init

convert-xlsx-to-csv.ts

@@ -0,0 +1,91 @@
+import XLSX from "xlsx";
+import { mkdir, readdir, writeFile } from "node:fs/promises";
+import path from "node:path";
+
+const INPUT_DIR = process.env.BIZMAX_OUTPUT_DIR ?? path.resolve(process.cwd(), "output");
+const CSV_DIR = process.env.BIZMAX_CSV_DIR ?? path.join(INPUT_DIR, "csv");
+const CSV_NAME = process.env.BIZMAX_CSV_NAME ?? "bizmax-all.csv";
+
+function toText(value: unknown): string {
+  if (value == null) {
+    return "";
+  }
+  return String(value);
+}
+
+function escapeCsv(value: string): string {
+  if (/[",\n\r]/.test(value)) {
+    return `"${value.replace(/"/g, "\"\"")}"`;
+  }
+  return value;
+}
+
+function rowToCsv(row: string[]): string {
+  return row.map(escapeCsv).join(",");
+}
+
+async function main(): Promise<void> {
+  const entries = await readdir(INPUT_DIR, { withFileTypes: true });
+  const xlsxFiles = entries
+    .filter((entry) => entry.isFile() && entry.name.toLowerCase().endsWith(".xlsx"))
+    .map((entry) => path.join(INPUT_DIR, entry.name))
+    .sort();
+
+  if (xlsxFiles.length === 0) {
+    throw new Error(`No .xlsx files found in ${INPUT_DIR}`);
+  }
+
+  const outputRows: string[] = [];
+  let wroteHeader = false;
+
+  for (const xlsxFile of xlsxFiles) {
+    const workbook = XLSX.readFile(xlsxFile, { cellDates: false });
+
+    for (const sheetName of workbook.SheetNames) {
+      const worksheet = workbook.Sheets[sheetName];
+      const rows = XLSX.utils.sheet_to_json<unknown[]>(worksheet, {
+        header: 1,
+        raw: false,
+        defval: "",
+      });
+
+      if (rows.length === 0) {
+        continue;
+      }
+
+      const headerRow = (rows[0] ?? []).map(toText);
+      if (!wroteHeader) {
+        outputRows.push(
+          rowToCsv(["source_file", "source_sheet", ...headerRow]),
+        );
+        wroteHeader = true;
+      }
+
+      for (const row of rows.slice(1)) {
+        const values = (row as unknown[]).map(toText);
+        outputRows.push(
+          rowToCsv([
+            path.basename(xlsxFile),
+            sheetName,
+            ...values,
+          ]),
+        );
+      }
+    }
+  }
+
+  if (!wroteHeader) {
+    throw new Error(`No worksheet rows found in ${INPUT_DIR}`);
+  }
+
+  await mkdir(CSV_DIR, { recursive: true });
+  const outputPath = path.join(CSV_DIR, CSV_NAME);
+  await writeFile(outputPath, `${outputRows.join("\n")}\n`, "utf8");
+  process.stdout.write(`${outputPath}\n`);
+}
+
+main().catch((error: unknown) => {
+  const message = error instanceof Error ? error.message : String(error);
+  process.stderr.write(`${message}\n`);
+  process.exitCode = 1;
+});