Re: dinput: joydevs[id].name (EVIOCGNAME) is product name, joydevs[id].device is a path

[Vitaliy Margolen]

Vincent Pelletier wrote:
> joydevs[id].name value comes from EVIOCGNAME iotcl, and is the product name.
>
> I think joystick device path qualifies as a "Friendly name for the instance" 
> (as defined for tszInstanceName on MSDN).
>
> If not, please tell me and I will extend the patch to generate a "Joystick %i" 
> name, or maybe put joydevs[id].name in both fields.
>
> -    strcpy(lpddi->tszInstanceName, joydevs[id].name);
> -    strcpy(lpddi->tszProductName, joydevs[id].device);
> +    strcpy(lpddi->tszInstanceName, joydevs[id].device);
> +    strcpy(lpddi->tszProductName, joydevs[id].name);

No this is not entirely correct. Native doesn't return any device paths. But
it does return proper joystick name:

Wine:
 Instance: Logitech Logitech Dual Action
 ProductName: /dev/input/event7

Win XP:
 Instance: Logitech Dual Action USB
 ProductName: Logitech Dual Action USB

Another joystick:
Wine:
 Instance: Logitech  Logitech MOMO Racing
 ProductName: /dev/input/event7

Win XP:
 Instance: Logitech MOMO Racing USB
 ProductName: Logitech MOMO Racing USB

So I'd suggest copying name into both places. I think some applications
looking for particular information there. And append "evdev?" at the end so
we can tell how it's handled.

Vitaliy.

dinput: joydevs[id].name (EVIOCGNAME) is product name, joydevs[id].device is a path

[Vincent Pelletier]

Le Saturday 07 February 2009 19:25:56 Vitaliy Margolen, vous avez écrit :
> So I'd suggest copying name into both places.

Ok.
> And append "evdev?" at the end so we can tell how it's handled.

See attached patch.
First version I locally wrote only stored an additional int in joydev, but:
- maybe it's not a good idea to compute the concatenation on each
  fill_joystick_dideviceinstance[AW] call
- ...I just don't know how to append to a widechar string (MSDN list of 
widechar funcs)

-- 
Vincent Pelletier

dinput: joydevs[id].name (EVIOCGNAME) is product name, joydevs[id].device is a path

[Vitaliy Margolen]

Vincent Pelletier wrote:
> Le Saturday 07 February 2009 19:25:56 Vitaliy Margolen, vous avez écrit :
> > So I'd suggest copying name into both places.
>
> Ok.
> > And append "evdev?" at the end so we can tell how it's handled.
>
> See attached patch.
> First version I locally wrote only stored an additional int in joydev, but:
> - maybe it's not a good idea to compute the concatenation on each
>   fill_joystick_dideviceinstance[AW] call
> - ...I just don't know how to append to a widechar string (MSDN list of 
> widechar funcs)

Sorry didn't reply earlier - no this is not what I meant. Native returns the
same string for both. So just drop the device path all together. Make it the
name returned from the kernel + "evdev%d" ala "USB" native adds.

Vitaliy.

dinput: joydevs[id].name (EVIOCGNAME) is product name, joydevs[id].device is a path

[Ben Klein]

2009/2/10 Vincent Pelletier <plr.vincent(a)gmail.com>:
> On Tue, Feb 10, 2009 at 1:00 AM, Ben Klein <shacklein(a)gmail.com> wrote:
> > You're adding a new field to the JoyDev struct called instance_name,
> > which if I read right will store the input device path, but what
> > happens to the existing device field? It looks like it's just taking
> > up space, not actually getting a value stored in it.
>
> There is a value stored in it: see the code just above the second
> chunk in my patch.

You're taking out the strcpy into device, and replacing it with a
strcpy into instance_name. What value will the device field hold?
> But maybe you mean that it might not be read any more... Which might
> be the case, I didn't check.
>
> About using space, I was more "worried" about the duplication of data
> between "name" (kernel-provided device string - not a path) and the
> new "instance_name" (could not find a better name for that without
> changing existing) which also contains the device string plus
> "evdev%d" as Vitaliy proposed.

When will the new "instance_name" field be accessed by Wine? Regarding
Vitaliy's earlier post, it looks ike "name" (product name) and
"instance_name" are the same on native Windows platform. If this is
always true, returning "name" when asked for instance name would be
valid (so, storing only one string). A counter-example of this would
be useful.

The other difference is that the evdev method returns the vendor as a
part of the product name (hence "Logitech Logitech MOMO Racing" etc.)
Should we be concerned about this?