That's the problem of all CSV formats, wrong or custom implementations. Someone will generate wrong usv e.g. escape or encapsulate values. And if you get that

That's the problem of all CSV formats, wrong or custom implementations. Someone will generate wrong usv e.g. escape or encapsulate values. And if you get that wrong format no one cares if they gave you a wrong format. => Your parser grows and grows...