Re: [sigrok-devel] [PATCH] Fix objectName notices and save/restore last used device

[Soeren A. <so...@ap...>]

Hi Joel, hi Peter,

I guess I shot myself in the foot with this one. My intention was
to not spam the list with tons of small patches but I guess it
backfired. Lesson learned :)

My intention also was to have something in PV that works now for a
lot of people, improving on it later on. If we want to include any
other means of identification then we need to have changes in
libsigrok as well since it currently doesn't supply anything else
that could be used.

As for the session file loading, it wouldn't interfere. The file is
loaded after the UI has been restored to its previous state, so if
anything, another device might be selected for a very short period
of time. You have a point, though. This is not clean and should be
prevented.

-Soeren


On Sun, 2014-09-14 at 08:37 +0100, Joel Holdsworth wrote:
> I would prefer that these two changes were not squashed together. It
> would certainly be better to have a more reliable way to identify
> different devices than just the name. Also how does this work when a
> file is load on the command line? Would it override the file-device?
> 
> On 14 September 2014 04:46:40 BST, Peter Stuge <pe...@st...> wrote:
>         so...@ap... wrote:
>                 My last patch introduced window state saving and restoring. It also
>                  introduced two unwanted messages in the debug output that I didn't
>                  notice before:
>                  
>                  QMainWindow::saveState(): 'objectName' not set for QToolBar 'Main Toolbar'
>                  QMainWindow::saveState(): 'objectName' not set for QToolBar 'Sampling Bar'
>         
>         Thanks for fixing that!
>         
>         
>                 As this change is so minor, I also added device selection persistence.
>         
>         I think that should be a separate commit.
>         
>         
>                 The device title is the only method of identification available, so I'm
>                  using it here despite its non-uniqueness when using multiple devices of
>                  the same kind.
>         
>         I don't think that's a very convincing solution. How about storing a
>         truly unique identifier for each device instead?
>         
>         Also - what if I move my device to another USB port. If my device has
>         a serial number I would like PV to still open the correct device,
>         even though it's now on a different port.
>         
>         
>         //Peter
>         
>         
>         ______________________________________________________________
>         
>         Want excitement?
>         Manually upgrade your production database.
>         When you want reliability, choose Perforce
>         Perforce version control. Predictably reliable.
>         http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>         
>         ______________________________________________________________
>         
>         sigrok-devel mailing list
>         sig...@li...
>         https://lists.sourceforge.net/lists/listinfo/sigrok-devel
> 
> -- 
> Sent from my Android device with K-9 Mail. Please excuse my brevity.
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________ sigrok-devel mailing list sig...@li... https://lists.sourceforge.net/lists/listinfo/sigrok-devel