Arne Welzel

> like f-strings They are in 3.6, luckily.. >Unless we need the new features, I am tempted to leave things as they are. While the python versions are EOL -...

From @0xxon over in #3021 : > In random questions - how difficult would it be to, instead, fix the scripts that currently require proxies to just work if we...

Quick skim indicates that feature uses the `Analyzer::ports` table to create a BPF filter. `ports` is populated via `register_for_ports()` so likely same reason as #3318. Current idea is to discourage...

This one timed out for me last week as well, misses `doc.zeekygen.identifier`. https://cirrus-ci.com/task/5446879278792704 Seems this is happen fairly regular and somehow `doc.zeekygen.*` related.

> We also use subprocess.check_call which also has a timeout argument. Hmm, skimming `btest` there are other places. For example, i think this isn't safe at all: https://github.com/zeek/btest/blob/master/btest#L379-L382 If a...

Hey, I think you're spot on. Including `fuid` in these scripts is little useful today. @0xxon , any thoughts on removing `fuid` from the `NOTICE()` invocations and possibly unconditionally adding...

> The reason for that would be if we think this could be useful in other circumstances where we have file hashes available for notices. :-) My proposal for `x509_fingerprint`...

@philhagen - thanks for pointing all this out. Linking #3160 as it's tangentially in the same area and came up recently. Wonder if that could be an overarching topic for...

My thoughts are to either deprecate `port` within `.evt` files, update the package template to render the boiler-plate and move on. The alternative could be for the spicy integration to...

I'm closing this. `ports` support in `.evt` has been deprecated for Zeek 7.0 and Tim is in the process of ripping out support for 7.1 :-)