Else the background job fails hard if the remote has for example an
invalid certificate.

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>

As opposed to hard-coding a list of excluded storages

Signed-off-by: Robin Appelman <robin@icewind.nl>

Signed-off-by: Bjoern Schiessle <bjoern@schiessle.org>

We have a function for it so better to override that. Also because other
codes that might check this should get the right value.

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>

Signed-off-by: Morris Jobke <hey@morrisjobke.de>

* Cache it for a day so we will retry eventually
* Cache the status.php response as well so we will try it once a day as
well

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>

Signed-off-by: Morris Jobke <hey@morrisjobke.de>

Signed-off-by: Morris Jobke <hey@morrisjobke.de>

* based on PR by @Ardinis
* see #4311

Signed-off-by: Morris Jobke <hey@morrisjobke.de>

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>

Signed-off-by: Bjoern Schiessle <bjoern@schiessle.org>

Signed-off-by: Robin Appelman <robin@icewind.nl>

Otherwise the unit test execution will do a ton of external HTTP requests which fail and then timeout…

See https://blackfire.io/profiles/compare/3c67acfa-a11e-4aec-bcd4-c945b006f01e/graph for reference

Pretty similar to https://github.com/nextcloud/server/pull/1565



Signed-off-by: Lukas Reschke <lukas@statuscode.ch>

Fixes an issue where retrying a previously failed federated share would
not properly reset the availability flag because the return value was
undefined instead of "true".

* More explicit http status codes

* handle unavailable fed shares while testing for availability

Fixes https://github.com/owncloud/core/issues/22973 and https://github.com/owncloud/core/issues/22987

If the check is negative it would depending on the used cache store the value as an empty string. When reading the value this check would thus return "false" even if a value exists.

Allows us to catch exceptions like 404's and thus prevents stuff like:

```
{"reqId":"fnmS0hc+fL4+WXKAruo5","remoteAddr":"192.168.39.100","app":"PHP","message":"file_get_contents(http:\/\/ckamm.blaucloud.de\/\/status.php): failed to open stream: HTTP request failed! HTTP\/1.1 404 Not Found\r\n at \/var\/www\/owncloud\/apps\/files_sharing\/lib\/external\/storage.php#260","level":3,"time":"2016-03-08T21:00:33+00:00","method":"PROPFIND","url":"\/remote.php\/webdav\/"}
```

Otherwise setting up the storage will result in a HTTP request and thus slowing down ownCloud.

Replaces https://github.com/owncloud/core/pull/22855

Adds autodiscovery support to server-to-server sharing as specified in the specification. If no discovery data is found it is using the fallback ownCloud endpoints for legacy support.

Whenever a remote fed share's shareinfo call returns a 404, don't
directly assume the storage is not available by throwing
StorageNotAvailableException. We need to properly throw
NotFoundException to trigger the later logic that calls testRemote()
that verifies that the 404 is not from a broken server but really from
an obsolete share.

Added in isSharable() in incoming remote share.
Added in isSharable() in regular incoming share.
Added in FileInfo to make sure the proper attributes are returned to the
clients.

This is even more important now that filesystem_check_changes is 0 by
default.

If the remote server is in maintenance mode, we must throw storage not
available exception instead of not found which might auto-remove the
share.

Most of the time it doesn't make sense to forward Guzzle's
RequestException, so we convert it to StorageNotAvailable instead.

This prevents unpredictable error codes to block access to unrelated
folders needlessly.

This commit consists of patches automatically generated for this project on https://scrutinizer-ci.com