Compare commits

..

No commits in common. "wkt/land_fsp_sync" and "main" have entirely different histories.

266 changed files with 2338 additions and 28295 deletions

View File

@ -1,78 +0,0 @@
---
name: piker-fsp-expert
description: Piker financial signal processing expertise. Apply when changing or debugging FSP definitions, Cascade execution, derived shared-memory arrays, history/realtime handoff, backfill synchronization, FSP graph composition, online financial algorithms, or FSP chart lifecycle and scaling.
---
# Piker FSP Expertise
Treat financial signal processing as causal, revision-aware processing of
event-time market data, not as generic batch array manipulation.
## Start Here
1. Identify the source `Flume`, timeframe, event clock, and writer actor.
2. State whether the operation is historical bootstrap, backfill revision,
current-sample mutation, or sample-step append.
3. Record source and destination absolute SHM bounds before reasoning from
lengths or `ShmArray.index`.
4. Separate compute, SHM publication, IPC wakeup, and graphics costs before
optimizing.
5. Preserve causality, warm-up behavior, missing-data semantics, and output
schema in every implementation and test.
Read [architecture.md](architecture.md) before modifying runtime code. Read
[signal-processing.md](signal-processing.md) before adding an algorithm. Use
[roadmap-and-research.md](roadmap-and-research.md) for design and dependency
work, and [test-map.md](test-map.md) for verification.
## Core Contracts
- `piker.data` owns provider ingestion, source `Flume`s, history publication,
quote fan-out, OHLC mutation, and `samplerd` step events.
- `piker.fsp` owns operator declarations, historical computation, realtime
state updates, and destination SHM publication.
- `piker.ui._fsp.FspAdmin` currently owns FSP worker selection, destination
allocation, cascade lifetime, and chart attachment.
- One logical writer owns each destination FSP array.
- Output event time and absolute indices derive from the source series, not
task scheduling or quote arrival order.
- A source prepend changes historical provenance even when the current sample
index is unchanged. Recompute or patch the affected destination range.
- One-step destination lag is normal between source sampling and destination
append. A destination lead or unexplained bound shift is not.
- Backfill notifications are source and timeframe specific even though the
current sampler transport broadcasts them globally.
- Only graphics backed by a rebuilt destination should refresh. Never use an
FSP update as an unconditional whole-chart redraw request.
## Editing Workflow
1. Trace the operator from `@fsp` declaration through `cascade()`,
`connect_streams()`, destination SHM, `FspAdmin`, and every consuming
`Viz`.
2. Write an event-order table covering source writes, bound publication,
notifications, cancellation, recomputation, and destination publication.
3. Add deterministic tests for history and realtime phases independently.
4. Add a controlled interleaving test for every race fix.
5. Profile representative visible ranges and feed rates before adding a
dependency or native boundary.
## Companion Skills
- Use `piker-conc-expert` for Tractor contexts, actor topology, cancellation,
service ownership, and cross-actor state.
- Use `timeseries-optimization` for NumPy/Polars lookup and vectorization.
- Use `piker-profiling` for distributed performance measurements.
- Use `pyqtgraph-optimization` for renderer and graphics batching work.
- Use `piker-clearing-expert` when a derived signal reaches EMS or accounting.
## Do Not
- Do not hide a historical mismatch merely to stop a resync log.
- Do not infer a complete synchronization state from array length alone.
- Do not run Python loops over historical numeric arrays without evidence.
- Do not apply regular-grid DSP blindly to irregular tick arrival times.
- Do not introduce an external stream runtime that duplicates Tractor before
measuring an isolated protocol or kernel prototype.
- Do not route strategy signals directly around EMS order-state, risk, or
accounting contracts.

View File

@ -1,117 +0,0 @@
# FSP Runtime Architecture
## Current Topology
```text
provider backend
-> datad feed bus
-> source Flume
-> rt ShmArray (typically 1 second)
-> hist ShmArray (typically 60 seconds)
-> samplerd sample and backfill events
chart actor
-> FspAdmin
-> chart.fsp_N worker actor
-> cascade()
-> connect_streams()
-> destination Flume / ShmArray
-> targeted Viz updates
```
The source path is implemented mainly by `piker.data.feed`,
`piker.data.flows`, `piker.data._sampling`, and `piker.tsp._history`. The FSP
path is implemented by `piker.fsp._api`, `piker.fsp._engine`, and operator
modules such as `_volume` and `_momo`. Worker and graphics lifecycle currently
live in `piker.ui._fsp` and `piker.ui._display`.
## Operator Contract
`@fsp` wraps an async generator. Its first yield is historical output computed
from the source array. Later yields are `(field, value)` realtime mutations.
`connect_streams()` converts the first yield into the destination structured
array, aligns time and absolute bounds, then updates the current destination
row from realtime quote events.
`samplerd` owns row advancement. At a new source sample, the cascade copies or
zeros the previous destination row and aligns destination timestamps to the
source. Quote-time mutations and sample-time appends are distinct operations.
## SHM Synchronization
Treat each array as an absolute half-open range:
```text
[first, last)
```
For a normal source history prepend:
```text
before: src=[10000, 12000) dst=[10000, 12000)
after: src=[ 8000, 12000) dst=[10000, 12000)
```
The current step is unchanged, but destination history is missing 2,000 rows.
The destination must be recomputed or incrementally repaired.
For a normal one-step source lead:
```text
src=[8000, 12001) dst=[8000, 12000)
```
The cascade should append exactly one destination row, not recompute history.
Use `_first.value` and `_last.value` for synchronization. `ShmArray.index` is
the last bound modulo capacity and is not a complete absolute position.
Snapshot bounds once per check; shared first and last counters do not form a
transactional pair.
## Backfill Event Ordering
1. `start_backfill()` receives an older provider frame.
2. `shm_push_in_between()` writes the frame and publishes the earlier first
bound.
3. `notify_backfill()` asks `samplerd` to broadcast the market and timeframe.
4. Relevant cascades treat the event as a history revision. A prepend changes
bounds; an in-place gap repair may not.
5. Each cascade cancels its current compute task and waits for completion.
6. The historical phase recomputes against the newest readable source range.
7. Destination bounds are published and the UI receives an `fsp_update`.
8. Only `Viz` objects backed by that destination token redraw.
Multiple provider frames may arrive while a recomputation runs. The cascade
must reject mixed-bound bootstrap results, replay in-place revisions, recheck
bounds after restart, and converge on the newest source range.
## Actor-Local State
`Fsp._flow_registry`, attached SHM handles, caches, and nurseries are local to
each actor. Never assume worker siblings share warmed registries or Python
objects. Graph dependencies currently rely on startup order, as demonstrated
by `flow_rates` starting only after `dolla_vlm`.
## Cross-Subsystem Boundaries
- Data ingest should normalize identity and event-time semantics before an FSP
consumes a stream.
- UI renderers should consume revision/range information without controlling
computation correctness.
- Clearing may consume a derived signal, but EMS remains responsible for order
intent, status, routing, fills, positions, and accounting.
- FSP output exposed as a feed needs the same fan-out, identity, lifecycle,
and backpressure contracts as provider feeds.
## Architectural Gaps
The current protocol infers history revisions from shared bounds. A future
revision message should carry at least:
```text
(fqme, timeframe, revision, first, last, start_ts, end_ts)
```
Destination publication should identify the source revision used. This allows
coalescing, range repair, stale-result rejection, and deterministic graph
dependency startup without turning every prepend into a global wakeup.

View File

@ -1,104 +0,0 @@
# Roadmap and Research
Issue states change. Re-read each tracker before planning work and preserve
human ownership of issue and checklist state.
## Guiding Roadmap
- [piker #270](https://github.com/pikers/piker/issues/270): typed,
stream-composable FSP graph and DSL direction.
- [piker #107](https://github.com/pikers/piker/issues/107): cross-actor shared
memory tick/ring-buffer direction.
- [piker #216](https://github.com/pikers/piker/issues/216): exposing FSP output
through feed-like interfaces.
- [piker #325](https://github.com/pikers/piker/issues/325): faster streaming
y-range computation and overlap with FSP reductions.
- [piker #536](https://github.com/pikers/piker/issues/536): Arrow/Parquet data
ingest and columnar boundaries.
- [piker #98](https://github.com/pikers/piker/issues/98): older realtime feed
architecture discussion. Treat it as likely superseded; compare every idea
against current `Flume`, feed-bus, history, and sampler code.
- [Tractor #339](https://github.com/goodboy/tractor/issues/339): shared-memory
array and localhost IPC evolution.
- [hotbaud issues](https://github.com/guilledk/hotbaud/issues): experimental
high-throughput local transport and ring-buffer work.
## Near-Term Protocol Research
Prototype an explicit source revision envelope and measure it with two FSPs
and two overlaid feeds. The prototype should prove source-specific wakeups,
coalescing of consecutive prepends, stale-result rejection, and targeted UI
range updates. Keep the protocol independent of the final IPC transport.
Characterize historical repair strategies:
1. Full replay from the new first bound.
2. Prefix-only calculation for pointwise operators.
3. Replay from an operator checkpoint for stateful recurrences.
4. Bounded overlap repair for finite-window operators.
An FSP declaration eventually needs repair metadata describing locality,
state checkpointability, warm-up/overlap, and whether old inputs can alter all
later outputs.
## Candidate Libraries
Candidates are experiments, not dependency recommendations.
### Bounded Near-Term Prototypes
- [Bottleneck](https://github.com/pydata/bottleneck): compare rolling min/max,
rank, and moving-window kernels against NumPy/Numba for chart y-ranges and
finite-window FSPs.
- [River](https://github.com/online-ml/river): study stateful online estimator
APIs, drift detectors, and reference semantics. Do not assume its object
model belongs in the hottest tick path.
- [nanoarrow](https://github.com/apache/arrow-nanoarrow): evaluate lightweight
Arrow C data interfaces at native/process boundaries already motivated by
`pyarrow` and piker #536.
- [DuckDB](https://github.com/duckdb/duckdb): use for historical Parquet query,
replay slicing, and validation, outside realtime tick execution.
### Watchlist
- [hotbaud](https://github.com/guilledk/hotbaud): track API and correctness
maturity for localhost SHM/ring transport.
- [iceoryx2](https://github.com/eclipse-iceoryx/iceoryx2): study zero-copy
pub/sub ownership and lifecycle, but avoid competing with Tractor without a
narrow adapter benchmark.
- [DataFusion Python](https://github.com/apache/datafusion-python): Arrow-native
historical query and expression execution if existing Polars paths cannot
meet a demonstrated need.
- [Feldera](https://github.com/feldera/feldera): study DBSP/incremental-view
semantics for revision-aware graphs; its service/SQL runtime is not a direct
piker integration target.
- [Bytewax](https://github.com/bytewax/bytewax) and
[Arroyo](https://github.com/ArroyoSystems/arroyo): study dataflow semantics,
watermarking, and recovery, not as replacements for the actor tree.
GPU stacks such as JAX and CuPy are poor defaults for low-batch per-tick work
because transfer, compilation, and scheduling can dominate. Reconsider only
for measured large-batch research or many-market matrix workloads.
## Benchmark Design
Report separate distributions for:
- provider decode and normalization;
- source SHM write and bound publication;
- historical kernel and realtime recurrence;
- actor notification and stream fan-out;
- destination publication;
- visible-range formatting and Qt paint.
Use p50, p95, p99, maximum, allocation rate, and resident-memory growth. Test
steady state, burst load, backfill overlap, cancellation, and slow consumers.
Throughput without event-time correctness and bounded teardown is not a win.
## Medium-Term Sequence
First establish revisioned range messages and operator repair semantics. Then
decouple FSP orchestration from chart ownership and expose destination flows
through feed-like subscriptions. Next benchmark localhost transport adapters
behind the same protocol. Only after those contracts stabilize should graph
placement, fusion, worker scaling, and remote partitioning be automated.

View File

@ -1,103 +0,0 @@
# Financial Signal Processing
## Model the Clock First
Market inputs can be quote events, trades, book changes, fixed-time bars, or
revised historical bars. Name both event time and processing time. For
irregular events, either use time-aware recurrences or define the resampling
and interpolation policy explicitly.
Never let backfilled information leak into a value presented as historically
tradable. Distinguish corrected analysis history from the signal values that a
live strategy could have observed.
## Useful Technique Families
### Online Moments and Robust Statistics
Use Welford/Chan recurrences for stable online moments, compensated summation
for long accumulations, and mergeable summaries when partitioning actors.
Rolling median, quantiles, MAD, and winsorized measures are useful under heavy
tails, but exact rolling order statistics need more state than moments.
### Time-Aware Exponential Filters
For irregular event intervals, derive decay from elapsed time:
```text
alpha(dt) = 1 - exp(-dt / tau)
y_t = y_prev + alpha(dt) * (x_t - y_prev)
```
Record initialization, session reset, gap, and backfill behavior. A fixed
sample alpha is valid only after an explicit regularization policy.
### State-Space Models
Kalman and robust state-space filters fit latent price, spread, volatility,
and lead/lag estimates. Prefer square-root or Joseph-form covariance updates
when conditioning is poor. State revisions after backfill require replay from
a checkpoint or a bounded smoother; silently prepending observations without
replaying state is incorrect.
### Point Processes and Microstructure
Trade and quote arrivals are events, not merely sampled amplitudes. Consider
intensity, duration, signed order flow, imbalance, spread, queue change, and
self-excitation models. Validate against provider-specific aggregation and
duplicate/out-of-order behavior before interpreting a statistic.
### Volatility and Covariance
Realized variance, bipower variation, EW covariance, and range estimators can
be updated online. Asynchronous cross-market covariance needs synchronization
or estimators designed for non-synchronous observations; naive row alignment
creates lead/lag and Epps-effect artifacts.
### Spectral and Multiscale Methods
FFT methods assume a regular grid and a window. Declare detrending, tapering,
overlap, normalization, and latency. For tick data, resample deliberately or
use irregular-time methods. Wavelets and multiresolution filters can separate
horizons, but boundary handling and causal delay must be visible to strategy
code.
### Change and Anomaly Detection
CUSUM, Page-Hinkley, sequential likelihood ratios, and robust z-scores are
cheap online tools. Calibrate false alarms under dependence and regime shifts;
do not treat IID thresholds as market guarantees.
## Numerical Implementation Order
1. Establish a scalar reference with explicit state and semantics.
2. Vectorize historical bootstrap with NumPy.
3. Keep realtime updates as O(1) recurrences where possible.
4. Use Numba for measured numeric kernels with stable dtypes and no Python
object traffic.
5. Use Polars or Arrow for columnar historical transforms and interchange,
not automatically for each tick.
6. Partition by independent market/operator state before adding threads or a
new distributed runtime.
## Existing Stack
- NumPy 2.x: canonical dense and structured-array kernels.
- Numba: compiled CPU loops and recurrences that do not vectorize cleanly.
- Polars: parallel/lazy columnar history preparation and validation.
- PyArrow: columnar interchange, storage, and ingest boundaries.
- Tractor/Trio: structured distributed execution and lifecycle.
- PyQtGraph: visible-range rendering, not a compute scheduler.
## Algorithm Acceptance Criteria
Every new FSP should define:
- input fields, clock, ordering, duplicate, gap, and revision policy;
- output dtype, units, identity, and parameterization;
- warm-up length and initialization bias;
- causal latency and strategy-visible publication point;
- historical/realtime equivalence tolerance;
- reset and session-boundary behavior;
- complexity, state size, and representative throughput;
- replay behavior when historical source data changes.

View File

@ -1,52 +0,0 @@
# FSP Test Map
## Operator Tests
- Historical output schema, dtype, units, and source timestamp alignment.
- Realtime update equivalence with the final row of a batch recomputation.
- Warm-up, NaN, zero, gap, duplicate, and out-of-order behavior.
- Session reset and elapsed-time behavior.
- Parameter and source identity isolation.
## Engine Tests
- Equal absolute source/destination bounds.
- Normal one-step destination lag.
- Destination lead and lag greater than one step.
- Source prepend by one, two, and a full provider frame.
- In-place gap repair with unchanged source bounds.
- Equal lengths with shifted absolute bounds.
- Source change during historical recomputation and convergence afterward.
- Cancellation while the quote stream, sample stream, or compute generator is
blocked.
- Foreign FQME and timeframe backfill broadcasts.
## Graph and Actor Tests
- Dependency startup before downstream operator lookup.
- Actor-local flow registry isolation.
- One destination writer and multiple readers.
- Consumer cancellation without orphaned cascade or worker tasks.
- Worker failure propagation and SHM cleanup.
- Feed-like fan-out with a slow or disconnected consumer.
## UI Tests
- An `fsp_update` redraws every field visualization sharing its destination
token and no unrelated market, OHLC, or FSP visualization.
- A prepend outside the visible range does not move or re-range the live view.
- A revised value inside the visible range invalidates only required formatter
and graphics caches.
- Hidden charts defer rendering without stopping compute correctness.
## Performance Tests
- Historical bootstrap across realistic SHM capacities.
- O(1) realtime recurrence cost under bursty quotes.
- Consecutive provider-frame prepends with two or more cascades.
- Multi-market sampler broadcasts and targeted wakeups.
- Visible-range redraw cost at several units-per-pixel levels.
Prefer deterministic local arrays and controlled Trio task interleavings.
Use live providers only as a separate integration layer because credentials,
rate limits, sessions, and changing history make poor regression oracles.

View File

@ -47,7 +47,7 @@ that's why it's like 1000x faster ya know?"
- Wrap all code symbols: `function()`, - Wrap all code symbols: `function()`,
`ClassName`, `field_name` `ClassName`, `field_name`
- File paths: `piker/ui/_remote_ctl.py` - File paths: `piker/ui/_remote_ctl.py`
- Commands: `git status`, `piker store shm` - Commands: `git status`, `piker store ldshm`
**Explain like you're pair programming:** **Explain like you're pair programming:**
``` ```

View File

@ -1 +0,0 @@
../../.agents/skills/piker-fsp-expert

View File

@ -1,143 +0,0 @@
# Split `brokerd.<broker>` into trading-only `brokerd` + new `datad.<broker>`
## Context
Today a single `brokerd.<broker>` actor hosts BOTH concerns:
- **data feed service tasks**: the `_FeedsBus` + `open_feed_bus()` ep (`piker/data/feed.py:464`), per-symbol `allocate_persistent_feed()` tasks (shm writers via `sample_and_broadcast()`, history backfill via `piker.tsp`), plus backend eps `stream_quotes`, `open_history_client`, `open_symbol_search`, `get_mkt_info` from `piker/brokers/<backend>/feed.py`/`symbols.py`,
- **live order-control tasks**: `open_trade_dialog()` from `piker/brokers/<backend>/broker.py`, driven by `emsd`.
The codebase already anticipates this split: `piker/data/validate.py:70-91` groups backend eps into `'datad'` vs `'brokerd'` kinds, `piker/brokers/ib/__init__.py:62-70` already declares `_brokerd_mods`/`_datad_mods`, and `piker/brokers/_daemon.py:62` carries the literal TODO *"rename the daemon to datad prolly once we split up broker vs. data tasks into separate actors?"*. This work executes that split.
**User-decided constraints:**
- `datad.<broker>` is a **sibling** of `brokerd.<broker>` under `pikerd`, spawned via the existing `Services` + `maybe_spawn_daemon()` machinery (`piker/service/_daemon.py:46`).
- **Hard cutover, staged by layer** — no dual-mode runtime flag; every stage lands fully working.
- Post-split `brokerd` is **trading-only and lazily spawned solely by emsd's** `open_brokerd_dialog` path; UI/CLI/feed code never spawns it. Chart-only and paper sessions run with **zero** brokerd processes.
## Target topology
```
pikerd
├── datad.ib feed bus, shm writers, tsp history, symbol search
├── brokerd.ib open_trade_dialog only (EMS-spawned, lazy)
├── emsd
│ └── paperboi.ib (paper mode; opens its feed via datad)
└── samplerd
```
## Key verified facts (load-bearing)
- `feed.portals` has exactly ONE trading consumer: `piker/clearing/_ems.py:671` (`portal = feed.portals[brokermod]` → handed to `open_brokerd_dialog`). This is the single coupling forcing feed + trading into one actor.
- `assert 'brokerd' in servicename` at `piker/data/feed.py:502` is the only actor-name assert in the tree.
- `piker ledger` (`piker/accounting/cli.py:100-157`) is a hidden consumer: it calls `broker_init()` directly, spawns its own ad-hoc actor, enters `_setup_persistent_brokerd`, then calls `open_brokerd_dialog(brokermod, portal, ...)` **with an explicit portal** — the new signature must keep a `portal:` override param.
- kraken's feed→broker coupling is mild: `kraken/broker.py:77-81` imports `NoBsWs`/`open_autorecon_ws` (really from `piker.data._web_bs`) and `stream_messages` (pure parser) from `feed.py`. In-process imports only — `enable_modules` gates RPC, not imports. No live shared state; each actor opens its own WS.
- ib: default `client_id=6116` (`ib/api.py:1320`) with linear retry `clientId=client_id + i` (`:1403`). Two ib actors connecting concurrently will collide → needs a role-based id offset. Both daemons need `_spawn_kwargs={'infect_asyncio': True}` (copied by both init fns).
- binance has no `symbols.py` (`get_mkt_info`/`open_symbol_search` live in `binance/feed.py`); kucoin/questrade/robinhood are flat single-file with NO `__enable_modules__`; deribit has no `broker.py`.
- `tests/test_services.py:80,185` assert `brokerd` actor names incl. the paper-mode flow asserting `brokerd.kraken` spawns (`:229,:237`) — must invert post-split.
- `_root_modules` (`piker/service/_actor_runtime.py:156`) must gain the new datad daemon mod so `pikerd_portal.run(spawn_datad, ...)` resolves.
- `piker/brokers/core.py:175` `symbol_search` does `portal.run(search_w_brokerd)` where the target fn lives in `piker.brokers.core` → that mod must stay RPC-enabled in datad.
## Module decision
New file **`piker/data/_daemon.py`** hosts all datad machinery (`_setup_persistent_datad`, `datad_init`, `spawn_datad`, `maybe_spawn_datad`, `_datad_service_mods`). Mirrors the `piker.brokers._daemon` / `piker.data._sampling` (samplerd) per-subsystem convention and satisfies the existing TODO at `brokers/_daemon.py:49` ("move this def to the `.data` subpkg"). `piker.brokers._daemon` keeps brokerd-only code, slimmed. Do NOT over-DRY the two ~40-line init fns into a shared factory yet (samplerd precedent accepts the duplication).
---
## Stage 0 — prep: backend module grouping + ep introspection (no topology change)
1. `piker/data/validate.py`: add a `get_eps(mod, kind) -> dict[str, Callable]` helper returning the backend's defined eps for a daemon-kind from `_eps` (missing eps excluded). Used later by both init fns + fail-fast checks.
2. Declare daemon-kind module groups in each split backend's `__init__.py`, keeping `__enable_modules__` as the (deduped) union so behavior is unchanged:
- `kraken`: `_brokerd_mods = ['api', 'broker']`, `_datad_mods = ['api', 'feed', 'symbols']`
- `binance`: `_brokerd_mods = ['api', 'broker']`, `_datad_mods = ['api', 'feed']`
- `deribit`: `_brokerd_mods = []`, `_datad_mods = ['api', 'feed']`
- `ib`: adjust existing groups so each includes `'api'`
- flat backends (kucoin etc.): no attrs — init fns fall back to enabling just `modpath` (existing behavior).
3. Hygiene: `kraken/broker.py:77-81` imports `NoBsWs`/`open_autorecon_ws` from `piker.data._web_bs` directly (keep the `stream_messages` import from `.feed`).
**Gate**: full pytest green, zero behavior change.
## Stage 1 — introduce `datad` daemon machinery (additive)
New `piker/data/_daemon.py`:
- `_datad_service_mods: list[str]` — datad-always-enabled mods (successor to the data side of `_data_mods` from `brokers/_daemon.py:52`): `['piker.brokers.core', 'piker.brokers.data', 'piker.data', 'piker.data.feed', 'piker.data._sampling', 'piker.data._daemon']`.
- `_setup_persistent_datad(ctx, brokername, loglevel)` — `@tractor.context` fixture; logging boilerplate (as `_setup_persistent_brokerd:81-88`), then allocates the actor-global feed bus exactly as the brokerd fixture does today (`brokers/_daemon.py:105-121`): `assert not feed._bus`, open service nursery, `feed.get_feed_bus(brokername, service_nursery)`, `ctx.started()`, `sleep_forever()`.
- `datad_init(brokername, ...)` — mirrors `broker_init()` (`brokers/_daemon.py:132`): actor name `f'datad.{brokername}'`, copies backend `_spawn_kwargs` (**critical for ib infect_asyncio**), builds `enable_modules` from `getattr(brokermod, '_datad_mods', getattr(brokermod, '__enable_modules__', []))`.
- `spawn_datad(brokername, ...)` — mirrors `spawn_brokerd()` (`brokers/_daemon.py:202`): `Services.actor_n.start_actor(dname, enable_modules=_datad_service_mods + backend_mods, ...)` + `Services.start_service_task(dname, portal, _setup_persistent_datad, ...)`.
- `maybe_spawn_datad(brokername, ...)` — wraps `maybe_spawn_daemon(service_name=f'datad.{brokername}', service_task_target=spawn_datad, ...)` exactly like `maybe_spawn_brokerd` (`brokers/_daemon.py:256`).
Supporting edits:
- `piker/service/_actor_runtime.py:156`: add `'piker.data._daemon'` to `_root_modules` (keep `'piker.brokers._daemon'`).
- `piker/service/__init__.py`: re-export `spawn_datad`/`maybe_spawn_datad` next to the brokerd ones (`:56-57`).
- `tests/test_services.py`: new `test_datad_spawn` — `open_test_pikerd` + `maybe_spawn_datad('kraken')` + `ensure_service('datad.kraken')`. (Do NOT route a feed through it yet — `open_feed_bus`'s assert still says brokerd.)
**Gate**: suite green + new spawn test; nothing routes through datad yet.
## Stage 2 — clearing layer: emsd self-spawns brokerd (decouple from `feed.portals`)
Sequenced BEFORE the feed cutover so live trading works at every boundary (otherwise `feed.portals` would hand emsd a datad portal).
`piker/clearing/_ems.py`:
1. `open_brokerd_dialog()` (`:336`) new signature: `(brokermod, exec_mode, fqme=None, portal: tractor.Portal|None = None, loglevel=None)`. Internals: keep trades-ep detection (`:400-417`); acquire the brokerd actor ONLY when a live ep will actually open, via a small inner `@acm _acquire_live_portal()` that yields the passed `portal` if given (the `piker ledger` path) else `async with maybe_spawn_brokerd(brokermod.name, loglevel=loglevel)` — **the single place brokerd boots post-split**. Move the eager `portal.open_context(trades_endpoint, ...)` construction (`:425`) inside that block; the paper short-circuit (`:432`) never touches it.
2. `Router.maybe_open_brokerd_dialog()` (`:581`) — drop the `portal` param; `Router.open_trade_relays()` (`:640`) — delete `portal = feed.portals[brokermod]` (`:671`) and the pass-through (`:685`).
3. `piker/accounting/cli.py:144`: switch to keyword form `open_brokerd_dialog(brokermod, exec_mode=..., portal=portal, loglevel=...)`.
Pre-cutover this is a pure refactor: emsd's `maybe_spawn_brokerd` just *finds* the already-running brokerd via the registry.
**Gate**: `tests/test_ems.py` + services suite green; manual paper order on kraken; `piker ledger` smoke.
## Stage 3 — feed layer hard cutover to datad
1. `piker/data/feed.py`: import `maybe_spawn_datad` (replacing `maybe_spawn_brokerd`, `:56-58`); `open_feed()` `:895` → `maybe_spawn_datad(...)` (rename `brokerd_ctxs` → `datad_ctxs`); `open_feed_bus()` `:502` → `assert 'datad' in servicename`; comment sweep.
2. `piker/brokers/_daemon.py` `_setup_persistent_brokerd()`: slim to trading-only fixture — logging setup, `ctx.started()`, `sleep_forever()`. Drop the bus alloc, `assert not feed._bus`, the service nursery (backend `open_trade_dialog` ctxs own their task trees), the `eg.ExceptionGroup` handler, and the stale `_FeedsBus` TYPE_CHECKING import. (`piker ledger`'s ad-hoc actor enters this same slimmed fixture — exactly what it needs.)
3. Repoint remaining data-flavored spawn sites `maybe_spawn_brokerd` → `maybe_spawn_datad`:
- `piker/ui/_app.py:40,57` (symbol search; optionally rename `install_brokerd_search` → `install_datad_search`, def at `piker/data/feed.py:766`)
- `piker/brokers/core.py:33,175` (`symbol_search`)
- `piker/brokers/cli.py:38,190,320` (`brokercheck`, `record`)
- `piker/ui/cli.py:30,72,121` (legacy kivy monitor/optschain) + best-effort `piker/ui/kivy/option_chain.py:498` `wait_for_actor('brokerd')` → `'datad'`
4. `tests/test_services.py`: `:80,:185` `actor_name = 'datad'`; paper-mode test asserts `datad.kraken` + `paperboi.kraken` + `emsd` AND adds the headline negative check: `assert 'brokerd.kraken' not in services.service_tasks`.
**Gate**: full suite with inverted assertions; manual: chart on binance/kraken shows `datad.<broker>` in `piker services` and NO brokerd; paper order fills; symbol search works; history backfill sane.
## Stage 4 — caps-sec slimming, validation, ib client-id
1. `piker/brokers/_daemon.py`: delete `_data_mods` (`:52`); `spawn_brokerd()` `:236` → `enable_modules=tractor_kwargs.pop('enable_modules')`; `broker_init()` `:184` reads `getattr(brokermod, '_brokerd_mods', getattr(brokermod, '__enable_modules__', []))`. Resulting brokerd enable set has no `piker.data.*` at all.
2. Fail-fast ep validation via Stage-0 `validate.get_eps()`: `broker_init` raises a clear error when `get_eps(mod, 'brokerd')` is empty (e.g. "kucoin is datad-only — use paper mode"); `datad_init` warns analogously.
3. ib client-id mitigation in `load_aio_clients()` (`ib/api.py:~1316`): role-based offset when `client_id` is the 6116 default (e.g. `+16` when `'datad' in tractor.current_actor().name`), optionally configurable via `brokers.toml` keys.
4. Docs/comment sweep: resolve `brokers/_daemon.py:62` TODO, `piker/cli/__init__.py:253` TODO, daemon list in `piker/service/__init__.py:21` docstring.
**Gate**: full suite; audit greps clean: `grep -rn maybe_spawn_brokerd piker/` hits only `clearing/_ems.py`, `service/__init__.py`, `brokers/_daemon.py`; no spawn-path `'brokerd'` literals left in `piker/data`, `piker/ui`, `piker/brokers/cli.py`.
---
## Risk register
| Risk | Sev | Mitigation |
|---|---|---|
| ib: `datad.ib` + `brokerd.ib` both connect to TWS/gw; `client_id=6116` collision burns `connect_timeout` retries | High (live ib only) | Stage 4 role-based id offset; both daemons inherit `infect_asyncio` via `_spawn_kwargs` copy in both init fns |
| `piker ledger` ad-hoc actor path silently broken by signature change | Med | `portal:` override kept on `open_brokerd_dialog` (Stage 2); explicit smoke test |
| datad-only backends (kucoin/deribit) → useless brokerd spawn | Low | already covered: `open_brokerd_dialog` forces `exec_mode='paper'` when no trades ep (`_ems.py:413-417`) BEFORE the lazy spawn; Stage 4 fail-fast |
| brokerd in-process symbology/ledger needs (ib `broker.py` imports `.symbols`; kraken uses `stream_messages`) | None | verified pure in-process imports; `enable_modules` gates RPC only |
| paper-mode test asserts brokerd spawns (`test_services.py:237`) | Low | Stage 3 deliberately inverts it |
| doubled per-broker clients (HTTP/WS, symcache loads) | Low | each actor already opens its own WS today; symcache is disk-read-mostly |
## Verification (per stage gates above, plus end-to-end)
- `pytest tests/test_services.py tests/test_feeds.py tests/test_ems.py` at every stage (kraken/binance public endpoints, no creds needed).
- Manual smoke matrix post-Stage-3: `pikerd -l info` + `piker chart btcusdt.spot.binance` → `piker services` shows `datad.binance`, no brokerd; submit paper order → `paperboi.binance` appears, still no brokerd; symbol search; `piker ledger <broker>.paper`.
- If ib creds/gateway available: live ib chart + small live order to exercise dual-actor client-id path (Stage 4).
## Critical files
- **new** `piker/data/_daemon.py` — all datad machinery
- `piker/brokers/_daemon.py` — slimmed brokerd fixture/init/spawn
- `piker/data/feed.py` — spawn cutover + actor-name assert
- `piker/clearing/_ems.py` — emsd lazy brokerd spawn (`open_brokerd_dialog`, `Router`)
- `piker/service/_actor_runtime.py`, `piker/service/__init__.py` — root mods + re-exports
- `piker/data/validate.py` — `get_eps()` helper
- backend `__init__.py`s (kraken/binance/deribit/ib) — `_datad_mods`/`_brokerd_mods`
- `piker/accounting/cli.py` — keyword-form dialog call
- `tests/test_services.py` — inverted + new assertions

View File

@ -1,233 +0,0 @@
# Bounded IB recovery candidate
Prepared 2026-10-22 on `wkt/land_ib_recovery`, in the main checkout
`/home/goodboy/repos/piker`. Base:
`dc618229bd8c9c90aa3160e26677b553a7d472c7`.
Changes are unstaged and uncommitted. No acceptance is implied.
## Patch selection and provenance
Keep these existing landing commits:
| Commit | Reason |
| --- | --- |
| `4133f073` and its ancestors | Packaging, Nix, runtime API and pin work |
| `e8e0730a` | Correlate MethodProxy requests and responses |
| `7bef50b0` | One response reader, per-request waiters |
| `d1571ddf` | Propagate daemon-handler exceptions |
| `dc618229` | Isolate cached feed stream consumers |
Proposed separate implementation boundaries, in this order:
1. Back out the code and original test from
`635314c757c74a1cd95c5cf3dc7dfd03be984b00`.
This restores the direct-owner `_quote_streams` implementation.
Retain the historical prompt records. Replace the removed
mocked-cache regression with the real-bridge restart reproducer.
This backout is not a claim that arbitrary concurrent direct
callers have safe broadcast ownership.
2. Backport **only** `piker/brokers/ib/feed.py` from
`0846cbd4a4197e47415689cf5e9d146ba3174c32`.
Retain its seven history-adapter regressions in
`tests/test_ib_history.py`. This includes reset serialization,
completion/result signalling and bounded cancellation/failure.
It does not include the original SHM/storage orchestration patch.
3. Backport **only** `piker/brokers/ib/feed.py` from
`1da9b760a3bc6b96501110cab2b5d09846d42a1d`.
Retain `test_empty_history_preserves_request_boundary` and its
six cases. This preserves no-data request boundaries and wakes
the waiter. It does not include the original history-layer
endpoint/filtering changes. Add the real-bridge history recovery
case after these adapter changes are present.
4. Import the code and original regression from
`12b0d66e854247e92cb5a8668be71d816c72d856`.
Add a separate real-broadcaster/production-loop regression.
5. Remove the single-null-sample debugger stop in `get_null_segs()`.
This is a local UX cleanup retaining the existing `None` result.
It does not backport the null-grouping/deduplication rewrite in
`976db6c33c4b8ed07eee3047226417d3928d9cca`.
6. Record cross-boundary regression results and live qualification
evidence, distinguishing the known callback failure from passes.
All three imported source commits were authored by
`goodboy <goodboy_foss@protonmail.com>`; preserve that authorship
and the full source SHA when committing each partial backport.
Use explicit partial-backport wording, not a whole-commit cherry
claim. New real-runtime tests are this session's additional work.
The current aggregate working diff contains these boundaries in
the same feed file. Staging that entire file would combine the
first three boundaries. Reconstruct/stage the original per-source
patches in the above order when preparing commits.
## Quote-cache history correction
`4af885d90f6e4bcfa963f2f7a1d40436d3c928f3` and landing commit
`635314c757c74a1cd95c5cf3dc7dfd03be984b00` have the same stable
patch ID: `bc54892c25e957be112b135bb9e383ec4dd123c5`.
They are equivalent replays, not different implementations.
`wkt/ib_downstream_pin_eval@e929fdc2` lacks that quote-cache
change and retains direct ownership. The original qualification
did not validate the combination assembled for the landing.
After this candidate's edits, `piker/brokers/ib/feed.py` matches
the source at `1da9b760` byte-for-byte. Compared with the evaluated
tip, only the later schedule/reservation APIs are absent.
The EMS code patch has the same stable patch ID as the original
`12b0d66e` EMS code patch:
`50fee047e94fd67c8d9feab66859fbe3ab25b64c`.
## Real-runtime evidence
`tests/test_ib_quote_restart.py` uses real TCP Tractor actors,
an asyncio-infected child, the installed pinned Tractor bridge,
production Piker quote setup, and real IB/eventkit ticker events.
Only the external IB connection/input is replaced with replay data.
Before the quote-cache backout, cancellation after ticker delivery
raised `RemoteActorError[TrioTaskExited]` through
`maybe_open_context()` / `_Cache.run_ctx` teardown, matching the
reported fatal path.
After the backout:
- Two quote generations finish; each subscription is cancelled,
each asyncio producer is done and each ticker handler removed.
- A second RPC repeats the sequence in the same actor successfully.
- The strict callback-cleanliness case still fails: Tractor's
`signal_trio_when_done()` raises `TrioTaskExited` into asyncio's
exception handler even though the Trio consumer recovers.
The history integration case keeps `Client.bars()`, MethodProxy,
both relays, `get_bars()`, `wait_on_data_reset()` and
`data_reset_hack()` real. It replaces external contract discovery,
IB history replies and VNC input. A real IB error event announces
farm recovery. The replay rejects retries before recovery and
proves one reset followed by completion for both 1s and 60s data.
The new EMS case runs the production dark-clearing loop over a
real Tractor broadcaster: both EMS and a sibling receive a tick,
EMS cancellation removes only its subscription, and the sibling
continues with an unchanged strict lag policy.
## Verification
Unique selected cases: **38 passed, 1 failed**.
The failure is intentionally visible, not skipped or xfailed.
Run from the checkout's working Nix environment:
```xsh
timeout -k 5 45 py313/bin/python -m pytest -p no:xonsh -q -x --tb=short tests/test_ib_history.py tests/test_ems_broadcast.py tests/test_ib_method_proxy.py tests/test_feed_broadcast.py
timeout -k 5 45 py313/bin/python -m pytest -p no:xonsh -q --tb=short tests/test_ib_quote_restart.py
timeout -k 5 45 py313/bin/python -m pytest -p no:xonsh -q -x --tb=short tests/test_logging.py tests/test_startup.py
timeout -k 5 45 py313/bin/python -m pytest -p no:xonsh -q -x --tb=short tests/test_services.py::test_runtime_boot
git diff --check
```
The adapter/proxy/feed/EMS selection initially passed 21 cases;
the added real EMS case then passed with its imported case.
The bridge file passed two cases and failed callback cleanliness.
Logging/startup passed 13 cases; runtime boot passed one with the
existing `Registrar.uid` deprecation warning.
The agent shell inherited another project's environment and lacked
native libraries. Tests used the existing Piker `py313` via
`/tmp/nix-shell.zfKbeB/opencode/piker_test_env.py`, which supplies
libraries from the pinned flake's existing Qt dependency closure,
excluding Nix Qt itself and glibc. No environment sync or shellHook
ran. Piker resolved to this checkout; Tractor resolved to its
installed package in `py313`.
Workflow-state preparation initially reported changed legacy
files/modes after migration. On 2026-10-23 the human authorized
reconciliation. Inspection found only two missing legacy guidance
snapshots; the neutral copies matched their saved hashes and modes.
Restoring those local snapshots resolved preparation without
changing archived messages or the original migration receipt.
## Human live qualification, 2026-10-23
The human ran these journeys with the installed Tractor pin,
the existing IB Gateway and pre-existing NativeDB. The agent did
not operate the broker connection, VNC, orders or database.
- Initial QQQ startup/backfill recovered after pacing error 162
with one reset attempt, then displayed live updates. The reset
waiter appears to have timed out; the log did not establish an
acknowledged farm reset or live quote-stream restart.
- Sequential chart exits and reattachment preserved all four
daemon PIDs; a read-only process/listener check confirmed this.
- The overlapping two-chart journey used `pikerd@961108`,
`brokerd.ib@964765`, `samplerd@965900` and `emsd@969015`.
B started at 08:16:31; C at 08:17:33. B closed at 08:18:11,
replacement B started at 08:18:24 and had FSPs up by 08:18:28,
then C closed at 08:18:37. The human reported the exercise worked;
survivor updates are a manual observation, while timestamps and
unchanged daemon identities corroborate overlap and service reuse.
- Startup in that journey hit the explicit single-null-sample
breakpoint several times before the charts overlapped. The small
debugger cleanup above was applied afterward; this live run does
not itself verify that later edit.
- Departing charts still report cancellation groups and an FSP
receiver-close `ClosedResourceError`, followed by the Qt close
handler's self-SIGINT. Peer-disconnected final result warnings
remain contained in the persistent services.
- Daemon SIGINT at 08:19:43 reproduced cached shared-client teardown:
`load_clients_for_trio()` / `_Cache.run_ctx` promoted
`TrioTaskExited` through `open_trade_dialog()` and entered PDB.
After the human continued, all three children were reported
exited with status 0 and the root terminated at 08:19:45.
Cancellation-only RPC error reporting and late cancel-ack
diagnostics also remain.
The human explicitly accepted deferring this daemon teardown fix
from #94 and requested loud attribution for the bounded backports.
This is an operational smoke result, not clean-shutdown coverage.
## Downstream reconciliation requirements
Before landing #94, the human wants an easy replay of the downstream
histories intended to follow it. Rehearse against the agreed #94
tree in disposable worktrees, then inspect range-diffs, final code
and relevant regressions. This note does not authorize rewriting
the original branches or changing acceptance markers.
- `0846cbd4` and `1da9b760` are partial backports. Their imported
adapter hunks may already be present, but their history/SHM/storage
changes and omitted integration tests remain necessary downstream.
Never drop either whole source commit merely because `feed.py`
matches or its adapter hunk becomes empty.
- `12b0d66e` has source-equivalent EMS code and its imported test,
plus an additional local real-broadcaster test. Preserve that extra
regression when resolving test-file overlap.
- The quote-cache backout intentionally restores direct ownership
for this landing. Reintroducing original `4af885d9` downstream
requires qualifying cached-owner teardown with the matching
Tractor revision; the backout is not a permanent ownership design.
- The local single-null breakpoint removal retains legacy behavior.
Prefer the full downstream `976db6c3` algorithm when it is replayed;
preserving this `return None` guard would lose that later fix.
- Real-bridge tests and manual qualification notes are local work,
not evidence that omitted portions of the source commits landed.
Commit messages must name full source SHAs, original authorship,
imported paths/tests, omissions, local additions and replay guidance.
Use explicit partial-backport language for adapter subsets; do not
claim a full cherry-pick or reuse the source's broader fix claims.
## Remaining decision
This candidate removes the reproduced fatal quote-restart
propagation and imports existing reset fixes. The human exercised
live QQQ startup, backfill and concurrent chart consumer lifetimes.
Completion-callback and teardown errors remain; the human accepts
them as follow-ups outside the bounded #94 landing.
The storage/SHM portions of the downstream fixes are not covered by
the adapter backport or its integration tests.
`ddac3553` is Deribit-specific and omitted. Qt lifecycle commit
`1a2d4893` stays a separate follow-up. Further Tractor teardown
changes or storage imports require reassessing the bounded landing
rather than silently expanding this candidate.

View File

@ -1,71 +0,0 @@
# IB recovery qualification, 2026-11-06
This records human-operated qualification of the bounded #94
landing on `wkt/land_ib_recovery`, starting from `495c0915` plus
the narrow history catch described below. It supplements the
historical evidence in `ai/ib-recovery-candidate.md`.
## Runtime and dependency source
The evaluated environment uses Python 3.13, Trio 0.34.0 and Tractor
`8fe574c1e447358dafd8ebf5ada3b0714c688f78`. Tractor was an editable
local checkout during evaluation. The human subsequently published
that exact commit as GitHub's `goodboy/tractor:piker_pin`; a read-only
remote query confirmed the published OID before restoring Piker's
Git source and recording the same OID in `uv.lock`.
Offline lock validation resolves 126 packages successfully. No
environment sync was performed. This validates the lock metadata;
it does not claim a fresh installation from the published source.
## Manual observations
The human operated Gateway, chart windows and paper orders. The
agent did not operate live services or the account.
- Same-feed overlap, close/reopen and paper order submit/cancel
worked after the existing order-dialog guard was imported.
- Different-feed overlap originally exposed a timed-out one-second
MNQ March 2027 gap query. PDB showed `timedout=True` and
`out is None`. Its uncaught `NoData` canceled sibling history and
quote tasks even though `pikerd` remained alive.
- After the narrow catch, the human reported the overlap journeys
working, then confirmed cases 5 and 6 with Ctrl+Alt+F: overlapping
live feeds survived the Gateway data reset; closing all charts
and reopening with `pikerd` left running restored live updates.
- Earlier Ctrl+Shift+R attempts left feeds live, but do not prove
that the intended Gateway data reset was triggered.
These are manual observations, not automated timing or latency
measurements. Ctrl+Alt+R was not qualified here; Piker's utility
labels it a connection reset.
## Narrow history backport
Only the `NoData`/`DataUnavailable` catch around `get_hist()` in
`maybe_fill_null_segments()` is copied from
`0846cbd4a4197e47415689cf5e9d146ba3174c32`, authored by
`goodboy <goodboy_foss@protonmail.com>`. The existing gap fallback
remains in place; this does not import broader history repair.
The new regression uses real null detection and checks that a
sibling in the same nursery still runs after either provider
exception. Both cases failed before the catch. With it, the two
focused history files passed: 17 tests. Earlier bounded-recovery
and Tractor qualification results remain separate evidence, not
fresh test runs of the final dependency commit tree.
## Remaining recovery limits
A full Gateway restart was triggered accidentally through its
auto-restart settings dialog. The disconnect delivered the string
`Peer closed connection.` to `deliver_trade_events()`. The string
branch then reached `err['reqid']`, raising `TypeError`; its `code`
local can also retain an earlier event's value. Full Gateway restart
recovery is outside this bounded landing.
The earlier history-failure teardown also exposed a separate
Tractor completion-callback assertion after `BrokenResourceError`.
The history catch does not establish that this callback race is
fixed. Broader history recovery and these disconnect paths remain
downstream work.

View File

@ -1,52 +0,0 @@
---
created_at: 2026-10-07T16:54:53Z
generated_by:
harness: opencode
provider: openai
model: gpt-6.1-sol
git_ref: bde2457b
scope: code
substantive: true
raw_file: 20261007T165453Z_bde2457b_prompt_io.raw.md
---
## Prompt
The user requested fixing all seven local PR94 review findings and
exporting the review for human inspection in Tuicr. They asked that a
Tractor pin adjustment be considered before adding Piker workarounds.
## Response summary
Correct the endpoint defaults, Click argument shape, daemon binding and
registrar option handling, chart fallback, and stray debugger pause.
Keep the absent datad implementation's test outside this landing
branch. Add offline startup coverage and export original findings
without modifying review states. The checked newer Tractor address
branch still rejects mixed transports, so leave the dependency pin.
## Files changed
- `config/conf.toml`: TCP-only default endpoints.
- `piker/cli/__init__.py`: repeatable and intact multiaddresses.
- `piker/service/_actor_runtime.py`: validate daemon listeners.
- `piker/service/_daemon.py`: initialize and consume registrar options.
- `piker/ui/cli.py`: fallback to automatic chart binding.
- `piker/brokers/binance/feed.py`: remove unconditional debugger entry.
- `tests/test_services.py`: defer test requiring missing datad API.
- `tests/test_startup.py`: CLI/config and service lifecycle regressions.
## Human edits
The human identified that datad tests arrived through the agent's
mixed cherry-picks and explicitly requested deferring that test. They
requested a chart default for existing configs, expanded the task to
all review findings, and directed checking whether a newer Tractor pin
could reduce the fix scope. No direct manual source edits were observed
during these changes; acceptance remains with the human.
During the human review pass, the human confirmed removing the Binance
pause and correcting CLI endpoint parsing. They requested an explicit
template note about this pin's single-transport limitation and detailed
comments distinguishing registrar discovery endpoints from daemon bind
endpoints. Those explanations were added at the human's direction.

View File

@ -1,61 +0,0 @@
---
created_at: 2026-10-07T16:54:53Z
generated_by:
harness: opencode
provider: openai
model: gpt-6.1-sol
git_ref: bde2457b
diff_cmd: git diff HEAD~1..HEAD
---
## Prompt
Fix all findings from the local PR94 review, prefer a Tractor pin
change if it reduces work, and export the original review for Tuicr.
## Generated changes
> `git diff HEAD~1..HEAD -- config/conf.toml`
Use TCP-only defaults, retaining commented UDS alternatives. Both
checked Tractor address branch tips still reject mixed transports.
> `git diff HEAD~1..HEAD -- piker/cli/__init__.py`
Make Click multiaddresses repeatable and normalize single strings.
> `git diff HEAD~1..HEAD -- piker/service/_actor_runtime.py`
Validate requested daemon bind addresses instead of remote registrars.
Normalize accepted endpoints and allow TCP ephemeral port requests.
> `git diff HEAD~1..HEAD -- piker/service/_daemon.py`
Resolve explicit or inherited registrars independently of other daemon
options and consume the explicit keyword before forwarding kwargs.
> `git diff HEAD~1..HEAD -- piker/ui/cli.py`
Default chart bind endpoints to None when network config is absent.
> `git diff HEAD~1..HEAD -- piker/brokers/binance/feed.py`
Remove the unconditional market-info debugger pause.
> `git diff HEAD~1..HEAD -- tests/test_services.py tests/test_startup.py`
Remove the prematurely cherry-picked datad test and add offline startup
regressions, including a real registrar/child daemon lifecycle check.
## Verification
- Startup and logging tests: 13 passed.
- Existing runtime boot and tagged registrar tests: 2 passed.
- Scoped native Ruff and diff checks pass.
- The first new actor test used a removed nursery API; it was corrected
to the installed start_actor/Portal.run interface before passing.
- Import initially failed without native libraries. Tests use the
existing flake's GCC, zlib, Qt and PyQt paths without executing its
environment-rebuilding shell hook.
- Original seven findings exported locally as Tuicr review-add payloads
bound to bde2457b; no session comments or remote PR were changed.

View File

@ -1,45 +0,0 @@
---
created_at: 2026-10-10T23:28:52Z
generated_by:
harness: opencode
provider: openai
model: openai/gpt-6.1-sol
git_ref: 1fa45dfa
scope: config
substantive: true
raw_file: 20261010T232852Z_1fa45dfa_prompt_io.raw.md
---
## Prompt
The human rejected keeping the April Tractor pin as a landing baseline.
They requested the reviewed `wkt/addr_unpacking` branch, removal of the
new rollback adaptations, reuse of original Piker histories where
possible, and a candidate prepared within their gateway relocation
window. They asked specifically whether the runtime validation review
hunk duplicated older endpoint history.
## Response summary
Select the reviewed address branch at an immutable locked revision and
preserve all committed application/test compatibility code. Verify the
minimal dependency boundary offline and preview existing consumer
fixes without importing unrelated datad/backfill changes. Distinguish
original-source test results from live teardown acceptance, which is
not available in this preparation window.
## Files changed
- `pyproject.toml`: select Tractor `wkt/addr_unpacking`.
- `uv.lock`: resolve reviewed `a966928d` and setproctitle 1.3.8 while
preserving Polars 1.36.0 and the rest of the locked graph.
## Human edits
The human redirected the task from an older known-good runtime to a
modern landing stack and insisted on reusing original history rather
than building new compatibility adapters. They proposed stricter
exception-group behavior as a hypothesis and set the live-test/time
constraints. Those directions materially changed the dependency
boundary. No direct manual edits to this final boundary were observed;
runtime acceptance and landing remain human-owned.

View File

@ -1,52 +0,0 @@
---
created_at: 2026-10-10T23:28:52Z
generated_by:
harness: opencode
provider: openai
model: openai/gpt-6.1-sol
git_ref: 1fa45dfa
diff_cmd: git diff HEAD~1..HEAD
---
## Prompt
Undo the April Tractor rollback and select the reviewed
`wkt/addr_unpacking` branch. Prefer original downstream compatibility
commits to new adaptations, and prepare an offline-verified candidate
while live IB verification is deferred for gateway relocation.
## Generated changes
> `git diff HEAD~1..HEAD -- pyproject.toml uv.lock`
Change Tractor's source from `piker_pin` to `wkt/addr_unpacking` and
lock it at `a966928dead4cbf369ad244834ce9dbe988ac8ca`. Restore application
and test sources byte-for-byte to the committed PR94 tree. Keep the
rejected rollback journal only in ignored local experiment reports.
Preserve Polars 1.36.0 by retaining its exact existing lock blocks;
setproctitle re-resolves from 1.3.7 to 1.3.8. `uv sync --locked` accepts
the resulting graph. Actual imports confirm the reviewed Tractor pin
and Trio 0.32.0, not the hypothesized Trio 0.33.
## Original-history previews
The existing MethodProxy stack is retained. Read-only merge previews
show original `4af885d9` and `b691422b` can be applied without the datad
cutover. Original-source regressions run through temporary in-memory
overlays against the real pinned runtime: seven cases pass and one
cache-hit case fails because `log.at_least_level('info')` looks up INFO
in a custom-level mapping. A separate standard logging correction
produces eight passing cases. Its patch and execution handoff are saved
under ignored `.ai/code-review/reports/pr94_modern_stack/`; no consumer
cherry-picks or source edits are performed by this boundary.
## Verification
- Startup and logging: 13 passed.
- Runtime boot and tagged registrar normalization: 2 passed.
- Locked sync and source whitespace checks pass.
- Original application/test trees are unchanged from PR94 HEAD.
- Consumer overlay with explicit logging correction: 8 passed.
- Correction patch validates against its exact preview tree using a
private temporary index; the real index is untouched.
- Live IB startup, SIGINT and surviving-pikerd reattachment are pending.

View File

@ -1,45 +0,0 @@
---
created_at: 2026-11-06T18:56:28.621577+00:00
generated_by:
harness: codex
provider: OpenAI
git_ref: 495c0915
scope: code
substantive: true
raw_file: 20261106T185628Z_495c0915_215e9b_prompt_io.raw.md
---
## Prompt
User reported a different-feed overlap failure and asked whether
a fix could be extracted without importing broader history work.
PDB confirmed MNQ March 2027 one-second history timed out and
returned None, producing NoData. After continuing PDB, sibling
history and quote tasks terminated while pikerd remained alive.
User requested continuing with the narrow backport.
## Response summary
Partial backport of the NoData/DataUnavailable catch from
0846cbd4a4197e47415689cf5e9d146ba3174c32 into
maybe_fill_null_segments(). The existing gap fallback remains in use.
A new parametrized regression uses real null detection and verifies
that a sibling in the same nursery survives the failed repair.
Both new cases failed before the catch. The focused history suites
passed after it: 17 passed. This does not qualify live overlap or fix
IB reset churn or the separate Tractor completion-callback assertion.
## Files changed
- `piker/tsp/_history.py`: existing downstream exception catch.
- `tests/test_history_backfill.py`: sibling survival regression.
## Human edits
User supplied live reproduction, PDB locals and cancellation logs,
and explicitly constrained scope to a small history extraction.
The production catch is copied from existing downstream commit
0846cbd4, not newly designed here. New regression is agent-authored;
no direct human source edits observed during this patch. Review and
live qualification remain with the user.

View File

@ -1,22 +0,0 @@
---
created_at: 2026-11-06T18:56:28.621577+00:00
generated_by:
harness: codex
provider: OpenAI
git_ref: 495c0915
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/tsp/_history.py`
> `git diff HEAD~1..HEAD -- tests/test_history_backfill.py`
Partial backport of the NoData/DataUnavailable catch from
0846cbd4a4197e47415689cf5e9d146ba3174c32 into
maybe_fill_null_segments(). The existing gap fallback remains in use.
A new parametrized regression uses real null detection and verifies
that a sibling in the same nursery survives the failed repair.
Both new cases failed before the catch. The focused history suites
passed after it: 17 passed. This does not qualify live overlap or fix
IB reset churn or the separate Tractor completion-callback assertion.

View File

@ -1,41 +0,0 @@
---
created_at: 2026-11-06T20:17:46.622316+00:00
generated_by:
harness: codex
provider: OpenAI
git_ref: 495c0915
scope: config
substantive: true
raw_file: 20261106T201746Z_495c0915_612dfc_prompt_io.raw.md
---
## Prompt
The human requested a commit plan, then a fast-forward landing of
the #94 branch and PR updates. They agreed the temporary local
Tractor source must be replaced, published piker_pin themselves,
and reported "pushed it". They also supplied the manual recovery
results and full Gateway restart failure diagnostics.
## Response summary
Restore the Git Tractor source after local qualification and lock
its published piker_pin at 8fe574c1e447358dafd8ebf5ada3b0714c688f78.
Retain the evaluated Trio 0.34.0 lock and require Trio >=0.34.0.
Record the human's overlap, data-reset and reopen observations,
the narrow history extraction, and full-restart recovery limits.
Offline uv lock validation passed without an environment sync.
## Files changed
- `pyproject.toml`: Trio minimum; restore existing Tractor Git source.
- `uv.lock`: published Tractor OID and evaluated Trio 0.34.0.
- `ai/ib-recovery-qualification-20261106.md`: scoped evidence.
## Human edits
The human selected the bounded landing scope, performed live tests,
identified the more aggressive reset distinction, and published
the tested Tractor ref. The agent applied the metadata edits and
wrote the qualification record from those observations. No human
acceptance or task-state transitions are inferred by this record.

View File

@ -1,21 +0,0 @@
---
created_at: 2026-11-06T20:17:46.622316+00:00
generated_by:
harness: codex
provider: OpenAI
git_ref: 495c0915
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- pyproject.toml`
> `git diff HEAD~1..HEAD -- uv.lock`
> `git diff HEAD~1..HEAD -- ai/ib-recovery-qualification-20261106.md`
Restore the Git Tractor source after local qualification and lock
its published piker_pin at 8fe574c1e447358dafd8ebf5ada3b0714c688f78.
Retain the evaluated Trio 0.34.0 lock and require Trio >=0.34.0.
Record the human's overlap, data-reset and reopen observations,
the narrow history extraction, and full-restart recovery limits.
Offline uv lock validation passed without an environment sync.

View File

@ -1,11 +0,0 @@
# AI Prompt I/O Log
This directory records substantive AI-assisted changes, prompts,
original harness identity, and unedited outputs. Human contributors
remain accountable for code decisions.
Records follow the [NLNet generative AI policy][policy]. Each
structured entry links its immutable `.raw.md` partner. Code output
uses Git diff references to avoid duplication.
[policy]: https://nlnet.nl/foundation/policies/generativeAI/

View File

@ -1,59 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:08:59Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T170859Z_75cefe10_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T170859Z_75cefe10_prompt_io.raw.md
---
## Prompt
Session-initiating instruction (driving all 7 commits in
this series):
> ok i want you to become the distributed runtime and
> concurrency expert for this project - namely acquire
> a deep understanding of tractor and how it's used.
> then i want you to attempt to factor our current
> brokerd service daemon into 2 daemons: a brokerd
> which only hosts live/paper trading endpoint tasks
> [..] a new `datad` subdaemon which in a separate
> subactor serves all the data feed service tasks [..]
> give me a mega detailed plan on how to approach this,
> and a staged approach for the implementation.
Proximate driver for THIS commit: during the approved
plan's stage-0 test gate the agent discovered test
subactors were writing into the user's REAL
`~/.config/piker/accounting/` files (a bogus paper fill
landed in `account.kraken.paper.toml`) because the old
test-dir override in `config.get_app_dir()` was
commented out and could never work in spawned
subactors anyway. Fix applied autonomously under the
approved plan; no explicit per-fix user prompt.
## Response summary
Restore `pytest` config-dir isolation by resolving the
per-test tmp dir lazily (at conf-path access time) from
`tractor` runtime-vars, which propagate down the actor
tree; route all conf-path derivation through
`config.get_conf_dir()` so the override is effective in
every (sub)actor.
## Files changed
- `piker/config.py` — add `_maybe_use_test_dir()`;
hook in `get_conf_dir()`; route `get_conf_path()` +
`load()` mkdir through it
- `piker/accounting/_ledger.py` — derive ledger dir
via `config.get_conf_dir()` (not the global)
- `piker/accounting/_pos.py` — same for account files
## Human edits
None — committed as generated.

View File

@ -1,46 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:08:59Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T170859Z_75cefe10_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); backfilled from the live dev
session transcript per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/config.py`
Generated: `config._maybe_use_test_dir()` — lazily
reads `piker_test_dir` from
`tractor.runtime._state._runtime_vars['piker_vars']`
(pre-loaded by `open_piker_runtime()` from the
`tests.conftest._open_test_pikerd()` overrides) and
calls `_override_config_dir()` when set. Hooked at the
top of `get_conf_dir()`; `get_conf_path()` and
`load()`'s dir-creation rerouted through
`get_conf_dir()`.
> `git log -1 -p --follow -- piker/accounting/_ledger.py`
> `git log -1 -p --follow -- piker/accounting/_pos.py`
Generated: ledger/account dir derivation switched from
the `config._config_dir` module global to
`config.get_conf_dir()` + `mkdir(parents=True,
exist_ok=True)` for nested tmp-dir creation.
Key diagnostic reasoning (verbatim from session):
The old (commented-out) `get_app_dir()` override gated
on `'pytest' in sys.modules` which can NEVER work in
spawned subactors (fresh procs, no pytest import); as
a result test `paperboi`/daemon actors were writing
into the user's REAL `~/.config/piker/accounting/`
files. Evidence: `account.kraken.paper.toml` gained a
single 0.001 xbtusdt clear stamped 2026-06-09T18:47Z
(a test fill), and `trades_kraken_paper.toml` (252B)
contained only that fill. The resolution must be
checked lazily at config-path access time (NOT import
time) since sub-actors only receive runtime-vars once
their `tractor` runtime has fully booted.

View File

@ -1,53 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:10:22Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171022Z_4485f2b9_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T171022Z_4485f2b9_prompt_io.raw.md
---
## Prompt
Same session-initiating `brokerd`-split instruction (see
`20260610T170859Z_75cefe10_prompt_io.md`). Proximate
driver: the approved plan's per-stage test gates could
not run AT ALL — the branch base was broken vs.
`tractor` git `main` (`AttributeError: 'tuple' object
has no attribute 'unwrap'` at `pikerd` boot, stale
`arbiter_sockaddr`/`.type`/`.size` API refs). The agent
fixed forward autonomously to (re)establish the gate
baseline, continuing the `repair_tests` branch lineage
(verified already merged in ancestry via `git cherry`
during a user-requested branch-overlap survey).
## Response summary
Port the service layer + test suites to current
`tractor` APIs: addr-type normalization in
`open_pikerd()`, `query_actor()`/`wait_for_actor()`
kwarg renames, modern self-cancel absorption semantics
in the cancel-method test harness, exc/position attr
renames, a paper-EMS startup-budget bump and a syntax
fix in `.deribit.api`.
## Files changed
- `piker/service/_actor_runtime.py` — `wrap_address()`
normalize before `.unwrap()` in `open_pikerd()`
- `piker/service/_registry.py` — `check_for_service()`
-> `query_actor(regaddr=)` + 2-tuple yield +
`open_registry(addrs=)`
- `piker/brokers/deribit/api.py` — missing comma in
`tractor.trionics` import tuple
- `tests/test_services.py` — `registry_addr=` kwarg,
raddr unwraps, cancel-semantics harness rewrite,
`fail_after` 9 -> 19s
- `tests/test_ems.py` — `.boxed_type`, `pp.cumsize`
## Human edits
None — committed as generated.

View File

@ -1,56 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:10:22Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171022Z_4485f2b9_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); backfilled from the live dev
session transcript per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/service/_actor_runtime.py`
Generated: normalize each registry addr via
`tractor.discovery._addr.wrap_address()` before
`.unwrap()`-ing for the `accept_addrs` bind check —
entries may be raw `tuple`s when passed in from (test)
client code. Import-path precedent taken from
`piker/cli/__init__.py:336`.
> `git log -1 -p --follow -- piker/service/_registry.py`
Generated: `check_for_service()` ported to
`tractor.query_actor(name, regaddr=...)` (kwarg was
`arbiter_sockaddr=`), unpacking the new
`(sockaddr, portal)` yield, and passing the
now-required `open_registry(addrs=Registry.addrs)`.
> `git log -1 -p --follow -- tests/test_services.py`
> `git log -1 -p --follow -- tests/test_ems.py`
> `git log -1 -p --follow -- piker/brokers/deribit/api.py`
Key diagnostic reasoning (verbatim from session):
- the "DID NOT RAISE ContextCancelled" failure: in this
test the client actor IS pikerd (in-proc), and
current `tractor` main absorbs a `ContextCancelled`
whose canceller is your own actor — self-requested
cancels now exit cleanly instead of raising; the
'sigint' variant propagates a bare collapsed
`KeyboardInterrupt` rather than a
`BaseExceptionGroup`.
- the hard-coded `trio.fail_after(9)` startup budget is
marginal — full stack boot (pikerd -> emsd ->
brokerd.kraken -> paperboi + live kraken symbology
fetch) occasionally exceeds 9s -> bumped to 19s.
- `RemoteActorError.type` -> `.boxed_type`;
`Position.size` -> `.cumsize` (the paper engine
populates `BrokerdPosition.size` from `pp.cumsize`).
- overlap survey (user-requested): all of the
`repair_tests` branch commits are already in this
stack's ancestry; this commit finishes that branch's
port mission (its f4c4f1e2 fixed `conftest.py`'s
`arbiter_sockaddr` usage; this fixes the remaining
`test_services.py` + `check_for_service()` sites).

View File

@ -1,54 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:11:05Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171105Z_bc6e18d7_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T171105Z_bc6e18d7_prompt_io.raw.md
---
## Prompt
Same session-initiating `brokerd`-split instruction (see
`20260610T170859Z_75cefe10_prompt_io.md`); this is the
approved plan's "stage 0" prep. Plan-shaping user
decisions captured via in-session Q&A:
- `datad.<broker>` topology: sibling of
`brokerd.<broker>` under `pikerd` (vs. child
subactor).
- migration: hard cutover, staged by layer (no
dual-mode runtime flag).
- post-split `brokerd` scope: trading-only +
EMS-lazy-spawned (charts/CLI never touch it).
## Response summary
Declare per-daemon-kind backend submod groups
(`_datad_mods`/`_brokerd_mods`) in the split-style
backends, keyed to the pre-existing
`piker.data.validate._eps` contract; add a
`validate.get_eps()` introspection helper; ws-import
hygiene in `.kraken.broker`. Zero behavior change
(`__enable_modules__` unions unchanged).
## Files changed
- `piker/data/validate.py` — add `get_eps()`
- `piker/brokers/kraken/__init__.py` — mod groups
- `piker/brokers/binance/__init__.py` — mod groups
(note: no `symbols.py`; search eps live in `.feed`)
- `piker/brokers/deribit/__init__.py` — datad-only
groups (no `broker.py` yet)
- `piker/brokers/ib/__init__.py` — add `'api'` to the
pre-existing `_datad_mods`
- `piker/brokers/kraken/broker.py` — import
`NoBsWs`/`open_autorecon_ws` from
`piker.data._web_bs` directly
## Human edits
None — committed as generated.

View File

@ -1,46 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:11:05Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171105Z_bc6e18d7_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); backfilled from the live dev
session transcript per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/data/validate.py`
Generated: `get_eps(mod, kind) -> dict[str, Callable]`
returning the daemon-kind's ep funcs defined by a
backend mod, keyed by ep name, missing eps excluded;
sourced from the existing `_eps` grouping table
(`'middleware' | 'datad' | 'brokerd'`).
> `git log -1 -p --follow -- piker/brokers/kraken/__init__.py`
> `git log -1 -p --follow -- piker/brokers/binance/__init__.py`
> `git log -1 -p --follow -- piker/brokers/deribit/__init__.py`
> `git log -1 -p --follow -- piker/brokers/ib/__init__.py`
> `git log -1 -p --follow -- piker/brokers/kraken/broker.py`
Key exploration findings driving the design (from the
planning phase, 3 parallel explore agents + 1 architect
agent):
- the codebase already anticipated this split:
`validate._eps` groups eps into 'datad' vs 'brokerd'
kinds and `ib/__init__.py` already declared
`_brokerd_mods`/`_datad_mods`; `brokers/_daemon.py`
carried the literal TODO "rename the daemon to datad
prolly once we split up broker vs. data tasks into
separate actors?".
- the feared kraken feed<->broker "shared ws state"
hazard is mild: `NoBsWs`/`open_autorecon_ws`
originate in `piker.data._web_bs` (re-exported via
`.kraken.feed`); only `stream_messages` is feed-local
and it's a pure parser — in-process imports only,
each actor opens its own ws conn.
- `enable_modules` gates RPC entry, NOT python imports,
so backend trading mods may keep importing feed-side
helpers in-process post-split.

View File

@ -1,47 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:11:42Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171142Z_119d2c04_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T171142Z_119d2c04_prompt_io.raw.md
---
## Prompt
Same session-initiating `brokerd`-split instruction (see
`20260610T170859Z_75cefe10_prompt_io.md`); this is the
approved plan's "stage 1": introduce the `datad` daemon
machinery additively (nothing routes through it yet).
User-decided constraint applied: `datad.<broker>` is a
SIBLING of `brokerd.<broker>` under `pikerd`, spawned
via the existing `Services` + `maybe_spawn_daemon()`
machinery.
## Response summary
New `piker/data/_daemon.py` hosting the
`datad.<broker>` feed-only daemon-actor: lifetime
fixture owning the actor-global `_FeedsBus`, init/spawn
fns mirroring `.brokers._daemon` conventions and the
`samplerd` sub-daemon precedent, plus root-mod
registration, `piker.service` re-exports and a spawn
test.
## Files changed
- `piker/data/_daemon.py` — NEW:
`_setup_persistent_datad()`, `datad_init()`,
`spawn_datad()`, `maybe_spawn_datad()`,
`_datad_service_mods`
- `piker/service/_actor_runtime.py` — add
`piker.data._daemon` to `_root_modules`
- `piker/service/__init__.py` — re-export spawn eps
- `tests/test_services.py` — add `test_datad_spawn`
## Human edits
None — committed as generated.

View File

@ -1,60 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:11:42Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171142Z_119d2c04_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); backfilled from the live dev
session transcript per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/data/_daemon.py`
Generated symbols + key design decisions:
- `_datad_service_mods: list[str]` — datad-always
enabled mods, the data-side successor to the old
`piker.brokers._daemon._data_mods` set; kept minimal
per the caps-sec model.
- `_setup_persistent_datad()` — `@tractor.context`
lifetime fixture: console-log setup then allocates
the actor-global feed bus via
`feed.get_feed_bus(brokername, service_nursery)`
exactly as the old brokerd fixture did, pinned open
with `ctx.started()` + `sleep_forever()`.
- `datad_init()` — actor name `f'datad.{brokername}'`;
copies backend `_spawn_kwargs` (CRITICAL for `ib`'s
`infect_asyncio=True`); builds `enable_modules` from
`getattr(brokermod, '_datad_mods',
getattr(brokermod, '__enable_modules__', []))` as
the flat-backend fallback.
- `spawn_datad()` — `Services.actor_n.start_actor()` +
`Services.start_service_task()` exactly mirroring
`spawn_brokerd()`; dedup-composes enable mods via
`list(dict.fromkeys(...))`.
- `maybe_spawn_datad()` — wraps `maybe_spawn_daemon(
service_name=f'datad.{brokername}', ...)`.
> `git log -1 -p --follow -- piker/service/_actor_runtime.py`
> `git log -1 -p --follow -- piker/service/__init__.py`
> `git log -1 -p --follow -- tests/test_services.py`
Design rationale (verbatim from session):
- `_root_modules` must gain `piker.data._daemon` so
`pikerd_portal.run(spawn_datad, ...)` resolves in
the root.
- the `Services`-based impl style deliberately mirrors
`spawn_brokerd()` so the eventual `tractor.hilevel`
`ServiceMngr` port (see the `service_mng_to_tractor`
branch's d8c21d44 prep, surfaced by the
user-requested branch-overlap survey) lands
symmetrically on both spawn fns.
- mod placement (`piker/data/_daemon.py` vs.
generalizing `piker.brokers._daemon`) follows the
per-subsystem daemon-mod convention
(`.clearing._ems`, `.data._sampling`) and resolves
the existing TODO at `brokers/_daemon.py:49` ("move
this def to the `.data` subpkg").

View File

@ -1,47 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:12:26Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171226Z_64181219_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T171226Z_64181219_prompt_io.raw.md
---
## Prompt
Same session-initiating `brokerd`-split instruction (see
`20260610T170859Z_75cefe10_prompt_io.md`); this is the
approved plan's "stage 2": decouple the clearing layer
from `feed.portals` BEFORE the feed cutover so live
trading works at every stage boundary. User-decided
constraint applied: post-split `brokerd` is
trading-only and spawned LAZILY only by `emsd`'s
`open_brokerd_dialog()` path.
## Response summary
Kill the single coupling forcing feed + trading eps
into one actor (`Router.open_trade_relays()` pulling
its trades portal from `feed.portals[brokermod]`):
`open_brokerd_dialog()` now (maybe) spawns/finds
`brokerd.<broker>` itself and ONLY when a live
trades-ep will actually open; paper mode never touches
it. Pre-cutover this is a pure refactor (registry
lookup finds the same feed-spawned daemon).
## Files changed
- `piker/clearing/_ems.py` — `open_brokerd_dialog()`
re-sig + inner `acquire_live_portal()`;
`Router.maybe_open_brokerd_dialog()` drops `portal`
param; `open_trade_relays()` drops the
`feed.portals` lookup
- `piker/accounting/cli.py` — keyword-form `portal=`
override kept for the `piker ledger` ad-hoc actor
## Human edits
None — committed as generated.

View File

@ -1,43 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:12:26Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171226Z_64181219_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); backfilled from the live dev
session transcript per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/clearing/_ems.py`
Generated: new `open_brokerd_dialog()` signature
`(brokermod, exec_mode, fqme=None, portal=None,
loglevel=None)` with an inner `@acm
acquire_live_portal()` that yields the caller-provided
`portal` override (the `piker ledger` path) else
`maybe_spawn_brokerd(brokermod.name)` — designated THE
one place a live, credentialed `brokerd.<broker>` gets
booted post-split. The eager
`portal.open_context(trades_endpoint, ...)`
construction moved inside that block so the paper
short-circuit never acquires a live portal.
> `git log -1 -p --follow -- piker/accounting/cli.py`
Key analysis (verbatim from session):
- `feed.portals` had exactly ONE trading consumer:
`piker/clearing/_ems.py:671` (`portal =
feed.portals[brokermod]`) — the single coupling
forcing feed + trading into one actor.
- `piker ledger` (`accounting/cli.py:100-157`) is a
hidden consumer: it calls `broker_init()` directly,
spawns its own ad-hoc actor and passes the portal
positionally — the new signature must keep an
explicit `portal:` override param for this path.
- stage sequencing: clearing decouple lands BEFORE the
feed cutover, otherwise `feed.portals` would hand
emsd a `datad` portal and live `open_trade_dialog`
RPC would fail.

View File

@ -1,48 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:12:59Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171259Z_59d5d9a6_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T171259Z_59d5d9a6_prompt_io.raw.md
---
## Prompt
Same session-initiating `brokerd`-split instruction (see
`20260610T170859Z_75cefe10_prompt_io.md`); this is the
approved plan's "stage 3": the feed-layer hard cutover
(user-decided: no dual-mode runtime flag, each stage
lands fully working).
## Response summary
The topology flip: all data-feed consumers route to
`datad.<broker>`; the `brokerd` lifetime fixture slims
to trading-only; UI/CLI data-flavoured spawn sites
repointed; test expectations inverted incl. an explicit
negative assert that paper sessions spawn NO
`brokerd.<broker>`.
## Files changed
- `piker/data/feed.py` — `open_feed()` ->
`maybe_spawn_datad()`; `open_feed_bus()` actor-name
assert -> `'datad'`; `install_brokerd_search` ->
`install_datad_search`; comment sweep
- `piker/brokers/_daemon.py` —
`_setup_persistent_brokerd()` slimmed (feed-bus
alloc moved to the datad fixture)
- `piker/ui/_app.py`, `piker/ui/cli.py`,
`piker/brokers/core.py`, `piker/brokers/cli.py`,
`piker/ui/kivy/option_chain.py` — spawn-site
repoints
- `tests/test_services.py` — datad assertions + the
no-`brokerd.kraken` negative check
## Human edits
None — committed as generated.

View File

@ -1,54 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:12:59Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171259Z_59d5d9a6_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); backfilled from the live dev
session transcript per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/data/feed.py`
Generated: `open_feed()` spawn cutover with the
`maybe_spawn_datad` import done "relative-direct" from
`._daemon` — NOT via `piker.service` — to dodge a
partial-init cycle: when `piker.data.feed` loads as
part of `piker.service.__init__` executing its own
`piker.data._daemon` import, the service pkg is
mid-init and its `maybe_spawn_datad` binding does not
exist yet. The `open_feed_bus()` local-state sanity
assert flips `'brokerd' in servicename` ->
`'datad' in servicename` (the only actor-name assert
in the tree, verified by grep).
> `git log -1 -p --follow -- piker/brokers/_daemon.py`
Generated: `_setup_persistent_brokerd()` slimmed to
console-log setup + pinned-open ctx; drops the bus
alloc, the `assert not feed._bus`, the service nursery
(backend `open_trade_dialog()` ctxs own their task
trees) and the `eg.ExceptionGroup` handler. The
`piker ledger` ad-hoc actor enters this same slimmed
fixture — exactly what it needs.
> `git log -1 -p --follow -- piker/ui/_app.py`
> `git log -1 -p --follow -- piker/ui/cli.py`
> `git log -1 -p --follow -- piker/brokers/core.py`
> `git log -1 -p --follow -- piker/brokers/cli.py`
> `git log -1 -p --follow -- piker/ui/kivy/option_chain.py`
> `git log -1 -p --follow -- tests/test_services.py`
Verification (from session): per-suite gates green at
this commit (services 5-passed incl. the new negative
assert, feeds 3-passed); a headless live smoke
(`maybe_open_pikerd` + `open_feed(['xbtusdt.kraken'])`
on an alt registry port) confirmed quotes flowing via
`datad.kraken` + `samplerd` with `check_for_service(
'brokerd.kraken') is None`. Known pre-existing flake
documented: `test_multi_fill_positions`' second
in-proc runtime boot wedges ~50% (zombie subactor w/
unread parent-IPC bytes); reproduced with the split
fully reverted so NOT a regression of this work.

View File

@ -1,43 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:13:44Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171344Z_eee19de0_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T171344Z_eee19de0_prompt_io.raw.md
---
## Prompt
Same session-initiating `brokerd`-split instruction (see
`20260610T170859Z_75cefe10_prompt_io.md`); this is the
approved plan's final "stage 4": caps-sec slimming of
the (live, credentialed) trading actor + the `ib`
dual-daemon `client_id` collision mitigation flagged in
the plan's risk register.
## Response summary
`brokerd` loses ALL `piker.data.*` (feed) RPC mods;
spawn fails fast for datad-only backends with a "use
paper-mode" error; `ib`'s default api-gw `client_id`
gets a per-daemon-kind offset so `datad.ib` +
`brokerd.ib` don't collide on connect.
## Files changed
- `piker/brokers/_daemon.py` — `_data_mods` -> minimal
`_brokerd_service_mods`; `broker_init()` reads
`_brokerd_mods` (fallback `__enable_modules__`);
`spawn_brokerd()` fail-fast via `validate.get_eps()`
- `piker/brokers/ib/api.py` — role-based `client_id`
offset in `load_aio_clients()`
- `piker/cli/__init__.py` — resolved "expose datad"
TODO
## Human edits
None — committed as generated.

View File

@ -1,41 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:13:44Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T171344Z_eee19de0_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); backfilled from the live dev
session transcript per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/brokers/_daemon.py`
Generated: the fail-fast originally landed in
`broker_init()` but was relocated to `spawn_brokerd()`
mid-implementation after realizing `piker ledger`
calls `broker_init()` directly even for paper accounts
on datad-only backends (would have crashed the cli);
the service-spawn path is the correct enforcement
seam. Error text:
Backend 'kucoin' offers NO `brokerd` (live
order-control) eps!? It is likely a datad-only
provider, use paper-mode for clearing instead.
(verified live via a `trio.run()` unit check.)
> `git log -1 -p --follow -- piker/brokers/ib/api.py`
Generated: in `load_aio_clients()`, when `client_id`
is the 6116 default: `datad`-named actors offset +16
(disjoint from `brokerd`'s linear `client_id + i`
retry range), other non-`brokerd` (ad-hoc test/cli)
actors +32. Rationale from the plan's risk register:
post-split BOTH per-broker daemons connect to the same
TWS/gw endpoint; a shared default id collides and
burns up to `connect_timeout * retries` (90s) in
retry cycles.
> `git log -1 -p --follow -- piker/cli/__init__.py`

View File

@ -1,66 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T17:33:09Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T173309Z_f15f8178_prompt_io.md
scope: docs
substantive: true
raw_file: 20260610T173309Z_f15f8178_prompt_io.raw.md
---
## Prompt
The session-initiating instruction, verbatim (this doc
IS its "mega detailed plan" deliverable):
> ok i want you to become the distributed runtime and
> concurrency expert for this project - namely acquire
> a deep understanding of tractor and how it's used.
> then i want you to atttempt to factor our current
> brokerd service daemon into 2 daemons:
>
> - a brokerd which only hosts live/paper trading
> endoint tasks as defined namely within all
> piker/brokers/<backend>/broker.py mods
> - a new `datad` subdaemon which in a separate
> subactor serves all the data feed service tasks
> namely delivered by endpoints in
> piker/brokers/<backend>/feed.py (or similar for
> less mature backends) and as more rigorously
> defined by the validation machinery in
> piker.data.validate.
>
> give me a mega detailed plan on how to approach
> this, and a staged approach for the implementation.
Plan-shaping user decisions (in-session Q&A): sibling
topology under `pikerd`; hard cutover staged by layer;
trading-only EMS-lazy-spawned `brokerd`. The human
staged the doc copy into `ai/claude-code/plans/` and
requested this commit after the 7 implementation
commits landed.
## Response summary
The staged design doc for the `brokerd` -> (`datad` +
`brokerd`) split: context, target supervision topology,
load-bearing verified facts, per-stage file-level
changes with gates, risk register and an end-to-end
verification matrix. Produced via 3 parallel explore
agents + 1 architect agent + human Q&A before any code
was written; all 7 implementation commits in this
branch reference its stages in their provenance
entries.
## Files changed
- `ai/claude-code/plans/datad_service.md` — the design
plan doc (human-staged copy of the AI-authored plan)
## Human edits
None to the doc content — committed as generated. NB:
the implementation deviated from the plan-as-written in
4 places, see the raw file's deviation log.

View File

@ -1,51 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T17:33:09Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T173309Z_f15f8178_prompt_io.md
---
NOTE: diff-ref mode entry; the AI output here is the
committed doc itself (a design plan, not code) so no
verbatim copy is duplicated:
> `git log -1 -p --follow -- ai/claude-code/plans/datad_service.md`
## Implementation deviation log
Where the landed 7-commit series differs from the plan
as written (recorded for reviewer transparency; the
plan doc is committed unmodified):
1. Commit ordering: the `pytest` config-dir isolation
fix lands BEFORE the `tractor`-API drift port. The
plan's stage gates assumed a runnable+isolated test
baseline; per-commit gating exposed that without
isolation the paper-EMS test reads the user's real
(polluted) `account.kraken.paper.toml` and reddens.
2. The plan's "stage 0: full pytest green" gate
required first repairing pre-existing branch
breakage vs `tractor` git `main` (boot
`AttributeError`, stale discovery/exc/position
APIs) — that repair became its own commit
("Port service+tests to latest `tractor` APIs")
rather than plan-stage work.
3. Stage-4 fail-fast placement: the plan said
`broker_init()` raises on brokerd-ep-less backends;
implementation moved the raise to `spawn_brokerd()`
since `piker ledger` calls `broker_init()` directly
even for paper accounts on datad-only backends.
4. `brokers/_daemon.py` change grouping: the
import-cleanup hunks (`exceptiongroup`, `_FeedsBus`
type-only import) landed with the caps-sec slim
commit instead of the fixture-slim commit, keeping
each intermediate tree import-clean without
sub-hunk surgery.
Also of record: the plan's "Verification" matrix was
executed as written (per-suite gates each stage, the
headless datad-feed smoke, the kucoin fail-fast unit
check); the known pre-existing ~50% second-runtime-boot
test wedge was characterized and excluded as a
regression via revert-testing.

View File

@ -1,60 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
session: 32d15f9a-b2d3-4c26-bdc9-190219141a25
timestamp: 2026-06-10T21:35:49Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T213549Z_f084e899_prompt_io.md
scope: code
substantive: true
raw_file: 20260610T213549Z_f084e899_prompt_io.raw.md
---
## Prompt
User bug report from live `ib` testing of the
(datad|brokerd)-split branch:
> ok doing some testing and noticing live orders do not
> work otb, pikerd show this on live submission,
> [pdb traceback: `RuntimeError("Can not order
> {symbol}, no live feed?")` at `ib/api.py:1152`
> `submit_limit()` with `self._contracts == {}` in the
> `brokerd.ib` actor + a `TrioTaskExited` teardown
> cascade]
Follow-up user direction fixing the design intent:
> i think the main thing here is that [brokerd] should
> be able to always submit orders without a live feed
> being up and registered in the same subactor right?
## Response summary
Root cause: pre-split the feed code populated the
actor-local `Client._contracts` cache (via
`get_mkt_info()` writing `mkt.bs_fqme` keys) in the
SAME process as order submission; post-split the
trading actor's client is never warmed. Fix: lazily
qualify + cache the contract per order request in
`handle_order_requests()` by running the same
`get_mkt_info(fqme, proxy=...)` ep the feed side uses,
plus per-order error relay (`BrokerdError`) so one bad
submission can't crash the whole trades dialog (the
`TrioTaskExited` storm was teardown cascade from the
original raise).
## Files changed
- `piker/brokers/ib/broker.py` — thread `proxies` into
`handle_order_requests()`; lazy contract qualify on
cache-miss; guard `submit_limit()` w/ `BrokerdError`
relay; uncomment the (anticipatory) `get_mkt_info`
import
- `piker/brokers/ib/api.py` — fix the non-f-string
raise msg + document the new qualification contract
## Human edits
None — committed as generated. Live `ib` order retest
performed by the human post-commit.

View File

@ -1,57 +0,0 @@
---
model: claude-fable-5[1m]
service: claude
timestamp: 2026-06-10T21:35:49Z
git_ref: datad_service
diff_cmd: git log -1 -p --follow -- ai/prompt-io/claude/20260610T213549Z_f084e899_prompt_io.md
---
NOTE: diff-ref mode entry (code committed in the same
commit as this log); recorded from the live debug
session per the `/prompt-io` skill rules.
> `git log -1 -p --follow -- piker/brokers/ib/broker.py`
> `git log -1 -p --follow -- piker/brokers/ib/api.py`
Key diagnostic chain (from session):
- pdb showed `Client._contracts == {}` inside
`brokerd.ib`'s `submit_limit()`; the cache has
exactly TWO write sites: `Client.find_contracts()`
(api.py, keys `f'{sym}.{exch}.ib'`) and
`symbols.get_mkt_info()` (key `mkt.bs_fqme`, eg.
'nvda.nasdaq' — NO `.ib` suffix).
- `BrokerdOrder.symbol` arrives as the bs_fqme form
('nvda.nasdaq') so ONLY the `get_mkt_info()` write
site produces the key `submit_limit()` reads —
ie. pre-split it was the feed's in-proc
`get_mkt_info(sym, proxy=proxy)` call keeping orders
working, NOT `find_contracts()`.
- the existing TODO at `symbols.py:642-644` literally
predicted this: "this is going to be problematic
if/when we split out the datad vs. brokerd actors
since the mktmap lookup table will now be
inaccessible.."
- instance identity verified: `proxy._aio_ns` IS the
same `Client` obj as `_accounts2clients[account]`
(both sourced from the `load_aio_clients()` cache via
`open_client_proxies()`), so a brokerd-side
`get_mkt_info(fqme, proxy=proxies[account])` warms
exactly the dict `submit_limit()` reads. It also
populates `client._cons2mkts` which the
position-audit path (`broker.py` backup-table code)
needs in this actor anyway.
- the `TrioTaskExited` storm in the user's log
(`recv_trade_updates`, `open_aio_client_method_relay`
aio tasks) is teardown cascade: the raise crashed
`handle_order_requests` -> nursery teardown ripped
the trio sides of still-running aio relay tasks.
Hence the added per-order try/except ->
`BrokerdError` relay hardening so a single bad
submission degrades to an EMS error msg instead of
killing the backend's entire order-ctl dialog.
Verification: `tests/test_services.py` (5 passed) +
`tests/test_ems.py` (6 passed) regression-green; live
`ib` submission retest delegated to the human (needs a
running TWS/gw).

View File

@ -1,34 +0,0 @@
# AI Prompt I/O Log — claude
This directory tracks prompt inputs and model
outputs for AI-assisted development using
`claude` (claude-code CLI).
## Policy
Prompt logging follows the
[NLNet generative AI policy][nlnet-ai].
All substantive AI contributions are logged
with:
- Model name and version
- Timestamps
- The prompts that produced the output
- Unedited model output (`.raw.md` files)
[nlnet-ai]: https://nlnet.nl/foundation/policies/generativeAI/
## Usage
Entries are created by the `/prompt-io` skill
or automatically via `/commit-msg` integration.
Each commit carrying AI-generated changes links
to its provenance entry via a `Prompt-IO:`
commit-msg trailer; entries use "diff-ref mode"
(pointers into `git log -p` instead of verbatim
code copies) to avoid duplicating committed
code.
Human contributors remain accountable for all
code decisions. AI-generated content is never
presented as human-authored work.

View File

@ -1,36 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
session: fsp-session-recovery-takeover-20260915
timestamp: 2026-09-15T15:52:29Z
git_ref: fadab3d2
scope: code
substantive: true
raw_file: 20260915T155229Z_fadab3d2_prompt_io.raw.md
---
## Prompt
Recover the interrupted OpenCode FSP session and take over its ongoing
work. The recovered child instruction records human authorization for
the minimal EMA omitted-default fix followed by a complete commit plan.
## Response summary
Removed the eager Numba signature that rejected Python's omitted
default arguments. Preserved nopython/nogil compilation and numerical
behavior. Added nine numerical regressions and verified all 38 focused
FSP cases. Rebuilt planning artifacts separately from production code.
## Files changed
- `piker/fsp/_momo.py`: drop the eager signature and unused type imports.
- `tests/test_fsp_momo.py`: omitted/explicit argument and seed coverage.
## Human edits
The human identified the omitted-default limitation in the earlier
session, authorized the minimal lazy-specialization approach, and
requested this takeover after OpenCode failed. Existing source changes
and older OpenCode provenance were retained. The new fix remains
unstaged and awaits human review; no acceptance is inferred.

View File

@ -1,50 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
timestamp: 2026-09-15T15:52:29Z
git_ref: fadab3d2
diff_cmd: git diff HEAD~1..HEAD
---
## Prompt
> i'm thinking maybe you can just take over that work and we can
> abandon that oc dialog for now?
Recovered child-session instructions describe the earlier human
authorization to remove EMA's eager Numba signature while retaining
`jit(nopython=True, nogil=True)`, verify defaults, and finish the commit
plan without staging or committing. This is recovered context, not a
verbatim quote of the earlier human authorization.
## Response
> `git diff HEAD~1..HEAD -- piker/fsp/_momo.py`
Removed EMA's eager Numba signature and now-unused Numba type imports.
Retained nopython/nogil compilation, the numerical kernel, historical
comments, and pre-existing staged and unstaged edits.
> `git diff HEAD~1..HEAD -- tests/test_fsp_momo.py`
Added production-dispatcher regressions for omitted positional and
keyword arguments, explicit None and numeric parameters, and the
previously corrected one-sample continuation. Expected outputs are
hand-computed recurrences, rather than the Python implementation.
Before the fix, the first case failed with:
```text
TypeError: No matching definition for argument type(s)
array(float64, 1d, C), omitted(default=None), omitted(default=None)
```
After the fix, the focused run outside the socket-restricted sandbox
reported:
```text
38 passed in 0.66s
```
Ruff passed for the modified momentum module and new regression file.
Combined-worktree results do not attest to isolated commit boundaries.

View File

@ -1,64 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
timestamp: 2026-09-23T16:46:56.830897+00:00
git_ref: efa72980
session: 01a0a5b9-f872-7ca3-ab8a-d1cffc14e895
scope: code
substantive: true
raw_file: 20260923T164656Z_efa72980_prompt_io.raw.md
---
## Prompt
Investigate existing branch fixes before repairing IB day-skipping.
Create an isolated worktree on ib_connectivity_harden and add a test
that checks actual persisted data, leaving live gap repair for later.
## Response summary
Preserve IB request boundaries on HMDS no-data responses. Retain the
existing earliest-history classification and bounded reset handling.
Separate requested and published boundaries during backfill so the
first response after an empty interval does not lose its end bar.
Prove supplied bars survive both SHM publication and a fresh parquet
load while a simulated venue closure remains empty.
## Files changed
- `piker/brokers/ib/feed.py`: complete HMDS no-data requests unchanged.
- `piker/tsp/_history.py`: filter overlap against published history.
- `tests/test_ib_history.py`: six empty-response boundary cases.
- `tests/test_history_backfill.py`: two SHM/parquet integration cases.
## Human edits
The human identified the live gap, required prior-branch discovery,
selected ib_connectivity_harden as the base, and requested stronger
end-to-end evidence. They corrected overstated reproduction claims,
insisted on plain explanations, and pointed out that valid venue
closures must remain gaps. Existing data repair was explicitly deferred.
No direct human source edits have been observed; changes await review.
## Review follow-up
The human's tuicr review asked whether None was part of the backfill
protocol and requested plain explanations of the two dates and overlap
filter. Clarified the IB-internal empty-response sentinel and its
translation into NoData/DataUnavailable before reaching the backfiller.
Renamed last_start_dt to request_end_dt and published_start_dt to
oldest_shm_dt. Added a concrete empty-interval example at the filter and
explained that only successful SHM writes move oldest_shm_dt.
These changes clarify naming and documentation without changing the
previously tested request or publication behavior.
## History-filter review follow-up
The human flagged the unused append_until_dt argument and confusing
append/prepend terminology in diff_history(). The helper had one
caller, using only prepend_until_dt. Removed the helper and inlined
its inclusive lower-time cutoff beside the SHM duplicate filter.
The human's nearby TODO comment is resolved by this removal; the
replacement comment describes the retained timestamps directly.
The storage comparison remains >= backfill_until_dt; the SHM
comparison remains < oldest_shm_dt.

View File

@ -1,87 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
timestamp: 2026-09-23T16:46:56.830897+00:00
git_ref: efa72980
diff_cmd: git diff efa72980 -- piker/brokers/ib/feed.py piker/tsp/_history.py tests/test_ib_history.py tests/test_history_backfill.py
---
## Prompt
The human requested the IB request-boundary repair in a new worktree
based on ib_connectivity_harden and a test exercising real persistence.
The existing MNQ parquet gap is outside this repair scope.
## Code output
> `git diff efa72980 -- piker/brokers/ib/feed.py`
> `git diff efa72980 -- piker/tsp/_history.py`
> `git diff efa72980 -- tests/test_ib_history.py`
> `git diff efa72980 -- tests/test_history_backfill.py`
get_bars() now completes HMDS no-data responses without changing the
requested date. open_history_client() retains its NoData and earliest
history classification. start_backfill() tracks the published boundary
separately from the request cursor after empty intervals.
## Unedited progress output
The new test failed on the original code as intended: the
temporary parquet contained only 4 bars instead of 2,004.
The day-skip caused 2,000 available bars to be omitted.
The test also includes a simulated venue closure that must remain
a gap. It checks that we fetch available bars without inventing
data for an empty interval.
The IB fix now preserves all 2,004 bars in parquet. The SHM
assertion exposed a second bug: after stepping over an empty
interval, the backfiller drops the last bar of the next response
as though it were already published.
I’ll separate the request date from the published-data boundary
so the same test can verify that both memory and disk retain
every available bar.
## Validation
Using the existing fsp_backfill_sync/py313 interpreter, with cwd in
ib_history_request_bounds and the Piker import checked against that cwd:
- tests/test_history_backfill.py: 11 passed in 1.92s.
- tests/test_ib_history.py: 17 passed in 0.36s.
- Ruff on all four changed Python files: all checks passed.
- git diff --check: passed.
The integration test uses simulated IB bars responses, real IB history
adapter and backfill logic, real SHM, and temporary NativeDB parquet.
It checks a fresh disk load. It does not test a live gateway, actor RPC,
or Qt rendering, and does not prove the origin of the user's live gap.
## Unedited review response
`None` is internal to the IB adapter. The backfiller receives
`NoData`, or `DataUnavailable` when the requested date has reached
the known beginning of history. I’ll make that distinction explicit.
The two dates answer different questions: “where should the next
history query end?” and “what is the oldest bar already in SHM?”
An empty reply changes only the first. I’ll rename them to
`request_end_dt` and `oldest_shm_dt` and explain the filtering with
a concrete example.
> `git diff efa72980 -- piker/brokers/ib/feed.py piker/tsp/_history.py`
## History-filter review follow-up
The human flagged the unused append_until_dt argument and confusing
append/prepend terminology in diff_history(). The helper had one
caller, using only prepend_until_dt. Removed the helper and inlined
its inclusive lower-time cutoff beside the SHM duplicate filter.
The human's nearby TODO comment is resolved by this removal; the
replacement comment describes the retained timestamps directly.
The storage comparison remains >= backfill_until_dt; the SHM
comparison remains < oldest_shm_dt.

View File

@ -1,51 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
timestamp: 2026-09-24T16:57:57.342784+00:00
git_ref: 1da9b760
scope: code
substantive: true
raw_file: 20260924T165757Z_1da9b760_prompt_io.raw.md
---
## Prompt
Document and test the existing null-segment repair, including its
flat-bar fallback, while distinguishing packed timestamp gaps.
## Response summary
Documented SHM-only repair, provider placement assumptions, fallback
fields, notifications, and concurrency limits. Added boundary and
no-history coverage and fixed epoch-dependent comparison tolerance
which could synthesize a duplicate timestamp at the right margin.
## Files changed
- piker/tsp/_history.py: helper contract and absolute tolerance.
- tests/test_history_backfill.py: four additional parameter/case runs.
## Human edits
The human identified cancellation as a source of stored gaps,
correctly described packed indexes and insertion costs, pointed to
maybe_fill_null_segments(), and requested immediate documentation
and tests. No direct source edits observed during this change;
the patch remains uncommitted for review.
## Close-copy factoring follow-up
The human requested stronger internal types, a reusable close-copy
helper, searchable comments for copying close and synthesizing
stamps, and research into historical venue closure queries.
Extracted _copy_close_to_null_rows(), preserving non-OHLC fields and
rejecting occupied slots before mutation. Added explicit callback,
array, segment-index and date annotations. Preserved the existing
provenance TODO and suggested avoiding volume sentinels.
Four helper cases cover successful fill and unchanged rejected
inputs. Backfill suite: 19 passed in 2.01s; Ruff and diff check pass.
IB documents historical SCHEDULE requests separately from contract
hours. The installed ib_async exposes reqHistoricalScheduleAsync.
No live schedule request or venue behavior change was performed.
Source: https://interactivebrokers.github.io/tws-api/historical_bars.html

View File

@ -1,48 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
timestamp: 2026-09-24T16:57:57.342784+00:00
git_ref: 1da9b760
diff_cmd: git diff 1da9b760 -- piker/tsp/_history.py tests/test_history_backfill.py
---
## Prompt
The human requested documentation and tests for
maybe_fill_null_segments(), identifying its flat OHLC fallback
and asking how packed SHM could accept later recovered bars.
## Code output
> `git diff 1da9b760 -- piker/tsp/_history.py`
> `git diff 1da9b760 -- tests/test_history_backfill.py`
Document the existing slot-based repair and its limitations.
Extend fallback coverage to NoData and DataUnavailable; verify
the observed packed MNQ gap remains unchanged. Reproduce the
relative-tolerance failure at a modern epoch, then disable relative
tolerance in the synthetic timestamp boundary check.
## Validation
Before the fix: modern-epoch boundary case failed, zero-epoch
case passed. The overfull synthetic run duplicated the right bar.
After the fix: 15 backfill tests passed in 1.99s.
## Close-copy factoring follow-up
The human requested stronger internal types, a reusable close-copy
helper, searchable comments for copying close and synthesizing
stamps, and research into historical venue closure queries.
Extracted _copy_close_to_null_rows(), preserving non-OHLC fields and
rejecting occupied slots before mutation. Added explicit callback,
array, segment-index and date annotations. Preserved the existing
provenance TODO and suggested avoiding volume sentinels.
Four helper cases cover successful fill and unchanged rejected
inputs. Backfill suite: 19 passed in 2.01s; Ruff and diff check pass.
IB documents historical SCHEDULE requests separately from contract
hours. The installed ib_async exposes reqHistoricalScheduleAsync.
No live schedule request or venue behavior change was performed.
Source: https://interactivebrokers.github.io/tws-api/historical_bars.html

View File

@ -1,91 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
timestamp: 2026-09-24T18:54:56.893490+00:00
git_ref: 1da9b760
scope: code
substantive: true
raw_file: 20260924T185456Z_1da9b760_schedule_prompt_io.raw.md
---
## Prompt
The human requested: first freeze a commit-plan for the existing
close-copy/null-repair changes, then implement ib_async historical
schedule support with tests, use it for datad.ib history-buffer gap
spacing, and plan a common on-demand segment interface for CLI and
chart selection. Avoid reindexing in this first pass and record
fallback insertion only as a TODO.
## Response summary
Implemented historical schedule normalization, bounded startup
reservations, explicit actor-local synthetic ranges, and two read
RPC functions. Added deterministic adapter tests and real SHM/parquet
startup coverage. Wrote the segment-backfill interface plan.
This entry covers the new schedule work only. The earlier frozen
null-repair plan has its own provenance. The shared _history.py file
contains changes from both scopes until the earlier commit is made.
## Human edits
The human identified cancellation during backfill as the reproducible
trigger, explained the packed-index insertion tradeoff, requested
reuse of venue closure knowledge, and chose placeholders to avoid
expensive rewrites. They required explicit placeholder annotation,
proposed later graphics compression, and asked to defer insertion
fallbacks. They also requested the existing null-repair commit plan
before the new schedule work. These are human-directed architecture
and scope choices; no acceptance or task completion state is inferred.
## Validation and limits
55 targeted tests passed; Ruff and diff checks passed. No live IB
qualification was performed. Reservations are synthetic, not recovered
observations. Repair execution, persistence of unfinished requests,
chart compression, FSP handling, export filtering, and request-level
IB cancellation remain follow-up work documented in the plan.
## Files changed
- `piker/brokers/ib/venues.py`
- `piker/brokers/ib/api.py`
- `piker/brokers/ib/feed.py`
- `piker/tsp/_reserve.py`
- `piker/tsp/_history.py`
- `tests/providers/test_ib_schedule.py`
- `tests/providers/test_history_reservations.py`
- `docs/history_segment_backfill.rst`
## Pre-commit module review
The human requested placing historical schedule normalization in the
existing venues.py module, naming the allocation module _reserve.py,
and adding the missing project license headers. Applied those moves,
updated imports and test patch targets, and added the standard AGPL
headers to the new allocation module and both new test modules.
## Local tuicr review follow-up
The human supplied eleven comments requesting a configurable newest
gap count, caller and metadata references, explanations of guards
and provider-specific limits, multiline boolean style, capacity
fallback rationale, an explanation of numpy split boundaries, and
clearer test cases, fixture types, and module docstrings.
Added newest_gaps=32, with zero disabling queries and negative values
rejected. Documented tsdb_backfill ownership and the local/RPC range
accessors. Explained recoverable overflow and half-open synthetic
range grouping. Typed tmp_path as pathlib.Path and monkeypatch as
pytest.MonkeyPatch. Added four request-budget cases.
Verification: reservation tests 11 passed; schedule tests 7 passed;
Ruff and git diff --check passed. Changes remain unstaged for review.
No forge review or CI state was queried; this was local feedback.
## Provider-test layout review
The human requested grouping both new history/schedule test modules
under tests/providers. Moved both files without changing test logic;
they retain the shared tests/conftest.py fixtures.

View File

@ -1,100 +0,0 @@
---
model: gpt-6 (exact variant not exposed)
service: codex
timestamp: 2026-09-24T18:54:56.893490+00:00
git_ref: 1da9b760
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/brokers/ib/_schedule.py`
> `git diff HEAD~1..HEAD -- piker/brokers/ib/api.py`
> `git diff HEAD~1..HEAD -- piker/brokers/ib/feed.py`
> `git diff HEAD~1..HEAD -- piker/tsp/_reservations.py`
> `git diff HEAD~1..HEAD -- piker/tsp/_history.py`
> `git diff HEAD~1..HEAD -- tests/test_ib_schedule.py`
> `git diff HEAD~1..HEAD -- tests/test_history_reservations.py`
> `git diff HEAD~1..HEAD -- docs/history_segment_backfill.rst`
## Generated output
Added Client.history_schedule() over the installed ib_async API,
normalizing dated sessions to UTC with explicit response coverage.
Reject ambiguous DST times, reversed sessions, and overlapping
sessions. Match the existing extended-hours bars policy.
Added startup-only schedule reservations to tsdb_backfill before
prepending older stored history. Reserve open-session sample slots
only where the reply covers the full candidate gap. Return synthetic
UTC ranges separately from OHLCV; copy preceding close into flat
placeholders without writing those placeholders to parquet.
Unknown coverage and capacity failures retain packed time gaps.
Added datad.ib schedule and reservation metadata RPC functions and
a proposed shared CLI/chart segment request protocol. The repair
queue, durable unresolved intervals, chart compression, FSP handling,
and coordinated insertion remain deferred. No live data was repaired.
Verification is complete: 55 targeted tests pass, including the
new schedule and real SHM/parquet tests. I also added checks for
reversed session bounds and the 25-hour day at the autumn DST
change.
The segment API plan uses UTC ranges for both CLI and chart
requests, a provider-owned queue, durable unfinished intervals,
and replacement of reserved slots without reindexing. Live IB
schedule coverage still needs qualification.
## Limits
Only stored-history gaps are reserved before prepend; published rows
are never shifted. Newest 32 gaps, at most 31 days each, are queried
serially with a 10-second schedule timeout. This can delay history
publication. Underlying IB request cancellation is still a TODO.
Manual whole-SHM export requires synthetic-range exclusion before
it can preserve provider-only storage. The chart currently displays
flat placeholders, and synthetic-aware FSP treatment is deferred.
Tests: history backfill 19, IB history 17, method proxy 5, schedule 7,
reservations 7. Sandbox local-socket failures were rerun with approval.
Ruff and git diff --check passed. No live broker requests were made.
## Module-layout review response
Schedule normalization now lives in piker.brokers.ib.venues beside
the existing session helpers. The allocation module is tsp._reserve.
Added the project AGPL headers to the new module and test files.
Earlier diff pointers record the original generated paths; use
the following paths for the reviewed layout:
> `git diff HEAD~1..HEAD -- piker/brokers/ib/venues.py`
> `git diff HEAD~1..HEAD -- piker/tsp/_reserve.py`
## Local review response
The fixes pass all 18 tests in the two affected files.
The new limit tests cover zero, negative, one-gap, and two-gap
budgets, including newest-first selection.
Capacity overflow remains a warning: no partial reservation is
published for that gap, and the original observations remain
available. The comments now explain that fallback explicitly.
Diff pointers above include the review edits to tsp._reserve, both
test modules, and the interface document. This review did not query
a forge or publish replies.
## Provider-test paths
Moved both new test modules under tests/providers as requested.
> `git diff HEAD~1..HEAD -- tests/providers/test_ib_schedule.py`
> `git diff HEAD~1..HEAD -- tests/providers/test_history_reservations.py`

View File

@ -1,45 +0,0 @@
---
model: gpt-6
service: codex
timestamp: 2026-09-24T23:24:05.525363+00:00
git_ref: 441358e2
scope: code
substantive: true
raw_file: 20260924T232405Z_441358e2_schedule_coverage_prompt_io.raw.md
---
## Prompt
User: "do it", approving the proposed bounded wider-window retry
for incomplete schedule coverage and its MNQ regression test.
Previous live qualification showed a one-day request began after
the gap's left edge; a wider query covered it and excluded the
one-hour venue closure.
## Response summary
Added bounded backwards widening under one timeout, preserving actual
coverage and fixed end bounds. Added captured-response allocation,
unknown-coverage, and retry-timeout tests, and updated API documentation.
39 selected tests and Ruff passed. Live data was not changed.
## Files changed
- piker/brokers/ib/api.py — bounded schedule coverage retries.
- tests/providers/test_ib_schedule.py — captured MNQ and retry regressions.
- docs/history_segment_backfill.rst — retry contract and limits.
## Human edits
The user established schedule-driven allocation, explicit unknown
coverage, and avoiding reindexing as the design constraints, supplied
the MNQ gap, and authorized this focused coverage fix. No direct human
source edits occurred during this patch; review is pending.
## Follow-up review
The user requested factoring and documenting the repeated 86400
conversion. Added piker.calc.SECONDS_PER_DAY, documented elapsed
versus calendar/session days once, and reused it in the IB query
and reservation span check. Named the client retry limit locally.
Schedule tests: 11 passed; Ruff passed after this adjustment.

View File

@ -1,66 +0,0 @@
---
model: gpt-6
service: codex
timestamp: 2026-09-24T23:24:05.525363+00:00
git_ref: 441358e2
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/brokers/ib/api.py`
> `git diff HEAD~1..HEAD -- tests/providers/test_ib_schedule.py`
> `git diff HEAD~1..HEAD -- docs/history_segment_backfill.rst`
Client.history_schedule() widens incomplete left coverage by doubling
numDays up to 31 while preserving endDateTime and actual returned
coverage. All attempts share an asyncio timeout. Missing right coverage
returns unchanged because extending the lookback cannot repair it.
Tests replay the captured MNQ schedules through the real client and
reservation allocator, check bounded incomplete coverage and retry
cancellation under one timeout. Documentation describes these limits.
## User-facing output
The captured MNQ case now passes through the real client method
and allocator: the request widens from one day to two, then
reserves 51,824 slots while leaving the venue closure compressed.
The tests also cover permanently incomplete replies and a timeout
during the retry. I’m running the related regression suites now.
All 39 selected tests passed, and Ruff passed. The reservation
suite needed a retry outside the sandbox because Trio’s local
wakeup socket was blocked.
The fix and regression tests are ready for review. I’m recording
the prompt provenance; live SHM and parquet remain unchanged.
## Validation commands and results
Worktree-local py313 in the existing Nix environment; import piker
resolved inside this worktree.
- py313/bin/python -m pytest -p no:xonsh -q -x --tb=short --no-header tests/providers/test_ib_schedule.py tests/providers/test_history_reservations.py
Sandbox: 11 passed then PermissionError creating Trio wakeup socket.
Escalated rerun: 22 passed in 1.02s; fake IB, disposable test SHM.
- py313/bin/python -m pytest -p no:xonsh -q -x --tb=short --no-header tests/test_ib_history.py
17 passed in 0.69s.
- ruff check piker/brokers/ib/api.py tests/providers/test_ib_schedule.py
All checks passed (existing deprecated configuration warning).
- git diff --check: passed.
No live service restart, live history repair, staging, or commit.
## Follow-up user review and generated changes
User: "feels like this same magic # from the prior commit should be
factored and commented once? `num_days: int = max(1, ceil(seconds /
86400))` other then that all looks decent"
> `git diff HEAD~1..HEAD -- piker/calc.py piker/tsp/_reserve.py`
Added SECONDS_PER_DAY for both consumers, with one comment explaining
fixed elapsed days versus calendar days and trading sessions. Named
the client's 31-day retry limit locally. After this change the
schedule suite passed 11 tests and Ruff passed all four Python files.

View File

@ -1,10 +0,0 @@
# AI Prompt I/O Log — codex
This directory tracks prompts and outputs for AI-assisted development
using Codex. Substantive contributions include model identification,
prompt context, output records, and human contribution accounting.
Code output uses Git diff references; non-code output is retained in
the paired `.raw.md` file. Human contributors remain responsible for
review and acceptance. Entries follow the
[NLNet generative AI policy](https://nlnet.nl/foundation/policies/generativeAI/).

View File

@ -1,41 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-21T21:50:55Z
git_ref: c6ec3d41
scope: docs
substantive: true
raw_file: 20260721T215055Z_c6ec3d41_prompt_io.raw.md
---
## Prompt
The user asked to boot a plan to continue work on the
time-series gap checker and bug fixing of the history
backfiller. The user authorized prompt-I/O capture for the
plan documentation after the repository audit.
## Response summary
Produced a bounded repair plan based on current source,
historical issue caches, existing profiling notes, and a
test-coverage audit. The plan prioritizes reproducible
correctness failures and storage durability before venue
classification, annotation performance, or manual repair
UX.
## Files changed
- `plans/opencode/gap-checker-backfiller-repair.md` -
phased implementation and verification plan
- `ai/prompt-io/opencode/20260721T215055Z_c6ec3d41_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260721T215055Z_c6ec3d41_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated plan has not yet been edited by the
human.

View File

@ -1,26 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-21T21:50:55Z
git_ref: c6ec3d41
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff -- plans/opencode/gap-checker-backfiller-repair.md`
Generated a phased implementation plan for continuing
piker's gap checker and history backfiller repairs. The
plan separates data correctness from annotation
performance and orders work by dependency: deterministic
tests, canonical frame/gap normalization, structured
backfill completion, non-destructive atomic NativeDB
writes, exact null repair, provider/venue classification,
annotation vectorization, and crash qualification.
The plan is grounded in current code paths, local issue
caches `#62`, `#71`, and `#75`, and the existing 1,285-gap
profiling handoff. It records confirmed present defects,
design invariants, per-phase acceptance gates, test
commands, commit boundaries, and explicitly deferred UX
and provider work.

View File

@ -1,46 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-22T00:13:57Z
git_ref: 16f3cd6f
scope: code
substantive: true
raw_file: 20260722T001357Z_16f3cd6f_prompt_io.raw.md
---
## Prompt
Continue the gap-checker and history-backfiller repair plan. Start
with deterministic analysis characterization, then fix canonical
sorting, dedupe-derived columns, positive gap detection, and null
grouping.
During review, the user challenged the replacement null-grouping
algorithm because its initial tests did not assert exact endpoints.
The user explicitly required retaining detailed, human-oriented
algorithm documentation such as the existing ASCII segment map.
## Response summary
Added deterministic analysis regressions and corrected normalization
so sorting and duplicate resolution precede derived-column
calculation. Unified null grouping for NumPy and Polars inputs,
documented its exact inclusive-boundary contract with an adapted
visual map, and verified all small null layouts against an independent
state-machine oracle. Retained sorted smart-dedupe output.
## Files changed
- `piker/tsp/_anal.py` - canonical gap and documented null grouping
- `piker/tsp/_dedupe_smart.py` - sorted duplicate resolution
- `tests/test_tsp_analysis.py` - exact and exhaustive regressions
- `ai/prompt-io/opencode/20260722T001357Z_16f3cd6f_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260722T001357Z_16f3cd6f_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not yet been edited by the human.

View File

@ -1,40 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-22T00:13:57Z
git_ref: 16f3cd6f
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/tsp/_anal.py`
Generated canonical gap and null-segment normalization. Derived
datetime and delta columns are recomputed after final sorting and
duplicate resolution, and negative deltas no longer become ordinary
gaps.
NumPy and Polars null grouping share one implementation while
retaining the human-oriented visual map of contiguous zero runs. The
documented contract defines absolute inclusive endpoints, margin
expansion, frame boundary clamping, contiguous-index validation, and
requested-column selection.
> `git diff HEAD~1..HEAD -- piker/tsp/_dedupe_smart.py`
Adjusted smart dedupe to honor sorted output and resolve duplicate
timestamps without retaining stale order.
> `git diff HEAD~1..HEAD -- tests/test_tsp_analysis.py`
Added deterministic synthetic OHLCV coverage for sorted dedupe, fresh
derived columns, and positive gap detection. Null grouping is checked
against readable endpoint examples and an independent state-machine
oracle over every layout through six rows, three margin widths, and
both NumPy and Polars inputs.
Verification generated with the patch:
`timeout -k 5 30 python -m pytest -q tests/test_tsp_analysis.py`
Result: 11 passed.

View File

@ -1,36 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-22T00:13:58Z
git_ref: 16f3cd6f
scope: code
substantive: true
raw_file: 20260722T001358Z_16f3cd6f_prompt_io.raw.md
---
## Prompt
Continue the structured backfill phase after the analysis regressions.
Characterize provider exhaustion and genuinely empty frames, then remove
completion paths that can hang after child-task exit.
## Response summary
Removed orphan-prone completion events from the structured nursery flow
and checked empty provider frames before indexing timestamp endpoints.
Added deterministic timeout-bounded regressions for both exit paths.
## Files changed
- `piker/tsp/_history.py` - structured completion and empty-frame fixes
- `tests/test_history_backfill.py` - deterministic exit regressions
- `ai/prompt-io/opencode/20260722T001358Z_16f3cd6f_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260722T001358Z_16f3cd6f_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not yet been edited by the human.

View File

@ -1,25 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-22T00:13:58Z
git_ref: 16f3cd6f
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/tsp/_history.py`
Removed redundant completion events whose consumers waited only after
their owning nurseries had joined. Added empty provider-frame handling
before timestamp endpoint indexing so provider exhaustion exits cleanly.
> `git diff HEAD~1..HEAD -- tests/test_history_backfill.py`
Added bounded, actor-free regressions for `DataUnavailable` and empty
provider-frame completion.
Verification generated with the patch:
`timeout -k 5 60 python -m pytest -q tests/test_history_backfill.py`
Result: 2 passed.

View File

@ -1,39 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-22T02:41:33Z
git_ref: 3a4f6737
scope: code
substantive: true
raw_file: 20260722T024133Z_3a4f6737_prompt_io.raw.md
---
## Prompt
The user challenged the analysis commit message's claim that only
positive sample-step gaps are detected and asked whether out-of-order
timestamps were simply ignored. After confirming the missing separate
reporting path, the user instructed the agent to implement the fix.
## Response summary
Separated missing-history gaps from timestamp-ordering errors without
changing the existing `dedupe()` return shape. Added a structured
detector and made normalization warn before sorting removes evidence of
duplicates or reversed timestamps.
## Files changed
- `piker/tsp/_anal.py` - ordering detector and dedupe reporting
- `piker/tsp/__init__.py` - public detector export
- `tests/test_tsp_analysis.py` - exact ordering-error regression
- `ai/prompt-io/opencode/20260722T024133Z_3a4f6737_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260722T024133Z_3a4f6737_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not yet been edited by the human.

View File

@ -1,30 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-22T02:41:33Z
git_ref: 3a4f6737
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/tsp/_anal.py`
Added `detect_time_ordering_errors()` to return non-positive timestamp
steps as structured rows. `dedupe()` now warns with those rows before
sorting and duplicate removal erase the original ordering evidence.
> `git diff HEAD~1..HEAD -- piker/tsp/__init__.py`
Exported the ordering detector through the public TSP package.
> `git diff HEAD~1..HEAD -- tests/test_tsp_analysis.py`
Changed the negative-delta regression to verify exact predecessor,
timestamp, and delta values, confirm exclusion from missing-history
gaps, and require explicit normalization reporting.
Verification generated with the patch:
- `tests/test_tsp_analysis.py`: 11 passed
- Ruff: passed
- `git diff --check`: passed

View File

@ -1,43 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-22T02:41:34Z
git_ref: 3a4f6737
scope: code
substantive: true
raw_file: 20260722T024134Z_3a4f6737_prompt_io.raw.md
---
## Prompt
The user asked why the in-progress `piker/storage` changes were omitted
from the earlier commit plan. After receiving an inventory of unfinished
Phase 3 requirements, the user instructed the agent to complete the work.
## Response summary
Finished non-destructive, atomic NativeDB updates and integrated them
with history startup and reverse backfill. Expanded tests through two
adversarial review rounds that caught and corrected cross-client lock
blocking, stale temp indexing, discontinuous persisted indexes, startup
snapshot races, missing latest-frame persistence, notification wedges,
and mutable/truncated SHM persistence.
## Files changed
- `piker/storage/__init__.py` - incremental storage contract
- `piker/storage/nativedb.py` - validated atomic merge persistence
- `piker/storage/marketstore/__init__.py` - non-duplicate update mode
- `piker/tsp/_history.py` - provider-delta and startup persistence
- `tests/test_storage_nativedb.py` - NativeDB durability regressions
- `tests/test_history_backfill.py` - persistence lifecycle regressions
- `ai/prompt-io/opencode/20260722T024134Z_3a4f6737_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260722T024134Z_3a4f6737_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not yet been edited by the human.

View File

@ -1,43 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-22T02:41:34Z
git_ref: 3a4f6737
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/storage/__init__.py piker/storage/nativedb.py piker/storage/marketstore/__init__.py`
Added an incremental storage update contract. NativeDB now validates
canonical numeric OHLCV data, serializes writers both actor-locally and
through cancellable filesystem locks, merges with incoming-wins seam
resolution, regenerates contiguous indexes, reopens and validates a
same-directory temporary parquet, fsyncs it, atomically replaces the
target, fsyncs the directory, and publishes cache only after replacement.
NativeDB indexing ignores lock and crash-left temporary files.
MarketStore uses its non-duplicate write mode for incremental updates.
> `git diff HEAD~1..HEAD -- piker/tsp/_history.py`
Separated full provider deltas from capacity-truncated SHM pushes.
Backfill persists before SHM mutation, publishes bounded notifications,
and delays latest-frame publication until the concurrent pre-update
storage snapshot completes.
> `git diff HEAD~1..HEAD -- tests/test_storage_nativedb.py tests/test_history_backfill.py`
Added deterministic coverage for merge preservation, conflict policy,
atomic failures, reopened-candidate validation, cache consistency,
schema and timestamp invariants, contiguous index regeneration,
separate series, stale temp indexing, cancellable file-lock contention,
full provider-delta persistence, latest-frame ordering, and bounded
notification teardown.
Verification generated with the patch:
- deterministic analysis/backfill/storage set: 31 passed
- Ruff: passed
- `git diff --check`: passed
- final adversarial review: no findings

View File

@ -1,40 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: unavailable
timestamp: 2026-07-27T20:47:11Z
git_ref: 84a6d47b
scope: code
substantive: true
raw_file: 20260727T204711Z_84a6d47b_prompt_io.raw.md
---
## Prompt
The user identified that `backfiller_deep_fixes` introduced persistent
`.parquet.lock` files, questioned whether Parquet or Arrow already handled
writer coordination, and clarified that piker's `tractor` structured
concurrency design already provides single-writer ownership through the
persistent `datad` feed task. The user then asked to open the backfiller
worktree and resolve the redundant locking.
## Response summary
Removed redundant actor-local and filesystem locking while preserving
atomic crash-safe Parquet replacement. Replaced the synthetic contention
test with regressions for the actual cross-branch failure: writes create
no lock sidecars, and legacy lock artifacts are ignored during indexing.
## Files changed
- `piker/storage/nativedb.py` - rely on `datad` writer ownership
- `tests/test_storage_nativedb.py` - lock-sidecar regressions
- `ai/prompt-io/opencode/20260727T204711Z_84a6d47b_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260727T204711Z_84a6d47b_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,30 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-27T20:47:11Z
git_ref: 84a6d47b
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/storage/nativedb.py`
Removed actor-local and filesystem writer locks from NativeDB. The
persistent `datad` feed task already owns writes for each series, while
its timeframe children write distinct files. NativeDB retains validated
temporary writes, file and directory synchronization, and atomic target
replacement for crash durability. The exact `.parquet` suffix filter is
retained so legacy lock sidecars are harmless.
> `git diff HEAD~1..HEAD -- tests/test_storage_nativedb.py`
Removed the synthetic cross-client lock-contention test. Added regression
coverage proving replacement and incremental writes create no lock
sidecars. Extended indexing coverage with the legacy
`.parquet.lock` artifact that caused upstream `flake_update` to crash.
Verification generated with the patch:
- NativeDB and history regression set: 20 passed
- Ruff: passed
- `git diff --check`: passed

View File

@ -1,42 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-27T21:24:54Z
git_ref: 689df816
scope: code
substantive: true
raw_file: 20260727T212454Z_689df816_prompt_io.raw.md
---
## Prompt
The user reran the known IB FQME on `backfiller_deep_fixes` and reported
that `publish_latest_frame()` still failed because NativeDB required an
`index` absent from IB's provider frame. The user called out the missed
tabular schema boundary and requested an actual end-to-end test suite
rather than storage fakes.
## Response summary
Repaired the IB-provider-to-NativeDB boundary by canonicalizing durable
fields and mapping reloads into provider-specific SHM buffers. Added a
local actor/SHM/Parquet integration regression covering first startup and
restart, plus focused conversion and timestamp-coercion tests. Two review
rounds found and resolved restart hydration and fractional-time hazards.
## Files changed
- `piker/storage/nativedb.py` - canonical storage and SHM field mapping
- `piker/tsp/_anal.py` - name-based Polars-to-NumPy conversion
- `tests/test_history_backfill.py` - actor/SHM/Parquet integration test
- `tests/test_storage_nativedb.py` - schema conversion regressions
- `ai/prompt-io/opencode/20260727T212454Z_689df816_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260727T212454Z_689df816_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,34 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-27T21:24:54Z
git_ref: 689df816
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/storage/nativedb.py piker/tsp/_anal.py`
Normalized index-less provider frames to NativeDB's canonical names,
order, and dtypes before merge and publication. Provider-only columns
are excluded from durable files, derived indexes are regenerated, and
fractional timestamps are rejected before integer conversion. Added a
canonical field map for hydrating provider-specific SHM buffers while
leaving provider-only fields at their defaults. Changed Polars-to-NumPy
conversion to select columns by field name rather than position.
> `git diff HEAD~1..HEAD -- tests/test_history_backfill.py tests/test_storage_nativedb.py`
Added deterministic coverage using IB's actual provider dtype and a real
local `tractor` root actor, IB-typed shared-memory buffers, NativeDB,
Parquet persistence, first-start publication, and restart hydration.
Added focused regressions for reordered/extra Polars columns and
fractional timestamp rejection.
Verification generated with the patch:
- NativeDB and history integration set: 23 passed
- Ruff: passed
- `git diff --check`: passed
- second adversarial review: no findings
- no live gateway or network tests run

View File

@ -1,41 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-27T21:59:06Z
git_ref: 689df816
scope: code
substantive: true
raw_file: 20260727T215906Z_689df816_prompt_io.raw.md
---
## Prompt
After the actor/SHM/Parquet integration test exposed a stale SHM triplet,
the user requested leak-cleaner machinery in the pytest harness using
Tractor's existing ownership and lifetime patterns.
## Response summary
Added creator-scoped SHM leak tracking to pytest without scanning
`/dev/shm` or treating attachments as owned. The fixture cooperates with
normal Tractor teardown, verifies POSIX object identity before fallback
cleanup, restores process-local token state, and fails after cleaning a
leak. Regressions cover the exact pre-actor-registration allocation window,
positional creators, and external attachment safety.
## Files changed
- `tests/conftest.py` - creator-scoped SHM leak tracking fixture
- `tests/test_shm_cleanup.py` - ownership and failure-window regressions
- `.claude/skills/run-tests/test-harness-reference.md`
- document current-process SHM cleanup behavior
- `ai/prompt-io/opencode/20260727T215906Z_689df816_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260727T215906Z_689df816_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,25 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-27T21:59:06Z
git_ref: 689df816
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- tests/conftest.py tests/test_shm_cleanup.py .claude/skills/run-tests/test-harness-reference.md`
Added function-scoped SHM ownership tracking around Tractor's test-process
factory. The fixture records only successful creators, removes ownership
records during normal Tractor teardown, verifies POSIX object identity
before fallback unlink, restores the token-cache baseline, and fails after
cleaning any leak. Regressions reproduce allocation failure before actor
lifetime registration and prove external attachments remain untouched.
Verification generated with the patch:
- SHM cleanup regressions: 2 passed
- combined SHM, NativeDB, and history set: 25 passed
- Ruff: passed
- `git diff --check`: passed
- final adversarial review: no findings

View File

@ -1,42 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: 24fb9765-a550-4570-8350-f0fc9b7e17db
timestamp: 2026-07-28T00:32:47Z
git_ref: ed85721c
scope: code
substantive: true
raw_file: 20260728T003247Z_ed85721c_prompt_io.raw.md
---
## Prompt
Build a safe xonsh workflow for qualifying NativeDB and reverse-backfill
correctness against a known gappy IB chart, beginning with
`mnq.cme.20260918`. Cover baseline, fresh, restart, append, and expansion
testing through implementation, adversarial review, and verification.
## Response summary
Implemented a private disposable xonsh qualification workflow. Evidence is
tied to a fixed case, report checksum, snapshot checksum, worktree import,
and phase metadata. Added explicit path and symlink safety, source
replacement handling, xonsh child environments, and lifecycle preservation
comparisons.
## Files changed
- `snippets/nativedb_backfill_audit.xsh` - disposable qualification helpers
- `docs/manual_backfill_qualification.rst` - operator lifecycle runbook
- `tests/test_backfill_audit_snippet.py` - xonsh and comparison regressions
- `.claude/skills/run-tests/test-harness-reference.md`
- qualification test-harness guidance
- `ai/prompt-io/opencode/20260728T003247Z_ed85721c_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260728T003247Z_ed85721c_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,32 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-28T00:32:47Z
git_ref: ed85721c
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- snippets/nativedb_backfill_audit.xsh docs/manual_backfill_qualification.rst tests/test_backfill_audit_snippet.py .claude/skills/run-tests/test-harness-reference.md`
Added sourceable xonsh helpers that create a private disposable root, bind
child processes to the reviewed worktree, capture checksum-bound phase
evidence, safely seed and clear test history, and compare lifecycle
preservation.
The workflow rejects path traversal, pre-existing symlinks, output
collisions, mismatched case/report/snapshot evidence, and same-phase
comparison. The accompanying runbook covers known-bad replay, fresh
backfill, restart, append, gap triage, and expansion cases.
Verification generated with the patch:
- xonsh qualification helper regressions: passed
- xonsh sourceability: passed
- Ruff: passed
- `git diff --check`: passed
Adversarial review drove additional evidence binding, worktree executable,
environment propagation, source replacement, and descriptor-level safety
fixes. Residual concurrent mutation risk is stated explicitly in the private
disposable-root runbook.

View File

@ -1,40 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: 24fb9765-a550-4570-8350-f0fc9b7e17db
timestamp: 2026-07-28T00:50:34Z
git_ref: ed85721c
scope: code
substantive: true
raw_file: 20260728T005034Z_ed85721c_prompt_io.raw.md
---
## Prompt
Add reusable read-only tooling to qualify raw NativeDB Parquet against a
known gappy IB chart, beginning with `mnq.cme.20260918`. Preserve malformed
evidence and distinguish structural validity from unclassified positive
gaps without opening or mutating the storage runtime.
## Response summary
Implemented raw Parquet audit reporting with human and JSON output, bounded
cadence details, exact integer arithmetic, strict status, collision-safe
snapshot capture, and source path protections. Regressions cover malformed
schema and values, timestamp precision, gap arithmetic, corrupt bytes,
copy failures, symlinks, traversal, and read-only CLI behavior.
## Files changed
- `piker/storage/_audit.py` - raw Parquet audit and snapshot implementation
- `piker/storage/cli.py` - `piker store audit` command
- `tests/test_storage_audit.py` - audit and CLI regressions
- `ai/prompt-io/opencode/20260728T005034Z_ed85721c_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260728T005034Z_ed85721c_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,27 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-28T00:50:34Z
git_ref: ed85721c
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/storage/_audit.py piker/storage/cli.py tests/test_storage_audit.py`
Added a raw, read-only NativeDB Parquet audit with human and JSON output,
bounded cadence evidence, exact integer arithmetic, pre-parse snapshot
capture, and strict qualification status.
The command rejects unsafe FQME paths, symlinked sources and outputs,
destination collisions, non-regular source nodes, malformed canonical
schema, non-finite values, invalid timestamps, noncanonical indexes, and
sub-period cadence. Copy-failure cleanup preserves source bytes while
completed corrupt snapshots remain available for diagnosis.
Verification generated with the patch:
- storage audit and CLI regressions: passed
- Ruff: passed
- `git diff --check`: passed
- final adversarial review: no findings

View File

@ -1,39 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: 24fb9765-a550-4570-8350-f0fc9b7e17db
timestamp: 2026-07-28T23:27:18Z
git_ref: ed85721c
scope: code
substantive: true
raw_file: 20260728T232718Z_ed85721c_prompt_io.raw.md
---
## Prompt
Diagnose the live `pikerd` failure where backfill publication raised
`ValueError: OHLCV timestamps must be finite and positive` while merging the
known gappy MNQ NativeDB baseline with fresh IB history. Determine whether to
commit the pending audit stack or iterate first.
## Response summary
Traced the primary failure through `publish_latest_frame()` to
`NativeStorageClient.update_ohlcv()`. The incoming frame was valid, but the
canonicalized persisted baseline retained its epoch-zero row and caused the
merged frame to fail validation. Added persisted-row repair before merge and
a regression using the observed legacy schema characteristics.
## Files changed
- `piker/storage/nativedb.py` - drop non-positive persisted rows during merge
- `tests/test_storage_nativedb.py` - live MNQ failure-shape regression
- `ai/prompt-io/opencode/20260728T232718Z_ed85721c_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260728T232718Z_ed85721c_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,27 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-28T23:27:18Z
git_ref: ed85721c
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/storage/nativedb.py tests/test_storage_nativedb.py`
Fixed incremental NativeDB updates against legacy persisted history with
non-positive timestamps. Incoming provider frames remain strictly validated,
while merge repair discards invalid persisted rows before concat, dedupe,
ordering, canonical-index rewrite, and atomic publication.
The regression reproduces the observed MNQ baseline shape: one epoch-zero
row, noncanonical absolute indexes, extra derived columns, and a valid fresh
IB frame. It proves the invalid row is removed while all valid old and new
bars survive with canonical indexes.
Verification generated with the patch:
- NativeDB regressions: 18 passed
- history backfill regressions: 6 passed
- focused Ruff: passed
- `git diff --check`: passed

View File

@ -1,40 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: 24fb9765-a550-4570-8350-f0fc9b7e17db
timestamp: 2026-07-29T03:41:06Z
git_ref: 922a5df8
scope: code
substantive: true
raw_file: 20260729T034106Z_922a5df8_prompt_io.raw.md
---
## Prompt
Diagnose the live IB reverse-fill failure where
`reqHistoricalDataAsync()` raised API error 10314 because its end date,
time, or timezone was invalid.
## Response summary
Traced the failure to `Client.bars()`, where an unfinished EST conversion
left the original datetime for `ib_async` to serialize as
`YYYYMMDD HH:MM:SS UTC`. Converted aware boundaries explicitly to UTC and
passed IB's accepted `YYYYMMDD-HH:MM:SS` string while retaining a blank
latest-request boundary. Added exact request-argument regressions.
## Files changed
- `piker/brokers/ib/api.py` - explicit UTC-dash history boundary format
- `tests/test_ib_history.py` - IB request-format regressions
- `.claude/skills/run-tests/test-harness-reference.md`
- deterministic IB history test mapping
- `ai/prompt-io/opencode/20260729T034106Z_922a5df8_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260729T034106Z_922a5df8_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,25 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-29T03:41:06Z
git_ref: 922a5df8
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/brokers/ib/api.py tests/test_ib_history.py .claude/skills/run-tests/test-harness-reference.md`
Fixed IB reverse-history requests to serialize aware end boundaries with
IB's explicit UTC-dash syntax, `YYYYMMDD-HH:MM:SS`. This bypasses
`ib_async`'s trailing-`UTC` representation, which the live Gateway rejected
with API error 10314, and removes the unfinished EST comparison breakpoint.
The regression captures the exact `reqHistoricalDataAsync()` argument for
latest, stdlib-aware, Pendulum UTC, and America/New_York inputs.
Verification generated with the patch:
- IB history request regressions: 4 passed
- focused Ruff: passed
- `git diff --check`: passed
- adversarial review: no findings

View File

@ -1,40 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
session: 24fb9765-a550-4570-8350-f0fc9b7e17db
timestamp: 2026-07-29T04:17:23Z
git_ref: ce33deb6
scope: code
substantive: true
raw_file: 20260729T041723Z_ce33deb6_prompt_io.raw.md
---
## Prompt
Diagnose why the chart still showed a zero first OHLCV row after the
epoch-zero timestamp repair, causing auto-y-ranging to include zero.
## Response summary
Audited live 60s and 1s NativeDB files and found one exact 60s first-slot
sentinel at epoch 60 with a completely zero OHLCV payload. Added narrowly
scoped read and merge repair, matching incoming validation, sentinel-only
startup handling, canonical reindexing, and explicit audit classification.
Preserved modern all-zero bars as unclassified evidence rather than assuming
all zero-priced instruments are corrupt.
## Files changed
- `piker/storage/nativedb.py` - first-slot sentinel repair and validation
- `piker/storage/_audit.py` - sentinel classification and zero-row evidence
- `tests/test_storage_nativedb.py` - hydration, merge, and boundary regressions
- `tests/test_storage_audit.py` - sentinel and modern-zero audit regressions
- `ai/prompt-io/opencode/20260729T041723Z_ce33deb6_prompt_io.raw.md`
- unedited response record
- `ai/prompt-io/opencode/20260729T041723Z_ce33deb6_prompt_io.md`
- provenance metadata and response summary
## Human edits
None - generated changes have not been edited by the human.

View File

@ -1,31 +0,0 @@
---
model: gpt-5.6-sol
provider: openai
service: opencode
timestamp: 2026-07-29T04:17:23Z
git_ref: ce33deb6
diff_cmd: git diff HEAD~1..HEAD
---
> `git diff HEAD~1..HEAD -- piker/storage/nativedb.py piker/storage/_audit.py tests/test_storage_nativedb.py tests/test_storage_audit.py`
Fixed the remaining legacy first-slot sentinel observed in live MNQ 60s
history: `time=60` with zero open, high, low, close, and volume. NativeDB now
repairs the exact sentinel before read hydration and merge, reindexes the
filtered view, and persists the repaired frame on the next valid update.
The repair remains narrowly scoped. Modern all-zero OHLC bars stay valid for
spreads or synthetic instruments, null-containing malformed rows still fail
validation, new writes can not recreate the first-slot sentinel, and a file
containing only sentinels loads as no history so fresh backfill can proceed.
The raw audit reports total all-zero price rows, classifies first-slot
sentinels as structural violations, and leaves later all-zero rows visible as
unclassified warnings.
Verification generated with the patch:
- NativeDB, audit, history, xonsh, and IB regressions: 61 passed
- focused Ruff: passed
- `git diff --check`: passed
- final adversarial review: no findings

View File

@ -1,34 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: 6409e3bb-1a9a-4ada-b6c4-c554115143e9
timestamp: 2026-07-29T18:54:06Z
git_ref: d359b1cb
scope: code
substantive: true
raw_file: 20260729T185406Z_d359b1cb_prompt_io.raw.md
---
## Prompt
The user reported that QQQ 60-second zeros appeared as a flat line after
reload, clarified that the outstanding concern was the prior `ldshm`
failure, and instructed the agent to implement the more focused fix.
## Response summary
Changed `ldshm` to classify invalid SHM snapshots before downstream
analysis. Cadence is inferred only from observed finite positive steps;
buffers containing non-finite or non-positive timestamps, or lacking a
positive cadence, are skipped rather than partially replacement-written
or reloaded. Added deterministic helper and command-level regressions.
## Files changed
- `piker/storage/cli.py` - inspect cadence and reject invalid SHM snapshots before mutation.
- `tests/test_ldshm.py` - cover QQQ-like zeros, corrupt timestamps, cadence selection, degenerate buffers, and mutation prevention.
- `.claude/skills/run-tests/test-harness-reference.md` - register the deterministic regression target.
## Human edits
None - generated output remains uncommitted.

View File

@ -1,58 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-07-29T18:54:06Z
git_ref: d359b1cb
diff_cmd: git diff HEAD~1..HEAD
---
The user clarified that the QQQ 60-second chart now displayed a flat
forward-filled segment, then asked whether the previously reported
`piker store ldshm` failure had been fixed. After being told it had not,
the user instructed: "ok but i read above you have a more focussed fix
todo now? so have at it!"
The implementation focused on the exact failure boundary. It does not
filter invalid rows into a partial replacement and does not mutate live
SHM. Instead, `ldshm` inspects one captured Polars snapshot, infers its
cadence from observed finite positive timestamp steps, and skips the
entire buffer before analysis, persistence, or reload when any timestamp
is non-finite or non-positive.
> `git diff HEAD~1..HEAD -- piker/storage/cli.py`
Generated `_shm_period_and_invalid_count()` and integrated its result in
`ldshm`. The helper selects the most frequent observed positive step,
using the smallest observed step as the deterministic tie-break. Short
or cadence-degenerate buffers are skipped. Invalid snapshots are logged
and cannot reach deduplication, NativeDB replacement, or SHM reload.
> `git diff HEAD~1..HEAD -- tests/test_ldshm.py`
Generated deterministic regressions for the QQQ-shaped interior zero
run, observed-step cadence selection, all invalid timestamp classes,
short and degenerate buffers, and command-level prevention of dedupe,
storage writes, and reload. The command-level test uses an immutable
Polars snapshot and failing spies, without live SHM mutation or timing.
> `git diff HEAD~1..HEAD -- .claude/skills/run-tests/test-harness-reference.md`
Added the deterministic `ldshm` regression target and its change-to-test
mapping to the repository test harness reference.
Verification output:
```text
.......................................................... [100%]
58 passed in 0.70s
```
`git diff --check` passed. Ruff was unavailable in the worktree's
existing environment. Three adversarial review passes found and drove
fixes for partial-replacement data loss, invented median timeframes,
mixed SHM observations, and pre-guard null-analysis failures. The final
review reported no remaining findings.
The patch intentionally does not claim that `ldshm` is safe to run as a
general concurrent repair writer. It handles invalid live snapshots by
refusing mutation rather than weakening NativeDB validation.

View File

@ -1,37 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: 253ade33-07ba-45b5-8839-be35d17ec164
timestamp: 2026-07-29T20:40:46Z
git_ref: ad6c560f
scope: code
substantive: true
raw_file: 20260729T204046Z_ad6c560f_prompt_io.raw.md
---
## Prompt
The user reported that the QQQ 1-second chart still contained a gap after
the `ldshm` fix, asked the agent to inspect it, and authorized continued
implementation after a short pause and shutdown of the live workspace.
## Response summary
Diagnosed a 653-row active RT SHM null reservation caused by inclusive IB
reverse-query overlaps and startup repair orchestration. Prevented extra
physical SHM overlap rows, serialized and bounded IB history recovery,
made blank/error responses terminate, and constrained synthetic repair to
valid interior timestamp boundaries. Added deterministic concurrency and
data-layout regressions.
## Files changed
- `piker/brokers/ib/feed.py` - bound and serialize IB history reset/retry behavior.
- `piker/tsp/_history.py` - prevent SHM overlap skew and safely finalize null repair.
- `tests/test_ib_history.py` - cover blank, cancellation, reset, and pacing interleavings.
- `tests/test_history_backfill.py` - cover overlap and null-repair invariants.
- `.claude/skills/run-tests/test-harness-reference.md` - register deterministic coverage.
## Human edits
None - generated output remains uncommitted.

View File

@ -1,72 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-07-29T20:40:46Z
git_ref: ad6c560f
diff_cmd: git diff HEAD~1..HEAD
---
The user reported that QQQ still displayed a gap on the 1-second chart
after the focused `ldshm` fix and asked the agent to inspect it. The user
then paused and resumed implementation while shutting down the live piker
trading workspace.
Read-only live inspection found the active datad generation held 653
contiguous zero-time rows in QQQ's 1-second RT SHM, while durable Parquet
contained no zero rows. The run occupied absolute indexes 168148 through
168800 between timestamps 1785348398 and 1785349051. The stale actor
generation and active 60-second history were separately identified.
> `git diff HEAD~1..HEAD -- piker/tsp/_history.py`
Generated history changes that exclude each inclusive reverse-query
endpoint already present in SHM while retaining the full provider frame
for NativeDB merge. Startup null repair now runs after reverse backfill,
uses bounded UI notifications, handles provider no-data without an
interactive pause, fills only exact interior zero groups, and refuses
synthetic timestamps unless both valid boundaries establish the expected
cadence.
> `git diff HEAD~1..HEAD -- piker/brokers/ib/feed.py`
Generated IB history reset changes that make blank responses complete,
bound repeated cancellation and reset failures per request, serialize
timeout and pacing farm resets with one actor-global Trio lock, carry
explicit reset results, and coordinate pacing handoff with queued timeout
waiters.
> `git diff HEAD~1..HEAD -- tests/test_history_backfill.py`
Generated regressions for inclusive endpoint exclusion, full NativeDB
provider deltas, strict synthetic gap boundaries, empty-provider local
fallback, bounded notifications, and preservation of valid margin rows.
> `git diff HEAD~1..HEAD -- tests/test_ib_history.py`
Generated deterministic regressions for blank IB responses, cancellation
limits, stalled reset cancellation, pacing reset failure limits,
cross-request reset serialization, timeout-to-pacing handoff, and
successful handoff timeout restart.
> `git diff HEAD~1..HEAD -- .claude/skills/run-tests/test-harness-reference.md`
Registered the deterministic history-backfill target and expanded the IB
history mapping to include `feed.py`.
Verification output:
```text
........................................................................ [ 93%]
..... [100%]
77 passed in 2.12s
```
`git diff --check` passed. Multiple adversarial review rounds identified
and drove fixes for unbounded IB empty/cancellation/reset paths, stale
reset ownership, actor-global failure poisoning, concurrent farm resets,
double-counted pacing handoffs, physical endpoint overlap, and unsafe
synthetic timestamp boundaries. Final review reported no findings.
Live validation was intentionally deferred because the user shut down the
trading workspace. A fresh datad/chart startup is required to regenerate
SHM and verify that QQQ no longer contains the 653-row null run.

View File

@ -1,41 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-07-31T16:52:15Z
git_ref: 0846cbd4
scope: code
substantive: true
raw_file: 20260731T165215Z_0846cbd4_prompt_io.raw.md
---
## Prompt
Evaluate and update `piker store` so bare groups and endpoints immediately
show help; prepare for eventual optional fuzzy selection in `anal` and
`ldshm`; aggregate recent time-series diagnostics into CLI endpoints. The
user selected the safe-foundation scope.
## Response summary
Added immediate help UX, exact NativeDB and SHM identities, read-only
`series` and `shm` diagnostics, selector-ready exact SHM names, and robust
discovery/attachment. Fixed related `ldshm` option and no-match failures.
Deferred optional `fzf` and mutation-default changes.
## Files changed
- `piker/storage/cli.py` - help UX and read-only series/SHM diagnostics.
- `piker/storage/nativedb.py` - canonical compound series identities.
- `piker/tsp/_history.py` - exact SHM discovery and dtype resolution.
- `piker/data/_sharedmem.py` - attach-only existing-segment helper.
- `piker/tsp/__init__.py` - SHM identity helper exports.
- `tests/test_store_cli.py` - command UX and endpoint regressions.
- `tests/test_ldshm.py` - SHM diagnostics and operational regressions.
- `tests/test_storage_nativedb.py` - series parsing/index regressions.
- `docs/manual_backfill_qualification.rst` - triage command documentation.
- `.claude/skills/run-tests/test-harness-reference.md` - test mapping.
## Human edits
None - generated output remains uncommitted.

View File

@ -1,93 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-07-31T16:52:15Z
git_ref: 0846cbd4
diff_cmd: git diff HEAD~1..HEAD
---
The user requested an evaluation and possible update of `piker store`,
including immediate help output when a group or endpoint is invoked with
no arguments. The user also proposed eventual optional `fzf` selection for
`anal` and `ldshm`, and authorized aggregating the recent time-series and
SHM debugging work into CLI endpoints. After reviewing three scopes, the
user selected the recommended safe foundation rather than implementing
interactive fuzzy selection or changing mutation defaults immediately.
> `git diff HEAD~1..HEAD -- piker/storage/cli.py`
Generated no-argument help behavior for the `store` group and every
endpoint. Added read-only `series` and `shm` commands, JSON and Rich output,
exact `--shm-name` plumbing, SHM cadence/invalid/order/gap summaries, and
safe corrupt-epoch formatting. Fixed existing `ldshm` no-write/no-reload
control flow and no-match debugger behavior.
> `git diff HEAD~1..HEAD -- piker/storage/nativedb.py`
Generated canonical `NativeSeriesRef` parsing and deterministic listing.
Changed the internal index to preserve `(fqme, period)` identities while
keeping unique-FQME compatibility listing and cardinality. Hardened
malformed, sidecar, Unicode-period, symlink, and traversal handling.
> `git diff HEAD~1..HEAD -- piker/tsp/_history.py`
Generated exact `ShmBufferRef` parsing and stable discovery, replacing
unsafe substring globbing. Added dtype resolution that mirrors history
allocation, exact-ref attachment, disappearance handling, and optional
exact-name iteration.
> `git diff HEAD~1..HEAD -- piker/data/_sharedmem.py`
Generated attach-only SHM access which never creates a segment when a
discovered object disappears.
> `git diff HEAD~1..HEAD -- piker/tsp/__init__.py`
Exported selector-ready SHM identity, parsing, listing, and dtype helpers.
> `git diff HEAD~1..HEAD -- tests/test_store_cli.py`
Generated command-level help, exact durable-series JSON, read-only SHM JSON,
and unknown exact-name regressions.
> `git diff HEAD~1..HEAD -- tests/test_ldshm.py`
Generated SHM identity, summary, ordering, dtype, forged-name, disabled
mutation-flag, and gap-markup regressions.
> `git diff HEAD~1..HEAD -- tests/test_storage_nativedb.py`
Generated compound index, canonical parser, malformed filename, and dotted
FQME regressions.
> `git diff HEAD~1..HEAD -- docs/manual_backfill_qualification.rst`
Documented `store series` and `store shm` as separate read-only durable and
live-memory triage layers.
> `git diff HEAD~1..HEAD -- .claude/skills/run-tests/test-harness-reference.md`
Registered the deterministic storage CLI UX target and mapping.
Verification output:
```text
........................................................................ [ 71%]
............................. [100%]
101 passed in 1.72s
```
Real `piker store` and `piker store audit` invocations printed complete
help without `--help`, exiting with Click's normal no-args help status 2.
`git diff --check` passed. Ruff was unavailable in the existing worktree
environment.
Adversarial review drove fixes for provider dtype reconstruction, observed
versus expected cadence, exact SHM grammar, discovery races, Typer direct
defaults, malformed NativeDB periods, compound cardinality compatibility,
undefined `ldshm` reload payloads, no-match debugger hangs, ordering
corruption reporting, and dtype precedence. Final review found no issues.
Optional external `fzf` or built-in numbered selection remains deferred.
The exact series and SHM identities added here are the intended foundation
for that later patch.

View File

@ -1,39 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-08-01T01:53:21Z
git_ref: 74ea2152
scope: code
substantive: true
raw_file: 20260801T015321Z_74ea2152_prompt_io.raw.md
---
## Prompt
Make `piker store ls` default to all NativeDB series as a compact counterpart
to detailed `series`, with sensible no-flag output and a visible help
description. Consolidate SHM inspection and repair under canonical `shm`,
because attachment to volatile segments is implied, and make inspection the
safe default while retaining explicit persistence and reload controls.
## Response summary
Replaced backend-oriented `ls` with compact NativeDB discovery and made both
listing commands default to all series. Consolidated SHM behavior under a
read-only-by-default `shm` command, guarded explicit repair options, updated
executable references and docs, and added command regressions.
## Files changed
- `piker/storage/cli.py` - listing defaults and unified SHM command modes.
- `tests/test_store_cli.py` - listing, help, and SHM mode regressions.
- `tests/test_ldshm.py` - explicit persistence-path regressions.
- `docs/manual_backfill_qualification.rst` - revised safe CLI workflow.
- `snippets/claude_debug_helper.py` - explicit SHM persistence commands.
- `.claude/skills/run-tests/test-harness-reference.md` - safety and test map.
- `.claude/skills/piker-slang/examples.md` - canonical command example.
## Human edits
None - generated output incorporates the user's design refinements.

View File

@ -1,78 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-08-01T01:53:21Z
git_ref: 74ea2152
diff_cmd: git diff HEAD~1..HEAD
---
The user requested command-surface refinements after reviewing the new
storage CLI. Bare `ls` should default to all series and act as a compact
form of detailed `series`; commands should produce sensible default output
without requiring trigger flags; and `ls` needed a parent-help description.
The user first proposed folding read-only SHM inspection into `ldshm -r`,
then clarified that `shm` is the better canonical command because attaching
to volatile segments is implied by the resource name.
The resulting design uses `piker store ls` as a compact NativeDB FQME and
period view, while `piker store series` reports each exact sub-series with
size and path details. Both default to all records and accept an optional
FQME substring.
`piker store shm FQME` is read-only by default. Persistence and reload enter
the prior repair workflow only through explicit `--write-parquet` and
`--reload-parquet-to-shm` flags. Reload requires persistence, and read-only
formatting flags are rejected in repair mode instead of being ignored.
> `git diff HEAD~1..HEAD -- piker/storage/cli.py`
Generated shared NativeDB payload collection, compact and detailed output,
numeric period ordering, default-all semantics, the canonical `shm` command,
safe mode dispatch, explicit repair validation, and debugger propagation in
read-only mode.
> `git diff HEAD~1..HEAD -- tests/test_store_cli.py`
Generated regressions for no-flag listing, parent-help descriptions, numeric
period display, default read-only SHM JSON, reload dependencies, and repair
versus read-only option conflicts.
> `git diff HEAD~1..HEAD -- tests/test_ldshm.py`
Adjusted persistence regressions for the renamed command and explicit repair
entry while preserving the historical failure rationale.
> `git diff HEAD~1..HEAD -- docs/manual_backfill_qualification.rst`
Documented compact versus detailed NativeDB discovery, default read-only SHM
inspection, and the explicitly mutating repair invocation.
> `git diff HEAD~1..HEAD -- snippets/claude_debug_helper.py`
Updated executable debug helper commands to select explicit SHM persistence.
> `git diff HEAD~1..HEAD -- .claude/skills/run-tests/test-harness-reference.md`
Updated operational safety guidance and SHM regression mappings.
> `git diff HEAD~1..HEAD -- .claude/skills/piker-slang/examples.md`
Updated the command example to the canonical `shm` spelling.
Verification output:
```text
........................................................................ [ 69%]
................................ [100%]
104 passed in 5.77s
```
Real command-help rendering was checked for the parent group, `ls`, `series`,
and `shm`. Real read-only `ls` and `series --json` output was also exercised
against the configured NativeDB. `git diff --check` passed.
Adversarial review identified and drove fixes for lexicographic period
ordering, stale runbook safety wording, ignored mode-specific options, and an
explicit-default `--max-gaps 10` validation bypass. Final review found no
remaining issues. Real OS-backed SHM attachment and successful repair/reload
remain integration-level validation concerns.

View File

@ -1,33 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-08-05T19:50:01Z
git_ref: f0f3c806
scope: code
substantive: true
raw_file: 20260805T195001Z_f0f3c806_prompt_io.raw.md
---
## Prompt
Inspect the completed NVDA 60s SHM backfill before workspace restart, explain
the persistent zero region, and continue fixing the indexing logic so sparse
IB history does not leave physical null reservations.
## Response summary
Captured a 42,720-row live SHM null span whose valid margins shared the same
timestamp. Replaced wall-clock row reservation with actual provider-boundary
placement, fixed stale SHM first-index, endpoint insertion and field-map
semantics, published the final storage prepend, and added a full multi-frame
`tsdb_backfill()` regression using a real `ShmArray`.
## Files changed
- `piker/tsp/_history.py` - contiguous provider and NativeDB prepends.
- `tests/test_history_backfill.py` - sparse multi-frame SHM regression.
## Human edits
None - generated output incorporates the user's live observations.

View File

@ -1,71 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-08-05T19:50:01Z
git_ref: f0f3c806
diff_cmd: git diff HEAD~1..HEAD
---
The user reported that a completed NVDA 60s backfill left a large zero
region in live SHM and asked for immediate inspection before restarting the
workspace. The captured buffer contained 42,720 invalid rows at absolute
indexes `3065895..3108614`. The valid rows on both sides had the identical
timestamp `1781271360`, proving the region was an erroneous physical-index
reservation rather than a real temporal gap.
The user then asked the agent to continue fixing the backfill logic.
> `git diff HEAD~1..HEAD -- piker/tsp/_history.py`
Removed wall-clock-based SHM reservation from `tsdb_backfill()`. Reverse
provider frames now extend the visible `ShmArray._first` boundary on every
prepend, beginning exactly at the current first index after inclusive query
endpoints are removed. Added `_prepend_tsdb_history()` to trim storage rows
against the actual earliest provider timestamp and prepend them directly
adjacent only after reverse provider retrieval completes.
This preserves sparse venue history as ordinal samples rather than treating
every elapsed calendar period as a physical SHM row. It also prevents the
old `_first + 1` insertion point from overwriting the earliest published bar.
> `git diff HEAD~1..HEAD -- tests/test_history_backfill.py`
Added a full `tsdb_backfill()` regression modeled on the live NVDA failure.
It uses a real `ShmArray`, NativeDB's field map, two inclusive reverse-provider
frames, and overlapping stored history. Assertions prove the complete
timestamp sequence is contiguous and lossless, contains no non-positive
timestamps, sends complete provider frames to durable storage, and notifies
chart/FSP consumers after the final storage prepend.
Live evidence captured before workspace restart:
```text
rows 1575296 first_index 1568233 last_index 3143528
invalid_count 42720
left boundary
[(3065893, 1781271300, 204.03, 204.82)
(3065894, 1781271360, 204.81, 204.72)
(3065895, 0, 0. , 0. )]
right boundary
[(3108614, 0, 0. , 0. )
(3108615, 1781271360, 204.81, 205.44)
(3108616, 1781271420, 205.46, 205.48)]
```
Verification output:
```text
........................................................................ [ 68%]
................................. [100%]
105 passed in 4.80s
```
Adversarial review caught stale `_first` semantics and an off-by-one
insertion point in the initial patch. The first live restart then exposed an
undefined `time_key` local at the new `tsdb_backfill()` call site, which
crashed `datad.ib` and propagated to the chart as `RemoteActorError`. The
helper now derives the source timestamp field from its field map, and the
test enters the full enclosing lifecycle so the original `NameError` can not
recur unnoticed. A final review also drove explicit publication after stored
history is prepended and deterministic final-stage synchronization. Final
review found no remaining issues; another live restart remains pending.

View File

@ -1,36 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: ses_0799212ebffe42arY96czXn89F
timestamp: 2026-08-11T20:58:26Z
git_ref: 06d5ea5d
scope: code
substantive: true
raw_file: 20260811T205826Z_06d5ea5d_prompt_io.raw.md
---
## Prompt
Diagnose and fix an IB live-order failure where lazy NVDA contract
qualification raised `tractor.trionics.Lagged` after the order reached
`brokerd.ib`. Determine whether recent history changes or cancelled symbol
searches caused the stale state.
## Response summary
Traced the failure to competing consumers of a one-slot Trio broadcast, not
the recent SHM change or stale asyncio client state. Replaced broadcast
fan-out with one response/status demultiplexer, retained request-ID
correlation, handled cancellation and teardown, documented the task/channel
ownership model, warned on diagnostic metadata eviction, and added
deterministic proxy regressions.
## Files changed
- `piker/brokers/ib/api.py` - sole-reader IB proxy response routing.
- `tests/test_ib_method_proxy.py` - status, concurrency, cancellation and EOF regressions.
- `.claude/skills/run-tests/test-harness-reference.md` - proxy test mapping.
## Human edits
None - generated output follows the user's live failure diagnosis.

View File

@ -1,66 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-08-11T20:58:26Z
git_ref: 06d5ea5d
diff_cmd: git diff HEAD~1..HEAD
---
The user reported that certain IB order entries might have broken and
provided a live failure for a first NVDA sell order. EMS delivered the
request to `brokerd.ib`, but lazy `get_mkt_info('nvda.nasdaq')` contract
qualification failed in `MethodProxy._run_method()` with:
```text
tractor.trionics._broadcast.Lagged:
Task `piker.brokers.ib.broker.handle_order_requests` overrun and
dropped `0` values
```
The user suspected a cancelled symbol-search task might have left stale
asyncio state. Cancellation can leave an orphaned method response, but the
observed failure was stale Trio broadcast state. `open_client_proxy()`
broadcast a one-slot `LinkedTaskChannel` to an always-running IB status relay
and an idle method receiver. Status traffic advanced only the relay, so the
first method call for an uncached symbol entered an already-lagged receiver.
Orders for eagerly cached contracts bypassed this path.
The immediately preceding sparse-history commit did not modify IB order,
symbol, or proxy code; it exposed no direct dependency on this failure.
> `git diff HEAD~1..HEAD -- piker/brokers/ib/api.py`
Generated a sole-reader message demultiplexer for each IB client proxy.
`relay_client_proxy_messages()` owns every asyncio-to-Trio receive, handles
status/error traffic, and dispatches `mid`-tagged responses into per-call
`MethodProxy` wait slots. Cancelled calls remove their pending slot, causing
late responses to be logged and dropped without skewing later calls. Clean
channel EOF is accepted during normal proxy teardown.
Follow-up review restored the masked reconnect TODO and per-message case
rationale, documented the one-reader/many-writer task hierarchy and separate
per-proxy channels, and added a warning with request method metadata when the
bounded unresolved-request diagnostic table evicts its oldest entry.
> `git diff HEAD~1..HEAD -- tests/test_ib_method_proxy.py`
Generated deterministic regressions for idle status traffic before first
qualification, two concurrent calls with reverse-order responses, cancelled
calls with late responses followed by a successful call, and graceful
channel EOF. Every synchronization wait is timeout-bounded.
> `git diff HEAD~1..HEAD -- .claude/skills/run-tests/test-harness-reference.md`
Registered the deterministic IB method-proxy target and source mapping.
Verification output:
```text
........................ [100%]
25 passed, 4 warnings in 3.44s
```
The warnings are existing Tractor/Trio deprecations from the EMS test.
Adversarial review drove a clean-EOF guard and bounded polling. Final review
found no actionable issues. Live IB order validation remains pending and
requires restarting `brokerd.ib`.

View File

@ -1,34 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: unavailable
timestamp: 2026-08-13T20:24:02Z
git_ref: 12b0d66e
scope: code
substantive: true
raw_file: 20260813T202402Z_12b0d66e_prompt_io.raw.md
---
## Prompt
Continue the paired piker broadcast-consumer work after Tractor PR 490
stabilization. Stop cached feed callers from mutating or consuming shared
quote-stream roots, then review, verify and prepare a commit boundary.
## Response summary
Made provider root streams private to the cached `Feed`, gave every caller a
local subscribed view, corrected feed cache identity, and made multi-provider
relays own synchronized child subscriptions and independent send-channel
clones.
## Files changed
- `piker/data/feed.py` - caller-local feed views, complete cache keys, and
subscribed multi-provider relays with independent channel lifetimes.
- `tests/test_feed_broadcast.py` - real broadcast ownership, lifecycle and
cache-identity regressions, including sibling survival after provider EOC.
## Human edits
None - generated output follows the requested cached-feed ownership item.

View File

@ -1,51 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-08-13T20:24:02Z
git_ref: 12b0d66e
diff_cmd: git diff HEAD~1..HEAD
---
The user asked to continue the paired piker broadcast-consumer work after
Tractor PR 490 stabilization. The next item was to stop cached feed callers
from mutating or consuming shared `Feed`/`Flume` quote-stream roots, then
review, verify and prepare a commit boundary.
> `git diff HEAD~1..HEAD -- piker/data/feed.py`
Changed `maybe_open_feed()` so the cached `Feed` exclusively owns provider
root streams and every caller, including the first cache owner, receives a
caller-local `Feed`/`Flume` view backed by one child subscription per provider.
The cache identity now includes normalized full symbol shape and stream policy.
Changed `Feed.open_multi_stream()` so every provider relay subscribes instead
of directly consuming a shared stream, and synchronizes relay startup before
yielding the merged channel. Each relay owns a cloned send channel so one
provider EOC cannot terminate its siblings. The module-level relay returns its
live child receiver through `TaskStatus` for typed retention and introspection.
Documented `Feed.pause()`/`Feed.resume()` as provider-context-wide controls
shared by local cache users.
> `git diff HEAD~1..HEAD -- tests/test_feed_broadcast.py`
Added real `Feed`, `Flume`, and Tractor broadcast regressions covering both
cache-miss owner and cache-hit paths, one-provider-many-market mapping,
multi-provider relay fan-out and startup, duplex controls, lexical child
closure, sibling survival after provider EOC, cached-root immutability, and
full cache-key differentiation.
Verification output:
```text
... [100%]
3 passed in 0.02s
. [100%]
1 passed in 0.01s
```
The second result is the existing EMS broadcast ownership regression. Ruff,
E501, compilation and whitespace checks passed. Repeated adversarial review
found and resolved shared `Feed.streams`, incomplete cache-key, fake-lifecycle,
relay-startup and first-owner asymmetry issues; final review reported no
findings.

View File

@ -1,32 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: unavailable
timestamp: 2026-08-13T23:07:44Z
git_ref: ddac3553
scope: code
substantive: true
raw_file: 20260813T230744Z_ddac3553_prompt_io.raw.md
---
## Prompt
Continue the piker broadcast-consumer audit after the Deribit ownership commit
and prepare the next smallest ownership fix.
## Response summary
Replaced IB's manual raw quote-channel registry with a symbol-keyed cached
context and gave every caller a lexical child from one retained
`LinkedTaskChannel` broadcaster root.
## Files changed
- `piker/brokers/ib/feed.py` - cached raw-channel lifetime and lexical child
ownership.
- `tests/test_ib_broadcast.py` - concurrent fan-out, cache-key and child
lifetime regression.
## Human edits
None - generated output follows the requested broadcast-consumer audit.

View File

@ -1,39 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-08-13T23:07:44Z
git_ref: ddac3553
diff_cmd: git diff HEAD~1..HEAD
---
The user asked to continue the piker broadcast-consumer audit after committing
the Deribit ownership fix. The next isolated item was IB's actor-local
`_quote_streams` cache, which retained a raw channel owned by its first caller
and constructed competing broadcaster roots for later callers.
> `git diff HEAD~1..HEAD -- piker/brokers/ib/feed.py`
Factored `_open_aio_quote_channel()` as the raw asyncio-channel context and
made `open_aio_quote_stream()` cache that context by symbol through
`maybe_open_context()`. Every caller now receives a lexical child from the
retained `LinkedTaskChannel` broadcaster, while the raw source remains alive
until its final cache user exits. Removed the manual `_quote_streams` registry
and teardown mutation.
> `git diff HEAD~1..HEAD -- tests/test_ib_broadcast.py`
Added a concurrent ownership regression using real Tractor broadcast state.
It proves miss and hit callers use the same symbol/context factory, receive the
same quote through distinct children, close independently, and leave the owner
live after the hit exits.
Verification output:
```text
. [100%]
1 passed in 0.01s
```
The new test passes through a source-isolated IB feed import. Test Ruff, E501,
Python compilation and whitespace checks pass. Normal package collection
remains blocked by the existing Qt environment mismatch.

View File

@ -1,43 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: unavailable
timestamp: 2026-08-14T03:27:16Z
git_ref: 2aa66c2c
scope: code
substantive: true
raw_file: 20260814T032716Z_2aa66c2c_prompt_io.raw.md
---
## Prompt
Prototype chart-actor-owned gap overlays from the current backfiller tip,
with actor-local startup and UI toggling. Preserve remote annotation IPC and
make the gap-specific path strict and `msgspec`-typed.
## Response summary
Adds a chart-local controller and vectorized SHM detector, starts historical
gap overlays with display state, toggles the focused timeframe through the
existing input loop and exposes the same renderer through a strict Tractor
payload context. Keeps all legacy remote annotation commands and separates
local graphics from each remote context's teardown ownership.
## Files changed
- `piker/ui/_gaps.py` - typed messages, detection and local controller.
- `piker/ui/_remote_ctl.py` - IPC clients and chart-actor endpoints.
- `piker/ui/_display.py` - actor-local startup integration.
- `piker/ui/_interaction.py` - focused `Ctrl+G` toggle.
- `piker/ui/_widget.py` - controller reference.
- `piker/ui/_annotate.py` - safe FQME-aware repositioning.
- `piker/tsp/_annotate.py` - typed annotation-client reference.
- `piker/storage/cli.py` - renamed annotation client context.
- `tests/test_gap_overlays.py` - detector and schema regression.
- `tests/test_ldshm.py` - renamed annotation-client test seams.
- `plans/opencode/chart-local-gap-overlays.md` - architecture and landing plan.
## Human edits
None - the work remains an uncommitted disposable-worktree prototype for
human review and live chart qualification.

View File

@ -1,72 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-08-14T03:27:16Z
git_ref: 2aa66c2c
diff_cmd: git diff HEAD~1..HEAD
---
## Prompt
The user asked what work had been done so far, then authorized autonomous
work while AFK. The target was to scheme and prototype an actor-local gap
annotator from the `backfiller_deep_fixes` tip without mutating existing
branches. The user then clarified twice that the annotation IPC API must
remain available and should become more rigorous and `msgspec`-typed rather
than being removed.
## Generated code
> `git diff HEAD~1..HEAD -- piker/ui/_gaps.py`
Adds tagged gap request/reply structs, render-ready gap specs, vectorized
OHLCV timestamp-gap detection and chart-local overlay ownership.
> `git diff HEAD~1..HEAD -- piker/ui/_remote_ctl.py`
Adds strict Tractor gap IPC, `AnnotClient` stream mapping/locking and
owner-scoped endpoint cleanup while retaining the legacy annotation endpoint.
> `git diff HEAD~1..HEAD -- piker/ui/_display.py`
Starts the local 60-second overlay after display-state registration.
> `git diff HEAD~1..HEAD -- piker/ui/_interaction.py`
Routes `Ctrl+G` from the focused chart's existing async input loop directly
to the local controller.
> `git diff HEAD~1..HEAD -- piker/ui/_widget.py`
Stores the chart actor's local controller reference.
> `git diff HEAD~1..HEAD -- piker/ui/_annotate.py`
Keeps redraw repositioning isolated by FQME and synchronizes arrow geometry.
> `git diff HEAD~1..HEAD -- piker/tsp/_annotate.py piker/storage/cli.py`
Renames the remote annotation control type and context manager to the
client-oriented `AnnotClient` / `open_annot_client` interface.
> `git diff HEAD~1..HEAD -- tests/test_gap_overlays.py`
Adds deterministic detector, directional payload-schema, typed error and
msgpack roundtrip regressions.
> `git diff HEAD~1..HEAD -- tests/test_ldshm.py`
Updates storage CLI test seams for the renamed annotation client context.
## Generated architecture output
> `git diff HEAD~1..HEAD -- plans/opencode/chart-local-gap-overlays.md`
Documents ownership, startup, UI, strict IPC, refresh and branch landing
semantics.
## Verification output
Source resolution points to this worktree. The targeted regression passes,
Ruff reports no changed-file findings, `compileall` succeeds, `git diff
--check` succeeds and Tractor's strict gap payload decoder initializes.

View File

@ -1,39 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: unavailable
timestamp: 2026-08-17T22:20:36Z
git_ref: f65df6a1
scope: code
substantive: true
raw_file: 20260817T222036Z_f65df6a1_prompt_io.raw.md
---
## Prompt
Add end-to-end coverage for chart-local gap annotations in the existing
gap-overlay worktree without disturbing concurrent manual backfiller testing
in the root checkout.
## Response summary
Adds real offscreen Qt and real Tractor actor integration coverage. The tests
drive startup registration, PyQtGraph insertion/removal, focused Ctrl-G input,
history-prepend repositioning, duplicate-FQME chart identity, typed endpoint
exchange and cancellation with stale-response recovery. Defects exposed by
those tests are fixed in manager cleanup, Qt geometry notification and PyQt6
event routing.
## Files changed
- `piker/ui/_gaps.py` - correct PyQtGraph ownership and bounds behavior.
- `piker/ui/_annotate.py` - notify Qt before reposition geometry changes.
- `piker/ui/_display.py` - expose deterministic startup registration.
- `piker/ui/_interaction.py` - focus toggles and PyQt6 event registration.
- `tests/test_gap_overlays.py` - real Qt and Tractor integration coverage.
- `plans/opencode/chart-local-gap-overlays.md` - update coverage boundaries.
- `plans/opencode/chart-local-gap-overlays.summary.md` - update test totals.
## Human edits
None - the changes remain uncommitted for human review.

View File

@ -1,59 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
timestamp: 2026-08-17T22:20:36Z
git_ref: f65df6a1
diff_cmd: git diff f65df6a1..HEAD
---
## Prompt
The user asked whether the chart-local gap annotation feature had complete
end-to-end coverage, then requested implementation in the existing gap-overlay
worktree while manual NativeDB/backfiller qualification continued in the root
checkout. The user initially declined a tests-only Prompt-IO entry. Real Qt
and Tractor integration tests then exposed production lifecycle and PyQt6
event-routing defects, making the resulting patch substantive code work and
requiring this full provenance entry.
## Generated code
> `git diff HEAD~1..HEAD -- piker/ui/_gaps.py`
Removes manager-owned graphics through their `PlotItem`, retains owning-plot
registrations, and excludes pixel-sized gap arrows from automatic data bounds.
> `git diff HEAD~1..HEAD -- piker/ui/_annotate.py`
Moves Qt geometry-change notification ahead of rectangle and arrow mutation so
scene spatial indexing follows history-prepend repositioning.
> `git diff HEAD~1..HEAD -- piker/ui/_display.py`
Extracts `_register_gap_overlays()` as the deterministic chart startup seam
which registers display states and renders default historical gap layers.
> `git diff HEAD~1..HEAD -- piker/ui/_interaction.py`
Adds focused-chart toggle routing, updates the gap keyboard path to PyQt6 event
enums, and keeps unrelated cached chart states untouched.
> `git diff HEAD~1..HEAD -- tests/test_gap_overlays.py`
Adds offscreen real-Qt manager, startup, keyboard, cached-chart and reposition
coverage plus a real Tractor actor endpoint/client dialog with deterministic
cancellation and stale-response recovery.
## Generated architecture output
> `git diff HEAD~1..HEAD -- plans/opencode/chart-local-gap-overlays.md plans/opencode/chart-local-gap-overlays.summary.md`
Records the automated integration boundary and remaining broker-backed manual
qualification requirements.
## Verification output
The source-isolated non-actor and related suites pass 24 tests with one actor
case deselected. The isolated actor case passes with one upstream Tractor/Trio
deprecation warning. Ruff, compileall, import resolution and whitespace checks
pass. Independent review reports no remaining P1/P2 findings.

View File

@ -1,43 +0,0 @@
---
model: gpt-5.6-sol
service: opencode
session: tractor-addr-unpacking
timestamp: 2026-08-20T16:35:19Z
git_ref: 5f631e19
scope: code
substantive: true
raw_file: 20260820T163519Z_5f631e19_prompt_io.raw.md
---
## Prompt
The human asked to implement the Piker compatibility work caused by
tractor's canonical tagged-address migration, run the previously
deferred tractor verification, and keep runtime-managed WireGuard
bindspaces in a later separate PR.
## Response summary
Made Piker normalize legacy and tagged registry inputs before semantic
comparison and process-global storage. Updated runtime address typing and
test assertions for canonical output, retained legacy caller input, and
removed a duplicate pytest log-level option that blocked downstream test
collection with current tractor.
## Files changed
- `piker/service/_registry.py` - canonical registry normalization and
shared address typing.
- `piker/service/_actor_runtime.py` - tagged-compatible runtime API
annotations.
- `tests/conftest.py` - canonical fixture comparison and shared pytest
option ownership.
- `tests/test_services.py` - registry regression and canonical runtime
assertions.
## Human edits
The human explicitly requested the downstream implementation and chose
to keep WireGuard bindspace lifecycle work in a separate future tractor
PR. The agent implemented and tested those directions; no direct manual
source edits were observed.

View File

@ -1,25 +0,0 @@
---
model: gpt-5.6-sol
service: opencode
timestamp: 2026-08-20T16:35:19Z
git_ref: 5f631e19
diff_cmd: git diff HEAD~1..HEAD
---
# Raw output - tractor tagged-address compatibility
The human requested downstream Piker qualification after tractor PR
#505 switched transport address emission to canonical tagged tuples.
> `git diff HEAD~1..HEAD -- piker/service/_registry.py piker/service/_actor_runtime.py tests/conftest.py tests/test_services.py`
Normalized legacy and tagged registry addresses through tractor's
address codec before Piker stores or compares them. Extended service
runtime annotations to cover both tuple shapes, updated runtime tests to
compare canonical serialized values, and removed Piker's duplicate
registration of tractor's shared `--ll` pytest option.
Ruff passed. The focused normalization regression and Piker runtime boot
passed against the local tractor checkout at `5d92595f`. The wider
service module reached an unrelated pre-existing logger-name assertion
in `maybe_spawn_daemon()` after both address-focused tests passed.

View File

@ -1,39 +0,0 @@
---
model: gpt-5.6-sol
service: opencode
session: tractor-addr-unpacking
timestamp: 2026-08-21T04:20:58Z
git_ref: b7aca4c0
scope: code
substantive: true
raw_file: 20260821T042058Z_b7aca4c0_prompt_io.raw.md
---
## Prompt
The human approved refining Piker's logger-name handling after the
wider service test exposed a stale parent-logger assertion during the
tractor tagged-address compatibility work.
## Response summary
Changed Piker's logger helpers to remove only an exact `piker.` prefix,
preserving complete requested submodule names. Removed daemon startup
assertions that incorrectly required parent logger names and added a
regression covering both logger factories and names that the old
character-set stripping silently corrupted.
## Files changed
- `piker/log.py` - exact relative logger-name derivation.
- `piker/service/_daemon.py` - remove stale parent-name assertion.
- `piker/data/_daemon.py` - remove stale parent-name assertion.
- `piker/brokers/_daemon.py` - remove stale parent-name assertion.
- `piker/clearing/_ems.py` - remove redundant logger assertion.
- `tests/test_logging.py` - package logger namespace regressions.
## Human edits
The human explicitly approved this follow-up after reviewing the
residual Piker service-suite gap. The agent implemented and verified
the refinement; no direct manual source edits were observed.

View File

@ -1,24 +0,0 @@
---
model: gpt-5.6-sol
service: opencode
timestamp: 2026-08-21T04:20:58Z
git_ref: b7aca4c0
diff_cmd: git diff HEAD~1..HEAD
---
# Raw output - preserve Piker logger namespaces
The human approved a follow-up refinement after the full Piker service
module exposed a stale logger-name assertion while qualifying tractor
PR #505.
> `git diff HEAD~1..HEAD -- piker/log.py piker/service/_daemon.py piker/data/_daemon.py piker/brokers/_daemon.py piker/clearing/_ems.py tests/test_logging.py`
Replaced character-set `str.lstrip()` calls with exact package-prefix
removal, removed assertion-only parent-logger expectations from daemon
startup paths, and added regression coverage for both logger factories
and prefix-sensitive submodule names.
Ruff passed after removing newly unused logger assignments. All four
logger regressions and all six service integration tests passed against
the local tractor PR #505 checkout.

View File

@ -1,46 +0,0 @@
---
model: openai/gpt-5.6-sol
service: opencode
session: unavailable
timestamp: 2026-08-26T22:26:00Z
git_ref: 5f631e19
scope: code
substantive: true
raw_file: 20260826T222600Z_5f631e19_prompt_io.raw.md
---
## Prompt
Diagnose and fix the hanging Binance USD-M perpetual feed for
`NVDAUSDT.USDTM.PERP.BINANCE` by comparing piker with Binance's current
API and changelog. Ensure unsupported or silent websocket feeds fail
loudly into `datad` and chart consumers instead of blocking startup.
## Response summary
Migrates Binance USD-M market data from the retired legacy websocket to
the split public and market routes, handles Binance's new aggregate-trade
schema, and introduces bounded connection and first-data startup. Direct
stream URLs avoid reconnect-time subscription-ACK races, while targeted
tests preserve routing and fail-loud behavior.
## Files changed
- `piker/brokers/binance/feed.py` - route, merge, normalize, and bound
Binance quote streams.
- `piker/brokers/binance/venues.py` - define current USD-M websocket
endpoints.
- `piker/data/_web_bs.py` - support bounded websocket context startup.
- `tests/test_binance.py` - cover routes, schemas, and startup failures.
- `ai/prompt-io/opencode/20260826T222600Z_5f631e19_prompt_io.raw.md` -
retain the generated-output record.
- `ai/prompt-io/opencode/20260826T222600Z_5f631e19_prompt_io.md` - record
structured provenance for the code changes.
## Human edits
The human explicitly required fail-loud behavior rather than merely
updating the endpoint, and supplied a direct CLI reproduction proving
both `datad.binance` and the chart caller blocked. Those requirements
drove the connection and first-data deadlines and their regressions. The
changes remain uncommitted for human review.

Some files were not shown because too many files have changed in this diff Show More