signed-off-by: Arnold Moyaux <arm@odoo.com>
Signed-off-by: Guillaume Merlin (megu) <megu@odoo.com>
set test env and remove username,password,token and token validity
so call never go to production server from test db
closesodoo/odoo#90084
X-original-commit: 955a83f9e773ad1bf7ce0aa4a834d9ca833cd580
Signed-off-by: William André (wan) <wan@odoo.com>
Before this commit: whenever we left the PaymentScreen when we were paying (either by reloading the page or by backing to the ProductScreen), the real payment status was not properly saved even tho the payment has been paid or cancelled.
With this commit: whenever we go to the PaymentScreen with a pending payment line with Adyen, we fetch the latest status from the back end. This way, the front end will always have the latest status of the payment.
opw-2802676
closesodoo/odoo#90083
X-original-commit: 07b30c519b101a2998f5320dde5203bfc964b7e9
Signed-off-by: Quentin Lejeune (qle) <qle@odoo.com>
Signed-off-by: Trinh Jacky (trj) <trj@odoo.com>
Purpose is to ensure tools are called like intended, notably by filtering
input and raising if an unexpected value check is asked. Some code is also
made a bit more generic to have the same kind of parameters when searching
for mail.mail.
In this commit we therefore fix some bad calls to assertSentEmails which
were not checking the right stuff.
closesodoo/odoo#90082
X-original-commit: d9759b2ee86b79b4fb2739a4aebd24a8044cf947
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>
This PR helps redducing the noise in the one introducing the new environment
in discuss. Indeed, the createWebClient helper will always be used so we won't
be able to use the form utils anymore. Moreover, since the clean up of the widget
is now handled by the start method, the calls to widget.destroy in those test was
not required.
task-2582313
closesodoo/odoo#90079
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
Before this commit, the public user would be prevented from creating,
reading, or using payment tokens entirely. This was put in place in an
effort to homogenize the way each application interacts with tokens,
and because it was considered more logical to only create tokens when
you can see actually them afterward. This however led to an undesirable
side effect in Subscriptions where customers would pay while being
logged out, thus preventing the token from being saved and failing the
automatic renewal of the subscription.
With this commit, we enable the public user to create tokens from any
payment flow. This also means that when a token is created, its owner
will not see it until they log in. This commit also reverts 2a084c48
which was intended to hide payments acquirers from the public user if
their payment would end up being tokenized.
opw-2789340
closesodoo/odoo#90061
X-original-commit: 4e03d4b6106f10522dd5f1839a0f941ca6b202fa
Related: odoo/enterprise#26741
Signed-off-by: Antoine Vandevenne (anv) <anv@odoo.com>
This commit introduces models that define records being 1:1 map with components,
as a step to move further to having essentially all business code in models.
Having code in models is desirable to have very maintainable code, thanks to
robust and declarative code with an ORM-like architecture.
Task-2831082
closesodoo/odoo#90057
Signed-off-by: Alexandre Kühn (aku) <aku@odoo.com>
For some unknown reason (part of a huge refactoring during the BS3 to
BS4 migration at [1]), a background color was applied to the <header>
(behind the navbar) once it becomes fixed (after scroll if the related
header scroll effect has been chosen).
That $light color made no sense as is visible if the header uses rounded
corners but is impossible to edit / remove.
This commit, for master, simply removes it. This is not possible to
migrate (for example to try and handle the case where the menu color is
transparent) as it would require to know the color of $light. It would
be overkill anyway, this just relies on users with that problematic
transparent header (probably nearly no one) to review their header in
the next Odoo version.
[1]: https://github.com/odoo/odoo/commit/09f40036b9c14161db90c6128861369c91c8c64e
opw-2686695
closesodoo/odoo#90058
X-original-commit: f721e6187933cde4366a7f987fb80612da581670
Signed-off-by: Quentin Smetz (qsm) <qsm@odoo.com>
This PR prepares the ground for the introduction of the wowlEnv in the discuss app.
Indeed, the local_storage service is not available as a service anymore so the one
from the browser is used instead. In order to reduce the noise in the main PR, this
change is done separately.
task-2582313
closesodoo/odoo#89851
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
Current behavior:
The invoice payment term of the report is a p tag with another p tag inside like:
<p name="payment_term">
<span>
<p> This is a payment term note</p>
</span>
</p>
but because nested p tags are not supported in html de result in the browser is something like:
<p name="payment_term">
<span></span>
</p>
<p> This is a payment term note</p>
<p></p>
Similar behavior can be observe for the fiscal position note.
Expected behavior:
The code should respect html format rules and not have nested p tags.
Solution: We replace the parent p tags by div tags, and we modify xml file that have xpath relying
on this tag.
opw-2804933
closesodoo/odoo#88660
Related: odoo/enterprise#26372
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
Before when reporting in the sale module in a multi-company
environment currencies where not converted to a unique currency
before mathematical operations (sum, avg, ...).
Now before mathematical operations currencies are converted to
the currency of the company currently selected.
Task - 2696759
closesodoo/odoo#83550
Signed-off-by: Victor Feyens (vfe) <vfe@odoo.com>
Current behavior:
When 2 steps delivery and product packing is activated, and have 2 kit products that have common components.
If you create a sale order with those 2 kits and pack them in the first step of the delivery. And in the second delivery step mark the pack as done the done quantity in the package
are not correct.
Before this fix, all the products quantities from different kits were on the same stock move line and the other lines were ignored,
this result in lines with too much products and lines with no product. When validating this
incorrect transfer an unnecessary backorder was created and the sale order lines were not correctly marked as delivered.
Steps to reproduce:
- Activate 2 steps delivery and product packing.
- Create Kit A with Component A and Component B
- Create Kit B with Component A and Component B
- Create a SO with 1 Kit A and 1 Kit B
- Confirm the SO
- Go in the first step of the delivery and put everything in a pack
- Go in the second step of the delivery and set the pack as done
- Go in the package details : The first component A has 1 reserved and 2 done, same for component B
opw-2754106
closesodoo/odoo#90033
X-original-commit: b81f70babc774eff2e04d2f7854d75ec7273a757
Signed-off-by: Engels Robin (roen) <roen@odoo.com>
Signed-off-by: William Henrotin (whe) <whe@odoo.com>
Step to reproduce:
- Create an expense for an expense product which has a cost
Current behaviour:
- Expense's unit price is modifiable which shouldn't be
the case according to the help message of hr.expense.unit_amount
Behaviour after PR:
- Expense unit price is only modifiable is there is no unit_amount
(unit_amount = 0)
opw-2781040
closesodoo/odoo#90014
X-original-commit: 44b6d60de301651e52df4c41e32e38d23b0a154e
Signed-off-by: Kevin Baptiste <kba@odoo.com>
It was impossible to create a new gift card or pay with a gift card from the front end.
With this fix, it's possible to use a gift card code in the front end and to generate a new gift card. The gift card product is now available in the pos.
closesodoo/odoo#89999
X-original-commit: 2a9b6a763b11084735cf46ac9aa83b39098b958c
Signed-off-by: Masereel Pierre <pim@odoo.com>
Since those tests are related to the legacy mockServer, they have been
moved to the legacy folder.
task-2582313
closesodoo/odoo#89883
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
This PR prepares the ground for the one introducing the new env in discuss.
The signature of the notify function has changed a lot from the one in the old env.
In order to ease the transition and reduce the noise, the discuss models now
rely on the messaging's notify function.
task-2582313
closesodoo/odoo#89877
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
*: calendar, hr, im_livechat, mail, mail_bot, snailmail, website_slides.
This PR prepares the ground for the one introducing the new env in discuss.
The signature of the rpc function has changed a lot from the one in the old env.
In order to ease the transition and reduce the noise, the discuss models now
rely on the messaging's rpc function.
task-2582313
closesodoo/odoo#89858
Related: odoo/enterprise#26676
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
Purpose:
At the moment, you can only find multi applications for the same applicant
based on the similar email.
Not every application comes through the same email.
We want to find same applicants based on the same email, phone or
mobile phone.
task - 2701472
closesodoo/odoo#86428
Signed-off-by: Kevin Baptiste <kba@odoo.com>
PURPOSE
Generic improvements for services.
SPECIFICATION
- For all reporting menus: removed the group by in the search view, and kept grouping
the grid, pivot and graph views by the corresponding field.
- For timesheet/attendance, switched the pivot and graph views from their places.
- Renamed 'billable type' into 'billing type'.
task-2766164
closesodoo/odoo#85726
Related: odoo/enterprise#24940
Signed-off-by: Laurent Stukkens (ltu) <ltu@odoo.com>
Some reports are specific to invoices and are raising an UserError if called without the correct business objects.
Since version 13.0, invoices have been merged into `account.move`.
The test is making a `search` on `account.move` to test the report generation but there is no guarantee there are invoices.
To prevent random failures in the futur, this commit allows custom domain to get the records in order to test these reports correctly.
closesodoo/odoo#85150
Task: 2726507
Signed-off-by: Quentin De Paoli <qdp@odoo.com>
For the Switzerland localization, we need to append some extra pages during the PDF generation such as QR barcodes or ISR payment slip.
There were multiple difficulties with the current codes:
- The ir.attachment was generated before the hook allowing to append the extra pages.
- There is no existing hook allowing to get the pages for each invoice when printing on multiple records at once.
This commit aims to dispatch the generation of PDFs from the creating of extra attachments:
- `_render_qweb_pdf_prepare_streams` is called first and creates a stream for each record containing the PDF pages.
- Then, the ir.attachment are generated if necessary.
- Finally, all streams are aggregated together to get the final PDF.
Task: 2726507
Part-of: odoo/odoo#85150
Globally ease event registrations followup by improving the related search
views and by allowing a quick access to statistics.
Specifications:
1. A stat button has been added on event that go to attendee reporting filtered
on the event
2. a) in the attendee reporting, the text search has been modified as follows:
- order: first search on event
- order + added: then search on responsible
- order + added: then search on organizer
- order: then search on participant
- then other field already present
2. b) still in the attendee reporting, filter "archived" has been moved as the
last filter
2. c) still in the attendee reporting, group by for campaign, medium and source
have been added
3. link to confirmed and expected attendee in the kanban view box now redirect
to the new stat view with the suitable filter
Technical remarks:
- view for a specific event launched from the event view uses a specific
search view which doesn't include search on event (name, organizer, ...).
Task 2761011
closesodoo/odoo#84546
Signed-off-by: Warnon Aurélien (awa) <awa@odoo.com>
Globally ease event registrations followup by improving the related search
views and by allowing a quick access to statistics.
Specifications:
1) A stat button has been added on event that go to attendee reporting filtered
on the event
2.a) in the attendee reporting, the text search has been modified as follows:
- order: first search on event
- order + added: then search on responsible
- order + added: then search on organizer
- order: then search on participant
- then other field already present
2.b) still in the attendee reporting, filter "archived" has been moved as the
last filter
2.c) still in the attendee reporting, group by for campaign, medium and source
have been added
3) link to confirmed and expected attendee in the kanban view box now redirect
to the new stat view with the suitable filter
Technical remarks:
- view for a specific event launched from the event view uses a specific
search view which doesn't include search on event (name, organizer, ...).
Task 2761011
Part-of: odoo/odoo#84546
The split allows to:
- have one file per model and shorter files
- ease finding of views
- ease tracking of changes
Task-2761011
Part-of: odoo/odoo#84546
# Main quest
Refactor the `load_views` API so it no longer sends multiple times the same
fields description.
e.g.
When `load_views` is called to get the kanban, tree and form views,
the list of fields of the model was sent 4 times:
- Once for each view, with only the fields used in the view,
in `['fields_views']['kanban']['fields']` for instance
- Once globally, with all the fields of the model, in `['fields']`
The goal of this revision is to change that so it sends the list of all fields
only once.
In addition, if a view contains x2many fields,
the fields description of the comodel is also sent.
It was sent in the `views` key of the view fields dict.
e.g.
When calling `load_views` of `res.partner` to get the kanban,
tree and form views,
the `res.partner` fields description was actually sent 6 times:
- Once for each view
- Once globally
- Once for each view of the many2many field `child_ids` of the form view, in
- `['fields_views']['form']['fields']['child_ids']['views']['kanban']['fields']`
- `['fields_views']['form']['fields']['child_ids']['views']['form']['fields']`
The change suggested in this revision is to:
- Remove the fields description for each view in `['fields_views']`.
As it no longer contains the fields,
the key becomes `['views']` instead of `['fields_views']`.
- Replace the dict key `['fields']` by `['models']`,
which is a dict with as key the model name and as values
the model fields description. It contains the fields description
for all models implied in the view:
the model of the main view and the model of all one2many and many2many fields.
With this change, the fields description will only be sent once by model
implied in the view.
In addition, the web client was getting the information about the fields
sometimes in the global fields description list (e.g. `['fields']`),
sometimes in the fields description list of the view type
(e.g. `['fields_views']['form']['fields']`),
making it a pain to try to make changes / performance gain
in these field description dictionaries, because you never knew in which dict
the web client was getting its info.
Now, as there is only one place to get the fields description from,
it's clearer and cleaner.
# Secondary quests
- one2many and many2many fields views are passed directly in the main view
architecture rather than being put in the `views` key
of the field description.
This is actually easier to treat by the web client,
and this will allow in a future work to cache an entire view in one block
of text rather than having to combine multiple cached blocks of text
to return one view.
- one2many and many2many fields which do not have directly embedded views
have their views directly injected in the architecture,
so the web client doesn't have to do RPC calls to `load_views`
for each one2many and many2many fields not having embedded views.
For instance, this allow to reduce the number of RPC calls to `load_views`
from 8 to 1 when loading the form of `product.product`.
Currently, this behavior is limited to 1 level deep but we consider making it
go all the way down in future works. We did not do it for the moment because
in certain cases it rises the processing time and the size (bytes) too much.
e.g. the sale.order view can be 5 levels deep,
meaning you can reach 4 dialogs on top the main view.
```
sale.order form > order_line > sale.order.line form > invoice_lines >
account.move.line form > asset_ids > account.asset form >
depreciation_move_ids > account.move form.
```
This will also benefit in future works to cache an entire view in one block
of text rather to having to combine multiple cached block of text
to get one view.
- `fields_view_get` becomes `get_view`.
As it no longer returns the fields description,
keeping the `fields` in the name `fields_view_get` no longer makes sense.
Hence removing `fields` from the method name, it becomes `view_get`.
As it gets renamed anyway, we take the opportunity to rename it `get_view`,
which is more in line with the general getter/setter guidelines
in the model object world.
- `_fields_view_get` becomes `_get_view`. For the same reasons than above.
- `load_views` becomes `get_views`.
This is not mandatory, there is no technical reason to rename `load_views` as
it practically sends the same info as before,
the view architectures and their fields description. Just in another way.
We just take the opportunity of this pull request to suggest a cleaner API:
`_get_view`, `get_view` and `get_views`.
- Arguments `toolbar=False, submenu=False` fo the methods
`_fields_view_get` and `fields_view_get` are converted to a kwargs `**options`
in `_get_view` and `get_view`.
The rationale is that submenu was already no longer used (deprecated)
and the mobile options is introduced.
The mobile options is necessary to tell the server to send the mobile views
for x2many fields (kanban instead of tree).
Instead of adding a new argument each time we add a new option to
`fields_view_get`, it seems wiser to have a kwargs `**options` to avoid
to re-write all overrides each time a new option is introduced.
- `_fields_view_get` returned a dict containing the arch in text and some of the
view information. Now, `get_view` returns a tuple with the view architecture
as an `etree` node, and the view as a browse record. The rationale is that all
overrides of `_fields_view_get` were about modifying the arch only
(e.g. changing the address format/re-organizing the address related field
nodes of the partner according to the company country).
To do so, all these overrides were doing `etree.fromstring` to parse the arch
which was sent in text to convert it to an `etree`,
then operations were done on the `etree`,
and then `etree.tostring` was called to convert back the arch to string.
With this change of signature to send the arch as an `etree`,
all these back and forth `etree.fromstring` -> `etree.tostring` are avoided,
allowing some performance gain and less code in the end.
- A cleanup of the keys returned in the dict of `fields_view_get`
has been performed in `get_view`:
- `fields` is removed, as explained above,
- `view_id` is renamed `id`,
- `name` is removed, it was unused by the web client,
- `type` is removed, it was unused by the web client,
- `field_parent` is removed, it was unused by the web client,
- `base_model` is removed, it was unused by the web client.
- `filters` is moved from the global dict returned by `load_views`
(now `get_views`) to the dict returned by `fields_view_get` (now `get_view`)
as it applies only to the `search` view type.
- Retro-compatible methods for the 3 methods
`fields_view_get`, `_fields_view_get` and `load_views` are provided,
with deprecation warnings in them.
# Future quests
- The web client could cache the model fields description
(as it already caches the views),
so it doesn't need to fetch them again if it asks for another view of a model
for which he already has the fields description.
If we do so, `get_views` could return only the list of models used by
the views, without the fields description as of now,
and the web client would then call `fields_get` independently only for
the models for which it doesn't have yet the fields description.
This would avoid the server to return the fields description
and to call `fields_get`, which is costly, for each `get_views`,
therefore gaining performances.
- Inject the views of the one2many and many2many fields all the way down,
unlimited depth level, as explained above.
- Cache with `ormcache` the architecture of back-end views.
This is already done for qweb views, it's not done for back-end views.
Therefore the postprocessing of the views is performed for each `get_views`,
which is costly, while the view architecture doesn't change for users
belonging to the same groups, according to the groups implied by the view.
### Credits
This pull request is co-authored by
Aaron Bohy (aab) for the web client part and
Denis Ledoux (dle) for the server part.
closesodoo/odoo#87522
Related: odoo/enterprise#25736
Related: odoo/upgrade#3478
Signed-off-by: Denis Ledoux (dle) <dle@odoo.com>
Before this commit, py_js was unabled to evaluate expressions
containing "true" or "false" (boolean values, JS syntax), which
seems reasonnable for a python interpreter. However in Odoo, we
often use that syntax in modifiers (e.g. required="true"). For
that reason, we need to add those two keywords in the evaluation
context.
Part-of: odoo/odoo#87522
Before this commit, the makeView test helper processed the given
arch and generated the fields to pass to the View component. This
led us to do a similar work as what is done in the view service,
i.e. duplicate code, or factorize/reuse code.
An alternative is to let the View component fetch whatever it needs
(view arch, fields, search view...). This is the purpose of this
commit. That way, the code of the view service is executed (and
thus tested as well) in each test doing a "makeView".
This commit makes that change, and adapts some tests accordingly
(mainly, tests doing assertions in mockRPC, since we do a get_view
in every test).
Part-of: odoo/odoo#87522
Before this commit, in form views, a call to "getViews" was done
for each field in the div with className "oe_chatter" (i.e.
"message_ids", "message_follower_ids" and "activity_ids"). Those
calls were unnecessary since those fields are displayed with a
custom widget (the Chatter) anyway.
Part-of: odoo/odoo#87522
Refactor the `load_views` API so it no longer sends multiple times the same
fields description.
e.g.
When `load_views` is called to get the kanban, tree and form views,
the list of fields of the model was sent 4 times:
- Once for each view, with only the fields used in the view,
in `['fields_views']['kanban']['fields']` for instance
- Once globally, with all the fields of the model, in `['fields']`
The goal of this revision is to change that so it sends the list of all fields
only once.
In addition, if a view contains x2many fields,
the fields description of the comodel is also sent.
It was sent in the `views` key of the view fields dict.
e.g.
When calling `load_views` of `res.partner` to get the kanban,
tree and form views,
the `res.partner` fields description was actually sent 6 times:
- Once for each view
- Once globally
- Once for each view of the many2many field `child_ids` of the form view, in
- `['fields_views']['form']['fields']['child_ids']['views']['kanban']['fields']`
- `['fields_views']['form']['fields']['child_ids']['views']['form']['fields']`
The change suggested in this revision is to:
- Remove the fields description for each view in `['fields_views']`.
As it no longer contains the fields,
the key becomes `['views']` instead of `['fields_views']`.
- Replace the dict key `['fields']` by `['models']`,
which is a dict with as key the model name and as values
the model fields description. It contains the fields description
for all models implied in the view:
the model of the main view and the model of all one2many and many2many fields.
With this change, the fields description will only be sent once by model
implied in the view.
In addition, the web client was getting the information about the fields
sometimes in the global fields description list (e.g. `['fields']`),
sometimes in the fields description list of the view type
(e.g. `['fields_views']['form']['fields']`),
making it a pain to try to make changes / performance gain
in these field description dictionaries, because you never knew in which dict
the web client was getting its info.
Now, as there is only one place to get the fields description from,
it's clearer and cleaner.
- one2many and many2many fields views are passed directly in the main view
architecture rather than being put in the `views` key
of the field description.
This is actually easier to treat by the web client,
and this will allow in a future work to cache an entire view in one block
of text rather than having to combine multiple cached blocks of text
to return one view.
- one2many and many2many fields which do not have directly embedded views
have their views directly injected in the architecture,
so the web client doesn't have to do RPC calls to `load_views`
for each one2many and many2many fields not having embedded views.
For instance, this allow to reduce the number of RPC calls to `load_views`
from 8 to 1 when loading the form of `product.product`.
Currently, this behavior is limited to 1 level deep but we consider making it
go all the way down in future works. We did not do it for the moment because
in certain cases it rises the processing time and the size (bytes) too much.
e.g. the sale.order view can be 5 levels deep,
meaning you can reach 4 dialogs on top the main view.
```
sale.order form > order_line > sale.order.line form > invoice_lines >
account.move.line form > asset_ids > account.asset form >
depreciation_move_ids > account.move form.
```
This will also benefit in future works to cache an entire view in one block
of text rather to having to combine multiple cached block of text
to get one view.
- `fields_view_get` becomes `get_view`.
As it no longer returns the fields description,
keeping the `fields` in the name `fields_view_get` no longer makes sense.
Hence removing `fields` from the method name, it becomes `view_get`.
As it gets renamed anyway, we take the opportunity to rename it `get_view`,
which is more in line with the general getter/setter guidelines
in the model object world.
- `_fields_view_get` becomes `_get_view`. For the same reasons than above.
- `load_views` becomes `get_views`.
This is not mandatory, there is no technical reason to rename `load_views` as
it practically sends the same info as before,
the view architectures and their fields description. Just in another way.
We just take the opportunity of this pull request to suggest a cleaner API:
`_get_view`, `get_view` and `get_views`.
- Arguments `toolbar=False, submenu=False` fo the methods
`_fields_view_get` and `fields_view_get` are converted to a kwargs `**options`
in `_get_view` and `get_view`.
The rationale is that submenu was already no longer used (deprecated)
and the mobile options is introduced.
The mobile options is necessary to tell the server to send the mobile views
for x2many fields (kanban instead of tree).
Instead of adding a new argument each time we add a new option to
`fields_view_get`, it seems wiser to have a kwargs `**options` to avoid
to re-write all overrides each time a new option is introduced.
- `_fields_view_get` returned a dict containing the arch in text and some of the
view information. Now, `get_view` returns a tuple with the view architecture
as an `etree` node, and the view as a browse record. The rationale is that all
overrides of `_fields_view_get` were about modifying the arch only
(e.g. changing the address format/re-organizing the address related field
nodes of the partner according to the company country).
To do so, all these overrides were doing `etree.fromstring` to parse the arch
which was sent in text to convert it to an `etree`,
then operations were done on the `etree`,
and then `etree.tostring` was called to convert back the arch to string.
With this change of signature to send the arch as an `etree`,
all these back and forth `etree.fromstring` -> `etree.tostring` are avoided,
allowing some performance gain and less code in the end.
- A cleanup of the keys returned in the dict of `fields_view_get`
has been performed in `get_view`:
- `fields` is removed, as explained above,
- `view_id` is renamed `id`,
- `name` is removed, it was unused by the web client,
- `type` is removed, it was unused by the web client,
- `field_parent` is removed, it was unused by the web client,
- `base_model` is removed, it was unused by the web client.
- `filters` is moved from the global dict returned by `load_views`
(now `get_views`) to the dict returned by `fields_view_get` (now `get_view`)
as it applies only to the `search` view type.
- Retro-compatible methods for the 3 methods
`fields_view_get`, `_fields_view_get` and `load_views` are provided,
with deprecation warnings in them.
- The web client could cache the model fields description
(as it already caches the views),
so it doesn't need to fetch them again if it asks for another view of a model
for which he already has the fields description.
If we do so, `get_views` could return only the list of models used by
the views, without the fields description as of now,
and the web client would then call `fields_get` independently only for
the models for which it doesn't have yet the fields description.
This would avoid the server to return the fields description
and to call `fields_get`, which is costly, for each `get_views`,
therefore gaining performances.
- Inject the views of the one2many and many2many fields all the way down,
unlimited depth level, as explained above.
- Cache with `ormcache` the architecture of back-end views.
This is already done for qweb views, it's not done for back-end views.
Therefore the postprocessing of the views is performed for each `get_views`,
which is costly, while the view architecture doesn't change for users
belonging to the same groups, according to the groups implied by the view.
This pull request is co-authored by
Aaron Bohy (aab) for the web client part and
Denis Ledoux (dle) for the server part.
Part-of: odoo/odoo#87522
PURPOSE
When opening the event page on the website, there is a service worker in place
that pre-fetches all links in order to be able to work offline.
This commit aims to disable the page tracking for those prefetch requests to
avoid polluting the statistics with "fake" page views.
SPECS
- Add a preparation commit to introduce a "X-Disable-Tracking" header
- Put the header on all requests in the event module that prefetch content
See underlying commit for details.
Task-2476513
closesodoo/odoo#86031
Signed-off-by: Warnon Aurélien (awa) <awa@odoo.com>
In 2 ways:
- by ignoring prefetched pages: indeed some page were reported inaccurately
as being viewed by the user when they were only prefetched.
- by adding the page /event in the tracked page as the page is obviously
important in the event business
Technical notes:
- The prefetched pages are ignored thanks to an header X-Disable-Tracking added
on each prefetch request in the service worker.
- Adding the page /event as a tracked page has increased the number of queries
when browsing the page /event. That's why some query count have been updated in
TestOnlineEventPerformance.
Task-2476513
Part-of: odoo/odoo#86031
In this commit we add a request header X-Disable-Tracking that enables the
client to ask the server not to track a request.
This is useful for example when the client prefetches pages to be able to work
offline. In that case, the prefetched pages must not be considered as viewed
by the user. In such scenario and without such header, it is not possible to
get an accurate page tracking.
Task-2476513
Part-of: odoo/odoo#86031