[Kst] [Bug 136780] New: Add quick curve creation dialog to plot RMB

[Nicolas Brisset]

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
         
http://bugs.kde.org/show_bug.cgi?id=136780         
           Summary: Add quick curve creation dialog to plot RMB
           Product: kst
           Version: unspecified
          Platform: unspecified
        OS/Version: Solaris
            Status: NEW
          Severity: wishlist
          Priority: NOR
         Component: general
        AssignedTo: kst kde org
        ReportedBy: nicolas.brisset eurocopter com


Version:           1.3.2_devel (using KDE 3.4.0, compiled sources)
Compiler:          gcc version 3.4.3
OS:                SunOS (sun4u) release 5.8

As announced on the list, this is the first of a few improvement suggestions coming from extensive discussions with users as to what they find cumbersome in the current UI.

This suggestion aims at streamlining one of the most common operations: adding one or more curve(s) to a given (or new) plot. Sure, the datawizard can do it and it is also possible to do it by creating a new vector, then a new curve. But all these processes take many clicks, and require the user to know which plot name corresponds to what contents (which is already a showstopper in many cases). And that's not all: as I described in bug #124942, adding curves from more than one datafile (e.g. when comparing two similar files) is very painful. 
Now imagine you can right-click on a plot, call a "quick curve" entry, and get the dialog attached to this report. You select the var(s) to add, click "Add to current plot" and you're done. Very easy :-) And the best is you can add curves from more than one source easily (for a more thorough discussion on this issue see bug #124942 as I won't repeat the whole argument here). In fact, this proposal is mostly a stripped-down datawizard, with a multi-file extension (and some ideas from gaiw). From the user feedback I get here, it would be a very useful addition usability-wise !

Some points are worth noting: 
- computing the intersection of variable names should be done intelligently as it can take very long (remember, we routinely use files with tens of thousands of vars). Probably with QDict or QMap (I never know) ?
- to speed things up we should cache the common variable list as it is likely that users will not change selected datasources so often
- selected (checked) datasources should be remembered
- the X vector combobox should contain the list of X vectors currently in use in the selected plot, provided that the corresponding field exists in all selected datasources. The X vectors should be reused.