When updating translations, the source (`src`) is irrelevant for
`field` and `help` translations. Theses translation types are only
matched through their `name`.
In the small cases where utf-8 is not escaped in the CSS of a module*, an error
could happen when breaking minified CSS on multiple page (for IE9).
For the issue #5050
*currenlty for 8.0 : https://gist.github.com/nle-odoo/e353b22f89031ced21a5
On internet explorer 6, 7, 8 and 9, the limit of CSS rules in a stylesheet is
4095 (http://blogs.msdn.com/b/ieinternals/archive/2011/05/14/10164546.aspx).
This commit breaks down a CSS bundle in several pages for these IE versions.
To do this, the CSS tag added is of the kind : /web/css.0/{xmlid}/{version} in
which there is:
- the whole CSS if there is no more than one page,
- a list of @import pointing to the multiple pages.
note: if a modification lowers the number of page, an old page may stay in
ir_attachment (e.g: go from 4 to 3 pages, the old 4th page of another version
will not be deleted untill the number goes again up to 4).
Note: the method css(self) previously returned an unicode variable (the first
time) or an str variable (the following times, if already cached), the fix
also correct this so an str variable is always returned.
fixes#5050
opw-627116
A cross-registry cache was introduced by e2ea691ce.
The initial idea was praiseworthy but sub-optimal for servers with a
lot of registries.
When there is lot of registry loaded, the cache size was huge and
clearing some entries took a lot of CPU time, increasing the chances
of timeout.
Also, the cache was not cleaned when a registry is removed from
registry LRU (this operation would also consume time).
fails due to ir.config_parameter permission
At rev. 80017b04c2
ir.config_parameter model has been restricted to employees
Getting parameter from this model should therefore
be done as SUPERUSER_ID where the uid could
be a user which is not an employee.
Closes#5280
This rev. is related to eb9113c04d
if a model or a resource id is not defined on an attachment
restrict access to employees only if the creator
of this attachment is not the current user.
So non-employees can access their attachments without
models/resource id, which includes attachment of
discussions threads.
Fixes#4309Closes#4310
The following case has shown the issue: extend the model `res.company` by
adding at least two fields F and G, where F has a default value defined as:
lambda self: self.env.user.company_id.name
If the column F is created before G in the database, the existing records will
be filled with the default value of F. When the default value is computed, the
field `name` from a `res.company` is read, and other fields are prefetched,
including G. This operation fails, because G does not exist in database yet!
Special case for binary attachments with an url, when there is no data
to serve. If the attachment name is an url, redirect to this url,
otherwise return a 204 HTTP error.
Do not load registry to backup a database. This is
allows backing up databases whose registry cannot
be instantiated at the moment (wrong version, missing
modules, etc.)
Closes#5775
When going to logged calls through opportunities,
and no logged calls were listed,
the mail alias of opportunities was displayed
within the placeholder help message,
wrongly, as we were in the logged calls list.
This was due to a context propagation, due
to have defined empty_list_help_* keys within
window actions. It was the case in several
modules.
Besides, we do not find the usefulness of the
custom_context passed to the get_empty_list_help
method: indeed, none of the methods get_empty_list_help
overriden in modules need any other keys than
'default_type', which do not require an evaluated
context.
We therefore remove the whole code regarding this
custom_context, and we therefore get rid of this
useless context evaluation thing within one of
the most accessed method of the web client.
opw-630673
Due to a compatibility problem between the new API
implementation and the @ormcache decorator, the
context parameter of ir.attachment._filestore()
was dropped at rev. 0beb14f0d2
However the need for caching this method has
disappeared in Odoo 8 (it used to require a
DB query). So it is even better to drop the
@ormcache decorator altogether, and keep the
context parameter.
This avoids a useless change of method signature,
even if that was on a private method.
Make sure pass a list of ids instead of a single id to write calls.
Some models (e.g. blog.post, fixed at 12fc5ea) are assuming that it got a list
of ids and is not checking the type.
Always pass a list of ids as it's the expected format for the orm.
When trying to duplicate a database, while
having opened connections to it, postgresql throws
```
OperationalError: source database "duplicate_test"
is being accessed by other users
DETAIL: There is 1 other session using the database.
```
Connections must be dropped before trying duplication
using TEMPLATE to avoid this.
Commit 540b753bf8 introduced
support for resources stored as ir.attachment records in
asset bundles too.
This is specifically useful for customizations.
However the HTTP route for reaching those resources
when they are *not* in a bundle was originally created
in the `website` module (as a special handling for
404 requests)
This means that these dynamic resources would only
be partially supported when `website` is not installed,
causing various problems:
- missing resources in debug mode where bundles are skipped
- errors when trying to define new client-side Qweb templates
via XML resources - which are loaded with a direct request
- ...
This commit moves back the supporting code to the web module.
The `mimetype` column is not present in ir.attachment without
the `website` module, but sniffing it based on the attachment
name works fine at serving time too.
Closes#6002
The implementation of `ormcache` does not work on methods that take a `context`
parameter. Because of the decorator `decorator`, the arguments of the call are
passed positionally to the method `ormcache.lookup`, and positional arguments
are used in the cache key.
The fix consists in removing the `context` parameter from the faulty methods,
either directly, or by caching a private method called by the public method.
Since revs 53582c2ea6 & f65c913027,
this was no more possible for portal users to read groups
on purpose, for privacy reasons.
fields_get of res.users model is overriden, for
the access rights form view features
(The groups selections and checkboxes).
At each call to fields_get, which happens at each call
to fields_view_get on the res.users model, operations are
done on the model res.groups (basically, to
build the selection groups and checkboxes). So,
each time a view of model res.users is displayed,
whatever the view, operations on res.group model were performed.
The thing is, these operations on res.groups
are actually needed only for the user access rights
view, or at least only for users having the group
Administration > Access rights. These group operations
aren't needed for the preferences view, nor for portal users.
We therefore avoid to do these if the user is not part of the
Administration > Access rights group, which lead to
performances improvment, and solves the fact
portal users couldn't access their user preferences view.
opw-627391
In ormcache, those statistic numbers were counted per cached method for all
registries. We introduce separate statistic counters, and create a counter per
(database, model, method). The existing attributes that are no longer used
have been removed.
The registry size is now assumed to be around 10Mb, and ormcache size is
proportional to the maximum number of registries. Statistics about ormcache
usage is shown to the log when receiving signal SIGUSR1.
In `__getattr__`, remove the test on accessing `_attrs`. This situation should
never happen, since the attribute `_attrs` is set in method `__init__()`.
Introduce slots on all field classes for common attributes; slots take much
less memory than a `__dict__`. The other attributes are stored in a dictionary
`_attrs`; all fields with an empty value for `_attrs` (common case) share the
same empty dictionary. This saves quite some memory (around 4.5Mb per
registry), given the number of field instances created for a registry.
Another mechanism is used for the default values of attributes, since slots
cannot be assigned on classes.
The optimization consists in using tuples for attributes `inverse_fields`,
`computed_fields` and `_triggers`, and to let them share their value when it is
empty, which is common. This saves around 1.8Mb per registry.
This saves about 400Kb per registry by not allocating those lists, which are in
most cases empty. The removal of the attribute will also simplify a bit the
management of free attributes.
The computed value of parameter digits is no longer stored into fields and
columns; instead the value is recomputed everytime it is needed. Note that
performance is not an issue, since the method `get_precision` of model
'decimal.precision' is cached by the orm. This simplifies the management of
digits on fields and saves about 300Kb per registry.
Sometimes, the expected mro of the model is not the same as the one built with
a binary class hierarchy. So we reorder the base classes in order to match the
equivalent binary class hierarchy. This also fixes the cases where duplicates
appear in base classes.
Instead of composing classes in a binary tree hierarchy, we make one class that
inherits from all its base classes. This avoids keeping intermediate data
structures for columns, inherits, constraints, etc. This saves about 600Kb per
registry.
The mappings model._all_columns takes quite some memory (around 2MB per
registry) because of the numerous dictionaries (one per model) and inefficient
memory storage of column_info. Since it is deprecated and almost no longer
used, it can be computed on demand.
The ormcache is now shared among registries. The cached methods use keys like
(DBNAME, MODELNAME, METHOD, args...). This allows registries with high load to
use more cache than other registries.
This will allow to speed up our migration platform: as funzip can pipe
the output of the first file of the archive, even if the file is not
fully unzipped yet. This is useful to plays with dump.sql without waiting
for the full archive decompression.
When a `noupdate` record is processed during an update,
the update_dummy method marks the record as "seen" so
it won't be deleted at the garbage collection step,
and seen as an obsolete external ID.
This needs to be done also for the parent records via
_inherits, because they have also received an implicit
external ID at creation, and must not be garbage
collected, even if their `noupdate` flag (db side)
is not set because the flag was added later.
This rare problem could be reproduced by creating a product
in a module, then referencing it in e.g. a sales order,
then updating the module after changing the product
record to be in a `noupdate` block.
At the end of the update the implicit product.template
record would be garbage collected, and trigger a cascade
deletion of its children - blocked by the SO reference
to the product.
Christophe Simonis
Nicolas Lempereur
Christophe Combelles
Denis Ledoux
Raphael Collet
Samus CTO
Olivier Dony
Robert Rübner
Jos De Graeve
FalcoBolger
Goffin Simon
Dharmraj Jhala