Let's talk about errors and warnings.

[GeneFC]

I think that PR misses the point.

There is no error with "CheckGeometry" itself, but there is an error shown in the result from "CheckGeometry".

Do you really want to suppress that error from the majority?

Gene

[vocx]

There is no error with "CheckGeometry" itself, but there is an error shown in the result from "CheckGeometry".

I think the idea is that an error should be reported only when the function itself has problems and cannot proceed. In this case, the function works correctly.

That the shape has an error is a different matter. That error is the result of the analysis of the shape, and as such, that analysis should be reported in the task panel window, but it should not be displayed as an error in the terminal, because no error was encountered in the execution of the command itself.

[chrisb]

I tend to agree with this paradigm, and disagree a bit with chrisb. In my opinion, the error should be raised whenever a tool or command cannot proceed due to some cause. It does not have to be a catastrophic error that needs to be reported on the forum, it just has to be something that stops the user from proceeding. So in this case, yes, we should instruct the user the proper workflow with certain messages.

I can very well accept such messages in the cases you describe. But we often see the accusation that FreeCAD's error messages are poor and the user guidance needs improvements.
So in a perfect world we would not have some more or less program internal error message in the report view, but either a wonderful and clear error dialog, or - even better - it would be impossible to reach such state while leaving all the freedom to the user.

That's what I mean, that it should be clear, that this is not the final state, albeit very well acceptable - even for years to come.

[sgrogan]

Change my mind.

The application checked for the availability of an "optional" library. If Path handles this gracefully its a warning "this is not possible, install ocl"

If ocl becomes a dependency and its not there, then its an error.

[dubstar-04]

Milling time should not be a warning, that could be an easy to implement relief, if you don't auto show report view on normal messages.

Related Post:
https://forum.freecadweb.org/posting.ph ... 5&p=405437

current logs:

No Tool Controller = Error - ("No Tool Controller selected.")
No Feed rate = Warning - ("Tool Controller feedrates required to calculate the cycle time.")
No Rapid Rates = Warning - ("Add Tool Controller Rapid Speeds on the SetupSheet for more accurate cycle times.")

Maybe the rapid rate shouldn't be a warning but IMO the other two are correct.

Thoughts?

[chrisb]

Milling time is shown in the Job itself, so a warning should indeed be issued if the number shown there is probably wrong. However, as said in the other post I would prefer a different technique for retrieving the milling time only on demand.
Another possibility would be to make the whole sum of milling times invalid if a single part is invalid. I don't know if this is possible without raising other errors.

[herbk]

The warnings itselve are OK.
The Problem is that FC opens the output panel although i have disabled "show warnings".

[Wsk8]

Disable "show report view on warning/error"

mfg

[herbk]

Disable "show report view on warning/error"

mfg

like said: it is disabled

[chrisb]

Disable "show report view on warning/error"

That's acceptable as a shorttime workaround for certain situations, but it can't be a solution.