A capture load now replays its queued tail, and that replay runs through
`batchTerminalWrite`, which samples `_wasAtBottomBeforeWrite` before it queues.
It runs inside `chunkedTerminalWrite`, before that promise resolves, with the
terminal freshly reset and rewritten — so the sample is always true. The caller
then restored the reader's position and the next `flushPendingWrites` scrolled
straight back to the bottom off the latched flag, undoing it. The only thing in
the way was `_hasRecentUserScrollUp()`, a 1500ms window a server-triggered
refresh is usually past.
`_syncStickyScrollBaseline()` re-takes the flag from wherever the viewport now
sits, and the two paths that restore a position call it right after doing so:
`_onSessionNeedsRefresh` and `_maybeRefetchFullHistory`. Those are the paths
#259 and #205 exist for, and they are also where a non-empty queue is most
likely, since a needsRefresh fires when output is flooding. Re-taking rather
than suppressing the sampling: suppressing leaves whatever stale value the flag
held from before the load, which on the full-history re-pull has no reason to
be false. `selectSession` and `_onSessionClearTerminal` deliberately end at the
bottom, so the sampled true is already the truth there and they do not call it.
`_bufferLoadFinishOpts` gains the coverage the CI gate can see: both mux
sources flush, `history` does not, and a payload naming no source does not.
Its only coverage was the browser suite, which CI does not run.
The JSDoc and the changeset now record the one duplicate window this cutoff
cannot close. The server appends output to the byte buffer in the same tick it
emits, but broadcasts on a batch timer — 8ms over WebSocket, 16 to 50ms over
SSE — so a batch pending when `capture-pane` ran leaves the server after the
reply and is replayed although the capture holds it. It is one batch interval
wide against a recovery window spanning the whole chunked write, and closing it
means flushing that batch server side before the capture.
The second browser test asserts its session was created, so a failed create
fails it instead of passing with zero hits.
docs/architecture-invariants.md no longer claims the replay leaves the
queued-event discard window alone. That clause now describes what decides how a
load ends, the baseline rule, the batch window, and the three covering tests.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Two render-polish fixes for codex sessions in the terminal write pipeline:
- flushPendingWrites uses a 32KB first-frame budget for codex (vs 64KB for
other modes). Codex's TUI emits dense synchronized redraws during
thinking/high-effort phases; a smaller first frame keeps per-frame
xterm/WebGL stalls short and avoids multi-second main-thread blocks.
- Sticky-scroll now honours a short grace window after a manual scroll-up
gesture (USER_SCROLL_STICKY_SUPPRESS_MS = 1500ms). High-frequency codex
"Working (Ns)" status ticks were snapping the viewport back to the bottom
while the user tried to read earlier output. The wheel/touch scroll
handlers record the gesture (_noteTerminalUserScroll); flushPendingWrites
suppresses the auto-scroll-to-bottom and restores the preserved viewport
via scrollToLine while the grace window is active.
Adds test/terminal-flush-budget.test.ts (vm-sandbox harness over
terminal-ui.js): codex vs non-codex first-frame budget, buffer-load
ownership, and the scroll-up suppression / viewport restore.
Co-Authored-By: Saqeb Akhter <saqeb.akhter@gmail.com>
