pá 9. 11. 2018 v 6:57 odesílatel Michael Paquier <michael@paquier.xyz> napsal:
On Thu, Nov 08, 2018 at 01:58:34PM +0900, Michael Paquier wrote: > Anyway, I am still going through the patch, so no need to send a new > version for now.
Okay, I have done a round of more in-depth review, and the patch looks to be in pretty good shape.
Relying on tuples_only to decide if the header should be printed or not looks good to me.
+ /* check for value being non-empty and with an MB length of 1 */ + if (*value == '\0' || value[PQmblen(value, pset.encoding)] != '\0')
It seems to me that this can just be replaced with that: if (strlen(value) != 1)
Attached is what I am finishing up with for the moment. Comments are welcome. I am still planning look at that stuff a bit more once again, particularly the printing part, but I am lacking of time now..
