Help has been moved to web_planner, but only in backend. The comportement needs to be defined in website also so it is now shared in frontend and backend (mixin in common)
Office Outlook doesn't take into account the [style] attribute, and it
doesn't manage width/height in percent value.
This commit set width/heigth attribute in pixels and also always set the
width/height in style attribute (keeping them in the original unit and
setting them to auto if not present).
In the former test, a stored binary field was recomputed. Changing its value
meant changing an attachment, and the latter was deleting a file in the
filestore, because that file was not used by another attachment. The problem
was: that operation is rolled back from the database (test transactions are not
committed), but not from the filestore!
Such a field is declared as:
image = fields.Binary("Image", attachment=True)
With this option, the value of a binary field F is stored in an attachment with
`res_field=F`. When attachments are stored into the filestore, binary fields
are consequently stored in the filestore as well. Note that the option does
not work on old-API function fields.
By default searching on `ir.attachment` filters out attachments for binary
fields: the condition `res_field=False` is added to the domain unless a
condition on field `id` or `res_field` is already present. The client uses
`search_read` with a domain like `[('id', 'in', ids)]` for reading the data on
a form view. In that case, we should not filter out attachments that store
binary fields.
Also, do not filter search results for the superuser.
If, in the same transaction, we need to render a template and read its
associated assets, they are unreachable because the writes in
ir.attachment are not yet commited. Workaround this issue by commiting
manually with new cursors.
This issue occurs when you want to print a pdf report for the first
time.
Added two dummy reports to print an empty internal and external report
and added direct link to edit header/footers template in the form view
(in technical features) for convenience.
Also moved the 'preview rml report' under the rml configuration and
tried to make the company form as clean as possible but i think it's
failed.
Everything has already been moved (read: duplicated) under "company >
report configuration" and only appears in technical features, plus
displaying "Report Font" in "general settings" is really confusing, as
it is a deprecated field.
This field has been removed by e57ac16476
(let's break everything, it is more fun!). But some people needeed it...
This patch proposes thus a new-API version of the sale_id field.
The stdlib version of the json library is more recent than the 3.5.3
version we are pinning in `requirements.txt`
There is no reason to use it.
Closes#6940
The `unittest2` package is simply a backport of `unittest` from the
Standard Library of Python 2.7 to previous versions.
There is no reason to use it any longer.
Closes#6941
Also, we should be able to set a customer on a project, whatever the access rights,
for example, if we're managing services by using internal projects.
When creating pos_orders from the form view, tax_ids on the
pos_order_lines were never set.
This was kind of fixed in 5327869356. But
in there it was in onchange_qty, because the way the fiscal position was
taken into account required it to be. Because we take the fiscal
position into account in a different way now we can assign the taxes in
onchange_product_id which makes a bit more sense.
Allows a user to set multiple fiscal positions per pos_config. Orders
created on that config can then be assigned one of these fiscal
positions. In the POS this can be done through a button that will appear
above the numpad when fiscal positions are set on the pos_config.
* Adding a new group in a group of fields is breaking enterprise layout
* Moreover the group was useless: useless colspan, unique field with
colspan>>>col in it, double access-right, ...
Calendar height was again a problem since the introduction of the
bootstrap layout (the .row had not height 100% and thus the calendar
was not able to compute its height anymore).
Before: if a user blur a editable area without editing, the editor is removed and all bind are remove.
The 'content_changed' bind are removed and the 'o_dirty' class is never apply.
Richard Mathot