CSV / XLSX Converter — CSV, TSV, Excel, JSON and NDJSON in any direction

For anyone who gets a file in the wrong format: a CSV that Excel shows as garbage, an Excel sheet a script cannot read, a JSON export you need as a table, or a scraper dataset that has to become a spreadsheet. Give this Actor file URLs (or an Apify dataset), pick the output format, and get the converted file plus a public download link, with encoding, delimiter and nested JSON handled for you. $0.05 per 1,000 rows converted (rounded up per file). Failed files and the demo are free. No scraping, no API keys.

What it does

• Converts between CSV, TSV, XLSX, JSON and NDJSON: any of them in, any of them out
• Detects the input format from the file extension, content type or content; you can also set it
• Detects the text encoding: UTF-8 (with or without BOM), Shift_JIS and Windows-1252 (cp1252); or choose one. Output is always UTF-8
• Detects the delimiter (comma, semicolon, tab, pipe) or lets you set it; choose the CSV output delimiter and an optional UTF-8 BOM so Excel opens non-English text correctly
• Reads a chosen XLSX sheet by name (first sheet by default) and tells you the sheet names if yours is missing
• Flattens nested JSON into dot-key columns (address.city); arrays stay as JSON text or are joined (a; b; c)
• Selects and renames columns (selecting address keeps every column under it)
• Turns CSV numbers into real numbers (so XLSX and JSON get numbers, not text); ZIP codes with leading zeros stay text
• Safe by default: text starting with =, +, -, @ is stored as plain text in XLSX, never as a formula
• Several files per run; each file gets its own output file and one summary row

Input example

{
  "fileUrls": ["https://raw.githubusercontent.com/mwaskom/seaborn-data/master/iris.csv"],
  "outputFormat": "xlsx",
  "select": ["species", "sepal_length", "petal_length"],
  "rename": {"sepal_length": "sepalLength"}
}

To convert an Apify dataset instead, use datasetIds (pick it from your runs). Empty input runs a free demo on a built-in sample CSV (not charged, rows carry _demo: true).

Output

• Dataset: one row per source file with status, inputFormat, outputFormat, encoding, rows, columns, columnNames, outputKey, outputUrl and, for failures, a plain-language error.
• Key-value store: the converted file, saved under outputKey (for example iris.xlsx). outputUrl is Apify's signed public link, so you can share the file without sharing your token.
• STATS record: files converted and failed, rows converted and the number of 1,000-row units charged.

{
  "source": "https://raw.githubusercontent.com/mwaskom/seaborn-data/master/iris.csv",
  "status": "ok", "inputFormat": "csv", "encoding": "utf-8", "outputFormat": "xlsx",
  "rows": 150, "columns": 3, "columnNames": ["species", "sepalLength", "petal_length"],
  "outputKey": "iris.xlsx", "outputBytes": 7312, "chargedUnits": 1,
  "outputUrl": "https://api.apify.com/v2/key-value-stores/…/records/iris.xlsx?signature=…"
}

Limits (cannot be raised by input)

• 50 MB per source file; 50 files or datasets per run
• 500,000 data rows per file or dataset, 1,000 columns, 100 MB per output file. A file over a limit fails with a clear message and is not charged; rows are never silently cut
• XLSX input: .xlsx / .xlsm only (not old .xls, not password-protected). Formulas are read as their last saved values
• Source URLs must be public http or https links. Addresses on internal networks (localhost, private ranges, cloud metadata) are blocked, also after redirects
• Empty rows are skipped; the first non-empty row is the header (duplicate or empty header names become name_2, column_3)

Pricing

Pay per event: $0.05 per 1,000 rows converted (event rows-1000). Rows are counted per file and rounded up: a 2,500-row file costs 3 units ($0.15), a 150-row file 1 unit ($0.05). A file that fails (download error, bad format, over a limit) is not charged. Set a maximum charge per run to cap spend; the Actor stops cleanly before the next file when it is reached.

Use cases

• Re-save a Shift_JIS or Windows-1252 CSV as UTF-8 so Excel, Sheets and databases show the text correctly
• Turn a CSV export into an XLSX workbook with real numbers, or an XLSX sheet into NDJSON for a script
• Convert a nested JSON API export into a flat table for a spreadsheet
• Hand a scraper dataset to a colleague as an Excel file with a download link

FAQ

Which encoding does auto choose? UTF-8 first. If that fails, Shift_JIS when the text contains Japanese kana, otherwise Windows-1252. The detected encoding is in the result row; set encoding if it guesses wrong. Why is my ID still text? Numbers with leading zeros, more than 15 digits or exponents stay text on purpose. Turn inferTypes off to keep every CSV value as text. Excel shows broken characters in my CSV. Turn on csvBom. Does it change my source? No. It only reads the file or dataset and writes a new file. Is the file link private? Anyone holding the signed link can download that file; it gives no access to anything else in your account.

More tools from the same developer

• Dataset Transform — flatten, filter, rename and convert any Apify dataset, export CSV / JSON / JSONL with a link
• Dataset Merge & Dedup — merge Apify datasets and remove duplicates by the fields you choose
• Google Sheets Import — read Google Sheet rows into a dataset as JSON/CSV, public link or your service account
• Google Sheets Export — export an Apify dataset to Google Sheets
• PDF Extractor — PDF to text, tables and metadata from public PDF URLs, page by page