Skip to content

Releases

csv-to-signal v0.5.0

Choose a tag to compare

@internalforces internalforces released this 23 Aug 08:13
dffdf6a

CSV to Signal 0.5.0 adds GitHub repository and CoinGecko market-data analysis to the existing deterministic CSV workflow.

Highlights:

  • Add github / command with optional GITHUB_TOKEN authentication.
  • Add coingecko command with optional COINGECKO_DEMO_API_KEY authentication.
  • Preserve analyze <file.csv>, JSON output, SQLite schema, and deterministic rerun behavior.
  • Refresh mutable external snapshots and isolate provider storage namespaces.
  • Support Node.js 22 and 24.

Install:
npm install --global csv-to-signal@0.5.0

Verified from merge dffdf6a on Node 22.22.3 and Node 24.19.0. The published four-file npm artifact matches the retained release tarball.

csv-to-signal v0.4.0

Choose a tag to compare

@internalforces internalforces released this 22 Aug 05:03
9b98ec9

Highlights

  • Supports Node.js 22 and 24; Node.js 20 is no longer supported.
  • Upgrades better-sqlite3 to 13.0.3 with N-API prebuilt binaries.
  • Preserves the existing CSV-to-signal CLI behavior and output format.

Install

npm install -g csv-to-signal@0.4.0

Published package: https://www.npmjs.com/package/csv-to-signal/v/0.4.0

Verification

  • Git tag commit: 9b98ec9
  • SHA-256: db1edfe4f79762ad616bdffbd53c663b81b405868ae62ddec6fdffa2130109ce
  • npm integrity: sha512-HEE2cDmU7Zz+NA35dzMMMliOb2y63mdKrQgGtPGyIguGn7EPbFHUKhcwG42gMUJkJ4xsRPnAoDNdG66I3cUkZQ==

csv-to-signal v0.3.0

Choose a tag to compare

@internalforces internalforces released this 08 Aug 03:18
59ec92e

CSV to Signal 0.3.0

This release exposes deterministic windowed change analysis through the CSV CLI.

Added

  • --window-hours <n> adds a WindowedChangeDetector alongside the default consecutive
    percentage-change detector and optional threshold detector.
  • Positive finite hour validation with the existing last-value-wins flag behavior.
  • Source-level, bundled-executable, package, and installed-consumer coverage for windowed analysis.

Compatibility

  • Existing commands without --window-hours keep their previous behavior.
  • The signal JSON shape, SQLite schema, database location, package name, and executable name are
    unchanged.
  • GitHub and CoinGecko connectors remain private workspace libraries and are not CLI commands.

Verification

  • Released from merge commit 59ec92e37dbd11226391f8eef59965b6821f8023.
  • Build, 90 tests, typecheck, full and production dependency audits passed.
  • The published four-file npm tarball is 8,906 bytes with integrity
    sha512-k1z2wk1Ub+9QE0yHLOv2iLBJCGLIhFnW7zTO1PcN4FNhTuvP0e0M5VSv7yA0CPZN3y7MGmy0nayMgR9JrExa6Q==.
  • A clean registry consumer installed csv-to-signal@0.3.0, ran --window-hours 24, produced the
    expected windowed signal, and created data.db in the consumer working directory.

Install from npm: https://www.npmjs.com/package/csv-to-signal

csv-to-signal v0.2.1

Choose a tag to compare

@internalforces internalforces released this 07 Aug 02:21
a3a0069

Highlights

  • First public npm release under the csv-to-signal package and executable name.
  • Deterministic CSV time-series analysis with ranked JSON output, SQLite persistence, score filtering, and optional threshold-crossing signals.
  • Standalone four-file npm package validated on supported Node.js 20, 22, and 24 release lines.

Install

npm install --global csv-to-signal

Package: https://www.npmjs.com/package/csv-to-signal

Verification

  • Tag v0.2.1 points to merge commit a3a00696d5526ea788199df2c1a3e1ce6a4217e3 (PR #11).
  • npm integrity: sha512-2yy8IYlFEohj3KxTJuG7JcHTrkU4yh5QTPClJQNXBazQ3QnFNj2YtwyaaDdi1F5IfNZiqzjt7oEVoWK3V+Ustg==
  • v0.2.0 remains the unpublished candidate for the npm-rejected signal-hub identity.

Known documentation issue

ISS-018: the immutable npm 0.2.1 tarball README still contains outdated pre-release wording. The source README is corrected, and the correction will be included in the next separately approved patch release. A 0.2.2 release will not be rushed solely for this low-impact wording issue.