First zellij log file format parser

[Zykino]

Hello,

I’m new to lnav and it looks like a very good project that I would have a hard time to learn since it does not understand the log format of tools I’m developing against regularly. So I went ahead and created a "parser" (not sure how you call them) for zellij which is a terminal multiplexer (a bit like tmux or screen but different).

I’m seeing 2 log variant and asked if there are others. I also asked if a "producer id" (lnav's opid-field) could be added to quickly filter the server, each clients and each plugin’s instances. I also asked if they have better name for some capture groups (module and module2 don’t feels good 😅)

Anyway I have this 1st implementation where I have some questions/issues:

1. I filled the "file-pattern" field with the most common log file path. I’m not sure it is necessary (or even a good idea), since almost no other formats appear to have it.
2. The cause_body field is always null, I’m not sure why since multiline default to true. This is an optional field of multi-line log that regex101.com correctly detected.
   1. This field contains a backtrace (or close enough). At first I wanted to cut it in lines or event components <cause_id>: <cause_log>. Is this useful in your experience or too much? I did not achieved it on regex101.com, since I’m not sure how dynamically increment capture group’s names.

Note: I did not compile this branch, I’m hoping that test done using the file installed in ~/.config/lnav/formats/installed give the same results as compiled in.

[Zykino]

I also read that lnav can "go to next/previous" linux’s boot, it can be interesting to go to next/previous zellij’s startup. I’m not sure how it is done. Can you point me in the right direction?