Re: workspace viewer layout

[Daniel J Sebald]

On 04/08/2013 11:27 AM, John W. Eaton wrote:
> Wile modifying the workspace connection/model/viewer, it occurred to
> me that it does not add much to have the separate collapsible lists
> for local, global, and persistent variables.
>
> The viewer is almost always displaying info for the top-level symbol
> table which can't have persistent values and probably rarely has many
> global values. It seems to me that it would make more sense to add a
> separate column to indicate global status.
>
> Would it be difficult to make the workspace view use a table instead
> of the tree view thing? I looked at it, but I'm not familiar enough
> with the Qt model classes and it wasn't obvious to me how to do it.
> Some help with this job would be much appreciated.

Not speaking about the layout, I think something has been lost here. 
The old code would correctly display the "argn" inside of a function, 
now it just shows ...

The behavior of the Workspace isn't quite what it should be.  I suggest 
some color coding of the row background color as a nice feature.  My 
thought on a GUI is that it needs to bring something that enhances and 
improves on CLI capabilities.  For example, the right mouse click 
"plot", "stem", sort of thing is kind of nice.  It could be more.  I 
think what I'd like is a right click inside the editor that will open a 
window with the help text in it.  Something that could be left on the 
screen while editing text.

Well, back to Workspace.  It would be nice to have a setup that shows 
variables according to context.  I've tried a little program with 
breakpoints.  Before entering the function, the variable space looks 
like the first image below (I'm going to try placing small PNGs here out 
of curiosity how that works through the list).  x is a vector of length 
fifty.

Upon entering the function and hitting a breakpoint, then stepping a few 
lines, the Workspace looks like the second image below.  x is now a 
"local" variable in the sense of its context and displays the value 
4.5000, because I used a variable x inside of the testfun() routine. 
It's easy enough to understand, but still it would be nice to retain the 
actual "x" namespace variable as well as the local function variable.

The third PNG is an example that uses some color to illustrate the 
different variables.  Could maybe even be some indentation on the 
variables that inside a function; maybe that is where the tree could be 
of use.

Dan

outside_function_call.png
Description: PNG image

inside_function_call.png
Description: PNG image

color_code_workspace.png
Description: PNG image