Encoding is a byte-to-text mapping
UTF-8 handles multilingual text well. Older files may use Windows-1252 or ISO-8859-1. They differ around byte values 128–159, so selecting the wrong one can change punctuation. UTF-16 stores 16-bit units in little-endian or big-endian order; a BOM can identify the order.
Review before saving
Changing the source encoding rereads the original bytes and can replace current edits after confirmation. Output encoding affects only the downloaded copy. If a character cannot fit in a legacy encoding, this tool stops the download instead of silently replacing it. Keep UTF-8 when your application supports it.
Line endings
LF is common on Unix systems and CRLF on Windows. CR is also available for legacy files. The tool normalizes the working text to LF and converts it on download. Mixed original endings are not preserved. BOM is optional, and is unavailable for single-byte encodings.