<A124>
I did look swiftly through. Depending on where it would and could be used it could speed up lookups in huge datasets. While having compact memory index that is faster then uncompressed one even.
<A124>
And given that many might want to share small files, like http pages, etc, the number of hashes for 10TB of data would be wastly above 40M. Just a thought.
<vtomole>
I'm thinking about starting an implementation of ipfs for common lisp, for netwroking, should i start with using UDP of TCP?
<vtomole>
*networking
<A124>
Not sure what to answer. If you implement UDP transport effectively, you get immense benefits over TCP.
<A124>
Russia -> USA TCP transfer stalls on about 3Mbit. TCP is slow for long distance.
<A124>
You got to have technique for re-sending lost packets, and ordering the packets, unless use use something that does that for you. Congestion control is just simple math that needs to be chosen well. If it was me, I would go UDP.
<lgierth>
vtomole: we don't have a crypto channel yet, and all the networked protocols in ipfs require a reliable transport, so i'd go with tcp
<lgierth>
there's also generally no support for udp based transports in any of the ipfs impls so you wouldn't be able to peer with anybody
<lgierth>
there's utp-over-udp in go-ipfs, but it's broken and needs a bit of love
<lgierth>
we're gonna gradually move to packet switching on top of transports based on udp but nothing of that is usable so far
<lgierth>
(regarding crypto channel, i meant we don't have one which would work without tcp or something else reliable underneath)
Guest78791[m] has joined #ipfs
deltab has quit [Ping timeout: 256 seconds]
deltab has joined #ipfs
qqtqt[dot]qt[at] has joined #ipfs
<vtomole>
Just for getting started, do you recommend that I use HTTP, BitTorrent or Bitswap for exchange? HTTP seems like the easier one.
<lgierth>
there is no other exchange than bitswap in ipfs currently
qqtqt[dot]qt[at] has quit [Client Quit]
lentan has joined #ipfs
<vtomole>
Okay so networking:TCP, routing:Kad Dht exchange:Bitswap should be a good start for libp2p?
<lgierth>
yep sounds good!
<lgierth>
:)
<lgierth>
also check out multiformats and ipld
<vtomole>
The way I understand it, libp2p is composed of networking,routing and exchange, So I shouldn't be thinking that far ahead until I've gotten those 3 thing implemented, correct?
<vtomole>
*things
<lgierth>
multiformats is a building block though
<lgierth>
you'll need at least multiaddr and multihash
<vtomole>
Ahh I see. They are in the whitepaper which is what basing this off of, Thank you :)
<A124>
So.. What are the plans of integrating reliable protocol on top of UDP? Time-wise. Just wondering.
Sigma has quit [Ping timeout: 250 seconds]
soloojos has quit [Ping timeout: 245 seconds]
kaotisk has quit [Ping timeout: 268 seconds]
sbruce has quit [Ping timeout: 250 seconds]
<lgierth>
i have a work in progress go-udp-transport, the blocker is a lack of a secure channel (e.g. dtls) in go
<lgierth>
sometime in Q1 it's gonna be usable
apiarian has joined #ipfs
<lgierth>
and the real fun starts when we can do packet switching and source routing on top of that
soloojos has joined #ipfs
Sigma has joined #ipfs
sbruce has joined #ipfs
kaotisk has joined #ipfs
ylp has quit [Ping timeout: 260 seconds]
<jbenet>
<3 <3 <3
<A124>
Great stuff. Thanks for info.
ylp has joined #ipfs
mguentner has quit [Ping timeout: 250 seconds]
mib_kd743naq has quit [Ping timeout: 260 seconds]
niekie has quit [Remote host closed the connection]
ylp has quit [Ping timeout: 260 seconds]
ylp has joined #ipfs
mguentner has joined #ipfs
anonymuse has joined #ipfs
pfrazee has quit [Remote host closed the connection]
pfrazee has joined #ipfs
niekie has joined #ipfs
pfrazee has quit [Ping timeout: 250 seconds]
mguentner has quit [Quit: WeeChat 1.6]
echo_oddly has quit [Remote host closed the connection]
apiarian has quit [Ping timeout: 252 seconds]
mguentner has joined #ipfs
jonnycrunch has quit [Ping timeout: 268 seconds]
apiarian has joined #ipfs
jonnycrunch has joined #ipfs
Qwertie has joined #ipfs
Qwertie has quit [Read error: error:1408F119:SSL routines:SSL3_GET_RECORD:decryption failed or bad record mac]
Qwertie has joined #ipfs
echo_oddly has joined #ipfs
anonymuse has quit [Remote host closed the connection]
Kamilion has quit [Ping timeout: 248 seconds]
anonymuse has joined #ipfs
gerald` has joined #ipfs
lentan has quit [Remote host closed the connection]
lentan has joined #ipfs
gerald` has quit [Remote host closed the connection]
kanzure has quit [Ping timeout: 260 seconds]
lkcl has quit [Ping timeout: 256 seconds]
lentan has quit []
robattila256 has joined #ipfs
Kamilion has joined #ipfs
jedahan has joined #ipfs
pfrazee has joined #ipfs
anonymuse has quit [Remote host closed the connection]
yoosi has quit [Remote host closed the connection]
yoosi has joined #ipfs
cubemonkey has joined #ipfs
zbyte64 has joined #ipfs
zbyte64 has quit [Read error: Connection reset by peer]
zbyte64 has joined #ipfs
pfrazee has quit [Ping timeout: 244 seconds]
cubemonkey has quit [Ping timeout: 256 seconds]
zbyte64 has quit [Read error: Connection reset by peer]
zbyte64 has joined #ipfs
tmg has quit [Ping timeout: 252 seconds]
Aranjedeath has quit [Quit: Three sheets to the wind]
<Mateon1>
Is there any research on how Kad DHT deals with cutting the network? It seems that part of the provider records would be lost on a split, but I might be misunderstanding the DHT
chriscool has joined #ipfs
ELLIOTTCABLE has quit [Ping timeout: 246 seconds]
henriquev has quit [Ping timeout: 245 seconds]
henriquev has joined #ipfs
ELLIOTTCABLE has joined #ipfs
ELLIOTTCABLE has quit [Excess Flood]
ELLIOTTCABLE has joined #ipfs
ylp1 has joined #ipfs
zbyte64 has joined #ipfs
<Kubuxu>
whyrusleeping, lgierth: I figured out how to rebase with dep change with much less conflicts
<Kubuxu>
you have to diff deps, update and commit same deps as for example master did and then rebase
zbyte64 has quit [Ping timeout: 244 seconds]
ygrek has quit [Ping timeout: 250 seconds]
<victorbjelkholm>
good morning earth!
jonnycrunch1 has joined #ipfs
jonnycrunch has quit [Ping timeout: 246 seconds]
jonnycrunch1 is now known as jonnycrunch
Oatmeal has quit [Quit: Suzie says, "TTFNs!"]
anderspree_ has quit [Ping timeout: 256 seconds]
anderspree_ has joined #ipfs
zbyte64 has joined #ipfs
zbyte64 has quit [Ping timeout: 245 seconds]
<dignifiedquire>
good morning
zbyte64_ has joined #ipfs
sametsisartenep has joined #ipfs
s_kunk has quit [Ping timeout: 260 seconds]
zbyte64 has joined #ipfs
zbyte64_ has quit [Ping timeout: 265 seconds]
zbyte64 has quit [Ping timeout: 258 seconds]
ylp1 has quit [Quit: Leaving.]
ylp1 has joined #ipfs
<daviddias>
morning! :)
<Kubuxu>
morning
Hijiri has quit [Read error: Connection reset by peer]
zbyte64 has joined #ipfs
<victorbjelkholm>
how is everyone doing? Here is raining and generally gray
<victorbjelkholm>
+ ticketmaster is the worst service out there, was waiting to buy tickets that got released at 10, got estimated queue of three minutes and five minutes later, the reached the maximum number of users and kicked us out :/
<victorbjelkholm>
/vent
zbyte64 has quit [Ping timeout: 258 seconds]
<victorbjelkholm>
anyone know which video jbenet is always using in his demos of IPFS? The one about earth
<haad>
hey go-ipfs, how do I get started to make a PR on go-ipfs? I don't understand from the documentation, how exactly I should checkout the project/branch and if I git clone it (as opposed to the usual go get), where is the repo then and how do I 'make' it?
<haad>
Kubuxu: ^ ?
<haad>
lgierth: ^
<rugu>
haad: Just fork the original repository, then git clone the your version of the repo on your comp. make changes and push it to your repo, and finally click on "Pull Request" and a pull request will appear on the original ipfs repository, referencing your code
zbyte64 has quit [Ping timeout: 258 seconds]
<Kubuxu>
haad: you need to have the go path seted up
<Kubuxu>
or use go get -d and change the remote
<Kubuxu>
to be your own
<Kubuxu>
or the ipfs's as you should have access
<Kubuxu>
to make it, just `make build`
<Kubuxu>
output is binary `cmd/ipfs/ipfs`
<rugu>
Kubuxu: Hey, are you involved in ipfs-libp2p-js or any of its dependencies?
<Kubuxu>
I am not on the js side of things
<Kubuxu>
haad might know a bit more
<rugu>
ohk, I wanted to understand libp2p more and CAD routers
<haad>
Kubuxu: so I now have go-ipfs branch wher the code I changed uses "gx/ipfs/Qmd6gKBjErWcfJLJ22wDJu1z2EqbFKb4wheNobZJtSxf8M/floodsub" which has also been modified. how do I make my go-ipfs code use my version of floodsub?
espadrine has quit [Ping timeout: 250 seconds]
<Kubuxu>
you have to publish your pubsub using `gx publish`
<haad>
Kubuxu: ok, now trying to 'make' go-ipfs in $GOPATH/src/github.com/haadcode/go-ipfs and get the following error:
<haad>
go-ipfs must be built from within your $GOPATH directory. expected within '/Users/haad/go' but got '/Users/haad/go/src/github.com/haadcode/go-ipfs'. make: *** [path_check] Error 1.
<haad>
Kubuxu: what to do?
taaem has quit [Read error: Connection reset by peer]
taaem has joined #ipfs
Oatmeal has quit [Ping timeout: 258 seconds]
anonymus_ has quit [Remote host closed the connection]
<victorbjelkholm>
haad: usually I just do `go get github.com/ipfs/go-ipfs` then `cd $GOPATH/src/github.com/ipfs/go-ipfs` then `make deps` and `make install`, that should verify you can use the project
zbyte64 has joined #ipfs
<victorbjelkholm>
haad: what's the output of `echo $GOPATH` ?
<haad>
victorbjelkholm: I did go get -u -d github.com/haadcode/go-ipfs
<haad>
victorbjelkholm: works fine when I pull the master (github.com/ipfs/go-ipfs)
vtomole has joined #ipfs
zbyte64 has quit [Ping timeout: 250 seconds]
<haad>
lgierth hsanjuan Kubuxu haaaaaalp
hpk has left #ipfs ["WeeChat 1.4"]
tmg has quit [Ping timeout: 250 seconds]
<lgierth>
ok
<lgierth>
you clone go-ipfs to $GOPATH/src/github.com/ipfs/go-ipfs
<hsanjuan>
haad: once you manage that, edit .git/config to point to your fork (or do via git remote... etc)
<Kubuxu>
haad: you have to be in github.com/ipfs/go-ipfs path
<haad>
aha! so the trick here is to change your git config to the fork. this is not documented anywhere.
zbyte64 has quit [Ping timeout: 244 seconds]
Oatmeal has joined #ipfs
lkcl has quit [Read error: Connection reset by peer]
<lgierth>
huh?
<lgierth>
oh that
kvda has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<lgierth>
it's the same dance as with any github fork
<lgierth>
except you must have it under github.com/ipfs/go-ipfs locally because go's import mechanism demands that
<haad>
right, so actually the trick is not to change the git path but *to clone to that specific directory, github.com/ipfs/go-ipfs* (lgierth this is very different from other projects)
<haad>
yup
<haad>
it'd be great to add a note to the docs about that
<haad>
it's a unique requirement for this particular project (or maybe it's a go thing?)
<lgierth>
it's a go thing
<lgierth>
and i'm pretty sure it's in the readme
<lgierth>
oh you mean in the context of a fork... yeah...
<Kubuxu>
whyrusleeping was worried about the streaming implementation having possible bugs, so we talked about it and agreed that fuzzy tests that prove that it gives the same output when streaming in chunks vs monolithic encryption by /x/crypto would increase his confidence greatly so I wrote this: https://github.com/davidlazar/go-crypto/pull/2/files
<lgierth>
Kubuxu: i have no idea about the crypto -- i can review the non-crypto things but for the crypto things i yield
<daviddias>
dignifiedquire: understood, I guess my question is: can you account to have that? +1000 if it is easy to change (like picking it from README or something)
<dignifiedquire>
right, yeah I'm pretty sure I can account for that and pull it in easily from markdown
bastianilso__ has quit [Remote host closed the connection]
<haad>
dignifiedquire: not having a front page like daviddias says above was one of the reasons I wasn't happy with documentation.js, so if that's possible, I'd definitely take another look
Guanin has joined #ipfs
Guanin has left #ipfs [#ipfs]
bastianilso__ has joined #ipfs
<dignifiedquire>
haad: sorry, you are not going to get dangling commas into our code base from my side :P
<haad>
dignifiedquire: any reason for that?
<dignifiedquire>
I left a comment, the main one is we would have a direct contradiction with standard
<dignifiedquire>
our current rules are only additions, not changes to standard
<haad>
dignifiedquire: "personally very opposed" is not a reason. I hear your argument re. standard, but you need to give more than "I don't personally like it".
<dignifiedquire>
I think the standard one is enough
<dignifiedquire>
the personal thing is just a note
kaotisk has quit [Read error: Connection reset by peer]
<haad>
keks: whyrusleeping: what floodsub got merged? please keep in the loop with anything pubsub as I'm working on it on JS side trying to get it work by this weekend for the hackaton
espadrine has joined #ipfs
<Kubuxu>
whyrusleeping: I will push my code and just take a look if it is good path
<whyrusleeping>
haad: keks multisubscribed PR that we've been working on for ~mo
<Kubuxu>
inb4 debugging, check is your tests are correct
<Kubuxu>
whyrusleeping: my problem with that is blakeX can produce any length of output
<Kubuxu>
X=B 1 to 64
<whyrusleeping>
haad: we've opted against having an 'unsub' call in pubsub a few different times
<Kubuxu>
and X=S 1 to 32
<whyrusleeping>
Kubuxu: hrm, but the multihash has the length embedded in it
maxlath has quit [Quit: maxlath]
zbyte64 has joined #ipfs
<haad>
whyrusleeping: did you read what I described on #ipfs-dev? I totally understand why it wouldn't be a good thing but as of yesterday things were not working in pubsub and that was the only way I can be sure my tests in js-ipfs-api are correct. haven't checked yet how keks' PR changes that but there's a tricky part there with closing the connections AND making sure the caller (ie. js-ipfs-api via HTTP API) can be sure that on system level the subscription has been
<haad>
so hopefully what keks did solves that. I can check and verify it against the tests easily :)
<haad>
(or dunno if that was even what you guys were working on)
<keks>
haad: I didn't see what you posted. can you copy a log somewhere
<pjz>
I got worried that daylight savings time had messed with my timing :)
herzmeister has quit [Quit: Leaving]
<haad>
pjz: I highly recommend to use/think in UTC times (UTC/Z): https://github.com/ipfs/pm/issues/258, we've improved a lot since we decided to switch using UTC times only :)
herzmeister has joined #ipfs
<pjz>
haad: well, sure, but that doesn't help scheduling lunch with non-geeks :P
ne1 has quit [Quit: rage quit]
<haad>
oh, sorry! the apps call is in 40mins (1830UTC), js-ipfs is in 10mins
<pjz>
ah, okay
<dignifiedquire>
haad: re including custom text in documentation.js, it's possible just needs some love in the design area and some small bug fixes
<haad>
keks: so on the pubsub issue: when this ctx.Done() happens, the http connection to the caller has been closed, and so the caller gets a "connection closed" on their side. however, at that point, Floodsub.Unsub(topic) hasn't been called, so within floodsub (the system state) it might still think that it's subscribed to the topic. so what happens then is that if that same caller tries to subscribe to the same topic again, they'll get "already subscribed" even thou
<haad>
dignifiedquire: ooooooh! that looks much better! \o/
<dignifiedquire>
haad: wait until my I'm done with the theme upgrade :)
<haad>
dignifiedquire: where are those About/Example/Quick Start/etc chapter written? in some code file or readme or smth?
<dignifiedquire>
currently you have to write them in "documentation.yml"
<dignifiedquire>
as markdown, or plain text
<haad>
dignifiedquire: ok. well yeah even that's fine.
<haad>
dignifiedquire: could it be documentation.md, then? :)
<dignifiedquire>
but I'm going to try to add an option to just point to markdown files
PseudoNoob has quit [Read error: Connection reset by peer]
PseudoNoob has joined #ipfs
<keks>
haad: i see. well the good news is that "already subscribed" is a thing of the past. also i think go-ipfs#3402 cleans up after itself, telling floodsub to cancel the subscription
<haad>
keks: it's already "cleaning up", the Floodsub.Unsub(topic) on the line below does that. the problem is that the API caller doesn't know that...
<haad>
keks: ...or rather that it happens "async" and after telling the caller API the connection has been closed
<haad>
keks: does that make sense?
<keks>
haad, floodsub.Unsub() is bad because it assumes you only subscribe to a topic once. I changed the floodsub API. sorry you were out of the loop but at the time I thought you were more of a js/client-sie person :(
<keks>
there is no floodsub.Unsub() anymore
<haad>
keks: is there a PR related to your go-floodsub/pull/10 in go-ipfs I could take a look?
<haad>
will take a look later tonight / tomorrow on all this. thank you.
<keks>
again, sorry for the communication messup :/
kaotisk has quit [Ping timeout: 252 seconds]
<haad>
keks: absolutely not a problem, you've been on this a lot longer and *I* didn't know and talk to you about it when I started looking into it on Friday. absolutely not your fault.
jedahan has quit [Ping timeout: 250 seconds]
kaotisk has joined #ipfs
ygrek has joined #ipfs
s_kunk has quit [Ping timeout: 268 seconds]
<Kubuxu>
jbenet: thanks for the feedback
<haad>
richardlitt: can you announce with your template the apps call?
<Kubuxu>
lgierth: in their from, separate canary, not transparency report, it is a form that they got gag order
<haad>
thank you everyone!
<jbenet>
thanks everyone! <
<jbenet>
<3
<victorbjelkholm[>
thanks! Fun to see pinbits pjz ! :)
<jbenet>
dignifiedquire victorbjelkholm[ is https://github.com/jbenet/depviz interesting to you at all? i imagine much of this may be super trivial to either of you
<dignifiedquire>
jbenet: it's super interesting, but there are so many interesting things :D
<victorbjelkholm[>
yeah, it is! Seems super interesting and would love to help out, I'm afraid I'm running out of time during the days though, so probably I can help out later...
se3000 has joined #ipfs
cubemonkey has quit [Ping timeout: 250 seconds]
<jbenet>
yeah that sounds good :)
<victorbjelkholm[>
jbenet: is there a runnable demo somewhere based on the current PR?
<Kubuxu>
victorbjelkholm[: I think currently it is graphical mockup
<victorbjelkholm[>
or maybe it's too early
<jbenet>
victorbjelkholm[ yeah-- clone the pr, cd webapp; npm start
<Kubuxu>
but I am not sure
<jbenet>
my branch (jbenet-webapp or something) has a different one too
<victorbjelkholm[>
jbenet: oh, awesome, I'll give it a try later today
<jbenet>
but wking's is much farther along so i'm going to push the styling changes from mine into his PR.
cubemonkey has joined #ipfs
<jbenet>
i dont understand why all node.js tooling is so slow. is it that spinning up v8 every time sucks? (like jvm?)
<jbenet>
nvm. i dont want to start that discussion.
<wking>
jbenet: you should probably base your work on wking/react-stubs (which is behind #11)
zbyte64 has joined #ipfs
<jbenet>
wking: thanks! will do that.
<jbenet>
whyrusleeping: we should have an easy way to review gx published versions and changes. how do i know the new go-crypto published by gx at QmaPHkZLbQQbvcyavn8q1GFHg6o6yeceyHFSJ3Pjf3p3TQ is not owned. we need a tool toadvertise new versions, and diff them.
<whyrusleeping>
gx diff <old> <new>
<jbenet>
yay. but where's the root?
<whyrusleeping>
root?
<jbenet>
<oldest>
jedahan has joined #ipfs
jedahan has quit [Client Quit]
palkeo has joined #ipfs
palkeo has quit [Changing host]
palkeo has joined #ipfs
zbyte64 has quit [Client Quit]
<whyrusleeping>
uh... i don't do three way diffs
<Kubuxu>
jbenet: clone git and publish on your own
<jbenet>
no need for three way, i can do individual diffs all the way back. what i'd like is some way to see the history of versions all the way back
<jbenet>
Kubuxu: yeah that's one way to deal with it in this case, i'll probably do that
<whyrusleeping>
oooh, its all in git. but thats a little more annoying to do i guess
<whyrusleeping>
Yeah, the only changes i made to the crypto lib past pulling it down is deleting one subpackage that had an external dep, and removing all the import directive require statements (otherwise we cant import things as gx/ipfs/...)
<whyrusleeping>
but please do verify thats correct
jedahan_ has quit [Ping timeout: 244 seconds]
<whyrusleeping>
I didnt want to update the crypto lib but the old one didnt have blake
<whyrusleeping>
alternatively, we could *just* publish the blake2 subpackage of the crypto lib
<Kubuxu>
I use also salas20 package from it
<whyrusleeping>
brb, planes taking off
<dignifiedquire>
jbenet: node tooling is slow because it takes a decent amount of work + knowledge to make it fast
<dignifiedquire>
it's not fast by default
<jbenet>
dignifiedquire: no the issue i'm talking about is that even this program: `console.log()` is too slow. the startup time is the big delay i'm annoyed by.
<dignifiedquire>
ah right, yeah thats v8 startup + parsing all the things
<dignifiedquire>
if you had an instance running that would just feed the file it would be a lot faster
<jbenet>
whyrusleeping gx diff is nice :)
<wking>
victorbjelkholm[, Kubuxu: pushed a fresh build of my depviz tip to https://wking.github.io/depviz/, if you want to look without running locally. Dragging around the viewBox is next, but I'm not familiar enough with react-dnd to have that working yet
<jbenet>
I think Kubuxu is on to something deeper -- what i want is a way to authenticate that this gx package comes from a given git hash, but ideally without having to do a gx publish myself. hm, maybe what i want is something like `gx-git diff <git-hash> <gx-hash>`
<jbenet>
basically, my verification steps should be: (1. verify go-crypto <git-hash> is now owned (which i can do by looking at Go's release process), and 2. verify go-crypto <gx-hash> is for <git-hash>)
maxlath has quit [Ping timeout: 244 seconds]
<jbenet>
Kubuxu whyrusleeping: what git tag does QmaPHkZLbQQbvcyavn8q1GFHg6o6yeceyHFSJ3Pjf3p3TQ come from? (so i can publish myself and verify that)
<Kubuxu>
jbenet: it might be master
<Kubuxu>
as it wasn't updated in 21 days
<jbenet>
Kubuxu whyrusleeping: master of golang/go? or is there an x/ package? can you point me directly to the tag i need to verify? (sorry if i spend time hunting this down, i cant spend time reviewing other things)
bastianilso__ has quit [Ping timeout: 260 seconds]
<jbenet>
whyrusleeping review done
bastianilso__ has joined #ipfs
jdloft has joined #ipfs
ianopolous_ has joined #ipfs
bastianilso__ has quit [Ping timeout: 258 seconds]
ianopolous has quit [Ping timeout: 256 seconds]
bastianilso__ has joined #ipfs
dlight has joined #ipfs
jokoon has joined #ipfs
bastianilso__ has quit [Ping timeout: 265 seconds]
<vtomole>
I'm not a js hacker, but I am following js-ipfs as an example for common lisp ipfs. I'm currently trying to run js-libp2p test, and I get this error when i run "npm test at the project directory:"sh: 1: dignified-test: not found".How do i fix this?
vtomole has quit [Quit: Page closed]
vtomole has joined #ipfs
maxlath has joined #ipfs
bastianilso__ has joined #ipfs
mildred has joined #ipfs
<lgierth>
vtomole: i figure dignifiedquire will know
tmg has joined #ipfs
<dignifiedquire>
vtomole: js-libp2p is outdated and not maintained, you will want to look at ipfs/js-libp2p-ipfs
<vtomole>
Thanks, how do i run the tests?
scde[m] has joined #ipfs
<dignifiedquire>
should be just git clone && npm install && npm test
<hsanjuan>
whyrusleeping: do you have RPis running ipfs? how's the experience?
ianopolous__ has joined #ipfs
fleeky__ has quit [Remote host closed the connection]
fleeky has joined #ipfs
ianopolous_ has quit [Ping timeout: 244 seconds]
evernite- has quit [Quit: ZNC 1.6.1 - http://znc.in]
rendar has quit [Quit: Leaving]
jdloft has quit [Ping timeout: 258 seconds]
bastianilso__ has quit [Quit: bastianilso__]
ulrichard has joined #ipfs
trvon has joined #ipfs
Encrypt has quit [Quit: Sleeping time!]
atrapado_ has joined #ipfs
maxlath has quit [Ping timeout: 256 seconds]
slothbag has joined #ipfs
PseudoNoob has quit [Remote host closed the connection]
jokoon has quit [Quit: Leaving]
<vtomole>
+dignifiedquire: When I run node-1.js, I get "Error: Missing peerId. Use Peer.create(cb) to create one" Do i have to code in a peerId?
grosscol has quit [Quit: Leaving]
ashark has joined #ipfs
cemerick has quit [Ping timeout: 260 seconds]
kvda has joined #ipfs
ashark_ has quit [Ping timeout: 252 seconds]
pfrazee has quit [Remote host closed the connection]
trvon has quit [Ping timeout: 245 seconds]
jdloft has joined #ipfs
espadrine has joined #ipfs
anonymuse has quit [Remote host closed the connection]
mildred has quit [Quit: WeeChat 1.5]
ulrichard has quit [Remote host closed the connection]
chriscool has quit [Ping timeout: 268 seconds]
pfrazee has joined #ipfs
anewuser has joined #ipfs
ashark has quit [Ping timeout: 250 seconds]
anonymus_ has joined #ipfs
Kane` has quit [Remote host closed the connection]
slothbag has quit [Ping timeout: 250 seconds]
anonymus_ has quit [Remote host closed the connection]
espadrine has quit [Ping timeout: 260 seconds]
slothbag has joined #ipfs
slothbag has quit [Ping timeout: 240 seconds]
slothbag has joined #ipfs
<Kubuxu>
whyrusleeping: can I extract dep changes from pnet PR and PR them separately? I don't want to resolve conflicts and it will take few more days.
robattila256 has quit [Ping timeout: 245 seconds]
<lgierth>
Kubuxu: sure definitely
<lgierth>
+++
atrapado_ has quit [Quit: Leaving]
<Kubuxu>
Hmm, I might just ran through and cleanup the deps in general.
<Kubuxu>
kvina will problably kill me
<Kubuxu>
kevina
anonymuse has joined #ipfs
PayasR_ has joined #ipfs
<lgierth>
maybe make it two commits, one for the pnet related imports and one for the rest
<Kubuxu>
yeah
<Kubuxu>
I am slowly going through it as I touch codebase
<Kubuxu>
when I finish cleaning it up I want to have goimport -d test in CI
<Kubuxu>
goimports -d
<PayasR_>
@jbenet, @whyrusleeping is anyone working on the ipfs/papers?
<Kubuxu>
I hoped to extend goimports to handle our style of gx imports {stdlib, local lib, gx}
<Kubuxu>
in order with separators
<Kubuxu>
but goimports is complex
<lgierth>
PayasR_: we're not working on updated papars *right now*, but we'll generally expand and update them soon
<Kubuxu>
it will be easier to to just sed it or something
<lgierth>
the joys of code gardening :')
<Kubuxu>
I might just write a quick tool which will test that imports are in correct order.
<Kubuxu>
also /me claps for why for 'context "context"'
<lgierth>
:)
_mak has quit [Remote host closed the connection]
peterix has quit [Remote host closed the connection]
<lgierth>
for anything that is not in the stdlib, having the imported named is really useful
<lgierth>
eeh s/useful/readable/
PayasR_ has quit [Ping timeout: 248 seconds]
<kevina>
Kubuxu: what's up? I need to eat will be back in 30 minutes or so if it can wait
anonymuse has quit [Ping timeout: 256 seconds]
anonymuse has joined #ipfs
<lgierth>
kevina: reordering imports all over the codebase
Boomerang has quit [Ping timeout: 245 seconds]
<kevina>
Kubuxu: well yeah that will create conflicts, but I could live with that if we document the new standard.
<kevina>
Kubuxu lgierth: and I will assume this is done in a way that gofmt won't reorder then again.
<kevina>
gotta run, be back in 30 or so....
<lgierth>
yeah gofmt doesn't cross double-newlines between imports
<Kubuxu>
yeah that is the point
<Kubuxu>
as currently some imports aren't even in order
<Kubuxu>
lgierth: where should I write the go import guideline?
<Kubuxu>
devtools? or community?
<lgierth>
put it in community for now
<lgierth>
all the js guides are there too
martinBrown has quit [Remote host closed the connection]
peterix has joined #ipfs
martinBrown has joined #ipfs
_mak has joined #ipfs
<seharder>
Kubuxu: I was excited to see your work with private networks today. When do you think it will be ready.
<Kubuxu>
seharder: it is functional, I am doing final UX adjustments and waiting on CR
<Kubuxu>
I am really want to have it done before the end of a week.