|
|
|
|
|
by mdaniel
1477 days ago
|
|
Congratulations on your launch! > This presented some character escaping issues, but those were manageable. In case you were interested, there's a section devoted to that in RFC 3548: https://datatracker.ietf.org/doc/html/rfc3548#section-4 with the tl;dr of `s/[+]/-/; s/[/]/_/` and then omitting the trailing "=" characters and just padding the incoming b64 since (AFAIK) one can add as many "=" as they'd like and it'll just ignore extra ones Depending on your interoperability interests, you can shave a bit of text out of it by shortening the JSON key names `"i"`, `"a"`, `"n"`, and then also `.filter(Boolean)` to nuke those two trailing empty strings |
|
> In case you were interested, there's a section devoted to that in RFC 3548: https://datatracker.ietf.org/doc/html/rfc3548#section-4 with the tl;dr of `s/[+]/-/; s/[/]/_/` and then omitting the trailing "=" characters and just padding the incoming b64 since (AFAIK) one can add as many "=" as they'd like and it'll just ignore extra ones
Thanks! I'll check this out.
> Depending on your interoperability interests, you can shave a bit of text out of it by shortening the JSON key names `"i"`, `"a"`, `"n"`, and then also `.filter(Boolean)` to nuke those two trailing empty strings
This is a good idea. I might already be filtering those empty strings but I'll have to check.