Commit Graph

57 Commits

Author SHA1 Message Date
daniel 39b995e824 Terminate the FX reclaimer's undo span on EVERY project-state load, not just undo
BeginLoadProjectState now routes unbranched through fxReclaimAbandonForStateLoad;
only the session's deferred reload stays undo-only.
2026-08-07 13:57:06 -04:00
daniel 9b6f725747 Close the FX reclaimer's undo span at the state-load hook, and expose the drain's direction
A Ctrl-Z mid-span used to close the pre-undo block over post-undo state; it now
terminates at BeginLoadProjectState and stops the drain where it stands.
fxReclaimActivity reports Online/Offline/Idle; fxReclaimPending is unchanged.
2026-08-07 13:45:58 -04:00
daniel 86aa8dabba Collapse a Design View mode switch and its background FX parking to ONE undo point
applyMode now returns with its undo block open; the reclaimer closes it when the drain
settles. A transport deferral or a tab switch ends it early -- two points, never a leak.
2026-08-07 12:53:42 -04:00
daniel 02790f1270 Gate the FX reclaimer on the transport, both directions; move the release's persist off the drain and onto the project save
The drain no longer dirties a project on its own initiative — a spent park record rides the user's next save via OnSaveExtensionConfig.
2026-08-06 22:31:47 -04:00
daniel 224dcf84df Fix the reclaimer stranding a tab-switched-into project: lift the pin and wakeup into FxReclaimQueue, retarget instead of going idle
Also validate the cached track in its project, and stop a drop-free tick resetting the report memo.
2026-08-06 22:16:41 -04:00
daniel 7687ea76fc Persist the reclaimer's FX release so a reload cannot re-drive a spent record, and report the FX-restore drops it was swallowing 2026-08-06 21:26:55 -04:00
daniel 5dcfd3a53f Reclaim drain owns its work: invalidation only bumps a generation, and a generation that moved aborts the tick's commit and its release. 2026-08-06 21:08:34 -04:00
daniel 5a35b71dc0 Add the FX reclaimer: a debounced, time-boxed convergence loop that offlines parked chains and restores returning ones off the switch path 2026-08-06 19:07:17 -04:00
daniel c4038a9e9c Strip FX writes from the Design View mode switch; split view.cpp into park/lanes/reconcile/tracks and add the record-scoped undo repair pass 2026-08-06 17:19:43 -04:00
daniel 2423225f0c Split FX out of the park plan: flag-only plans, derived online-first FX work list, two-half record release, applied-state fingerprint 2026-08-06 16:31:59 -04:00
daniel ab14ae769f Guard poll() against a mode-apply mid-flight; two doc corrections
Extends the OnTimer guard to skip session.poll() itself while modeApplyInProgress(), preventing an undo/redo or project-switch reload from replacing the view model under an in-flight applyMode.
2026-08-06 05:23:58 -04:00
daniel 2269fd80b6 Close the collapse's two dropped guarantees: latch applyMode against re-entrancy, pin its project once; correct three overclaiming doc lines 2026-08-06 05:07:31 -04:00
daniel 8c06383fe7 Collapse a Design View mode switch to ONE undo point: FX writes run inline in applyMode's block, deferred park queue deleted 2026-08-06 04:32:26 -04:00
daniel bd86ede5b3 Merge FX-park snapshot guard: refuse to park when the pre-park truth is gone
A snapshot is never taken from a chain a park has touched, on either
half -- flags or per-FX offline. Stranded tracks are refused, named
once in the console, and left recoverable by hand.
2026-08-05 20:17:00 -04:00
daniel 5376ab085c Extend the park trust test to the FX chain, name the refused tracks, and stop reprinting an unchanged refusal 2026-08-05 17:10:19 -04:00
daniel 67524206fb 2nd Pass Render 2026-08-05 16:21:17 -04:00
daniel 8487833342 Revert "Merge Ω-W2-T4: split the deferred FX-park drain by kind"
This reverts commit 70bd29dc89, reversing
changes made to 84afd1b545.
2026-08-03 17:46:17 -04:00
daniel 4b19556513 Merge Insert as FX: load ReaSampler 9000 onto the selected track from the panel
# Conflicts:
#	src/shell/actions/CLAUDE.md
2026-08-03 16:46:40 -04:00
daniel 51c7505dc0 Fix Insert-as-FX review findings: id rename, selection-count drift, refusal tooltip, success msg, master-only, reserve test
Renames the permanent action id out of the placement family, makes the button's
painted state and pressed outcome share one selection count, surfaces
panel-known refusals in the tooltip, reports the target track on success,
distinguishes a master-only selection, and pins the overflow-reserve test to the
real bar spec.
2026-08-03 16:43:10 -04:00
daniel fcd1ed022c feat(panel): add Insert as FX — ReaSampler 9000 onto the selected track, preloaded with the focused capture, over the existing instrument-drop body 2026-08-03 16:06:22 -04:00
daniel 6e2128e937 view: split the deferred FX-park drain by kind — restores stay forced, parks go one FX per idle tick behind a 1s coalescing delay
A rapid A→B→A flip now costs no plugin work: the restore cancels the still-pending
park outright. A park that has already written one FX carries a `partial` flag and is
superseded by its inverse rather than cancelled, so a half-parked chain is never stranded.
2026-08-03 15:00:40 -04:00
daniel 78e5f06928 Drain deferred FX parks before the view model is serialized; re-validate track handles per intent; pin the restore-plan round trip 2026-08-03 14:08:28 -04:00
daniel 7169d7f22b Clear the park snapshot where the restore is planned, not where it drains; make the drain re-entrant and reload-aware 2026-08-03 13:35:30 -04:00
daniel a4a1c3860f Defer Design View's per-FX park to an idle tick; honest undo mask, compare-before-write, PreventUIRefresh bracket, O(1) handle resolve 2026-08-03 12:39:22 -04:00
daniel 77eb7c6680 Merge dev into phase-g: Phase Ε/Ρ and the 1.5.0 bump meet Phase Gamma's instrument work; 120/120 green
The per-directory CLAUDE.md count is re-derived at twenty-seven rather than
carried from either side. The "Decouple the instrument reload from VST3
activation" TODO entry does not survive: Γ-W3-T1 landed it, and COMPLETED.md
carries the discharge.
2026-08-02 21:57:47 -04:00
daniel a3698972db docs: close three review nits — the build-shape index, a redundant header restatement, and the CMake guard comment
Bump the product doc's track count and list for T3; trim the header's
pre-limiter restatement now that bake_render.cpp carries it; add limiter
to the extension's not-linked enumeration.
2026-08-02 21:24:00 -04:00
daniel f9b191bd36 Merge Ε-W2: bank export and bank import, both verbs and both panel rows
Union of two parallel tracks. Both action rows, both menu rows, both link
edges survive; the two package CLAUDE.md files now describe the post-merge
reality rather than either side's pre-merge scope.
2026-08-02 17:19:57 -04:00
daniel 454f67b3bc Close bank-export review findings: name-cap underflow, double overwrite prompt, test scope
Clamps insertSuffix's underflow, floors uniqueEntryName's validity guard, suppresses
the redundant overwrite confirm via a picker out-param, adds a PayloadBuffer
high-water mark, and corrects stale CLAUDE.md/CMake claims.
2026-08-02 17:19:30 -04:00
daniel a927dad2f4 import: a .rsbank lands as a new bank, whole or not at all
Four collisions answered explicitly: ids reminted, names never overwritten,
content deduped before the write, bank name auto-suffixed. Degraded ledger
refuses before the picker.
2026-08-02 17:19:30 -04:00
daniel 081b6f1028 package: one bank leaves the project as one .rsbank, or the export refuses and says why
Pure planner classifies missing/unreadable/unrepresentable and repairs transport
names; the verb digests, streams and commits atomically over a const session.
2026-08-02 17:19:30 -04:00
daniel 5c0f5f1591 Render in place: a track's output to a new sibling, source to the bench 2026-08-02 14:47:59 -04:00
daniel 0afb885987 Make a bake's Banked landing reachable only through the persist upgrade
The dedup path claimed it directly, so a key could be answered Ok pointing at an entry a failed persist never wrote. Write-back is now guarded and three-valued, and landing one request split out of the scan pass.
2026-08-02 12:35:06 -04:00
daniel 2005f90c66 capture: state the bounds tolerance as empirical, refuse unmeasurable renders, keep refused ones for diagnosis
The one-frame bound is not provable for a per-edge renderer; the test now shows where it breaks. Refused renders move out of the bank instead of being deleted, so the DAW experiment has something to read.
2026-08-02 07:23:30 -04:00
daniel 3278b4eced Name captures after their source track: label and filename both, on every interactive mint site, and show the name on the panel card 2026-08-01 21:46:53 -04:00
daniel 09d64c9f46 Merge Ψ-W1-T1: a ranged item capture renders the window, with children and receives silenced for it
# Conflicts:
#	docs/TODO.md
#	src/app/CMakeLists.txt
2026-08-01 21:25:37 -04:00
daniel 7dc80e7a4f capture: refuse the multi-track ranged item render, silence the track's children and receives for it, and gate every exact-bounds capture on its frame count 2026-08-01 20:58:33 -04:00
daniel a0f9a23711 Merge Ψ-W1-T4: resolve drop targets per move, not once
# Conflicts:
#	src/shell/actions/CLAUDE.md
2026-08-01 20:57:55 -04:00
daniel 1044f5418f Merge Ψ-W1-T3: publish the Media Explorer import into the Media Explorer action section as well as Main 2026-08-01 20:37:30 -04:00
daniel d589b99705 Ψ-W1-T3 review remediation: soften the double-fire claim to unspecified-by-SDK, mark the unload mirror [verify — DAW], fix stale comment, drop redundant assertions, dedupe hook-partitioning comments, drop what-comment; file the action_registry test-seam deferral. 2026-08-01 20:13:13 -04:00
daniel fe3ac79ab5 Ψ-W1-T4: resolve drop targets per move, not once — every surface gets a defined outcome, a cue, and no silent no-op 2026-08-01 19:43:37 -04:00
daniel 72b870459c Publish the Media Explorer import into REAPER's Media Explorer action section as well as Main — custom_action + hookcommand2, second forever-stable id, one handler. 2026-08-01 19:43:27 -04:00
daniel 9c234c2e6b Ψ-W1-T2: disjoint per-mode solo surfaces and a playback-gated mode switch
Solo is cached, cleared and replayed per mode on a real switch only; the switch is refused visibly while the transport runs. The footer segment now routes through the activate actions, so a panel switch finally persists.
2026-08-01 19:43:18 -04:00
daniel 5e3ea6c851 capture: render a ranged item capture time-bounded — the selected-items source can't narrow a window, only a full-extent one uses it
sourceModeForScope now takes the item extent vs. the requested window. Full-extent item captures and the batch keep the old path unchanged.
2026-08-01 19:43:06 -04:00
daniel 60308a3655 Ξ-W2-T1: the resample bake chain — instrument renders, extension banks, one click re-points and resets 2026-08-01 16:26:28 -04:00
daniel 7f70d94228 tracking: one ledger, one authority — prune protection and replace-vs-add answered from the same records, fail-safe on unreadable state 2026-07-30 19:44:11 -04:00
daniel 5420550ff3 fix: correct app CMakeLists comments, drop dead bridge_marshal link edge 2026-07-30 12:40:37 -04:00
daniel 68765da031 build: route every core/ TU into reaper_reasampler through exactly one static-library link edge, never a direct compile 2026-07-30 12:28:39 -04:00
daniel d04045be69 build: split the 1423-line root CMakeLists into per-directory files with shared target helpers 2026-07-30 12:19:18 -04:00
daniel 0800760833 Fix drag-out losing audio and FX-container drops losing the capture; re-home ingest under shell/actions 2026-07-29 23:49:53 -04:00
daniel 58c6d49261 Cut shell/actions, bank_ops, app comment bloat ~48% (comments only, zero code change) 2026-07-29 20:49:31 -04:00