scripts/refresh.mjs

#!/usr/bin/env node
// CalcFi Open Data — refresh script
// Pulls every series from calcfi.app/api/llms/rates manifest, fetches
// per-series history, writes CSV + datapackage.json + README per subdir.
// Idempotent. Safe to run on cron.

import { writeFile, mkdir, readFile } from 'node:fs/promises'
import { existsSync } from 'node:fs'
import { dirname, join } from 'node:path'
import { fileURLToPath } from 'node:url'

const __dirname = dirname(fileURLToPath(import.meta.url))
const ROOT = join(__dirname, '..')
const DATASETS = join(ROOT, 'datasets')

const MANIFEST_URL = 'https://calcfi.app/api/llms/rates'
const HISTORY_CSV = (slug) => `https://calcfi.app/api/data/${slug}/history.csv`

const SCHEMA_FOR_UNIT = {
  percent: { type: 'number', description: 'Value as a percent (e.g. 6.36 means 6.36%).' },
  index: { type: 'number', description: 'Index value (base period varies; see source).' },
  usd: { type: 'number', description: 'Value in US dollars.' },
  usd_per_barrel: { type: 'number', description: 'US dollars per barrel.' },
  usd_per_gallon: { type: 'number', description: 'US dollars per gallon.' },
  ratio: { type: 'number', description: 'Exchange rate (units of currency A per 1 unit of currency B).' },
}

const PROVIDER_NAME = {
  fred: 'Federal Reserve Economic Data (FRED)',
  fdic: 'Federal Deposit Insurance Corporation (FDIC)',
  coingecko: 'CoinGecko',
  worldbank: 'World Bank',
  bls: 'Bureau of Labor Statistics (BLS)',
  bea: 'Bureau of Economic Analysis (BEA)',
  eia: 'Energy Information Administration (EIA)',
  treasury: 'US Treasury',
}

async function fetchJson(url) {
  const res = await fetch(url, { headers: { 'User-Agent': 'calcfi-open-data/1.0 (+https://calcfi.app/developers)' } })
  if (!res.ok) throw new Error(`${res.status} ${url}`)
  return res.json()
}

async function fetchText(url) {
  const res = await fetch(url, { headers: { 'User-Agent': 'calcfi-open-data/1.0 (+https://calcfi.app/developers)' } })
  if (!res.ok) throw new Error(`${res.status} ${url}`)
  return res.text()
}

function csvEscape(v) {
  if (v == null) return ''
  const s = String(v)
  return /[",\n]/.test(s) ? `"${s.replaceAll('"', '""')}"` : s
}

function buildCsv(rows, columns) {
  const lines = [columns.join(',')]
  for (const r of rows) lines.push(columns.map((c) => csvEscape(r[c])).join(','))
  return lines.join('\n') + '\n'
}

function buildDatapackage(rate, columns, schemaCols, today) {
  const provider = PROVIDER_NAME[rate.source.provider] || rate.source.provider
  return {
    name: rate.slug,
    title: rate.name,
    description: `${rate.name}. Source: ${rate.source.name}. Live current value at https://calcfi.app/rates/history/${rate.slug}. Cadence: ${rate.cadence}. Unit: ${rate.unit}.`,
    homepage: `https://calcfi.app/rates/history/${rate.slug}`,
    licenses: [
      {
        name: 'CC-BY-4.0',
        path: 'https://creativecommons.org/licenses/by/4.0/',
        title: 'Creative Commons Attribution 4.0 International',
      },
    ],
    keywords: ['finance', 'calcfi', rate.category, rate.source.provider, rate.unit].filter(Boolean),
    sources: [
      { title: provider, path: rate.source.url || `https://fred.stlouisfed.org/series/${rate.source.series_id}` },
      { title: 'CalcFi live page', path: `https://calcfi.app/rates/history/${rate.slug}` },
    ],
    resources: [
      {
        name: 'data',
        path: 'data.csv',
        format: 'csv',
        mediatype: 'text/csv',
        encoding: 'utf-8',
        schema: { fields: schemaCols },
      },
    ],
    version: today,
  }
}

function buildReadme(rate) {
  const provider = PROVIDER_NAME[rate.source.provider] || rate.source.provider
  return `# ${rate.name}

${rate.cadence === 'daily' ? 'Daily' : rate.cadence === 'weekly' ? 'Weekly' : rate.cadence === 'monthly' ? 'Monthly' : rate.cadence === 'quarterly' ? 'Quarterly' : 'Periodic'} observations of **${rate.name}** (${rate.short_name}).

## At a glance

| Field | Value |
|---|---|
| Slug | \`${rate.slug}\` |
| Category | ${rate.category} |
| Unit | ${rate.unit} |
| Cadence | ${rate.cadence} |
| Source | [${rate.source.name}](${rate.source.url}) |
| Source series ID | \`${rate.source.series_id}\` |
| License | [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) |

## Live data on CalcFi

- **Latest value + chart:** https://calcfi.app/rates/history/${rate.slug}
- **JSON API (no auth):** https://calcfi.app/api/rates/${rate.slug}
- **CSV (this file, daily-refreshed via cron):** [data.csv](data.csv)

## Quick start

\`\`\`bash
# Latest single value from the API
curl https://calcfi.app/api/rates/${rate.slug}

# Full history from this repo
curl https://raw.githubusercontent.com/jerehere/calcfi-open-data/main/datasets/${rate.slug}/data.csv
\`\`\`

\`\`\`python
import pandas as pd
url = "https://raw.githubusercontent.com/jerehere/calcfi-open-data/main/datasets/${rate.slug}/data.csv"
df = pd.read_csv(url, parse_dates=["date"])
print(df.tail())
\`\`\`

## Source + provenance

This dataset mirrors **${rate.source.name}** (\`${rate.source.series_id}\`).
Primary source: ${provider}.

Values are pulled daily by a GitHub Actions workflow (\`.github/workflows/refresh.yml\`) from CalcFi's read-only data API, which itself pulls from primary sources. No transformations or imputations are applied; values are passed through verbatim.

## Citation

If you use this dataset, please cite:

> CalcFi (${new Date().getFullYear()}). ${rate.name} dataset. https://calcfi.app/rates/history/${rate.slug}

## License

CC BY 4.0 — attribution requested. The data itself is sourced from ${provider}; consult their terms for any additional restrictions.
`
}

async function processRate(rate) {
  const slug = rate.slug
  const dir = join(DATASETS, slug)
  await mkdir(dir, { recursive: true })

  const schemaCols = [
    { name: 'date', type: 'date', format: '%Y-%m-%d', description: 'Observation date (YYYY-MM-DD).' },
    { name: 'value', ...(SCHEMA_FOR_UNIT[rate.unit] || { type: 'number', description: 'Observed value.' }) },
    { name: 'unit', type: 'string', description: 'Unit of observation (matches manifest).' },
  ]

  let rowCount = 0
  try {
    const csv = await fetchText(HISTORY_CSV(slug))
    await writeFile(join(dir, 'data.csv'), csv)
    rowCount = csv.split('\n').filter((l) => l && !l.startsWith('#') && !l.startsWith('date,')).length
  } catch (e) {
    console.warn(`  [WARN] ${slug}: history.csv unavailable (${e.message}); writing manifest-only snapshot`)
    if (rate.value != null && rate.observation_date) {
      const csv = `# ${rate.name}\n# Source: ${rate.source.name}\n# Snapshot from manifest only — history endpoint unavailable\ndate,value,unit\n${rate.observation_date},${rate.value},${rate.unit}\n`
      await writeFile(join(dir, 'data.csv'), csv)
      rowCount = 1
    } else {
      await writeFile(join(dir, 'data.csv'), `date,value,unit\n`)
    }
  }

  const today = new Date().toISOString().slice(0, 10)
  await writeFile(join(dir, 'datapackage.json'), JSON.stringify(buildDatapackage(rate, ['date', 'value', 'unit'], schemaCols, today), null, 2) + '\n')
  await writeFile(join(dir, 'README.md'), buildReadme(rate))

  console.log(`  [OK] ${slug.padEnd(32)} ${rowCount} rows`)
  return { slug, rows: rowCount }
}

async function main() {
  console.log(`Fetching manifest: ${MANIFEST_URL}`)
  const manifest = await fetchJson(MANIFEST_URL)
  console.log(`  ${manifest.count} series in manifest\n`)

  const summary = []
  for (const rate of manifest.rates) {
    try {
      const res = await processRate(rate)
      summary.push(res)
    } catch (e) {
      console.error(`  [FAIL] ${rate.slug}: ${e.message}`)
      summary.push({ slug: rate.slug, rows: 0, error: e.message })
    }
  }

  const indexPath = join(ROOT, 'datasets', 'INDEX.json')
  await writeFile(indexPath, JSON.stringify({ generated_at: new Date().toISOString(), datasets: summary }, null, 2) + '\n')
  console.log(`\nWrote ${summary.length} datasets. Index: datasets/INDEX.json`)
}

main().catch((e) => {
  console.error(e)
  process.exit(1)
})