Jack Christensen

@mchandler-plato Take a look at https://github.com/jackc/pgx/pull/2422. I suspect that would make a much bigger impact on network usage than short statement names. Not sure if I'm willing to accept the...

> Interesting PR, is describe needed for the internal workings of the driver? Yes. pgx uses the result column types as part of the scanning process. > Also I wonder...

This restriction is due to the difficulty of safely interpolating parameters into the query string when different encodings are used. While in theory I am not opposed to lifting this...

@LKaemmerling Yes, if it can be cleanly resolved I'm interested. However, there might be a bit of a mismatch between how pgx uses fallback configs and ideal `connect_timeout` and `target_session_attrs`...

Can you use a normal `time.Time`? pgx doesn't know anything about "arrow" types.

Regarding the Go version, 1.24 is due any day now so that wouldn't be a problem. It would also would have been possible to use build tags to provide backwards...

> I personally don't understand how to use ForEachRow to scan rows into structures of.. say 20 fields, without introducing templating engine and codegeneration (which sound odd to me by...

> I personally don't understand how to use ForEachRow to scan rows into structures of.. say 20 fields, without introducing templating engine and codegeneration (which sound odd to me by...

Your `ForScannedRows` is the idea of what I was thinking of. Returning from the outer scope does seem to be the major functional difference. But the error handling inside the...

I've reviewed this again, and I'm starting to lean toward we will want iterators in pgx eventually, but I'm not ready to commit to any particular interface yet. My reluctance...