/** * Archive check: reads the report attached to one or more Faunosteo PDFs and * prints it next to its v3 reading on the bundled reference sheet. * * pnpm report:compare ... [--summary] [--identity] * [--locale fr|en] [--diff a.pdf b.pdf] * * A v2 archive PDF is shown legacy id → canonical id, line by line; a v3 PDF * is checked against the sheet. Items the sheet cannot place are marked * `!! UNRESOLVED` with the reason. `--diff` compares the v3 content of two * PDFs (an archive and its re-export after import, typically). Names are * printed only with `--identity`. Reads only; never writes a file. */ import { readFileSync, readdirSync, statSync } from 'node:fs' import { basename, join, resolve } from 'node:path' import { fileURLToPath } from 'node:url' import { PDFDocument } from 'pdf-lib' import { readReportDocument, REPORT_VERSION, UnsupportedReportVersionError, } from '@faunosteo/domain' import { readAttachedDocument } from '../src/pdf/attachment' import { MAX_PDF_BYTES } from '../src/pdf/import' import { parseBundledSheet } from '../src/sheets/provider' import { buildSheetIndex, type SheetIndex } from '../src/services/sheetIndex' import { compareReport, diffLines, parseArgs, renderComparison, renderSummary, type CliArgs, type Comparison, } from './compare/report' const SEED_DIR = resolve( fileURLToPath(import.meta.url), '../../../../supabase/seed/anatomical-sheets/equine/draft', ) /** pnpm runs the script from apps/app; paths are typed from the caller's folder. */ const CWD = process.env.INIT_CWD ?? process.cwd() function loadSheet(): SheetIndex { const json = (name: string) => JSON.parse(readFileSync(join(SEED_DIR, name), 'utf8')) as never // The comparison needs the document and the map, not the drawings. return buildSheetIndex( parseBundledSheet({ status: 'draft', ref: json('sheet-ref.json'), document: json('document.json'), migrationMap: json('report-migration-map.json'), assets: {}, }), ) } function expandFiles(args: string[]): string[] { const out: string[] = [] for (const arg of args) { const path = resolve(CWD, arg) if (statSync(path).isDirectory()) { const pdfs = readdirSync(path) .filter((f) => f.toLowerCase().endsWith('.pdf')) .sort() out.push(...pdfs.map((f) => join(path, f))) } else out.push(path) } return out } async function compareFile( path: string, sheet: SheetIndex, args: CliArgs, ): Promise { const bytes = new Uint8Array(readFileSync(path)) if (bytes.byteLength > MAX_PDF_BYTES) throw new Error('file too large') const pdf = await PDFDocument.load(bytes, { ignoreEncryption: true }) const raw = await readAttachedDocument(pdf) const version = (raw as { version: number }).version try { const outcome = readReportDocument(raw, sheet.sheet.migrationMap) return compareReport(raw, outcome, version !== REPORT_VERSION, sheet, args) } catch (err) { if (err instanceof UnsupportedReportVersionError) { throw new Error(`report version ${version} cannot be read`) } throw err } } async function main(argv: string[]): Promise { const args = parseArgs(argv) const files = expandFiles(args.files) if (files.length === 0 || (args.diff && files.length !== 2)) { console.error( 'usage: pnpm report:compare ... [--summary] [--identity] [--locale fr|en] [--diff a.pdf b.pdf]', ) return 1 } const sheet = loadSheet() const results: Comparison[] = [] let failures = 0 for (const file of files) { const name = basename(file) try { const c = await compareFile(file, sheet, args) results.push(c) if (!args.diff) { const lines = args.summary ? [renderSummary(name, c)] : [...renderComparison(name, c), ''] console.log(lines.join('\n')) } } catch (err) { failures += 1 console.log(`!! ${name}: ${err instanceof Error ? err.message : err}`) } } if (args.diff && results.length === 2) { const diff = diffLines(results[0]!.normalized, results[1]!.normalized) console.log(diff.length ? diff.join('\n') : 'identical v3 content') } return failures ? 2 : 0 } main(process.argv.slice(2)).then( (code) => process.exit(code), (err) => { console.error(err instanceof Error ? err.message : err) process.exit(1) }, )