Repository navigation
Conversation
… and spell the assertion's count and owners to the reaction's own types
…ersation module, take the name resolver as a reborrowed FnMut, and let who read a shared context
…nference is unambiguous
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changed and why
A reaction now keeps who the network already said gave it, and the window shows them:
hovering a chip names them, right-clicking it (desktop) or long-pressing the message
(phone) opens the list grouped by emoji, and a row opens the person's card. Networks
whose payloads carry only a count — Discord's message payload and Tiny's history rows —
keep a plain chip and no list, and Telegram's hidden lists (channels, large groups) stay
hidden. Nothing is asked of any network: every name shown was already in a payload chatd
fetches today.
Type
Platforms
Proof
Scenario: e2e/scenarios/reaction-reactors
Lab run:
Compatibility
Only the protocol grew:
Reactiongainedwho: Vec<String>(
crates/protocol/src/message.rs), with#[serde(default)]andskip_serializing_if = "Vec::is_empty", so an older frame reads and an older shell neversees the field; an older chatd's frames read unchanged in the new shell.
Message::boundcaps the list at
REACTION_REACTORS(50) ids of 200 characters before the store writes arow, so no network can size it past the existing caps. No request, route, or host is new:
no adapter makes a call it did not make before. Discord and Tiny deliberately stay
count-only; their follow-ups (a per-emoji Discord endpoint, a tinyserver wire change) are
named in the issue's staging, not smuggled in here.
Per-network honesty:
reactions[].usersand the reaction eventsemotions[].users[].mriReplyTo.personalready carriescan_see_listis set and the message is notmin; counts only otherwiseMarkalready carrieswhoA person the shell has not resolved yet reads "Someone", never a bare id; nothing from an
encrypted Tiny envelope is read, and reactions were never inside it.
Risk and rollback
The list is display-only: no send path reads
who, so a wrong name costs nothing but alook. The store's event merge is the shared piece — a wrong removal could leave a stale
name until the next payload replaces the list, which every network here sends on its next
message read. Rollback is a revert; old cached rows read back as count-only chips with no
migration.
Release note
Hover or long-press a reaction to see who gave it — on Slack, Teams, Telegram (where
Telegram shows it), Signal and Messages; a count-only chip stays a count.