23 Sep
2020
23 Sep
'20
1:54 p.m.
Vinnie Falco wrote:
I had that originally, it was called number_parser, it was a public interface, and it was one of the first things to go during our 3-month journey of optimization.
I suspect that something similar would have happened to the hypothetical pull parser lowest level, if there were one. It seems to me that the pull parser would be very similar to the current parser, except it would need to suspend on each token, instead of on each buffer boundary. This probably has a performance cost.