Skip to content

Bring integrate/design-polish into main (to-main-7: #702, #704, #705) - #706

Merged
ufna merged 12 commits into
mainfrom
integrate/to-main-7
Oct 31, 2026
Merged

ufna merged 12 commits into
mainfrom
integrate/to-main-7

Conversation

@ufna

@ufna ufna commented Oct 31, 2026

Copy link
Copy Markdown
Owner

This brings integrate/design-polish into main through the frozen snapshot integrate/to-main-7: design-polish at 6220865, after the to-main-6 sync (#703). With it, main holds the complete Accounts work.

Local gate: the full pre-cut gate passed at b467ae7, apart from the known artefacts below and the account_actions race that #705 fixes. At 6220865:

  • client/core passes all 1,450 tests and its fmt check;
  • the frontend passes 2,171 tests (one existing skip), npm run check, the build and the bundle budget (451,904 bytes);
  • separators and privacy pass.
  • At b467ae7, the gate also passed root clippy and tests, Tauri clippy and tests, Windows cross-clippy (root, core and Tauri), the locks, gitleaks and the Python suites.

Known non-CI artefacts: root and Tauri rustfmt drift, and the terminal_launch failure under a symlinked TMPDIR.

🤖 Generated with Claude Code

ufna and others added 12 commits October 31, 2026 05:21
The to-main-6 Windows native job failed: "/source with spaces" is not
absolute on Windows (no drive), so setup() correctly refused it. The
picker supplies native paths in production; only the test needed a
per-platform absolute source.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
test(client): use a platform absolute source path in the setup test
#693 archives the private source snapshot with 'tar -C snapshot .', so
members are './install.sh' and so on. tests/test_machines.py was
updated, but the Windows-only client test still looked names up without
the prefix and failed the to-main-6 Windows zip job.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
test: read the setup bundle's ./-prefixed names on Windows
feat(client): complete Accounts Add to machine (.25.2.4)
Main now carries the to-main-6 snapshot (4fd10c9). Its two extra commits
are cherry-picks of design-polish's own #700 and #701, so this merge
changes no content; it records main's history here.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Merge main into integrate/design-polish after to-main-6
Single toasts showed the full session name (codex/example/extra-0) while
batch toasts showed a shortened path (extra-6). Both now use the session
list's own label, its catalog title or else its name, for at most three
sessions, found in one bounded pass over the catalog rows.

Closes zerus-7s3s.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fix(client): name Undo sessions the way the session list does
The to-main-7 pre-cut gate hit ETXTBSY in account_actions: setup() and
the transfer test copied a fresh executable fake, and a parallel test
thread forking at that moment inherits the write fd, so exec fails with
"Text file busy". Link the checked-in fakes as tests/common already does;
the edits fake becomes executable in Git. 15 consecutive runs pass.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
test(client): link account fakes instead of copying executables
@ufna
ufna merged commit 240fab8 into main Oct 31, 2026
31 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant