Simplify base64URLStringToBuffer padding #566
Conversation
| * (4 - (87 % 4 = 3) = 1) % 4 = 1 padding | ||
| * (4 - (88 % 4 = 0) = 4) % 4 = 0 padding | ||
| */ | ||
| const padLength = (4 - (base64.length % 4)) % 4; |
There was a problem hiding this comment.
I actually like padEnd approach better because padEnd is there so you don't have to write a while loop and avoid multiple assignments and multiple concatenations
the only thing I'd change there is I would remove the last % 4
I do realize though that impact here is minuscule as the loop runs at most two times
|
|
||
| // Pad with '=' until it's a multiple of four | ||
| while (base64.length % 4 != 0) { | ||
| base64 += "="; |
There was a problem hiding this comment.
Actually, did you know that atob() works with un-padded base64 strings? check this out:
> btoa("padding")
'cGFkZGluZw=='
> atob("cGFkZGluZw")
'padding'
because if it's trivial to calculate amount of padding needed, it's trivial to change decoding logic to not need any padding.
I'd say it's not really needed here
|
Thank you for your effort @qalandarov but this project is currently not open to external contributions as per the README. If you believe you have found an issue or identified an optimization then please open a new issue to begin discussing it. |
|
I just gotta ask, is this because of |
|
It's typically because reviewing PRs from external contributors takes time and energy and isn't fun for me when I do so much of that as my day job. And sometimes, if I don't feel like reviewing something right away, I think it can seem disrespectful for making the potential contributor wait when really I just haven't found time to thoroughly review the content of their PR. |
No description provided.