mark not matching bytes red in received packet

[fridolinsiegmund]

Hi,
it would be great to determine which bytes exactly are not matching between the received and expected packet with "graphical support", therefore i propose this solution where the not matching bytes at the received packet are marked red.

[fridolinsiegmund]

Hi @antoninbas and @jafingerhut ,
I updated my PTF fork to the most recent commit in the main branch in this repository.
Any updates on my pull request? I still think having the wrong bytes marked red helps debugging.

Best Wishes,
Fridolin

[jafingerhut]

I have not tried out your changes yet, but do they insert the terminal escape sequences that change color on some kinds of virtual terminals unconditionally? That is, if I were to run PTF in batch mode with output saved in a text file, would the escape sequences be inserted there, too?

Having some kind of command line option to enable/disable the insertion of these escape sequences seems better to me than enabling them unconditionally. I'm not sure what the default should be.

I believe Linux commands like ls often somehow automatically determine whether their output is to a terminal or not at run time, and insert colorization when their output to a terminal, but do not insert colorization when their output is to a file or pipe. That seems preferable to me.

[jafingerhut]

I have attempted to send an email to the address ipu.ci.triage.team@intel.com, but I got back an error message containing this text. Fun! :-)

Address not found
Your message wasn't delivered to ipu.ci.triage.team@intel.com because the address couldn't be found, or is unable to receive mail.
The response from the remote server was:550 #5.1.0 Address rejected.

[antoninbas]

@jafingerhut if you can't get a hold of anyone at Intel, I think it is possible to block the bot account at the p4lang-org level, so that it can no longer comment.

[jafingerhut]

@antoninbas Good to know, in case that turns out to be useful. There are a couple of people within Intel looking into this starting yesterday or today -- hopefully they can see what they can do at their end.

[grg]

@antoninbas @jafingerhut Intel is working to address this. The comments on the open-source repos were unintentional: there working on adjusting tools to only comment on internal repos, not the p4lang open-source ones.

They also plan to have someone delete the unintentional comments from the p4lang repos.