FreeTDS Development Group

["ZIGLIO, Frediano, VF-IT" <Frediano.Ziglio AT vodafone.com>]

>
> ZIGLIO, Frediano, VF-IT wrote:
> >
> > Mmm... tds_send_login is not that fine, perhaps it's better
> to rename
> > tds_login structure to  _tds_login (cause we only use
> TDSLOGIN) and add
> > a tds_login.
>
> Good observation. s/tds_login/tds_login_t/g perhaps?
>

Good.

>
> > auto as protocol version, now tds_connect connect and disconnect
> > as it wants...
>
> Does the server disconnect if the login fails?  If not, we can keep
> re-trying login packets on the same connection.
>
> Otherwise, I imagine this (simplified):
>
>
>       if ((conn = tds_connect()) == NULL)
>               return tdserror();
>       else
>               tds_login();
>

can be

> and in tds_login():
>
>       if (tdsver == 0) {
>               for( i=0; i < maxver; i++ ) {
>                       tdsver = versions[i];
>                       int fOK = tds_login(tdsver);
>                       if( fOK )
>                               return fOK;
>
>                       if ((conn = tds_connect()) == NULL)
>                               return tdserror();
>               }
>       }
>       if (i == maxver)
>               return tdserror();
>
>       /* call tdsX_send/do_login function  ... */
>

I don't like this that much... connect on second but not for first...
I never liked that name resolution is in config.c, I thing that should be in 
net.c. Also I 
think it would be better to put name resolution during connection. To detect 
problems 
related to tcp/ip connect in tsql we could test for TDSSOCKET->s, is valid 
only if there is 
a connection (connection is closed before tdserror, but perhaps is closed 
also on login 
errors)...

About naming convention we have

tds_submit/put_XXX  send packets (query.c) there is also a mix of 
submit/put/send... any 
difference ??
tds_process_XXX     read and process packets (token.c)

There is also a tds8_do_login

freddy77

Attachment: smime.p7s
Description: S/MIME cryptographic signature