Base64 Encoding in Python and JavaScript

JavaScript: btoa, atob, and Their Limitations

btoa(str) encodes a string; atob(str) decodes it. The critical limitation: btoa only works with 'binary strings' — strings where every character's code point is 0-255. For true binary data (ArrayBuffer, Uint8Array), convert to a binary string first. For Unicode strings with high code points, btoa throws — use TextEncoder to convert to UTF-8 bytes first.

Python: The base64 Module

Python's base64 module is clear and complete: base64.b64encode(bytes_obj) returns bytes; b64decode(str_or_bytes) returns bytes. URL-safe variants use urlsafe_b64encode and urlsafe_b64decode. Always encode bytes, not strings — call str.encode('utf-8') first.

Cross-Language Interoperability

Base64 is a standard, so encoded data from one language decodes correctly in another — as long as you use the same variant (standard vs URL-safe) and handle padding consistently. The most common interoperability bug is URL-safe Base64 produced by one language decoded with a standard decoder in another.

Common Encoding Pitfalls

Double-encoding (encoding an already-encoded string) produces garbage output. Missing character set normalization when encoding text strings corrupts multi-byte characters. Always track where encoding happens and where decoding happens in your data flow to prevent these issues.