<A124>
Could anyone address/comment my issue please? What happens if you throw bunch of requests at it? In my experience it keeps at "connecting". Doing that on public gateway (more peers) it took longer to choke it up then on gateway with 70 peers.
palkeo_ has quit [Ping timeout: 252 seconds]
cemerick has joined #ipfs
<whyrusleeping>
A124: not sure what youre asking about.
dignifiedquire has quit [Quit: Connection closed for inactivity]
cketti has quit [Quit: Leaving]
anewuser has joined #ipfs
palkeo_ has joined #ipfs
<sdgathman>
A non-enumerable cache seems pretty doable. My idea was to truncate or rehash the CID to a 32-bit value, and look that up in a hash table. Entries are encrypted using the full hash, and collisions are resolved by decrypting the entries to see which matches the full hash.
<sdgathman>
Thus you can enumerate all the 32-bit truncated hashes (or hashes of hashes), but not all the full hashes.
<sdgathman>
To avoid decrypting entire contents to resolve a collision, contents can be prepended with the CID before encrypting - so that only the header need be decrypted to verify a match.
<sdgathman>
My question concerning that last point:
<sdgathman>
Does encrypting a key with the same key make it more vulnerable to recovery from ciphertext?
JesseW has joined #ipfs
<sdgathman>
Perhaps the CID should be transformed in some easily reversible way before encrypting along with contents (but randomly choosing such schemes can make things worse).
<whyrusleeping>
sdgathman: hrm... encrypting a key with itself (what some people call 'convergent encryption') is definitely weak to confirmation attacks
<whyrusleeping>
although i dont think you can get around that if youre building a cache
<whyrusleeping>
and i don't think it makes it any 'more' vulnerable
sametsisartenep has joined #ipfs
<sdgathman>
I'm thinking that while the non-enumerable cache may not protect an individual, it could protect a corporation with expensive lawyers.
<sdgathman>
And it would be a good thing for corporations to run general ipfs caches (excluding government ban-lists, of course).
<kevina>
whyrusleeping: I am trying to change EnumerateChildren to take in a CID but ran into a problem with a test case.
<whyrusleeping>
kevina: oh?
<kevina>
Have a look at TestFetchGraph in merkledag_test.go
<sdgathman>
If govt bans your manifesto (or evil image), bumping the version (or changing one pixel) gives it an entirely different hash - so it seems any govt ban-list could be easily overwhelmed.
<sdgathman>
But that process would also make it hard for would be consumers to find a working CID.
<kevina>
whyrusleeping: It creates a Node, but doesn't store it anywhere.
<kevina>
whyrusleeping: at least I think that is what it is doing...
<whyrusleeping>
kevina: doesnt it? you could just store that node in dservs[0] if it doesnt already
<whyrusleeping>
hrm... no. BuildDagFromReader definitely stores the object it returns
<whyrusleeping>
i think the issue here is that FetchGraph doesnt fetch the root node
Kane` has joined #ipfs
<whyrusleeping>
heh, FetchGraph should be called 'FetchGraphExceptForThatOneNodeAtTheTop'
<whyrusleeping>
(i'd call that a bug)
<kevina>
I'm confused, so should I just be able to get the CID of `root` and pass that in instead of root.Links?
<whyrusleeping>
Yes, except on line 234 where we call FetchGraph, its fetching that graph to node 1 from node 0
<whyrusleeping>
but the issue is that FetchGraph fetches all nodes below the given node
<whyrusleeping>
so dserv[1] never ends up having 'root'
<whyrusleeping>
so when you get to EnumerateGraph, and call 'Get' on the root cid, it cant find it
<kevina>
Okay, so how best to fix?
<whyrusleeping>
this worked before because you never asked EnumerateChildren to fetch the root node
<whyrusleeping>
i would probably make fetch graph also take a cid
<whyrusleeping>
especially since FetchGraph is literally just a call through to EnumerateChildrenAsync...
* whyrusleeping
runs git blame and finds only himself at fault
<kevina>
but then EnumerateChildrenAsync will also need to be changed
<kevina>
whyrusleeping: And in any case the root node must still exist in the LinkService/DAGService passed in,
<kevina>
which if I understand you correctlly is not the case.
<whyrusleeping>
changing ECAsync is fine
<whyrusleeping>
if you do that, then it will pull the root node over
<whyrusleeping>
the 'FetchGraph' operation is 'online'
<whyrusleeping>
where the second 'EnumerateChildren' operation is 'local'
bsm117532 is now known as Guest95984
<kevina>
I'm sorry but I am not following you.
cemerick has quit [Ping timeout: 264 seconds]
<kevina>
whyrusleeping: I am also getting another unit test failure.
<kevina>
whyrusleeping: if I push the code with the broken go tests do you think you could fix them?
sametsisartenep has quit [Quit: leaving]
<kevina>
whyrusleeping: but hold off on that, shareness tests are failing, so its likely my fault....
<whyrusleeping>
kevina: sure, just point me to it when youre ready
sametsisartenep has joined #ipfs
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
sametsisartenep has quit [Client Quit]
structuralist has quit [Remote host closed the connection]
<whyrusleeping>
should be pretty easy to grab all things we've asked pinbot to pin from there
<A124>
Yeah but botbot.me has no export.
<A124>
This "dataset" should simulate diverse load of different users on a loaded gateway node.
<A124>
Of course until cache fill.
<whyrusleeping>
kevina: i fixed one of the test issues
<whyrusleeping>
although another one is being weird still
anonymuse has joined #ipfs
<kevina>
okay, can you just push to that branch when ready
<kevina>
I won't do anything else until I get the all clear....
<whyrusleeping>
just pushed
anonymuse has quit [Remote host closed the connection]
<whyrusleeping>
TestEnumerateChildren is failing
<whyrusleeping>
(which explains why the t0080 sharness tests are failing too)
kulelu88 has quit [Quit: Leaving]
herzmeister has quit [Quit: Leaving]
<kevina>
Um, okay, so I just found my really stupid error!
<whyrusleeping>
lol, what was it?
<whyrusleeping>
because i'm still looking
<kevina>
I will push my fix.
herzmeister has joined #ipfs
<whyrusleeping>
kevina: haha, nice
<kevina>
whyrusleeping: your code is not compiling....
<whyrusleeping>
its not?
<whyrusleeping>
ah, other callers of fetchGraph need updating
<kevina>
yep
<kevina>
do you want to fix? or should I?
<whyrusleeping>
i can if you want, should be pretty quick
<kevina>
I'll do it. Well see if things pass this time and will ping if I run into more problems.
<whyrusleeping>
sounds good
<whyrusleeping>
i'm seeing another weird failure in pin_test.go though
anewuser_ has joined #ipfs
<whyrusleeping>
got it, TestPinRecursiveFail adds 'b' before the final pin check, it also needs to add 'a' now since we're using cids and fetching them internally
anewuser has quit [Read error: Connection reset by peer]
PrinceOfPeeves has quit [Quit: Leaving]
ygrek_ has quit [Ping timeout: 248 seconds]
JesseW has joined #ipfs
lkcl has joined #ipfs
<kevina>
just saw this, pushed the compile fix, could you push your fix
Mateon1 has quit [Ping timeout: 264 seconds]
<whyrusleeping>
yeap, one sec
<whyrusleeping>
pushed
anewuser_ has quit [Quit: anewuser_]
xeed has quit [Ping timeout: 248 seconds]
mgue has quit [Quit: WeeChat 1.5]
xeed has joined #ipfs
<kevina>
whyrusleeping: all right this is looking good, I am going to squash are commits
mgue has joined #ipfs
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
<whyrusleeping>
interesting... running ipfs on a raspberry pi, spends a lot of its time just doing crypto
matoro has quit [Remote host closed the connection]
structuralist has quit [Remote host closed the connection]
matoro has joined #ipfs
<Igel>
across all cores tho?
<A124>
What is the simplest way to make a "directory+filenames" from existing hashes?
<whyrusleeping>
Igel: just one
<whyrusleeping>
Igel: occasionally spread across two or so though
<whyrusleeping>
A124: what to you mean?
<A124>
whyrusleeping Upload stuff via pipe is possible? And even if not, let say I have hashes pointing to files, and I want to give it names and wrap in directory.
<A124>
Aka content management.
HastaJun_ has quit [Ping timeout: 240 seconds]
<achin>
A124: i think you are proposing that ipfs-add gains a new command line argument, to specify the filename while reading data from stdin?
<A124>
Not exactly, but that would be great too.
<A124>
But imagine you want to take a bunch of files aready in blockstore (or someones else node) and create links on top of that
<achin>
maybe you're looking for `ipfs object patch` or something like it?
<whyrusleeping>
A124: you can wrap a given hash in a directory by taking an empty directory: QmUNLLsPACCz1vLxQVkXqqLX5R1X345qqfHbsf67hvA3Nn
<whyrusleeping>
and doing: ipfs object patch add-link QmUNLLsPACCz1vLxQVkXqqLX5R1X345qqfHbsf67hvA3Nn foo $YOUR_HASH
<whyrusleeping>
and that will spit out a hash of a directory containing 'foo' that points to '$YOUR_HASH'
<A124>
Indeed. Which when doing two step, would also allow pipe. But optionaly providing name could be easier for some user, idk.
<A124>
whyrusleeping Yeah, that is single file. If it is a tree structure, that takes kind of work.
<A124>
The --help gives at least impression you have to do it one file by one.
<whyrusleeping>
Yeah, add-link adds one at a time
<whyrusleeping>
or you could manually create the object in JSON
<achin>
another option would be to construct the object in the language of your choice and write out json
<whyrusleeping>
and 'put' it
<whyrusleeping>
achin on point tonight :P
<achin>
\o/
<achin>
just in time for bed
<A124>
whyrusleeping How do I do that though?
<A124>
I mean, I looked at the structure, but there is magic data field.
<whyrusleeping>
A124: don't touch the data field, its a magic number
<A124>
Tried to look at IPLD and stuff (not that in depth) and missed all specs.
<achin>
extract the magicness from `ipfs object get QmUNLLsPACCz1vLxQVkXqqLX5R1X345qqfHbsf67hvA3Nn`
<A124>
So I just make json and data is created automatically?
<achin>
it's a protobuf encoding of an empty PBNode structure
<A124>
Yeah, but ... then I cannot construct it.
<achin>
what do you mean?
<A124>
Or I just use that magic same each time, no matter how many files and no matter how nested?
<achin>
just use the same magic
<achin>
the only stuff you'll be changing are the links, if i understand your usecase correctly
<achin>
Links is just an array of objects, each object having a 'Name', 'Hash' and 'Size' key
<achin>
i confess i don't know exactly what 'Size' is supposed to be. i think it's suppose to be the size of the data linked, assuming it was a file or a directory of files
<A124>
Size is the whole file or data including all protobuf wrapping.
<A124>
Thanks then! So I did it "right" last time. And now I know I can just copy that or make a go tool for the same.
<achin>
with that successful note, i'm off to bed o/
slothbag has joined #ipfs
lkcl has quit [Ping timeout: 265 seconds]
matoro has quit [Remote host closed the connection]
nothingmuch has quit [Ping timeout: 276 seconds]
nothingmuch has joined #ipfs
HastaJun_ has joined #ipfs
matoro has joined #ipfs
ploop has quit [Ping timeout: 276 seconds]
wallacoloo has joined #ipfs
em-ly has quit [Quit: Connection closed for inactivity]
Foxcool has joined #ipfs
ploop has joined #ipfs
codebam has quit [Quit: later :)]
pfrazee has quit [Remote host closed the connection]
chriscool has joined #ipfs
kvda has joined #ipfs
ygrek_ has joined #ipfs
captain_morgan has joined #ipfs
captain_morgan has quit [Remote host closed the connection]
captain_morgan_ has joined #ipfs
JesseW has quit [Ping timeout: 272 seconds]
captain_morgan_ has quit [Ping timeout: 264 seconds]
ulrichard has joined #ipfs
Tv` has quit [Quit: Connection closed for inactivity]
tokyo_jesus has joined #ipfs
Mateon1 has joined #ipfs
rgrinberg has quit [Ping timeout: 264 seconds]
dignifiedquire has joined #ipfs
Aranjedeath has quit [Quit: Three sheets to the wind]
structuralist has joined #ipfs
rendar has joined #ipfs
structuralist has quit [Remote host closed the connection]
JesseW has joined #ipfs
zombu2 has quit [Ping timeout: 272 seconds]
structuralist has joined #ipfs
corvinux has joined #ipfs
mildred has joined #ipfs
tooldels is now known as toop
lacour has quit [Quit: Leaving]
taaem has quit [Ping timeout: 264 seconds]
lkcl has joined #ipfs
<haad>
dignifiedquire: is that merkledag.proto fix already in in ipfs-merkle-dag npm package?
cjd has quit [Ping timeout: 248 seconds]
zorglub27 has joined #ipfs
<haad>
dignifiedquire: is there a way for me to write this: `const IpfsApi = require('exports?IpfsApi!ipfs-api/dist/index.js')
<haad>
...in webpack config instead of using that weird syntax?
<oed>
is there any documentation availiable for the pubsub protocol?
<dignifiedquire>
haad: no
<dignifiedquire>
(it is not merged yet)
<dignifiedquire>
haad: there is, but then you have to write that weird syntax in the config :D
Kane` has quit [Remote host closed the connection]
mgue has quit [Ping timeout: 252 seconds]
<dignifiedquire>
haad: I think Orbit releases should be both on github and on dist.ipfs.io, githbub releases are just the place everyone is going to look for to download if they are familar with github projets
corvinux has quit [Remote host closed the connection]
palkeo_ has quit [Ping timeout: 252 seconds]
<dignifiedquire>
daviddias: can we try to get a first batch of review and releases out today for webcrypto? *pretty please*
kingflurkel has joined #ipfs
<kingflurkel>
hi guys
<kingflurkel>
good morning from belgium
<daviddias>
we can definitely try :)
<kingflurkel>
daviddias who's working on the js-ipfs-api ?
<daviddias>
dignifiedquire: re: reviews, won't be able to commit today (given all the other stuff), but I've a pretty good feeling about being able to go through all of them tomorrow. Will try to start today at least :)
<dignifiedquire>
daviddias: sounds good, it would be great if you can focus on the crypto one first, as that is the first one that needs release to get many oft the others to a green ci
<kingflurkel>
So I'm trying to 'correct' that in my fork
<kingflurkel>
I was hoping someone here has 5 minutes time to look at that
<kingflurkel>
It's the only thing we're stuck at to launch our prototype for Arcade City
mgue has joined #ipfs
<kingflurkel>
Do I sound desperate yet
keks has quit [Ping timeout: 272 seconds]
<dignifiedquire>
you sound like you are trying :)
<dignifiedquire>
kingflurkel: you could try to use `url.parse`(https://www.npmjs.com/package/url) on line 18 to get the correct protocol
kenshyx has joined #ipfs
<lgierth>
aww come one github, let me comment on lines in PRs which aren't deletions or additions... :/ this used to work
cjd has joined #ipfs
wallacoloo has quit [Quit: wallacoloo]
<lgierth>
now it works, wtf. did i get a timeout to cool down?
zorglub27 has quit [Quit: zorglub27]
espadrine has joined #ipfs
<demize>
:p
cketti has joined #ipfs
cketti has joined #ipfs
cketti has quit [Changing host]
<hsanjuan>
richardlitt: hey
structuralist has quit [Remote host closed the connection]
<kulelu88>
how is protocol labs funded? VC? government grants?
Tv` has joined #ipfs
soloojos has quit [Ping timeout: 258 seconds]
<lgierth>
a bit from here and a bit from there -- i'm not free to talk about it :)
<kulelu88>
interesting.
dmr has joined #ipfs
Boomerang has joined #ipfs
dmr has quit [Max SendQ exceeded]
dmr has joined #ipfs
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
jedahan has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
JesseW has quit [Ping timeout: 264 seconds]
ruby32 has joined #ipfs
ruby32 has quit [Remote host closed the connection]
soloojos has joined #ipfs
jedahan has joined #ipfs
Stebalien has joined #ipfs
dmr has quit [Quit: Leaving]
richardlitt[cli] has joined #ipfs
matoro has quit [Ping timeout: 248 seconds]
lacour has joined #ipfs
richardlitt[cli] has quit [Ping timeout: 244 seconds]
galois_d_ has joined #ipfs
cketti has quit [Quit: Leaving]
galois_dmz has quit [Ping timeout: 248 seconds]
richardlitt[cli] has joined #ipfs
Monokles has quit [Quit: No Ping reply in 180 seconds.]
Oatmeal has quit [Ping timeout: 252 seconds]
Monokles has joined #ipfs
richardl1tt[cli] has joined #ipfs
Stebalien has quit [Remote host closed the connection]
jedahan has quit [Remote host closed the connection]
jedahan has joined #ipfs
richardlitt[cli] has quit [Ping timeout: 264 seconds]
ylp has quit [Ping timeout: 252 seconds]
ygrek_ has joined #ipfs
jedahan has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ylp has joined #ipfs
jedahan has joined #ipfs
jedahan has quit [Client Quit]
rgrinberg has quit [Ping timeout: 264 seconds]
espadrine has quit [Ping timeout: 265 seconds]
cemerick has joined #ipfs
Oatmeal has joined #ipfs
<dignifiedquire>
daviddias: good that you didn't review today :D I realised earlier that I hadn't tested the full down interop and had to fix some more issues, but now all is good and 💯 interop is back
kvda has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
rgrinberg has joined #ipfs
lkcl has joined #ipfs
ulrichard has quit [Remote host closed the connection]
captain_morgan_ has quit [Ping timeout: 244 seconds]
<daviddias>
ahah :D perfect
gmcquillan__ has joined #ipfs
richardl1tt[cli] has quit [Ping timeout: 244 seconds]
jedahan has joined #ipfs
ygrek_ has quit [Ping timeout: 244 seconds]
taaem has joined #ipfs
G-Ray has quit [Remote host closed the connection]
s_kunk has quit [Ping timeout: 265 seconds]
cemerick has quit [Ping timeout: 244 seconds]
chriscool has joined #ipfs
Encrypt has joined #ipfs
<haad>
updated: https://github.com/haadcode/orbit-db - now with less servers and more distributed pubsub \o/ let me know if there's anything you would like to see in the readme/examples, or any feedback whatsoever
anonymuse has quit [Remote host closed the connection]
cemerick has joined #ipfs
<dignifiedquire>
haad: does orbit have reactions? like slack on individual messages? that would be super awesome, I'm really missing that from irc
Qwertie has quit [Ping timeout: 272 seconds]
Qwertie- has joined #ipfs
kenshyx has quit [Remote host closed the connection]
<Mateon1>
I tried to build on AMD64 linux, and failed (due to something with Node) and using a prebuilt dist binary (one of the ones linked/pinned recently) didn't launch (sigsegv)
<Mateon1>
Oh goddamnit
ylp has quit [Ping timeout: 252 seconds]
<Mateon1>
Now I can't reproduce that sigsegv
<Mateon1>
It works semi-perfectly
<Mateon1>
AKA, the Orbit binary instantly returns to shell and doesn't seem to do anything, but keeps running in the background
<Mateon1>
So.. How do I actually use electron?
Encrypt has quit [Quit: Quit]
anonymuse has joined #ipfs
ylp has joined #ipfs
rgrinberg has quit [Ping timeout: 272 seconds]
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
kingflurkel has quit [Quit: Page closed]
arnav has quit [Ping timeout: 265 seconds]
rendar has quit [Ping timeout: 248 seconds]
palkeo_ has joined #ipfs
CHOCODARCHEVILLE has joined #ipfs
<apiarian>
is there something wrong with https://ipfs.io/ipns/ipn.io/ ? is the dns text value of a domain supposed to point to an IPFS based mirror of the same website? is that a loose convention?
<whyrusleeping>
apiarian: either the TXT on the domain itself, or the TXT on _dnslink.YOURDOMAIN
Mateon3 has joined #ipfs
CHOCODARCHEVILLE has quit [Excess Flood]
Mateon1 has quit [Ping timeout: 272 seconds]
Mateon3 is now known as Mateon1
rgrinberg has joined #ipfs
cjb has joined #ipfs
pratch has quit [Read error: Connection reset by peer]
rendar has joined #ipfs
Encrypt has joined #ipfs
galois_d_ has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
galois_dmz has quit [Ping timeout: 248 seconds]
* A124
has node that is going berserk on CPU. 200+%
<whyrusleeping>
A124: wanna grab a profile for me?
<whyrusleeping>
send me that file and your ipfs binary, would be very useful :)
G-Ray has joined #ipfs
<A124>
whyrusleeping I already killed it. Can you give me how to update to latest dev ?
<A124>
I run dev on localhost, and stable else. I have suspicion that ... it has something to do either with disconnections, packet loss, or not being able to listen on 4001 publicaly.
Boomerang has quit [Quit: Leaving]
<A124>
rm -rf should do the trick I hope.
Cris_ has joined #ipfs
captain_morgan has joined #ipfs
mildred has quit [Ping timeout: 264 seconds]
Encrypt has quit [Quit: Quit]
matoro has joined #ipfs
se3000 has joined #ipfs
Mateon3 has joined #ipfs
richardlitt[cli] has joined #ipfs
Mateon1 has quit [Ping timeout: 264 seconds]
Mateon3 is now known as Mateon1
chriscool has quit [Quit: Leaving.]
chriscool1 has joined #ipfs
richardlitt[cli] has quit [Ping timeout: 252 seconds]
jan has joined #ipfs
jan is now known as Guest3354
<Guest3354>
does anyone know an easy way to validate a multihash in Python? A valid hash in this case would be any hash that is formatted correctly.
<lgierth>
whyrusleeping: ok but what's the matter with the lack of deps
Oatmeal has joined #ipfs
<cjd>
Well then how come he's on irc, ay ?
<whyrusleeping>
lgierth: because it has two dependencies?
<lgierth>
oh, go-base58
<whyrusleeping>
and they arent in the package.json
<whyrusleeping>
and go-crypto
<lgierth>
i might have been lazy and ignorant *whistles innocently*
* whyrusleeping
grumbles about packaging nonsense
<lgierth>
simply forgot that all packages down that leaf need to be gx'd...
<whyrusleeping>
lol
<cjd>
is there no automatic CI for testing these types of issues ?
<whyrusleeping>
lgierth: fixing it
<whyrusleeping>
cjd: there should be
<whyrusleeping>
not sure why it didnt get caught
<whyrusleeping>
ah
<lgierth>
cjd: i think it breaks in a non-obvious way
<lgierth>
yeah how does it actually break?
<whyrusleeping>
we had old style CI scripts
<lgierth>
go get -d?
xavivives has quit [Read error: Connection reset by peer]
<whyrusleeping>
no, just go get and go test
<lgierth>
oh, yeah, i remember even running them
<lgierth>
and generally fixing ci
<vmatekole>
hello
<lgierth>
good $timeOfDay
_rht has quit [Quit: Connection closed for inactivity]
<vmatekole>
I am working on a simple membership registry stored using IPFS. The registry only stores changes to it e.g. adding a member or revoking a member.
<vmatekole>
Each state change is stored in JSON and will always link to the previous one.
<vmatekole>
Can anyone point me to some resources on the use of IPLD and linking IPFS objects?
<vmatekole>
Furthermore, what is the best way to persist JSON in IPFS?
<lgierth>
we don't have ipld in go-ifps yet
<whyrusleeping>
lgierth: the multihash tests don't even compile
<whyrusleeping>
wtf
<whyrusleeping>
its not your fault either
<lgierth>
vmatekole: so for now it's `echo "$json" | ipfs add`
<vmatekole>
What about whitespace etc?
Oatmeal has quit [Ping timeout: 244 seconds]
<vmatekole>
should I be stripping it?
<vmatekole>
lgierth: Also, what is the recommended way of linking ipfs objects?
matoro has joined #ipfs
<dansup>
whyrusleeping, where are the multihash tests? How would i test my own implentation with it?
<lgierth>
vmatekole: an object has links and data. the data is your json, the links can be modified using the object patch add-link/rm-link commands
<vmatekole>
lgierth: e.g. "@link": "/ipfs/QmVF73qEL9zLbFzLfAwZPehduFhsDFQTKK4yCVfDW9EtrZ"
<lgierth>
vmatekole: that's ipld, but we don't have it in go-ipfs yet
<lgierth>
once ipld is release it's definitely the recommended way
<vmatekole>
lgierth: but it is safe to start linking data in this way? Right?
<lgierth>
sure
<lgierth>
go-ipfs just currently can't make sense of these links
<lgierth>
0.5.0 will
<vmatekole>
lgierth: cool :) ... And where do multihashes come in to play - still can't make sense of where this comes into play in terms of navigating data - https://github.com/ipfs/js-ipfs-ipld
rsync has joined #ipfs
<lgierth>
vmatekole: dignifiedquire can tell you more about js-ipfs-ipld during UTC daytime
<vmatekole>
lgierth: Thanks!
<lgierth>
:)
<lgierth>
multihashes are just the hash in /ipfs/Qmsomething
<lgierth>
a format that fits all kinds of different hash algorithms output
<vmatekole>
lgierth: Ok. My concern is this something I have to account for in my data format to take advantage of?
<lgierth>
mh i can't give you a good answer to that at the moment
<lgierth>
i haven't dug into ipld too much myself
<vmatekole>
lgierth: Thanks for your advice. I'll get in touch with dignifiedquire.
<whyrusleeping>
dansup: in the go-multihash repo there are a few different types of tests