Regex to Match Base64 Encoded Strings

This regex validates strings encoded in standard Base64 format according to RFC 4648. Here is the token-by-token breakdown:

^ — Anchors the match at the start of a line, ensuring the entire string is validated from the beginning.

(?:[A-Za-z0-9+/]{4})* — A non-capturing group that matches zero or more complete 4-character Base64 blocks. The character class [A-Za-z0-9+/] contains the 64 characters of the standard Base64 alphabet: uppercase letters, lowercase letters, digits, plus sign, and forward slash. The {4} quantifier requires exactly four characters per block, and the * allows this to repeat for the full-length encoded data.

(?: — Opens a non-capturing group for the optional padding section.

[A-Za-z0-9+/]{2}== — First alternative: matches the final block when only one byte of data remains, producing two Base64 characters followed by two equals-sign padding characters.

| — Alternation operator.

[A-Za-z0-9+/]{3}= — Second alternative: matches the final block when two bytes of data remain, producing three Base64 characters followed by one equals-sign padding character.

)? — Makes the padding section optional, since some encoded strings have lengths that are exact multiples of three bytes and need no padding.

$ — Anchors the match at the end of a line.

The m flag enables multiline mode. This pattern validates that a string conforms to Base64 encoding rules including proper padding. It is useful for input validation, data format detection, and API request verification.