Hi! On Mon, May 6, 2019 at 1:21 PM Tom Lane <tgl@xxxxxxxxxxxxx> wrote: > Getting us to deviate from the RFC so blatantly would be a very hard sell. > A large part of the point of the JSON datatype is to be interoperable; > once you give that up you may as well use some not-standard-at-all > representation. Python supports that, enabled by default: https://docs.python.org/3/library/json.html#infinite-and-nan-number-values > There is not, and never has been, any claim that JSON numbers correspond > to the IEEE spec. There is note [1], but yes, it does not claim that nor I claimed that. I am just saying that the reality is that most people these days use IEEE spec floating numbers so it is sad that those cannot be easily stored in JSON, or a database. Mitar [1] https://tools.ietf.org/html/rfc7159#page-7 -- http://mitar.tnode.com/ https://twitter.com/mitar_m
