cctally 1.22.2 → 1.22.4

package/CHANGELOG.md

@@ -5,6 +5,27 @@ based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
 ## [Unreleased]
 
+## [1.22.4] - 2026-06-01
+
+### Fixed
+- **A surprise mid-week Anthropic usage reset is now reflected in the 7d percentage even when you were below ~25% usage.** When Anthropic zeroes the weekly counter mid-window (same reset timestamp, usage drops to ~0 — e.g. the 2026-06-01 incident), the reset detector previously only fired when the drop was at least 25 percentage points, so any account reset from a lower base (the observed 14% → 0%) slipped through: no `week_reset_events` row was written, the monotonic high-water-mark clamp kept reporting the stale pre-reset percentage across the statusline, `weekly`/`report`, and the dashboard, and post-reset 0% reads were silently dropped at the write site. The detector now ALSO fires on a reset-to-zero — when the post-reset value collapses to ~0 (≤ 1%) with at least a 3pp drop — independently of the 25pp magnitude gate, since a clean drop to zero is an unambiguous reset regardless of size (a lagging API replica reports a slightly-lower number, never a clean 0 against real usage), while the 3pp floor rejects 1%→0% replica jitter that would otherwise spuriously segment the week. The new discriminator is a single shared `_is_reset_drop()` helper wired into all four 7d detection sites (live + backfill, boundary-advance + in-place-credit branches) so they stay byte-identical; the 25pp partial-credit path and the separate 5h detector are unchanged. Regression: three new cases in `tests/test_in_place_credit_detection.py` (live reset-to-zero fires below 25pp, the 3pp floor rejects 1%→0% jitter, and backfill parity).
+
+### Changed
+- **Internal refactor (no user-facing change): extracted the shared formatting/color/table render primitives — the boxed-table renderer, the color/unicode capability detectors, the ANSI styling + display-width helpers, the integer/width-budget number formatters, the compact timestamp/week-window formatters, and the `_ANSI_ESC_RE` regex (plus the small `_parse_iso_datetime_optional` helper they depend on) — out of the `bin/cctally` single-file program into a new stdlib-only pure-function sibling module, `bin/_lib_fmt.py` (#126, the final tranche of the `bin/cctally` split).** This trims the main script by ~250 lines (2,969 → 2,715), and additionally converts the eight back-references these primitives had in `bin/_lib_render.py` and `bin/_lib_diff_kernel.py` from `cctally`-namespace shims into honest direct imports of the new kernel — with no change to any command's behavior, flags, output, or exit codes: every reporting/`diff`/`project`/`cache-report`/`forecast`/`blocks`/`percent-breakdown`/`five-hour-blocks` golden test is byte-identical and the full harness + pytest suite (1,329 checks) passes, a new `tests/test_lib_fmt_extraction_invariants.py` locks the extraction's import discipline, and the new module ships in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+
+## [1.22.3] - 2026-06-01
+
+### Changed
+- **Internal refactor (no user-facing change): extracted the OpenAI/Codex parity command family (`codex-daily`/`codex-monthly`/`codex-weekly`/`codex-session` plus their speed/timezone resolvers and the `--debug` cost-stats cluster) and the Claude reporting command family (`daily`/`monthly`/`weekly`/`session`) out of the `bin/cctally` single-file program into two new stdlib-only sibling modules — `bin/_cctally_codex.py` and `bin/_cctally_reporting.py` — continuing the ongoing `bin/cctally` split.** This trims the main script by ~980 lines (8,843 → 7,864) with no change to any command's behavior, flags, output, or exit codes — every `codex-daily`/`codex-monthly`/`codex-weekly`/`codex-session`, `daily`/`weekly`/`session`, `daily --instances`, and subgroup golden test is unchanged and the full harness + pytest suite passes — and both modules ship in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): extracted the two read-only diagnostics commands — `doctor` (`doctor_gather_state` + `cmd_doctor`) and `pricing-check` (the LiteLLM/Anthropic fetchers + `cmd_pricing_check` + its text renderer) — out of the `bin/cctally` single-file program into two new stdlib-only sibling modules, `bin/_cctally_doctor.py` and `bin/_cctally_pricing_check.py` (continuing the ongoing `bin/cctally` split).** This trims the main script by ~700 lines (11,441 → 10,744) with no change to either command's behavior, flags, output, or exit codes — every `doctor` and `pricing-check` golden test is unchanged and the full harness + pytest suite passes — and both modules ship in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): extracted the two 5-hour-window command families and the statusline command out of the `bin/cctally` single-file program into two new stdlib-only sibling modules — `bin/_cctally_five_hour.py` (`blocks` / `five-hour-blocks` / `five-hour-breakdown` plus the shared recorded-window resolution layer) and `bin/_cctally_statusline.py` (`statusline` glue plus the per-model context-window tables) — continuing the ongoing `bin/cctally` split.** This trims the main script by ~1,900 lines (10,741 → 8,843) with no change to any command's behavior, flags, output, or exit codes — every `blocks`, `five-hour-blocks`, `statusline`, subgroup, and 5h-canonical golden test is unchanged and the full harness + pytest suite passes — and both modules ship in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): extracted the `report`/`forecast`/`budget` command family — `cmd_report`, `cmd_forecast`, `cmd_budget`, the forecast core cluster (window resolvers, `ForecastInputs`/`ForecastOutput`/`BudgetRow`, the projection math, and the terminal/JSON/status-line renderers), and the ten budget render/snapshot helpers — out of the `bin/cctally` single-file program into a new stdlib-only sibling module, `bin/_cctally_forecast.py` (continuing the ongoing `bin/cctally` split).** This trims the main script by ~1,860 lines (7,864 → 6,000) with no change to any command's behavior, flags, output, or exit codes — every `forecast`, `budget`, and share golden test is byte-identical and the full harness + pytest suite passes — the budget write-path helpers and the week-boundary join layer stay in the main program (reached via the runtime namespace accessor), and the new module ships in the npm/brew/public packages (promoted to the mirror allowlist). A new namespace-binding regression test guards the module-boundary wiring. Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): split the `cache-report` command out of the `bin/cctally` single-file program — the pure data kernel was renamed to the conventional `bin/_lib_cache_report.py` and the command/render glue moved into a new stdlib-only sibling, `bin/_cctally_cache_report.py`, so `cache-report` now matches the `_lib_<x>`/`_cctally_<x>` shape of `statusline` and `doctor` (continuing the ongoing `bin/cctally` split).** This trims the main script by ~1,170 lines (6,000 → 4,833) with no change to the command's behavior, flags, output, or exit codes — every `cache-report` golden test is unchanged and the full harness + pytest suite passes — and both modules ship in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): extracted the weekly cost-snapshot + milestone DB cluster and the `percent-breakdown` command handler out of the `bin/cctally` single-file program into two new stdlib-only sibling modules — `bin/_cctally_milestones.py` and `bin/_cctally_percent_breakdown.py` (continuing the ongoing `bin/cctally` split).** This trims the main script by ~530 lines (4,831 → 4,297) with no change to any command's behavior, flags, output, or exit codes — every `percent-breakdown` and milestone-idempotency golden test is unchanged and the full harness + pytest suite passes — and both modules ship in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): extracted the `diff` command (`cmd_diff` plus its two-window debug-sample reporter) out of the `bin/cctally` single-file program into a new stdlib-only sibling module, `bin/_cctally_diff.py`, and folded the `range-cost` command into the existing `bin/_cctally_reporting.py` sibling (continuing the ongoing `bin/cctally` split).** This trims the main script by ~340 lines (4,297 → 3,954) with no change to either command's behavior, flags, output, or exit codes — every `diff` and `range-cost` golden test is unchanged and the full harness + pytest suite passes — and the new module ships in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): extracted the WeekRef / reset-event cluster — the canonical week-boundary resolvers, `get_recent_weeks`, the reset-event backfill/apply/overlap-clamp helpers, the per-WeekRef cost computer, and the two reset-drop-threshold constants — out of the `bin/cctally` single-file program into a new stdlib-only sibling module, `bin/_cctally_weekrefs.py`, and folded the 5-hour-block backfill into the existing `bin/_cctally_five_hour.py` sibling (continuing the ongoing `bin/cctally` split).** This trims the main script by ~660 lines (3,954 → 3,293) with no change to any command's behavior, flags, output, or exit codes — every affected `weekly`/`report`/`five-hour-blocks` golden test is unchanged and the full harness + pytest suite passes — and the new module ships in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+- **Internal refactor (no user-facing change): extracted the pricing-observed-models scan into the existing `bin/_cctally_pricing_check.py` sibling, the pure pricing-mismatch-debug kernel into a new stdlib-only module `bin/_lib_pricing_debug.py`, and the setup/hook settings-file glue into the existing `bin/_cctally_setup.py` sibling — the final batch of the `bin/cctally` single-file split (#125).** This trims the main script by ~320 lines (3,293 → 2,969) with no change to any command's behavior, flags, output, or exit codes — every `pricing-check`, `doctor`, and `setup` golden test is unchanged and the full harness + pytest suite (1,328 checks) passes — and the new module ships in the npm/brew/public packages (promoted to the mirror allowlist). Purely a maintainability / code-organization change; nothing to do on upgrade.
+
 ## [1.22.2] - 2026-05-30
 
 ### Changed