Re: [PATCH #8316] Variable Editor

[John W. Eaton]

On 02/20/2014 02:00 AM, Daniel J Sebald wrote:

> Yes, before continuing some words of caution.  I don't want to
> discourage you from working on this, but at the same time it probably
> would be best to hold off perfecting things until the issue of sending
> data from one thread to another is worked out.
>
> John Eaton made a good start on a variable editor, so you may want to
> see if there is anything in common there at some point.

I thought my solution was quite clumsy and required way too much copying 
of data and translating data structures from things that Octave uses 
(octave_value, etc.) to things that Qt understands (QList or whatever).

> In my view, I think that really all that is need is a way to process an
> Octave command in the "background", i.e., the result doesn't appear in
> the command window.  (And of course, the results are sent back to the
> GUI via signal using octave_value pointer...I've don't this, removing
> callbacks, seems to work.)  For example, when a file is opened in the
> editor, the GUI could request in the background all the breakpoints for
> that file.  There is such an Octave script/command already.

My original reason for going with the octave-qt-link thing was that I 
hoped to create an interface between Octave and a GUI that would say 
"provide these functions and your GUI will be fully functional".  But 
I'm not sure that it's possible or even reasonable to try to enforce the 
requirements of a GUI this way.  I agree that we need a way for the GUI 
to say "Hey, Octave, run this function and, optionally, return a result 
to me."  Since Octave is not thread safe, we need a way to make these 
function calls in a safe way.  Similarly, Octave may need to have a way 
to notify the GUI that something has changed when the GUI did not 
specifically request the change (current working directory, for example).

How should functions called through this mechanism pass data?  The 
natural thing for Octave is octave_value and octave_value_list.  But 
that means passing these objects across threads, doesn't it?  If so, is 
Michael Goffioul's reference counting change the best solution?

Dan, now would probably be a good time to take another look at the 
changes you proposed last year.  Can you tell me where they are again? 
I thought they were in the patch tracker but I don't see them now.

> In the case of the variable editor, I think after the completion of a
> command the editor just asks for the contents of the variable and
> updates those that have changed, then that is all that is needed.  But I
> would suggest to do so efficiently.  Rather than asking for all
> variables, just ask for those displayed, e.g., x(6:30,17:25) if it is
> only those that are displayed.

Agreed, because you can only display a small segment of a Matrix at once 
on any display, no matter how large.  But we would still like to be able 
to edit variables with millions of elements.

jwe