Support for COMPARAM-SUBSET

[kayoub5]

No description provided.

[kayoub5]

@andlaus I can add a test by using the sample PDX from https://github.com/sebastianwilczek/CANoe-Configurations/tree/master/Ethernet/Diagnostics/EthernetDiagnostics/ODX

Creating a full com param configuration using python sounds like an overkill.

[andlaus]

@andlaus I can add a test by using the sample PDX from https://github.com/sebastianwilczek/CANoe-Configurations/tree/master/Ethernet/Diagnostics/EthernetDiagnostics/ODX

if @sebastianwilczek can confirm that the file can be distributed under the MIT license, yes. If we can't get that affirmation, a second repository (odxtools-additional-tests?) could be created for that purpose. (this repo must not be affiliated with mercedes-benz, though.)

Creating a full com param configuration using python sounds like an overkill.

I agree. at least in the short and medium term ;)

[sebastianwilczek]

Hey everyone, not sure about the context here, but the linked repository is only a modified version of the configuration files used by CANoe, developed and owned by Vector Informatik GmbH: https://www.vector.com/int/en/download/?tx_vecdownload_downloaddetail%5Bdownload%5D=92521

As far as I can tell, they do not list any form of license (in context of MIT or otherwise), so unfortunately I can not tell you whether you may use any of the files in your processes.

[andlaus]

I tested this PR with our internal files written by candela studio, and it seems that after this PR, the send and receive IDs (ecu.get_send_id() and ecu.get_receive_id()) are swapped...

[andlaus]

@sebastianwilczek: okay, I suppose we should ask vector then. I'll keep you updated.

[kayoub5]

I tested this PR with our internal files written by candela studio, and it seems that after this PR, the send and receive IDs (ecu.get_send_id() and ecu.get_receive_id()) are swapped...

should be correct now

[kayoub5]

@andlaus I think you forgot to commit the ISO_11898_2_DWCAN.odx-cs file, among others

[kayoub5]

@andlaus FYI, you push directly into a fork if the author enabled it (checked "Allow edits by maintainers")

So feel free to push changes directly to my branch.

[andlaus]

@andlaus I think you forgot to commit the ISO_11898_2_DWCAN.odx-cs file, among others

oops. I had some unpacked somersault.pdx file lying around in the directory where I tested this, i.e. I loaded these fragments from the incorrect location. should be fixed now: kayoub5#3

[andlaus]

So feel free to push changes directly to my branch.

thanks for the offer, but I think it is less confusing if I don't do this (I always get slightly dizzy if some of my PRs change without my intervention...)

[andlaus]

btw, I now get the following warning:

/home/and/src/odxtools/odxtools/diaglayer.py:453: OdxWarning: Communication parameter `CP_UniqueRespIdTable` specified more than once. Using first occurence.

I assume that this is caused by the fact that the ECU can be interacted with using multiple protocols (DoIP and CAN). I guess that fixing this requires a full understanding of the protocol concept and should thus be left for another day?

[andlaus]

ok, I've tested this with our internal files, and it seems to work without the enable_candela_workarounds hack. can you remove all traces of this blemish?

[kayoub5]

ok, I've tested this with our internal files, and it seems to work without the enable_candela_workarounds hack. can you remove all traces of this blemish?

We can do that in a different PR, this one is too loaded as it is

[kayoub5]

btw, I now get the following warning:

/home/and/src/odxtools/odxtools/diaglayer.py:453: OdxWarning: Communication parameter `CP_UniqueRespIdTable` specified more than once. Using first occurence.

I assume that this is caused by the fact that the ECU can be interacted with using multiple protocols (DoIP and CAN). I guess that fixing this requires a full understanding of the protocol concept and should thus be left for another day?

You can cross check this by checking that the different com params point to different proto stacks

It could even be different CAN buses.

[kayoub5]

@andlaus anything blocks merging this?

[andlaus]

@andlaus anything blocks merging this?

no. nice work, thanks!