Client::get_url is there for SDK consumers to be able to use the
existing HTTP stack (configuration and all) however in practice it was a
bit weird:
- Responses came back as strings limiting the types of resource that
could be fetched (as well as requiring the string to be converted back
to data before handed to a JSON decoder).
- HTTP errors weren't being raised and instead you would find the (e.g.
404) error response in the Ok case.
This patch fixes both of these issues.
This patch updates our `Raw` API usage since the newly added `JsonCastable` that disallows Raw casts that are known to fail deserialization.
Signed-off-by: Kévin Commaille <zecakeh@tedomum.fr>
This patch changes the query used by
`SqliteEventCacheStore::load_all_chunks_metadata`. It was the cause of
severe slowness. The new query improves the throughput by +1140% and the
time by -91.916%. The benchmark will follow in the next patch.
Metrics for 10'000 events (with 1 gap every 80 events).
- Before:
- throughput: 20.686 Kelem/s,
- time: 483.43 ms.
- After:
- throughput: 253.52 Kelem/s,
- time: 39.478 ms.
This query will visit all chunks of a linked chunk with ID
`hashed_linked_chunk_id`. For each chunk, it collects its ID
(`ChunkIdentifier`), previous chunk, next chunk, and number of
events (`num_events`). If it's a gap, `num_events` is equal to 0,
otherwise it counts the number of events in `event_chunks` where
`event_chunks.chunk_id = linked_chunks.id`.
Why not using a `(LEFT) JOIN` + `COUNT`? Because for gaps, the entire
`event_chunks` will be traversed every time. It's extremely inefficient.
To speed that up, we could use an `INDEX` but it will consume more
storage space. Finally, traversing an `INDEX` boils down to traverse a
B-tree, which is O(log n), whilst this `CASE` approach is O(1). This
solution is nice trade-off and offers great performance.
It uses the statically-known type from the `StaticEventContent`
implementation to call `fetch_account_data()`.
This is the equivalent of `Account::account_data()` but for fetching
from the server. It avoids the need to cast to the proper type after.
Signed-off-by: Kévin Commaille <zecakeh@tedomum.fr>
Currently Element X can't distinguish the cases where a homeserver only
supports legacy SSO without OIDC (and password login isn't avaliable),
and other server unreachable scenarios.
This patch exposes legacy SSO support infomation so that Element X side
can give a dedicated error message when it encounters a homeserver that
can only support legacy SSO.
Signed-off-by: Yorusaka Miyabi <23130178+ShadowRZ@users.noreply.github.com>
Historic room key bundles are uploaded as an encrypted file to the media
repo and the key to decrypt the file is sent as a to-device message to
the recipient device.
In the nominal case, the invite and this to-device message should arrive
at the same time and accepting the invite would download and import the
bundle.
If the to-device message arrives after the invite has already been
accepted we would never download and import the bundle.
To mitigate this problem, this patch introduces a task that listens for
bundles that arrive. If the bundle is for a room that we have joined we
will consider importing the bundle.
It is going away in MSC4291, which should be accepted next week.
Removing it now makes sure that no one uses it.
Signed-off-by: Kévin Commaille <zecakeh@tedomum.fr>
For custom integrations it might be necessary to allow the SDK to
request additional scopes for the OAuth2 authorization code flow.
Currently, only the MSC2967 client API and client device scopes are
requested statically.
Signed-off-by: fl0lli <github@fl0lli.de>
