<dignifiedquire>
trying to finish the integration of your new network code
<daviddias>
sweet :D
<dignifiedquire>
but it turns out to be tricker than I hoped
<daviddias>
you mean, when it comes to the mock network?
<dignifiedquire>
before I was just mocking the two calls the network exposes
<daviddias>
that has to have mock libp2p?
<dignifiedquire>
yes
<daviddias>
the mockLibp2p is quite simple
<dignifiedquire>
but now I have to mock all the libp2p calls wich are much more complicated
ghtdak has quit [Ping timeout: 260 seconds]
<dignifiedquire>
because I have to return streams and manage all of that :/
<daviddias>
but you should still be able to mock connecTo and sendMessage
<daviddias>
and just give that simple libp2pMock
<dignifiedquire>
I could, but that means I'm not testing the network code in those instances
<daviddias>
I'm just not aware how that return {connectTo, sendMessage} calls the constructor of network
<dignifiedquire>
it doesn't, that's the trick, it fakes the whole network objec
<daviddias>
shouldn't we have a set for full bitswap tests
<daviddias>
like, with no mocks
<dignifiedquire>
if we have a libp2p version that works in the browser we can do that yes
<dignifiedquire>
we need an easy and quick way to generate libp2p instances that represent a network with `x`nodes
<dignifiedquire>
that works in the browser and node
<daviddias>
we can create x nodes
<daviddias>
but we will always have to tell the topology to the thing
<daviddias>
and make the connections
<daviddias>
A good smoke tests at the module level, would be to make 5 or 10 nodes, all connected between each other (full mesh) and run the bitswapping
<daviddias>
Then in js-ipfs, make the super versatile tests
<dignifiedquire>
yes, and we should have a nice function that we can use to generate that
<dignifiedquire>
daviddias: yeah that's a very odd api
<dignifiedquire>
daviddias: one more question, are you okay with pulling out the individual parts in core/index.js into their own files? at the current rate the file is growing quickly and it would be nice to have them in separate files like, core/repo.js
<daviddias>
I am :D
<daviddias>
actually, had that on my map, as soon as the Files thing voxelot has been doing is merged
<voxelot>
will push a commit with the details so people can look if i cant figure out a better system in the next hour
<daviddias>
dignifiedquire: not quite, the thing is that the blobstore will check for the existence of a blob, which will fail if the file/folder is not even there
<dignifiedquire>
yeah that's super weird
computerfreak has joined #ipfs
<voxelot>
daviddias: think i fixed stream bug, going to push and squash commits for CR
<daviddias>
wowowow
<daviddias>
:D
<dignifiedquire>
daviddias: that
<dignifiedquire>
is not what the interface definition says
<dignifiedquire>
voxelot: cool
Combined2857 has joined #ipfs
Combined2857 has quit [Client Quit]
ghtdak has joined #ipfs
<dignifiedquire>
voxelot: when can I see the finished code ?
<voxelot>
dignifiedquire: just linting now, 5 minutes
<voxelot>
daviddias: cat replies with a stream, i'm not sure how the hapi server is chunking it but go-ipfs isn't having problems receiving it
<daviddias>
voxelot: even for a 1Gb file?
<dignifiedquire>
daviddias: lol, I'm changing it right now and you already merged it :P
<daviddias>
I'm just skeptical, because I know go-ipfs sets custom headers, so hapijs can be very smart, but it will not guess what go-ipfs needs
<daviddias>
dignifiedquire: ahah I thought your answer was more 'yeah, we could do it in several ways, let me know if I should change'
<dignifiedquire>
:D
<dignifiedquire>
no matter, it works so I'm happy
<daviddias>
I can now notice little details of when your codestyle changes
Encrypt_ has joined #ipfs
<daviddias>
or at least, I've been noticing some patterns over time
Protocol_under has joined #ipfs
* dignifiedquire
feels watched
<daviddias>
ahaha
<daviddias>
it is cool :)
<dignifiedquire>
but yes I have a lot of small patterns I'm using everywhere, but those are not checked by any linter and I try to always improve on them
<daviddias>
ipfs-repo 0.8.0 released
Arakela0_ has joined #ipfs
Arakela007 has quit [Read error: Connection reset by peer]
<voxelot>
daviddias: js-ipfs still can't add files > 1g
<daviddias>
dignifiedquire: voxelot why?
<voxelot>
block-stream2 is still saying invalid buffer array length
Arakela0_ has quit [Read error: Connection reset by peer]
<daviddias>
got it
Arakela007 has joined #ipfs
<daviddias>
ok, nevertheless, can go-ipfs add a 500MiB file to js-ipfs http-api
<daviddias>
and vice versa
<daviddias>
dignifiedquire: do you need anything else from me to get this one https://github.com/ipfs/js-ipfs-bitswap/pull/4 merged. I'm going to work on the libp2p test network, and wanted to know if I can branch from the branch (preferably, branch from master with this merged PR)
<dignifiedquire>
ipfs-repo has quite the turn around :D
<ipfsbot>
[js-ipfs] greenkeeperio-bot opened pull request #191: Update ipfs-repo to version 0.8.0
<dignifiedquire>
daviddias: we can merge it and work on the test improvements from master
<dignifiedquire>
daviddias: merged
<daviddias>
awesome :D
<voxelot>
go can't add anything to js-ipfs http-api until i figure out what type of response the go-cli is expecting from the go core so i can send that to go-cli
<voxelot>
breaking up 900 lines of code to take out the http stuff for a ne PR
<ipfsbot>
[js-ipfs] dignifiedquire deleted greenkeeper-ipfs-repo-0.8.0 at b3b5943: https://git.io/vwbUa
<ipfsbot>
[js-ipfs] dignifiedquire force-pushed fix/exists from d762025 to cd126dc: https://git.io/vwbUw
<nicolagreco>
:( how do I make the bot to pin my content?
<mmuller_>
pinbot: help
<mmuller_>
It was worth a try :-)
i_am_ss has joined #ipfs
zorglub27 has joined #ipfs
<lgierth>
!befriend nicolagreco admin
<pinbot>
Hey nicolagreco, let's be friends! You can admin
elasticdog has joined #ipfs
<daviddias>
!friends
M-spudz has joined #ipfs
M-spudz has left #ipfs [#ipfs]
Encrypt has quit [Quit: Quitte]
Encrypt has joined #ipfs
i_am_ss has quit [Remote host closed the connection]
<nicolagreco>
I never felt this admin
<voxelot>
daviddias: problem was the event emitter was sending the stream to the consumer, and the consumer tried reading data before any had been pushed
<daviddias>
Yep, that is why you should just push inside a read context, because if outside of that context, you can't really be sure there will be data to be consumed
pfraze has quit [Remote host closed the connection]
<voxelot>
well, either way if you try to push data inside a read context or not, if you fire an event emitter to a write stream before you push any data it will cry
<voxelot>
i assume this worked before because the event emitter was slow enough to pause for the next line to push data, and most of the testing i do is on nodes with links
<dignifiedquire>
voxelot: what timezone are you in? have you moved your sleep schedule to gmt??
<voxelot>
dignifiedquire: lol, pst, it;s 9am here, woke up at 2am so pretty much
<daviddias>
voxelot: pretty sure it is more of the case that if you do push before a .read, it will create a stub .read for you
<daviddias>
but that is an antipattern
<daviddias>
what you want to do is the .push inside your impl of .read
<daviddias>
and it will not throw
<daviddias>
there is no rule that a readable stream has to have data buffered to be pipe'pable
<voxelot>
if you push inside of the read implementation, you have write function read from it to hit the first tick of the .read correct?
<voxelot>
before it hits the first data push
<voxelot>
i had written the test for no links in exporter to just look for data coming in instead of using a write function like bl to start that process, so it failed, so i got the wrong idea i believe
pfraze has joined #ipfs
<voxelot>
i assumed that would be okay since no link data should always be under a chunk size, so the buffer wouldnt be too large, but i had fired the event emitter with the stream to the cli before the buffer and sometimes it would catch up, other times it wouldnt
<whyrusleeping>
noffle: lgierth got time for some CR?
rgrinberg has quit [Read error: Connection reset by peer]
rgrinberg has joined #ipfs
rgrinberg has quit [Read error: Connection reset by peer]
rgrinberg has joined #ipfs
libman has quit [Remote host closed the connection]
rgrinberg has quit [Read error: Connection reset by peer]
afisk has joined #ipfs
martinkl_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
afisk has quit [Read error: Connection reset by peer]
Dunkhan has joined #ipfs
Wolf480pl has quit [Ping timeout: 252 seconds]
rgrinberg has joined #ipfs
<Dunkhan>
I have a lot of noob questions about how ipfs works, is there a source where I can get a basic high level understanding before bothering people with my questions?
kaiza has joined #ipfs
kaiza has quit [Max SendQ exceeded]
<Dunkhan>
I was at the talks at c-base so the details I am interested in are more detailed than what was discussed there
<dignifiedquire>
Dunkhan: the videos at ipfs.io should give you a good overview
<Dunkhan>
I was looking for less highlevel than that, I basically am looking for a better understanding of how the underlying technologies work
<Dunkhan>
I guess I should read up on git and torrent
<Dunkhan>
wait that second video might be helpful, I will watch that more carefully first
cemerick has quit [Ping timeout: 240 seconds]
<Dunkhan>
Hmm no, that seems to just be the same presentation again
<Dunkhan>
I will ask one question, that will likely lead to a source for answers
<Dunkhan>
If someone has a piece of data, and I request it via ipfs, how does that request find the person that has it. If they are many nodes away, say on the other side of the world. Is there some kind of broadcast signal that just propagates until it finds the node with that data? Or is there some kind of address table stored locally?
<Dunkhan>
I understand how a had addresses content, but not how that translates to network traffic
<Dunkhan>
hash*
<baccenfutter>
Dunkhan: perhaps you wan to study the whitepaper?
<geir_>
Is any one working on a youtube clone using ipfs as a backend? That would be an amazing project
hugooconnor has joined #ipfs
cemerick has joined #ipfs
rgrinberg has quit [Ping timeout: 246 seconds]
hugooconnor has quit [Ping timeout: 240 seconds]
jaboja has quit [Ping timeout: 276 seconds]
Boomerang has joined #ipfs
rendar has quit [Ping timeout: 276 seconds]
<baccenfutter>
seems I am having a verz hard time figuring out the correct szntax for plazing a video in the browser. >:/ I added a file, a straight file, no parenting directory. QmNnXAM5pAVh944V1XgT2rxQSdiwAGyygt97bN588p22gQ how do I play that in my browser?
<baccenfutter>
very(
<baccenfutter>
setxkbmap us
<baccenfutter>
arghs!>!?!?!
<baccenfutter>
sometimes I get chain-reactions of things that go wrong...
<baccenfutter>
I hate that
corvinux has joined #ipfs
<baccenfutter>
let me re-phrase: how do I play the video behind this hash in my browser using ipfs.io/ipfs? QmNnXAM5pAVh944V1XgT2rxQSdiwAGyygt97bN588p22gQ
<voxelot>
dignifiedquire: made some comments
corvinux has quit [Client Quit]
rendar has joined #ipfs
corvinux has joined #ipfs
corvinux has joined #ipfs
corvinux has quit [Changing host]
<voxelot>
daviddias: files PR is ready for CR
<daviddias>
voxelot: thank you, I'm having dinner with fam to celebrate my dad's birthday, will CR when I get back :)
<dignifiedquire>
voxelot: more comments
ashark has quit [Ping timeout: 240 seconds]
<voxelot>
daviddias: ahh enjoy!
Aeon has quit [Ping timeout: 246 seconds]
Aeon has joined #ipfs
Aeon has quit [Changing host]
Aeon has joined #ipfs
hugooconnor has joined #ipfs
Encrypt has quit [Quit: Quitte]
mildred has joined #ipfs
ashark has joined #ipfs
chriscool has quit [Quit: Leaving.]
chriscool has joined #ipfs
ashark has quit [Ping timeout: 250 seconds]
b421 has joined #ipfs
b421 is now known as be42
<baccenfutter>
the fact that I can not grab my objects from http://ipfs.io/ipfs is what has been causing my confusion. loclahost:8080/ipfs/<hash> works perfectly fine vs. ipfs.io/ipfs/<hash>
<whyrusleeping>
baccenfutter: the ipfs.io gateways have having some annoying problems the last couple days
kran has quit [K-Lined]
<whyrusleeping>
I think i've found the root cause, and i'm working on pushing out a fix
<ipfsbot>
[js-ipfs] diasdavid created greenkeeper-joi-8.0.5 (+1 new commit): https://git.io/vwN8i
<ipfsbot>
js-ipfs/greenkeeper-joi-8.0.5 69c9c45 greenkeeperio-bot: chore(package): update joi to version 8.0.5...
<ipfsbot>
[js-ipfs] dignifiedquire created refactor/core (+1 new commit): https://git.io/vwN85
<kevina>
whyrusleeping: I am working creating a pull request to make Block an interface and, am having a build problem, that I hope you can help
<whyrusleeping>
kevina: i sure can!
<conway>
hello :) I have a fun problem. I'm on a win10 machine running 0.4.0 and getting that cpu bug. Currently eating 60% cpu (on a surface 4 pro). Does anyone have the compiled windows executables? and need logs?
<conway>
whyrusleeping: I tried that and got this: "Path Resolve error: no link named "go-ipfs_v0.4.1_windows-amd64.zip" under QmdcPDVzQQpaAZpqCrGAp2fuzP5Kb44uiEcMbB8mNgBozf"
<kevina>
can't load package: package gx/ipfs/go-libp2p-secio: cannot find package "gx/ipfs/go-libp2p-secio" in any of:
<kevina>
whyrusleeping: let me know if you need more info and I will copy and paste the output somewhere else
<whyrusleeping>
kevina: yeah, could you put the entire output in a gist for me?
<kevina>
okay
<whyrusleeping>
conway: thats... concerning
<conway>
I know :/ I crawled that key and there's no files in there for windows at all. There are 2 buildlog files for windows for i386 and x64, but no execs.
patcon has quit [Read error: Connection reset by peer]
afisk has quit [Remote host closed the connection]
<ipfsbot>
[go-ipfs] kevina opened pull request #2633: Make blocks.Block an interface. (master...blocks-interface) https://git.io/vwNPc
computerfreak has quit [Quit: Leaving.]
<whyrusleeping>
npm update failed
<whyrusleeping>
dignifiedquire: does distributions work for you?
afisk has joined #ipfs
<lgierth>
dignifiedquire: while we're at it, can dist vendor fira.css?
<dignifiedquire>
lgierth: file an issue
<lgierth>
hehe
<lgierth>
PRs welcome
<dignifiedquire>
but it's gonna be a lot of work
<dignifiedquire>
well medium work, once I've actually quired all the files
<lgierth>
but there's no licensing restrictions involved?
<lgierth>
cool
<dignifiedquire>
lgierth: no restrictions on that side, just need to get all the font files and download the css file
ashark has quit [Ping timeout: 260 seconds]
<dignifiedquire>
whyrusleeping: nope it's broken on node 6, go back to 5 and it'll work just fine
afisk has quit [Ping timeout: 250 seconds]
<whyrusleeping>
meh
KatzZ has joined #ipfs
KatzZ has quit [Read error: Connection reset by peer]
alusion has joined #ipfs
alusion has quit [Read error: Connection reset by peer]
Akaibu has quit [Quit: Connection closed for inactivity]
KatzZ has joined #ipfs
palkeo has quit [Quit: Konversation terminated!]
<kevina>
whyrusleeping: I just submitted a pull request https://git.io/vwNPc, could you look it over, I would like to get this is ASAP so I can rebase by commits on it
<whyrusleeping>
kevina: looking at the circleCI for that PR
<whyrusleeping>
did you cancel the build?
<whyrusleeping>
its not letting me restart it
Aeon has joined #ipfs
Aeon has quit [Changing host]
Aeon has joined #ipfs
<whyrusleeping>
eh, no matter
libman has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping pushed 2 new commits to master: https://git.io/vwNyw
<ipfsbot>
go-ipfs/master b84cbec Kevin Atkinson: Make blocks.Block an interface....
<pinbot>
[host 1] failed to grab refs for /ipfs/QmXZQzBAFuoELw3NtjQZHkWSdA332PyQUj6pQjuhEukvg8: Failed to get block for QmXZQzBAFuoELw3NtjQZHkWSdA332PyQUj6pQjuhEukvg8: context canceled
patcon has joined #ipfs
chris6131 has joined #ipfs
matoro has quit [Ping timeout: 240 seconds]
ashark has joined #ipfs
ashark has quit [Ping timeout: 260 seconds]
martinkl_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]