initMountPoints is marking a user as successfully initialized too
early. If the user was not found an NoUserException was thrown, the
second time initMountPoints is called would not rethrow the exception
and happily continue.
This fix makes sure that we consistently throw NoUserException when
initMountPoints is called repeatedly with invalid users.
In some scenarios initMountPoints is called with an empty user, and
also there is no user in the session.
In such cases, it is unsafe to let the code move on with an empty user.
In theory, if your instance ever creates more jobs then your system cron can
handle, the default background jobs get never executed anymore. Because
everytime when the joblist returns the next job it looks for the next ID,
however there is always a new next ID, so it will never wrap back to execute
the low IDs. But when we change the sort order to be DESC, we make sure that
these low IDs are always executed, before the system jumps back up to
execute the new IDs.
The hook now calls the share manager that will call the responsible
shareProvider to do the proper cleanup.
* Unit tests added
Again nothing should change it is just to cleanup old code
If we do not allow public upload we should limit the permissions on
links shares upon retrieval.
* Added unit test
* Allow fetching federated shares by token as well
This makes the post_userDelete hook call the sharemanager. This will
cleanup to and from this user.
* All shares owned by this user
* All shares with this user (user)
* All custom group shares
* All link share initiated by this user (to avoid invisible link shares)
Unit tests are added for the defaultshare provider as well as the
federated share provider
in order to create a 1:1 copy of a file if a version gets created
we need to store this information on copyBetweenStorage(). This
allows us to by-pass the encryption wrapper if we read the source file.
The code path called when using external storage with WebDAV is using `\OC\Files\Storage\Wrapper\Encryption::getMetaData` which did not contain the actual encrypted version inside the cache entry version. This lead to the following:
1. User uploaded a file
2. File is created and `\OC\Files\Storage\Wrapper\Encryption::getMetaData` is called. It has an empty `encryptedVersion` but sets `encrypted` to either `true` or `false`.
3. The call when updating the file cache will use the old version.
Fixes#23536
The new sharing code is much stricter in checking permissions. However
for non moveable mounts the permissions UPDATE+DELETE are not reported
on the mount point.
This is just a quick fix.
* Updated unit tests
In 9.0 we converted the old shares to the new shares. So for 9.1 we can
savely remove the fallback code.
This code was required when there was no initiator set.
* Fixed unit tests
The old behaviour was that only languages could be used for an app
that are already present in the apps/$app/l10n folder. If there is
a themed l10n that is not present in the apps default l10n folder
the language could not be used and the texts are not translated.
With this change this is possible and also the l10n files are
loaded even if the default l10n doesn't contain the l10n file.
In case of a move operation from an unencrypted to an encrypted
storage the old encrypted version would stay with "0" while the
correct value would be "1". Thus we manually set the value to "1"
for those cases.
See also https://github.com/owncloud/core/issues/23078
A PreconditionNotMetException must only be thrown if explicit
preconditions are specified for setValues(), not if the value is merely
the same as was already in the DB.
The setup uses `\OCP\IRequest::getInsecureServerHost` which in some cases can also include a port. This makes the trusted domain check fail thus.
I've decided to add this here that way because adjusting the setup would require parsing the host properly. This is not something that can be done very good in PHP. Check the following example for why `parse_url` is not our friend: https://3v4l.org/k501Z
When calling `\OC\Files\View::copy` we should also keep the version to ensure that the file will always have the correct version attached and can be successfully decrypted.
To test this the following steps are necessary (from https://github.com/owncloud/core/issues/22781#issuecomment-191328982):
1. setup a new ownCloud 9.0 beta2
2. enable encryption
2. upload a docx (5.7MB large)
3. upload the same file again and overwrite the existing file
4. I can download the original file and the first version
5. I restore the first version
6. restored version can no longer be downloaded with the error described above
The manual cache operation in `\OCA\Files_Versions\Storage` is unfortunately necessary since `\OCA\Files_Versions\Storage::copyFileContents` is not using `\OCP\Files\Storage::moveFromStorage` in the case when an object storage is used. Due to the workaround added in 54cea05271 the stream is directly copied and thus bypassing the FS.
Sharing of the users root folder should not be allowed as it is very
weird UX. Also many of our clients have no proper way of displaying
this.
Added unit test
Also added intergration tests to make sure we won't allow it in the
future.
Thomas Müller
Thomas Müller
Robin McCorkell
Lukas Reschke
Robin Appelman
Christoph Wurst
Morris Jobke
Roeland Jago Douma
Vincent Petry
Joas Schilling
Roeland Douma
Joas Schilling
Stefan Weil
Jörn Friedrich Dreyer
Frank Karlitschek
Bjoern Schiessle