Having trouble decoding url-encoded base64

[marshallpierce]

Having trouble decoding url-encoded base64

[marshallpierce]

This is correct behavior, actually. The issue is that URL_SAFE base64 is not the same as URL encoding/decoding.

URL encoding is what maps a space to %20, / to %2F, etc, and is applied, piece by piece, when assembling a URL. This is necessary to be able to represent a path containing a / without interpreting the / as the separator between path components, and other such issues.

URL-safe Base64 is a Base64 variant that uses - and _ instead of + and / to avoid any further transformation when being used in a URL.

So, what you have above is a complete URL that includes as part of it a URL-encoded non-url-safe Base64 section. You can see it's been URL encoded because it has things like %2F (hex 0x2F is the ASCII encoding of /) and %2B (which would be +).

More specifically, that's a URL with a query string composed of query key/value pair, and the part you want is the query value. The RFCs around URL encoding are a mess, and the conventions used by HTML forms go beyond what the base URI/URL RFCs do (e.g. using + to represent -- what if you really wanted a plus, not a space?), so there isn't a single answer for how to parse or construct these things, but basically what you want to do is get to this part: payload=FZkVZv, extract the FZkVZv... part, URL-decode that (undo the % encoding), strip whitespace (which base64 can optionally do for you), and then base64 decode the result with base64::STANDARD.