Error messages in "graph.cpp" slow down debugging

[ldpgh]

While doing debug using QtCreator and latest commit ....

• My Cartesian diagram is showing just a part (5n...6n) of the simulated data range (0n->6n)
• In this context the console-output is flooded with error message. The messages are associated to negative values given/returned by the listed functions in graph.cpp
  • void Graph::ScrPt::setScrX(float x)
  • void Graph::ScrPt::setScrY(float x)
  • float Graph::ScrPt::getScrX() const
• Those function are called very very often if the diagram is redrawn (e.g. clipping algorithm).

It's a heavy performance drop in case of debugging.
Normal use case (non-debug) ...

• latest commit & release build
  • built executable started by QtCreator environment ... behaves similar to the debugging build, but just stderr is printed (qdebug is suppressed)
  • already built executable, but started from cmd.exe ... does not print anything to the commandline
  • already built executable, but started from C:\msys64\mingw64.exe ... prints like the debug session.
• latest release (0.0.19) @ Win7Pro ... does not print anything to the commandline

Test schematic: test_negativeScreenCoordinates.sch.

[in3otd]

I think there is already a PR including a fix for this, see #314 (comment) and #370 , please check if that works ok .
(yes, sometimes all those debug messages slow down the graphs redraw quite a lot)

[ldpgh]

Now I'm definitely nervous ... maybe just confused ...
PR #314 is closed, but I do not see any of the changes made in 2015 while cloning the repo 2016/Nov (same for successor PR #370)
PR #370 may not easily merge with PR #631 ... (changes in rect3ddiagram.cpp)

Btw. ... have seen already graph-segments drawn outside of the regular 2D Cartesian diagram region. Unfortunately the test schematic has been thrown away.

[in3otd]

#314 was closed "by mistake" and #370 is its continuation, so the code there was not yet merged anywhere. My comment above was just to say that the debug messages should be reduced once that PR is merged.

I don't remember seeing segments out of place in 2D graphs, please open an issue if you happen to see that again.