00:35
alcymd has quit [Ping timeout: 272 seconds]
01:07
sir_guy_carleton has joined #nixos-chat
01:28
lassulus_ has joined #nixos-chat
01:29
jtojnar has joined #nixos-chat
01:30
lassulus has quit [Ping timeout: 252 seconds]
01:30
lassulus_ is now known as lassulus
01:48
jtojnar has quit [Ping timeout: 252 seconds]
02:45
jtojnar has joined #nixos-chat
03:45
matthewbauer has quit [Ping timeout: 240 seconds]
03:45
matthewbauer has joined #nixos-chat
04:08
matthewbauer has quit [Ping timeout: 252 seconds]
05:13
Lisanna has quit [Quit: Lisanna]
05:40
<
sphalerite >
jasongrossman: thank you :)
06:02
jackdk has joined #nixos-chat
07:00
sir_guy_carleton has quit [Ping timeout: 268 seconds]
07:03
lopsided98 has quit [Ping timeout: 240 seconds]
07:04
lopsided98 has joined #nixos-chat
07:54
jackdk has quit [Ping timeout: 244 seconds]
10:00
__monty__ has joined #nixos-chat
10:47
jasongrossman has quit [Read error: Connection reset by peer]
11:03
jasongrossman has joined #nixos-chat
11:08
jasongrossman has quit [Ping timeout: 252 seconds]
11:10
nckx has quit [Ping timeout: 240 seconds]
12:18
nckx has joined #nixos-chat
12:25
Synthetica has quit [Quit: Connection closed for inactivity]
13:07
jasongrossman has joined #nixos-chat
13:45
Synthetica has joined #nixos-chat
14:14
sphalerite has quit [Ping timeout: 250 seconds]
14:16
sphalerite has joined #nixos-chat
16:11
* sphalerite
wonders if there is a human who can remember C's function pointer syntax
17:09
jtojnar has quit [Remote host closed the connection]
17:10
Lisanna has joined #nixos-chat
17:13
jtojnar has joined #nixos-chat
17:32
<
infinisil >
sphalerite: Having seen Haskell type syntax, anything else doesn't make any sense anymore
17:36
<
sphalerit >
infinisil: sure, but even go or rust syntax is just worlds better than C as far as referring to functions/procedures is concerned
17:37
<
infinisil >
I'm a bit concerned about go's syntax regarding the new proposal
17:37
<
infinisil >
The contracts thing
17:38
<
infinisil >
Goddamnit
17:38
<
infinisil >
Try finding the go contracts proposal by searching for "go contract" or similar
17:40
<
gchristensen >
samueldr: so what is next on the release process?
17:40
<
samueldr >
go figure (heh), search engine company's project, totally ungooglable
17:40
<
samueldr >
gchristensen: looking at the PR that fixes the last blocker, then it should be good to go AFAIK
17:40
<
gchristensen >
yay!
17:41
<
infinisil >
Example of proposed syntax: "func Keys(type K, V Equal(K))(m map[K]V) []K"
17:41
<
samueldr >
(sorry, first thing that popped in mind seeing that many k and m together)
17:42
<
LnL >
sphalerit: blocks in objective-c are even worse
17:42
<
infinisil >
Another gem: "func Transform(type T1, T2)(lst *List(T1), f func(T1) T2) *List(T2)"
17:43
<
infinisil >
Compare that to Haskell: [T1] -> (T1 -> T2) -> [T2]
17:44
<
sphalerit >
C++ lambdas are pretty bad too
17:44
<
gchristensen >
hmm I think I could make the docs I'm writing link to the definition...
17:45
<
samueldr >
gchristensen: that would be lovely
17:47
<
samueldr >
hey, in case anyone's curious, #47842 is the PR; I'll be taking a closer look this evening after work. (pretty limited scope, but now's not the time to test it in full)
18:41
<
jasongrossman >
Fastly++
18:41
<
{^_^} >
Fastly's karma got increased to 1
20:30
Lisanna has quit [Quit: Lisanna]
20:31
jtojnar has quit [Remote host closed the connection]
20:32
jtojnar has joined #nixos-chat
20:32
matthewbauer has joined #nixos-chat
20:38
<
samueldr >
("that link" I think isn't the one written here, but the link inside the documentation linking to the function)
20:40
jasongrossman has quit [Ping timeout: 246 seconds]
20:42
<
gchristensen >
samueldr: do you have display/formatting suggestions for that?
20:43
<
gchristensen >
it looks pretty lame as-is, and I don't want to do it for all those functions without feeling like it is pretty good
20:43
matthewbauer has quit [Ping timeout: 264 seconds]
20:45
<
samueldr >
I'll need to dig into docbook shenanigans to know what could be done
20:45
<
gchristensen >
well if you just have ideas on what it looks like, I can try and map to docbook
20:46
jtojnar has quit [Read error: Connection reset by peer]
20:46
<
samueldr >
is this all manually mapped?
20:46
<
gchristensen >
to line locations?
20:46
<
gchristensen >
oh dear no
20:47
<
samueldr >
I didn't really doubt, but just had to know for sure
20:49
<
samueldr >
already, a good thing would be to write on the same line
*what* this filename + line number is, something like "Location in <nixpkgs>: ..."
20:49
<
gchristensen >
and then to bring in the link in the docs, you use <xi:include href="./locations.xml" xpointer="lib.attrsets.mapAttrsRecursive" />
20:50
<
samueldr >
for a tool initially made to browse hypertext document, it sure has issues browsing that hypertext document
20:50
<
samueldr >
(the gist)
20:50
<
gchristensen >
lol.
20:51
matthewbauer has joined #nixos-chat
21:05
matthewbauer has quit [Ping timeout: 240 seconds]
21:11
<
gchristensen >
s/ion/ed/
21:12
Lisanna has joined #nixos-chat
21:14
Lisanna has quit [Client Quit]
21:15
<
samueldr >
yeah, looks more like "something"
21:15
<
samueldr >
not a floating link
21:17
<
gchristensen >
I wonder if putting the type signature in the title is too much / obscuring it too much
21:18
<
gchristensen >
ehhh I should stop futzing and get it merged :)
21:19
<
LnL >
did you create w pr already? can't remember
21:19
<
{^_^} >
#47837 (by grahamc, 19 hours ago, open): nixpkgs: Start documenting library functions in XML
21:21
<
LnL >
let's get that in then :)
21:22
* gchristensen
can't stop tinkering
21:23
<
LnL >
you don't have to
21:23
<
gchristensen >
right. right.
21:24
<
LnL >
also how would you create a stable link to the implementation?
21:24
<
LnL >
not sure if the commit available
21:25
<
gchristensen >
commit isn't certain to be available :/
21:25
<
gchristensen >
I could use commit if possible, and a branch reference if not?
21:27
<
LnL >
what branch? :p
21:27
<
LnL >
but yeah, a fallback might be fine
21:28
<
LnL >
I'm just not sure if the commit is available in hydra builds
21:28
<
gchristensen >
release-${nixpkgs.lib.trivial.release}
21:29
<
LnL >
but probably, channels must be doing something similar
21:31
<
gchristensen >
yeah it is available
21:31
<
gchristensen >
(ish)
21:46
<
gchristensen >
LnL, samueldr: is it critical I link to the hash, or is linking to the branch sufficient?
21:46
<
gchristensen >
if it isn't sufficient, I won't include this linking part in this PR
21:48
<
LnL >
well, it's pretty unlikely for something to move files
21:48
<
gchristensen >
a few lines off isn't a big deal I think
21:49
<
LnL >
yeah, slightly confusing but there's a reasonable chance it's still on screen
21:50
jtojnar has joined #nixos-chat
21:59
<
gchristensen >
I can easily add some mild smarts
21:59
<
gchristensen >
I'll go that route first
22:02
Lisanna has joined #nixos-chat
22:08
__monty__ has quit [Quit: leaving]
22:14
<
gchristensen >
ehhh it is very annoying to add mild smarts, so I'm only going to do good-enoughlinks
22:22
<
samueldr >
gchristensen: the hash will cause issues with local builds
22:22
<
gchristensen >
bingo :)
22:22
<
samueldr >
the release branch will always match
22:22
<
samueldr >
tough with varia according to local changes
22:23
<
samueldr >
(I had the though process earlier today)
22:23
<
samueldr >
thought*
22:23
* samueldr
is more tired from the cycle commute than expected
22:27
<
samueldr >
(to prevent hijacking #nixos-dev) LnL those URL paths suspiciously look like nix store paths
22:28
<
samueldr >
I'm wondering how worrying it could be :)
22:29
<
LnL >
it's a separate store
22:29
<
samueldr >
yeah, I was writing "coming from you it probably is fine"
22:30
<
LnL >
nix-store --store 'local?root=/path/to/prefix' --add-fixed sha256
22:31
<
LnL >
this way I can upload files using nix
22:32
matthewbauer has joined #nixos-chat
22:41
<
samueldr >
can't see any reason to change anything there
22:41
<
samueldr >
so yeah, looks good
23:04
<
gchristensen >
guhhh using the release name makes broken URLs for unstable
23:05
<
gchristensen >
since that says 19.03
23:06
<
samueldr >
gchristensen: just had a slow cache
23:06
<
samueldr >
it may be too late to diagnose :/
23:07
<
samueldr >
do you perchance have a new poke-and-prod-fastly-diagnostics.sh?
23:07
<
samueldr >
though now I'm doubting since
*nothing* loads
23:09
<
gchristensen >
no I don't
23:10
<
samueldr >
I think gremlins are doing something weird here
23:10
<
samueldr >
this SSL TCP connection with my own server works fine for chatting, 17ms lag which is nominal
23:11
<
samueldr >
but uh, can't load sites :/ so probably not a fastly issue
23:13
<
samueldr >
gremlins! I say
23:14
* samueldr
turns it off and on again hoping it will help
23:17
<
samueldr >
it helped
23:45
jackdk has joined #nixos-chat
23:47
matthewbauer has quit [Remote host closed the connection]
23:55
jtojnar has quit [Ping timeout: 268 seconds]