Skip to content

refactor(mobile): drop client-side starter channel setup - #9067

Draft
brow wants to merge 1 commit into
mainfrom
remove-mobile-starter-channels
Draft

brow wants to merge 1 commit into
mainfrom
remove-mobile-starter-channels

Conversation

@brow

@brow brow commented Nov 4, 2026 •

Copy link
Copy Markdown
Collaborator

The relay now joins new community members to default channels itself (#8594). A relay owner or admin flags a channel auto_join, and the relay joins each new member on invite claim or admission. Mobile no longer needs its own starter-channel setup, where after an invite claim (or a hosted sign-in from an invite) it created general and welcome-everyone if they were missing and joined them.

This PR removes that client-side setup:

  • An invite claim now goes straight to the "You joined" sheet. There's no "Finish setting up" or "Retry setup" step, and the success sheet no longer offers "Continue to #welcome-everyone".
  • Mobile no longer creates the starter channels. A community gets default channels only when its operator flags them.
  • The persisted Community.starterSetupIncomplete flag is gone. Saved communities that still have the key load normally, and the key is simply ignored.
  • The hosted flow no longer reruns starter recovery or forces activation for an invite community. Reopening an invite for a saved community still reuses it without a second claim.

Validation

  • just mobile-check and just mobile-test pass. Starter-only tests were removed. The invite and hosted journeys now cover claim success, switching to an existing community, errors, policy acceptance, and the hosted flow without starter steps. I broke the no-reclaim guard once to confirm its test fails.
  • I tested the whole app on an iPhone 17 Pro simulator (iOS 27) against a local relay built from main, with membership required. general and welcome-everyone were flagged auto-join and random was not.
    • A fresh install opened the invite link and joined. The relay logged both auto-joins, and the channel list showed general and welcome-everyone but not random.
    • I opened welcome-everyone from a channel deep link and sent a message, which the relay received.
    • After a relaunch the app restored the community. Reopening the same invite stayed in the community, and the relay recorded only the one claim.
    • Limitation: I did not test the hosted (Block SSO) invite path on a device.
invite.mp4

🤖 Generated with Claude Code

The relay now joins new members to channels flagged auto_join on invite
claim and admission, so mobile no longer creates or joins the `general`
and `welcome-everyone` starter channels itself.

Remove MobileInviteJoinRecovery, the starter channel id derivation, the
persisted Community.starterSetupIncomplete flag (old saved JSON that still
carries the key keeps loading), the retry and recovery UI, and the focus
on welcome-everyone after joining. A successful claim now goes straight
to the success state, and the hosted flow no longer reruns setup for an
incomplete community.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Tom Brow <tomb@squareup.com>
@brow

brow commented Nov 4, 2026

Copy link
Copy Markdown
Collaborator Author

@codex review

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Nov 4, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-11-04T18:34:20.978648Z 4a3abc6 Manual request
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4a3abc6163

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment on lines +204 to +205
state = state.copyWith(
status: InviteJoinStatus.success,

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Add mobile reconciliation for failed relay auto-joins

When an invite claim commits relay membership but any default-channel join has a transient failure, this path now reports success and removes the only client recovery mechanism. auto_join_default_channels explicitly catches these failures and still lets the claim return 2xx (crates/buzz-relay/src/handlers/side_effects.rs:3044-3047), while a later claim is already_member and does not rerun the joins (crates/buzz-relay/src/api/invites.rs:639-643). Since a repo-wide search finds no mobile handling of the advertised auto_join metadata, affected users can remain permanently absent from default channels; retain durable pending state or add session-time reconciliation before treating setup as complete.

AGENTS.md reference: AGENTS.md:L219-L226

Useful? React with 👍 / 👎.

@brow brow added the parked Not being shepherded toward review or merge; sweeps skip it label Nov 4, 2026

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

parked Not being shepherded toward review or merge; sweeps skip it

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants