[08:05] <mup_> PR operator#246 opened: Made breakpoint handling code to use already parsed JUJU_DEBUG_AT, and isolated the message showing logic <Created by facundobatista> <https://github.com/canonical/operator/pull/246>
[11:01] <jam> "morning" all
[11:04] <Chipaca> moin moin
[11:04] <niemeyer> o/
[11:34] <facubatista> Muy buenos días a todos!
[11:49] <Chipaca> facubatista: heyyy
[11:49] <facubatista> hola Chipaca
[11:49] <Chipaca> facubatista: q'acelga caezn
[11:50] <facubatista> jejej
[11:55] <jam> morning facubatista  and Chipaca
[11:55] <jam> "morning" :)
[11:55] <facubatista> hola jam
[11:57] <facubatista> Chipaca, juju (maybe) never will set JUJU_DISPATCH_PATH as absolute, but a user may; if it's absolute, are we ok with it?
[12:24] <jam> facubatista, are you sure we don't handle abspath correctly? Because if I do:
[12:24] <jam> 0
[12:24] <jam> https://paste.ubuntu.com/p/7rdWThBDjc/
[12:25] <jam> base = Path('/var/run')
[12:25] <jam> print(base / '/something/else')
[12:25] <jam> I just get
[12:25] <jam> '/something/else'
[12:25] <jam> so self.charm_dir / dispatch_path
[12:25] <jam> will give an abspath if it is set
[12:26] <facubatista> jam, it depends on what "correctly" means... if we're ok with using an abspath if it's given, yes, we're handling it correctly, but then it bothers me a previous comment saying that the path "will be relative"
[12:27] <jam> facubatista, so Juju will always pass a relative path
[12:27] <facubatista> jam, the user may not
[12:27] <jam> facubatista, but that is only if someone is intentionally not treating the object as a Juju charm and running something direcly
[12:27] <facubatista> FTR, I'm +1 to improving the comment
[12:27] <jam> in which case, they can do what they need to
[12:28] <facubatista> let's fix the comment so when somebody reads it don't get confused with the code
[12:31]  * Chipaca updates the comment to say "if the path is not relative, the absolute value of the path's eigenvector is used to compute its state in P-space before proceeding"
[12:36] <jam> Chipaca, what about fourier domain ?
[12:36] <Chipaca> so first fourier transform, *then* get the eigenvector?
[12:36] <Chipaca> seems reasonable to me
=== __chip__ is now known as Chipaca
[13:59] <jam> o/ Chipaca
[13:59] <jam> so chip is chipaca but isn't chip...
[14:00] <Chipaca> Chipaca, chipakeitor, __chi__, ChipaPhone, and ChipAway are all grouped to me
[14:50] <jam> "has been a joy to work with" \o/
[14:50] <niemeyer> !!!
[15:17]  * facubatista -> lunch
[15:48]  * facubatista -> *really* lunch now
=== Dmitrii-Sh0 is now known as Dmitrii-Sh
[16:52]  * facubatista is back
[17:01]  * Chipaca goes to have dinner
[17:04] <jam> heading out, bbiab
=== mup__ is now known as mup_
[18:06] <jam> back
[18:30] <Chipaca> facubatista: you there?
[18:30] <facubatista> Chipaca, yes, with my head inside shotcut, but heading to the meeting
=== Dmitrii-Sh1 is now known as Dmitrii-Sh
[19:18] <Chipaca> woooooooooooo⁰⁰⁰⁰⁰⁰⁰⁰⁰⁰
[21:43]  * facubatista eods