The catalog's context window is now a text field that reads a decimal K or M
suffix — 1M is 1000K, matching how model capacities are quoted — and stores
the plain token count, so settings.yaml and the adapter are unchanged.
A stored count reads back in the shortest form that round-trips: 1000000 as
1M, 256000 as 256K, and 131072 written out, because it is not a whole number
of thousands. The field holds the typed text while its row has focus, since
re-deriving it from the parsed count on every keystroke would rewrite 1000
to 1K mid-word; text that does not parse stays on screen so the save-time
rejection names a row the user can still see and correct.
- WorkspaceRegistry.archiveSession no longer wraps persistence-listing
failures as WorkspaceUnknownSessionError: only a definite miss (live
lookup, header index, then a fresh list) maps to session-not-found;
storage faults propagate as internal errors, with a negative test.
- The archived-current sweep moves from the unary path into the
projection: any install path (local echo, another tab's frame, a
reconnect baseline) clears a selection that landed in the archive set.
- An archive set installed while workspace.list is in flight supersedes
the stale baseline's set instead of being rolled back by it.
- The workspace-management e2e anchors the archived row by its session
actions button and asserts the single-stray fixture assumption loudly.
- Drop the stale touchSession rows from the workspace READMEs (the
method was removed with its Agent Note).
The visual-only Delete session placeholder becomes a wired Archive
session action: no confirmation dialog (non-destructive), failures stay
console diagnostics. tree.ts hides archived sessions in every
derivation (workspace groups, Ungrouped, search, flat list) through the
sessionVisible predicate. The workspace-management e2e pins the archive
round trip across reload.
Review flagged that "equal by construction" rested on an engine behaviour I
had not measured: `scrollbar-gutter: stable` only equalizes the layers if the
engine applies it to `overflow: hidden` the way it does to `overflow-y: auto`.
Measured it on the running app across the three engines Playwright ships,
and the property does not hold up.
engine .input / .backdrop / .mirror wrap width
chromium 776 / 776 / 776 (768 / 768 / 768 with the declaration)
firefox 776 / 776 / 776 (unchanged by it — overlay scrollbar)
WebKit 768 / 776 / 776 (unchanged by it)
WebKit reserves for `overflow-y: auto` and not for `overflow: hidden`, so the
declaration left .input at 768 against 776 — exactly the gap it was meant to
close — on the one engine where that gap is observable at all, while costing
every chromium user 8px of text column unconditionally. Reverted: the
composer's metrics are now the same as before this PR.
The WebKit gap predates this change and is not closed here. It is recorded in
the Agent Note with the numbers, and the browser scenario asserts the equality
on the lane's engine so a regression into that state fails loudly. The mirror
is unaffected on WebKit for the drafts measured — the extents still agree — but
a draft whose wrapping turns on those 8px would clamp it.
The review's monotonicity concern resolves the same way: the declaration was
never worse than master, because WebKit already measured 768 against 776
without it. It simply was not better.
Also from this round: the wrap-width assertion now covers .mirror as well as
the two glyph layers — it is the height authority, so a mirror alone wrapping
wider would measure the box short and clip content below the 14-line cap with
every other assertion green. Plus `renderGeometry`'s missing `@param
trailingNewline`, and both e2e tsconfig lists restored to alphabetical order.
imccyu
NI0317
Yichen Jiang
Chinesezjc
Tianyi Cui
creatixchu
Hypatia May
kingwl