On a database where many modules are installed (e.g. runbot all), with
the option "manage available packaging options per products", doing a groupby
"product" on the "package type" list view triggers a js error.
It comes from the fact that the fixed_group variable contains a "length"
attribute; the _().each function uses this to determine (erroneously)
that the object is an array.
It thus uses fixed_group[0] to get an undefined, and then tries to access
a method of undefined.
We replace the each by a simple js loop.
opw 1863576
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.
opw-691070
When we drag and drop a record in a folded column, the number of elements
in this column displayed in the column title is not updated.
This issue is present because when a record is droped in a column
its id is not added to the dataset of its new column.
To fix this, we had to add the id of the record in the dataset, and
override the method 'add_ids' in 'DataSetSearch' to have the size
of the dataset equal to the number of ids in it. We have replicated
the behaviour of the function 'remove_ids' in the new function 'add_ids'.
This is a following of revision
ab8a93c852
This revision solves a new case:
when quickly performing a second search
after one which has not been resolved yet in a kanban view, and the second
search is resolved before the first, it was not possible to load the form
by clicking the card of the kanban, it leaded to a
`Could not find id in dataset` web client error
This was because the ids of the dataset were altered with
the ids of the first search (the one that lasted longer), while
the records displayed were the ones of the second search
(the one that finished first)
opw-682362
M2O use the autocomplete lib to display its content. The problem is
that if a result is empty (or only composed of whitespaces), a
divider is displayed in the dropdown.
Commit 1d9f7d279a was fixing this by only displaying non-empty
name results... but it did not work as many models implement the
name_search method the wrong way.
The new solution is to add a default display for record values which
are empty strings, at the right places in the webclient components.
Avoid duplicating web addon in enterprise by extracting a common basis.
Enterprise features stay in enterprise, but use that common basis.
Mainly:
- JS refactoring and linting
- Conversion of .sass into .less split into multiple files
- Templates cleaning and DOM simplification
- Re-generation of web.pot, and update of .po files
The read_slice method in dataset is problematic, because it alters the
dataset ids. So, if 2 consecutive read_slice are started, but the first
one ends after the second one, we don't have the correct ids in the
dataset.
This commit just make sure that previous read_slice do not alter the ids
of the next read_slice.
- Issue: Edit sale.order form view, add new line, change product who have
taxes and click quickly on save (with slow network) => error message.
Because some deferreds are lost.
- reduce number of reads and onchange for x2m widgets
- remove before_save method and use commit_value (the issue was that if the
before_save and the change trigger an onchange, the onchange deferred is not
checked before saving the content.
The method evict_record calls _update_cache with an empty dictionary,
which will not replace any cached value. Therefore, the cache is never
cleaned.
opw-652393
Some widgets like list editable use and change values dictionary. When the write is called, the values in the cache is equal to the changes send => field no dirty => no change triggered.
The dataset implementation has been improved to handle the result of a `read`
on ids that are already modified in the dataset: the lists `to_create` and
`to_write` are updated properly.
The module system needs to know the dependencies of a given module
before executing the function. This is why the dependencies were
defined once in an array, and then were described one more times in the
call to require.
But a trick can simplify this: the boot function can parse the string
representation of the module and extract the calls to require from it.
It is more work for the processor, but it leads to simpler module
definitions.
This commit introduces a new module system for the javascript code.
Instead of using global (namespaced) variables everywhere (like
openerp.web.whatever, ...), each file declare one (or more) module
with its dependencies. Each module will be loaded after its
dependencies, and will be given a reference to the return values.
This change is a massive departure from the way the client worked
previously. It will sadly break some/most history, but I am sure that
the future git ecosystem will allow you to easily browse the history
past this revision (one can dream).
This commit feature:
* main namespace is now odoo instead of openerp
* a way to define modules and dependencies (inspired by requirejs)
* many files have been splitted in simpler subfiles
* basic support for debugging in odoo.__DEBUG__
* no global variable (i.e. nothing in window.openerp)
* many bug fixes, many new bugs
Future vision: this modularization should simplify future code
development, simplify debugging and will allow a serious testing effort
Christophe Simonis