brk/docs/CHANGELOG.md

Changelog

All notable changes to the Bitcoin Research Kit (BRK) project will be documented in this file.

This changelog was generated by Claude Code

v0.2.3 - 2026-03-26

New Features

brk_computer

• Added new investing module with complete Dollar-Cost Averaging (DCA) and lump-sum investment analysis — computes DCA stack (cumulative sats purchased at $100/day), cost basis, returns, and CAGR by rolling period (1w, 1m, 3m, 6m, 1y through 10y); lump-sum stack and returns by rolling period; and DCA stack, cost basis, and returns by year class (2015 through 2026). Extracted from market::dca into a standalone top-level module with parallel import and compute (source)
• Added RealizedEnvelope indicator that computes the tightest percentile bounds across 10 pricing models (realized price, investor price, STH/LTH variants, cointime vaultedness/activity/true market mean/cointime price), producing 8 envelope price bands (0.5th through 99.5th percentile), a zone index (-4 to +4) based on spot price position relative to envelope bands, and a composite score (-40 to +40) summing band crossings across all models (source)
• Expanded RatioPerBlockPercentiles from 6 to 8 percentile bands, adding 0.5th and 99.5th percentiles alongside the existing 1st/2nd/5th/95th/98th/99th — ExpandingPercentiles Fenwick tree computation also widened from 6-quantile to 8-quantile output (source)
• Consolidated timestamp data into a new indexes::timestamp::Timestamps struct — moved timestamp_monotonic, date, and per-resolution timestamp lookups from blocks::time::Vecs into the indexes module, making timestamp data accessible without depending on the blocks module (source)

brk_indexer

• Added take_all_pending_ingests() to batch all KV store puts/deletes into closures that can be executed on a background thread, enabling the indexer to hand off fjall store commits to run_bg() instead of blocking the main indexing loop (source)
• Extracted stamped_write() from flush() so vec stamping can happen on the main thread while compaction and store ingests run in the background (source)

brk_store

• Added take_pending_ingest() method that drains buffered puts/deletes and returns a Send closure for background ingestion into the fjall keyspace (source)

website

• Added range preset buttons (1w, 1m, 3m, 6m, 1y, 4y, 8y, YTD, all) to the chart that automatically select the appropriate sub-day resolution index (30min through 1w) and scroll the visible range to that time window (source)
• Added value-dependent coloring support (colorFn) for histogram and baseline chart series, enabling per-bar colors based on data values (source)
• Added "Realized Envelope" chart to market indicators showing 8-level percentile price bands as dashed lines, a color-coded zone index histogram, and a score baseline (source)
• Added percentileBands() and priceBands() shared helpers for displaying 8-level percentile bands with dedicated colors per level, reusable across ratio and price chart types (source)
• Added subtle grid lines to charts using a new offBorder color variable (source)

Internal Changes

brk_computer

• Switched all module compute() methods to use db.run_bg() for background compaction instead of synchronous db.compact(), and added db.sync_bg_tasks() at compute entry to await prior background work — applies to blocks, cointime, distribution, indicators, inputs, investing, market, mining, outputs, pools, positions, prices, scripts, supply, and transactions (source)
• Removed market::dca sub-module — DCA types (ByDcaClass, ByDcaPeriod, ByDcaCagr) and computation logic moved to the new top-level investing module; market::returns now imports ByDcaPeriod from investing (source)
• Removed blocks::time sub-module — timestamp_monotonic compute logic and per-resolution timestamp imports relocated to indexes::timestamp::Timestamps (source)
• Changed lookback::compute() to take &indexes::Vecs instead of &time::Vecs, sourcing monotonic timestamps from the indexes module (source)
• Changed market::compute() parameter order: now takes (prices, indexes, blocks) instead of (indexes, prices, blocks) (source)

brk_indexer

• Moved indexer store ingest and fjall persist to background via run_bg(), running compaction, store commit, and fjall SyncData persist concurrently with the next indexing batch (source)

brk_client

• Added Pct0Pct1Pct2Pct5Pct95Pct98Pct99Pattern replacing Pct1Pct2Pct5Pct95Pct98Pct99Pattern to include the new 0.5th and 99.5th percentile bands (source)
• Added SeriesTree_Investing with full period/class sub-trees for DCA stack, cost basis, returns, CAGR, and lump-sum series (source)
• Added SeriesTree_Indicators_RealizedEnvelope with percentile price bands, index, and score series (source)
• Added SeriesTree_Indexes_Timestamp with monotonic and per-resolution timestamp series (source)
• Removed date and timestamp_monotonic from SeriesTree_Blocks_Time, simplified to only expose timestamp as a per-resolution pattern (source)

brk_query

• Updated all mining query endpoints (day1_iter, difficulty, epochs, hashrate) and series timestamp resolution to read from indexes.timestamp instead of blocks.time.timestamp (source)

website

• Refactored investing chart options to use the new investing tree path instead of market.dca (source)
• Added dedicated percentile color palette (_0_5 through _99_5) replacing the previous 6-color scheme with 8 distinct colors (source)
• Added createRatioChart() and groupedWindowsCumulative() shared chart builders for reuse across ratio-based chart options (source)

docker

• Switched health check from curl http://localhost:3110/health to pgrep -x brk, replacing curl package dependency with procps and reducing start period from 60s to 10s (source)

workspace

• Bumped Rust toolchain from 1.94.0 to 1.94.1 (source)
• Bumped quickmatch-js from 0.4.0 to 0.4.1 (source)

View changes

v0.2.2 - 2026-03-23

New Features

website

• Created a comprehensive LLM-readable API reference (llms-full.txt) documenting every endpoint with curl examples, response shapes, parameter descriptions, and series categories — enabling LLM agents to discover and use the BRK API via the llms.txt standard (source)
• Rewrote llms.txt from a minimal overview into a structured quick-start guide with live links to key endpoints (search, series data, blocks, transactions, addresses, fees, live price) and links to the full reference, OpenAPI specs, and interactive docs (source)
• Added llms-full.txt to robots.txt as a sitemap entry so crawlers and LLM agents can discover it (source)

brk_server

• Added Open Graph and Twitter Card meta tags to the Scalar API documentation page, providing proper title, description, and image for social media link previews (source)
• Renamed the API docs page title from "BRK API" to "BRK API — Bitcoin Research Kit" with a richer meta description emphasizing the free, no-auth nature of the API (source)

Internal Changes

workspace

• Re-enabled cargo build --workspace --release in the release script (was previously commented out) (source)

View changes

v0.2.1 - 2026-03-23

Breaking Changes

brk_rpc

• Made corepc the default feature — brk_rpc now uses the brk-corepc-client RPC backend by default instead of requiring explicit feature opt-in; users relying on bitcoincore-rpc must now explicitly enable it (source)

View changes

v0.2.0 - 2026-03-23

This is a major architectural release. The most significant changes are:

1. Self-contained, oracle-only pricing: The computation pipeline no longer uses exchange APIs for price data. All prices are now derived on-chain via the brk_oracle crate from block 550,000 onward, with pre-computed historical prices embedded for earlier blocks. brk_fetcher remains as a crate but is no longer used by brk_computer or brk_cli. Bitcoin Core is now the only runtime dependency — the toolkit runs fully offline and produces deterministic, reproducible results.

2. Height-only storage: All series now store only the per-block (height-indexed) vec on disk. Previously each series stored both a height vec and a date-indexed vec. All non-height resolutions (minute10 through year10, halving, epoch — 15 total, up from 8) are now computed lazily on demand from the height data using LazyAggVec with flat height-to-period mappings. This roughly halves on-disk vec count while nearly doubling the number of available resolutions.

3. Flat index architecture: The old cascading index hierarchy (height → date → week → month → quarter → year → decade) is replaced by a flat system where every period type maps directly to Height. This eliminates intermediate stored mappings and enables sub-day resolutions (10min, 30min, 1h, 4h, 12h) that were previously impossible.

4. Profitability cohorts: New UTXO bucketing by unrealized profit/loss percentage — 25 ranges from ">1000% in profit" to "90-100% in loss", plus 14 aggregate profit thresholds and 9 loss thresholds, powered by a 190K-bucket Fenwick tree for O(log N) per-block computation.

As a consequence of height-only storage, temporal aggregation shifted from date-based (daily sum, daily last, daily cumulative) to rolling windows (24h, 1w, 1m, 1y) computed from height data using precomputed lookback start positions (CachedWindowStarts). Most series now expose a block value, cumulative sum, and RollingComplete (rolling sums + rolling averages + rolling distribution with min/max/median/percentiles per window) — all derived from the single height vec.

Resolution data is served through a popularity-based caching layer: up to 256 height-source vecs are cached in memory with LRU eviction based on access frequency, while period-to-height index mappings are always cached since they're small and shared across all series.

Breaking Changes

brk_types

• Renamed all Address/Addresses types to Addr/Addrs across the entire public API: Address → Addr, AddressBytes → AddrBytes, AddressHash → AddrHash, AddressStats → AddrStats, AddressChainStats → AddrChainStats, AddressMempoolStats → AddrMempoolStats, AddressParam → AddrParam, AddressValidation → AddrValidation, AnyAddressIndex → AnyAddrIndex, FundedAddressData → FundedAddrData, EmptyAddressData → EmptyAddrData, etc. (source)
• Renamed all Metric/Metrics types to Series: MetricParam → SeriesParam, MetricData → SeriesData, MetricOutput → SeriesOutput, MetricCount → SeriesCount, MetricSelection → SeriesSelection, MetricsPaginated → SeriesPaginated, MetricWithIndex → SeriesNameWithIndex (source)
• Renamed DateIndex to Day1 and removed DecadeIndex, WeekIndex, MonthIndex, QuarterIndex, SemesterIndex, YearIndex — replaced with a comprehensive resolution system: Minute10, Minute30, Hour1, Hour4, Hour12, Day1, Day3, Week1, Month1, Month3, Month6, Year1, Year10 (source)
• Renamed DifficultyEpoch to Epoch and HalvingEpoch to Halving (source)
• Renamed CentsUnsigned to Cents and CentsUnsignedCompact to CentsCompact (source)
• Adopted snake_case for all module file names (e.g. blockinfo → block_info, blocktimestamp → block_timestamp, feeratepercentiles → feerate_percentiles, txstatus → tx_status, poolslug → pool_slug, etc.)

brk_computer

• All series now store height-indexed data only on disk — previously each metric stored both a height: EagerVec and a dateindex: EagerVec (and sometimes cumulative/sum variants at each level). Now PerBlock<T> has height: EagerVec + resolutions: Box<Resolutions<T>> where all 15 resolutions are LazyAggVec instances computed on demand. This roughly halves the on-disk vec count per series (source)
• Replaced cascading index hierarchy with flat height-to-period mappings — old system chained height→dateindex→weekindex→monthindex→yearindex→decadeindex; new system maps every period type directly to Height via EagerIndexes, enabling all resolutions to derive from height data without intermediate stored vecs (source)
• Restructured internal/ SDK from single/multi hierarchy to concept-oriented modules: per_block/, per_tx/, algo/, transform/, containers/, indexes/ (source)
• Moved blocks/mining/ and blocks/rewards/ into a dedicated top-level mining/ module with rewards/ and hashrate/ sub-modules (source)
• Moved market/indicators/ to a top-level indicators/ module with expanded scope (source)
• Renamed price module to prices with restructured sub-modules: by_unit.rs (multi-currency structs), ohlcs.rs (OHLC aggregation), compute.rs (oracle-based price computation) (source)
• Renamed cointime/pricing to cointime/prices (source)
• Renamed distribution/address to distribution/addr and distribution/cohorts/address to distribution/cohorts/addr
• Replaced all Clone-derived Vecs structs with generic StorageMode parameter (Vecs<M: StorageMode = Rw>) across every computation module (blocks, transactions, market, distribution, supply, etc.) for type-safe read-only sharing
• Removed brk_fetcher dependency — price computation is now fully on-chain via the oracle, eliminating exchange API calls (Binance, Kraken) and the 12-hour retry/HAR-import fallback logic
• Removed traits/ module (pricing traits) and utils.rs
• Computer::forced_import no longer accepts a Fetcher parameter — signature simplified from (path, indexer, fetcher) to (path, indexer)
• Restructured all market sub-modules with new type-safe containers replacing ad-hoc naming: price_1w_sma → sma._1w, price_ath → high, price_drawdown → drawdown, price_returns → periods, price_ago → price_past, etc.
• Restructured ATH vecs: price_ath → high.cents, price_drawdown → drawdown, days_since_price_ath → days_since, max_days_between_price_aths → max_days_between (source)
• Restructured DCA vecs: split into PeriodVecs (stack, cost_basis, return, cagr, lump_sum) and ClassVecs (stack, cost_basis, return), removing explicit days-in-profit/loss and min/max return fields (source)
• Restructured volatility vecs: replaced individual price_Xw_volatility, sharpe_Xw, sortino_Xw fields with Windows<LazyPerBlock<StoredF32>> (source)
• Restructured returns vecs: replaced individual period fields with ByLookbackPeriod<PercentPerBlock>, added sd_24h: Windows<StdDevPerBlock> replacing explicit 1w/1m/1y standard deviation and downside fields (source)
• Restructured range vecs: replaced 8 individual min/max price fields with PriceMinMaxVecs containing min/max groups, and added choppiness index as PercentPerBlock (source)
• Restructured moving average vecs: split into SmaVecs and EmaVecs using PriceWithRatioPerBlock<M> for each period, added Mayer Multiple bands (_200d_x2_4, _200d_x0_8) (source)
• Restructured burned supply: merged opreturn and unspendable into single total: AmountPerBlockCumulative with inline computation (source)
• Restructured transaction volume: replaced sent_sum/received_sum/annualized_volume with single transfer_volume: AmountPerBlockCumulativeRolling, wrapped throughput metrics in Windows<PerBlock> (source)
• Removed distribution/compute/aggregates.rs — aggregate computation moved into metric tiers
• Removed distribution/range_map.rs — range mapping moved to brk_types::RangeMap

brk_query

• Renamed address()/address_txids()/address_utxos() methods to addr()/addr_txids()/addr_utxos() (source)
• Renamed metrics query module to series (source)
• Removed resolved.rs — resolution handling moved inline

brk_server

• Renamed /api/address/* routes and AddressRoutes trait to AddrRoutes (redirects preserved from old paths) (source)
• Renamed /api/series/* routes (previously /api/metrics/*) — legacy /api/metrics/* routes preserved as redirects (source)
• Removed brk_fetcher dependency
• Removed Client from AppState — server no longer holds a direct RPC client reference

brk_client

• Switched HTTP backend from minreq to ureq
• Removed fetch_prices.rs example

brk_fetcher

• Switched HTTP backend from minreq to ureq

brk_rpc

• Added corepc feature integrating brk-corepc-client and brk-corepc-jsonrpc as alternative RPC backends (source)

brk_cohort

• Renamed AddressGroups to AddrGroups with renamed fields: ge_amount → over_amount, lt_amount → under_amount, ByGreatEqualAmount → OverAmount, ByLowerThanAmount → UnderAmount, ByAmountRange → AmountRange (source)
• Renamed ByAgeRange to AgeRange with renamed fields: up_to_1h → under_1h, from_15y → over_15y (source)
• Renamed file names from concatenated (by_age_range.rs, by_amount_range.rs) to snake_case (age_range.rs, amount_range.rs)

New Features

brk_computer

• Added indicators/ module computing on-chain indicators: Puell Multiple, RHODL Ratio, NVT Ratio, Thermocap Multiple, Supply-Adjusted CDD, Supply-Adjusted CYD, Supply-Adjusted Dormancy, Dormancy Flow, Stock-to-Flow, Seller Exhaustion Constant, and Gini coefficient (source)
• Added mining/ module with dedicated hashrate computations: hash rate with 1w/1m/2m/1y SMAs, hash rate ATH and drawdown, hash price and hash value per TH/s and PH/s, hash price/value minimums and rebound metrics (source)
• Added mining/rewards/ with coinbase rewards, subsidy, fee dominance, unclaimed rewards — each with block/cumulative/rolling views and sats/btc/usd units (source)
• Added market/technical/ module implementing RSI with Stochastic RSI (RSI, StochRSI, K-line, D-line with configurable RMA/SMA periods) and MACD (fast/slow EMA, signal line, histogram) (source)
• Added Fenwick tree (Binary Indexed Tree) data structure supporting O(log N) point-update, prefix-sum, and multi-target kth walk-down queries (source)
• Added ExpandingPercentiles tracker using Fenwick tree for O(log N) percentile queries with 0.1% resolution, supporting bulk-load initialization (source)
• Added CostBasisFenwick — 4-field Fenwick tree (all_sats, sth_sats, all_usd, sth_usd) for per-block accurate cost basis percentiles, supply density queries (±5% of spot price), and profitability range computation across 25 ranges with all/STH splits (source)
• Added sliding distribution, sliding median, sliding window, and drawdown algorithms in internal/algo/ (source)
• Added Resolutions<T> — a PerResolution of 15 LazyAggVec instances that lazily derive all supported resolutions from a single height-indexed source vec, using flat height-to-period mappings and a "last value in period" aggregation strategy (source)
• Added DerivedResolutions<T, S1T> — doubly-lazy resolution system for LazyPerBlock that chains UnaryTransform on top of source resolutions, enabling zero-storage derived series at all 15 resolutions (source)
• Added popularity-based resolution caching: CachedVec wrapping with a global budget of 256 slots, minimum 2 accesses before caching, and LRU eviction that replaces the least-accessed cached vec when a more popular one requests caching (source)
• Added prices/ module with SplitByUnit (open/high/low/close in USD/cents/sats), OhlcByUnit, and PriceByUnit structured types for multi-currency price access (source)
• Added ProfitabilityMetrics computing per-block supply, realized cap, unrealized PnL, and NUPL across 25 profitability ranges plus 14 profit thresholds and 9 loss thresholds, each with all/STH splits (source)
• Restructured distribution metrics into trait-based tiered system with ActivityLike, RealizedLike, UnrealizedLike trait hierarchies — each metric type has core/basic/minimal/full variants enabling flexible per-cohort metric composition (source)
• Added distribution/metrics/relative/ sub-module with 6 variants: RelativeToAll, RelativeForAll, RelativeWithExtended, RelativeFull, RelativeExtendedOwnMarketCap, RelativeExtendedOwnPnl for cohort-to-total ratio metrics (source)
• Added distribution/addr/activity.rs tracking per-block address activity counts (receiving, sending, reactivated, both) globally and per address type, with rolling averages (source)
• Added distribution/addr/delta.rs computing rolling address count deltas with change magnitude and basis-point rates (source)
• Added distribution/state/pending.rs with PendingDelta, PendingCapDelta, PendingInvestorCapDelta for batched cost basis state updates
• Added maturation tracking to age cohorts: tick_tock_next_block now returns AgeRange<Sats> tracking how many sats mature from each age range into the next older one per block, stored as matured: AgeRange<AmountPerBlockCumulativeRolling> on UTXOCohorts with cumulative and rolling window views for all 21 age ranges (source)
• Added pools/major.rs and pools/minor.rs separating major pools (with cumulative rolling rewards and dominance rolling windows) from minor pools (blocks mined with cumulative rolling and dominance ratio) (source)
• Added blocks/lookback.rs for precomputed rolling window lookback start vectors used across all windowed computations
• Added cache_budget.rs system for managing memory cache allocation across vecs
• Added tree.rs example for inspecting the full series tree structure

brk_types

• Added 11 new resolution index types expanding from 8 resolutions (day, week, month, quarter, semester, year, decade, difficulty epoch) to 15: Minute10, Minute30, Hour1, Hour4, Hour12 (sub-day — entirely new), Day3, Month3, Month6 (new intermediate periods), Year10 (replaces decade), plus renamed Day1 (was DateIndex), Week1 (was WeekIndex), Month1 (was MonthIndex), Year1 (was YearIndex), Halving (was HalvingEpoch), Epoch (was DifficultyEpoch) (source)
• Added FromCoarser trait enabling automatic derivation of finer-grained resolution data from coarser sources (source)
• Added basis points types: BasisPoints16, BasisPoints32, BasisPointsSigned16, BasisPointsSigned32 for precision ratio/percentage storage without floating point (source)
• Added StoredI64 type for signed 64-bit stored values (source)
• Added RangeIndex and RangeMap types for range-based lookups (source)
• Added SearchQuery type for series full-text search (source)
• Added SeriesInfo type carrying metadata about each data series (source)
• Added DataRange as a dedicated type (previously inline) (source)
• Added pools-v2.json with expanded mining pool identification data
• Added SeriesList and SeriesName types for typed series enumeration

brk_server

• Added structured JSON error responses with RFC 9457-inspired format: type (error category), code (machine-readable), message (human-readable), doc_url (link to API docs) — covers all BRK error variants with proper HTTP status mapping (source)
• Added X-Response-Time header (in microseconds) on all responses for latency observability
• Added Vary header consolidation middleware merging duplicate Vary headers into a single comma-separated value
• Added custom panic handler extracting panic messages into structured JSON error responses instead of empty 500s
• Added ContentEncoding module with server-side content encoding negotiation (zstd > br > gzip > identity) and transparent pre-compression of cached responses via brotli, flate2, and zstd (source)
• Added pre-compressed Brotli encoding for the Scalar API documentation viewer (scalar.js.br) (source)
• Extracted finish_openapi() and generate_bindings() as public functions for external use (source)
• Made ApiRoutes trait public for external router composition

brk_query

• Added list.rs example for enumerating all available series names (source)

brk_cohort

• Added ProfitabilityRange<T> with 25 buckets from ">1000% in profit" to "90-100% in loss", with boundary prices computed dynamically from spot price via compute_profitability_boundaries() (source)
• Added Profit<T> with 14 aggregate "at least X% profit" thresholds (prefix sums over ranges) and Loss<T> with 9 aggregate "at least X% loss" thresholds (suffix sums over ranges) (source)
• Added AgeRange::from_array() constructor for building age range structs from fixed-size arrays (source)
• Derived Traversable on AddrGroups via proc macro (previously manual Traversable impl on AddressGroups)

brk_traversable

• Added iter_any_visible() method to the Traversable trait — allows structs to expose all vecs for disk operations while hiding some from the public series tree (source)
• Added Traversable impls for LazyAggVec, LazyDeltaVec, ReadOnlyCompressedVec, ReadOnlyRawVec — enabling lazy and read-only vecs to appear in the series tree

brk_traversable_derive

• Added #[traversable(hidden)] attribute at struct and field level — hidden structs return empty tree nodes and empty iter_any_visible() while still participating in iter_any_exportable() for disk I/O (source)
• Added automatic ReadOnlyClone derive generation — the proc macro now generates impl ReadOnlyClone alongside impl Traversable, mapping Rw to Ro storage mode for all fields with StorageMode parameters
• Added multi-level wrap paths (#[traversable(wrap = "a/b")]) for creating nested tree structure
• Preserved struct field declaration order in tree key ordering (previously normal and flattened fields were separated)

brk_oracle

• Lowered START_HEIGHT from 575,000 to 550,000 — oracle begins computing 25,000 blocks earlier (source)

brk_reader

• Added timestamp-based pre-filtering in block decoding — blocks are skipped by header timestamp before computing block hash, avoiding expensive RPC calls for out-of-range blocks (source)
• Added find_magic() function with XOR-decoded sliding window for blk file scanning, replacing byte-by-byte loop
• Replaced batch-based block decoding pipeline (accumulated blocks, sleep-based backpressure) with direct par_bridge() streaming — simpler, no sleep delays

brk_bindgen

• Added template-based pattern system with discriminator support — try_detect_template() handles suffix discriminators (e.g. ratio_sd vs ratio_sd_4y) and embedded discriminators (e.g. ratio_pct99_bps vs ratio_pct1_bps) for smarter client code generation (source)
• Added two-pass field part analysis: first pass collects instance analyses bottom-up, second pass fills mixed-empty field parts using shortest leaf names
• Added owned_expr(), disc_arg_expr(), template_expr() to LanguageSyntax trait for language-specific discriminator formatting
• Fixed pattern normalization to only erase leaf types when all leaves share the same type (prevents incorrect genericization of mixed-type signatures)

brk_client (Rust)

• Switched HTTP backend from minreq to ureq
• Removed fetch_prices.rs example

brk_client (JavaScript)

• Added tests/consistency.js — verifies all series sharing the same index have the same length (catches stale state after reorg rollback)
• Upgraded quickmatch 0.3.1 → 0.4.0: expanded default separators (_- :/), added prefix indexing and compound word indexing (e.g. "hashrate" now matches "hash_rate"), added minScore threshold to reduce false positives, replaced per-match object allocation with pre-allocated Uint32Array scoring

brk_client (Python)

• Added DateSeriesData subclass with date-aware methods (.dates(), .date_items(), .to_date_dict()) — date-based indexes return DateSeriesData, non-date indexes return SeriesData
• Added new API methods: get_series_data(), get_series_latest(), get_series_len(), get_series_version(), date_to_index()
• Added DetailedSeriesCount with distinct_series, total_endpoints, lazy_endpoints, stored_endpoints fields
• Added ErrorDetail/ErrorBody types matching the server's structured error responses
• Range parameters now accept date and datetime objects in addition to integers

Internal Changes

brk_computer

• Rewrote the internal/ SDK: replaced the single/multi hierarchy (160+ files with per-combination types like lazy_binary_computed_full, value_lazy_sum_cum, ComputedFromHeightAndDateLast, ComputedFromHeightSumCum) with a composable system organized by concept (per_block/, per_tx/, algo/, transform/, containers/, indexes/). Concrete example: block count went from ~15 on-disk vecs (ComputedFromHeightSumCum + 4 rolling start vecs + 4 ComputedFromHeightLast) down to ~2 on-disk vecs (PerBlockCumulativeRolling) with all resolutions lazy
• Added UnaryTransform trait system in internal/transform/ replacing ad-hoc closures: arithmetic.rs, bps.rs, currency.rs, derived.rs, ratio.rs, specialized.rs
• Added container types in internal/containers/: per_resolution, windows, windows_from_1w, windows_to_1m, window_24h, distribution_stats, constant, percent
• Added per-block type variants: PerBlock, PerBlockFull, PerBlockCumulative, PerBlockCumulativeWithSums, PerBlockRolling, PerBlockAggregated, LazyPerBlock, LazyPerBlockRolling, plus AmountPerBlock, AmountPerBlockCumulative, AmountPerBlockCumulativeRolling, AmountPerBlockWithDeltas, FiatPerBlock, PercentPerBlock, RatioPerBlock, PerBlockRollingAverage, PercentRollingWindows, PriceWithRatioPerBlock, StdDevPerBlock
• Added rolling types: RollingComplete, LazyRollingComplete, RollingDistribution, LazyRollingSums, LazyRollingDeltasFromHeight
• Added per-tx types: PerTxDerived, PerTxDistribution, LazyPerTxDerived, LazyPerTxDistribution, LazyPerTxDistributionTransformed
• Added eager/lazy index helpers in internal/indexes/
• Introduced CachedWindowStarts for shared precomputed lookback window start positions across all modules
• Split distribution UTXO cohort vecs into core.rs, minimal.rs, type.rs sub-modules for tiered storage
• Added sum_others macro in distribution metrics for computing aggregate cohort fields by summing constituent cohorts
• Removed brk_fetcher dependency — oracle prices are computed directly in the prices/ module
• Removed distribution/compute/aggregates.rs — aggregate computation replaced by compute_from_stateful / compute_sum_of_others within each metric tier

brk_indexer

• Bumped reindex version 24 → 25 (forces full reindex on upgrade)
• Added StorageMode generic parameter on Indexer<M> and Vecs<M> with ReadOnlyClone impl producing Indexer<Ro> for type-safe read-only sharing
• Parallelized block processing pipeline: inputs and outputs now processed in parallel via rayon::join, finalization of outputs+inputs runs in parallel with tx metadata storage
• Added BlockBuffers struct with reusable FxHashMap/FxHashSet allocations cleared per block, eliminating per-block allocation overhead
• Added AddrTypeVecs<I, B> and ScriptTypeVecs<I> generic sub-structs for uniform per-type vec organization, replacing individually named fields
• Added TxMetadataVecs sub-struct with split_for_finalize() for separating tx metadata vecs from index vecs during parallel finalization
• Added AddrReaders with script_pubkey(&self, output_type, type_index) -> ScriptBuf for typed address-to-script lookup
• Block size and weight now computed from pre-computed ComputedTx.base_size/total_size fields instead of re-serializing transactions
• Removed 512MB thread stack size workaround — main function runs directly
• Added schemars and serde dependencies for OpenAPI-compatible schema generation

brk_server

• Applied NormalizePathLayer as an outer wrapping layer (via ServiceExt) instead of an inner router layer, fixing route matching for trailing slashes
• Reduced in-memory response cache from 5,000 to 1,000 entries
• Moved CatchPanicLayer after compression in the middleware stack with a custom handler
• Restructured response caching to include content encoding in cache keys, pre-compressing responses per-encoding for cache-friendly serving
• Added brotli, flate2, and zstd dependencies for server-side compression
• Added color-eyre for panic reporting

brk_query

• Removed thread stack-size workaround in the query example — main function now runs directly
• Added parking_lot and serde_json dependencies

brk_rpc

• Added serde and serde_json dependencies for corepc feature

brk_store

• Removed insert_if_needed() and remove_if_needed() — callers now use plain insert()/remove()
• Changed clear_caches() to drop allocations entirely (FxHashMap::default()) instead of clearing in place

brk_reader

• Removed batch-based block decoding pipeline — simplified to streaming par_bridge()
• Removed AnyBlock enum, replaced with decode_block() standalone function
• Added find_magic() with XOR-decoded sliding window for efficient blk file scanning

brk_logger

• Replaced EnvFilter with Targets for simpler log filter configuration
• Separated LOG (default level) from RUST_LOG (full directive control) environment variables
• Added corepc=off to default filter list

brk_error

• Added brk-corepc-client and ureq error conversions, removed minreq error conversions

brk_cli

• Removed --fetch CLI flag and brk_fetcher dependency — price fetching handled by the oracle
• Removed 512MB thread stack size workaround

docker

• Switched base image from rustlang/rust:nightly to rust:1.93-bookworm (stable, pinned)
• Replaced pgrep + nc healthcheck with curl -sf http://localhost:3110/health
• Removed docker-build.sh helper script and docker-specific .dockerignore
• Removed all environment: variables from docker-compose.yml — configuration now via command: args only

scripts

• Added pool_major_threshold.py — utility to determine major vs minor pool classification based on rolling dominance thresholds across multiple time windows

workspace

• Upgraded Rust toolchain to edition 2024
• Updated .dockerignore to exclude docker/ directory

View changes

v0.1.9 - 2026-02-13

New Features

brk

• Added oracle feature flag re-exporting brk_oracle from the brk umbrella crate, making the on-chain price oracle available via brk::oracle (source)
• Added brk_oracle to the crates table in the umbrella crate README (source)

Internal Changes

brk

• Added brk_oracle to the Rust publish script ordering (source)

View changes

v0.1.8 - 2026-02-13

Breaking Changes

brk_types

• Renamed Sats::is_round_btc() to is_common_round_value() and rewrote the algorithm: replaces a hardcoded array of 19 specific round sat values with a general integer-based approach using ilog10() to detect any value matching d × 10^n where d ∈ {1, 2, 3, 5, 6} (source)

brk_server

• Renamed CacheStrategy::MaxAge(u64) to CacheStrategy::MempoolHash(u64), switching mempool endpoints from time-based Cache-Control: max-age=N to content-addressed ETags derived from the next projected block hash (source)

New Features

brk_oracle

• Added new brk_oracle crate: a pure on-chain BTC/USD price oracle that derives the bitcoin price from round-dollar transaction output patterns ($1, $5, $10, ... $10,000) without any exchange data (source)
• Algorithm maps outputs to a log-scale histogram (2,400 bins, 200 per decade), smooths with an EMA over a 12-block ring buffer, then slides a 19-point stencil representing round-dollar spikes to find the best-fit price with parabolic sub-bin interpolation
• Operates from height 575,000 (May 2019) onward; pre-oracle exchange prices for heights 0–630,000 included in prices.txt for bootstrapping
• Median per-block error 0.10%, 95th percentile 0.55%; zero blocks exceed 10% error since 2022
• Inspired by UTXOracle by @SteveSimple, with per-block resolution, single-pass stencil scoring, and rolling EMA operation

brk_computer

• Enabled oracle price computation with full OHLC data at all period levels (date, week, month, quarter, semester, year, decade, height, difficulty epoch) (source)
• Added live_oracle() method that seeds an oracle from the last committed price and replays recent blocks for real-time price estimation (source)
• Added market_cap_growth_rate metric: 365-day percentage change of market cap (source)
• Added realized_cap_growth_rate metric: 365-day percentage change of realized cap
• Added cap_growth_rate_diff metric: lazy difference between market cap and realized cap growth rates
• Added net sentiment computation for UTXO cohorts: computes greed minus pain per separate cohort, weighted average for aggregate cohorts, and dateindex for all cohorts (source)

brk_server

• Added server-side response caching via get_or_insert() on AppState, backed by quick_cache with 50ms guard timeout for request deduplication (source)
• Added GET /api/mempool/price endpoint returning the current BTC/USD price in dollars, derived from on-chain round-dollar output patterns in the last 12 blocks plus mempool (source)
• Added CF-Connecting-IP header detection for requests arriving from loopback addresses through Cloudflare, rewriting the socket address to apply non-loopback rate limits (source)
• Switched address mempool endpoint to content-based cache invalidation using a hash of the address's mempool state

brk_query

• Added live_price() method returning the oracle-derived BTC/USD price, seeded from the last committed price with replay of the last window of blocks and mempool transactions (source)
• Added address_mempool_hash() method returning a u64 hash of an address's mempool state for content-based cache invalidation (source)

brk_mempool

• Added Snapshot::next_block_hash() returning a hash of the first projected block for content-based cache keys (source)
• Added Sync::next_block_hash() and Sync::address_hash() for exposing content hashes at the sync layer (source)

brk_types

• Derived Hash on Sats for use in content-based cache invalidation (source)
• Derived Hash on AddressMempoolStats for per-address mempool hashing (source)

brk_client

• Added get_live_price() method for the /api/mempool/price endpoint (source)
• Added MetricsTree_Price_Oracle struct exposing oracle price metrics (price_cents, ohlc_cents, split, ohlc, ohlc_dollars)
• Added market_cap_growth_rate, realized_cap_growth_rate, and cap_growth_rate_diff fields to MetricsTree_Supply

website

• Replaced Kraken WebSocket price feed with HTTP polling to the oracle-based /api/mempool/price endpoint every 5 seconds, with re-poll on tab visibility change (source)
• Added Oracle candlestick chart under the Market section showing on-chain derived BTC/USD price (source)
• Expanded Capitalization into a tree with Market Cap, Realized Cap, and Growth Rate sub-charts showing market cap growth rate, realized cap growth rate, and their difference
• Added dynamic color assignment via at(index, length) for palette-aware spacing and seq(keys) for named color maps, replacing hardcoded color mappings across all chart options (source)

Internal Changes

brk_computer

• Flattened supply submodules (circulating/, inflation/, market_cap/) into the parent supply module, inlining their logic directly into vecs.rs, import.rs, and compute.rs (source)
• Changed CostBasisDistribution::serialize_iter to accept merged.into_iter() and write to a path variable

brk_server

• Centralized response caching in state.rs with cached_json(), cached_text(), and cached_bytes() now accepting a Uri parameter for server-side cache keying, removing per-handler GuardResult boilerplate
• Removed ResponseExtended::new_text_cached() and new_bytes_cached() in favor of inline handling in the state caching methods
• Updated all API handlers to destructure State(state) and pass &uri for cache keying

website

• Split monolithic components.css into separate files: main.css, nav.css, search.css, chart.css for better organization
• Converted chart option top and bottom properties from eager values to lazy functions with memoized evaluation for faster startup (source)
• Refactored chart component: simplified pane management, converted legends from named object to array, consolidated time data into a single object with setIndex() and fetch() methods (source)
• Optimized navigation highlight tracking to only clear previously highlighted items instead of iterating all entries
• Removed shadow divs from HTML and <div id="simulation"> pane
• Removed unused utility files (ws.js, date.js, serde.js) and functions (range(), numberToDollars(), numberToPercentage(), createShadow(), localhost)

clients

• Bumped JavaScript and Python client versions

View changes

v0.1.7 - 2026-02-07

New Features

brk_computer

• Added lower_price_band metric per cohort, computed as realized_price² / investor_price, representing a floor price estimate (source)
• Added upper_price_band metric per cohort, computed as investor_price² / realized_price, representing a ceiling price estimate
• Added LazyBinaryPriceFromHeight composite type for fully lazy binary price metrics providing both USD and sats representations at all time levels (source)
• Added DollarsSquaredDivide binary transform implementing a² / b for Dollars values (source)

brk_types

• Added Log200 variant to CostBasisBucket for logarithmic bucketing with 200 buckets per decade (source)
• Added From<CentsSigned> for i128 and From<i128> for CentsSigned conversions for wider arithmetic operations (source)

brk_website

• Added ETag-based HTTP 304 Not Modified responses for HTML pages in release mode, reducing bandwidth by returning empty body when content hasn't changed (source)
• Compute and cache a hash-based ETag for index.html on first request, used for subsequent conditional requests (source)

website

• Added upper price band (I²/R) and lower price band (R²/I) series to the price comparison chart (source)

Bug Fixes

brk_types

• Fixed unrealized value calculation in cost basis distribution to compute (spot - cost_basis) × supply instead of spot × supply, correctly reflecting unrealized profit/loss per bucket (source)
• Fixed Dollars * Sats multiplication to use i128 intermediate values, preventing overflow on large values (source)

Internal Changes

brk_computer

• Moved price rounding logic and unrealized state cache from CohortState into CostBasisData, encapsulating all cost basis concerns in a single structure (source)
• Added compute_unrealized_states() method on CostBasisData that combines cache management and unrealized computation in one call
• Removed CachedUnrealizedState from public API, making it pub(super) only
• Removed range() method from CostBasisData (unrealized computation now accesses the map directly)
• Added binary transform constructors (from_block_last_and_lazy_block_last, from_lazy_block_last_and_block_last, from_binary) across height, date, and height-derived lazy metric levels to support mixed computed/lazy binary operations

website

• Changed shared time data type from number[] to MetricData<number> throughout the chart system, propagating stamp metadata alongside data
• Added lastTimeStamp tracking to prevent redundant chart updates when time data hasn't changed

clients

• Bumped JavaScript and Python client versions

View changes

v0.1.6 - 2026-02-05

New Features

brk_types

• Added CostBasisDistribution struct with pco-compressed serialization/deserialization for storing price-to-sats distributions, and a format() method supporting bucket aggregation and value type transformation (source)
• Added CostBasisBucket enum with 7 aggregation strategies: raw (no aggregation), lin200/lin500/lin1000 (linear buckets by dollar amount), and log10/log50/log100 (logarithmic buckets per decade) (source)
• Added CostBasisValue enum with three output modes: supply (BTC at each price), realized (USD cost basis), unrealized (USD at spot price) (source)
• Added Cohort, CostBasisParams, CostBasisCohortParam, and CostBasisQuery types for cost basis API path and query parameters (source)
• Added round_to() and round_to_dollar() methods on CentsUnsigned and CentsUnsignedCompact for rounding prices to N significant digits (source)
• Implemented FromStr for Date type enabling parsing from "YYYY-MM-DD" format (source)

brk_computer

• Added daily cost basis snapshot writing for all, sth, and lth UTXO cohorts, persisting the merged price-to-sats distribution to disk per date for historical querying (source)
• Added configurable price rounding for cost basis data: UTXO cohorts use 5 significant digits and address cohorts use 4 significant digits, reducing unique price points for more efficient storage and aggregation (source)

brk_query

• Added cost basis query methods: cost_basis_cohorts() to list available cohorts, cost_basis_dates() to list available dates per cohort, cost_basis_distribution() to retrieve raw data, and cost_basis_formatted() to get bucketed/value-transformed output (source)

brk_server

• Added three new API endpoints for cost basis distribution: GET /api/metrics/cost-basis (list cohorts), GET /api/metrics/cost-basis/{cohort}/dates (list dates), and GET /api/metrics/cost-basis/{cohort}/{date}?bucket=&value= (get distribution data) (source)

brk_client

• Added get_cost_basis_cohorts(), get_cost_basis_dates(), and get_cost_basis() methods to the Rust client for querying cost basis distributions (source)

brk_bindgen

• Added wildcard (*) and Object type mapping to serde_json::Value in Rust type generation (source)

Bug Fixes

website

• Fixed chart range restoration on Safari by wrapping the range-set logic in requestAnimationFrame for browser compatibility (source)

Internal Changes

brk_computer

• Reorganized cost basis state storage: height-based snapshots now stored under by_height/ subdirectory, renamed cost_basis_data module to data
• Moved pco compression dependency from brk_computer to brk_types for reuse across crates
• Refactored the K-way merge in aggregate percentile computation to collect merged entries with price deduplication during the merge pass

brk_error

• Added pco feature flag and Pco error variant for pco compression errors (source)

workspace

• Added *.csv to .gitignore

clients

• Bumped JavaScript and Python client versions

View changes

v0.1.5 - 2026-02-05

New Features

website

• Added in-memory data cache for chart metric and time data, enabling instant visual feedback when switching between chart indexes without re-fetching from the server (source)
• Implemented cache-first loading strategy: cached data is rendered immediately while fresh data is fetched in the background, making index switching feel instantaneous

Bug Fixes

website

• Fixed URL range being overwritten during chart initialization by adding an initialLoadComplete guard that prevents range persistence until the first data load completes, and resets on rebuild (source)
• Fixed temporal dead zone (TDZ) error by moving zoom handler registration (onZoomChange.add) to after series creation instead of before it (source)

Internal Changes

website

• Simplified showLine visibility calculation to use the already-tracked visibleBarsCount variable instead of re-reading the visible logical range from the chart

clients

• Bumped JavaScript and Python client versions

View changes

v0.1.4 - 2026-02-04

Breaking Changes

brk_computer

• Changed VOCDD (Value of Coin Days Destroyed) calculation to be supply-adjusted: price × (CDD / circulating_supply) instead of price × CDD, accounting for growing supply over time (source)
• Changed Reserve Risk calculation to use 365-day rolling median instead of 365-day SMA for VOCDD smoothing, providing more robust values resistant to outliers (source)
• Changed net_sentiment computation for aggregate cohorts to use capital-weighted average of component cohorts instead of simple greed - pain formula, producing values that properly represent the weighted contribution of each cohort (source)
• Changed drawdown computation formula from (ath - current) / ath * -100 to (current - ath) / ath * 100, where positive values now indicate performance above ATH and negative values indicate drawdown below ATH (source)

New Features

brk_computer

• Added hash_rate_ath metric tracking the all-time high hash rate at each block (source)
• Added hash_rate_drawdown metric showing percentage drawdown from hash rate all-time high (source)
• Added _30d_change metric for all address type counts (p2pkh, p2sh, p2wpkh, p2wsh, p2tr, p2pk33, p2pk65, p2a), tracking 30-day percentage change in address counts (source)
• Added utxo_count_30d_change metric to OutputsMetrics tracking 30-day percentage change in UTXO count per cohort (source)

brk_types

• Added From<StoredU64> conversion for StoredF64 enabling direct u64 to f64 stored value conversion (source)

Internal Changes

brk_computer

• Generalized ComputeDrawdown trait to work with any value types convertible to f64, not just Close<Dollars> (source)
• Removed NegRatio32 transform (no longer needed after drawdown formula change) (source)
• Separated net_sentiment computation into compute_net_sentiment_height for separate cohorts and compute_net_sentiment_from_others for aggregate cohorts (source)
• Added AddrCountVecs struct grouping address count with its 30d_change metric (source)
• Refactored aggregate UTXO cohort computation with for_each_aggregate helper method for cleaner code organization
• Added unit tests for HODL Bank and Reserve Risk formula verification (source)

brk_indexer

• Removed debug logging from transaction metadata storage for cleaner logs (source)

website

• Reorganized chart options into new module structure with dedicated files for activity, profitability, cost-basis, holdings, prices, and valuation
• Consolidated color utilities into website/scripts/utils/colors.js module

clients

• Bumped JavaScript and Python client versions

View changes

v0.1.3 - 2026-02-03

Breaking Changes

brk_types

• Renamed LoadedAddressIndex to FundedAddressIndex throughout the codebase (source)
• Renamed LoadedAddressData to FundedAddressData (source)
• Renamed OHLCCents to OHLCCentsUnsigned for consistency with unsigned cents types (source)
• Removed Cents, CentsCompact, and CentsSignedCompact types (consolidated into unsigned variants)

New Features

brk_types

• Added SatsSigned type (i64-based) for signed satoshi values used in changes, deltas, and profit/loss calculations (source)
• Added CentsSats type (u128-based) for price × sats calculations with widening multiplication to prevent overflow (source)
• Added CentsSquaredSats type (u128-based) for price² × sats calculations used in investor cap metrics (source)

brk_computer

• Added investor_price metric - dollar-weighted average acquisition price with both USD and sats representations (source)
• Added cap_raw metric storing raw Σ(price × sats) for precise realized cap aggregation across cohorts
• Added investor_cap_raw metric storing raw Σ(price² × sats) for precise investor_price aggregation
• Added realized_profit_7d_ema, realized_loss_7d_ema, and net_realized_pnl_7d_ema metrics for smoothed realized PnL analysis
• Added profit_value_created and profit_value_destroyed metrics splitting value flows by profit transactions
• Added loss_value_created and loss_value_destroyed metrics splitting value flows by loss transactions
• Added capitulation_flow as a lazy alias for loss_value_destroyed (coins sold at a loss)
• Added profit_flow as a lazy alias for profit_value_created (coins sold at a profit)
• Added peak_regret metric: Σ((peak - sell_price) × sats) measuring how much more could have been made by selling at peak (source)
• Added peak_regret_rel_to_realized_cap expressing peak regret as percentage of realized cap
• Added invested_capital_in_profit and invested_capital_in_loss metrics tracking cost basis of UTXOs by profit/loss status (source)
• Added raw aggregation values for unrealized metrics: invested_capital_in_profit_raw, invested_capital_in_loss_raw, investor_cap_in_profit_raw, investor_cap_in_loss_raw
• Added pain_index metric: investor_price_of_losers - spot, measuring average distance underwater weighted by USD invested
• Added greed_index metric: spot - investor_price_of_winners, measuring average distance in profit weighted by USD invested
• Added net_sentiment metric: greed_index - pain_index, positive values indicate greedy market, negative indicate painful market
• Added unrealized_peak_regret metric for age-range UTXO cohorts
• Added invested_capital percentiles providing USD-weighted cost basis distribution (complementing sat-weighted percentiles) (source)
• Added spot_cost_basis_percentile metric showing what percentile of cost basis is below current spot price (sat-weighted)
• Added spot_invested_capital_percentile metric showing what percentile of invested capital is below current spot price (USD-weighted)
• Added ValueChangeFromDate module for computing N-day value changes stored as signed sats (source)
• Added LazyValueChangeDateDerived module for lazy computation of bitcoin and dollar values from signed sats changes (source)
• Added LazyPriceFromCents module providing lazy price wrapper with both USD and sats representations from a cents base (source)
• Added CentsUnsignedToDollars transform for converting cents to dollars (source)
• Added CentsUnsignedToSatsFract transform for converting cents to sats exchange rate (source)

brk_cohort

• Added compute_relative() method to Filter that returns false for edge-case output types (Empty, P2MS, Unknown) which have insufficient volume for meaningful ratio analysis (source)

Internal Changes

brk_computer

• Replaced PriceToAmount struct with CostBasisData providing more efficient storage with raw cap and investor_cap values for precise aggregation (source)
• Added Percentiles struct that computes both sat-weighted and USD-weighted percentiles in a single pass (source)
• Changed UnrealizedState to use CentsUnsigned instead of Dollars for unrealized profit/loss values
• Changed realized_cap to derive from realized_cap_cents via lazy transform for consistent precision
• Aggregate percentile computation now supports both sat-weighted and USD-weighted metrics with spot price parameter
• Removed ClosePriceTimesRatio transform (functionality consolidated into other transforms)

brk_bindgen

• Added pattern deduplication by name to prevent duplicate patterns when different signatures normalize to the same name (source)
• Changed to BTreeMap/BTreeSet for deterministic pattern generation order
• Changed pattern naming to generate stable names from first word of each field (deduped, sorted)

brk_fetcher

• Changed return types from OHLCCents to OHLCCentsUnsigned for consistency (source)

brk_traversable

• Changed TreeNode::Branch internal storage from BTreeMap to IndexMap to preserve insertion order in API tree responses (source)

website

• Removed website/scripts/options/market/ subdirectory structure (averages.js, bands.js, index.js, investing.js, momentum.js, onchain.js, performance.js, volatility.js)
• Consolidated market options into top-level market.js and investing.js files
• Added new mining.js and network.js option modules
• Removed chain.js module

View changes

v0.1.2 - 2026-01-29

New Features

brk_bindgen

• Added indexToDate(index, i) and isDateIndex(index) helper functions to JavaScript client for converting index values to Date objects (source)
• Added index_to_date(index, i) and is_date_index(index) helper functions to Python client (source)
• Added rich helper methods to JavaScript MetricData response object:
  • dates() - convert index range to Date array
  • indexes() - get index range as number array
  • toDateMap() / toIndexMap() - convert to Map keyed by date or index
  • dateEntries() / indexEntries() - get as [key, value] pair arrays
  • iter() / iterDates() - generator functions for iteration
  • [Symbol.iterator]() - enable for...of loop support
• Added rich helper methods to Python MetricData dataclass:
  • dates() - convert index range to date list
  • indexes() - get index range as list
  • to_date_dict() / to_index_dict() - convert to dict
  • date_items() / index_items() - get as tuple lists
  • iter() / iter_dates() - iterators for (index, value) or (date, value)
  • to_polars(with_dates) / to_pandas(with_dates) - convert to DataFrame with optional date column
• Added index field to MetricData JSON response indicating the index type used for the query

brk_types

• Added CentsSigned type (i64-based) for signed cent values used in profit/loss calculations (source)
• Added CentsSignedCompact type (i32-based) for memory-efficient signed cents in map keys (source)
• Added CentsUnsigned type (u64-based) for unsigned cent values used in invested capital and realized cap (source)
• Added CentsUnsignedCompact type (u32-based) for memory-efficient unsigned cents (source)
• Added is_date_based() method to Index enum to check if an index type represents dates (source)
• Added index_to_date(i) method to Index enum for converting index values to Date (source)
• Added From implementations to convert date-based index types (WeekIndex, MonthIndex, YearIndex, QuarterIndex, SemesterIndex, DecadeIndex) to Date (source)

brk_computer

• Added invested_capital_in_profit and invested_capital_in_loss fields to UnrealizedState for tracking cost basis of positions (source)

brk_server

• Added higher API weight limit (50MB) for localhost requests compared to remote requests (650KB) to support larger local data queries (source)

brk_logger

• Added LOG environment variable support for simple log level control (e.g., LOG=debug brk) (source)

brk_cli

• Added ENVIRONMENT and CONFIG sections to help output documenting LOG/RUST_LOG variables and config file location (source)

workspace

• Added GitHub Actions workflow to check for outdated dependencies daily (source)
• Added SIMD instruction flags (BMI1, BMI2, AVX2) for x86_64 targets to accelerate pcodec decompression (source)

Bug Fixes

brk_indexer

• Fixed rollback consistency: deduplicate addressindex_txindex removals to prevent double deletion when same address appears multiple times in a transaction (source)
• Fixed rollback cache consistency: clear txid caches after rollback to prevent stale reads (source)

brk_computer

• Fixed rollback recovery to use actual consistent height instead of target height for state imports (source)
• Fixed detection of failed rollbacks when height lands ahead of target, now falls back to fresh start (source)
• Fixed address index lookups to use get_any_or_read_at_unwrap instead of get_pushed_or_read_at_unwrap to properly check updated layer after rollback (source)
• Fixed height subtraction overflow using saturating_sub in computer example (source)

brk_cli

• Fixed config file not being overwritten on every startup, preserving user modifications (source)

Internal Changes

brk_bindgen

• Switched from HashMap to BTreeMap in pattern analysis for deterministic code generation order (source)
• Added write_if_changed() utility to skip file writes when content is unchanged, avoiding unnecessary cargo rebuilds (source)

brk_store

• Added clear_caches() method to clear all partition caches after bulk removals like rollback (source)

brk_indexer

• Removed unused COLLISIONS_CHECKED_UP_TO constant (source)
• Added debug logging for transaction metadata storage showing inserted/skipped counts (source)
• Added error logging for unknown TXID cases with full context including store result and current txindex (source)

brk_computer

• Added extensive debug and warning logging throughout rollback recovery process (source)

brk_server

• Added middleware to extract client socket address and pass to API handlers (source)

brk_logger

• Changed log file naming format from {prefix}.{date} to {prefix}_{date}.txt (source)

brk_cli

• Increased server startup delay from 3 to 10 seconds when indexing is far behind (source)
• Added separator lines around "Indexing blocks" message for visibility (source)

brk_types

• Added comprehensive unit tests for Date from index type conversions (source)

View changes

v0.1.1 - 2026-01-27

Bug Fixes

brk_website

• Fixed HTML file caching rules to properly detect .html files and apply must-revalidate cache policy instead of immutable caching (source)

brk_client

• Fixed metric name mappings in ClassDaysInLossPattern (changed from max_drawdown to days_in_profit)
• Regenerated client code with corrected metric patterns

Internal Changes

scripts

• Split release scripts into separate components for better modularity

View changes

v0.1.0 - 2026-01-27

New Features

brk_computer

• Added sats-denominated versions of all price metrics alongside USD versions (source)
  • New Price wrapper struct provides both .dollars and .sats representations
  • Derefs to dollars for backward compatibility with existing code
  • Sats values computed via DollarsToSatsFract transform
• Added address activity tracking metrics (source)
  • receiving - Unique addresses that received funds in each block
  • sending - Unique addresses that sent funds in each block
  • reactivated - Addresses that were empty and now have funds
  • both - Addresses that both sent AND received in same block
  • balance_increased - Receive-only addresses
  • balance_decreased - Send-only addresses
  • Available globally and per address type (P2PKH, P2SH, P2WPKH, etc.)
• Added address growth rate metrics (source)
  • Computes new_addr_count / total_addr_count ratio globally and per type
  • Includes distribution statistics (min, max, average, percentiles)
• Added new address count metrics per block
• Added total address count tracking
• Added RatioU64F32 transform for u64 to f32 ratio calculations
• Added U64Plus transform for u64 addition operations
• Added price-aware multi modules: lazy_binary_price, lazy_price, unary_last
• Added height-based distribution and price modules

View changes

v0.1.0-beta.1 - 2026-01-26

New Features

brk_types

• Added SatsFract type for fractional satoshi representation (source)
  • Stores f64 values representing sub-satoshi amounts
  • Useful for expressing USD prices in sats when price is high (e.g., at $100k, $0.0001 = 0.1 sats)
  • Includes arithmetic operations and NaN handling

brk_computer

• Added DollarsToSatsFract transform for converting USD prices to sats exchange rate (source)
• Added CloseDollarsToSatsFract variant for close price conversion

website

• Added more investing data and charts

Internal Changes

scripts

• Updated release workflow scripts

View changes

v0.1.0-beta.0 - 2026-01-25

Breaking Changes

brk_mcp

• Removed brk_mcp crate from workspace (MCP server functionality removed)

New Features

brk_computer

• Added Reserve Risk metric to cointime module (source)
  • Computes VOCDD 365-day SMA (Value-weighted Outstanding Coin Days Destroyed)
  • Calculates HODL Bank accumulation metric
  • Derives Reserve Risk indicator for market cycle analysis
• Added return_i8 transform for small integer returns

brk_website (New Crate)

• Created standalone website serving crate without BRK data layer dependencies (source)
• Supports both embedded website and filesystem path serving
• Provides router() function for Axum integration
• Added example demonstrating standalone server usage

brk_playground (New Standalone Crate)

• Created experimental playground crate for on-chain price analysis (not published to crates.io) (source)
• Implements phase histogram analysis of UTXO patterns
• Provides filter-based output selection for price signal extraction
• Includes on-chain OHLC price oracle derivation with confidence metrics
• Features anchor-based price calibration using weekly OHLC data

brk_logger

• Added log hook system for custom log message handling (source)
• Added rate-limited file writer (100 writes/second max) to prevent log flooding (source)
• Added custom log format module

brk_client

• Added fetch_prices.rs example demonstrating price data fetching

brk_types

• Added StoredI8 type for signed 8-bit integer storage

brk_server

• Added Scalar API documentation JavaScript

Internal Changes

brk_computer

• Refactored vecs.rs files to remove verbose comments
• Reduced verbosity and switched to vecdb cumulative function for cleaner code

website

• Replaced uFuzzy search library with quickmatch for better performance
• Major cleanup and refactoring of frontend code

scripts

• Updated release scripts with improved workflow

View changes

v0.1.0-alpha.6 - 2026-01-18

Bug Fixes

brk_client

• Fixed minreq HTTP client feature flag: changed from serde_json to json-using-serde for proper JSON serialization support
• Regenerated client code with corrected HTTP handling

scripts

• Added full workspace compilation check to release script to catch feature flag errors before publishing

View changes

v0.1.0-alpha.5 - 2026-01-17

Bug Fixes

brk_query

• Fixed tokio feature flag to properly propagate to brk_error dependency, enabling async error handling when tokio feature is enabled

View changes

v0.1.0-alpha.4 - 2026-01-17

New Features

brk_computer

• Implemented Phase Oracle V2 algorithm using round USD template cross-correlation for on-chain price discovery (source)
  • Uses 200-bin phase histogram with weighted round USD template ($1, $5, $10, $20, etc.)
  • Weekly OHLC anchors constrain search range for accuracy
  • Cross-correlates satoshi values with expected USD fingerprint patterns
• Added cents_to_dollars transform for price conversions (source)

brk_query

• Added ResolvedQuery struct for efficient metric query resolution with ETag generation (source)
• Two-phase query processing: resolve metadata (cheap) then format output (expensive, cached)

brk_server

• Added bulk metrics API endpoint with ETag-based caching and 304 Not Modified support (source)
• Implemented CacheStrategy module with Height, Static, and MaxAge caching strategies (source)
• Added compact OpenAPI spec generator optimized for LLM consumption (source)
  • Removes redundant fields (error responses, descriptions, examples)
  • Simplifies schemas and flattens allOf/anyOf constructs
  • Compacts responses to "returns": "Type" format
• Added dedicated error handling module
• Added website file serving module with symlink to website directory

brk_types

• Added Etag type for HTTP ETag header handling (source)
• Added Output enum for JSON and CSV metric output formats (source)
• Added MetricOutput struct wrapping output with metadata for ETag generation (source)
• Added OracleBins type for 100-bin phase histogram storage used in price discovery (source)
• Added PairOutputIndex for indexing 2-output transactions (oracle pair candidates) (source)
• Added Port type with default server port 3110 (source)

brk_fetcher

• Added fetcher example demonstrating price data fetching

Internal Changes

brk_server

• Reorganized OpenAPI generation into dedicated openapi/ module
• Added Scalar API documentation HTML template

modules/brk-client

• Regenerated documentation with new types (OracleBins, PairOutputIndex, SyncStatus)
• Added README with usage examples

packages/brk_client

• Regenerated documentation and updated README

View changes

v0.1.0-alpha.3 - 2026-01-14

Breaking Changes

brk_bundler

• Removed brk_bundler crate from workspace (JavaScript bundling functionality removed)

New Features

brk_computer

• Implemented UTXOracle trustless on-chain price discovery algorithm that derives Bitcoin prices from transaction data without external price feeds (source)
  • Detects round USD amounts ($10, $20, $50, $100, etc.) in transaction outputs
  • Uses histogram-based stencil matching with Gaussian smoothing
  • Refines prices via geometric median convergence
  • Filters "clean" transactions: 2 outputs, ≤5 inputs, no OP_RETURN, simple signatures
• Added price/cents/ module for cent-based price storage with split components (source)
• Created dedicated index modules for time-based metric aggregation:
  • indexes/dateindex.rs - Daily index with date, first_height, height_count, weekindex, monthindex mappings
  • indexes/weekindex.rs - Weekly aggregation
  • indexes/monthindex.rs - Monthly aggregation
  • indexes/quarterindex.rs - Quarterly aggregation
  • indexes/semesterindex.rs - Semester (half-year) aggregation
  • indexes/yearindex.rs - Yearly aggregation
  • indexes/decadeindex.rs - Decade aggregation
  • indexes/halvingepoch.rs - Halving epoch indexing
  • indexes/difficultyepoch.rs - Difficulty adjustment epoch indexing
  • indexes/height.rs - Block height indexing
  • indexes/txindex.rs, txinindex.rs, txoutindex.rs - Transaction-level indexing
  • indexes/address.rs - Address type indexing with vecs for all output types (P2PK33, P2PK65, P2PKH, P2SH, P2TR, P2WPKH, P2WSH, P2A, P2MS, empty, unknown, OP_RETURN)
• Added comprehensive internal/multi/ module with 100+ metric aggregation strategies (source):
  • from_height/ - Height-based metric derivation with lazy, binary, and value variants
  • from_date/ - Date-based aggregation with percentiles, ratios, statistics
  • from_height_and_date/ - Combined height+date metrics including OHLC
  • from_tx/ - Transaction-level distributions
  • height_derived/ - Height-to-higher-index derivations
  • date_derived/ - Date-to-higher-index derivations (average, first, last, min, max, spread, sum)
  • tx_derived/ - Transaction-derived distributions
  • height_and_date/ - Combined aggregations for OHLC and bytes
• Added internal/single/ module with specialized metric computations (source):
  • transform/ - 35+ value transformations (price conversions, ratios, percentages, RSI, volatility)
  • lazy/ - Lazy evaluation for sum, average, cumulative, first, last, min, max, spread
  • lazy_transform/ - Lazy binary operations with percentiles and statistics
  • vec/ - Vector-based aggregations (percentiles, sum, average)
  • group/ - Group statistics (min/max, average, percentiles, distribution)
  • height/ - Height-indexed value derivations
  • tx/ - Transaction-level distributions
  • difficultyepoch/ - Difficulty epoch lazy values
• Added internal/compute.rs for unified computation orchestration
• Added internal/traits.rs for shared metric computation traits
• Added distribution/metrics/outputs.rs for output-specific distribution metrics

brk_client

• Added basic.rs example demonstrating typed metrics API with method chaining (source)
• Added tree.rs example for navigating the metrics tree structure

brk_fetcher

• Implemented retry mechanism for network requests with exponential backoff (source)
• Added permanent network error detection to skip retries for blocked endpoints

brk_indexer

• Added vecs/addresses.rs for address-specific vector storage
• Added vecs/inputs.rs for input vector organization
• Added vecs/outputs.rs for output vector organization
• Added vecs/transactions.rs for transaction vector organization
• Added vecs/scripts.rs for script vector organization

brk_types

• Added DiskUsage type for tracking storage consumption per component (source)
• Added FormatResponse type for API response formatting (source)
• Added SyncStatus type for indexer synchronization status (source)

brk_bindgen

• Added generate/constants.rs for constant generation
• Added generate/tree.rs for tree structure generation
• Added catalog test suite (source)

brk_server

• Added /api/server/sync endpoint returning SyncStatus with indexed_height, tip_height, blocks_behind, and last indexed timestamp (source)
• Added /api/server/disk endpoint returning DiskUsage with storage consumption per component
• Added /api/server/health endpoint for health checks

brk_store

• Added any.rs for type-erased store access

modules/brk-client

• Generated comprehensive TypeScript-style documentation with 300+ interface and type definitions
• Added test suites: basic.js and tree.js
• Added CLAUDE.md with development instructions

packages/brk_client

• Added Python documentation generation with pydoc-markdown
• Added DOCS.md with generated API documentation
• Added test suites: test_basic.py, test_tree.py

website

• Restructured from websites/bitview/ to unified website/ directory
• Added Progressive Web App (PWA) support with manifest and service worker (source)
• Added 40+ Apple splash screen images for iOS home screen support
• Added search pane (panes/search.js) and navigation pane (panes/nav.js)
• Added units utility (utils/units.js) for value formatting
• Reorganized options system with dedicated modules for chain, cohorts, cointime, market, and series configuration

scripts

• Added js-docs.sh for JavaScript documentation generation
• Added js-publish.sh for JavaScript package publishing
• Added python-docs.sh for Python documentation generation
• Added python-publish.sh for Python package publishing to PyPI
• Added rust-publish.sh for Rust crate publishing
• Added release.sh for coordinated multi-language releases

Internal Changes

workspace

• Added build.rs scripts to all crates for consistent build configuration
• Shortened percentiles path in metric tree for cleaner API

brk_computer

• Reorganized traits into dedicated traits/ module with pricing.rs
• Removed specialized percentiles, ratio, stddev modules from internal/specialized/

brk_logger

• Added example demonstrating logger usage

View changes

v0.1.0-alpha.2 - 2026-01-04

Breaking Changes

brk_binder → brk_bindgen

• Replaced brk_binder crate with completely rewritten brk_bindgen crate featuring new architecture with separate generators/, backends/, and analysis/ modules (source)

brk_grouper → brk_cohort

• Migrated cohort filtering functionality from brk_grouper to new brk_cohort crate with expanded filtering capabilities (source)

brk_computer

• Replaced brk_grouper dependency with brk_cohort throughout the crate
• Removed monolithic chain/ module containing compute.rs, import.rs, and mod.rs
• Removed monolithic cointime.rs in favor of modular cointime/ directory structure
• Removed stateful/ module tree including address cohorts, UTXO cohorts, and state management
• Removed fetched.rs, grouped/, indexes.rs, and price.rs modules

New Features

brk_alloc (New Crate)

• Created global allocator crate using mimalloc for consistent memory allocation across all brk crates (source)
• Added Mimalloc::collect() function for eager memory release back to OS at natural pause points

brk_bindgen (New Crate)

• Implemented ClientOutputPaths builder for configuring output file paths per language (source)
• Added generate_clients() function that generates all configured language clients from OpenAPI spec
• Created analysis module with name resolution, pattern detection, position tracking, and tree traversal (source)
• Implemented language-specific backends for JavaScript, Python, and Rust code generation (source)
• Created generators module with separate API, client, tree, and types generation per language (source)
• Added syntax utilities for code generation formatting
• Implemented schema extraction and type metadata handling (source)

brk_client (New Crate)

• Created Rust client crate with build-time CPU optimization for release builds (source)
• Generated client library using brk_types for strong typing

brk_cohort (New Crate)

• Implemented comprehensive UTXO and address cohort filtering system (source)
• Added address type filtering for all output types: P2PK65, P2PK33, P2PKH, P2SH, P2WPKH, P2WSH, P2TR, P2A (source)
• Added spendable and unspendable type filtering (source)
• Implemented CohortContext for automatic metric name prefix generation based on cohort type (source)
• Added CohortName system for generating consistent cohort identifiers
• Implemented StateLevel tracking for cohort state management
• Added year-based filtering with ByYear grouper
• Added ByAnyAddress for filtering across all address types
• Migrated and extended age range, amount range, epoch, term, min/max age, ge/lt amount filters from brk_grouper

brk_computer

• Created modular blocks/ module with separate submodules for count, difficulty, halving, interval, mining, rewards, size, time, and weight metrics (source)
• Created modular cointime/ module with submodules for activity, adjusted, cap, pricing, supply, and value calculations (source)
• Created distribution/ module for UTXO distribution metrics with address, block, cohorts, compute, metrics, and state submodules (source)
• Reorganized indexes/ into modular structure with address, block, time, and transaction submodules
• Reorganized inputs/ into count and spent submodules
• Reorganized outputs/ into count and spent submodules
• Reorganized market/ into ath, dca, indicators, lookback, moving_average, range, returns, and volatility submodules
• Reorganized price/ into ohlc, sats, and usd submodules
• Reorganized supply/ into burned, circulating, inflation, market_cap, and velocity submodules
• Reorganized transactions/ into count, fees, size, versions, and volume submodules
• Reorganized scripts/ into count and value submodules
• Added internal module with builder, computed, lazy, specialized, and value submodules for metric computation infrastructure

brk_indexer

• Refactored processor into modular structure with separate metadata, tx, txin, txout, and types submodules (source)

brk_bundler

• Added bundle example demonstrating bundler usage (source)

brk_types

• Added Age type for UTXO age representation
• Added CentsCompact type for compact cent storage
• Added Hex type for hexadecimal string handling
• Added Indexes module for index type collections
• Added OptionExt trait for option extensions
• Added Percentile type for percentile values
• Added SupplyState type for supply tracking state
• Added Term type for short-term/long-term holder classification

packages/brk_client (Python)

• Created Python client package with httpx for HTTP requests (source)
• Added typed dataclass-based client with full metric catalog support
• Included test suite for client functionality

Internal Changes

brk_computer

• Removed debug_indexer.rs and pools.rs examples
• Simplified constants.rs by removing unused constant definitions
• Consolidated address data structures into distribution module

brk_indexer

• Removed BENCH.md benchmarking documentation

brk_mempool

• Removed DESIGN.md design documentation

workspace

• Added brk_alloc as dev-dependency for benchmarks using mimalloc

View changes

v0.1.0-alpha.1 - 2025-12-21

New Features

brk_binder

• Implemented complete multi-language client generation system that produces typed API clients from the metric catalog and OpenAPI specification (source)
• Added JavaScript client generator with full JSDoc type annotations for IDE autocomplete support across 20k+ metrics (source)
• Added Python client generator with type hints and httpx for HTTP requests (source)
• Added Rust client generator with strong typing using reqwest (source)
• Implemented OpenAPI 3.1 integration using the oas3 crate for parsing endpoint definitions (source)
• Created structural pattern detection system that identifies repeating tree structures for type reuse (source)
• Added generic pattern detection that groups type-parameterized structures to reduce generated code size
• Created ClientMetadata for extracting catalog structure, patterns, and index sets from brk_query::Vecs (source)

brk_types

• Added MetricLeaf struct containing metric name, value type, and available indexes (source)
• Added MetricLeafWithSchema that wraps MetricLeaf with a JSON Schema for client type generation
• Changed TreeNode::Leaf from String to MetricLeafWithSchema to carry full metadata through the tree
• Added JsonSchema derive to byte array types (U8x2, U8x20, U8x32) with custom implementations for U8x33 and U8x65
• Added JsonSchema derive to OHLC types (OHLCCents, OHLCDollars, OHLCSats, Open, High, Low, Close)
• Added JsonSchema derive to all index types (date, week, month, quarter, semester, year, decade, halving epoch, difficulty epoch)
• Added JsonSchema derive to address index types (P2PKH, P2SH, P2WPKH, P2WSH, P2TR, P2PK33, P2PK65, P2A, OP_RETURN, empty, unknown)
• Implemented index merging when collapsing tree nodes with the same metric name

brk_traversable

• Added make_leaf helper function that generates MetricLeafWithSchema nodes with schemars-based JSON Schema generation (source)
• Updated all Traversable implementations to require JsonSchema bound and generate full leaf metadata

brk_computer

• Added neg_realized_loss metric (realized_loss * -1) for charting convenience (source)
• Added net_realized_pnl metric (realized_profit - realized_loss)
• Added realized_value metric (realized_profit + realized_loss)
• Added total_realized_pnl at both height and dateindex levels
• Added realized_price metric (realized_cap / supply)
• Added realized_cap_30d_delta for 30-day realized cap changes
• Added sopr (Spent Output Profit Ratio) with 7-day and 30-day EMA variants
• Added adjusted_sopr with EMA variants for coinbase-adjusted analysis
• Added sell_side_risk_ratio (realized_value / realized_cap) with 7-day and 30-day EMAs
• Added realized profit/loss ratios relative to realized cap
• Added net_realized_pnl_cumulative_30d_delta with ratios relative to realized cap and market cap
• Added neg_unrealized_loss_rel_to_market_cap and net_unrealized_pnl_rel_to_market_cap metrics (source)
• Added supply_in_profit/loss_rel_to_circulating_supply metrics
• Added unrealized profit/loss relative to own market cap at both height and indexes levels
• Added full benchmark example with indexer, computer, and bencher integration (source)

brk_grouper

• Added public constants for age boundaries (DAYS_1D, DAYS_1W, DAYS_1M, etc.) for external use (source)
• Added get_mut_by_days_old() method to ByAgeRange for O(1) direct bucket access by days
• Added AmountBucket type for O(1) amount range classification (source)
• Added amounts_in_different_buckets() helper for cheap bucket comparison
• Optimized get() and get_mut() in ByAmountRange using match on bucket index instead of if-else chain

brk_server

• Integrated automatic client generation on startup when brk_binder/clients/ directory exists (source)
• Added panic catching for client generation to prevent server startup failures

brk_cli

• Enabled distribution via cargo-dist (dist = true in package metadata)
• Added performance benchmarks section to README with timing and resource usage data

workspace

• Added rust-toolchain.toml pinning toolchain to Rust 1.92.0
• Created comprehensive scripts/publish.sh for automated crate publishing in dependency order (source)
• Updated scripts/update.sh to also update rust-toolchain version

Internal Changes

brk_computer

• Removed FenwickTree state module (O(log n) prefix sum data structure)
• Removed PriceBuckets state module (logarithmic price bucket distribution)
• Simplified cohort state management after removal of Fenwick-based percentile computation

brk_indexer

• Renamed push_if_needed() to checked_push() throughout for API consistency (source)
• Renamed Indexes::push_if_needed() to Indexes::checked_push()

brk_cli

• Updated zip dependency from 6.0.0 to 7.0.0

View changes

v0.1.0-alpha.0 - 2025-12-18

Breaking Changes

brk_structs → brk_types

• Renamed brk_structs crate to brk_types with complete reorganization of all Bitcoin data types into 150+ dedicated modules (source)

brk_parser → brk_reader

• Renamed brk_parser crate to brk_reader with simplified scope focused exclusively on BLK file reading and XOR decryption (source)
• Moved block iteration logic to new brk_iterator crate

brk_interface

• Removed brk_interface crate entirely, replacing it with the more comprehensive brk_query crate

brk_store

• Upgraded storage backend from fjall v2.11.2 to fjall v3.0.0-rc.6 with breaking API changes (source)

brk_indexer

• Replaced brk_parser::Parser with brk_iterator::Blocks for block iteration (source)
• Renamed OutputIndex and InputIndex to TxOutIndex and TxInIndex throughout the codebase for clarity
• Changed address indexing from addressindex->txoutindex to addressindex->txindex and addressindex->outpoint patterns for more flexible queries

brk_computer

• Replaced brk_parser::Parser with brk_reader::Reader for block file access
• Reorganized stateful computation architecture with new cohort-based grouping system
• Removed states module, integrating state management directly into stateful module

New Features

brk_mempool (New Crate)

• Created comprehensive mempool monitoring crate with real-time transaction tracking and fee estimation (source)
• Implemented CPFP-aware block builder algorithm using dependency graph analysis and heap-based transaction selection (source)
• Created Selector component that greedily selects highest fee-rate transactions while respecting ancestor dependencies
• Implemented Partitioner that groups selected transactions into projected blocks of ~1MB vsize each (8 blocks projected)
• Added Package abstraction for atomic transaction groups where children pay for parents (CPFP)
• Created Graph structure tracking transaction dependencies with ancestor fee/size aggregation
• Implemented RecommendedFees computation providing fee estimates for fastest, half-hour, hour, economy, and minimum confirmation targets (source)
• Created AddressTracker for monitoring mempool transactions affecting specific addresses with add/remove tracking (source)
• Implemented EntryPool with slot-based entry management using generation counters for efficient reuse
• Added Snapshot system capturing point-in-time mempool state with per-block statistics (tx count, vsize, fees, fee range) (source)
• Created TxStore for mempool transaction storage with prefix-based lookups using TxidPrefix
• Implemented MempoolInfo tracking aggregate statistics: tx count, total vsize, total fees
• Added automatic synchronization loop with 1-second update interval and rate-limited rebuilds (minimum 1 second between rebuilds) (source)
• Created dirty flag mechanism for efficient change detection and rebuild triggering

brk_query (New Crate)

• Created unified query interface combining indexer, computer, reader, and mempool access into single Query struct (source)
• Implemented address query methods (source):
  • address() - Returns AddressStats with chain stats (funded/spent txo counts and sums, tx count) and optional mempool stats
  • address_txids() - Returns paginated transaction IDs for an address with after_txid cursor support
  • address_utxos() - Returns all unspent outputs for an address with full status information
  • address_mempool_txids() - Returns up to 50 unconfirmed transaction IDs affecting an address
  • resolve_address() - Helper converting address string to output type and type_index
• Implemented block query methods (source):
  • block() / block_by_height() - Returns BlockInfo with hash, height, timestamp, difficulty, size, weight, tx count
  • blocks() - Returns last 10 blocks, optionally starting from a specific height
  • block_status() - Returns BlockStatus with in_best_chain flag, height, and next_best hash
  • block_timestamp() - Returns BlockTimestamp for timestamp-based block lookup
  • block_txids() - Returns all transaction IDs in a block
  • block_txs() - Returns paginated full transactions (25 per page)
  • block_txid_at_index() - Returns single txid at specific index within block
  • block_raw() - Returns raw block bytes read directly from blk file with XOR decoding
  • block_by_timestamp() - Finds block closest to given UNIX timestamp using binary search
• Implemented transaction query methods (source):
  • transaction() - Returns full Transaction with inputs (including prevout values), outputs, fee, status; checks mempool first for unconfirmed txs
  • transaction_by_index() - Internal method reading tx from blk file and reconstructing full transaction object
  • transaction_status() - Returns TxStatus with confirmed flag, block height/hash/time
  • transaction_hex() - Returns raw transaction as hex string
  • outspend() - Returns TxOutspend for single output with spending tx details if spent
  • outspends() - Returns spend status for all outputs in a transaction
  • outspend_details() - Helper computing spending txid, vin, and status from txinindex
• Implemented mempool query methods (source):
  • mempool_info() - Returns MempoolInfo with aggregate statistics
  • mempool_txids() - Returns all transaction IDs currently in mempool
  • recommended_fees() - Returns RecommendedFees with fee rate targets
  • mempool_blocks() - Returns Vec<MempoolBlock> with projected block statistics
• Implemented mining query methods (source):
  • mining_pools() - Returns PoolsSummary with pool statistics for time period (24h, 3d, 1w, 1m, 3m, 6m, 1y, 2y, 3y)
  • all_pools() - Returns Vec<PoolInfo> with all 166 known mining pools
  • pool_detail() - Returns PoolDetail with block counts/shares for all time, 24h, and 1 week
  • hashrate() - Returns HashrateSummary with network hashrate and difficulty data
  • difficulty_adjustment() - Returns current DifficultyAdjustment with epoch progress and retarget estimate
  • difficulty_adjustments() - Returns historical Vec<DifficultyAdjustmentEntry> (timestamp, height, difficulty, change%)
  • block_fees() - Returns Vec<BlockFeesEntry> with average fees per time period
  • block_rewards() - Returns Vec<BlockRewardsEntry> with average coinbase rewards
  • block_sizes_weights() - Returns BlockSizesWeights with average sizes and weights
  • reward_stats() - Returns RewardStats for last N blocks
• Implemented metrics query methods (source):
  • metric_count() - Returns count of available metrics
  • indexes() - Returns all available index types with accepted aliases
  • metrics() - Returns paginated list of metrics with offset/limit support
  • metrics_catalog() - Returns hierarchical TreeNode of all metrics organized by category
  • match_metric() - Fuzzy search for metrics by name with configurable result limit
  • metric_to_indexes() - Returns supported indexes for a specific metric
  • metric_data() - Returns MetricData with values for metric/index combination and date range
• Added async query support via optional tokio feature for non-blocking database access

brk_iterator (New Crate)

• Created Blocks factory for iterating over Bitcoin blocks with smart source selection (source)
• Implemented three source modes:
  • Smart - Auto-selects RPC for ranges ≤10 blocks, Reader for larger ranges
  • Rpc - Always uses Bitcoin Core RPC for block retrieval
  • Reader - Always uses BLK file reader for block retrieval
• Created flexible range specifications:
  • range(start, end) - Iterate over specific block range (inclusive)
  • start(height) - Iterate from height to chain tip
  • end(height) - Iterate from genesis to height
  • last(n) - Iterate over last n blocks
  • after(hash) - Iterate after a specific block hash (for incremental syncing)
• Implemented BlockIterator with chain reorganization detection that stops iteration on prev_hash discontinuity
• Added BlockRange enum for type-safe range specification
• Created State struct managing iteration progress with RPC or Reader backend

brk_rpc (New Crate)

• Created Bitcoin Core RPC client wrapper with automatic retry logic and configurable timeouts (source)
• Implemented comprehensive RPC methods:
  • get_blockchain_info() - Chain state and sync status
  • get_block() / get_block_count() / get_last_height() - Block retrieval and chain height
  • get_block_hash() / get_block_header() / get_block_info() / get_block_header_info() - Block metadata
  • get_transaction() / get_raw_transaction() / get_raw_transaction_hex() - Transaction retrieval
  • get_tx_out() - UTXO lookup with optional mempool inclusion
  • get_raw_mempool() - Mempool transaction IDs
  • get_raw_mempool_verbose() - Returns Vec<MempoolEntryInfo> with fee/ancestor data for all mempool entries
  • get_mempool_transaction() - Returns TxWithHex with full transaction data including resolved prevouts
  • is_in_main_chain() - Check if block is in best chain
  • get_closest_valid_height() - Find closest valid ancestor for reorg recovery
  • wait_for_synced_node() - Block until node headers == blocks
• Created ClientInner with retry mechanism (default 1M retries, 1s delay) (source)
• Added default path resolution for macOS (~/Library/Application Support/Bitcoin) and Linux (~/.bitcoin)

brk_types (New Crate)

• Created comprehensive type library with 150+ Bitcoin-related types organized into dedicated modules (source)
• Added mempool types: MempoolBlock (tx_count, total_vsize, total_fee, fee_range), MempoolEntryInfo, MempoolInfo, RecommendedFees (fastest_fee, half_hour_fee, hour_fee, economy_fee, minimum_fee)
• Created API parameter types with serde deserialization: AddressParam, AddressTxidsParam, BlockHashParam, BlockCountParam, HeightParam, TxidParam, TxidVout, LimitParam, TimePeriodParam, PoolSlugParam, TimestampParam, ValidateAddressParam
• Added mining types: Pool (id, name, slug, addresses, tags, link), PoolDetail, PoolDetailInfo, PoolInfo, PoolStats, PoolBlockCounts, PoolBlockShares, PoolsSummary, HashrateSummary, HashrateEntry, DifficultyAdjustment, DifficultyAdjustmentEntry, DifficultyEntry, DifficultyEpoch, RewardStats
• Implemented metric types: Metric, MetricCount, MetricData, MetricParam, Metrics, MetricSelection, MetricSelectionLegacy, MetricWithIndex, PaginatedMetrics, DataRange, DataRangeFormat
• Added block types: BlockInfo, BlockStatus, BlockTimestamp, BlockSizesWeights, BlockFeesEntry, BlockFeeRatesEntry, BlockRewardsEntry, BlockSizeEntry, BlockWeightEntry
• Created transaction types: Transaction, TxIn, TxOut, TxStatus, TxOutspend, TxWithHex, TxIndex, TxInIndex, TxOutIndex
• Added address types: Address, AddressBytes, AddressHash, AddressChainStats, AddressMempoolStats, AddressStats, AddressValidation, AnyAddressIndex, EmptyAddressIndex, LoadedAddressIndex, with separate types for each output type (P2PKH, P2SH, P2WPKH, P2WSH, P2TR, P2PK33, P2PK65, P2A, P2MS, OpReturn, Unknown)
• Implemented index types: Index enum (Height, DateIndex, WeekIndex, MonthIndex, QuarterIndex, SemesterIndex, YearIndex, DecadeIndex, TxIndex, P2PKHIndex, etc.), IndexInfo
• Added stored numeric types: StoredBool, StoredF32, StoredF64, StoredI16, StoredU8, StoredU16, StoredU32, StoredU64, StoredString
• Created utility types: Health, TreeNode, Unit, Pagination, PaginationIndex, Limit, OutPoint, Utxo, Vin, Vout

brk_grouper (New Crate)

• Created cohort filtering and grouping infrastructure for UTXO and address analysis (source)
• Implemented filter modules:
  • ByAddressType - Filter by output type (P2PKH, P2SH, P2WPKH, etc.)
  • ByAgeRange - Filter by UTXO age in blocks
  • ByAmountRange - Filter by satoshi value range
  • ByEpoch - Filter by halving epoch
  • ByTerm - Filter by holder term classification (STH/LTH)
  • ByYear - Filter by creation year
  • ByMinAge / ByMaxAge - Filter by minimum/maximum UTXO age
  • ByGeAmount / ByLtAmount - Filter by amount thresholds
  • BySpendableType / ByUnspendableType - Filter by spendability
  • ByAnyAddress - Filter by specific address membership
  • ByType / ByValue - Generic type and value filters
• Created AmountFilter and TimeFilter base traits for composable filtering
• Implemented CohortContext for managing filter state during cohort analysis
• Added Filtered wrapper for filtered iterators with lazy evaluation
• Created StateLevel enum for controlling computation granularity
• Implemented Term type for STH (Short Term Holder) / LTH (Long Term Holder) classification
• Added AddressCohort and UtxoCohort traits for cohort membership testing

brk_traversable and brk_traversable_derive (New Crates)

• Created Traversable trait for vector tree navigation (source):
  • to_tree_node() - Converts struct to hierarchical TreeNode representation
  • iter_any_exportable() - Returns iterator over all exportable vectors in the tree
• Implemented trait for all vecdb vector types: BytesVec, ZeroCopyVec, PcoVec, LZ4Vec, ZstdVec, EagerVec, LazyVecFrom1/2/3
• Added implementations for container types: Box<T>, Option<T>, BTreeMap<K, V>
• Created derive macro for automatic Traversable implementation on structs (source)
• Added TreeNode type with Leaf(String) and Branch(BTreeMap<String, TreeNode>) variants

brk_bencher and brk_bencher_visualizer (New Crates)

• Created benchmarking infrastructure with multi-metric monitoring (source):
  • DiskMonitor - Tracks directory size changes over time
  • MemoryMonitor - Tracks process RSS memory using system APIs
  • IoMonitor - Tracks read/write bytes using process I/O statistics
  • ProgressionMonitor - Tracks computation progress via log message pattern matching
• Implemented Bencher struct with:
  • Automatic benchmark directory creation under workspace_root/benches/{crate_name}/{timestamp}/
  • Background monitoring thread with 5-second sample interval
  • Logger hook registration for progress tracking
  • CSV output for all metrics with millisecond timestamps
• Created brk_bencher_visualizer for generating SVG charts from benchmark data (source):
  • Chart struct for SVG chart generation with configurable dimensions
  • Data module for CSV parsing and data transformation
  • Format module for human-readable byte/number formatting
  • Command-line tool for batch chart generation from benchmark directories

brk_reader (Renamed from brk_parser)

• Implemented thread-safe BLK file reader with parallel block decoding (source)
• Created multi-stage pipeline:
  • Stage 1: Sequential BLK file reading with magic byte detection and XOR decoding
  • Stage 2: Parallel block parsing using dedicated rayon thread pool (half of available threads)
  • Stage 3: Block ordering with future block buffering for out-of-order handling
• Added binary search optimization for finding starting BLK file index when parsing from specific heights
• Created read_raw_bytes() method for reading arbitrary byte ranges from blk files with XOR decoding
• Implemented chain continuity verification that detects prev_hash discontinuities and logs errors
• Added BlkIndexToBlkPath mapping with automatic BLK file discovery and caching
• Created XORBytes and XORIndex types for efficient XOR decryption state management

brk_server - REST API

• Added comprehensive OpenAPI documentation using aide with Scalar UI at /api endpoint (source)
• Created Address API endpoints (source):
  • GET /api/address/{address} - Address information with chain and mempool stats
  • GET /api/address/{address}/txs - Paginated transaction IDs with after_txid and limit params
  • GET /api/address/{address}/utxo - Unspent transaction outputs
  • GET /api/address/{address}/txs/mempool - Unconfirmed transactions (max 50)
  • GET /api/validate-address/{address} - Address validation with type detection
• Created Block API endpoints (source):
  • GET /api/blocks - Last 10 blocks
  • GET /api/blocks/{height} - 10 blocks backwards from height
  • GET /api/block/{hash} - Block information by hash
  • GET /api/block/{hash}/status - Block status (in_best_chain, next_best)
  • GET /api/block/{hash}/txids - All transaction IDs in block
  • GET /api/block/{hash}/txs/{start_index} - Paginated transactions (25 per page)
  • GET /api/block/{hash}/txid/{index} - Single txid at index
  • GET /api/block/{hash}/raw - Raw block bytes
  • GET /api/block-height/{height} - Block by height
  • GET /api/v1/mining/blocks/timestamp/{timestamp} - Block by timestamp
• Created Transaction API endpoints (source):
  • GET /api/tx/{txid} - Full transaction data with inputs/outputs
  • GET /api/tx/{txid}/status - Confirmation status
  • GET /api/tx/{txid}/hex - Raw transaction hex
  • GET /api/tx/{txid}/outspend/{vout} - Single output spend status
  • GET /api/tx/{txid}/outspends - All output spend statuses
• Created Mempool API endpoints (source):
  • GET /api/mempool/info - Mempool statistics (tx count, vsize, fees)
  • GET /api/mempool/txids - All mempool transaction IDs
  • GET /api/v1/fees/recommended - Recommended fee rates for confirmation targets
  • GET /api/v1/fees/mempool-blocks - Projected blocks with fee statistics
• Created Mining API endpoints (source):
  • GET /api/v1/difficulty-adjustment - Current difficulty adjustment info
  • GET /api/v1/mining/pools - List all known mining pools
  • GET /api/v1/mining/pools/{time_period} - Pool statistics for time period
  • GET /api/v1/mining/pool/{slug} - Detailed pool information
  • GET /api/v1/mining/hashrate / GET /api/v1/mining/hashrate/{time_period} - Network hashrate
  • GET /api/v1/mining/difficulty-adjustments / GET /api/v1/mining/difficulty-adjustments/{time_period} - Historical adjustments
  • GET /api/v1/mining/blocks/fees/{time_period} - Block fee statistics
  • GET /api/v1/mining/blocks/rewards/{time_period} - Block reward statistics
  • GET /api/v1/mining/blocks/sizes-weights/{time_period} - Block size/weight statistics
  • GET /api/v1/mining/reward-stats/{block_count} - Reward stats for last N blocks
• Created Metrics API endpoints (source):
  • GET /api/metrics/count - Total metric count
  • GET /api/metrics/indexes - Available indexes with aliases
  • GET /api/metrics/list - Paginated metric list
  • GET /api/metrics/catalog - Hierarchical metric tree
  • GET /api/metrics/search/{metric} - Fuzzy metric search
  • GET /api/metric/{metric} - Supported indexes for metric
  • GET /api/metric/{metric}/{index} - Metric data with date range filtering
  • GET /api/metrics/bulk - Bulk metric query (max 650KB response weight)
  • GET /api/vecs/{variant} - Legacy variant endpoint (deprecated, sunset 2027-01-01)
  • GET /api/vecs/query - Legacy query endpoint (deprecated)
• Created Server API endpoints:
  • GET /version - API version
  • GET /health - Health check with timestamp
  • GET /api.json - OpenAPI specification
  • GET /api - Scalar API documentation UI
• Implemented CacheStrategy enum for response caching: Static (immutable), Height (ETag based on block height), MaxAge(seconds) (Cache-Control max-age)
• Added response transformations in extended module: HeaderMapExtended, ResponseExtended, TransformResponseExtended

brk_computer

• Achieved ~3x computation speed improvement through optimized vecdb iterators and major stateful processing optimizations
• Added Traversable derive macro support for automatic vector tree generation
• Implemented parallel import for indexes, fetched data, and blks vectors using 512MB stack-size threads (source)
• Created new blks module with txindex_to_position and txindex_to_len vectors for transaction location tracking

brk_indexer

• Achieved ~4x indexing speed improvement (~12h → ~3h for full chain) through parallel TXID computation, parallel input/output processing, efficient same-block spend detection, fjall v3, and mimalloc
• Implemented parallel import for vecs and stores using scoped threads (source)
• Added Readers struct providing typed vector iterators for efficient batch data access
• Created Processor for block processing with address and transaction indexing
• Added automatic file handle limit increase at startup (minimum 10,000 via rlimit)

brk_binder

• Added Rust binding generator for type-safe metric access alongside existing JavaScript generator (source)

Internal Changes

brk_indexer

• Split indexer code into dedicated modules: constants, indexes, processor, readers, stores, vecs
• Moved address vector handling into nested vecs/address.rs module
• Separated block, transaction, txin, and txout vectors into dedicated submodules

brk_computer

• Major refactoring of stateful computation architecture with new module structure
• Split cohorts into address.rs, address_cohorts.rs, traits.rs, utxo.rs, and utxo_cohorts/ (mod, receive, send, tick_tock)
• Created compute module with aggregates, block_loop, context, readers, recover, and write submodules
• Reorganized metrics into activity, config, price_paid, realized, relative, supply, and unrealized submodules
• Added process module with address_updates, cache, inputs, lookup, outputs, range_map, received, sent, tx_counts, with_source
• Created states module with address_cohort, block, cohort, fenwick, price_buckets, price_to_amount, realized, supply, transacted, unrealized, utxo_cohort
• Added address module in stateful with address_count, any_address_indexes, data, height_type_vec, type_index_map, type_vec
• Created addresstype module with addresscount, height_to_addresscount, height_to_vec, indexes_to_addresscount, typeindex_tree, vec

brk_fetcher

• Restructured fetcher into dedicated source modules: binance.rs, brk.rs, kraken.rs
• Added retry.rs module for fetch retry logic with exponential backoff
• Created ohlc.rs module for OHLC data aggregation and transformation
• Added source.rs module for abstracting data source selection

brk_store

• Simplified store implementation to use brk_types instead of brk_structs
• Added rustc-hash for faster FxHashMap operations

brk_logger

• Added hook registration system via register_hook() for external log message monitoring (source)

Documentation

Benchmarks

• Added benchmark data and visualizations for brk_computer and brk_indexer across multiple hardware configurations
• Created benchmark directories with disk, I/O, memory, and progress CSV/SVG files:
  • benches/brk_computer/ - Computer benchmark visualizations
  • benches/brk_indexer/mac_mini_m4_16gb_ext_ssd/ - Mac Mini M4 with external SSD
  • benches/brk_indexer/mbp_m3_pro_36gb_int_ssd/ - MacBook Pro M3 Pro with internal SSD

View changes

v0.0.111 - 2025-10-03

Breaking Changes

brk_interface

• Renamed metric_count() method to distinct_metric_count() and vec_count() to total_metric_count() for clearer API semantics distinguishing between unique metric names versus total metric-index combinations (source)
• Consolidated indexes data structure from separate indexes vector and accepted_indexes map into single indexes BTreeMap containing index names as keys and their accepted variants as values
• Removed redundant index_count() method in favor of accessing indexes.len() directly

brk_mcp

• Removed get_index_count() tool from Model Context Protocol interface as index count can be derived from indexes map length (source)
• Removed get_accepted_indexes() tool, consolidating functionality into enhanced get_indexes() tool that returns complete mapping
• Updated get_metric_count() description to clarify it returns count of unique metrics
• Updated get_vec_count() description to explain it returns total metrics (distinct metrics multiplied by supported indexes)

New Features

Website (bitview)

• Enhanced "Coins Destroyed" chart organization by separating "Sum" and "Cumulative" views into distinct sub-charts when comparing multiple cohorts, improving visual clarity and reducing chart clutter (source)
• Maintained backward compatibility with single-cohort view showing both sum and cumulative series in one chart

Internal Changes

brk_fetcher

• Migrated JSON parsing from sonic_rs::from_str() to standard serde_json with updated response handling using .as_str() instead of .json() for improved compatibility and maintainability (source)

brk_interface

• Switched from sonic_rs to serde_json for JSON value handling in metrics deserialization for consistent JSON processing across the codebase
• Streamlined internal data structures eliminating duplicate index tracking mechanisms

brk_server

• Updated metrics API endpoint /api/metrics/indexes to use consolidated get_indexes() method returning complete index-to-variants mapping (source)

Workspace

• Updated all crate versions from 0.0.110 to 0.0.111 across the workspace

View changes

v0.0.110 - 2025-10-02

New Features

brk_binder

• Created new crate for generating language binding files to facilitate integration with other programming languages (source)
• Implemented JavaScript binding generator with compressed metric-to-index mappings using base62 encoding for optimized frontend bundle sizes (source)
• Added automatic generation of TypeScript-compatible pool ID mappings with sorted pool names for improved developer experience
• Implemented word frequency analysis to create compressed metric names, reducing JavaScript bundle size and improving load times
• Added version file generation to track binding file compatibility with backend versions

brk_server

• Implemented comprehensive address lookup API endpoint supporting all Bitcoin address types including P2PK, P2PKH, P2SH, P2WPKH, P2WSH, P2TR, and P2A (source)
• Added transaction lookup endpoint with raw transaction decoding from blk files including position-based seeking and XOR decryption
• Created address balance and statistics API providing UTXO counts, sent/received amounts, realized value, and average cost basis
• Implemented real-time USD balance calculation using latest price data for address endpoints
• Reorganized API structure into dedicated metrics/ and explorer/ modules for better code organization and maintainability (source)
• Added metrics count endpoint exposing both distinct and total metric counts for API discovery
• Created bulk metrics query endpoint for efficient batch data retrieval

brk_interface

• Implemented MaybeMetrics type for handling multiple metric queries with automatic sanitization and validation (source)
• Added support for both comma-separated string and array formats for metric parameters with configurable size limits (max 32 metrics, 2KB total)
• Implemented automatic metric name normalization converting hyphens to underscores and enforcing lowercase for consistent API usage
• Added metric-to-indexes mapping functionality for discovering available data dimensions per metric

brk_bundler

• Restructured bundler to use centralized modules/ directory instead of website-specific packages for better code reuse (source)
• Implemented automatic module copying from modules/ to website source directories before bundling
• Added inline constant optimization configuration for improved runtime performance through compile-time constant evaluation
• Enhanced HTML generation with dynamic script path injection based on bundled entry point hashes

Frontend Modules

• Created brk-client module providing type-safe BRK API client with metrics catalog, pool mappings, and idle callback utilities
• Implemented brk-resources module for managing shared resources and dependencies across frontend applications
• Added brk-signals module for reactive state management with SolidJS signals compatibility
• Centralized third-party dependencies including lightweight-charts 5.0.9, lean-qr 2.6.0, ufuzzy 1.0.19, and solidjs-signals 0.6.3

Internal Changes

brk_cli

• Removed legacy bridge generation code in favor of new brk_binder crate for cleaner separation of concerns (source)
• Integrated brk_binder for automated JavaScript binding file generation during CLI operations

brk_computer

• Removed unused profit/loss metrics from stateful computations to streamline calculation pipelines
• Cleaned up cohort analysis code removing obsolete breakeven metrics for better performance

brk_interface

• Removed legacy MaybeIds type in favor of MaybeMetrics for more accurate API parameter naming
• Enhanced deserialization logic with better error messages for invalid metric formats
• Restructured interface to expose metric catalog and index mappings for API documentation

brk_mcp

• Updated Model Context Protocol integration to use new metrics infrastructure with improved parameter handling

Workspace

• Reorganized project structure moving website packages to centralized modules/ directory for better maintainability
• Removed historical asset files reducing repository size and focusing on active development
• Updated solidjs-signals from 0.4.1 to 0.6.3 with enhanced TypeScript definitions and reactive primitives

View changes

v0.0.109 - 2025-09-20

New Features

brk_parser

• Implemented new typed block parsing architecture with AnyBlock enum supporting Raw, Decoded, and Skipped states (source)
• Added ParsedBlock wrapper that combines block data with position metadata and transaction metadata vectors
• Introduced binary search optimization for finding starting block indices when parsing from specific heights
• Added concurrent parsing pipeline with parallel block decoding using rayon for improved performance

brk_structs

• Added comprehensive Block and ParsedBlock data structures with proper height, hash, and metadata tracking (source)
• Implemented BlkPosition and BlkMetadata types for precise block and transaction positioning within blk files
• Added coinbase_tag() method to Block for extracting coinbase transaction script signatures
• Migrated Pool, PoolId, and Pools structs from brk_computer for better code organization and reusability

brk_computer

• Created dedicated blks.rs module for tracking block positions and lengths with compressed vector storage (source)
• Added height_to_position, height_to_len, txindex_to_position, and txindex_to_len compressed vectors for efficient block data access
• Implemented forced import functionality with automatic region retention for optimized storage management

brk_server

• Added scaffolding for new explorer API endpoints in dedicated explorer.rs module (implementation commented for future development)
• Renamed interface.rs to vecs.rs and restructured API module organization for better separation of concerns

Internal Changes

brk_parser

• Removed complex BlkIndexToBlkRecap state tracking system and associated recap file management to simplify parsing logic
• Eliminated BlkMetadata, BlkRecap, BlockState, and error handling modules that were causing parsing complexity
• Streamlined XOR decryption handling with direct byte manipulation instead of state machine patterns
• Replaced tuple-based return types with structured ParsedBlock objects for better type safety

brk_interface

• Removed table formatting functionality and simplified output handling for cleaner API design
• Updated format and output modules to work with new parsing architecture

Workspace

• Updated all crate versions from 0.0.107 to 0.0.109 across the workspace
• Optimized development profile settings: reduced opt-level to 2, increased codegen-units to 16, and added split-debuginfo for faster builds
• Updated dependencies: serde (1.0.219 <20> 1.0.225), serde_json (1.0.143 <20> 1.0.145), vecdb (0.2.14 <20> 0.2.16)
• Added sonic-rs dependency for high-performance JSON processing
• Removed allocative_derive dependency that was no longer needed

Documentation

All Crates

• Comprehensive rewrite of all crate README files with detailed API documentation, usage examples, and installation instructions
• Standardized README format across workspace with consistent structure: overview, features, installation, usage, API reference
• Added professional crates.io badges and proper technical documentation for each crate's purpose and capabilities
• Enhanced code examples with practical use cases and proper error handling patterns

Project Structure

• Reorganized documentation into dedicated docs/ directory with CHANGELOG.md, README.md, TODO.md, and LICENSE.md
• Added comprehensive project-level documentation explaining the Bitcoin Research Kit architecture and component relationships
• Moved and updated scripts to scripts/ directory for better project organization

View changes

v0.0.107 - 2025-09-13

New Features

brk_computer

• Added comprehensive hash rate analytics with THS (TeraHash/Second) and PHS (PetaHash/Second) pricing models for mining profitability analysis (source)
• Implemented hash price and hash value calculations including minimum thresholds and rebound metrics for market analysis
• Added difficulty adjustment tracking with block and day countdown to next adjustment for network monitoring
• Implemented halving countdown functionality with blocks and days remaining until next halving event
• Added 24-hour rolling coinbase sum tracking in both BTC and USD for mining reward analysis
• Created comprehensive SOPR (Spent Output Profit Ratio) analytics with multi-timeframe standard deviation analysis
• Implemented inflation rate calculations and annualized volume metrics in BTC and USD
• Added transaction velocity, per-second transaction metrics, and input/output rate tracking for network activity analysis

brk_structs

• Enhanced OHLC (Open, High, Low, Close) data structures with StoredF64 conversion capabilities for improved financial data handling (source)
• Added CheckedSub implementation for Bitcoin struct to enable safe arithmetic operations between Bitcoin amounts
• Enhanced StoredF32 and StoredF64 types with additional mathematical utility methods

Internal Changes

brk_computer

• Removed complex ComputedVecBuilder system and streamlined computation architecture for better maintainability
• Refactored lazy and eager vector builders to eliminate redundant computed builder patterns
• Reorganized stateful computation modules with improved address and UTXO cohort management
• Simplified chain computation logic by consolidating version handling and removing unused computation paths
• Enhanced market data processing with better price integration and SOPR calculation efficiency

Workspace

• Added new clippy development profile with optimized settings for faster linting during development

Website Enhancements

Frontend

• Improved ratio-based chart options with cleaner naming conventions (removed "_in_usd" suffix for better readability)
• Enhanced SOPR visualization options with multi-timeframe standard deviation displays
• Added new data series configurations for hash rate analytics and mining profitability charts
• Updated chart legends and color schemes for better user experience and data clarity

View changes

v0.0.106 - 2025-09-09

New Features

brk_computer

• Added mathematical constants constant_38_2 and constant_61_8 for precise financial ratio calculations (source)
• Standardized percentile naming from p10, p25, p75, p90 to pct10, pct25, pct75, pct90 for improved code clarity (source)

brk_structs

• Enhanced Dollars struct with comprehensive OHLC (Open, High, Low, Close) conversion support for better financial data integration (source)
• Added proper type conversions between OHLC price components and dollar amounts

Internal Changes

Workspace

• Added explicit Rust toolchain specification pinned to version 1.89.0 for consistent compilation across development environments
• Improved statistical calculation accuracy with standardized percentile naming conventions

Website Enhancements

Frontend

• Enhanced chart visualization capabilities with improved data presentation
• Updated statistical displays to use standardized percentile naming for better user understanding

View changes

v0.0.105 - 2025-09-08

New Features

Distribution System

• Added cargo-dist configuration for automated cross-platform binary distribution with GitHub CI integration
• Enabled pre-built binary releases for aarch64-apple-darwin (macOS ARM64), aarch64-unknown-linux-gnu (Linux ARM64), and x86_64-unknown-linux-gnu (Linux x86_64)
• Configured automated release workflow with Rust toolchain version 1.89 for consistent builds

Internal Changes

Workspace

• Integrated cargo-dist v0.30.0 for streamlined release management and binary distribution
• Added distribution metadata configuration to simplify installation process for end users

View changes

v0.0.104 - 2025-09-08

Internal Changes

CI/CD Infrastructure

• Updated GitHub Actions release workflow to use cargo-dist v0.30.0 for improved release automation (source)
• Added foundational cargo-dist metadata configuration to prepare for automated binary distribution
• Enhanced release pipeline reliability with updated tooling and workflow improvements

View changes

v0.0.103 - 2025-09-08

New Features

brk_computer

• Added comprehensive price volatility analytics with 1-week, 1-month, and 1-year standard deviation calculations for returns analysis (source)
• Implemented advanced market risk metrics including rolling volatility indicators for better risk assessment
• Enhanced standard deviation calculations with optional z-score support and configurable statistical analysis (source)
• Refactored statistical computation modules to make calculations optional for improved performance and flexibility

brk_bundler

• Updated bundler configuration for optimized JavaScript asset compilation and improved frontend performance

Internal Changes

brk_computer

• Converted fixed standard deviation bands to optional calculations, reducing computational overhead when specific metrics aren't needed
• Streamlined market analysis architecture with better separation between volatility calculations and price analysis
• Enhanced statistical accuracy in ratio and standard deviation computations

Website Enhancements

Frontend

• Added support for new volatility and risk metrics in chart configurations
• Enhanced data visualization capabilities for displaying standard deviation bands and volatility indicators
• Improved chart options to accommodate advanced market analysis features

View changes

v0.0.101 - 2025-09-07

New Features

brk_structs

• Added Sum trait implementation for StoredF64 to enable collection aggregation and improved mathematical operations (source)
• Implemented division by Dollars for StoredF64 to support ratio calculations in financial analysis

brk_computer

• Upgraded SOPR (Spent Output Profit Ratio) calculations from StoredF32 to StoredF64 for enhanced precision in profit/loss analysis (source)
• Enhanced SOPR-related metrics including 7-day and 30-day exponential moving averages with double precision
• Improved adjusted SOPR calculations for more accurate short-term holder analysis

Internal Changes

brk_computer

• Removed unused imports and streamlined module dependencies for improved compilation efficiency
• Updated SOPR computation version handling to ensure proper recalculation with enhanced precision

brk_interface

• Simplified interface ID handling and reduced code complexity for better maintainability

Website Enhancements

Frontend

• Improved chart titles and interface labels for better user clarity, particularly for mining dominance displays
• Enhanced user experience with more descriptive chart naming conventions

View changes

v0.0.100 - 2025-09-07

New Features

brk_computer

• Created comprehensive mining pool analytics system with dedicated pools tracking module (source)
• Implemented mining pool dominance analysis with 1-day, 1-week, 1-month, and 1-year timeframes
• Added mining pool subsidy, fee, and coinbase reward tracking for economic analysis
• Created days-since-block mining for pool activity monitoring and mining frequency analysis

brk_structs

• Enhanced mathematical operations across all stored numeric types with comprehensive arithmetic support (source)
• Added StoredU32 to StoredF32 conversions and cross-type division operations for better data interoperability
• Implemented utility methods including is_zero() checks and mathematical constants for improved API consistency
• Added allocative trait derivations across all data structures for detailed memory usage tracking and performance optimization

brk_cli

• Expanded bridge system with automated pools data export for website integration (source)
• Enhanced website generation workflow with comprehensive data bridge file creation
• Improved CLI configuration and data export capabilities for better development workflow

Internal Changes

brk_computer

• Restructured pools module architecture with dedicated vector storage and computation systems
• Enhanced pool identification and tracking systems with improved data organization
• Streamlined computation workflows with better separation of mining pool analytics

brk_indexer

• Simplified indexer architecture and reduced code complexity for better maintainability
• Enhanced data processing efficiency with streamlined storage operations

Multiple Crates

• Added comprehensive allocative memory tracking across all major data structures for performance monitoring
• Standardized arithmetic operations and type conversions across the entire numeric type system
• Enhanced error handling and validation in mathematical operations

Website Enhancements

Frontend

• Major improvements to mining pool visualization with enhanced chart displays and dominance tracking
• Expanded table functionality with better data presentation and user interaction
• Enhanced chart options with comprehensive mining pool analytics integration
• Improved performance and user experience with updated data visualization libraries

View changes

v0.0.98 - 2025-09-05

New Features

brk_computer

• Created foundational mining pools system with comprehensive database of 166 predefined mining pools including identification by coinbase tags and addresses (source)
• Implemented target block count analytics for all time intervals including daily, weekly, monthly, quarterly, semester, yearly, and decade targets (source)
• Added mining pool identification system using coinbase transaction analysis and known pool addresses
• Created pools example demonstrating mining pool analysis capabilities

brk_interface

• Implemented robust MaybeIds system for handling multiple ID parameter formats with comprehensive validation and sanitization (source)
• Enhanced API parameter handling with support for both string and array formats
• Added input validation with configurable limits for string size and array length

Internal Changes

brk_computer

• Enhanced stateful computation organization with improved efficiency for complex analytics
• Restructured chain computation constants with better mathematical precision for target calculations
• Streamlined pools module architecture for better data organization and access patterns

brk_store

• Optimized store operations with reduced complexity and improved data access efficiency
• Removed unused functionality and streamlined storage interfaces

brk_fetcher

• Minor improvements to BRK API integration for better data fetching reliability

Website Enhancements

Frontend

• Added foundational support for mining pool visualization in chart configurations
• Enhanced website structure with mining pool data integration capabilities
• Improved chart options to accommodate future mining pool analytics displays

View changes

v0.0.96 - 2025-09-03

New Features

brk_structs

• Created comprehensive StoredString data type with ByteView integration for efficient string storage and serialization (source)
• Enhanced FeeRate calculations with improved mathematical operations and precision handling
• Added utility methods to Date and Timestamp structs for better temporal data manipulation
• Expanded StoredF64 with additional mathematical operations and conversion capabilities

brk_computer

• Major module restructuring: consolidated blocks and mining functionality into unified chain module for better Bitcoin blockchain analysis organization (source)
• Renamed transactions.rs to chain.rs to better reflect comprehensive blockchain transaction and block analysis
• Enhanced market analysis with improved price calculations and volatility metrics
• Streamlined stateful computations with better organization of address and UTXO cohort analysis

brk_interface

• Enhanced API interfaces with improved data access patterns and better error handling
• Added support for new data types and expanded parameter handling capabilities
• Improved index management with more robust validation and processing

Internal Changes

brk_computer

• Consolidated blockchain analysis modules for better code organization and reduced complexity
• Enhanced computation efficiency across market, price, and stateful analysis modules
• Improved memory management and data flow in complex analytical operations

brk_indexer

• Streamlined indexing operations with better data processing efficiency
• Enhanced vector storage and retrieval operations for improved performance
• Improved indexer examples with clearer demonstration of functionality

brk_parser

• Enhanced block parsing capabilities with better error handling and validation
• Improved parser examples demonstrating advanced blockchain data extraction techniques

brk_server

• Enhanced API module organization with improved interface handling
• Better integration with updated interface and data access patterns

Website Enhancements

Typography and Design

• Updated to modern GeistMono and Lilex font families for improved readability and professional appearance
• Enhanced asset management with optimized font loading and better file organization
• Improved service worker functionality for better offline capabilities

Chart and Visualization

• Major improvements to chart rendering capabilities with enhanced interactivity and data visualization
• Expanded chart options with new data series types and better configuration flexibility
• Enhanced table functionality with improved data presentation and user interaction
• Added support for new data types including hash rate (H/s) and address data visualization

Development Environment

• Added Zed editor configuration for improved development workflow
• Enhanced development tooling and editor integration

Documentation

Multiple Crates

• Comprehensive README updates across brk, brk_indexer, brk_mcp, brk_parser, and brk_server crates
• Improved documentation clarity with better examples and usage instructions
• Enhanced project-level documentation with updated architecture descriptions

View changes

v0.0.95 - 2025-08-28

New Features

brk_computer

• Added comprehensive unrealized profit and loss analytics with advanced relative metrics calculations (source)
• Implemented unrealized profit plus loss aggregation metrics for comprehensive holder analysis
• Created relative profitability metrics including ratios to own market cap and own unrealized profit plus loss
• Added multi-dimensional profit/loss analysis with separate positive and negative unrealized loss tracking
• Enhanced holder profitability analysis with extensive cross-referencing between different calculation methods

Website Enhancements

Chart and Visualization

• Added new unit types for advanced profit/loss visualization: %cmcap (relative to own market cap) and %cp+l (relative to own unrealized profit plus loss)
• Enhanced price line functionality with configurable colors and line styles for improved chart customization
• Improved chart options with better visual distinction between different profitability metrics
• Added support for displaying complex relative profitability calculations with proper unit labeling

User Interface

• Enhanced chart configuration with more flexible price line customization options
• Improved data visualization clarity for sophisticated profit/loss analytics
• Better integration of advanced metrics into existing chart framework

View changes

v0.0.94 - 2025-08-28

New Features

brk_computer

• Added comprehensive SOPR (Spent Output Profit Ratio) analytics with 7-day and 30-day exponential moving averages (source)
• Implemented adjusted SOPR calculations with exponential moving averages for enhanced transaction profitability analysis
• Created sell-side risk ratio metrics with 7-day and 30-day exponential moving averages for advanced market risk assessment
• Enhanced unrealized profit/loss analytics with relative calculations to market cap for comprehensive holder analysis
• Added extended configuration options for stateful computations with improved flexibility and performance

Website Transformation

Major Rebranding

• Complete rebrand from "default" to "bitview" with new domain reference (bitview.space) and professional identity
• Updated README and documentation to reflect new branding and simplified domain structure
• Reorganized website assets and structure for better organization and scalability

Progressive Web App (PWA) Implementation

• Added comprehensive PWA support with complete Apple touch icon and splash screen sets for all device sizes
• Implemented web app manifest with proper configuration for native app-like installation experience
• Created device-specific splash screens for optimal mobile user experience across iPhone and iPad variants
• Added proper PWA metadata and configuration for improved mobile usability

Asset and Structure Organization

• Moved website from websites/default to websites/bitview for better project organization
• Added modern font assets (GeistMono and Lilex) for improved typography and readability
• Reorganized packages and dependencies with better structure and maintainability
• Simplified TypeScript configuration and reduced development complexity

Internal Changes

brk_computer

• Enhanced UTXO cohort analysis with improved parameter handling and configuration flexibility
• Streamlined stateful computation architecture with better organization of advanced metrics
• Improved computational efficiency for complex profit/loss and risk calculations

Development Workflow

• Simplified frontend development approach with reduced TypeScript complexity
• Improved asset management and build processes for better development experience
• Enhanced website configuration and deployment processes

View changes

v0.0.93 - 2025-08-26

New Features

Website Screenshot Capabilities

• Added comprehensive screenshot functionality using modern-screenshot package for high-quality chart and data visualization capture (source)
• Implemented 2x scale rendering for crisp, professional-quality screenshot output
• Added automatic screenshot opening in new tabs with proper URL cleanup for optimal user experience

Chart Enhancement and Branding

• Added domain watermarking to charts for proper attribution and source identification (source)
• Enhanced chart presentation with integrated branding for professional appearance
• Improved chart layout and visual organization for better data presentation

Website Enhancements

User Interface Improvements

• Enhanced website styling with dedicated screenshot mode support for clean capture
• Improved visual presentation with better element spacing and layout optimization
• Added screenshot button integration with conditional display logic for better user experience

Package and Dependency Updates

• Updated ufuzzy dependency from v1.0.18 to v1.0.19 for improved search functionality
• Added modern-screenshot v4.6.6 package for advanced screenshot capabilities
• Enhanced package management with better organization and dependency handling

Internal Changes

Development Workflow

• Streamlined TODO list by removing completed tasks and outdated development items
• Simplified project roadmap focus with cleaner task organization
• Improved development priorities and task tracking

Website Structure

• Enhanced HTML structure with better CSS organization for screenshot functionality
• Improved element styling and layout for optimal visual presentation
• Better integration of new features with existing website architecture

View changes

v0.0.91 - 2025-08-26

New Features

brk_computer

• Added comprehensive Exponential Moving Average (EMA) calculations for all major timeframes matching SMA counterparts (source)
• Implemented EMA support for 1w, 8d, 13d, 21d, 1m, 34d, 55d, 89d, 144d, 200d, 1y, 2y, 200w, and 4y periods
• Enhanced technical analysis capabilities with both Simple Moving Averages (SMA) and Exponential Moving Averages (EMA) for sophisticated trend analysis
• Added mathematical constants 144 and 600 for Bitcoin-specific calculations (blocks per day and standard analysis periods) (source)
• Simplified constants computation with loop-based implementation for better maintainability and performance

brk_structs

• Enhanced OHLC (Open, High, Low, Close) data structures with improved mathematical operations for better financial analysis
• Improved Sats and StoredF32 types with additional utility methods for enhanced precision in calculations

Website Enhancements

Typography and Design

• Added Lilex font family for improved code readability and professional data presentation
• Enhanced typography with dedicated monospace font for better numeric data display and code visibility

Chart Configuration and Customization

• Enhanced price line creation with customizable naming capabilities for better chart personalization
• Improved average series configuration with optional title customization for clearer data presentation
• Better chart options organization with more intuitive naming conventions (e.g., "100K+ btc" instead of "100K btc+")
• Enhanced chart configuration flexibility with improved parameter handling

Internal Changes

brk_computer

• Streamlined constants computation architecture with more efficient loop-based calculations
• Enhanced market analysis module with comprehensive EMA implementation alongside existing SMA calculations
• Improved ratio computations with better organization and performance optimizations

brk_indexer

• Enhanced indexer functionality with improved data processing capabilities
• Better integration with new constants and mathematical operations

Development Workflow

• Simplified TODO management with better task organization and priority tracking
• Enhanced development workflow with improved mathematical operations and constants handling

View changes

v0.0.90 - 2025-08-24

Documentation

Release Management

• Converted "Unreleased" changelog section to properly formatted v0.0.89 release with comprehensive source code links
• Enhanced changelog structure with professional formatting and detailed feature documentation
• Added direct links to source code for all major features and architectural changes
• Improved documentation organization with better categorization and technical details

View changes

v0.0.89 - 2025-08-24

Documentation

Comprehensive README Overhaul

• Enhanced comprehensive rewrite of all crate README files for improved clarity and developer experience across all workspace crates (source)
• Updated main project README with better structure and comprehensive architecture documentation (source)
• Standardized README format across all workspace crates with consistent styling and improved developer experience
• Added more detailed descriptions for each crate's purpose and functionality with examples

New Features

brk_computer - Major Refactoring

• Refactored ComputedFrom pattern to LazyFrom pattern for improved performance and memory efficiency (source)
• Added new LazyVecBuilder implementation for on-demand computation with support for first, average, sum, max, min, last, and cumulative operations
• Restructured Computer module stateful operations with improved organization and rollback functionality (source)
• Enhanced address type organization into dedicated module structure (source)
• Improved stateful rollback functionality for better error recovery and blockchain reorganization handling

brk_structs

• Added new StoredI16 data type for efficient 16-bit signed integer storage with compression support (source)
• Enhanced StoredF32 with additional utility methods for mathematical operations (source)

Website Frontend

Package Management Overhaul

• Updated solid-signals from v0.3.2 to solidjs-signals v0.4.1 for improved reactivity and performance (source)
• Enhanced frontend package management and dependency organization
• Improved chart rendering performance with updated signal library
• Reorganized package structure with better organization and maintainability

Internal Changes

Build System

• Updated all crate versions from 0.0.88 to 0.0.89 across the workspace
• Enhanced Cargo.toml dependency management and version consistency (source)
• Added comprehensive CHANGELOG.md with professional formatting and detailed documentation

Development Infrastructure

• Improved development workflow with better package management
• Enhanced project organization with standardized documentation structure
• Better integration between frontend and backend components

View changes

v0.0.88 - 2025-08-10

Build System

Rust Environment Standardization

• Added explicit Rust version specification (1.89) to workspace configuration for consistent compilation across development environments
• Enhanced build reliability with standardized toolchain requirements
• Improved development environment consistency and reduced compilation variance

Website Enhancements

SOPR Visualization Improvements

• Enhanced SOPR (Spent Output Profit Ratio) chart configuration with improved key handling and better code organization
• Added conditional display logic for adjusted SOPR (aSOPR) charts when data is available
• Improved chart organization with cleaner key generation and more robust data handling
• Better separation between standard SOPR and adjusted SOPR visualization options

Internal Changes

Version Management

• Updated all crate versions from 0.0.87 to 0.0.88 across the workspace for consistent release management
• Enhanced workspace version coordination and dependency consistency

View changes

v0.0.87 - 2025-08-10

Breaking Changes

• Vector Storage Architecture Refactor: Replaced internal brk_vecs crate with externalized vecdb crate, moving vector storage functionality to a separate project while maintaining the same core functionality (Cargo.toml)

Architecture Improvements

• Crate Externalization: Migrated vector storage implementation from internal brk_vecs and brk_vecs_macros to external vecdb and vecdb_derive crates, enabling independent development and versioning of the vector storage system
• Sequential Database Foundation: Added seqdb as the underlying sequential data storage engine powering the vector database functionality
• Workspace Simplification: Removed brk_vecs and brk_vecs_macros from the workspace, reducing codebase complexity and focusing on Bitcoin-specific functionality (crates/brk/Cargo.toml)

API Changes

• Vector Interface Consistency: Updated all vector references from brk_vecs to vecdb while maintaining identical API surface for seamless transition (crates/brk/src/lib.rs)
• Documentation Updates: Updated README and project documentation to reflect the new external vector database dependency (README.md)

Development Experience

• Modular Development: Enables independent development of vector storage functionality separate from Bitcoin analysis logic
• Local Development Support: Added commented local path configuration for easier development workflow when working on both projects simultaneously

View changes

v0.0.85 - 2025-08-07

Bug Fixes

• Logging Filter Correction: Fixed typo in logger filter configuration, changing rmcp=off to brk_rmcp=off to properly suppress unwanted logs from the RMCP protocol module (crates/brk_logger/src/lib.rs)

Configuration Changes

• MCP Server Configuration: Removed stateful mode setting from MCP server configuration, enabling more flexible protocol handling and improved compatibility with different client implementations (crates/brk_mcp/src/route.rs)

View changes

v0.0.84 - 2025-08-07

Breaking Changes

• Major Architecture Restructuring: Complete reorganization of the codebase with extraction of core functionality into dedicated crates (brk_structs, brk_error) and comprehensive modularization of all components for better maintainability and reusability
• Struct System Overhaul: Moved all Bitcoin-related structures from brk_core to new brk_structs crate with enhanced grouping capabilities, comprehensive address types, and improved serialization support (crates/brk_structs/src/lib.rs)

New Features

• Centralized Error Handling: Introduced brk_error crate providing unified error handling across all BRK components with comprehensive error types for different subsystems (crates/brk_error/src/lib.rs)
• Advanced Address Analytics: Enhanced address grouping system with comprehensive filtering capabilities including by address type, amount ranges, age ranges, epochs, terms, and spendable/unspendable classifications (crates/brk_structs/src/groups/mod.rs)
• Vector Macro System: Added brk_vecs_macros crate providing procedural macros for automated vector trait implementations and boilerplate reduction (crates/brk_vecs_macros/src/lib.rs)
• Enhanced Data Types: Comprehensive Bitcoin data type system including transaction IDs, block hashes, raw locktime, version numbers, and specialized index types for all Bitcoin primitives (crates/brk_structs/src/structs/mod.rs)

Architecture Improvements

• Modular Design: Complete separation of concerns with dedicated crates for different functionality areas, enabling better code organization and independent versioning
• Build System Enhancement: Added comprehensive build scripts across all crates with cargo-dist integration for automated release management and distribution (.github/workflows/release.yml)
• Price Analysis: New dedicated price analysis module with enhanced market data processing and price-to-amount state tracking (crates/brk_computer/src/price.rs)

Developer Experience

• Comprehensive Documentation: Added README files and examples for all major crates, providing clear usage instructions and API documentation
• Enhanced Examples: Expanded example collection demonstrating various BRK components and usage patterns across all crates
• Development Tools: Improved development workflow with better build scripts, dependency management, and testing infrastructure

Performance Improvements

• Vector System: Enhanced vector system with improved compression, better memory management, and optimized I/O operations through the redesigned brk_vecs architecture
• State Management: Improved state tracking with dedicated modules for block states, realized/unrealized values, supply tracking, and transaction analysis

Infrastructure

• CI/CD Enhancement: Updated GitHub Actions workflow with improved Ubuntu runner (22.04) and enhanced release automation
• Docker Improvements: Enhanced Docker configuration with better environment handling and deployment options

Social Media

• Platform Updates: Updated social media links removing outdated platforms and focusing on active community channels (README.md)

View changes

v0.0.83 - 2025-07-26

New Features

• Cross-Platform Hole Punching: Added Linux support for file hole punching operations using fallocate system call, expanding platform compatibility beyond macOS for efficient sparse file management (crates/brk_vecs/src/file/mod.rs)

Performance Improvements

• Platform-Specific Optimization: Implemented platform-specific hole punching strategies with optimized system calls for both macOS (fcntl with F_PUNCHHOLE) and Linux (fallocate), improving storage efficiency and I/O performance on different operating systems

Internal Changes

• Code Organization: Refactored hole punching implementation with unified punch_hole_impl() method that dispatches to platform-specific implementations, improving code maintainability and reducing duplication
• Error Handling: Enhanced error handling for unsupported platforms with clear error messages for hole punching operations

View changes

v0.0.82 - 2025-07-26

Breaking Changes

• Vector System Refactoring: Replaced brk_vec with new brk_vecs crate, introducing a completely redesigned vector storage system with improved file layouts, better compression, and enhanced performance characteristics (crates/brk_vecs/src/lib.rs)

New Features

• Advanced Vector Architecture: Introduced comprehensive vector system with multiple storage variants including RawVec, CompressedVec, ComputedVec, EagerVec, LazyVec, and StampedVec for different use cases and performance requirements (crates/brk_vecs/src/variants/mod.rs)
• File-Based Storage: New file-based storage system with dedicated File and Reader implementations, featuring page-based architecture and memory mapping for efficient disk I/O (crates/brk_vecs/src/file/mod.rs)
• Stamped Vectors: Added timestamped vector support with StampedVec and Stamp structures for versioned data management and temporal analysis (crates/brk_vecs/src/variants/stamped/mod.rs)

Architecture Improvements

• Lazy Computation: Enhanced lazy evaluation system with support for 1, 2, and 3-source computed vectors, enabling complex data transformations with minimal memory overhead (crates/brk_vecs/src/variants/lazy/mod.rs)
• Compression Optimization: Redesigned compression system with page-based metadata management and improved compression ratios for large datasets (crates/brk_vecs/src/variants/compressed/mod.rs)
• Type System Enhancement: Improved trait system with dedicated interfaces for different vector operations, providing better type safety and performance (crates/brk_vecs/src/traits/mod.rs)

Performance Improvements

• Build Optimization: Updated build scripts to focus on release builds for better performance in production deployments (update.sh)
• Memory Management: Enhanced memory usage patterns with better file mapping strategies and reduced memory footprint for large-scale Bitcoin data processing

Infrastructure

• Comprehensive Examples: Added extensive examples demonstrating various vector types and usage patterns for better developer onboarding (crates/brk_vecs/examples/)
• Documentation: Enhanced documentation with detailed explanations of vector architectures and storage strategies

Internal Changes

• Address Handling: Improved address byte hash management and height-based indexing for better address analytics performance
• Group Processing: Enhanced address and UTXO grouping capabilities with better filtering and processing mechanisms
• Error Handling: Refined error handling throughout the vector system for more robust data operations

View changes

v0.0.81 - 2025-07-17

Breaking Changes

• Computer Module Restructuring: Major reorganization of the computer module from nested vecs structure to flat module layout, moving vector implementations directly into separate modules (all, blocks, cointime, constants, fetched, grouped, indexes, market, mining, stateful, transactions) for better maintainability (crates/brk_computer/src/lib.rs)

New Features

• Docker Support: Added comprehensive Docker setup with Dockerfile, docker-compose configuration, environment templates, and detailed documentation for containerized BRK deployment (docker/README.md)
• Address Index System: Implemented new address indexing architecture with AnyAddressIndex, EmptyAddressIndex, and LoadedAddressIndex structures for more efficient address data management (crates/brk_core/src/structs/anyaddressindex.rs)
• Enhanced Address Grouping: Added comprehensive address filtering and grouping capabilities with by_address_type, by_any_address, and flexible filter systems (crates/brk_core/src/groups/mod.rs)

Architecture Improvements

• Stateful Processing: Enhanced stateful vector processing with dedicated traits and improved address data source handling (crates/brk_computer/src/stateful/trait.rs)
• Address Analytics: Added sophisticated address counting and analysis capabilities with support for different address types and cohort analysis
• Module Organization: Flattened module hierarchy eliminating nested structures for clearer code organization and easier navigation

Infrastructure

• Container Deployment: Full Docker containerization support with multi-stage builds, environment configuration, and production-ready setup
• Build Optimization: Enhanced build scripts and profiling tools for better development workflow

Documentation

• Website Updates: Added brekit.org as primary domain in README alongside existing mirrors (README.md)
• Docker Guide: Comprehensive Docker setup and deployment documentation with prerequisites, configuration, and usage instructions

View changes

v0.0.80 - 2025-07-13

Performance Improvements

• Vector Length Caching: Added local length caching in raw vectors with local_stored_len field, reducing atomic operations and improving read performance by storing frequently accessed length values locally (crates/brk_vec/src/variants/raw.rs)

API Enhancements

• Length Management: Introduced set_stored_len() method for direct length management, providing better control over vector metadata and synchronization between local and shared length values

Documentation

• README Cleanup: Removed detailed AI example links from main README to streamline content and focus on core functionality, improving readability and reducing maintenance overhead (README.md)
• MCP Documentation: Added dedicated README for Model Context Protocol integration with comprehensive usage examples and setup instructions (crates/brk_mcp/README.md)

Internal Changes

• Vector Architecture: Enhanced raw and compressed vector implementations with improved length tracking mechanisms for better performance and consistency

View changes

v0.0.79 - 2025-07-13

Performance Improvements

• File Handle Optimization: Reduced minimum file handle limit from 250,000 to 10,000, significantly lowering system resource requirements while maintaining adequate file access for most Bitcoin datasets (crates/brk_core/src/utils/rlimit.rs)

Breaking Changes

• Vector Memory Architecture: Replaced ArcSwap<Mmap> with Arc<AtomicUsize> for stored length tracking in raw vectors, eliminating memory mapping overhead and improving concurrent access patterns while reducing memory usage (crates/brk_vec/src/variants/raw.rs)

Internal Changes

• Vector Storage Simplification: Removed arc-swap dependency from vector variants, streamlining memory management and reducing complexity in concurrent scenarios
• Store Integration: Enhanced indexer store integration with better error handling and resource management
• Header Management: Improved vector header handling for more efficient metadata operations

View changes

v0.0.78 - 2025-07-13

New Features

• Semester Index Support: Added SemesterIndex struct for 6-month interval datasets, enabling bi-annual analysis of Bitcoin metrics with proper serialization and mathematical operations (crates/brk_core/src/structs/semesterindex.rs)
• Enhanced Vector Architecture: Refactored vector computation infrastructure with improved VecBuilderOptions and Source handling for more flexible dataset generation (crates/brk_computer/src/vecs/grouped/builder_computed.rs)

Architecture Improvements

• Vector Builder Refactoring: Renamed StorableVecGeneatorOptions to VecBuilderOptions for better naming consistency and enhanced vector generation capabilities
• Cointime Integration: Enhanced cointime computation with better indexing support and computation parameter handling
• Profiling Tools: Added dedicated flamegraph and samply profiling scripts for the computer module to aid performance analysis

Development Experience

• Documentation Updates: Expanded TODO list with additional feature requests including chopiness datasets, reused address analytics, z-score chart improvements, and HTTPS support
• Profiling Support: Added flamegraph.sh and samply.sh scripts to the computer module for better performance debugging capabilities

View changes

v0.0.76 - 2025-07-09

Build System Updates

• Rolldown Bundler: Updated brk_rolldown from 0.1.0 to 0.1.1 with improvements across all bundler components including enhanced ECMAScript utilities, better plugin architecture, and improved string manipulation capabilities (crates/brk_bundler/Cargo.toml)

View changes

v0.0.75 - 2025-07-09

Performance Improvements

• Chunk Size Optimization: Adjusted parallel processing chunk size from length/4 to length/3, increasing parallelization granularity for better performance on high-end storage systems while maintaining external drive protection (crates/brk_computer/src/vecs/stateful/mod.rs)

View changes

v0.0.74 - 2025-07-09

Performance Improvements

• Parallel Processing Optimization: Refactored stateful vector computation to use chunked parallel processing with controlled chunk sizes (length/4) instead of unbounded parallelization, preventing external drive bottlenecks and improving performance on Thunderbolt 4 SSDs (crates/brk_computer/src/vecs/stateful/mod.rs)
• Thread Scope Elimination: Replaced complex thread scope management with streamlined Either-based collection processing, reducing threading overhead and improving memory usage patterns during computation

Bug Fixes

• Memory Table Rotation Logging: Fixed missing log message for memory table rotation operations, improving debugging and monitoring capabilities during store operations (crates/brk_computer/src/stores.rs)

Internal Changes

• Computation Architecture: Unified UTXO and address vector processing using Either enum for type-safe handling of different vector types while maintaining performance through parallel iteration
• Chunk-based Processing: Implemented intelligent chunking strategy to prevent overwhelming external storage devices while maintaining parallelization benefits

View changes

v0.0.73 - 2025-07-09

Breaking Changes

• Architecture Restructuring: Removed brk_state crate entirely, consolidating its functionality into brk_computer with direct store and serialization dependencies, simplifying the overall architecture and reducing module complexity
• Build System Upgrade: Updated brk_rolldown bundler from 0.0.1 to 0.1.0 with significant API changes including removal of deprecated loader utilities and addition of hot module replacement support

New Features

• Hot Module Replacement: Added HMR plugin support for improved development experience with live code updates and faster iteration cycles during bundling operations
• Advanced Hashing: Integrated Blake3 hashing algorithm for improved performance in debug operations and file integrity verification

Internal Changes

• Computer Module Enhancement: Added bincode serialization, zerocopy optimizations, and derive_deref functionality directly to the computer module for better performance
• Store Integration: Enhanced store module with logging capabilities for better debugging and monitoring
• Plugin Architecture: Expanded bundler plugin system with data URI support and improved ECMAScript utilities

View changes

v0.0.71 - 2025-06-25

New Features

• Enhanced Schema Documentation: Added comprehensive descriptions to all Index enum variants and API parameters, providing clear explanations for date/day index, height/block index, transaction indexes, address types, and time periods for better API usability (crates/brk_interface/src/index.rs, crates/brk_interface/src/params.rs)

API Changes

• MCP Tool Modernization: Upgraded Model Context Protocol implementation to use new tool_router and Parameters wrapper for better type safety and parameter handling, replacing legacy tool_box approach (crates/brk_mcp/src/lib.rs)
• Endpoint Renaming: Renamed get_variant_count to get_vec_count for better consistency with API terminology

Internal Changes

• Development Flexibility: Added commented local development path for brk_rmcp to facilitate easier testing and development workflows

View changes

v0.0.70 - 2025-06-24

New Features

• Model Context Protocol Default: Enabled MCP (Model Context Protocol) by default for LLM integration, changing from opt-in to opt-out behavior to make AI-powered Bitcoin data analysis more accessible (crates/brk_cli/src/config.rs)

API Changes

• Simplified Endpoints: Streamlined API endpoints by replacing complex query parameters with dedicated paths (/height-to-ohlc and /dateindex-to-ohlc) for cleaner and more intuitive data access (crates/brk_fetcher/src/brk.rs)

Internal Changes

• Module Restructuring: Flattened fetcher module structure by moving fetcher implementations directly into the main crate, eliminating nested module hierarchy for better maintainability
• Documentation: Updated API badge URL to reflect new vec-count endpoint and improved MCP tool description clarity (README.md)
• Code Cleanup: Removed commented debug code and unused file handle references from raw vector implementation

View changes

v0.0.69 - 2025-06-24

Performance Improvements

• Vector Initialization Optimization: Removed thread-based parallel initialization of indexes and fetched vectors, replacing with sequential initialization to reduce overhead and improve memory usage patterns during startup (crates/brk_computer/src/vecs/mod.rs)
• File Handle Management: Refactored vector file handling architecture to use on-demand file opening instead of persistent file handles, reducing memory footprint and file descriptor usage (crates/brk_vec/src/variants/raw.rs)

Internal Changes

• Resource Limit Adjustment: Reverted file handle limit back to 210,000 from 420,000, optimizing for more conservative resource usage
• Header Write Logic: Simplified vector header writing by removing conditional logic and always performing write operations when needed, improving code clarity and reducing potential race conditions (crates/brk_vec/src/structs/header.rs)
• Memory Map Updates: Refactored memory map update logic to accept file references directly, eliminating need for persistent file handle storage and improving resource management
• Code Cleanup: Removed unused imports and simplified vector trait implementations for better maintainability

View changes

v0.0.68 - 2025-06-24

New Features

• CLI Version Information: Added version command support to CLI configuration, enabling users to check the current version of BRK directly from the command line (crates/brk_cli/src/config.rs)

Performance Improvements

• File Handle Limits: Doubled the minimum file handle limit from 210,000 to 420,000 to accommodate larger Bitcoin datasets and prevent resource exhaustion during intensive blockchain processing operations (crates/brk_core/src/utils/rlimit.rs)

View changes

v0.0.67 - 2025-06-24

New Features

• Data Interface Layer: Added brk_interface crate providing a comprehensive data interface layer with serialization, schema generation, and tabular formatting capabilities for exporting datasets in multiple formats (JSON, CSV, TSV, Markdown)
• Model Context Protocol Support: Introduced brk_mcp crate implementing Model Context Protocol (MCP) for seamless integration with Large Language Models, enabling AI-powered Bitcoin data analysis and querying
• External MCP Integration: Added brk_rmcp external dependency for robust MCP client/server communication capabilities

Breaking Changes

• Dependency Restructuring: Removed direct brk_query dependency from CLI, replaced with modular brk_interface and brk_mcp architecture for better separation of concerns
• Storage Simplification: Removed arc-swap and fjall dependencies from brk_state, streamlining state management architecture

Internal Changes

• Build System: Added auto-generated vecid-to-indexes.js file to gitignore to prevent version control conflicts
• Documentation Updates: Updated historical changelog entries to reflect correct repository URLs and asset paths, migrating from kibo-money references to bitcoinresearchkit organization
• Dependency Updates: Updated multiple core dependencies including async-compression, autocfg, errno, and syn to their latest versions for improved performance and compatibility

View changes

v0.0.66 - 2025-06-19

New Features

• Cointime Analysis: Added comprehensive cointime economic metrics including coinblocks creation/storage tracking, liveliness and vaultedness calculations, activity ratios, and specialized price metrics (vaulted price, active price, true market mean, cointime price) with corresponding market cap calculations (crates/brk_computer/src/vecs/cointime.rs)
• Advanced Supply Analysis: Implemented vaulted and active supply tracking with dollar value computations, enabling analysis of Bitcoin storage behavior and economic activity patterns

Documentation

• CLI Documentation: Improved CLI README with better formatting, clearer installation instructions, enhanced usage examples, and restructured content with proper markdown syntax and information blocks (crates/brk_cli/README.md)
• Installation Guidance: Added prominent installation notes for Ubuntu users and improved download section formatting with better link formatting and structure

Internal Changes

• Ratio Analysis Integration: Connected cointime price metrics with ratio analysis system for comprehensive trend analysis across vaulted price, active price, true market mean, and cointime price ratios

View changes

v0.0.65 - 2025-06-17

Bug Fixes

• Chart Data Update Logic: Simplified chart data update algorithm by removing complex comparison logic and replacing it with a streamlined approach that updates all data points at or after the last known time, significantly reducing complexity and potential update conflicts (websites/default/packages/lightweight-charts/wrapper.js)

Internal Changes

• Cohort Analysis Versioning: Incremented version numbers for cumulative net realized profit/loss metrics from VERSION::TWO to VERSION::new(3), ensuring proper cache invalidation and data consistency for cohort analysis computations (crates/brk_computer/src/vecs/stateful/cohort.rs)

View changes

v0.0.64 - 2025-06-17

New Features

• 4-Year Statistical Analysis: Added comprehensive 4-year moving averages and standard deviation calculations for ratio analysis, enabling long-term trend analysis with ratio_4y_sma, ratio_4y_sd, and ratio_4y_zscore metrics (crates/brk_computer/src/vecs/grouped/ratio_from_dateindex.rs)
• Z-Score Computation Optimization: Refactored z-score calculations to use a dedicated compute_zscore() method, improving code maintainability and performance by centralizing statistical computations

Internal Changes

• Vector Naming Convention: Renamed ratio_standard_deviation to ratio_sd for consistency with other abbreviated metric names
• Statistical Method Extraction: Extracted z-score computation logic into reusable methods, reducing code duplication and improving mathematical accuracy
• Vector Builder Enhancement: Added conditional naming logic in grouped vector builder for better handling of sum operations with optional suffixes

View changes

v0.0.63 - 2025-06-16

New Features

• Static File Caching: Implemented intelligent HTTP caching strategy for static assets, setting immutable cache headers for long-term assets (images, fonts, JavaScript, map files) and must-revalidate headers for HTML and service worker files to ensure optimal performance and cache invalidation (crates/brk_server/src/files/file.rs)
• Progressive Web App Assets: Added comprehensive PWA assets including Apple Touch icons, splash screens for all iOS device sizes and orientations, and maskable icons for Android, enabling native app-like experience across all mobile platforms

Internal Changes

• File Extension Detection: Refactored file extension checking from string-based ends_with() pattern matching to more robust Path::extension() method for better accuracy and maintainability
• Cache Logic Simplification: Consolidated cache control logic by removing redundant file path checks and grouping related file types for cleaner code organization

View changes

v0.0.62 - 2025-06-16

Bug Fixes

• TypeScript Bridge Generation: Fixed JavaScript/TypeScript bridge generation by properly handling optional return values when reading hashed main.js entry files, preventing crashes when entry files don't exist (crates/brk_bundler/src/lib.rs)
• TypeScript Type Generation: Restructured TypeScript type definitions for VecIdToIndexes, moving typedef comments outside function declaration for proper type recognition (crates/brk_server/src/api/query/bridge.rs)

New Features

• Chart Time Scale Optimization: Improved time scale spacing for different chart time periods with more balanced bar spacing (quarters: 3→2, years: 12→6, decades: 120→60) for better visual density (websites/default/packages/lightweight-charts/wrapper.js)
• Progressive Web App Support: Added service worker scope configuration and improved PWA manifest handling for better standalone app experience

Internal Changes

• Chart Data Processing: Refactored chart data handling to use object destructuring instead of arrays, improving code clarity and type safety in effect callbacks
• HTML Standards: Updated HTML5 DOCTYPE declaration and improved syntax consistency
• Error Handling: Added try-catch protection for localStorage operations to prevent crashes in restricted environments

View changes

v0.0.61 - 2025-06-15

New Features

• OHLC Price Continuity: Added logic to maintain price continuity between days in OHLC data computation by setting today's opening price to the previous day's closing price, ensuring no gaps in price series when computing daily OHLC from fetched data (crates/brk_computer/src/vecs/fetched.rs)

Internal Changes

• Struct Mutability: Added DerefMut trait to all OHLC struct variants (OhlcF64, OhlcBtc, OhlcSats, OhlcUsd) enabling mutable access to underlying values
• Error Handling: Improved file creation error handling in block recap export with more descriptive panic messages

View changes