Repository navigation
Polish session tag click areas, filters and new-session tags - #2
Merged
Merged
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Three small follow-ups to the session-tag feature:
Existing no-filter creation, folder/worktree options, row actions, modifier selection and mobile hold actions are preserved. Unrelated text or Archived constraints can still hide a new session.
This PR targets jeroendee/pi-pocket:main for inspection. It contains only the polish commit; it does not update upstream PR #118 or its source branch.
Placement
Core under C2/C5: this corrects existing native session-list interactions and stores initial shared metadata through the existing authorized creation path. No new extension, setting, document kind or migration.
Proof
Revision:
7a264a3de67f3f2a69d1d9dc56c94d9477f4698e(one commit above the feature already merged into fork main). Real localPocketApp, HTTP routes and browser assets, isolated synthetic data, scripted model; Chromium desktop and phone-size emulation. No mocked application behavior or live session data.The 22-second native demo below shows all 3 interactions. It samples real browser screenshots at 5fps and encodes at 10fps; no application behavior is simulated.
API example, commands and observed results
The existing
POST /api/sessionsrequest adds optionaltags: string[];cwd,titleandworktreeretain their existing meaning. Response shape stays{ id: number }. The demo's actual creation returned HTTP 200 and opened session 29; its server metadata and visible row both containedpolishandreview.Equivalent call for that isolated fixture (temporary server now stopped):
The displayed command is an equivalent example, not a separately executed curl call; the native UI issued the demonstrated request. Initial names are normalized with the existing rules. Over 10 canonical tags returns 409; malformed tags return 400. Viewer/scoped creation and missing X-Pocket remain 403. Tests verify no session is created on rejection.
npm run format,npm run lint,npm run check, web-module syntax checks andgit diff --checkpass.The complete
npm testrun passed on the byte-identical candidate immediately before commit; the 12 changed-file hashes were verified against the committed files. An earlier tool-limited invocation stopped after 300 seconds; the completed background run is the result reported here:After commit,
npm run check,npm run lintand all 17 new regressions were rerun on this exact revision:Before repair, the new regressions reproduced non-opening blank space, failed Ctrl-selection, the 10px hover gap, missing inherited/persistent initial tags and invalid initial tags wrongly accepted. The same relevant assertions now pass. Existing atomic tag, worktree and tag/filter UI tests also pass.
Limits: local integration evidence, not CI or production rollout. Mobile input is Chromium emulation; physical devices, other engines and screen readers were not audited. No live app, existing preview or upstream PR was changed. The short video is a demonstration; the tests establish the failure boundaries and persistence checks beyond what the video shows.
session-tags-polish-22s.mp4