Re: fix regression in .pgpass handling

Neil, thanks for the fix. I saw PQsetdbLogin() calling connectDBStart()
and thought that was the same as PQconnectStart(), where
PasswordFromFile() was already handled; especially thanks for fixing
this befor 7.3.2 was released.

---------------------------------------------------------------------------

Neil Conway wrote:
> The handling of .pgpass files in libpq was recently "fixed" to work with
> all connection methods (a patch applied to both REL7_3_STABLE and HEAD),
> but that change broke .pgpass usage in PQsetdbLogin().
>
> This patch fixes the regression, so .pgpass should now be able to work
> with all of the appropriate database connection methods in libpq.
>
> I also fixed a bug in the parsing of .pgpass: if the file was not
> terminated by a newline, the password would be slightly mangled. I
> didn't bother to account for a CRLF line ending -- is getenv("HOME")
> defined on Win32?
>
> I also refactored a bit of code in PasswordFromFile(), and added some
> missing environmental variables to the libpq docs.
>
> This patch is against CVS HEAD, but it should apply to REL7_3_STABLE
> with just a few offsets. Please apply to both.
>
> Cheers,
>
> Neil
> --
> Neil Conway <neilc(at)samurai(dot)com> || PGP Key ID: DB3C29FC
>
>

[ Attachment, skipping... ]

>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
> (send "unregister YourEmailAddressHere" to majordomo(at)postgresql(dot)org)

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073