Thorsten

I would consider the sc-32 handling in the spy as broken, or to the least not finished. Firstly, I just spent a day to find a reliable source how to calculate the sc-32 correctly. The standard has an example value for "123456789", which is wrong, as it relates to the AUTOSAR crc, which from what I understand, uses different data reflections ... I think. The testcase in tcnopen isn't any help either, as it is not checking against an expectation. Couldn't find any examples for Profisafe neither, and...

@John I cannot remember whether this was a bug that I fixed. Alternatively, did you accidentally set the dissector preference? You can choose between LE and BE interpretation, see screenshot 2. If this does not work as in the screen shots, you'll have to wait for the spy to respawn with the latest fixes in upstream-wireshark. It may be 4.7 or 4.8 ... And thanks for providing the attachments.

MD messages look better in the latest updates (once they go public). I think it might had an issue with packet reassembly.

I believe this is fixed once the latest updates go upstream in wireshark

You can add a whole folder of XML files now - this also requires a bit of discipline. Available: https://github.com/T12z/TCNopen/tree/master/trdp/spy/plugins/3.6/epan

Done: Plugin for 3.6 available providing support for multiple XML files and trying to be helpful finding inconsistencies. https://github.com/T12z/TCNopen/tree/master/trdp/spy/plugins/3.6/epan

Please do not checkin build artifacts

Didn't notice this. I am not actively listening. I am currently switching to wireshark 3.6 and will pull your/Florian's fix.