Hi, > > This was briefly brought up in the discussion that ultimately led to > > protocol.h [0]. I don't have a particularly strong opinion on the matter, > > but I will note that protocol.h was intended to be easily usable in > > third-party code, and changing it from characters to an enum from v17 to > > v18 might cause some extra code churn. > > We could avoid that issue by back-patching into 17; I don't think > it's quite too late for that, and the header is new as of 17. > > However, I'm generally -1 on the proposal independently of that, > > [...] > > (This applies only to the s/char/enum/ proposal; I've not read > the patchset further than that.)
That's fair enough. Also it's not clear how much type safety enums give us exactly. E.g. after applying 0002 pq_putmessage() a.k.a. PQcommMethods->putmessage() silently casts its first argument from `enum PqMsg` to `char` (shrug). -- Best regards, Aleksander Alekseev
