<samueldr>
>> A patch commit message MUST consist of a single short (less than 50 characters) line stating the problem (“Problem: …") being solved, followed by a blank line and then the proposed solution (“Solution: …").
<samueldr>
uh
<samueldr>
that's extremely tight with the problematic function name
<samueldr>
and funnily enough, none of the commit messages from the repo follow that
<samueldr>
danderson: it's not examples, it's test cases under the test suite
<samueldr>
they're validating specific cases, e.g. age restricted VEVO video
<samueldr>
all of VEVO is music
<samueldr>
and not like you uploading music
<samueldr>
but big cartel music
<samueldr>
so (1) it's not even proeminent!
<samueldr>
and (2) not documentation
<samueldr>
though they can definitely remove the test cases
<samueldr>
thought it's not great not testing specifics
<samueldr>
so yeah, until we see the obvious rehosts like debian's salsa being acted against, I don't think we'll see build instructions being acted against
<clever>
samueldr: i have seen a github project taken down by automation before, because they hosted source outside of github, and compiled binaries on github (for weird reasons)
<clever>
and the github automation thought it was malware, too many binaries, not enough src
cjpbirkbeck has quit [Quit: Goodbye, take care]
<samueldr>
since I saw an issue on the Nixpkgs repo of people being all worried, I wrote up something
<samueldr>
basically, we can carry on for a while until upstream gives news
<samueldr>
we could even apply patches while using our old source if need be!
<samueldr>
(please tell me if I made factual errors in that comment)
<MichaelRaskin>
Should we make a statement that youtube-dl is the recommended way of downloading the CC-licensed NixCon recording from Youtube?
<MichaelRaskin>
BTW, does anyone still clone NixOS repos with all PRs included? I guess if there is a couple of clones, then a GitHub takedown won't be that much of a loss
<gchristensen>
sphalerite: that is how you know it is good code
<sphalerite>
gchristensen: didn't you have the same problem with something the other day?
<gchristensen>
sorta yea
<gchristensen>
someone in #systemd has been trying to get them to add easter monday support to systemd timers for 4 days
<sphalerite>
uuuuh okay
<gchristensen>
12:32 <gchristensen> what do you find very important about being able to use systemd list-timers to see the next easter monday?
<gchristensen>
12:33 <spiri> it would be nice to get a message "let the bunny hop" as a notification
<gchristensen>
y
<sphalerite>
gchristensen: help why am I feeding the troll
<gchristensen>
lol
avn has quit [Ping timeout: 260 seconds]
maxdevjs has joined #nixos-chat
<MichaelRaskin>
When I am in mood for a lot of pain, I should try pushing an RFC on recognising the fact that tristate logic describes half of the meta stuff better than booleans.
<gchristensen>
yeah
<MichaelRaskin>
Is Easter Monday even computable from databases that a normal system ships?
<MichaelRaskin>
Isn't it tied to a judgement call whether the full moon moment is properly ascribed to one week or the next, or something like that?
<gchristensen>
something like that
cole-h has joined #nixos-chat
<hexa->
first monday after the first sunday after the first full moon in spring
<__monty__>
So if the full moon is on a sunday in spring, it's not the day after but a week after?
<hexa->
apparently
<__monty__>
(Having trouble with *first* monday after *first* sunday.)
<hexa->
oh right, my bad
<hexa->
easter is the first sunday after the first full moon in spring
<hexa->
easter monday is day++
<hexa->
so it is the first monday after that first sunday
<sphalerite>
hexa-: you mean ++day? :p
<hexa->
obviously
<hexa->
we wouldn't make our easter calculations unefficient, right?
<hexa->
s/un/in/
<hexa->
we wouldn't wanna make our easter calculations inefficient, right?
<hexa->
fixed.
<sphalerite>
well, day++ isn't just inefficient, it's incorrect ;)
<immae>
It still doesn’t answer the question of whether "after" is inclusive or not. If the full moon happens on "sunday Xth", does "first sunday after" correspond to "sunday Xth" or "sunday (X+7)th"
<__monty__>
I think the latter, because otherwise you'd just say first monday after full moon, no?
<__monty__>
Unless first monday after first sunday was really supposed to be first monday after easter sunday.
<immae>
which is as much imprecise as the previous formulatoin :)
<sphalerite>
easter was a terrible idea.
<__monty__>
immae: Well, after isn't on in my book.
<MichaelRaskin>
sphalerite: it kind of began well! So they come to the empty grave. A bit of a reference to Ananda's enlightenemnt, I guess. Then it got worse.
Baughn has joined #nixos-chat
Baughn has quit [Client Quit]
Baughn has joined #nixos-chat
<infinisil>
,launch
<{^_^}>
Ping for space stuff: infinisil Taneb ldlework etu philipp[m] eyJhb
<infinisil>
In 30 minutes, stream starts in 15 minutes, SpaceX starlink :D
<infinisil>
I'll probably miss this one
<philipp[m]>
Same, Vuelta is more important while cooking!
<joepie91>
(the trick is to look for alternatives to some obscure VCS, not alternatives to git :P)
<hexa->
displays a calendar and the date of Easter
<hexa->
man cal(1)
<hexa->
The problem is apparently solved
<hexa->
The program that you likely know as "cal" is itself a clone, named "ncal" and written by Wolfgang Helbig in the 1990s. It was preceded by a clone written for BSD by Kim Letkeman in 1989, which was in turn a replacement for a "cal" written by Bill Joy.
<samueldr>
MichaelRaskin: by ordering by last update and only *completing* your info, I have a feeling that *with a login* you can run it maybe every 15 minutes
<samueldr>
MichaelRaskin: maybe even tighter loops
<samueldr>
MichaelRaskin: though my implementation uses octokit which handles waiting for you if needed
<samueldr>
MichaelRaskin: so you could do it in tight loops, but then you're burning requests when there might not have been any activity
<eyJhb>
What.. The.. Hell...
<eyJhb>
Holy hell gchristensen , I didn't know that openconnect was SO WELL. Not really used to that...
<gchristensen>
huh?
<eyJhb>
Meant to sent that before I connected to the VPN
<gchristensen>
:)
<eyJhb>
Got DC'ed from IRC when I connected to the VPN, so it didn't send the message :)
<gchristensen>
ah :)
<eyJhb>
Hoping I can do some nice iptables + VPN stuff, to only send specific ranges down the VPN...
<gchristensen>
well... yeah, so that script I sent you is written the way it is so that ONLY the RDP program is using the VPN
<MichaelRaskin>
samueldr: do you know if anyone runs mirroring?
<samueldr>
I don't know
<samueldr>
WHY oh why did I not leavea README with notes about how to use my own damn tool?
<MichaelRaskin>
Docs are for weaklings
<cole-h>
gchristensen: Hmm... that script looks familiar :P
<samueldr>
hi, it me, weakly remembering how I did things
<MichaelRaskin>
And that happens weekly?
<eyJhb>
gchristensen: Well, I guess testing for today is over since I apparantly got banned for DoS on the authservice
<samueldr>
MichaelRaskin: that was a word I tried to pun on, but the puns were weak
<MichaelRaskin>
… at that time of the week
<drakonis>
__monty__: soon we'll have nickel...
<drakonis>
but are we really going to still have nixpkgs or nipkgs?
<cole-h>
nclpkgs
<kraem>
anyone had any experience with a streacom db4 chassi? :)
<samueldr>
oh, I totally had forgotten that the script I use tries to handle direct pushes on branches!
<__monty__>
drakonis: I'm secretly still rooting for dhall : )
<samueldr>
and uh, it's definitely not working well since I ran it in the wrong order
<samueldr>
it's telling me
<samueldr>
20.03..20.09: 21363 commits on master, (99.34%) 21506 total commits.
<samueldr>
since I accidentally broke fetching the data beforehand :)
<samueldr>
the amount of people that are just willy-nilly mirroring and sharing mirrored unverified youtube-dl around is... not going to be any help in the end :|
<samueldr>
how many of those are trustable?
<samueldr>
how many of those are accidentally going to be big SEO-wise?
<samueldr>
so many reactionary responses that may harm the project in the end :/
<gchristensen>
I'll trust any of them as long as their 2020.09.20 release hashes to 1pkw3hnkddk1kqv0in152q1k4jjgbmf2xvc9j3r5nd38z6f7j6mc
<gchristensen>
btw, the sharing the source is silly since it is availableon the youtube-dlwebsite :)
<gchristensen>
nice to get copies around but the reactionary sharing is silly since the website is unaffected
<samueldr>
yeah
<samueldr>
(though it wasn't serving downloads yesterday)
<gchristensen>
oh?
<samueldr>
(as they were served through github)
<gchristensen>
ah
<samueldr>
but yes, wait for upstream
<hexa->
I wonder who will pick up that legal battle though
<samueldr>
hopefully the EFF can, and knows how to help
<supersandro2000>
he maintains a fork since a bit of time
<samueldr>
still doesn't look official to me, but I don't know
<samueldr>
though I don'T know who were the owner/authors!
<etu>
It appears that emacs has changed the response data of (directory-files-recursively) between 26 and 27... Fix was easy but meh. Took time to find that. :/
<samueldr>
there's no reason to see bad intents, but that doesn't look authoritative
cjpbirkb1 has joined #nixos-chat
<etu>
__monty__: The type was the same, it was just the difference that I got a relative path instead of a full path so my regex didn't handle it properly.
numkem has joined #nixos-chat
cjpbirkbeck has quit [Ping timeout: 258 seconds]
<__monty__>
samueldr: Maybe they were the person behind youtube-dlc?
<__monty__>
etu: Fair enough. Though you could differentiate relative and absolute paths with types ; )
<samueldr>
yeah, but that's not youtube-dl :)
<samueldr>
so, the opinion of "just a fork"
<__monty__>
Well many people were already using youtude-dlc for some reason. If the HN thread is anything to go by.
<__monty__>
Just saying maybe it's not *just* another fork.
<samueldr>
what I meant by "just a fork" is "not a fork in reaction to the DMCA issue", a genuine fork
<samueldr>
but still, not the upstream
cjpbirkbeck has joined #nixos-chat
<__monty__>
Ah.
cjpbirkb1 has quit [Ping timeout: 240 seconds]
cjpbirkb1 has joined #nixos-chat
xd1le has quit [Read error: Connection reset by peer]
cjpbirkbeck has quit [Ping timeout: 240 seconds]
cjpbirkbeck has joined #nixos-chat
xd1le has joined #nixos-chat
tilpner has quit [Remote host closed the connection]
tilpner has joined #nixos-chat
cjpbirkb1 has quit [Ping timeout: 258 seconds]
cjpbirkb1 has joined #nixos-chat
cjpbirkbeck has quit [Ping timeout: 240 seconds]
tilpner has quit [Ping timeout: 240 seconds]
cjpbirkbeck has joined #nixos-chat
cjpbirkb1 has quit [Ping timeout: 240 seconds]
tilpner has joined #nixos-chat
naivettes has joined #nixos-chat
__monty__ has quit [Quit: leaving]
<naivettes>
Hello! Beginner here. I'd like to add support for add-ons to the tome4 game. Add-ons are meant to be downloaded to into a directory within the game, but it's in the nix store. https://github.com/NixOS/nixpkgs/tree/master/pkgs/games/tome4
<naivettes>
Could anyone recommend a good example of how this might be done?
<samueldr>
naivettes: I would suggest you ask on #nixos, this channel is for off-topic uses :)
<naivettes>
thanks!
naivettes has left #nixos-chat [#nixos-chat]
hmpffff has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
supersandro2000 has quit [Disconnected by services]
supersandro2000 has joined #nixos-chat
<samueldr>
yay 502 - Server Error (Octokit::BadGateway)
<MichaelRaskin>
Yay. Can you hold it down now?
<samueldr>
it was a sarcastic yay
<MichaelRaskin>
I mean, ability to bring down Github cheaply and at will could be interesting!
<samueldr>
looks like it went through the second time around
<samueldr>
luckily, since GitHub is not a paragon if stability all of it is built assuming things will fail
<samueldr>
paragon of*
<MichaelRaskin>
Does it also assume that eventual consistency is very eventual?
<samueldr>
yes
<samueldr>
well
<samueldr>
what do you have in mind?
<samueldr>
there is no real consistency in the database here
<samueldr>
it's a collection of factoids that happen to generally relate together
<samueldr>
and the name of the database file shows what I have in mind
<samueldr>
it's named: cache.db
<samueldr>
the main thing you trade-in if you delete it, si time for a later re-hydration
<samueldr>
[if we forget about issues and people vanishing off GitHub]
<MichaelRaskin>
Well, consequtive loads can give you sets of stuff that are not included in either direction, even without deletion
<samueldr>
care to expand?
<samueldr>
I think it's handled through just using whatever state is currently given by GitHub for any PR at any point in time
<MichaelRaskin>
On bad days the changes seems to propagate… weirdly