KiCad Crashes on reloading deleted footprint

[Ted Strassburg III]

Deleting a footprint from a library in the Footprint Editor, and saving , while Pcbnew is open causes KiCad to crash when selecting to place another footprint in Pcbnew. The previously used footprint no longer exists and can not be shown in Recently Used footprints by Pcbnew. This causes all open KiCad modules to close with no warnings or errors.

This occurs when Pcbnew is open at the time of editing or closed then reopened to add another footprint.
This is repeatable, as I've just done it several times trying to see if there was a way around the issue.

Application: kicad
Version: (5.1.2)-1, release build
Libraries:
    wxWidgets 3.0.4
    libcurl/7.61.1 OpenSSL/1.1.1 (WinSSL) zlib/1.2.11 brotli/1.0.6 libidn2/2.0.5 libpsl/0.20.2 (+libidn2/2.0.5) nghttp2/1.34.0
Platform: Windows 8 (build 9200), 64-bit edition, 64 bit, Little endian, wxMSW
Build Info:
    wxWidgets: 3.0.4 (wchar_t,wx containers,compatible with 2.8)
    Boost: 1.68.0
    OpenCASCADE Community Edition: 6.9.1
    Curl: 7.61.1
    Compiler: GCC 8.2.0 with C++ ABI 1013

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

[John Beard (john-j-beard)]

I think that's a high because you can crash an actively edited file quite easily just by opening the dialog.

[jean-pierre charras (jp-charras)]

Hi John:

All crashes, especially when confirmed, can be flagged as critical.

[John Beard (john-j-beard)]

Whoops, didn't see that changing! Damn...

[jean-pierre charras (jp-charras)]

Is my fix not working?

[John Beard (john-j-beard)]

It is working. I was trying to claim it, but didn't refresh this page, and LP doesn't tell you you're overwriting things that have been changed since you loaded the issue.

[John Beard (john-j-beard)]

Hope you don't mind, I also pushed my patch that I was working on when I changed the status.

It's basically the same fix, but it keep symmetry with the same check in eeschema (and prevents any future users of historyInfos from catching a nullptr unexpectedly).

I'll keep the Critical status in mind!