patcon has quit [Read error: Connection reset by peer]
palkeo has quit [Quit: Konversation terminated!]
seafood has joined #ipfs
<seafood>
Hey peeps. Do you think you could help me out with IPFS file hashes?
<seafood>
Specifically, decoded the hash of a file I uploaded and discovered (once I had stripped the multihash header bytes) that the SHA256 didn’t match the hash I got with Linux utility `sha256`
<seafood>
Anyone got any ideas why?
<seafood>
Is it because you hash the contents of the file in a particular endianess?
wallacoloo___ has joined #ipfs
<Kubuxu>
seafood: we are chunking files so they can be retrived in parts, the hash you see at the end is hash of hashes linking to parts of the file
<seafood>
Ahah!
<seafood>
Thank you so much. It was hard for me to find that information on the Internet using Google.
<seafood>
Could you provide a reference to some documentation on that?
<Kubuxu>
I am off to bed right now (2am here).
lkcl_ has quit [Ping timeout: 258 seconds]
<seafood>
Okay, good night.
jooni has joined #ipfs
ianopolous_ has quit [Read error: Connection reset by peer]
se3000 has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
infinity0 has quit [Remote host closed the connection]
cemerick has joined #ipfs
infinity0 has joined #ipfs
<haad>
!pin QmSbnqWPbuqrbovMiUs6uU5R9VEpYiym3d6EvrqwLAh77P test
<pinbot>
now pinning /ipfs/QmSbnqWPbuqrbovMiUs6uU5R9VEpYiym3d6EvrqwLAh77P
galois_dmz has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
lkcl_ has quit [Ping timeout: 246 seconds]
lkcl_ has joined #ipfs
chungy has quit [Ping timeout: 258 seconds]
wallacoloo___ has joined #ipfs
chungy has joined #ipfs
HostFat_ has joined #ipfs
HostFat has quit [Ping timeout: 248 seconds]
chris6131 has joined #ipfs
chris613 has quit [Ping timeout: 258 seconds]
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
Alaura_ has joined #ipfs
Alaura has quit [Ping timeout: 258 seconds]
wallacoloo___ has quit [Quit: wallacoloo___]
xa0 has quit [Ping timeout: 246 seconds]
SuperPhly has joined #ipfs
xa0 has joined #ipfs
xa0 has joined #ipfs
xa0 has quit [Changing host]
s_kunk has quit [Ping timeout: 260 seconds]
wallacoloo___ has joined #ipfs
fiatjaf has quit [Ping timeout: 252 seconds]
fiatjaf has joined #ipfs
ygrek_ has quit [Ping timeout: 258 seconds]
bwerthmann has quit [Ping timeout: 245 seconds]
slothbag has quit [Quit: Leaving.]
i[m] has left #ipfs ["User left"]
Jon30 has quit [Read error: Connection reset by peer]
s_kunk has joined #ipfs
ygrek_ has joined #ipfs
lkcl_ has quit [Ping timeout: 248 seconds]
lkcl has quit [Ping timeout: 252 seconds]
mguentner2 has quit [Quit: WeeChat 1.6]
mguentner has joined #ipfs
bmoriarty has quit [Ping timeout: 250 seconds]
bmoriarty has joined #ipfs
HostFat_ has quit [Quit: Leaving]
lkcl_ has joined #ipfs
mguentner2 has joined #ipfs
mguentner has quit [Ping timeout: 240 seconds]
SuperPhly has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
SuperPhly has joined #ipfs
vtomole has joined #ipfs
chris6131 has quit [Quit: Leaving.]
pfrazee has quit [Remote host closed the connection]
vtomole has quit [Ping timeout: 260 seconds]
patcon has joined #ipfs
SuperPhly has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
ploop has quit [Ping timeout: 250 seconds]
ploop has joined #ipfs
rendar has joined #ipfs
ulrichard has joined #ipfs
<wking>
jbenet, victorbjelkholm: jbenet/depviz#66 will unblock a number of outstanding issues, so if either of you has time to land it (or the other two non-WIP depviz PRs) tomorrow, that would be great :)
leeola has quit [Quit: Connection closed for inactivity]
lkcl_ has quit [Ping timeout: 256 seconds]
konubinix has joined #ipfs
konubinix has quit [Ping timeout: 248 seconds]
konubinix has joined #ipfs
ulrichard has quit [Ping timeout: 268 seconds]
konubinix has quit [Ping timeout: 264 seconds]
konubinix has joined #ipfs
ulrichard has joined #ipfs
ylp has joined #ipfs
konubinix has quit [Ping timeout: 252 seconds]
kvda has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
cemerick has quit [Ping timeout: 258 seconds]
jchevalay has joined #ipfs
konubinix has joined #ipfs
wallacoloo____ has joined #ipfs
wallacoloo___ has quit [Ping timeout: 264 seconds]
bastianilso_____ has joined #ipfs
bwerthmann has joined #ipfs
ianopolous_ has joined #ipfs
Guest18 has joined #ipfs
s_kunk has quit [Ping timeout: 258 seconds]
ylp has quit [Ping timeout: 258 seconds]
d6e has quit [Ping timeout: 250 seconds]
d6e has joined #ipfs
ylp has joined #ipfs
espadrine has joined #ipfs
ianopolous_ has quit [Ping timeout: 250 seconds]
palkeo has joined #ipfs
chungy has quit [Ping timeout: 258 seconds]
d6e has quit [Ping timeout: 252 seconds]
<victorbjelkholm>
wking: sweet, configuration page! I'll take a look later today
Kingsquee has quit [Quit: Konversation terminated!]
<widmaster>
Hi, I don't understand the link between unixfs and ipns: I would like to publish my unixfs / to a ipns name. How do I do that ?
ofdm has quit [Ping timeout: 258 seconds]
lkcl has joined #ipfs
<jchevalay>
hi widmaster for ipns the goals is associate human entry to hash ressource for example can access your ressource /ipns/my-ressource instead of /ipfs/your hash. but link between unixfs and ipns i'm sorry i can't explain that or explain how to use
<widmaster>
Well actually I understand now how it works: we can create any directory structure and when we want to make it availlable, publish it.
<Kubuxu>
ipfs name publish
<jchevalay>
ipfs name publish <your_hash_ressource> get the hash after ipfs add <your_path_to_ressource>
ylp has quit [Ping timeout: 258 seconds]
<widmaster>
Yes it works.
<jchevalay>
i think you can add somes stucture in your directory
<jchevalay>
it's good news ^^
corvinux has joined #ipfs
<widmaster>
This is what I have done: echo test | ipfs files write /test1 --create ; ipfs name publish `ipfs files stat --hash /`
<widmaster>
This way I can change my unixfs in any way I want then republish it with: ipfs name publish `ipfs files stat --hash /`
<widmaster>
And I get at the same time mutability and more namespaces than just a single file under my id
corvinux has quit [Quit: Leaving]
<jchevalay>
yep
<jchevalay>
Are there any French people who would be interested in translating the IPFS documentation into French ?
widmaster has quit [Ping timeout: 260 seconds]
ylp has joined #ipfs
seafood has quit [Quit: seafood]
<Kubuxu>
So far we are quite reluctant about translations as our docs are iterated quite quickly.
<jchevalay>
ah ok
seafood has joined #ipfs
<jchevalay>
There are french communities around IPFS then?
<Kubuxu>
no idea, sorry
<ansuz>
ou pas
<ansuz>
I think most of the french speakers who might be interested in IPFS at this point already speak english quite well
<Kubuxu>
ansuz: you are not french, :p
<ansuz>
^
* ansuz
just lives in Paris
<ansuz>
I think the valuable things to translate are user interfaces on ipfs apps
<ansuz>
things that will be used by non-devs
<jchevalay>
i agree
<jchevalay>
To interess non-dev we must already be able to communicate with non-devs. I think that we must be able to present the project to the dev and not-devs almost everywhere and in many languages
<ansuz>
is there something like a flagship gui app for ipfs?
<ansuz>
I've only ever used the cli
<Kubuxu>
not really, we are working on a few right now
<jchevalay>
i think haadcode/orbit
<Kubuxu>
the biggest one is ^^
<jchevalay>
a chat
<apiarian>
jchevalay: agreed! i'm plugging along on a system that devs can share IPFS with their non-dev friends
disgusting[m] has quit [Quit: Client limit exceeded: 5000]
<jchevalay>
Another example a member of the community does not know the flagship application of the project
<jchevalay>
Local communities can also create synergies around the project or its
red5d[m] has quit [Quit: Client limit exceeded: 5000]
<lgierth>
first-proposal.md is bit more detailed than the readme
<ansuz>
cool
jchevalay has quit [Ping timeout: 260 seconds]
<frood>
lgierth: unless I'm misreading, only the CID's <mh> portion would be used for DHT routing.
<lgierth>
frood: yeah but it's intended to support the whole CID. the dht is from a time when CIDs didn't exist yet -- one of the spots that we missed while designing everything for pluggability
<lgierth>
it's a bit tricky to get the compatibility between old and new dht protocols right
<frood>
lgierth: I'm still not seeing the applicability of CIDs to DHT routing. Using the CID as a key feels like a sugar layer over using the hash as a key. The KV store should be codec and CID version agnostic, no?
<frood>
i.e. the CID is useful to agents looking up values for a purpose, but not to the DHT routing algorithm.
Kane` has quit [Ping timeout: 258 seconds]
Kane` has joined #ipfs
draynium_ has quit [Ping timeout: 248 seconds]
ashark has joined #ipfs
cjb has joined #ipfs
lkcl has quit [Ping timeout: 252 seconds]
musicmatze[m] has quit [Quit: Client limit exceeded: 5000]
<Mateon1>
frood: Yep, and it also makes it deduplicate less data, and more difficult to switch to other multihahes. I said pretty much the same thing yesterday or 2 days aho
PseudoNoob has joined #ipfs
<frood>
my followup question is whether there are plans for the DHT routing to support digest lengths besides 256 bits
<frood>
because that seems like a hard problem for kademlia routing
<Mateon1>
frood: Isn't it possible to truncate or pad the hash for kademlia? Not quite sure about how it works, so I might be wrong
<frood>
you can truncate, but I don't think you can pad elegantly without biasing routing
brodo has joined #ipfs
<brodo>
hi, are there some slides i can use for a "how IPFS works" talk?
<Mateon1>
jbenet: ^
jchevalay has joined #ipfs
<frood>
I'm curious why sha256 was chosen as well. been browsing old discussions, but haven't stumbled on it yet. Kademlia usually uses 160 bits, which should be more than enough.
<jchevalay>
if you find a good presention how it works i'm very intresting ^^
<brodo>
ah, i havent seen that one yet
<brodo>
and you can download the slides. thanks!
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<lgierth>
frood: note that the dht and choice of hash aren't related to each other
<lgierth>
the fact that the dht currently doesn't support anything but sha256 multihashes is a design bug
<frood>
well, the choice of hash for kademlia determines things like number of k-buckets and message routing
<lgierth>
wait i think actually that design bug is in bitswap (the block exchange above the dht)
<lgierth>
yes in the strict sense of the kademlia paper -- the dht is based on that and s/kademlia
ylp has quit [Quit: Leaving.]
robattila256 has quit [Ping timeout: 250 seconds]
<lgierth>
what i mean is that the choice of sha256 as the the current default multihash had nothing to do with what the original kademlia paper mandates
<lgierth>
the default hash function might be switched to blake2 eventually
<lgierth>
we have multihash exactly for being able swawp out different hash algorithms
shizy has joined #ipfs
ulrichard has quit [Ping timeout: 258 seconds]
pfrazee has joined #ipfs
s_kunk has quit [Ping timeout: 258 seconds]
bastianilso_____ has quit [Ping timeout: 246 seconds]
GandalfTheWhite has joined #ipfs
GandalfTheWhite has quit [Max SendQ exceeded]
GandalfTheWizard has joined #ipfs
GandalfTheWizard has quit [Max SendQ exceeded]
AragonTheStrider has joined #ipfs
jchevalay has quit [Quit: Page closed]
<Kubuxu>
lgierth: DHT hashes the hashes
<Kubuxu>
it supports other hashes
<Kubuxu>
also DHT doesn't need strong guarantees about hash
<Kubuxu>
even MD5 would be fine
<lgierth>
i keep forgetting aboutt the hashing-again
Boomerang has joined #ipfs
patcon has joined #ipfs
infinity0 has quit [Remote host closed the connection]
<whyrusleeping>
so something bugged out on one side
<whyrusleeping>
there should be two streams open
<Kubuxu>
already killed the node unfortunately
<whyrusleeping>
i'm 90% sure this has been fixed in newer versions
<Kubuxu>
kk, just updated
<Kubuxu>
it was the reason why it didn't pin, because I was propagating hashes through my server to gexpin (which doesn't have cjdns), and it stopped there
<_mak>
the cmd ipfs stats repo is taking a ridiculous amount of time
<_mak>
is it normal?
<_mak>
NumObjects 12210036
<_mak>
RepoSize 730080273898
jokoon has quit [Quit: Leaving]
<Kubuxu>
yes because it calculates size and counts objects
<_mak>
I see
<_mak>
thanks then :)
arpu has joined #ipfs
Ronsor has quit [Ping timeout: 258 seconds]
se3000 has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
cemerick has quit [Ping timeout: 265 seconds]
<richardlitt>
dignifiedquire: I honestly don't know
<richardlitt>
I am having a hell of a time trying to understand how to build even the simplest remark plugin
<richardlitt>
I recall now why I only built a couple, based heavily on other people's. It requires a pretty deep knowledge of the ecosystem, and the amount of effort needed to do something like add a section is monumental. It's faster for me to use a regex linter and just manually edit READMEs, although that's not a long term solution.
Ronsor has joined #ipfs
se3000 has joined #ipfs
<richardlitt>
wooorm: you don't have a tip for levelling up in remark faster, do you?
chris613 has joined #ipfs
cemerick has joined #ipfs
<richardlitt>
To be fair, remark is great. It is just very architectured, and it's hard to navigate for someone who wants to do simple things
lkcl has joined #ipfs
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
Boomerang has quit [Quit: Lost terminal]
<_mak>
how can I check how many nodes are seeding a given file
<Mateon1>
_mak: ipfs dht findprovs <hash>, but note that this only shows the providers of a single block. You probably want to check providers of all blocks below the root hash as well.
<dignifiedquire>
richardlitt: I will take a look myself tomorrow and think a little about it
<_mak>
Mateon1: how do I know the difference between a block and the root hash?
<Mateon1>
_mak: If you type `ipfs refs -r <hash>`, you will get all block hashes that belong to the hash you gave. You can then check every hash with `ipfs dht findprovs <hash>`
<Mateon1>
I feel like this should be in the FAQ, I answered this question three times to date. Let me check
Kingsquee has joined #ipfs
<_mak>
Mateon1: thanks mate
<_mak>
Mateon1: if the result of refs is nothing it means that the hash is the root hash?
<Mateon1>
_mak: That means the hash has no child blocks.
cemerick has quit [Ping timeout: 246 seconds]
<Mateon1>
Most likely it's a file less than 256 kB in size
<_mak>
so on ipfs pin ls 'recursive' means that the hash is a child block and indirect means that it is the root hash?
<Mateon1>
Uhh.. Not exactly. 'recursive' means all hashes that are children to the recursive hash are pinned 'indirect'. So if I have a directory with a file, and I pin the directory, the directory is pinned recursively, and the file is pinned indirectly
<_mak>
hmm I guess I need to re-read the docs :/
<_mak>
thanks mate :)
Aranjedeath has quit [Quit: Three sheets to the wind]
kvda has joined #ipfs
slothbag has joined #ipfs
patcon is now known as patcon|afk
galois_dmz has quit []
pfrazee has quit [Remote host closed the connection]
patcon|afk has quit [Ping timeout: 268 seconds]
galois_dmz has joined #ipfs
<slothbag>
is the relay functionality broken in go-ipfs right now?
<lgierth>
slothbag: probably -- i don't know of anyone actually using it
<lgierth>
slothbag: happy about bug reports :)
<richardlitt>
dignifiedquire: cool. I tried for a while today to do something as simple as replacing badges based on the language of a file
beta_ignatz has quit [Quit: Leaving]
<richardlitt>
after a few hours, I wasn't much further along than when I started. Navigating the factories, and keeping in mind remark, remark lint, mdast, unist, and their differences is not easy.
<Kubuxu>
I don't know if CoreAPI is best place for it
<Kubuxu>
lgierth, whyrusleeping, daviddias ^^
bwerthmann has joined #ipfs
herzmeister has quit [Quit: Leaving]
<kevina>
Kubuxu: Not sure where else to put the call to read the contents of a symbolic link...
<kevina>
the coreapi has an "Ls" call so I thought a "ReadSymLink" would also be okay, but am open to suggestions.
<Kubuxu>
it might be ok
<lgierth>
coreapi is an okay place i think, for me the question is only whether resolution should happen implicitly (within Cat() and Ls()), or explicitly
herzmeister has joined #ipfs
<Kubuxu>
thing is, we should spec it out and confirm with js-team
<alterego>
Really wish I could integrate STUN into ipfs somehow
<lgierth>
i think this is the first time we're attempting the read-end of symlinks
<Kubuxu>
alterego: currently we are using TCP mostly but we are working on UDP based connections
<alterego>
whyrusleeping: so it's half way there for me.
anonymuse has quit [Ping timeout: 260 seconds]
<whyrusleeping>
ah, i see. You want to abuse ipfs to do discovery for an external application
<alterego>
whyrusleeping: sort of, it's for web apps distributed on ipfs on devices that have ipfs daemon running.
<whyrusleeping>
hrmmm
<alterego>
whyrusleeping: so "abuse" is slightly harsh way of putting it :P
<whyrusleeping>
haha
<whyrusleeping>
yeah, i think what would work really well for you is if we managed to get the corenet dialing api stuff done
<alterego>
whyrusleeping: that would be great, especially if it had a good interface via the api
<alterego>
web api, as that's what I'm using.
<alterego>
Everything for my apps has to run in a browser.
<whyrusleeping>
yeap
<whyrusleeping>
Kubuxu: lets prioritize that for 0.4.6
<alterego>
This is kind of why I was thinking WebRTC, because if we could have that layered on IPFS, then you've got some pretty easy porting work for all manner of apps
<alterego>
Where as having an ipfs specific channel, which goes through the ipfs daemon http api, seems a bit round-the-houses.
<alterego>
(from my perspective) it's bloody cool otherwise.