Take the name pool and the rule matrix from @gryt/core - #375
Merged
Conversation
Both were copies of the phone's, and one of them had drifted without anybody noticing. randomName.ts is gone. Its own header said "This is a copy of the mobile app's src/profile/randomName.ts ... Change one, change the other. GRYT-847." The 152 words were byte-identical either side, so nothing anybody gets called changes. common/index.ts re-exports from the package, so the consumers that reach it through `@/common` do not move. channelPermissionRules.ts becomes a re-export. The implementation it held keyed a cell by role and permission joined with a NUL, where the phone used a space — both consistent, neither broken, and only one of the two safe from a value containing the separator. The package keeps the NUL. That file also held two literal NUL bytes, which is why git had been treating a permissions file as binary and showing no diffs for it. The package writes the same separator as a `\u0000` escape, and this file is text again. Needs @gryt/core 0.2.0. Verified against a packed tarball of it before release. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
sivert-io
added a commit
that referenced
this pull request
Sep 4, 2026
`client:main` has been failing CI since #375. `yarn lint` is the job, and it fails on two `simple-import-sort/exports` errors: `common/index.ts:65` and `settings/src/channelPermissionRules.ts:13`. Both are the autofix. No behaviour, no reordering of anything that runs — the rule sorts the names inside an export list and nothing reads them in order. Kept separate from the GRYT-406 work that found it, because a red main blocks every other branch and should not wait behind a feature. Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
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
1 participant
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.
Second tranche, against
@gryt/core@0.2.0. 83 lines out.The phone side is Gryt-chat/mobile with the same package.
The name pool
This file asked for it:
The 152 words were byte-identical either side — I checked before moving — so
nothing anybody gets called changes.
common/index.tsre-exports from thepackage so consumers reaching it through
@/commondon't move.The rule matrix, where the two had drifted
channelPermissionRules.tsbecomes a re-export. Its implementation keyed a cellby role and permission joined with a NUL; the phone used a space. Both
internally consistent, neither broken today, and only one of the two safe from a
value containing the separator. The package keeps the NUL.
Worth a look: this file held two literal NUL bytes, which is why git had
been treating a permissions file as binary and showing no diff for it. The
package writes the same separator as a
\u0000escape, so it's text again — andthis PR's diff for it is readable, which the last few changes to it were not.
Something I got wrong, and should fix next
Moving the whole file swept five exports into the package that only this app
uses:
CUSTOM_VALUE,EVERYONE_VALUE,scopeChoiceFromValue,scopeChoiceValue,scopeOptions. They're<select>adapters — web-shaped, andthey fail the package's own test of "would both apps need a copy".
They're pure and harmless, so I'm not churning a release for it, but they should
come back here in tranche 3. Flagging rather than leaving the package to quietly
become a dumping ground.
Checks
tsc -bclean. Verified first against a packed tarball, then against thepublished package — the lockfile resolves shasum
edf6dfe3…, the same artifact.