Closed GoogleCodeExporter closed 8 years ago
Show inxi -Sxxx, extra data like mate or cinnamon show only with extra data, as the helm menu and documentation should show.
As far as I know, neither mate nor cinnamon are currently actually identifiable
using the xprop output that we use to see what desktop is running, but inxi
does get that data later with the extra flags.
Now, you can upload the inxi debugger output: inxi -xx@14
for each desktop when it's running and inxi is run inside that desktop.
and I can see if there's anything there that requires adjusting, but this issue
was fixed a while ago.
Original comment by inxi-...@techpatterns.com
on 15 Sep 2012 at 11:58
Alright, so I have the .tar.gz for cinnamon, I don't have MATE though. I took a
look at the xprop_root.txt, and on the first line it mentions muffin
(cinnamon's window manager), so shouldn't that be enough? Attaching it, even
though it said it uploaded automatically.
inxi-rarity-all-20120915.tar.gz is the name.
Original comment by dan...@the-beach.co
on 16 Sep 2012 at 3:56
Attachments:
I've also had somebody running mate to run it as well. They reported the
filename was: inxi-buntupc-all-20120915.tar.gz
Original comment by dan...@the-beach.co
on 16 Sep 2012 at 3:57
Turns out this is a valid issue, but it will require more information.
Currently inxi checks this: $GNOME_DESKTOP_SESSION_ID which is a gnome global,
try it, do:
echo $GNOME_DESKTOP_SESSION_ID
and if you got some result, that is what inxi used to determine that you run
gnome.
To handle the two sub cases of gnome, cinnamon and mate, would require changing
that core identification method.
I didn't see anything that identifies mate in xprop output for Mate, the
cinnamon thing now that you identified it shows the one identifier, but I have
no idea how to get the subsequent version information, or what to query from
cinnamon to get its version information.
So there's a few things happening here, some easier than others to fix. I will
need to find out how to query cinnamon for its version information however, ie,
once inxi detects that it is a subset of gnome, cinnamon, it needs to kick in a
different thing. Plus cinnamon as I understand it is in fact more of a wrapper
for gnome 3 than an actual separate desktop, isn't that right?
And Mate, that is a different system so that's worth looking more at as a
unique desktop, that is afterall an actual fork. Mate I have no information on
at all, I saw nothing in the xprop data I received that id it as mate, but I
could simply not know what to look for.
I attach the xprop file in case someone knows what to look for.
But in both cinnamon and mate's case, I need to verify that
$GNOME_DESKTOP_SESSION_ID
is either null or not null, since that is the test inxi currently uses to id
gnome desktops. If mate and cinnamon are both not null, then I need to add in
some sub logic to detect if either is running, and then I need to know what the
version detection for mate and cinnamon are, ie, what is the command line
command to get its version, like: cinnamon --version or whatever it is.
At that point it's not too hard for cinnamon, since I can simply do the xprop
test for the muffin string and if present, report desktop as gnome 3 running
cinnamon, or something. But keep in mind, cinnamon is not really a complete
desktop, or is it now? Mate is, so that would be good to get id'ed correctly,
but cinnamon seems to be more of a fix of gnome 3 than an actual desktop, so
I'm not sure if inxi should id it as cinnamon over gnome 3.0 (cinnamon) for
example.
Anyway, the issue is valid, thanks for following up.
Original comment by inxi-...@techpatterns.com
on 16 Sep 2012 at 5:32
Attachments:
Actually, Cinnamon is a gnome fork as well, and muffin is a fork of mutter. You
may be thinking of the shell extensions that mint used for a release that more
or less did the same thing. It's definitely a separate desktop at this point
though.
Now, `echo $GNOME_DESKTOP_SESSION_ID` returns "this-is-deprecated".
Getting the cinnamon version is as simple as running "cinnamon --version".
Which, in my case returns "Cinnamon 1.5.8".
I can't speak too much for MATE, as I don't run it, but I'm guessing since it's
more or less just gnome 2 with a different name, it may be a bit harder to
identify.
Anyways, thanks, if you need any more info, just let me know.
Original comment by dan...@the-beach.co
on 16 Sep 2012 at 5:48
inxi 1.8.16 should have working cinnamon detection now on the primary -S level,
and it's turned off as a secondary info output.
I hope this doesn't break older Mint cinnamon handling, but we'll see.
I'm leaving Mate unhandled except for the -Sxxx output until enough data is
collected to do that reliably. So this issue is half resolved now, or should be.
On the plus side, $GNOME_DESKTOP_SESSION_ID is deprecated, but still useful, so
it's good to move away from depending on that for all gnome instances, over
time inxi will adjust to whatever gnome presents in current versions while
leaving the legacy handling for old systems.
Original comment by inxi-...@techpatterns.com
on 16 Sep 2012 at 6:03
That's wonderful, I'll try it out tomorrow, as I'm out of bandwidth for the
month. If you need input as far as MATE, I can find somebody on IRC to
help. But, thanks for fixing this, I love when developers are this
responsive towards bug reports/feature requests :D
Original comment by dan...@the-beach.co
on 16 Sep 2012 at 10:26
MATE is now handled as well, it has an ID in xrpop too, _MARCO which is the
renamed metacity window manager.
The version command is mate-about -v
Got that info from MATE irc support. So I think that closes this issue, unless
there's a bug with it.
So now primary -S info should correctly show the forked new desktops Cinnamon
and MATE.
Unless there's a bug, of course.
Original comment by inxi-...@techpatterns.com
on 17 Sep 2012 at 12:49
Original issue reported on code.google.com by
dan...@the-beach.co
on 12 Sep 2012 at 2:13