Quoted-Printable Encoder & Decoder

Quoted-Printable limits lines to 76 characters. When a logical line is longer, the encoder ends a physical line with a lone '=' and continues on the next line; that trailing '=' plus the line break is a soft line break and represents no data. On decode it's removed, rejoining the line. A real line break in the text is left as an actual line break.

A literal space or tab is allowed in the middle of a line, but not at the end — trailing whitespace can be silently stripped by mail servers. So the encoder escapes a space at the end of a line as =20 and a tab as =09, guaranteeing it survives. In the middle of a line spaces are left as-is for readability.

Yes. On encode, each byte of the UTF-8 representation that isn't printable ASCII becomes a =XX escape, so '€' becomes =E2=82=AC and 'café' becomes caf=C3=A9. On decode the bytes are collected and interpreted as UTF-8, so the original text comes back exactly. The encoding assumes UTF-8, which is the modern default.

Both are MIME transfer-encodings, but Quoted-Printable keeps mostly-ASCII text human-readable — only the unusual bytes are escaped — whereas Base64 turns everything into opaque blocks. Quoted-Printable is best for text that's almost all ASCII; Base64 is better for binary or heavily non-ASCII data.