guest234234 has quit [Read error: Connection reset by peer]
Matoro has joined #ipfs
devbug_ has quit [Ping timeout: 240 seconds]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
domanic has quit [Ping timeout: 250 seconds]
domanic has joined #ipfs
<roguism>
Is there a spec for the git-like system on IPFS?
<jbenet>
hey roguism -- not quite yet. there's two paths: (1) is importing/wrapping actual git. there's a few approaches to this, whyrusleeping made a git-http-clonable rehost https://github.com/whyrusleeping/git-ipfs-rehost -- cryptix made an _actual git remote_ plugin https://github.com/cryptix/git-remote-ipfs -- then (2) there's adding commits to ipfs. this has
<jbenet>
been meaning to happen for a well, but i wanted to get identity right first. that still hasnt happened, but we can now count on identity being objects, so we can probably implement https://github.com/ipfs/notes/issues/23 very soon
<jbenet>
hey everyone can i ask we do more Q/A over the FAQ ? it's a much better resource for newcommers than irc logs.
reit has joined #ipfs
<lgierth>
.
hoony has quit [Ping timeout: 246 seconds]
nham has quit [Ping timeout: 260 seconds]
<roguism>
Alright
Qwertie- has joined #ipfs
<Qwertie->
Hi
Matoro has quit [Ping timeout: 255 seconds]
Senji has quit [Disconnected by services]
Senj has joined #ipfs
<Qwertie->
I have had the ipfs daemon running in the background for a while and I just noticed its using an unusually large amount of data even though I havent used it to fetch anything for a while. Its using 10-20KB/s continuously. What could be causing that?
M-prosodyContext has quit [Ping timeout: 264 seconds]
M-cryptix1 has quit [Ping timeout: 264 seconds]
kitcambridge_ has joined #ipfs
friendly12345 has quit [Ping timeout: 264 seconds]
Stskeeps has quit [Ping timeout: 264 seconds]
bret has quit [Ping timeout: 264 seconds]
davidar_ has quit [Ping timeout: 264 seconds]
M-staplemac has quit [Ping timeout: 264 seconds]
M-matthew has quit [Ping timeout: 264 seconds]
Myagui has quit [Ping timeout: 264 seconds]
bret_ has joined #ipfs
jfred has quit [Ping timeout: 264 seconds]
sseagull has joined #ipfs
M-jfred1 has joined #ipfs
jfred has joined #ipfs
rabbitface3 has joined #ipfs
bret_ is now known as bret
Stskeeps has joined #ipfs
M-edrex1 has joined #ipfs
Teeed_ has joined #ipfs
Remjey_ has joined #ipfs
locusf_ has joined #ipfs
zrl_ has joined #ipfs
M-jgrowl has joined #ipfs
M-hash has joined #ipfs
Obamatron has joined #ipfs
substack_ has joined #ipfs
M-prosodyContext has joined #ipfs
friendly12345 has joined #ipfs
Qwertie has quit [*.net *.split]
RJ2 has quit [*.net *.split]
zrl has quit [*.net *.split]
leeola has quit [*.net *.split]
mg- has quit [*.net *.split]
mungojelly has quit [*.net *.split]
go1111111 has quit [*.net *.split]
xicombd has quit [*.net *.split]
eater has quit [*.net *.split]
ebarch has quit [*.net *.split]
bigpox has quit [*.net *.split]
kanzure has quit [*.net *.split]
M-rschulman1 has quit [*.net *.split]
alu has quit [*.net *.split]
M-alien has quit [*.net *.split]
kanzure_ has joined #ipfs
astrocyte has quit [*.net *.split]
substack has quit [*.net *.split]
zz_r04r has quit [*.net *.split]
CarlWeathers has quit [*.net *.split]
rabbitface2 has quit [*.net *.split]
jatb has quit [*.net *.split]
compleatang has quit [*.net *.split]
M-jfred has quit [*.net *.split]
rektide has quit [*.net *.split]
locusf has quit [*.net *.split]
dvn has quit [*.net *.split]
wiedi has quit [*.net *.split]
emery has quit [*.net *.split]
felixn has quit [*.net *.split]
M-edrex has quit [*.net *.split]
cSmith_ has quit [*.net *.split]
Remjey has quit [*.net *.split]
Teeed has quit [*.net *.split]
|GonZo2000| has quit [*.net *.split]
erikj` has quit [*.net *.split]
infinity0 has quit [*.net *.split]
arpu has quit [*.net *.split]
M-nated has quit [*.net *.split]
AtnNn has quit [*.net *.split]
cypher_ has quit [*.net *.split]
jager has quit [*.net *.split]
Matoro has quit [*.net *.split]
Tv` has quit [*.net *.split]
e-lima has quit [*.net *.split]
gperrin has quit [*.net *.split]
__konrad_ has quit [*.net *.split]
niekie has quit [*.net *.split]
gunn has quit [*.net *.split]
M-osener has quit [*.net *.split]
[0__0] has quit [*.net *.split]
alterego has quit [*.net *.split]
cblgh has quit [*.net *.split]
thelinuxkid has quit [*.net *.split]
ThomasWaldmann has quit [*.net *.split]
kandinski has quit [*.net *.split]
Sargun has quit [*.net *.split]
Pharyngeal has quit [*.net *.split]
felixn has joined #ipfs
GonZo2000 has joined #ipfs
M-staplemac has joined #ipfs
GonZo2000 has quit [Changing host]
GonZo2000 has joined #ipfs
M-matthew has joined #ipfs
rektide has joined #ipfs
Qwertie has joined #ipfs
zrl_ is now known as zrl
Myagui has joined #ipfs
go1111111 has joined #ipfs
dvn has joined #ipfs
xicombd has joined #ipfs
alu has joined #ipfs
bigpox has joined #ipfs
M-jgrowl has quit [Ping timeout: 250 seconds]
M-oddvar has quit [Ping timeout: 250 seconds]
vakla has quit [Ping timeout: 250 seconds]
M-giodamelio has quit [Ping timeout: 250 seconds]
whyrusleeping has quit [Ping timeout: 250 seconds]
substack_ has quit [Ping timeout: 250 seconds]
rabbitface3 has quit [Ping timeout: 250 seconds]
fleeky has quit [Ping timeout: 250 seconds]
barnacs has quit [Ping timeout: 250 seconds]
wemeetagain has quit [Ping timeout: 250 seconds]
Tristitia has quit [Ping timeout: 250 seconds]
trn has quit [Ping timeout: 250 seconds]
gendale_ has quit [Ping timeout: 250 seconds]
mmuller_ has quit [Ping timeout: 250 seconds]
trn has joined #ipfs
astrocyte has joined #ipfs
RJ2 has joined #ipfs
gendale_ has joined #ipfs
substack has joined #ipfs
predeficient has joined #ipfs
erikj` has joined #ipfs
infinity0 has joined #ipfs
M-nated has joined #ipfs
cypher_ has joined #ipfs
jager has joined #ipfs
AtnNn has joined #ipfs
arpu has joined #ipfs
Pharyngeal has joined #ipfs
M-davidar has joined #ipfs
M-cryptix has joined #ipfs
Tv` has joined #ipfs
niekie has joined #ipfs
[0__0] has joined #ipfs
Matoro has joined #ipfs
thelinuxkid has joined #ipfs
ThomasWaldmann has joined #ipfs
r04rs has joined #ipfs
e-lima has joined #ipfs
alterego has joined #ipfs
kandinski has joined #ipfs
Sargun has joined #ipfs
__konrad_ has joined #ipfs
gperrin has joined #ipfs
cblgh has joined #ipfs
M-osener has joined #ipfs
gunn has joined #ipfs
M-cryptix has joined #ipfs
M-davidar has joined #ipfs
M-cryptix has quit [Changing host]
M-davidar has quit [Changing host]
ebarch has joined #ipfs
M-oddvar has joined #ipfs
M-jgrowl has joined #ipfs
M-giodamelio has joined #ipfs
barnacs has joined #ipfs
Tristitia has joined #ipfs
leeola has joined #ipfs
wemeetagain has joined #ipfs
rabbitface3 has joined #ipfs
mmuller has joined #ipfs
vakla has joined #ipfs
whyrusleeping has joined #ipfs
fleeky has joined #ipfs
joshbuddy has joined #ipfs
joshbuddy has quit [Client Quit]
domanic has joined #ipfs
locusf_ is now known as locusf
wrouesnel2 has quit [Ping timeout: 240 seconds]
Senji has joined #ipfs
ygrek has joined #ipfs
sseagull has quit [Quit: Lost terminal]
Jeff_ has joined #ipfs
Jeff_ is now known as Guest97452
Guest97452 is now known as Jeph
Jeph is now known as Jeffy
domanic has quit [Ping timeout: 272 seconds]
Jeffy has left #ipfs ["Leaving"]
mg- has joined #ipfs
simonv3 has quit [Quit: Connection closed for inactivity]
slothbag has joined #ipfs
<ipfs-gitter-bot>
(void4) We have the /ipfs/<hash>(/path) and /ipns/<id>(/path) notation. Is there already an agreed upon notation to describe protocol channels? (e.g. /peer/<id>/<protocolstring>)
<jbenet>
@void4: yep that works, well use /ipfs/<peer-id>/<rest-of-protocol>. it's like a https://github.com/jbenet/multicodec identifier or a https://github.com/jbenet/multiaddr. maybe open an issue in notes to document ideas. and whatever you do should be embeddable, so i could write it in a multiaddr: /ip4/1.2.3.4/tcp/1234/ipfs/<peerid>/<protocol>
<ipfs-gitter-bot>
(void4) Because right now, it is not possible to differ between IPNS paths without the prefix and a channel (whose protocol string could be part of a path layout itself)
null_rad- has quit [Ping timeout: 240 seconds]
<ipfs-gitter-bot>
(void4) I am not sure if both device dependent and independent identifiers should have the same prefix. Since you told me of libp2p and especially after I've heard of Ethereums view of Whisper and Swarm, both seem logically separate to me. This notation basically mixes immutable data and channels. If a program that only uses libp2p receives an IPFS path and it represents a hash it only
<jbenet>
i think it works, for example, suppose a pub/sub protocol uses "/pubsub/<channel-id>" eg /pubsub/chanid -- you can tell people to connect to /ipfs/<peerid>/pubsub/<chanid> (their ipfs node can find /ipfs/<peerid> in the network, and connect, and ask for /pubsub/<chanid>)
<jbenet>
but you can ALSO say connect to /ip4/1.2.3.4/ipfs/<peerid>/pubsub/<chanid> to an ipfs node _without_ a peer-routing protocol, and it would now know how to connect to you.
Tv` has quit [Quit: Connection closed for inactivity]
<ipfs-gitter-bot>
(void4) It seems to depend on whether the identity resolution mechanism is part of libp2p or IPFS. Is ipfs/specs up-to-date with this information?
<ipfs-gitter-bot>
(void4) I don't know if your terminology has changed in the last months, but for me, libp2p is responsible for direct messaging with another device (not necessarily another node in the IPFS network), which may include a peer routing mechanism, while IPFS builds on top of it and is concerned about source independent data - which requires additional capabilities such as the Distributed Record Store, BitSwap etc.
<ipfs-gitter-bot>
(void4)
<ipfs-gitter-bot>
(void4) While libp2p has a flat identity namespace, IPFS has a flat data namespace.
<jbenet>
yeah void4 that makes sense. it depends on whether other p2p systems want to share systems like dhts and connectivity in general. maybe so :)
__mek__ has joined #ipfs
<ipfs-gitter-bot>
(void4) In the end, it's also about defining where the following is preferable over the other: path elements, prefixes, internal data formats. Immutable paths are structured like this: /ipfs/<prefix+hash>(/path), mutable paths are /ipns/<identity>(+possibly /path) and channels are /ipfs/<identity>/<protocol>. Unless identities get their own prefix (multikey) libp2p cannot discern what the next component in a /ipfs path is. And even then: both multike
<ipfs-gitter-bot>
multihashes occupy the same namespace.
Matoro has quit [Ping timeout: 255 seconds]
null_radix has joined #ipfs
slothbag has quit [Read error: Connection reset by peer]
voxelot has quit [Ping timeout: 260 seconds]
frabrunelle has joined #ipfs
frabrunelle has quit [Client Quit]
hoony has joined #ipfs
gperrin has quit [Remote host closed the connection]
disgusting_wall has quit []
hoony has quit [Quit: hoony]
disgusting_wall has joined #ipfs
jhulten_ has joined #ipfs
grahamperrin has joined #ipfs
Matoro has joined #ipfs
jhulten_ has quit [Ping timeout: 240 seconds]
<kandinski>
I seem to recall a discussion about extending the multihash protocol to prepend a byte with the encoding (bytestring, hex, base64, bas58, which base58 alphabet, etc)
<kandinski>
what was the conclusion?
<whyrusleeping>
kandinski: if you dont know what the encoding is, how will you know what a 'byte' is?
<whyrusleeping>
seems silly
niekie has quit [Ping timeout: 255 seconds]
<ipfs-gitter-bot>
(void4) Technically, we could go full circle and prepend a byte describing what multi-prefix is used next.
predeficient has quit [Ping timeout: 276 seconds]
crocheter has joined #ipfs
<kandinski>
whyrusleeping: it would be a byte-restricted-to-printable describing the encoding of the rest of the multihash
HostFat has quit [Read error: Connection reset by peer]
<kandinski>
that's a good point
<ipfs-gitter-bot>
(davidar) prefix it with the largest character in that encoding
<whyrusleeping>
kandinski: okay, so we need to limit ourselves to 'only bytes that can be printed the same in all encodings' ?
<whyrusleeping>
davidar, but there are multiple implementations of a given encoding
<whyrusleeping>
base64 has at least three
<whyrusleeping>
base58 has two that i know of
<kandinski>
but you would take that byte out before decoding the rest
<kandinski>
so imagine the encoding is hex
<kandinski>
the code for "hex encoding" could be X
<kandinski>
X<currentmultihashinhex>
<kandinski>
whyrusleeping: mind you, I'm not suggesting it; I was merely asking whether there was a conclusion reached.
<ipfs-gitter-bot>
(void4) The multi-prefix relates to the problem mentioned above: You need something else to indicate what type of format the identifier is. Right now they are the path elements /ipfs and /ipns, whereby /ipfs is a bit ambigous. A multi-prefix would effectively specify whether the following data represents an identity, data hash or channel.
<ipfs-gitter-bot>
(davidar) whyrusleeping: ok, largest character, and then disambiguating characters :p
<kandinski>
I see no conclusion reached yet.
niekie has joined #ipfs
<ipfs-gitter-bot>
(davidar) In some sense, I don't understand why we don't just have /ipfs/sha256/hex/deadbeef
<ipfs-gitter-bot>
(davidar) More verbose, but fits in with the path naming system
nham has quit [Ping timeout: 260 seconds]
<ipfs-gitter-bot>
(void4) The problem with the path naming system is that we cannot publish files which have the path of another file as name.
<ipfsbot>
[go-ipfs] rht created reuse-transport (+1 new commit): http://git.io/vBeHk
<ipfsbot>
go-ipfs/reuse-transport aeab3e6 rht: http client: reuse transport previously created in NewClient...
<kandinski>
void4, can you give an example?
ygrek has quit [Ping timeout: 240 seconds]
<ipfs-gitter-bot>
(void4) If I want to distribute my address as a file which has that name, the path to the file itself becomes ambigous
<ipfsbot>
[go-ipfs] rht opened pull request #1991: http client: reuse transport previously created in NewClient (dev0.4.0...reuse-transport) http://git.io/vBeHl
<kandinski>
I see, thanks
<ipfs-gitter-bot>
(void4) /ipfs/Qm...//ipfs/<id>
<ipfs-gitter-bot>
(davidar) I fail to see the problem
<ipfs-gitter-bot>
(void4) If the filename would have a multi-prefix (or whatever it should be called) we wouldn't even need /ipfs, so the above would become something like a1Qm/a0<id>
astrocyte has quit [Ping timeout: 240 seconds]
<ipfs-gitter-bot>
(davidar) I don't think removing /ipfs is going to happen
mungojelly has joined #ipfs
<ipfs-gitter-bot>
(davidar) Given jbenet wants everything to be a path...
<ipfs-gitter-bot>
(davidar) If anything i think we should replace multihash with path components
<ipfs-gitter-bot>
(void4) That sounds interesting, too.
<ipfs-gitter-bot>
(void4) davidar: Does my general argument for the separation of the identity and data namespaces make sense (/ipfs)?
<kandinski>
but a blob's multihash is also used in kad routing, for finding a distance between the file and "its node", right?
<kandinski>
you xor the blob's hash and the node's address
<kandinski>
so both have to have the same shape or similar
<whyrusleeping>
mmm, i dont like gitter
<whyrusleeping>
matrix was nicer for me
<whyrusleeping>
i could more easily tell who you all are
<whyrusleeping>
also, whered my monacle go?
<whyrusleeping>
there we go
<deltab>
hmm, irc clients could add support for bridging, given a convention for signalling it
astrocyte has joined #ipfs
<kandinski>
hi deltab
<deltab>
hi
<deltab>
I'm half asleep, so my apologies if I'm not making sense
kahiru has joined #ipfs
doublec has quit [Read error: Connection reset by peer]
<deltab>
I see some appeal in /ipfs/sha256/hex/deadbeef, but does it confer any practical benefit?
doublec has joined #ipfs
<ipfs-gitter-bot>
(davidar) whyrusleeping: yeah, only using this temporarily ;)
<deltab>
at the level at which it's used in a path, it's just an arbitrary string, and its substructure isn't that interesting or worth exposing in separate path components
<ipfs-gitter-bot>
(void4) and it would make the "path as filename" problem even worse
<deltab>
IP addresses and domain names are hierarchical, but that's not typically reflected in directory structures (Java class naming and the like aside)
<whyrusleeping>
guys, os.Exit(1) is bad
<whyrusleeping>
its kills defers
<whyrusleeping>
just b t dubs
<deltab>
yeah, the python equivalent raises a SystemExit exception allowing handlers to run
<ipfs-gitter-bot>
(void4) @deltab: One might want to store a file which has its multihash as its name
<ipfsbot>
[go-ipfs] rht pushed 1 new commit to reuse-transport: http://git.io/vBebk
<whyrusleeping>
deltab: yeah, the go version just calls exit
<whyrusleeping>
it does exactly what you'd expect if you wrote C for a living before coming to go
<whyrusleeping>
"GTFO right now"
rendar has joined #ipfs
e-lima has quit [Ping timeout: 255 seconds]
NightRa has joined #ipfs
grahamperrin has quit [Remote host closed the connection]
<ipfsbot>
[go-ipfs] whyrusleeping created wip/fix-close-notify (+1 new commit): http://git.io/vBepi
<ipfsbot>
go-ipfs/wip/fix-close-notify 2245858 Jeromy: WIP: working on making close notify work
hoony has joined #ipfs
Not_ has joined #ipfs
<Not_>
so what can i already do with ipfs?
<ipfsbot>
[go-ipfs] whyrusleeping opened pull request #1992: WIP: working on making close notify work (dev0.4.0...wip/fix-close-notify) http://git.io/vBepb
<Stskeeps>
what can't you do :) it works, download it, try it out
<Not_>
like what do people use it for
<Not_>
not was its theoretically possible to do with it
evanmccarter has quit [Quit: Connection closed for inactivity]
amstocker has joined #ipfs
disgusting_wall has quit [Quit: Connection closed for inactivity]
pfraze has joined #ipfs
pfraze has quit [Remote host closed the connection]
wiedi has joined #ipfs
doublec_ is now known as doublec
Encrypt_ has joined #ipfs
Encrypt_ is now known as Encrypt
joshbuddy has quit [Quit: joshbuddy]
rombou has joined #ipfs
amstocker has quit [Ping timeout: 240 seconds]
grahamperrin has joined #ipfs
Deciph has joined #ipfs
ogzy has joined #ipfs
ogzy has quit [Changing host]
ogzy has joined #ipfs
<ogzy>
hi, i am trying to run ipfs at my private network to understand its working, , removed te bootstrap nodes, does superouting servers required?
Deciph has quit [Quit: Leaving]
<ogzy>
i meaned removed from .ipfs/config
ljhms has joined #ipfs
<kahiru>
ogzy, if i understand it correctly, you could remove the superrouting servers and set the daemon to use dht routing
<ogzy>
kahiru, what is dht routing?
<ogzy>
kahiru, distributed hash table routing?
<kahiru>
ogzy, something like that. I'm not really sure how it works, I just saw something like that in the daemons options
17SAD39SD has joined #ipfs
jatb has joined #ipfs
emery has joined #ipfs
jhulten_ has joined #ipfs
compleatang has joined #ipfs
rombou has quit [Ping timeout: 255 seconds]
jhulten_ has quit [Ping timeout: 265 seconds]
guest234234 has joined #ipfs
dignifiedquire has joined #ipfs
<dignifiedquire>
good morning everyone
rombou has joined #ipfs
rombou has quit [Ping timeout: 244 seconds]
machrider has quit [Ping timeout: 240 seconds]
machrider has joined #ipfs
<ipfs-gitter-bot>
(void4) Good morning!
kandinski has quit [Ping timeout: 255 seconds]
kandinski has joined #ipfs
Teeed_ is now known as Teeed
dignifiedquire has quit [Quit: dignifiedquire]
[BNC]geir is now known as geir_
<ipfs-gitter-bot>
(void4) Some time ago there was a request for libp2p functionality to be accessible from the console and from the APIs (e.g. $ ipfs channel send /ipfs/<id>/<protocol> <data>). Are there any updates for that? It would simplify the development for devices that cannot run nodes themselves.
__mek__ has quit [Ping timeout: 240 seconds]
<ogzy>
after removing superrouting and bootstrap nodes, i run the ipfs daemon at the nodes, to make them connect, what should i run, ipfs swarm connect? or ipfs bootstrap add?
locusf has quit [Ping timeout: 250 seconds]
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
locusf has joined #ipfs
<daviddias>
dignifiedquire I'm here :)
dignifiedquire has joined #ipfs
<dignifiedquire>
daviddias: welcome :D
<daviddias>
thank you :)
<daviddias>
looking at your PR's :)
voxelot has joined #ipfs
voxelot has joined #ipfs
voxelot has quit [Ping timeout: 244 seconds]
dignifiedquire has quit [Quit: dignifiedquire]
master8787 has joined #ipfs
<ipfsbot>
[js-ipfs-api] diasdavid pushed 14 new commits to master: http://git.io/vBvMq
<ipfsbot>
js-ipfs-api/master b138ee1 Connor Keenan: add opts to constructor
<ipfsbot>
js-ipfs-api/master 8babda6 Connor Keenan: parse mocha test files from karma config
<ipfsbot>
js-ipfs-api/master bbcdd5b Connor Keenan: new test/constructor.js parameter tests
besenwesen has quit [Ping timeout: 240 seconds]
master8787 is now known as itradarcz
itradarcz is now known as itradar
besenwesen has joined #ipfs
besenwesen has quit [Changing host]
besenwesen has joined #ipfs
itradar has left #ipfs [#ipfs]
devbug_ has joined #ipfs
itradar has joined #ipfs
<daviddias>
My wifi went out :( will take me a bit more to get this last merge released
keroberos has joined #ipfs
kerozene has quit [Ping timeout: 260 seconds]
keroberos has quit [Ping timeout: 252 seconds]
dignifiedquire has joined #ipfs
kerozene has joined #ipfs
<dignifiedquire>
daviddias: yeah it passes on my local machine, log/tail is super slow at times :(
<dignifiedquire>
sometimes completely hangs for minutes
<dignifiedquire>
(even on the cli)
arpu has quit [Remote host closed the connection]
r04rs has quit [Changing host]
r04rs has joined #ipfs
r04rs is now known as r04r
<ipfsbot>
[js-ipfs-api] diasdavid pushed 2 new commits to master: http://git.io/vBvbR
<ipfsbot>
js-ipfs-api/master 2032805 David Dias: chore: build
<ipfsbot>
js-ipfs-api/master fe1a912 David Dias: release v2.9.2
<ipfsbot>
[js-ipfs-api] diasdavid tagged v2.9.2 at ec1f27f: http://git.io/vBvbu
jhulten_ has joined #ipfs
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #98: Update ipfs-api to version 2.9.2
<daviddias>
.ask whyrusleeping why `ipfs log tail` is so slow //cc dignifiedquire
<multivac>
daviddias: I'll pass that on when whyrusleeping is around.
<ipfsbot>
[js-ipfs-api] Dignifiedquire deleted support-multiaddr at 639b73a: http://git.io/vBftj
<ipfsbot>
[js-ipfs-api] diasdavid deleted buffer-readme at cd543dd: http://git.io/vBfqO
itradar has quit [Quit: Leaving]
<ipfsbot>
[js-ipfs-api] diasdavid deleted issue70 at ccbf5ef: http://git.io/vBfqR
<ipfsbot>
[js-ipfs-api] diasdavid deleted add-vinyl-stream at d6e93e0: http://git.io/vBfqz
<ipfsbot>
[js-ipfs-api] diasdavid deleted feat/name-publish at 6df2261: http://git.io/vBfqa
<daviddias>
done :)
<dignifiedquire>
daviddias: I’m thinking of splitting up index.js in the same way I did with the tests and have something similar like ronin does for cli stuff loading it dynamically from a specified folder, what do you think?
<dignifiedquire>
daviddias: spring cleaning :)
<daviddias>
sounds good :)
<daviddias>
one question, what `standard` are you using? Mine complains with these global vars scheme on tests
<ipfsbot>
[js-ipfs-api] diasdavid pushed 1 new commit to ref-tests: http://git.io/vBfYd
<ipfsbot>
js-ipfs-api/ref-tests d3a5a98 David Dias: travis test log
<daviddias>
ahahah
<dignifiedquire>
looks like you can continue using syntastic, just enable the eslint checker instead of standard
guest234234 has quit [Ping timeout: 250 seconds]
<daviddias>
"Hang tight, the log cannot be shown until the build has started."
* daviddias
grabs popcorn
<dignifiedquire>
hah
<dignifiedquire>
it’s a firefox issue Oo
<dignifiedquire>
I can reproduce on my local machine
<daviddias>
re: but it would be nice if people don't have to configure more stuff for one project (therefore putting the /* globals at the top
<dignifiedquire>
but there are so many files now, you’d have to put it in every one of those :(
baselab has quit [Quit: Leaving]
<dignifiedquire>
and it’s not like eslint is the rarest used linter ;) for example for emacs there isn’t a standard integration I always have to use the eslint version
<dignifiedquire>
I officially hate “chunked response"
<daviddias>
it is the problem with 'ping'?
<dignifiedquire>
yes
<dignifiedquire>
it behaves differently in firefox
<dignifiedquire>
Firefox is combining two chunks into one, leaving with me unparsable json
<dignifiedquire>
-.-
<dignifiedquire>
grabbing some food
<daviddias>
fuuuuuuu
<daviddias>
/o\
<daviddias>
why is PING being chunked anyway it should be a stream, the most
survi has joined #ipfs
mildred has joined #ipfs
Senji has quit [Ping timeout: 240 seconds]
<dignifiedquire>
daviddias: can’t we dropped chunked resonse, and just use ndjson streams everyhwere and depending on the command we either buffer it into a full json object or return the stream of parsed objects
<dignifiedquire>
that would make it much less error prone
<daviddias>
a file isn't really a ndjson stream
<ipfsbot>
[js-ipfs-api] diasdavid pushed 1 new commit to ref-tests: http://git.io/vBfn4
<dignifiedquire>
but we are using chunked responses for json objects
<dignifiedquire>
and files should just be streamed
<daviddias>
re: "but we are using chunked responses for json objects" I believe there might be some mis{understanding,implementation} problem there
<dignifiedquire>
ideally I think it should go-ipfs should send
<dignifiedquire>
1. json based response: ndjson stream
<dignifiedquire>
2. all other responses: text based stream
<dignifiedquire>
js-ipfs-api would handle these like this:
<dignifiedquire>
1. buffer ? aggregate json objects into array and return it : return res.pipe(ndjson)
<dignifiedquire>
2. buffer ? aggregate into new Buffer() : return res
<daviddias>
From my understanding, an ndjson stream should be a collection of independent datablobs, that can be "sent" asynchronously
<daviddias>
there can be JSON responses that aren't `streams`
<dignifiedquire>
as long as we send stream-channels=true all responses should be streams
<dignifiedquire>
and for stream-channels=false no response should be a stream
<dignifiedquire>
this would be much more predictable for everyone involved compared to the current behaviour where it’s more guess work than anything else what will happen
nham has joined #ipfs
<daviddias>
We need to have a solid HTTP API spec, that go-ipfs can execute and client libraries can follow
dignifiedquire_ has joined #ipfs
<dignifiedquire>
yes
dignifiedquire_ has quit [Client Quit]
<dignifiedquire>
daviddias: is there any start on this? like test cases, auto generated specs, etc?
survi has quit [Remote host closed the connection]
<ipfsbot>
[js-ipfs-api] diasdavid closed pull request #134: Refactor tests into separate files (master...ref-tests) http://git.io/vBft3
jhulten_ has joined #ipfs
emery has quit [Disconnected by services]
ehmry has joined #ipfs
<dignifiedquire>
daviddias: I think my next task is then focusing on getting this spec done and implemented, as this blocking and distubing the work in ipfs-api and the stuff I want to do in the webui everytime I touch it
<ipfsbot>
[js-ipfs-api] diasdavid tagged v2.9.3 at 1e9339e: http://git.io/vBflb
NightRa has quit [Quit: Connection closed for inactivity]
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #99: Update ipfs-api to version 2.9.3
<daviddias>
dignifiedquire: yeah, I'm also feeling that I keep having conversations (and sometimes repeated) about the HTTP API and something that is poorly documented, badly implemented
jhulten_ has quit [Ping timeout: 272 seconds]
<dignifiedquire>
daviddias: maybe we can use the apps hangout tomorrow to get a plan of action together on how to best tacle this?
<daviddias>
dignifiedquire: sounds good
<daviddias>
I'll be arriving at Lisbon at 18:55h , 35 mins before the Apps hangout, if nothing wrong happens, I should make it just in time :)
<dignifiedquire>
daviddias: great, otherwise we steal your ipfs-js time :P
<multivac>
dignifiedquire: I'll pass that on when richardlit is around.
<ipfsbot>
[webui] Dignifiedquire closed pull request #98: Update ipfs-api to version 2.9.2
<victorbjelkholm>
daviddias, life is grabbing me by the balls :/ You're in Barcelona now right?
<victorbjelkholm>
you have any time for a beer tonight perhaps?
mildred has quit [Ping timeout: 252 seconds]
<daviddias>
victorbjelkholm: are you ok? No problem if you are super busy
<daviddias>
I'm in BCN till tomorrow :)
<victorbjelkholm>
daviddias, yeah, everything is good here :) Would love to have a beer tonight if you have time
voxelot has joined #ipfs
cemerick has joined #ipfs
simonv3 has joined #ipfs
<ipfsbot>
[js-ipfs-api] diasdavid force-pushed feat/files-api from f5ae644 to 13f2c92: http://git.io/vBf2m
<ipfsbot>
js-ipfs-api/feat/files-api d4e8f0c David Dias: add test-usn to be able to use own local node (good for when testing new IPFS versions)
<ipfsbot>
js-ipfs-api/feat/files-api 0ecdd2b David Dias: files api tests
<ipfsbot>
js-ipfs-api/feat/files-api 13e29ea David Dias: fix rm -r bug
kanzure_ is now known as kanzure
<nicolagreco>
anyone in boston?
cemerick has quit [Ping timeout: 272 seconds]
Tv` has joined #ipfs
<ipfsbot>
[go-ipfs] rht created wire/ctx-bitswap-cancelwant (+1 new commit): http://git.io/vBfw9
<ipfsbot>
go-ipfs/wire/ctx-bitswap-cancelwant 5717463 rht: Bitswap Getblocks: CancelWant on ctx cancel...
<ipfsbot>
[go-ipfs] rht opened pull request #1995: Bitswap Getblocks: CancelWant on ctx cancel (dev0.4.0...wire/ctx-bitswap-cancelwant) http://git.io/vBfrL
<ipfsbot>
[go-ipfs] rht force-pushed wire/ctx-bitswap-cancelwant from 5717463 to ad84e8b: http://git.io/vBfr2
<ipfsbot>
go-ipfs/wire/ctx-bitswap-cancelwant ad84e8b rht: Bitswap Getblocks: CancelWant on ctx cancel...
jhulten_ has joined #ipfs
jhulten_ has quit [Ping timeout: 240 seconds]
HoboPrimate has joined #ipfs
forth has joined #ipfs
<ipfsbot>
[js-ipfs-api] diasdavid pushed 1 new commit to feat/files-api: http://git.io/vBfXI
<ipfsbot>
js-ipfs-api/feat/files-api f2b72db David Dias: fix linting
<ipfsbot>
[js-ipfs-api] diasdavid pushed 1 new commit to feat/files-api: http://git.io/vBfXE
<ipfsbot>
js-ipfs-api/feat/files-api 85942b9 David Dias: remove standard ignore from package.json
<nicolagreco>
how is `/ipfs/dht/1.0.0` resolved? is dht a reserved name?
NightRa has joined #ipfs
ogzy has quit [Remote host closed the connection]
e-lima has joined #ipfs
<ipfsbot>
[js-ipfs-api] Dignifiedquire created ref/api-split (+1 new commit): http://git.io/vBfbN
<ipfsbot>
js-ipfs-api/ref/api-split ebd915e dignifiedquire: refactor: Split api definitions into separate files
<dignifiedquire>
daviddias: api split done
<ipfsbot>
[js-ipfs-api] Dignifiedquire opened pull request #136: refactor: Split api definitions into separate files (master...ref/api-split) http://git.io/vBfNJ
<nicolagreco>
coming back to a discussion I had in the past few days, is there a way for me to use /bittorrent/infohash right now?
e-lima has quit [Ping timeout: 240 seconds]
josh has joined #ipfs
josh is now known as Guest14215
mildred has joined #ipfs
jhulten_ has joined #ipfs
pfraze has joined #ipfs
jhulten_ has quit [Ping timeout: 265 seconds]
Tristitia has quit [Ping timeout: 250 seconds]
eaterof is now known as eater
rombou has joined #ipfs
vacuu has joined #ipfs
voxelot has quit [Ping timeout: 272 seconds]
mildred has quit [Ping timeout: 250 seconds]
pfraze has quit [Remote host closed the connection]
pfraze has joined #ipfs
pfraze has quit [Remote host closed the connection]
s_kunk has quit [Ping timeout: 264 seconds]
rombou has quit [Ping timeout: 255 seconds]
voxelot has joined #ipfs
voxelot has joined #ipfs
Encrypt has quit [Quit: Dinner time!]
SebastianCB has joined #ipfs
e-lima has joined #ipfs
mildred has joined #ipfs
<SebastianCB>
Hi all! Is any data encryption currently implemented in IPFS?
jhulten_ has joined #ipfs
<lgierth>
SebastianCB: not yet -- it's not too far away though
<SebastianCB>
thanks!
<SebastianCB>
lgierth: any idea if we are talking about weeks or months?
<lgierth>
definitely not this year ;)
<SebastianCB>
(depending on that I'd look for encryption elsewhere)
<SebastianCB>
ok thanks!
<lgierth>
you can add already encrypted data to ipfs
<SebastianCB>
sure, was just wondering whats the fastest way for me to go
rombou has joined #ipfs
jhulten_ has quit [Ping timeout: 244 seconds]
<SebastianCB>
so if, e.g. everything that I ipfs add would be encrypted already I'd be super happy :)
<SebastianCB>
but of course you're right I will simply encrypt it myself before that
<James_Epp>
I'm torturing IPFS right now. ipfs init ; ipfs get $hash ; (it downloads to 43MB) ; (ctrl+C the daemon) ; ipfs init ; ipfs get $hash ; . Doesn't look like it's downloading anymore. Do I have to wait more or can I start it from scratch?
<cryptix>
James_Epp: why are you doing that to begin with? (the restarting)
<James_Epp>
cryptix: See how well it manages a service disruption.
andresgutgon has quit [Remote host closed the connection]
voxelot has quit [Ping timeout: 240 seconds]
reit has joined #ipfs
James_Epp has quit [Quit: Leaving.]
leer10 has quit [Ping timeout: 240 seconds]
dignifiedquire has quit [Quit: dignifiedquire]
jhulten_ has joined #ipfs
jhulten_ has quit [Ping timeout: 244 seconds]
Matoro has quit [Ping timeout: 255 seconds]
joshbuddy has joined #ipfs
dignifiedquire has joined #ipfs
dignifiedquire has quit [Client Quit]
rendar has quit [Ping timeout: 250 seconds]
leer10 has joined #ipfs
reit has quit [Quit: Leaving]
rendar has joined #ipfs
SebastianCB has quit [Ping timeout: 240 seconds]
joshbuddy has quit [Quit: joshbuddy]
voxelot has joined #ipfs
voxelot has quit [Changing host]
voxelot has joined #ipfs
dignifiedquire has joined #ipfs
Matoro has joined #ipfs
<ipfsbot>
[go-ipfs] gamemanj closed pull request #1907: Add "privkey" parameter to ipfs name publish (master...master) http://git.io/vW5IE
amstocker has quit [Ping timeout: 240 seconds]
HostFat has joined #ipfs
dignifiedquire_ has joined #ipfs
voxelot has quit [Ping timeout: 250 seconds]
joshbuddy has joined #ipfs
vacuu has quit [Quit: Leaving]
fazo has joined #ipfs
voxelot has joined #ipfs
voxelot has quit [Changing host]
voxelot has joined #ipfs
* drathir
wonder if that really need to be downloaded ? github.com/ipfs/go-ipfs/Godeps/_workspace/src/github.com
<drathir>
and that one github.com/ipfs/go-ipfs/vendor/QmQg1J6vikuXF9oDvm4wpdeAUvvkVEKW1EYDw9HhTMnP2b/go-log
<fazo>
who wants to see a discussion board platform built on top of ipfs? :)
voxelot has quit [Ping timeout: 240 seconds]
<fazo>
I've been working on one for a few weeks. It's got a prototype that works and a protocol well laid out. Just looking for anyone who's interested in contributing
<nicolagreco>
btw, I am quite scared of everyone using localstore in ipfs webapps since if served through the ipfs gateway they all share the same origin, so different single page web app can get information from each other pages
<nicolagreco>
fazo I am from Rome, what about you?
<ion>
nicolagreco: The gateway uses suborigins for that.
<fazo>
nicolagreco: I'm from a small town near Milan :)
<fazo>
nicolagreco: ion: I forgot about that localstorage issue, I should probably put a warning... The plan is to use that to store everything since really there's no other way that I know of
<ion>
nicolagreco: yeah
<fazo>
ion: well then it's a non-issue. Should still document that in the FAQs.
<ion>
I don’t know how many of the popular browsers support suborigins yet, though. But it’s coming.
<nicolagreco>
ion I havent read much about that, can you give me an example of how it would work?
<ion>
Non-matching Suborigin headers (given the same hostname) in the response are handled like non-matching subdomains would be if the requests had a different hostname AFAIU.
<fazo>
nicolagreco: basically the server adds a header or something like that to pages that identifies them, so the browser uses different localStorages even if the domain is the same.
<fazo>
I guess the gateway could use the IP(F/N)S address as a suborigin. It would work fine
<nicolagreco>
very interesting! This is the first time I saw suborigin, I think this is really great (I was sure someone would have come up with something similar)
<nicolagreco>
I wonder when surfing an eventual fs:/ how origins are handled
<ion>
The hash in /ipfs/<hash>/path is used as the Suborigin indeed.
<ion>
nicolagreco: The same way, I’d assume.
<nicolagreco>
ion do you know by any chance how content-types are handled by ipfs? is it just by looking at the extension or there is some clever way?
<ion>
AFAIU go-ipfs just uses magic (no pun intended) to estimate the MIME type when serving a HTTP request.
<ion>
Determining the MIME type based on the filename (if any) would be a bit evil.
<fazo>
can confirm it's magic. Been using the http api extensively for a while, and when the file has valid json it justs knows. That's easy, but I think it also applies to other stuff
nham has joined #ipfs
<fazo>
if you have a directory structure you can use the filenames, but it wouldn't work with just the hash
leer10 has joined #ipfs
<fazo>
I'm trying to figure out a way to index content in my discussion board platform. Content could be distributed in hundreds or thousands of user profiles. How do you get "the latest 10 posts"?
<fazo>
or the most commented
<fazo>
the only way is to build an index from scratch every time or cache it
<fazo>
the protocol currently uses list to aggregate every post a user makes in a specific board, but I was thinking I should switch to a linked list (only the last is indexed, then each one points to the previous)
ygrek has quit [Ping timeout: 252 seconds]
<fazo>
it would allow huge lists without having to download the whole list. But then what if I want to figure out if a user actually posted a post? I need to traverse the list thus downloading unneeded content
<dignifiedquire>
survi: also looking at the makefile helps to figure some things out
<survi>
dignifiedquire: the link uses "go get", but I want to use the tagged version (non-development) to build the binary. And the Makefile always fails me...
<whyrusleeping>
survi: cd $GOPATH/src/github.com/ipfs/go-ipfs
<multivac>
whyrusleeping: 2015-11-22 - 13:28:04 <daviddias> ask whyrusleeping why `ipfs log tail` is so slow //cc dignifiedquire
guest234234 has joined #ipfs
<dignifiedquire>
survi: that link just pulls down master from github, so if you want to build a tagged version you need to cd into go-ipfs and run git checkout <tag> and build afterwards
<whyrusleeping>
daviddias: what do you mean log tail is slow?
<dignifiedquire>
whyrusleeping: it takes sometimes > 10s to respond, and sometimes it doesn’t even respond after > 1min
<dignifiedquire>
(cli + api)
Seka has joined #ipfs
<dignifiedquire>
but after restarting the daemon the first call is fast again, and after becomes slow again
Not_ has joined #ipfs
<Not_>
how can i pin an upload?
<Not_>
and is it possible to give it a name so on my files list i know what is what?
e-lima has quit [Ping timeout: 255 seconds]
<whyrusleeping>
dignifiedquire: interesting... mind filing an issue?
<dignifiedquire>
whyrusleeping: sure will do
xicombd has quit [Quit: Connection closed for inactivity]