Add a new report in order to display:
- bom structure and cost
- bom structure
- bom cost
The user is allowed to print multiple bom reports at once.
Print the report dynamically depending what have been unfolded.
In html the rendering is lazy, when unfolding the client make a request
to the server in order to get the html that will be append to the DOM.
For the pdf the kwargs unfolded could be used in order to render everything
at once.
With the help of:
Dhaval Suthar <dut@odoo.com>
Jigar Patel <jat@openerp.com>
task-36741
Since report refactoring the function to override is get_report_values
that only returns a dict of value that are used in rendering.
In order to render a pdf with a landscape orientation, a context key could
be added however since it's not possible to return a report object in
get_report_values, it's difficult to add a contexti without using some tricks.
This commit add a report option to use in the template that will be used
in order to modify the pdf orientation. The process used was already existing
for other options like dpi,...
Current demo data do not have advanced configuration.
This commit provides new demo data in order to test
bom report or advanced MRP configuration with sub bom
bom with kit,...
Impacted modules: hr_timesheet, sale_timesheet and project.
This merge commit provides feature for service flow (selling
timesheet). The most important points are
1/ selling project based on template: on product configuration,
user can now set a template for the Sales Order to create a
project by duplicating this ones (stages, tasks, ... are duplicated
from the template). If several Sale Lines are based on the same
template, only one will be created. Same mecanism if multiple product
creating a project without template: only one project per SO.
2/ Project Overview Revamp: before this, it was based on a filter
domain on analytic line (aka timesheet). This causes many problem
as if you are selling prepaid services, you project plan was empty
as long as no timesheet was created. Moreover, this kind of domain
is not very usefull for the user (nobody want to see project plan of
timesheet containing "analysis" for instance). The project plan is
now a filter on project, to ease fetching data from the project (even
if no timesheet or forecast), and allow cross project analysis.
The overview table part is now available even when forecast is not
installed.
3/ Assign Tasks Massively: a wizard is provided to ease linking
task to the same Sale Order Line, to ease billing their related
timesheets.
4/ Usability Details: filter task per SO or SO line, stat button
on Sale Order now redirect to kanban project instead of global
project plan, generated task/project names are cleaner, ...
5/ Installing only timesheet module, user will be able to track
cost without selling time. Fields of cost and employee hourly cost
are moved from sale_timesheet to hr_timesheet.
To make this feature works, some technical efforts was required:
1/ Project and Analytic Account: To allow SO to generate multiple
project, we wanted them to share the same analytic account (the one
of the Sale Order). To do so, we had to remove the inheritS between
project.project and account.analytic.account models.
To allow timesheet on a project, an analytic need to be set manually,
or it will be automatically generated (like before with the inheritS),
if allow_timesheet is checked.
2/ Refactoring of SOL service generation: to allow SOL to generate
only one project (per template) for the SO, the part of service creation
on sale.order.line needed a refactoring.
3/ Revamping the overview also improve its performances: some part
are now based on SQL queries (table part), or on report (profitability
report). Doing a report allow the user to create its own filters, and
give him a more accurate tools to extract statistics then project plan.
The idea is to get only one source of information for reporting and
analysis.
Please see sub commits for more details.
Thanks for @tde-banana-odoo for his review and advices during
development,
and thanks to @bst-odoo for his functionnal ideas, testing and support.
Tasks #34702Closes#22268
'remaining_hours' is defined in project as a simple
float field, manually updateable, only displayed
when the group "Time Estimation on Task" is activated.
But for now, there is no setting to active this group.
Those are legacy useless stuff, so we can remove it.
The purpose is to make 'time estimation' feature only
available when timesheet is installed.
Thie commit also split the compute methods of
'remaining_hours' and 'progress' fields to make
'remaining_hours' updateable, though inverse method.
Without splitting the compute method, the calculation
of the progress was done correctly during onchange
but not when saving.
We want to split the sale order line description as follow:
- first line will be the title of the task, prefixed by SO
reference
- the rest of SO line description should be the task
description
The purpose is to avoid duplicating the first SO line
description in task title and description.
Task #1831934
From task stat button from SO form view, it
is relevant to see the task of the SO only.
So, we add a filter to do so. AS the kanban
task use a read_group, it is required to store
the sale_order_id field on task.
The code of the project overview is split to be used when only
timesheet is installed (no forecast). The table part only
display column for timesheeted time per month.
When forecast is installed and activated, other columns
comes to complete this table.
This commit also makes visible SO line having no timesheet
linked in project overview.
When selling a project from a sale order, the SO line
creates the project. When we create a task in that
kind of project, by default, the task will be linked
to the SO line of its project.
However, for the SO line field to be visible on the
task form view, the customer must be set. This commit
set the customer from the SO line when changing
the project (billable project only).
Selling product creating project can now be based
on project template. When setting a project template
on product, this will copy the project (like a duplicate)
and replace its customer, ... with information from the
SO line, on SO confirmation.
There is only one project without template per SO, and
one project per template per SO (so 2 SO lines with the
same project template on their product will only create
one project with this template).
When confirming a SO, only one project will be created, even if
some so lines would create a project. Only the first line will
create the project, the other will not be linked to its.
`project_id` field is added on so line, to keep track of the
generated project, as already done for the task.
The '_timesheet_service_generation' method has been rewritten
to be execute in batch and improve performance and prefetching.
On SO confirmation, no message are now post in chatter to avoid
having one message per so line that should have created a task.
A message is only posted when user manually add SO lines on
confirmed SO.
Now that an analytic account can be linked
to several projects, you might want to change
the analytic on the SO.
Before this commit, the user was allow to create
an analytic account on a confirmed SO but not
modify it. This was usefull when you wanted to
reinvoice an expense on a SO without AA set.
This is now solved (AA is automatically created
when validated expense).
Now, we want to allow set manually the AA before
confirming SO to match the one of potential
projects.
This commit removes the inheritS between account.analytic.account and
project.project. This means :
- When a project is created, an analytic account is not necessarily
genareted
- project has now its own name, company_id and partner_id fields
- an analytic account can be linked to several project
- project module does not depend anymore of analytic, but hr_timesheet
does.
This change implied
- an analytic account is required on project to timesheet on
its tasks. So, when checking 'allow_timsheets', if an AA is not
provided at project creation, an AA will be generated.
- to timesheet on project, an active AA is required. We don't want
timesheeting on project linked to unused AA to avoid bad analytic
entries.
This prepare the fact a sale order can create multiple project, sharing
the same analytic account (the one from the SO).
The project overview is kind of badly design. This
commit does not solve all the issue but tries to make
it simpler and more efficient:
- Does not depends on AAL anymore: the overview search
is now about projects. You can see the overview of multiple
projects. Even when no timesheet are encoded, the plan will
not be empty.
- "Time by people" displays each employee assigned to task,
even if no timesheet are recorded.
- Code now use reporting view to keep consistency and simple
code in overview controller.
The cost of an employee should be in hr_timesheet for a long time, to
allow user to track time on project/task: hours spent and their cost.
This commit simply moves fields and related views. It also moves the
'postprocess' method from sale_timesheet to hr_timesheet since this
method is responsible to set the amount (cost) on AAL.
User can now track their time and cost, without having sale_timesheet
installed.
The project of a SO will not be used anymore, as we planned
to have a sale order creating multiple projects.
The concept of 'only project from SO' does not make sense.
This commit provide a real reporting tool
for project profitability: graph and pivot
views.
You can now have a look at cost and revenues
per project, so line, product, see which
project manager brings the more money, which
project costs many human time, ...
The project overview is based on this report
analysis to be consistent.
However, a change of computation for 'amount
invoiced' is introduced: we now based the amount
on the price_reduce of the SO line, and not from
the invoice lines. Also, once a draft invoice is
generated, the quantity is considered as invoiced
at the price of the SO line (no matter is extra
discount happens when validating invoices).
- Redesign all mimetype icons
- use SVG to reduce file size and improve rendering quality.
- Add a data attribute to attachments to detect the file extension (in somes
cases just mimetype is not accurate)
In some case, (many JS like runbot with all modules), the onload event was
binded after that the image was loaded. So the JS code that make the zoom
conditionnal was not executed.
Now we apply the function on all images already loaded, or that will is loading
Courtesy of @ged-odoo for help and review
An empty div `o_attachment_preview` is added in expense and vendor bills form
views. It will be used as a container to preview the attachment.
The actual behaviour will be implemented in another commit.
Task 40663
Purpose
========
The links on the project kanban cards are not always at the same place : https://drive.google.com/file/d/1aONUYkZerqqD7fSVmgozRScgLKR-QHtA/view
============
Specification
============
I get rid of the flex element to allow the placing of links using width
attribute. Create the .o_project_timesheet_box class to assure that the
timesheet/overview link is always placed in the middle.
I also deleted the needaction_count button to avoid breaking the display.
Task: 1817850
Before this rev:
* Take 2 domains that contain at least two leaves, and at least one of
them must use the implicit `&` operator
e.g.:
d1 = [('so_line', 'in', [91]), ('amount', '<=', 0.0)]
d2 = ['&', ('so_line', 'in', []), ('project_id', '!=', False)]
* Perform osv.expression.OR() between both domains
Expected result:
d3 = ['|', d1, d2]
Actual result (after normalization):
d3 = ['&', '|', d1, d2]
This is because, since the `&` is implicit for the first domain, when we
OR it, we give it an explicit `|` operator, so when we pass this domain
through the normalize_domain function, d1 no longer contains an implicit
`&` operator but instead the implicit operator is the one between d1 and
d2, therefore giving us a completely wrong domain.
The `combine` function states that it only accepts normalized domains,
however neither the OR nor AND functions do, this leads to a lot of
developers putting non-normalized domains into these functions, and
there's no error checking or anything that obviously indicates that the
domain is incorrect, so we might as well normalize all domains being
passed since it's already pretty optimized.
Chat window having higher z-index value then attachment preview model and in the
mobile chat windows are covers whole screen so attachment preview is hidden behind
chat window. so increasing preview model's z-index by 1 will fix the issue.
(PR #24065)
The read_group method was recently modified to allow more flexibility.
With this commit, we add the possibility of aggregating many2one fields.
These fields can be aggregating by counting the number of distinct IDs,
directly in postgres.
Also, note that we had to modify the mock server to implement the
corresponding functionality.
For x2many relational fields(of type kanban), we shoud be able to set custom label for
'Create / Add' button by adding 'create_text' node option on particual field from xml.
Right now, 'create_text' is only handled for many2many fields, preventing users to set
custom label for one2many fields.
This commit allows users to set Create Button label for both m2m and o2m fields.
This commit fixes two issues with discuss tests (JS):
1. Improved mocked `message_fetch` rpc
Some tests were incorrectly mocking the `message_fetch` rpc.
With this commit, the mock server correctly mock this route,
so that tests should simply pass the records as data, and the
mock server correctly handles this rpc.
2. Patch `_.debounce` and `_.throttle`
The test "older messages are loaded on scroll" failed with
Google Chrome right after destroying the widget discuss.
This is due to a debounced function that called its inner
function at this moment.
With this commit, any discuss test patch `_.debounce` and
`_.throttle` so that the inner functions are executed
immediately and synchronously.
Before this patch the registry and cache signaling was only activated
for PreforkServer. In case Odoo was deployed in a multi process/multi
threaded architecture the signaling was not ensured, causing registry
de-synchronisation amongst threaded servers.
This fix will be backported in v10.0 and v11.0 as soon as it has been
battle tested.
Also when there is both a docstring and directive-level content put
the docstring first. That seems to match what Sphinx's autodoc does
and looks less odd.
Possible future improvement: a parameter to suppress the docstring?
Possibly a way to reorder the non-docstring content (e.g. members
documentation)?
- Improve design
- LESS: use mixing and variables when possible, code formatting
- Solve a minor issue causing images to be not perfectly centered
- Improve visual feedbacks (tooltips, cursors, hover states...)
The read_group method (from the server) is implemented in the mock
server, to allow easy unit testing in javascript. However, the
behaviour was not exactly the same: a read_group call on a domain with
no records returns null for aggregates. But the mock implementation
returned a value of 0.
We fix this in this commit, to allow more precise testing of Odoo
javascript.
The state of the JS documentation was quite sad, not updated in a long
time. With this commit, we rework a lot of the JS doc. Most notably,
we add a section 'Javascript Cheatsheet' (help on some common
customization tasks) and rewrite the 'Javascript Reference' totally.
Note that the JS reference page also contains a reference on all field
widgets.
This work is just the beginning, it is probably not complete, but at
least we will have a foundation.
This wizard allows the stock users to replenish a product using the
routes applied to this product. A specific route can also be applied
to bypass the default route.
This replaces the "request procurement" wizard from v10, lost during the
stock refactoing.
Task ID: 47938
related to #22041
This commit allows to define a putaway strategy by product. Before
this commit, it was only possible to define a putaway strategy by
category.
We add a second one2many on the model 'product.putaway': the original
(fixed_location_ids) is used for the putaway per category, the new one
(product_location_ids) is used for the putway per product.
The two one2many are displayed on the putaway strategy and a domain on
the fields is used to filter the records between them. Inline tree view
add a required on the product_id on the one2many per product and a
required on category_id on the one2many per category.
If a putaway strategy per product AND per category match for a product,
the putaway per product takes precedence
This task also reword some tooltips.
We add some testcases to display the behaviour
- when a putaway per product is applied
- the conflict between a putaway per product and category
- the fallback to the category when nothing is found on the product
Task 23495
PR #20143
* don't use FS to do export
* update translations export dialog to fix outdated links (launchpad
& 6.1 translations doc)
Note: can't use ``TarFile.gettarinfo`` to create the TarInfo object
because it wants an actual file object with a ``name`` attribute and
possibly even an fd.
closes#22798
This merge replaces followers subscription code in order to be optimized
and perform less queries. This improvement is done in several steps :
* mail followers model has new low-level and optimized methods to compute
followers to create or update. It uses SQL queries to fetch data in a
single request;
* mail thread message_subscribe method is rewritten to use this new method.
Its API is also simplified and message_subscribe_user is removed to
simplify the whole subscription API;
* auto subscription is also rewritten. Fetching subtype data about parent /
child like project / task is done in a single query that prepares data.
It then uses new method to create followers;
Various addons are updated because of message_subscribe or auto subscription
change. Functionally behavior should stay the same as purpose is to be
equivalent in term of functional coverage.
Performance counters show this optimized code works well when dealing with
large set of followers to subscribe or update. Most tests perform better
of a few queries. Most complex ones involving (auto) subscription of several
followers perform way better. About 30% of query are saved in the three top
tests, aka between 80 and 90 queries with auto subscription of 10 followers.
On overall community runbot we have a gain of about 11K queries. As Odoo
data has few followers impact on runbot is probably less interesting than
impact on real-life use case.
Thanks to @rco-odoo for reviewing, and cheers !
Adrian Torres