| Age | Commit message (Collapse) | Author |
|---|
|
|
|
- remove premature optimization for fast-notifies. blobs will cover
that, no point in maintaning the optimization at the price of creeping
complexity.
|
|
- Create the 'Sent folder' ourselves to avoid pixelated hitting a bug in
mailbox creation.
- I believe there's still a problem with bitmask desing for the adaptor
(in get-or-create mailbox). This needs further tests.
- Case manipualation to avoid having a 'Sent' and 'SENT' folder when
Thunderbird and Pixelated write to those.
- Further hacks to monkeypatch the leap-mail-adapter that Pixelated
uses (make them reuse the account instance!). This is getting insane,
I am really looking forward to the fork.
- Duly note our technical debt in the area of Pixelated integration.
Keeping the Pixelated codebase untouched for a long time will
backfire. As far as I've noticed, we have a basic violation of the
assumptions about a single-instance writes and notifications to all
listeners. As commented in the commit, this should go either for a
guarantee that only one account object is created per user (creating
it in the bootstrapping process in bitmask), or for the opposite
direction in which the listeners are communicated in some other way
(zmq events, for instance).
- In any case, it's strongly recommended to deduplicate the Pixelated
libraries as soon as possible and make Pixelated use a better defined
set of Bitmask's public apis.
- Modify the wrapper create methods so that they return the modified
wrapper itself.
- Resolves: #8903, #8904
|
|
If we do fast-notifies with pixelated running, the poll notifications
for new mail on inbox break on pixelated side, because the email is not
ready.
- Related: #8798
|
|
When setting the listeners in the IMAP Folder, we avoid setting more
than one listener for the same imap mailbox (because in some situations
we were registering way too many listeners).
this was making the pixelated inbox registering the notification and
therefore the imap mailbox not being registered.
this MR also refactors the way pixelated is initialized, so that it
avoid creating a second Account instance. In this way, we make sure that
the pixelated mua and the imap server share the same collections for a
given mailbox, and therefore any of the two is able to get a
notification whenever the other adds a message to the mailbox.
- Resolves: #8846, #8798
|
|
- Resolves: #8870
|
|
|
|
|
|
I changed most of the logger statements to use a class attribute, in
this way it's easier to identify which class it's logging them.
in some cases I leave a module-level logger, when we're either using
functions or when the module it's too small.
at the same time I did a general review and cleanup of the logging
statements.
|
|
|
|
the migration from the stdlib python's logger to twisted logger is
unable to format properly the failures unless the call is adapted
- Related: #8849
|
|
- Related: #8798
|
|
In addition to the UnicodeError exception in _process_decrypted_doc
function, we have added ValueError to the exception list so that we can
catch any error in JSON parsing, specially a 'NODATA' error that we were
getting with some of the emails.
This is in reference to issues -
https://github.com/pixelated/pixelated-user-agent/issues/908 &
https://github.com/pixelated/pixelated-user-agent/issues/981 - to ignore
documents that have this problem and not have Soledad try to sync them
again and again.
with @deniscostadsc
|
|
with @deniscostadsc
|
|
with @deniscostadsc
|
|
The offline toggle added a race condition in which the client could end
up offline forever. We decided to remove it for now and then decide if
soledad client needs to know about network connectivity (check:
https://0xacab.org/leap/soledad/issues/8789).
|
|
a bit hacky and all, but this should launch the service
and allow interacting from the default site (localhost:9090).
this is the first example of a pyqt-js bridge, it's an interesting
mechanism that we can use more in the future.
no efforts made so far in authenticating the app.
|
|
|
|
- Resolves: #8721
|
|
|
|
- Resolves: #8754
|
|
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
Previously, if an email was signed with a new key, that was also sent
as an attachment, the verification of the signature could fail if the
only available source of this new key is the attachment ifself.
I changed to extract the attachment before adding the leap header,
which is responsible for the signed/encrypted flags. Also, if the
previous verification failed and a new key was successful imported,
it's going to decrypt the original email again, just to update the
verify status.
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
Signed-off-by: Ruben Pollan <meskio@sindominio.net>
|
|
- some pep8 problems
- some confusion with old and new code after merging
from old keymanager
|
|
|
|
|
|
Twisted 16.5 sends a BytesIO as message, which was unexpected in this
types list.
|
|
this looks like a careless migration to twisted failures.
who knows how long this was like this, this should be covered by tests.
|
|
|
|
Following Twisted documentation, I found out that the warning method
doesn't exist on logger class. That class has warn method instead. And I
got a error because of that.
https://twistedmatrix.com/documents/16.4.1/api/twisted.logger.Logger.html
|
|
Before we used to have content_doc.content_type = 'text/plain;
charset="utf-8"'.
Currently, we have content_doc.content_type = 'text/plain'.
This change will give us, on top of that: content_doc.charset = 'utf-8'.
One can reconstruct the original one using .content_type and .charset.
|
|
|
|
|
|
|
|
|
|
|
|
not using this anymore, was a vile hack.
and a very bad idea too, happy to see it gone.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
