Reply is highlighted even without actual mention when MXID substring matches username

[pacien]

Description

My username matches my domain (MXID of the form @<username>:<username>.tld).
As a result, all replies to messages from any other user on my homeserver are highlighted in my Riot session, flooding me with urgent notifications without intended mentions.

Steps to reproduce

Let @alice:alice.tld, @bob:alice.tld and @caroline:alice.tld be three users on the homeserver alice.tld. All three are members of a room in which:

1. Bob sends a message that does not mention Alice.
2. Caroline replies to Bob's message using the reply feature, still not mentionning Alice.
3. Alice's Riot highlights Caroline's message and Alice is notified, even though nobody intended to mention her.

Remediation?

Do not allow partial matches on MXIDs contained in messages for highlighting purposes.
A highlight should require the whole MXID to match.

Version information

• Platforms: all
• Version: 1.4.1

[t3chguy]

Highlights are calculated server side, not in riot

[pacien]

I haven't tested it, but the issue may probably also arise when someone's username is an infix of someone else's username: @pierre:domain.tld might get notified when someone replies to a message from @jean.pierre@domain.tld. To confirm.

Highlights are calculated server side, not in riot

Should I submit this to the Synapse bug tracker instead?
Submitted on Synapse's bug tracker: matrix-org/synapse#6202

[t3chguy]

I think it's more of a matrix spec thing

[turt2live]

Closing in favour of #7874

[pacien]

Closing in favour of #7874

#7874 looks like a different issue in which messages contain pings hidden in the reply chain.

The current issue is about the way mentions are being matched incorrectly in the body field. In the given example, Alice gets notified even though no actual mention of her has ever been sent by anyone.

Please re-open.

[turt2live]

Ah, well: in that case you need to adjust your notification preferences. Disable notifications for your username in settings, not just your display name. I recommend using keywords instead of the default options (it's how I survive).

Unfortunately notifications are complicated, but this is working as intended.

[pacien]

The same problem appears with user-defined keywords too: they're still generating highlights/notifications when present in someone else's MXID.

I think that MXIDs strings should be excluded when scanning messages for mentions using strings. It makes no sense to try to find a substring in those as they already are explicit mentions as a whole.

Is there any chance to have the specification fixed for this?

[turt2live]

The specification could be fixed, yes. It's a well known issue to the whole team that notifications are a bit annoying to handle.

If an issue doesn't exist already, please open one on matrix-doc