Stephen Frost <[EMAIL PROTECTED]> writes:
> * Tom Lane ([EMAIL PROTECTED]) wrote:
>> Right offhand I like the idea of pushing it into connectOptions2 --- can
>> you experiment with that?  Seems like there is no reason to call
>> Kerberos if the user supplies the name to connect as.

> Patch attached.  After looking through the code around this I discovered
> that conninfo_parse is also called by PQconndefaults.  This patch
> changes the 'default' returned for user to NULL (instead of whatever
> pg_fe_getauthname returns).

This is probably not a good idea --- changing the API behavior in
pursuit of saving a few cycles is just going to get people mad at us.

After looking more closely, I see that there is inefficiency only in the
PQsetdbLogin case --- in the PQconnectdb case, we only bother to compute
"fallback" values for parameters where they're actually needed.  I think
that at the time this code was last restructured, the assumption was
that PQsetdbLogin would die soon and there wasn't any need to worry
about whether it wastes a few cycles.  If you don't subscribe to that
argument, probably the best answer is to keep the fallback-computation
loop (fe-connect.c lines 2681-2735 in CVS tip) more or less as-is, but
split it out from conninfo_parse, and somehow arrange for PQsetdbLogin
to be able to insert the values that it wants before the fallbacks are
computed.  The trick here is that PQsetdbLogin wants to work on a PGconn
not a PQconninfoOption array, and that isn't going to work nicely.

I think we'd have to refactor the code so that PQsetdbLogin gets a
PQconninfoOption array, overrides values *in that array*, then calls the
fallback-substitution code etc.  Not sure if it's worth the trouble.
The extra complexity of searching the array for values to override could
eat up the cycles we're hoping to save, too :-(

                        regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 4: Have you searched our list archives?


Reply via email to