Before this commit, `getTZOffset` and `user_has_group` weren't
correctly reset to their initial values after tests. As a
consequence tests executed afterwards might behave unexpectedly.
closesodoo/odoo#69497
X-original-commit: b7b83db88c61d7db87045d87b09d7348f3a8a9eb
Signed-off-by: Géry Debongnie (ged) <ged@openerp.com>
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
A lot of tests in different QUnit modules need this delay to be
patched to 0. Some of them (e.g. the 'ActionManager > Misc' module)
didn't patch it correctly. It worked before because a patch was
done in a previous test, and wasn't unpatch. The previous commit
ensures the patch is removed, so now we have at least 2 tests of
the above mentionned module that fail. This commit ensures that
the delay is patched in every tests, and allows to specify a
custom delay if necessary.
The Many2ManyCheckboxes widget displays all values that could be
in the many2many relation, with a checkbox indicating whether each
value is in the relation or not. It is designed to be set on fields
where the comodel contains a few records (typically, we don't want
to see dozens of checkboxes in the form view). This widget shouldn't
be used on many2manys with a large comodel, as we have better tools
to handle them (like a tree view).
We deal with extreme cases (when the widget is, by mistake, set on
a field where the comodel is huge) by using the name_search limit
of 100: at most 100 checkboxes are displayed.
Before this commit, this extreme situation wasn't correctly handled.
If there were in the relation records that weren't displayed
(because they weren't inside the 100 limit), then, editing the value
by (un)selecting a checkbox would automatically remove all non
displayed values from the relation.
This commit ensures that we keep in the relation all values that
aren't displayed.
Issue spotted when working on opw~2439041
closesodoo/odoo#67400
X-original-commit: 9e9d3aa78c42ad4ffca3b56a28382ed84078cde3
Signed-off-by: Géry Debongnie (ged) <ged@openerp.com>
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Since commit odoo/odoo@c96e3b96f3
the TouchEvent constructor was added to the test utils.
In FireFox (no touch mode) and Safari (desktop) this constructor
doesn't exist and so the test suite won't start anymore.
This commit, inserts TouchEvent constructor only when it's supported by
the browser. So now we can run the tests in FireFox and Safari again.
closesodoo/odoo#66846
X-original-commit: 296fd86ddf90cff8d5b8f095d0c51b1c5a8b59ca
Signed-off-by: Géry Debongnie (ged) <ged@openerp.com>
Signed-off-by: rfr-odoo <rfr-odoo@users.noreply.github.com>
Warning: this commit savagely patches qunitjs sourcecode. I know... I
feel bad.
Since we updated the way debug=assets work, we have a new problem in the
qunit test suite: the tracebacks displayed by QUnit are relative to the
bundle file, not the original file, which is annoying in practice.
There is really no good way that I could find to integrate with QUnit to
perform that task, so I had to do it the ugly way: modify QUnit from the
inside to use the StackTrace library to annotate the traceback with the
proper information.
closesodoo/odoo#66771
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
The support for "active_test" context key has been introduced in c2918c1
However, the context key is ignored in many mocked ORM methods (if not all).
This commit adds support for the `search` and `search_read` methods.
Note: this is done to easily write some tests in odoo/enterprise#16363
and it might be usefull to test futures features (or fixes) using this
context key.
closesodoo/odoo#66393
X-original-commit: 0ad1c1631eb233b3e8f98b40f8c0813d0ad7acbe
Related: odoo/enterprise#16447
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Signed-off-by: Lucas Lefèvre (lul) <lul@odoo.com>
The recent change in the way debug=assets works (which now bundles all
files in a bundle instead of serving them statically) had a negative
impact on the stacktrace displayed in the error dialog in debug=assets:
it now display the bundle/linenumber instead of the actual file/line
number.
This is not a huge deal, most of the time, because the errors displayed
in the console display the correct information, and the debugging
process should work as before. But it can certainly be annoying in some
cases.
With this commit, we use the Stacktrace.js library to dynamically fetch
the sourcemaps and to decorate the displayed information with the
correct file and line numbers.
closesodoo/odoo#66318
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
The previous work on adding support for native JS modules needs to adapt
some existing files, which have an incompatible name (with a '/').
Also, we convert a few JS file in /web to the native JS module system,
to show that it can work.
Part of PR 63177
Co-authored-by: Simon Genin (ges) <ges@odoo.com>
This commit add the support to simulate a "TAP" (touch) using coordinate.
This utils will be used in a future commit in web_enterprise to be able
to test some specific case for FullCalendar.
Task ID: 1891957
Disabled in odoo/odoo@b992619840
by mistake.
Also, restore the "failfast" for the QUnit tests (disabled in the same
commit...).
X-original-commit: 7f8e0fe4d2a47968bdcb136fd7e6d96a184f9eae
When you create a new record, you will have in this order:
- default_get
- onchange (can trigger a warning)
- _pushController => close all dialogs since ea2207afea
This is an issue since the possible warning is directly hidden to the
user.
Without the change, the added test failed with:
Warning modal should be opened
"executing a window action with onchange warning do not hide it":
Found 0 elements to click on, instead of 1
opw-2342273
opw-2374051
closes#61732closesodoo/odoo#63415
X-original-commit: ff61ab121e3fb3c0c9b269c0dc9a65b3f1c8e664
Signed-off-by: Nicolas Lempereur (nle) <nle@odoo.com>
Before this commit, the "Import" button was available in the
favorite menus of dialogs opened from many2one and many2many field
widgets. It makes no sense from those dialogs.
Task 2376279
closesodoo/odoo#62079
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
The mock implementation of `name_get` does not fully reflect
the real server implementation.
The real implementation is described in tests added in
this commit.
Note: in my wildest dreams, the mock server tests would be run
against the real *and* the mock implementation to ensure they
do not diverge.
closesodoo/odoo#62430
X-original-commit: c8f59bfd5c8ddb7c7abf1108111507e667393208
Related: odoo/enterprise#15017
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Signed-off-by: Lucas Lefèvre (lul) <lul@odoo.com>
The mock server was slicing the records before sorting.
Now mocking a search read would sort the records before slicing them.
This will ensure the records to be loaded with the correct sorting order
when e.g. a load more is applied on a subset.
Taskid: 2196661
When running the webclient in a regular environment, a 'o_touch_device'
class is set when ran on a touch-screen enabled device.
This commit adds the ability to simulate touch-screen when creating a
view inside a test (independently of the device), allowing tests to
match the reality.
Task ID: 1879566
Introduced a function that awaits for an additionnal rendering frame
initiated by the Owl compatibility layer processing.
By default a simple "nextTick" will handle the rendering of any widget/
component stuctures having at most 1 switch between the type of
entities (Component > Widget or Widget > Component). However more time
must be spent rendering in case we have additionnal switches. In such
cases this function must be used (1 call for each additionnal switch)
since it will be removed along with the compatiblity layer once the
framework has been entirely converted, and using this helper will make
it easier to wipe it from the code base.
Before this commit: the action menu registry items were rendered
regardless of their inner state. The problem is that these item
components can choose not to render themselves, leading to a blank
dropdown menu if no other item was present.
To fix this, the "getProps" function of the registry has been altered to
act as validation: the item will not be rendered if the return value is
falsy. The function has also been made asynchronous to allow RPC calls.
This solution is still flawed and hides a bad design choice which is to
register components and not items or item getters; this API is too
complex for its use case(s) and adds pointless depth to the DOM. The
shape of the registry should be simplified in an improvement in master.
closesodoo/odoo#59259
Task: 2325587
X-original-commit: 45c7d881a5d130f6b3222749c8c87b6c6b45e910
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Co-authored-by: Julien Mougenot <jum@odoo.com>
Co-authored-by: Mohammed Shekha <msh@odoo.com>
After each test, a check is done that there is no leftover in the
DOM. If there are, the suite fails, but there is no indication on
the test that left elements in the DOM.
This commit logs that missing information. We also stop dumping
the DOM when this happens, as it was rather creating noise than
providing useful information (the remaining elements are still
logged though).
closesodoo/odoo#58806
X-original-commit: 598d036b59012af0a95894d759fe3b6276fc8ccb
Signed-off-by: Christophe Monniez (moc) <moc@odoo.com>
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Also fix an issue with livechats not being considered in 'chat'
filter of messaging menu.
Task-Id 2282426
closesodoo/odoo#58468
X-original-commit: 631e52536964763bbfe857305f023e5e67084e95
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
When one constucts a custom filter in the search view and clicks
on the "ADD A CONDITION" button, the new condition dropdowns always
start with the top of the lists.
This commit will instead default a new condition to the same field and
operator as the previous one.
Presetting the dropdowns used in the previous condition is a better
guess for what a user tries to filter than starting the dropdowns each
time at the top of their lists.
closesodoo/odoo#56918
Taskid: 2222545
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Was not super visible, especially without HR as the dialog immediately
closes (limitation of the web client), or when an administrator user
massively disables totp.
Show a toast notification indicating the success of the action.
Also fix the logging of `totp_disable` so it correctly handles being
called on more than one record, and change the logging data to
<browse_record> (<logins>) across the board.
Also add a way to provide an action to execute after a notification:
by default nothing happens, which leads to dialogs not closing and
forms not reloading, and there is no good way to show a notification
and/then do some other thing, which is inconvenient.
This commit adapts the BasicModel to combine calls to `default_get` and
(first) `onchange`. When creating a new record, we now only call
`onchange`, which thus return default values and potential onchange
values.
Tests (and the MockServer) have been adapted accordingly.
NOTE 1. If the `default_get` within the `onchange` returns a value for
a field that is not in the view, we ignore it, and it won't be saved.
Before, that value was kept and sent upon save. This change in behavior
may prove problematic, although the overall risk is small. Decision has
been made to keep heavy comments and code snippets if we were to revert
back somehow to the previous situation.
NOTE 2. Putting a context on a many2one field may change the value
returned by `name_get` for that field. By default, the calls to
`name_get` are done by `onchange`. If the context on a field must be
used for `name_get`, one has to set the option `always_reload` to `True`
on the field. In that case, every `onchange` that changes the value
will trigger an extra `name_get`.
NOTE 3. Suppose that a one2many field has a list view with field A, and
a form view without field A. When adding a line, we now send all known
fields (main view and inline views) to the `onchange`, which may return
a default value for A. The value will appear on the list view, but not
in the form view. The former behavior was to call `default_get` with
the fields that occur in the form view only, and therefore field A would
be left to value `False`.
NOTE 4. A test surprisingly adds an extra call to `read`. The test was
actually wrong before. With the changes in MockServer, we now correctly
receive a command `[6, false, [1]]`, whereas before we received `[1]`,
which isn't a valid command, and which was ignored. As a consequence,
an extra `read` is done, whereas the test asserts it shouldn't. But it
already didn't work before (I checked by sending the correct command).
This needs to be bugfixed elsewhere (task-id-2323491): in a o2m with a
onchange and default order records on an other page than the first
should not trigger a `read`.
Task 2261084
Co-authored-by: Raphael Collet <rco@odoo.com>
Co-authored-by: Aaron Bohy <aab@odoo.com>
Improve mock server:
- add support for mocked `fetch`
- add support for `active_test`
- add support for x2m `in` in domains
- add support for default values computed from a function
- implement a more natural "next id" compute
- allow initial data without ids
- ensure write and x2m commands integrity
- improve bad data/bad commands error messages
- always warn for failing RPC, not only in debug mode
- fix all existing tests that had inconsistency data
Other changes done in mail (or dependents) that are not just related to tests:
- remove `direct_partner` from formatter result
->`correspondent` can be computed from other keys, especially `members`
- fix `livechat_visitor` convertData
-> only process if there is value
- add `current_partner` and `current_user_id` as `init_messaging` result
-> easier to mock than session
- remove usage of `need_moderation`
-> that was just a search indirection to `moderation_status`
- adapt `partner_id` -> `res_partner_id` key in `_notification_format`
-> to be consistent with field name
- add name in result of `mail_partner_format`
-> sometimes display_name is not the same
- remove usage of `is_moderator`
-> that was just an indirection to `moderation_channel_ids`
Enterprise counterpart: https://github.com/odoo/enterprise/pull/11523
task-2287171
closesodoo/odoo#55854
X-original-commit: 7ba3fecb3377a720d1eb70e7515a0c45da73836d
Related: odoo/enterprise#12391
Signed-off-by: Sébastien Theys (seb) <seb@odoo.com>
This commit introduces a new semantics to define action and method buttons
in the list view.
The aim of this feature is to be able to do some actions on selected records, or on the entire
records' domain with the help of buttons located in the controlPanel, much like the form view header buttons.
To define such buttons, one must add a `header` node in the list view's xml definition
```
<tree>
<header>
<button type="action|object" name="actionId|methodName" string="button's text" context="{}"/>
</header>
<field name="fieldName" />
</tree>
```
task-2246383
Adaptation of the tests to the changes regarding the refactored search
panel and model.
Part of task 2228968
Co-authored-by: Julien Mougenot <jum@odoo.com>
Co-authored-by: Mathieu Duckerts-Antoine <dam@odoo.com>
Extraction of the findItem function from the control panel test utils
to the dom utils.
This function helps find an item matching a given selector and:
- having the given text as its inner text
OR
- being the item at the given index
An error is thrown if no item matches the predicate
Support for the "allow exports" group was implemented by checking the
group in init(), which is sync. Move that check over to willStart
instead.
Also add a default mock for user_has_group (makes it so it always
replies that the user doesn't have the group), and override that
specifically for the group we're interested in in the export tests, so
those tests have access to the Export action / option.
closesodoo/odoo#54254
X-original-commit: 30968ac93373990338c23f49e678d272b8e60c6b
Signed-off-by: Xavier Morel (xmo) <xmo@odoo.com>
After changing the organisation of the test bundles, some elements seem
to have disappeared from most of the tests, leaving free space for other
undesirable elements to remain on the DOM after tests unnoticed.
Now, any element not allowed in the body or qunit fixture will be
properly detected and removed.
X-original-commit: 65da01e5b96ff77d25e4421d654585bc08e09d75
In search view, the "invisible" attribute can use complex expression,
and so we must rely on the "modifiers" attribute set by the ir.ui.view
post-processing to determine if a filter should be invisible or not.
For example, considering the context {'hide_myfilter': False}, the
following view:
```xml
<search>
...
<filter name="myfilter" string="My Filter" invisible="context.get('hide_myfilter')"/>
...
</search>
```
Will be returned as such by the server:
```xml
<search>
...
<filter name="myfilter" string="My Filter" invisible="context.get('hide_myfilter')"
modifiers="{"invisible": false}"/>
...
</search>
```
After this commit, the filter "myfilter" will correctly be shown.
OPW-2243639
closesodoo/odoo#53690
X-original-commit: 00969326e24dadd9b3f6e026789a1e575dd1c8dc
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Signed-off-by: Julien Mougenot (jum) <jum@odoo.com>
Signed-off-by: Xavier ALT <xavieralt@users.noreply.github.com>
With this commit, it is now possible to display sample (fake)
data in empty views, in the hope of easing user onboarding.
This can be enabled (in list and kanban views) with attribute
sample="1" on the arch root element. In this case, if there is
no data to display, sample data will be generated based on
heuristics (depending on field types and names). This can be
used in addition to the no content helper.
Task 2232801
X-original-commit: 7c8e627ff5029cb539d702705e5ce53d658c76ed
Co-authored-by: Aaron Bohy <aab@odoo.com>
Co-authored-by: Mathieu Duckerts-Antoine <dam@odoo.com>
Co-authored-by: Jérémy Hennecart <jeh@odoo.com>
Co-authored-by: Julien Mougenot <jum@odoo.com>
Leftover tooltips may make following tests fail, which isn't what
we want as a test shouldn't impact another one.
X-original-commit: 17a4c72b9a86186c7690bdba5e62ccd64ab2f77a
This commit is a significant rewriting of client-side discuss, chatter,
chat window, and messaging menu using OWL. The behavior should be broadly
the same, with some slight functional changes here and there.
From a technical standpoint, the code of messaging is mainly organized in 2
main groups of modules:
- models, which are logical entities that depict the client-side state of
messaging as a whole.
- components, which are in charge of displaying information from models.
This refactoring also introduces new JS guidelines regarding folder structure
(/static) and naming rules for JS modules.
Community PR: https://github.com/odoo/odoo/pull/39023
Enterprise PR: https://github.com/odoo/enterprise/pull/6249
Task-1914207
This PR is a collaborative work by Alexandre, Julien, Sébastien and Xavier,
with the precious help of Lucas to speed it up towards the end.
closesodoo/odoo#39023
Related: odoo/enterprise#6249
Signed-off-by: Alexandre Kühn (aku) <aku@odoo.com>
Co-authored-by: Alexandre Kühn <aku@odoo.com>
Co-authored-by: Julien Giannone <jgi@odoo.com>
Co-authored-by: Lucas Perais <lpe@odoo.com>
Co-authored-by: Sébastien Theys <seb@odoo.com>
Co-authored-by: Xavier Dubuc <xdu@odoo.com>
The counters for categories were not updated when filter values were
selected. This commit fixes that situation.
This might impact the global performances of the search panel but
there is still room for improvement. Actually, it could be
possible to reload categories and filters less often by carefully
track the internal changes and the categories/filters attributes
(enable counters, expand,...).
closesodoo/odoo#49307
Related: odoo/enterprise#9795
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
In the search panel, the domain used to compute the field image values
(when the attribute expand is false) was the same as the domain used
to compute counters (if enabled). The idea was that a value bringing an
empty domain is totally useless and should not displayed.
That being true, it turns out that if one does not allow them and
several fields are used, selecting a value in the search panel will very
often totally transform the search panel. From a UI perspective,
this turns out to be bad: the search panel 'moves'.
Let us give an example:
Let us start from a search panel that looks like to:
first_field
A 1
B 3
second_field
C 1
D 2 <--- mouse above D
E 1
with first_field and second_field both with expand="0" and
enable_counters="1". Let us also assume that no record in the global
domain has both first_field=A and second_field=D.
Click on D would make the search panel look like to something like
first_field
B 1
second_field
C 1
D 2
E 1 <--- mouse here
(the selection of D does not impact the values for the second_field but
does for the other first_field values).
This has led us to use basically only the domain comming from
outside of the search panel to compute field image values.
This means that we might now have value with zero count even if expand
is false. In the above situation, a click on D would give us
first_field
A
B 1
second_field
C 1
D 2 <--- mouse still above D
E 1
Task ID: 2154749
Co-authored-by: Mathieu Duckerts-Antoine <dam@odoo.com>
Co-authored-by: Julien Mougenot <jum@odoo.com>
This commit introduces a new attribute 'limit' for search panel fields
that allows to avoid performance issues. That integer attribute (with
default 200) allows to fix a maximal number of values to display for the
fields. When the number of field values to display reaches the limit,
no values will be displayed. Instead, a warning message will be shown
in the corresponding search panel section.
Note it is possible to have no limit using limit="0" on a field.
This commit reintroduces in a better way the principle brought by the
fix 8d57153b34c04952a85f6642951cb2697016da84.
Task ID: 2154749
The commit introduces two new attributes for search panel fields:
- hierarchize: boolean attribute (default True) available for
many2one fields with select="one". It allows to choose whether
to hierarchize the field values using the _parent_name (if set)
on the field comodel.
Note that a sanitization of the parent hierarchy takes place.
Basically, it ensures that parent chains are
completely in the domain (on comodel) accessible by the user.
See _search_panel_sanitized_parent_hierarchy documentation for
more information.
- expand: boolean attribute (default False) available for many2one
and many2many fields. If set to true, all field values are fetched
and displayed in the search panel. If set to false, only the
values that have at least one corresponding value in the field
model (and in some domain) are fetched.
An exception in the case of an hierarchized field
(hierarchize=True and _parent_name set): more/less values can be
displayed in order to have a good representation of the parent
hierarchy. That means we complete and sanitize the set of initial
field image values.
Note that the fix 8d57153b34c04952a85f6642951cb2697016da84 bringing the
notion of limit in search panel has been reverted in the present commit.
An upcomming commit will reintroduce the limit principle in a better way.
Task ID: 2154749
Co-authored-by: Mathieu Duckerts-Antoine <dam@odoo.com>
Co-authored-by: Julien Mougenot <jum@odoo.com>
Accessing the "Create and edit" option can be bothersome when testing:
- you must click on the m2o dropdown to open it
- input a value
- click on the "Create and edit" m2o item
Now, a helper exists to compile all these tasks into a single helper:
the many2one 'createAndEdit' function.
Usage:
`await testUtils.fields.many2one.createAndEdit('m2o_field', "ABC");`
Before this commit, the mock server "_mockDefaultGet" method was a bit
naive:
- it accepted any 'default_' field coming from the 'kwargs.context'
regardless of the existence of the field
- it accepted any value for these fields
Now the method is looks a bit more like its original server-side
counterpart:
- the context keys are added if they correspond to actual fields
- a pseudo validation removes many2one fields values not corresponding
to any record
The control panel test helpers related to the "time range" menu have
been removed and a new helper for the "comparison" menu has been added.
Task ID: 2245719
Co-authored-by: Mathieu Duckerts-Antoine <dam@odoo.com>
Co-authored-by: Julien Mougenot <jum@odoo.com>
This commit fixes the method GET in httpRequest and
adds it in test_env.
Before this commit, the GET method crashed because
of the forced body param.
closesodoo/odoo#51980
X-original-commit: 7c4fba660a13c38048020faf595bd348d1bca225
Signed-off-by: Aaron Bohy (aab) <aab@odoo.com>
Having _lt in the env doesn't make much sense, as it is useless
at run time, and it would incite people to require 'web.env' to
use it (at load time). The env should never be directly imported,
it is automatically propagated from parents to children.
To sum up, in Owl components:
- when terms have to be translated at run time (e.g. in a function
called when the user clicks on a button), the env is available
and env._t must be used
- when terms have to be translated at load time (e.g. a JS module
defining a component with a translatable static property), one
must require 'web.translation' and use function _lt it exports.
closesodoo/odoo#50656
Signed-off-by: Géry Debongnie (ged) <ged@openerp.com>
This commit introduces several changes in the search panel with
respect to record counts:
- the record counts are now also available for
the fields with select="one" attribute
(if not disabled explicitely).
- the record counts are better computed using the idea that
selected values within a group should not impact the counts
for the group values but only the counts for the other
group values.
On the way we have changed two keys in the values returned
by the server:
- 'count' becomes '__count'.
It has been done to avoid a possible clash in case a model would
have a field named 'count' and that the field values would be
wanted for some reason.
- 'name' (multi case) becomes 'display_name'.
It has been done in order to make the select one and multi cases
more similar and factorize some code.
TASK-ID: 2166814
Co-authored-by: Raphaël Collet <rco@openerp.com>
Co-authored-by: Mathieu Duckerts-Antoine <dam@odoo.com>
Co-authored-by: Alexis Lacroix <laa@odoo.com>
Co-authored-by: Julien Mougenot <jum@odoo.com>
Xavier Morel