20 Mar
2024
20 Mar
'24
11:40 a.m.
FWIW the boost serialization library has included a module for handling this for over 20 years.
I've been looking at the implementation and it wouldn't fit my use case (parsing P-H-C strings as per https://github.com/P-H-C/phc-string-format/blob/master/phc-sf-spec.md) because: * It's hardwired to use exceptions. My use case is a parser for a data format with several base64 fields. It's not exceptional for them to be malformed, so I'd rather use error codes. * The format I'm parsing mandates to not output padding characters. I don't see any configuration option in the documentation to disable these. How does serialization compare in performance when doing base64 to other libraries?