spacenav-users Mailing List for Free Spacenav

On Fri, Jul 19, 2024 at 08:03:02PM -0600, Lynn Grant wrote:
> 
> /* TODO until we flesh out the USB code on FreeBSD, let's fill the structure
>          * with fake but plausible information.
>          */
>         dev->bnbase = 0;
>         dev->num_buttons = 2;
>         dev->num_axes = 6;
> 
> I don't know if I am studly enough to figure out the USB data, but I will
> look into it tomorrow. 

Sorry for not being able to respond earlier to save you some debugging,
I just woke up and saw your thread. Yeah USB support on FreeBSD is very
rudimentary at this point. I didn't write it myself so I can't help with
any details about the current implementation, but if you're interested
in improving it, you're very welcome.

> And if that doesn't work, I can always set dev_>num_buttons to 15 for
> the time being.

Keep in mind that other than being able to configure button mappings
from the GUI, having spacenavd report 2 buttons shouldn't affect the
functionality of most programs. They will still be able to receive
input from all buttons regardless, and you can still configure button
mappings for them from the config file manually.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

OK, I found it.

In the open_dev_usb() routine of the dev_usb_freebsd.c module, it says

/* TODO until we flesh out the USB code on FreeBSD, let's fill the structure
         * with fake but plausible information.
         */
        dev->bnbase = 0;
        dev->num_buttons = 2;
        dev->num_axes = 6;

I don't know if I am studly enough to figure out the USB data, but I will
look into it tomorrow. And if that doesn't work, I can always set
dev_>num_buttons to 15 for the time being.

Have a great evening.

Lynn


On Fri, Jul 19, 2024 at 7:55 PM Lynn Grant <lg...@ne...> wrote:

> But wait!
>
> Now that I think about it, request() is probably talking to spacenavd,
> which I haven't analyzed yet. The number 2 could be coming from there I
> suppose.
>
> I will look into spacenavd tomorrow.
>
> Lynn
>
> On Fri, Jul 19, 2024 at 5:54 PM Lynn Grant <lg...@ne...> wrote:
>
>> This seems kind of odd. I put some print statements in the request()
>> function of spnav.c, printing the reqrsp structure at entry and after
>> the call to wait_resp() for spnav_dev_name(), spnav_dev_buttons(), and
>> spnav_dev_axes() calls.
>>
>> In spnav_dev_name
>> Entry to request, type = 0
>> Data: 0, 0, 0, 0, 0, 0, 0
>> After wait_resp, type = 7faa2000
>> Data: 0, 0, 0, 0, 0, 0, 0
>>
>> In spnav_dev_buttons
>> Entry to request, type = 0
>> Data: 0, 0, 0, 0, 0, 0, 0
>> After wait_resp, type = 7faa2003
>> Data: 2, 0, 0, 0, 0, 0, 0
>>
>> In spnav_dev_axes
>> Entry to request, type = 0
>> Data: 0, 0, 0, 0, 0, 0, 0
>> After wait_resp, type = 7faa2002
>> Data: 6, 0, 0, 0, 0, 0, 0
>>
>> If I understand things correctly, request() is talking directly to the
>> device. If that is true, then it would appear that the device is actually
>> reporting that it has two buttons, when in reality, it has something like
>> 14.
>>
>> Lynn
>>
>>
>> On Fri, Jul 19, 2024 at 4:01 PM Lynn Grant <lg...@ne...> wrote:
>>
>>> John:
>>>
>>> Thank you. That was it.
>>>
>>> Would it be worth adding something to the README to say that for FreeBSD
>>> you need to set UIC, MOC, and RCC in the Makefile to uic-qt5, moc-qt5, and
>>> rcc-qt5? (This may be obvious to qt5 developers, but for those of us who
>>> are essentially end users, it is not.)
>>>
>>> Everything is working well, except that spnavcfg has a blank device
>>> name, and the default values of 6 and 2 for axes and buttons. It's got the
>>> right icon, though. (I am using a Spacemouse Pro, with a bunch of buttons.)
>>> So it appears that spnav_dev_name(), spnav_dev_axes(), and
>>> spnav_dev_buttons() in spnav.c are not getting back good data when they
>>> call request() to get the information from the Spacemouse. The libspnav
>>> examples all work, so libspnav *is* communicating with the device.
>>>
>>> I am looking into seeing if I can track down the problem, but it will be
>>> slow, because I don't have a lot of experience with FreeBSD debugging. (I
>>> come from a mainframe background; if this were in mainframe assembler, I'd
>>> be all over it :-))
>>>
>>> I am also working on getting an rc.d initialization script set up. I
>>> will let you know when that is working.
>>>
>>> Thanks,
>>>
>>> Lynn
>>>
>>> On Fri, Jul 19, 2024 at 2:13 PM John Tsiombikas <nu...@me...>
>>> wrote:
>>>
>>>> On Fri, Jul 19, 2024 at 12:16:10PM -0600, Lynn Grant wrote:
>>>> >
>>>> > moc -o src/ui.moc.cc -I. -I/usr/local/include src/ui.h
>>>> > gmake: moc: No such file or directory
>>>> ...
>>>> > rcc -o res.cc
>>>> > make: exec(rcc) failed (No such file or directory)
>>>> >
>>>> > I remember running into this error when I was working on this before,
>>>> but
>>>> > that was over a year ago, and I do not remember what the solution was.
>>>>
>>>> I think it was something like FreeBSD has these binaries as moc-qt5 or
>>>> something like that. See what the names are and set the MOC and RCC
>>>> variables accordingly while building.
>>>>
>>>> --
>>>> John Tsiombikas
>>>> http://nuclear.mutantstargoat.com/
>>>>
>>>>
>>>> _______________________________________________
>>>> Spacenav-users mailing list
>>>> Spa...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/spacenav-users
>>>>
>>>

But wait!

Now that I think about it, request() is probably talking to spacenavd,
which I haven't analyzed yet. The number 2 could be coming from there I
suppose.

I will look into spacenavd tomorrow.

Lynn

On Fri, Jul 19, 2024 at 5:54 PM Lynn Grant <lg...@ne...> wrote:

> This seems kind of odd. I put some print statements in the request()
> function of spnav.c, printing the reqrsp structure at entry and after the
> call to wait_resp() for spnav_dev_name(), spnav_dev_buttons(), and
> spnav_dev_axes() calls.
>
> In spnav_dev_name
> Entry to request, type = 0
> Data: 0, 0, 0, 0, 0, 0, 0
> After wait_resp, type = 7faa2000
> Data: 0, 0, 0, 0, 0, 0, 0
>
> In spnav_dev_buttons
> Entry to request, type = 0
> Data: 0, 0, 0, 0, 0, 0, 0
> After wait_resp, type = 7faa2003
> Data: 2, 0, 0, 0, 0, 0, 0
>
> In spnav_dev_axes
> Entry to request, type = 0
> Data: 0, 0, 0, 0, 0, 0, 0
> After wait_resp, type = 7faa2002
> Data: 6, 0, 0, 0, 0, 0, 0
>
> If I understand things correctly, request() is talking directly to the
> device. If that is true, then it would appear that the device is actually
> reporting that it has two buttons, when in reality, it has something like
> 14.
>
> Lynn
>
>
> On Fri, Jul 19, 2024 at 4:01 PM Lynn Grant <lg...@ne...> wrote:
>
>> John:
>>
>> Thank you. That was it.
>>
>> Would it be worth adding something to the README to say that for FreeBSD
>> you need to set UIC, MOC, and RCC in the Makefile to uic-qt5, moc-qt5, and
>> rcc-qt5? (This may be obvious to qt5 developers, but for those of us who
>> are essentially end users, it is not.)
>>
>> Everything is working well, except that spnavcfg has a blank device name,
>> and the default values of 6 and 2 for axes and buttons. It's got the right
>> icon, though. (I am using a Spacemouse Pro, with a bunch of buttons.) So it
>> appears that spnav_dev_name(), spnav_dev_axes(), and spnav_dev_buttons() in
>> spnav.c are not getting back good data when they call request() to get the
>> information from the Spacemouse. The libspnav examples all work, so
>> libspnav *is* communicating with the device.
>>
>> I am looking into seeing if I can track down the problem, but it will be
>> slow, because I don't have a lot of experience with FreeBSD debugging. (I
>> come from a mainframe background; if this were in mainframe assembler, I'd
>> be all over it :-))
>>
>> I am also working on getting an rc.d initialization script set up. I will
>> let you know when that is working.
>>
>> Thanks,
>>
>> Lynn
>>
>> On Fri, Jul 19, 2024 at 2:13 PM John Tsiombikas <nu...@me...>
>> wrote:
>>
>>> On Fri, Jul 19, 2024 at 12:16:10PM -0600, Lynn Grant wrote:
>>> >
>>> > moc -o src/ui.moc.cc -I. -I/usr/local/include src/ui.h
>>> > gmake: moc: No such file or directory
>>> ...
>>> > rcc -o res.cc
>>> > make: exec(rcc) failed (No such file or directory)
>>> >
>>> > I remember running into this error when I was working on this before,
>>> but
>>> > that was over a year ago, and I do not remember what the solution was.
>>>
>>> I think it was something like FreeBSD has these binaries as moc-qt5 or
>>> something like that. See what the names are and set the MOC and RCC
>>> variables accordingly while building.
>>>
>>> --
>>> John Tsiombikas
>>> http://nuclear.mutantstargoat.com/
>>>
>>>
>>> _______________________________________________
>>> Spacenav-users mailing list
>>> Spa...@li...
>>> https://lists.sourceforge.net/lists/listinfo/spacenav-users
>>>
>>

John:

Thank you. That was it.

Would it be worth adding something to the README to say that for FreeBSD
you need to set UIC, MOC, and RCC in the Makefile to uic-qt5, moc-qt5, and
rcc-qt5? (This may be obvious to qt5 developers, but for those of us who
are essentially end users, it is not.)

Everything is working well, except that spnavcfg has a blank device name,
and the default values of 6 and 2 for axes and buttons. It's got the right
icon, though. (I am using a Spacemouse Pro, with a bunch of buttons.) So it
appears that spnav_dev_name(), spnav_dev_axes(), and spnav_dev_buttons() in
spnav.c are not getting back good data when they call request() to get the
information from the Spacemouse. The libspnav examples all work, so
libspnav *is* communicating with the device.

I am looking into seeing if I can track down the problem, but it will be
slow, because I don't have a lot of experience with FreeBSD debugging. (I
come from a mainframe background; if this were in mainframe assembler, I'd
be all over it :-))

I am also working on getting an rc.d initialization script set up. I will
let you know when that is working.

Thanks,

Lynn

On Fri, Jul 19, 2024 at 2:13 PM John Tsiombikas <nu...@me...>
wrote:

> On Fri, Jul 19, 2024 at 12:16:10PM -0600, Lynn Grant wrote:
> >
> > moc -o src/ui.moc.cc -I. -I/usr/local/include src/ui.h
> > gmake: moc: No such file or directory
> ...
> > rcc -o res.cc
> > make: exec(rcc) failed (No such file or directory)
> >
> > I remember running into this error when I was working on this before, but
> > that was over a year ago, and I do not remember what the solution was.
>
> I think it was something like FreeBSD has these binaries as moc-qt5 or
> something like that. See what the names are and set the MOC and RCC
> variables accordingly while building.
>
> --
> John Tsiombikas
> http://nuclear.mutantstargoat.com/
>
>
> _______________________________________________
> Spacenav-users mailing list
> Spa...@li...
> https://lists.sourceforge.net/lists/listinfo/spacenav-users
>

This seems kind of odd. I put some print statements in the request()
function of spnav.c, printing the reqrsp structure at entry and after the
call to wait_resp() for spnav_dev_name(), spnav_dev_buttons(), and
spnav_dev_axes() calls.

In spnav_dev_name
Entry to request, type = 0
Data: 0, 0, 0, 0, 0, 0, 0
After wait_resp, type = 7faa2000
Data: 0, 0, 0, 0, 0, 0, 0

In spnav_dev_buttons
Entry to request, type = 0
Data: 0, 0, 0, 0, 0, 0, 0
After wait_resp, type = 7faa2003
Data: 2, 0, 0, 0, 0, 0, 0

In spnav_dev_axes
Entry to request, type = 0
Data: 0, 0, 0, 0, 0, 0, 0
After wait_resp, type = 7faa2002
Data: 6, 0, 0, 0, 0, 0, 0

If I understand things correctly, request() is talking directly to the
device. If that is true, then it would appear that the device is actually
reporting that it has two buttons, when in reality, it has something like
14.

Lynn


On Fri, Jul 19, 2024 at 4:01 PM Lynn Grant <lg...@ne...> wrote:

> John:
>
> Thank you. That was it.
>
> Would it be worth adding something to the README to say that for FreeBSD
> you need to set UIC, MOC, and RCC in the Makefile to uic-qt5, moc-qt5, and
> rcc-qt5? (This may be obvious to qt5 developers, but for those of us who
> are essentially end users, it is not.)
>
> Everything is working well, except that spnavcfg has a blank device name,
> and the default values of 6 and 2 for axes and buttons. It's got the right
> icon, though. (I am using a Spacemouse Pro, with a bunch of buttons.) So it
> appears that spnav_dev_name(), spnav_dev_axes(), and spnav_dev_buttons() in
> spnav.c are not getting back good data when they call request() to get the
> information from the Spacemouse. The libspnav examples all work, so
> libspnav *is* communicating with the device.
>
> I am looking into seeing if I can track down the problem, but it will be
> slow, because I don't have a lot of experience with FreeBSD debugging. (I
> come from a mainframe background; if this were in mainframe assembler, I'd
> be all over it :-))
>
> I am also working on getting an rc.d initialization script set up. I will
> let you know when that is working.
>
> Thanks,
>
> Lynn
>
> On Fri, Jul 19, 2024 at 2:13 PM John Tsiombikas <nu...@me...>
> wrote:
>
>> On Fri, Jul 19, 2024 at 12:16:10PM -0600, Lynn Grant wrote:
>> >
>> > moc -o src/ui.moc.cc -I. -I/usr/local/include src/ui.h
>> > gmake: moc: No such file or directory
>> ...
>> > rcc -o res.cc
>> > make: exec(rcc) failed (No such file or directory)
>> >
>> > I remember running into this error when I was working on this before,
>> but
>> > that was over a year ago, and I do not remember what the solution was.
>>
>> I think it was something like FreeBSD has these binaries as moc-qt5 or
>> something like that. See what the names are and set the MOC and RCC
>> variables accordingly while building.
>>
>> --
>> John Tsiombikas
>> http://nuclear.mutantstargoat.com/
>>
>>
>> _______________________________________________
>> Spacenav-users mailing list
>> Spa...@li...
>> https://lists.sourceforge.net/lists/listinfo/spacenav-users
>>
>

On Fri, Jul 19, 2024 at 12:16:10PM -0600, Lynn Grant wrote:
> 
> moc -o src/ui.moc.cc -I. -I/usr/local/include src/ui.h
> gmake: moc: No such file or directory
...
> rcc -o res.cc
> make: exec(rcc) failed (No such file or directory)
> 
> I remember running into this error when I was working on this before, but
> that was over a year ago, and I do not remember what the solution was.

I think it was something like FreeBSD has these binaries as moc-qt5 or
something like that. See what the names are and set the MOC and RCC
variables accordingly while building.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

After a long absence, I have a working FreeBSD 14.1 machine, and am working
on installing spacenav. Things are going much better than last time. All
the examples for libspnav are working fine. (I just have to convert the
spacenavd script to a FreeBSD rc.d script.)

I am having some trouble building spnavcfg. I have libspnav and the QT5
prerequisites (core, gui, and widgets) installed. When I try to build it
using gmake, I get the following error:

moc -o src/ui.moc.cc -I. -I/usr/local/include src/ui.h
gmake: moc: No such file or directory
gmake: *** [Makefile:46: src/ui.moc.cc] Error 127

 Using make, I get the following error:

rcc -o res.cc
make: exec(rcc) failed (No such file or directory)
*** Error code 1

I remember running into this error when I was working on this before, but
that was over a year ago, and I do not remember what the solution was.

Any suggestions?

Thanks,

Lynn

On Tue, Jul 02, 2024 at 01:39:04PM +0000, Keith Hoard via Spacenav-users wrote:
>     
> First off, thank you for the warm welcome.  I'll see what I can do
> about the 72 column thing. . . 

Whatever you attempted to do, failed.

> When I type "spacenavd" in a terminal, the command executes, but
> returns "failed to open log file: /var/log/spnavd.log"
>
> The blue LEDs do illuminate in the mouse after executing the command
> in a terminal.  Also, "spacenavd" shows up in htop.  
>
> However, I cannot get FreeCAD to respond to the mouse's commands.

Spacenavd is designed to be run as a daemon, started by init during
boot, and running as root. It's not meant to be started by the user
before use.

Having said that, while you're testing things it might be more
convenient to start it as you did from a terminal, however make sure to
run it as root. When started interactively like that for testing, make
sure to pass the -d flag (do not daemonize), and ideally also -v
(verbose output). In that mode instead of logging any messages to a log
file, they are written to stdout/stderr, which makes it easier to keep
track of what's going on.

To summarize:
 - in normal use, arrange for your init script to start spacenavd during
   boot.
 - for testing/debugging, you can run it like so:
   $ sudo spacenavd -d -v

To test if it works, beyond monitoring the spacenavd log output, try
running the example programs which come with libspnav. Especially try
both of the simple_af_unix and simple_x11 example programs.

> I've read the FAQ and think items 9 and 10 may be applicable here, but
> I don't know how to proceed to fix them.

FAQ 9 probably not, FAQ 10 might be an issue further down the line, you
will know it if `simple_af_unix` works, but `simple_x11` doesn't.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

    
   First off, thank you for the warm welcome.  I'll see what I can do about the 72 column thing. . . 

   I have a 3D Connexion Space Navigator mouse that works well in Windows. I just swapped to Manjaro a few 
months ago and use FreeCAD and Fusion (w/Wine when it works) for 3D print modelling.  

   I have installed "spacenavd" AUR version 1.3-1 using the Manjaro Package Manager system.

   When I type "spacenavd" in a terminal, the command executes, but returns "failed to open log file: /var/log/spnavd.log"
The blue LEDs do illuminate in the mouse after executing the command in a terminal.  Also, "spacenavd" shows up in htop.  
However, I cannot get FreeCAD to respond to the mouse's commands.  Fusion w/Wine may be another battle for another day.

   I've read the FAQ and think items 9 and 10 may be applicable here, but I don't know how to proceed to fix them.

   Any help I could get would be greatly appreciated.  Thanks!!

-Keith Hoard
-kl...@pr...

Sent with Proton Mail secure email.

On Monday, July 1st, 2024 at 20:13, John Tsiombikas - nuclear at member.fsf.org <nuc...@pa...> wrote:

> This email failed anti-phishing checks when it was received by SimpleLogin, be careful with its content.
> More info on https://simplelogin.io/docs/getting-started/anti-phishing/
> 
> ------------------------------
> On Mon, Jul 01, 2024 at 10:14:02PM +0000, Keith Hoard via Spacenav-users wrote:
> 
> > Just checking if this list has folks subscribed. . . Need some help
> > getting my 3D mouse working. .
> 
> 
> When faced with such a conundrum, there are always two distinct
> possibilities. Either the list is empty, in which case you will not
> receive any answer, or there are people subscribed to the list, in which
> case you just annoyed everyone by posting a redundant question.
> 
> In both cases, it would have been best if you just asked your real
> question from the start.
> 
> Also, hard-wrap your emails at 72 columns.
> 
> --
> John Tsiombikas
> http://nuclear.mutantstargoat.com/
> 
> 
> _______________________________________________
> Spacenav-users mailing list
> Spa...@li...
> https://lists.sourceforge.net/lists/listinfo/spacenav-users

On Mon, Jul 01, 2024 at 10:14:02PM +0000, Keith Hoard via Spacenav-users wrote:
> Just checking if this list has folks subscribed. . . Need some help
> getting my 3D mouse working. .

When faced with such a conundrum, there are always two distinct
possibilities. Either the list is empty, in which case you will not
receive any answer, or there are people subscribed to the list, in which
case you just annoyed everyone by posting a redundant question.

In both cases, it would have been best if you just asked your real
question from the start.

Also, hard-wrap your emails at 72 columns.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

Just checking if this list has folks subscribed. . . Need some help getting my 3D mouse working. .

-Keith Hoard
-kl...@pr...

Sent with [Proton Mail](https://proton.me/) secure email.

Hi John,

I feel so sorry that I did so much inconvenience to you. By following your instructions, it
seems that some updated packages only work after I reboot the system. So now I could
see the axes moving in config GUI and the generated motions from “simple_af_unix”.

Thank you for your help again! Have a nice weekend!

Best regards,
Haocheng

Please make an effort to post correctly.
 1. You accidentally responded to my personal email instead of the list
 2. You fail to quote relevant sections *before* your response, instead
    include the whole text of my message *after* your response, which is
    unacceptable.
 3. You fail to hard-wrap your messages to 72 columns, which is the
    standard email nettiquette, instead you let the text run off the
    side of the screen.
 4. You post multi-part messages, both text and html, please send text
    emails only to this mailing list.

I'm responding to the mailing list, please continue the conversation
here, and try to fix the issues I mentioned above.

On Thu, Jun 27, 2024 at 04:59:03PM +0000, Yin, Haocheng wrote:
>
> When I run spnavcfg, I open a new terminal window and run “sudo
> /etc/init.d/spacemavd start”, the result shows that “Spavenav
> daemon already running (pid: xxxxx).  Aborting.” which means daemon
> is running normally.

If the daemon crashes, it might leave the pid file on the disk, and
further invocations will complain that it's already running. Use `ps` to
verify that it is indeed running.

During testing it might be easier to run spacenavd from a root shell
directly (as opposed to telling init to start it as a daemon), and
instruct it to not daemonise by passing the -d option (also pass -v for
verbose output). When run this way, spacenavd does not release the
terminal, and uses it to print all its messages instead of logging them
to a file. If it crashes, the crash will be obvious.

> One thing I need to clarify is that spnavcfg could only identify my
> device but no matter how I move the SpaceMouse, the axes are not
> moving.

It could be a bug in device handling. The spacemouse wireless should be
usable both wireless and wired, with a USB cable. Did you try both ways?

> And just now I realized that the config file “/etc/spnavrc” did
> not change after I did some changes from spnavcfg and clicked “save
> config”. Did you meet any similar case before?

When you click the save button in spnavcfg, it sends a command to
spacenavd to save the current configuration. It's not done by spnavcfg
itself. That it didn't work, is another indication that spacenavd is not
running at that stage.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

On Thu, Jun 27, 2024 at 03:38:10PM +0000, Yin, Haocheng wrote:
>
>   1.  However, when I tried the given examples “simple_af_unix”,
>   “simple_x11” and “cube”, it seems that no matter how I move the
>   SpaceMouse, there’s no response from the program and the cube was
>   never rotated.

>From the log you posted, it looks like the device was detected correctly
and should be working. Also you mentioned you used spnavcfg to configure
it. If it was working on spnavcfg (device identified, axes moving and so
on), it doesn't make any sense that the example programs would fail to
work also. spnavcfg uses libspnav to communicate with spacenavd, same as
the `simple_af_unix` example program. If one works, the other should
too.

>   2.  Since I’m very sure that I have set the correct “device_id” into
>   spnavrc, I strongly doubted that the button mapping from my
>   SpaceMouse is wrong, but I have tried all the provided example
>   mappings into spnavrc and nothing changed.

You shouldn't have to set device_id. That's only for forcing spacenavd
to use unsupported devices like joysticks. 3Dconnexion spacemice should
be detected automatically.

Button mappings only affect what happens when you press a button on the
device, nothing else.

> Do you know how could I fix this problem? Thank you in advance and
> looking forward to your reply!

The only possible cause I can think of, based on your report, is that it
worked initially when you run spnavcfg, and then the spacenavd daemon
crashed or was stopped, and wasn't running at all when you tried the
example programs. I can't see what else could be happening based on your
story.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

Dear Spacenav team,

I hope this email finds you well. I’m a master student who is working on a research project which requires SpaceMouse as input device. I’m writing to kindly ask how could I receive the correct motion after my SpaceMouse is successfully connected to my Linux system?

My driver problem could be described as follows:

  1.  After following the instructions from FreeCAD wiki<https://wiki.freecad.org/3Dconnexion_input_devices/en>, I installed spacenavd==1.3<https://github.com/FreeSpacenav/spacenavd/releases>, libspnav==1.1<https://github.com/FreeSpacenav/libspnav/releases>, spnavcfg==1.1<https://github.com/FreeSpacenav/spnavcfg/releases> from source (the official github repos) to make sure every package is the latest. Besides, I also did the trick to copy “run/user/1000/gdm/Xauthority” to the “/root/” folder. Then I copied the provided example “spnavrc” file under “/etc/” folder and used the spnavcfg GUI to adjust the config. I have double checked that “/etc/spnavrc” could be sync to exactly the same setting in the GUI.
  2.  Then I used “tail -n100 -f /var/log/spnavd.log” to check if the installed driver works well. Based on the results of unplugging and plugging, my PC could successfully show the connection to the SpaceMouse:

Spacenav daemon 1.3
reading config file: /etc/spnavrc
trying to open X11 display ":1"
   XAUTHORITY=/run/user/1000/gdm/Xauthority
Using XTEST to send key events
found usb device [256f:c63a]: "3Dconnexion SpaceMouse Wireless BT" (/dev/input/event2)
adding device (id: 0).
device name: 3Dconnexion SpaceMouse Wireless BT
  Number of axes: 6 (6a 0r)
  Number of buttons: 2 (evdev offset: 256)
  Axis 0 value range: -350 - 350 (fuzz: 0)
  Axis 1 value range: -350 - 350 (fuzz: 0)
  Axis 2 value range: -350 - 350 (fuzz: 0)
  Axis 3 value range: -350 - 350 (fuzz: 0)
  Axis 4 value range: -350 - 350 (fuzz: 0)
  Axis 5 value range: -350 - 350 (fuzz: 0)
using device: 3Dconnexion SpaceMouse Wireless BT (/dev/input/event2)
  device flags: swap y-z invert y-z


  1.  However, when I tried the given examples “simple_af_unix”, “simple_x11” and “cube”, it seems that no matter how I move the SpaceMouse, there’s no response from the program and the cube was never rotated.
  2.  Since I’m very sure that I have set the correct “device_id” into spnavrc, I strongly doubted that the button mapping from my SpaceMouse is wrong, but I have tried all the provided example mappings into spnavrc and nothing changed.

Do you know how could I fix this problem? Thank you in advance and looking forward to your reply!

Best regards,
Haocheng

Hello,

Thank you for your very clear explanations an quick reply.

As you recommended, I created a /etc/spnavrc file.
I added the line: "serial = /dev/ttyS1" and, when I ran the command: 
"sudo spacenavd -v -d", it successfully found my spacemouse !

Here is the output (| added for clarity):
| USER@HOST:~$ sudo spacenavd -v -d
| Spacenav daemon v1.1
| reading config file: /etc/spnavrc
| adding device (id: 0).
| Magellan SpaceMouse detected:
| v  MAGELLAN  Version 6.70  3Dconnexion GmbH 05/11/02
|
| 9 buttons
| using device: /dev/ttyS1
| failed to bind unix socket: run/spnav.sock: No such file or directory
| trying to open X11 display ":1"
|    XAUTHORITY=/run/user/1000/xauth_AAYSWt
| Using XTEST to send key events

At this point the spacemouse still didn't work with FreeCAD.

I then tried creating a blank .Xauthority file in my home directory and 
copying it to /root like explained in the FreeCAD wiki:
"cp ~/.Xauthority /root/"

After running the command: "sudo spacenavd -v -d", IT WORKED !

Thank you so much for your help, I was starting to think I wouldn't be 
able to get the spacemouse to work.

I can now enjoy moving in an intuitive way in my favorite CAD and I also 
have tons of macro keys !

Thanks again.




On 01/06/2024 00:42, John Tsiombikas wrote:
> On Fri, May 31, 2024 at 08:05:32PM +0200, Gabriel PORTIER via Spacenav-users wrote:
>> Hello,
>>
>> I have a Spacemouse Plus Magellan connected to one my PC's serial ports.
>> I have installed the spacenavd driver according to this guide:
>> https://wiki.freecad.org/3Dconnexion_input_devices
> 
> That guide doesn't seem to say anything about serial spacemice. Serial
> devices cannot be auto-detected. You need to explicitly specify the
> correct device file in /etc/spnavrc like so (obviously use the device
> file which corresponds to the correct serial port):
> 
>    serial = /dev/ttyS0
> 
> You can refer to doc/example-spnavrc in the spacenavd source tree for
> detailed descriptions and syntax examples of all of the options, or you
> can use the spnavcfg GUI to configure most options interactively. In
> your case for starters, just create an /etc/spnavrc with just the serial
> line.
> 
> Also the part of the guide about running spnavd_ctl is unnecessary since
> spacenavd v0.4 released in 2009.
> 
>> The FreeCAD log file states:
>> "Connected to spacenav daemon"
>>
>> Unfortunately, the spacemouse doesn't work at all with FreeCAD.
>>
>> During installation, I noticed that the step:
>> "cp ~/.Xauthority /root/"
>> returned an error because the file .Xauthority does not exist.
>>
>> Could this be the reason why the spacemouse isn't working ?
> 
> Possibly, if spacenavd cannot connect to the X server and FreeCAD
> relies on the X11-based protocol. Check the log file (by default
> /var/log/spnavd.log) for any relevant error messages.
> 
> Also to make sure the device works, run any of the example programs
> which come with libspnav. Specifically at least simple_af_unix and
> simple_x11, which excercise either of the two supported communication
> protocols.
> 

On Fri, May 31, 2024 at 08:05:32PM +0200, Gabriel PORTIER via Spacenav-users wrote:
> Hello,
> 
> I have a Spacemouse Plus Magellan connected to one my PC's serial ports.
> I have installed the spacenavd driver according to this guide:
> https://wiki.freecad.org/3Dconnexion_input_devices

That guide doesn't seem to say anything about serial spacemice. Serial
devices cannot be auto-detected. You need to explicitly specify the
correct device file in /etc/spnavrc like so (obviously use the device
file which corresponds to the correct serial port):

  serial = /dev/ttyS0

You can refer to doc/example-spnavrc in the spacenavd source tree for
detailed descriptions and syntax examples of all of the options, or you
can use the spnavcfg GUI to configure most options interactively. In
your case for starters, just create an /etc/spnavrc with just the serial
line.

Also the part of the guide about running spnavd_ctl is unnecessary since
spacenavd v0.4 released in 2009.

> The FreeCAD log file states:
> "Connected to spacenav daemon"
> 
> Unfortunately, the spacemouse doesn't work at all with FreeCAD.
> 
> During installation, I noticed that the step:
> "cp ~/.Xauthority /root/"
> returned an error because the file .Xauthority does not exist.
> 
> Could this be the reason why the spacemouse isn't working ?

Possibly, if spacenavd cannot connect to the X server and FreeCAD
relies on the X11-based protocol. Check the log file (by default
/var/log/spnavd.log) for any relevant error messages.

Also to make sure the device works, run any of the example programs
which come with libspnav. Specifically at least simple_af_unix and
simple_x11, which excercise either of the two supported communication
protocols.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

Hello,

I have a Spacemouse Plus Magellan connected to one my PC's serial ports.
I have installed the spacenavd driver according to this guide:
https://wiki.freecad.org/3Dconnexion_input_devices

The FreeCAD log file states:
"Connected to spacenav daemon"

Unfortunately, the spacemouse doesn't work at all with FreeCAD.

During installation, I noticed that the step:
"cp ~/.Xauthority /root/"
returned an error because the file .Xauthority does not exist.

Could this be the reason why the spacemouse isn't working ?

Regards.

On Thu, Jan 18, 2024 at 04:03:59PM +0000, L. Mollicone wrote:
> Hey guys, any tips for building this on IRIX?
> 
> Platform is an Octane, version 6.5.30

It should just work, but you need to use GNU make and GCC. So I assume
something like `./configure && gmake CC=gcc` will probably work.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

Hey guys, any tips for building this on IRIX?

Platform is an Octane, version 6.5.30

Thanks!

On Sun, Nov 12, 2023 at 10:58:41AM +0100, Manfred Zagler wrote:
>
> Nov 12 10:03:56 CAD spacenavd[810]: Authorization required, but no
> authorization protocol specified

This does sound like an X authentication problem. If it is, spacenavd
should be working fine, but programs written to use the 3Dconnexion
X-based protocol will not work. Presumably your CAD program is one such,
though it would be nice to verify all that by running the "simple"
example programs which come with libspnav, as Maximilian suggested.

>              └─810 /usr/bin/spacenavd -d -v
...
> I tried to follow your descriptions and solutions, but it was not sucessful.
> Tried to do a symbolic link to the root directroy, but .Xauthority file
> does not exit.
> /var/log/spnavd.log does not exit.

The `-d` command line argument instructs spacenavd to not daemonize, and
*not* use the regular log file, but rather use standard output and
standard error streams for logging. I don't know why the debian
maintainers do that, but I guess it's so that systemd can intercept the
messages and log them somewhere centrally. I can't help with finding
them, as I've never used systemd, and I'm not familiar with it, but if
you remove the `-d` argument from the systemd file, it will log to
/var/log/spnavd.log normally.

What is the value of the XAUTHORITY environment variable, while you're
logged into X normally?

Make sure to run the example programs before your next reply, to have
something concrete to be going on, instead of basing everything onto
assumptions.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

Did you try the example programs to check if there's a problem at system level or something related to a specific software (Varicad in your case)?

On November 12, 2023 10:58:41 AM GMT+01:00, Manfred Zagler <man...@gm...> wrote:
>Dear all,
>
>i use a CAD "Varicad" within Linux which worked well for me together
>with Space Navigator 3D Mouse.
>Recently i updated to Debian 12. Now it is no longer possible for me to
>use the 3D mouse as before.
>
>*_My system:_*
>Linux CAD 6.1.0-13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.55-1
>(2023-09-29) x86_64 GNU/Linux
>
>*_USB recognize the 3D mouse:_*
>Bus 001 Device 003: ID 046d:c626 Logitech, Inc. 3Dconnexion Space
>Navigator 3D Mouse
>
>*_Installed package from Debian with apt:_*
>Package: spacenavd
>Version: 1.1-1
>Priority: optional
>Section: utils
>Maintainer: Debian Science Maintainers
><deb...@li...>
>Installed-Size: 123 kB
>Pre-Depends: init-system-helpers (>= 1.54~)
>Depends: lsb-base, libc6 (>= 2.34), libx11-6, libxi6 (>= 2:1.2.99.4),
>libxtst6
>Suggests: freecad
>Homepage: http://spacenav.sourceforge.net
>Tag: hardware::input, interface::daemon, role::program
>Download-Size: 42,0 kB
>APT-Manual-Installed: yes
>APT-Sources: http://deb.debian.org/debian bookworm/main amd64 Packages
>
>*_systemd.service status:_*
>root@CAD:~# systemctl status spacenavd.service
>spacenavd.service - 3Dconnexion Input Devices Userspace Driver
>
>     Loaded: loaded (/lib/systemd/system/spacenavd.service; enabled;
>preset: enabled)
>     Active: active (running) since Sun 2023-11-12 10:03:34 CET; 36min ago
>       Docs: man:spacenavd(8)
>   Main PID: 810 (spacenavd)
>      Tasks: 1 (limit: 18300)
>     Memory: 1.6M
>        CPU: 15ms
>     CGroup: /system.slice/spacenavd.service
>             └─810 /usr/bin/spacenavd -d -v
>Nov 12 10:03:56 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:03:57 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:03:58 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:03:59 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:04:00 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:04:01 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:04:02 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:04:03 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:04:04 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>Nov 12 10:04:05 CAD spacenavd[810]: Authorization required, but no
>authorization protocol specified
>
>I tried to follow your descriptions and solutions, but it was not sucessful.
>Tried to do a symbolic link to the root directroy, but .Xauthority file
>does not exit.
>/var/log/spnavd.log does not exit.
>
>I hope some of you can help.
>
>BR
>
>Manfred

Dear all,

i use a CAD "Varicad" within Linux which worked well for me together
with Space Navigator 3D Mouse.
Recently i updated to Debian 12. Now it is no longer possible for me to
use the 3D mouse as before.

*_My system:_*
Linux CAD 6.1.0-13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.55-1
(2023-09-29) x86_64 GNU/Linux

*_USB recognize the 3D mouse:_*
Bus 001 Device 003: ID 046d:c626 Logitech, Inc. 3Dconnexion Space
Navigator 3D Mouse

*_Installed package from Debian with apt:_*
Package: spacenavd
Version: 1.1-1
Priority: optional
Section: utils
Maintainer: Debian Science Maintainers
<deb...@li...>
Installed-Size: 123 kB
Pre-Depends: init-system-helpers (>= 1.54~)
Depends: lsb-base, libc6 (>= 2.34), libx11-6, libxi6 (>= 2:1.2.99.4),
libxtst6
Suggests: freecad
Homepage: http://spacenav.sourceforge.net
Tag: hardware::input, interface::daemon, role::program
Download-Size: 42,0 kB
APT-Manual-Installed: yes
APT-Sources: http://deb.debian.org/debian bookworm/main amd64 Packages

*_systemd.service status:_*
root@CAD:~# systemctl status spacenavd.service
spacenavd.service - 3Dconnexion Input Devices Userspace Driver

      Loaded: loaded (/lib/systemd/system/spacenavd.service; enabled;
preset: enabled)
      Active: active (running) since Sun 2023-11-12 10:03:34 CET; 36min ago
        Docs: man:spacenavd(8)
    Main PID: 810 (spacenavd)
       Tasks: 1 (limit: 18300)
      Memory: 1.6M
         CPU: 15ms
      CGroup: /system.slice/spacenavd.service
              └─810 /usr/bin/spacenavd -d -v
Nov 12 10:03:56 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:03:57 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:03:58 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:03:59 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:04:00 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:04:01 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:04:02 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:04:03 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:04:04 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified
Nov 12 10:04:05 CAD spacenavd[810]: Authorization required, but no
authorization protocol specified

I tried to follow your descriptions and solutions, but it was not sucessful.
Tried to do a symbolic link to the root directroy, but .Xauthority file
does not exit.
/var/log/spnavd.log does not exit.

I hope some of you can help.

BR

Manfred

On Thu, Oct 26, 2023 at 10:29:48AM +0000, Alan Munro wrote:
> Hi all,
> 
> Good to be on the group; I was searching for support / drivers for my
> Cadmouse Pro on Linux mint, so any advice would be most appreciated

I have a cadmouse pro wireless, and it worked perfectly in wireless mode
out of the box. It stops working when I plug the USB cable for some
reason but I haven't bothered to look into it so far.

In any case this is not relevant to this project. Spacenavd only deals
with 6dof devices. It's up to the X server to support regular mice.

-- 
John Tsiombikas
http://nuclear.mutantstargoat.com/

Hi all,

Good to be on the group; I was searching for support / drivers for my Cadmouse Pro on Linux mint, so any advice would be most appreciated

Regards,

Alan Munro
+27 82 477 1461

Sent from my iPhone