Segfault on creating square filled zone with python interface

I am unable to recreate this. Do you still observe the same behavior?

Application: Pcbnew
Version: (5.1.0-1256-g39954b20f), release build
Libraries:
    wxWidgets 3.0.4
    libcurl/7.64.0 OpenSSL/1.1.1c zlib/1.2.11 libidn2/2.0.5 libpsl/0.20.2 (+libidn2/2.0.5) libssh2/1.8.0 nghttp2/1.36.0 librtmp/2.3
Platform: Linux 4.19.0-5-amd64 x86_64, 64 bit, Little endian, wxGTK
Build Info:
    wxWidgets: 3.0.4 (wchar_t,wx containers,compatible with 2.8) GTK+ 3.24
    Boost: 1.67.0
    OpenCASCADE Community Edition: 6.9.1
    Curl: 7.64.0
    Compiler: GCC 8.3.0 with C++ ABI 1013

Build settings:
    KICAD_SCRIPTING=ON
    KICAD_SCRIPTING_MODULES=ON
    KICAD_SCRIPTING_PYTHON3=ON
    KICAD_SCRIPTING_WXPYTHON=ON
    KICAD_SCRIPTING_WXPYTHON_PHOENIX=ON
    KICAD_SCRIPTING_ACTION_MENU=ON
    BUILD_GITHUB_PLUGIN=ON
    KICAD_USE_OCE=ON
    KICAD_USE_OCC=OFF
    KICAD_SPICE=ON

My apologizes, I didn't see this response earlier.

I have just tested with:
  * Version: (5.1.0-rc2-24-gdb31b7902)
  * Version: (5.1.0-1305-gc84c10fa0)
both with python 3. Full version info tested is attached.

I can confirm that I do *not* see this bug with either of these versions and can't replicate.

However, the original bug I reported was with a 6.0 pre-release (r15606) which I was testing back then and no longer have access to.

If you could point me to the equivalent 6.0 branch for testing I would be happy to build it and test again. I am confused about the branches and version numbering.

Just speculating here, but my hunch is that this is probably something to do with detecting if a shape is open or closed or something similar; perhaps failing on some exact case with exactly equal successive coordinates in one dimension. If that is the case, then would that be something handled by OpenCascade? I notice that the OpenCascade version has bumped up a lot (6.9.1 -> 7.3.0) between the build I saw the bug with and now. So perhaps they have fixed it.

(5.1.0-1305-gc84c10fa0) is the newer version of the 6.0 development that you were using previously.

The major difference looks like Python version here. This doesn't crash in Python3, which is used in the new Ubuntu builds.

We should test with Python2 before marking this fixed.

In fact the script is incorrect.
a working script (although not a very good script) is:

import pcbnew
from pcbnew import FromMM
board = pcbnew.GetBoard()
zone = pcbnew.ZONE_CONTAINER(board)

sps = zone.Outline() #pcbnew.SHAPE_POLY_SET()
chain = pcbnew.SHAPE_LINE_CHAIN()
chain.Append( FromMM(0.0), FromMM(0.0) )
chain.Append( FromMM(20.0), FromMM(0.0) )
chain.Append( FromMM(20.0), FromMM(20.0) )
chain.Append( FromMM(0.0), FromMM(20.0) )
chain.SetClosed(True)
sps.AddOutline(chain)
#zone.SetOutline(sps)
board.Add(zone)
pcbnew.Refresh()

Incorrect lines are commented.
The crash is for me something like a double deletion:
AddOutline() is an accessor to a private variable, and its use here is fully incorrect.

KiCad bug tracker has moved to Gitlab. This report is now available here: https://gitlab.com/kicad/code/kicad/-/issues/1750