sphh

Oh, well I managed to construct a test case, which fails: ```python3 import numpy as np from geomdl import fitting from geomdl.visualization import VisMPL def ellipse(x, a, b): """Elliptic function."""...

Well, it's true, that I used elliptical sections, section widths and thickness here. But this was only used to construct the simple test case and in real life different sections...

@portnov: As mentioned earlier, this is just a testcase, the actual case is much more complicated. So let us not get side-tracked by figuring out, how the test case can...

Back to the original problem … I managed to create a singularity. I succeeded, but needed another change in the `fitting.py` code. The first one is mentioned in https://github.com/orbingol/NURBS-Python/issues/132#issuecomment-862412256, which...

Well, monkey-patching is always an option, until the upstream source is changed :wink:

I tried your approach and removed `[:]`. That also works for me. Thanks!

Ok, I can see your point. But I am already at MPL 3.4.2 and do not want / cannot downgrade. Oh, I guess, I have to live with that warning...

Here is one of my iCalendar files. It was created by Thunderbird/Lightning: ``` BEGIN:VCALENDAR VERSION:2.0 PRODID:-//PYVOBJECT//NONSGML Version 1//EN BEGIN:VTIMEZONE TZID:Europe/London BEGIN:STANDARD DTSTART:20001029T030000 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10 TZNAME:GMT TZOFFSETFROM:+0100 TZOFFSETTO:+0000 END:STANDARD BEGIN:DAYLIGHT DTSTART:20000326T010000 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3...

I can confirm, that it works with read access to the root collection as suggested by @Unrud. Please, can this information be added to the https://radicale.org/rights/ documentation? Thanks.

Thanks for considering this a missing feature. Since I use a `dtype`, disabling the Qt widget would already be a good indication, that the input is not used. How would...