readViaStream (used for files ≥8 MB) reconstructs the full file as a
single string via chunks.join('\n'), giving the same peak allocation as
readFile. Callers then call content.split('\n'), creating a second copy.
With FILE_READ_CONCURRENCY=16 and files up to 128 MB this can exhaust
the V8 heap (~6 GB theoretical peak).
readSessionLines already exists as a proper async generator that yields
one line at a time. Switch both hot-path callers to iterate it directly
so the full file string is never held in memory.
Adds two tests: a spy test confirming readSessionLines is called (not
readSessionFile), and a 500-entry correctness test.
Fixes#131
The PR #59 merge landed on main (ac31883) before the 0.8.7 bump
(cb44cc6), so the code shipped in 0.8.7 but the initial changelog entry
only credited MiniMax and the menubar fixes. Adds three entries under
Added for the OMP provider, the model-alias command, and the built-in
Anthropic proxy aliases. Adds one entry under Fixed (CLI) for the
prototype-pollution guard in resolveAlias. Updates the CLI parity note
so users on 0.8.5 know whether they can stay there.
Filed before 0.8.7 hits npm.
Skips 0.8.6 to match mac-v0.8.7 and keep CLI + menubar versions aligned.
CLI change: MiniMax-M2.7 and MiniMax-M2.7-highspeed pricing. macOS menubar
change: three stability fixes culminating in the App Nap opt-out that keeps
the refresh loop ticking while the icon is idle in the background.
Confirmed in the system log: while the menubar icon sits idle in the
background, macOS flips _kLSApplicationWouldBeTerminatedByTALKey to 1,
which is the Automatic Termination and App Nap subsystem declaring the
app a suspend candidate. Once that happens the 15s refresh Task's sleep
stretches arbitrarily, so the status bar label freezes until the user
clicks the icon (which triggers NSApp.activate and wakes everything up).
Hold a ProcessInfo.beginActivity handle for the life of the app with
.userInitiated + .automaticTerminationDisabled + .suddenTerminationDisabled
so macOS leaves the app alone. Released implicitly when the app exits.
Adds Oh My Pi support by parameterizing the Pi JSONL reader to accept a
providerName, so sessions at ~/.omp/agent/sessions/ are discovered and
tracked alongside Pi. Ships a model-alias CLI command plus five
built-in aliases for the anthropic--claude-X.Y-tier double-dash format
that some Anthropic-compatible proxies emit, so cost rows no longer
read $0.00 for those names.
Contributed by @cgrossde.
Second merge of main since the PR was opened. Main moved 30+ commits
(0.8.5 bump, plan tracking feature, MiniMax pricing, menubar
prefetchAll walk-back, aicrowd cache rewrite revert) so the branch
needed another reconciliation before merging to main.
Two new conflicts resolved. Took main's text in both cases per the
policy of favoring main when the feature work is neutral:
README.md Kept main's Node 20+ / better-sqlite3
Requirements wording and main's shorter src/
tree listing. Added OMP to the Requirements
line.
src/providers/pi.ts Main dropped the discovery-cache snapshot and
the rich source-metadata fields as part of the
aicrowd revert. Took main's simpler structure
and only kept the providerName parameter so
OMP sources still report the correct provider
in the session source and dedup key.
Earlier fixups carried forward from the prior merge commit:
- Object.hasOwn guards in resolveAlias against prototype-pollution
via a model literally named '__proto__'.
- source.provider in the dedup key prefix so OMP rows no longer
stamp 'pi:'.
- Combined pi.js imports in providers/index.ts.
- Trailing newline on pi.ts.
- Unknown-model fallback in cursor-agent.ts from yesterday's PR #117
fixup (preserved via main).
353 tests pass (count dropped from 378 because main deleted the
parse-progress / parser-cache / provider-colors / source-cache test
files alongside the cache-rewrite revert).
Feature work by @cgrossde.
Two independent causes for the stuck-label / only-refreshes-on-click
behaviour, both fixed here.
1. NSStatusItem button defers the status bar paint for accessory apps
that are not foreground, so after refreshStatusButton sets the new
attributed title the menu bar visually froze until the user opened
the popover (which triggers NSApp.activate and a forced redraw
cycle). Explicit needsDisplay + display() forces the paint every
cycle.
2. The codeburn subprocess inherited the accessory app's default QoS,
which macOS background-throttles. That could stretch a sub-1-second
parse into tens of seconds on large corpora and overrun the 15s
refresh cadence. Set .userInitiated so the CLI runs at the same
priority it does from a user-interactive terminal.
The detached prefetch spawned four codeburn subprocesses (week, 30days,
month, all) that competed with the main refresh loop's Today fetch for
disk and parser time. On large session corpora that starved the Today
refresh for minutes at a time, and the status label drifted stale until
the prefetch queue finally drained. Users perceived this as the menubar
"stopping" until they clicked a provider tab, which coincided with the
queue clearing.
Drop prefetchAll. Period tabs in the popover now fetch lazily on first
click (3-10 seconds once per app lifetime per period), same as they did
in 0.8.1. Today's label stays fresh because nothing else is hammering
the parser in the background.
Adds pr-number input and workflow_dispatch trigger so existing open PRs
can be scanned without waiting for a new push. Leave pr-number empty to
loop all open PRs; supply a number to rescan one.
Runs on every PR open/reopen/synchronize against getagentseal/firstlook
and fails the check when the author's score is 'unknown' (new or
untracked accounts). Skips bot accounts so dependabot and renovate pass
through.
This screens the same drive-by pattern that landed in PR #118 (octo-patch,
fresh automation account) without requiring a manual tier check on every
submission.
Adds FALLBACK_PRICING entries plus display names so MiniMax sessions
show up with the right cost and readable labels when users route MiniMax
through providers like OpenCode. Pricing verified against the live
MiniMax paygo page:
MiniMax-M2.7 input $0.3/M output $1.2/M cache-read $0.06/M cache-write $0.375/M
MiniMax-M2.7-highspeed input $0.6/M output $2.4/M cache-read $0.06/M cache-write $0.375/M
prefetchAll was awaited inside the refresh loop after the first cycle.
On large corpora the All Time parse takes 60+ seconds and kept the Today
refresh blocked for that whole window, so the status label got stuck for
minutes at a time while the CLI was returning fresh numbers.
Detach prefetchAll into its own @MainActor task. The popover still gets
the benefit of pre-warmed period caches, the refresh loop ticks every 15
seconds regardless.
Also adds a regression test for the midnight-straddle bucketing invariant
that was flagged by the pre-push review: if someone reverts the assistant-
timestamp bucketing back to user-timestamp, this test will catch it.
Reverts the persistent source cache added in 0.8.2 back to v0.8.1's
full-reparse path for Claude sessions; keeps the plan pill, pricing fix,
cursor-agent provider, and the Mac menubar prefetch+timezone work added
between 0.8.2 and 0.8.4.
A turn that straddles midnight (user typed at 23:58, assistant responded
at 00:30) was bucketed and filtered inconsistently across call sites.
parseSessionFile filtered entries by timestamp, producing orphan assistant
calls that groupIntoTurns pushed as turns with empty timestamp. Some
downstream code counted those (buildPeriodData summing project totals)
and other code dropped them (renderStatusBar's empty-timestamp skip).
The menubar showed today = $32 while the terminal status showed today = $27
for the same dataset; each was internally consistent but used a different
turn-bucket rule.
Fix both: parseSessionFile now builds all turns first, then filters each
turn by its first assistant call timestamp (the moment cost was incurred).
renderStatusBar buckets the same way. day-aggregator.ts already bucketed
on assistant time, so it is now consistent too.
Net effect: a turn is counted in the day the API call actually ran in.
cursor-agent was authored on top of the Sharada cache rewrite and referenced
fingerprintPath, cacheStrategy, progressLabel, and parserVersion. With the
persistent source cache reverted, these fields no longer exist on SessionSource.
Strip the references; cursor-agent continues to work on the v0.8.1 discover +
parse path like every other provider.
Loading overlay no longer flashes on every 15s poll. isLoading now only
toggles when the cache is cold, and all periods prefetch once on launch
so tab switching is instant.
Heatmap tooltip, trend bars, forecast, and all-time stats were computing
on UTC dates while the CLI reports on local dates, so the two disagreed
at day boundaries. Switched every date formatter and calendar in these
paths to .current so the menubar matches codeburn today output.
The fallback path in modelDisplayName returned "Auto (Sonnet est.) (est.)"
for any model not listed in modelDisplayNames, double-tagging the est.
suffix and hiding the real model ID. New Cursor model IDs now surface as
their raw name with a single (est.) suffix until the display map is
updated. Adds a regression test.
Discovers transcripts at ~/.cursor/projects/*/agent-transcripts/*.txt
and joins against ~/.cursor/ai-tracking/ai-code-tracking.db for model
attribution. Token counts are estimated from transcript character
length since the attribution DB does not carry them; the model label
surfaces the estimation with an (est.) suffix on every row.
Deduplication keys prefix cursor-agent: to stay disjoint from the
existing cursor: prefix so the two providers do not cross-dedupe
on shared conversationId namespaces.
Tests cover: empty ~/.cursor/projects/, single transcript, multiple
projects, missing ai-code-tracking.db, unrecognized transcript format
skip, non-UUID filename fallback, and sqlite metadata join.
Closes#55
Two pre-existing type errors surfaced during the rebase against main:
1. JsonPlanSummary.id was hardcoded to four plan ids, but PlanId now
includes 'none' (PLAN_IDS was extended when 'codeburn plan clear'
was added). toJsonPlanSummary only runs for active plans at runtime,
but the static type still had to be widened. Use PlanId directly
instead of the hand-rolled union.
2. isActivePlan used Boolean(plan) as the nullish guard, which doesn't
narrow plan's type in TypeScript. Switch to an explicit
'plan !== undefined' so the subsequent .id and .monthlyUsd accesses
type-check.
npx tsc --noEmit is now clean; all 285 tests still pass.
Address review feedback on #74:
1. TUI plan row previously used the active tab's filtered projects as
plan spend, so 'Today' showed today's cost as plan spent. Switch
renderDashboard and reloadData to getPlanUsageOrNull(), which uses
the plan's own billing period regardless of tab.
2. Plan row rendered via a local formatUsd that hardcoded USD. Replace
every call with formatCost so 'codeburn currency EUR' flows through.
Removes the adjacent '$3,425.52' vs '$32.07' style mismatch.
3. renderPlanBar capped filled width at 100%, so 105% and 1700% looked
identical. Past 100%, render a full bar plus chevron tail sized by
order of magnitude (log10): 1.05x -> 1 chevron, 17x -> 2, 170x -> 3.
4. 'running on API overage pricing' is wrong for Claude Pro/Max (rate
limited, not charged overage). Drop that claim; keep the Nx-over
multiplier and match the under/near projection line structure.
5. Spell out 'equiv' as 'API-equivalent' in the plan label.
Dead code cleanup: getPlanUsageOrNullForProjects is now unused; remove
it. getPlanUsageFromProjects stays (unit tests still use it).
Adds `codeburn plan set <id>` to configure a subscription plan (Claude Pro,
Claude Max, Cursor Pro, or custom). When set, the Overview panel renders
an API-equivalent progress bar against subscription price with a
projected month-end cost.
Closes the loudest demand signal on the repo: issue #11 ("Subscription
vs API Use") from two independent voices, plus the routing-decision use
case raised in #12.
- src/config.ts: extends CodeburnConfig with Plan, adds readPlan/savePlan/clearPlan
- src/plans.ts: presets (claude-pro $20, claude-max $200, cursor-pro $20)
- src/plan-usage.ts: getPlanUsage, resetDay-aware period math (1-28),
median-of-7-day-trailing projection
- src/cli.ts: `codeburn plan [show|set|reset]` subcommand, plan wired
into JSON outputs for report/today/month/status (only when active)
- src/dashboard.tsx: Plan row in Overview, color-coded (green under 80%,
orange near, red over), with days-until-reset
- README.md: Plans section with honest framing (API-equivalent vs
subscription price, not token allowance)
- tests/plan-usage.test.ts, tests/plans.test.ts, tests/cli-plan.test.ts:
period math, presets, CLI round-trip
Resets respect resetDay across month boundaries. Uses median daily spend
(not mean) so one huge day doesn't distort the month-end projection.
Fixes#11
- Remove bidirectional fuzzy match in getModelCosts that could return
wrong pricing when a short canonical name prefix-matched a longer key
- Use explicit undefined check in parseLiteLLMEntry so free models with
zero cost are not silently dropped from the LiteLLM pricing database
- Destroy read stream in finally block of readSessionLines to prevent
file descriptor leaks when the generator is abandoned early
- Extend CSV injection escaping to cover tab and carriage-return prefixes
- Add optional chaining fallback for empty periods in exportCsv/exportJson
- Add regression tests for all fixes (models, export, fs-utils)
Brings the PR branch up to the current main so the OMP provider and the
model-alias command can land cleanly. Resolves six merge conflicts and
applies a handful of small fixups alongside the resolution so the
feature matches the conventions set by the cursor-agent merge earlier
today.
Conflict resolutions:
README.md Combine cursor-agent and OMP rows in provider
list, Requirements, and data-location table;
take main's Node 22+ and node:sqlite text.
src/cli.ts Keep both new commands: model-alias and plan.
src/config.ts Add modelAliases alongside plan on the config
type.
src/providers/index.ts Keep the cursor-agent lazy-loader from main
and add omp to coreProviders. Fold the two
pi-module imports into one statement.
src/providers/pi.ts Keep the discovery-cache snapshot path from
main and the providerName parameterization
from the PR. Propagate providerName through
saveDiscoveryCache, loadDiscoveryCache, the
parserVersion tag, and the dedup key prefix
so OMP sources no longer stamp 'pi:' inside
their cache entries or dedup keys.
tests/models.test.ts Keep main's pricing-and-short-name tests and
add the PR's alias tests alongside, sharing a
single loadPricing setup and an afterEach
alias reset.
Fixups in the same commit:
src/models.ts Replace ?? chain in resolveAlias with
Object.hasOwn checks. The previous form
returned Object.prototype for a model named
'__proto__' and broke downstream
canonical.startsWith calls. Caught by the
existing prototype-pollution test suite.
src/providers/pi.ts Use source.provider in the dedup key prefix
and add a trailing newline to the file.
tests/providers/omp.test.ts Expect 'omp:' in the dedup key for OMP
sources, matching the fix above.
Feature work by @cgrossde.
Adds a cursor-agent provider that reads sessions from
~/.cursor/projects/*/agent-transcripts/*.txt and joins against the
ai-code-tracking.db for model attribution, closing issue #55. Token
counts are estimated from transcript character length since neither the
transcript nor the attribution DB carries real counts, and every row
carries an (est.) label to surface that.
Contributed by @mvanhorn.
Loading overlay no longer flashes on every 15s poll. isLoading now only
toggles when the cache is cold, and all periods prefetch once on launch
so tab switching is instant.
Heatmap tooltip, trend bars, forecast, and all-time stats were computing
on UTC dates while the CLI reports on local dates, so the two disagreed
at day boundaries. Switched every date formatter and calendar in these
paths to .current so the menubar matches codeburn today output.
The fallback path in modelDisplayName returned "Auto (Sonnet est.) (est.)"
for any model not listed in modelDisplayNames, double-tagging the est.
suffix and hiding the real model ID. New Cursor model IDs now surface as
their raw name with a single (est.) suffix until the display map is
updated. Adds a regression test.
Source cache fixes (empty-session guards, date range reordering) make
the cache safe for menubar use. Forcing noCache on every 15s poll was
re-parsing 5800+ files each time, causing the menubar to hang.
