<lgierth>
ipfs daemon's output is still there though -- same as what was the log file
<slothbag>
lgierth: thanks, whats the best way to capture it? via the API?
<noffle>
I guess I could expose --local for 'mount'. hacky
<lgierth>
the output? just run ipfs daemon
<lgierth>
oh and you can set IPFS_LOGGING
<lgierth>
to e.g. debug
<slothbag>
mine just says Daemon is ready and nothing more
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
<noffle>
whyrusleeping: thanks -- this has me on the right track now
felixn has quit [Read error: Connection reset by peer]
Oatmeal has quit [Ping timeout: 276 seconds]
<whyrusleeping>
noffle: wo0oo!
<noffle>
whyrusleeping: in general we need a way to setup the daemon s.t. namesys resolves happen locally, for tests
<whyrusleeping>
the --local flag could work
<noffle>
a CLI-wise flag is the easiest/hackiest way
<whyrusleeping>
also, check the tests with iptb
<noffle>
CLI-wide*
<whyrusleeping>
they are able to resolve just fine
<whyrusleeping>
(because they set up multiple nodes)
<noffle>
oh, so they resolve to each other?
<noffle>
how does it work?
<noffle>
whyrusleeping: also, +1 for a testbed app :)
elima_ has quit [Remote host closed the connection]
<noffle>
wait, sanity check: if namesys resolution fails on mount, shouldn't we fallback to whatever's in the local cache? it seems bizarre to make this a hard failure
jhulten_ has quit [Ping timeout: 272 seconds]
<whyrusleeping>
eh...
<whyrusleeping>
theres weird safety implications around that
<whyrusleeping>
falling back to an old version might lead to 'corrupted' data
pfraze has joined #ipfs
kerozene has quit [Ping timeout: 256 seconds]
kerozene has joined #ipfs
tmg has quit [Ping timeout: 264 seconds]
tmg has joined #ipfs
HostFat has joined #ipfs
fiatjaf_ has joined #ipfs
kerozene has quit [Client Quit]
kerozene has joined #ipfs
Oatmeal has joined #ipfs
reit has quit [Quit: Leaving]
boxxa has quit [Quit: Connection closed for inactivity]
IlanGodik has quit [Quit: Connection closed for inactivity]
Matoro has quit [Ping timeout: 256 seconds]
<noffle>
whyrusleeping: you can never guarantee you have the latest ipns data anyways
<noffle>
we need to be very tolerant of having stale data anyways
<richardlitt>
achin: you around?
<noffle>
brb, they're killing the wifi soon at hanahaus on us :'(
<achin>
richardlitt: ohi
<richardlitt>
can you get people who have committed in a certain time range?
<achin>
aye
<richardlitt>
cool
<achin>
i think i'll only need to make a tiny tweak to get_commits.py
<richardlitt>
Want to throw me everyone from Jan 11 to jan 25?
<richardlitt>
Yeah, I was looking at it, I think I know the tweak to do.
<achin>
roger; one sec
<richardlitt>
I've spent today adding ranges to my js stuff, so now I can do that.
<richardlitt>
Going from 2016-01-11T00:01:01Z until 2016-01-25T18:00:01Z
Guest17506 has quit [Ping timeout: 240 seconds]
<achin>
oh! this reminds me that there was something i wanted to talk to you about
<achin>
let me try to remember :)
TheWhisper has quit [Read error: Connection reset by peer]
fiatjaf_ has quit [Remote host closed the connection]
<achin>
oh yeah! ok, so when people to big merge work (like whyrusleeping merging dev040 into master) there is often a lot of cherrypicking and rewriting. this causes my script to think that a commit that happned a long time ago is a new commit (and thus included in the report)
<achin>
i don't think this is a big deal, but just something that i wanted to mention
<richardlitt>
hmm
<richardlitt>
Good point.
<richardlitt>
Yeah, that's interesting. There's at least one name on here that I can't place
<richardlitt>
Attached to IPFS code, anyway
<achin>
what's the name?
Matoro has joined #ipfs
<richardlitt>
Make that... four?
Matoro has quit [Read error: Connection reset by peer]
Matoro has joined #ipfs
joshbuddy has quit [Quit: joshbuddy]
hellertime has joined #ipfs
<ipfsbot>
[go-ipfs] rht opened pull request #2261: Fix LinksSize calculation in node.Stat() (master...linksSize) https://git.io/vzFKa
Matoro has quit [Ping timeout: 256 seconds]
besenwesen has quit [Ping timeout: 256 seconds]
besenwesen has joined #ipfs
besenwesen has quit [Changing host]
besenwesen has joined #ipfs
Matoro has joined #ipfs
jhulten_ has joined #ipfs
voxelot has joined #ipfs
pierssturley has quit [Quit: Connection closed for inactivity]
r04r is now known as zz_r04r
<whyrusleeping>
noffle: no, we're incredibly careful of making sure we have the 'latest' ipns data
<whyrusleeping>
if the newest record exists on the network, we'll get it
<whyrusleeping>
(which is really why ipns is slow right now)
<achin>
whyrusleeping: is your DHT-only thingus capable of keeping IPNS entries published?
immethodize has quit [Ping timeout: 240 seconds]
Cary has joined #ipfs
dignifiedquire has quit [Quit: Connection closed for inactivity]
<slothbag>
is it possible to "pin" a raw block?
<slothbag>
i can see this block "ipfs block get QmXWXR8ku6MmxeZxtZPf4dTK3VwesKQJsFdZnNx7NhpN2A"
<slothbag>
but i cant pin it
<lgierth>
no?
<slothbag>
bugger.. so no way to tell ipfs to remember this data.. do not drop from cache
<lgierth>
pin the object that it belongs to
<slothbag>
it doesn't belong to any.. its a raw block I created
<lgierth>
the blockstore is one layer below
<lgierth>
i'm not sure, maybe it should be able to pin blocks
<voxelot>
unless you haven't published in ~24 then the dht forgets your entry
<slothbag>
hmm. .. this example doesn't seem to work
<slothbag>
echo '{ "Data": "abc" }' | ipfs object put
<slothbag>
Error: Argument 'data' is required
<b0at>
maybe it's case-sensitive
<slothbag>
nope, triede both
<achin>
what version of ipfs?
<achin>
that works fine for me
<slothbag>
latest master
<slothbag>
on windows
<achin>
(i'm running 0.4.0-dev)
<achin>
hmm, i'm on linux, but i would not have expected that to matter
pfraze has quit [Remote host closed the connection]
Matoro_ has joined #ipfs
the193rd has quit [Ping timeout: 256 seconds]
the193rd has joined #ipfs
simonv3 has quit [Quit: Connection closed for inactivity]
<slothbag>
yup, works fine on my linux too
<achin>
quoting on windows is stupid and funky. any chance that echo command isn't actually echoing the right thing?
Matoro has quit [Ping timeout: 256 seconds]
<slothbag>
yeah, just playing around with that now.. funky is a polite way to put it :(
pfraze has joined #ipfs
reit has joined #ipfs
pixeljoe has joined #ipfs
kaiza has joined #ipfs
pixeljoe has left #ipfs [#ipfs]
bitseph has joined #ipfs
joshbuddy has joined #ipfs
<deltab>
slothbag: which shell are you using?
HostFat has quit [Quit: Leaving]
bitseph has quit [Quit: Leaving]
simonv3 has joined #ipfs
joshbuddy has quit [Quit: joshbuddy]
joshbuddy has joined #ipfs
voxelot has quit [Remote host closed the connection]
patcon has quit [Ping timeout: 250 seconds]
<slothbag>
ConEmu which wraps a standard command prompt i thought
jhulten_ has quit [Ping timeout: 264 seconds]
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to feat/gx-libp2p: https://git.io/vzFF8
jfis has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<whyrusleeping>
daviddias: that seems normalish
<whyrusleeping>
probably
Tv` has quit [Quit: Connection closed for inactivity]
<daviddias>
whyrusleeping: I managed to load the hash on the gateways, not sure why you and richardlitt can't get it from me
<whyrusleeping>
the pictures?
<whyrusleeping>
i think its just slow
<daviddias>
yep
<whyrusleeping>
i've got most of them
<whyrusleeping>
just taking a while
<daviddias>
aaaa
<whyrusleeping>
i need to do bitswap work
<whyrusleeping>
after everything else on my list gets done
<daviddias>
I thought you said you only managed to get 120MB and then it would hang
<whyrusleeping>
nah
<whyrusleeping>
it kept climbing
jfis has joined #ipfs
<daviddias>
sweet, so it works :D
<whyrusleeping>
yep! :D
<whyrusleeping>
also, the gx install stuff should be a bit faster now
<whyrusleeping>
for good measure, do 'rm -rf $GOPATH/src/gx'
<whyrusleeping>
before starting it again
<whyrusleeping>
the transition from 'works on my machine' to 'works for everyone' is always rough
<daviddias>
but it is a pretty damn good sticker! :P
<daviddias>
(running it again)
<whyrusleeping>
lol
<daviddias>
woa, it is printing nicer messages
<whyrusleeping>
yay
<daviddias>
whyrusleeping: when doing removeRecursive, do you fetch node links from bitswap to make sure you don't miss any node to be removed by missing a link in the chain, or you stop when you don't have any more of the links in your local blocks store?
<daviddias>
wa7son I can ask, they typically hire someone to do that, but probably the deal might have been already closed
<wa7son>
daviddias: let me know - I can send our video guy from Opbeat
<wa7son>
… maybe
randomguy has joined #ipfs
<daviddias>
wa7son pinging them as we speak :)
r04r is now known as zz_r04r
drnet has joined #ipfs
<daviddias>
woot! wa7son just heard back
<daviddias>
they not only haven't closed the deal yet, but also were looking for a new partner because they haven't had much luck with the provider on the last editions
<computerfreak>
any idea hoe to replace couchDB with ipfs? for Hoodie ?
consentfully has joined #ipfs
chriscool has quit [Quit: Leaving.]
zz_r04r is now known as r04r
consentfully has quit [Remote host closed the connection]
<wa7son>
daviddias: cool - I’ll just check that he’s free on those dates before I promise too much
<daviddias>
ok :)
hashcore has quit [Ping timeout: 240 seconds]
<richardlitt>
haha
<richardlitt>
wa7son: I was so confused for a few seconds there, thought I was in arctifjs
<richardlitt>
New edition of the IPFS weekly! Can you please take a look and let me know if it sums up the past two weeks?
<richardlitt>
daviddias: maybe that's for this last week, not the two weeks before?
<richardlitt>
As in, nothing done in the past four days
<richardlitt>
I know, my dilatory writing of this has resulted in lots of confusion. :(
IlanGodik has joined #ipfs
<daviddias>
richardlitt: oh I see, so will we ship another one before next week?
<daviddias>
or merge 2?
<richardlitt>
This is for anything before this monday.
<richardlitt>
We'll ship one for This Week next Wednesday or Tuesday.
gaboose has joined #ipfs
<richardlitt>
I was supposed to write and send this on Tuesday or Wednesday, but travel kept me from the intensive amount of time I needed to prepare for this
<richardlitt>
The weekly is a LOT of work. :(
<achin>
i wonder how we can fix that
<richardlitt>
Given the systems we have right now in place for it.
<richardlitt>
achin: I've got a few ideas.
<richardlitt>
I just want to publish this one before moving on to that.
<richardlitt>
This is mostly set-up work at the moment. It'll be easier in the futuree.
<richardlitt>
" IPFS Never Forgives, IPFS Never Forgets" - best quote.
<achin>
IPFS Is Forever
<daviddias>
richardlitt: I understand, thank you for keeping doing it though <3
<richardlitt>
daviddias: trying. Today I read 1500 lines of IRC logs before I gave up in despair
ulrichard has quit [Quit: Ex-Chat]
ashark has joined #ipfs
jaboja has quit [Ping timeout: 272 seconds]
<dignifiedquire>
daviddias: how is the js-ipfs hacking coming along?
<daviddias>
doing some progress, got the Data Importing creating the MerkleDAG layout for a file (that required chunking) and accessed to it's DAG objects succesfuly through go-ipfs
<dignifiedquire>
<daviddias>
now working on ipfs-merkle-dag and ipfs-blocks to complete the dag-service and make more exhaustive tests/make the API nicer
<dignifiedquire>
I got my first api request handled through redux + redux-saga yesterday
<daviddias>
I noticed that the encoding of the data for the same file was being stored differently
<daviddias>
figuring out way by creating stronger foudnations
<daviddias>
dignifiedquire: oh nice! do you like it?
<dignifiedquire>
it's definitely the best "flux" architecture I've seen so far
<dignifiedquire>
still quite a lot of overhead and a steep learining curve though
<dignifiedquire>
but I think the best choice for now as long as we are using react
<ipfsbot>
[go-ipfs] whyrusleeping closed pull request #1776: fix http output to never have both transfer-encoding and content-length (master...fix/cat-http-output) https://git.io/vczXs
<daviddias>
richardlitt hmm :) well, having worked remotely my whole life and also having had the chance to manage teams, I learned how it is important to write things down and how the life of someone that is working with us gets so much easier if they have a way to peak into our status, so I just do that naturally now
<ipfsbot>
[go-ipfs] whyrusleeping closed pull request #1864: Add systemd unit to misc/ (master...systemd) https://git.io/vWfri
<dignifiedquire>
richardlitt: daviddias I do the same thing, just on a private document where I note what I worked on every day to keep track for myself and for reports
<voxelot>
daviddias: is it odd that running 'jsipfs version' shows 0.3.11-dev when my go-ipfs repo version is 0.4.0-dev
<daviddias>
I use wunderlist to keep track of my TODOs
Matoro_ has joined #ipfs
<daviddias>
if something seems like a reasonable action item, I try to push it somewhere
<voxelot>
daviddias: ohh wait, maybe it's odd that updating to v0.4.0-dev didn't update my config version that jsipfs reads from =/
<voxelot>
yet 'ipfs version' still shows 0.4.0-dev
<daviddias>
there is a repo migrations thing
<voxelot>
yeah i didn't properly migrate
<richardlitt>
daviddias: I was half joking, I think that's awesome and agree. I do the same thing that dignifiedquire does
<whyrusleeping>
Kubuxu: blech, i was hoping i wouldnt see that one again
<whyrusleeping>
its nothing to do with your code
<Kubuxu>
I know, wanted to let you know but you seem to have met it before, repushed so tests run again.
jaboja64 has quit [Remote host closed the connection]
mildred has quit [Ping timeout: 256 seconds]
<ansuz>
ohai whyrusleeping
<whyrusleeping>
ansuz: hello again
<ansuz>
are you back in the land of freedom again?
<whyrusleeping>
MERICA
<whyrusleeping>
I FEEL SO FREE RIGHT NOW
Kane`_ has joined #ipfs
jhulten_ has quit [Ping timeout: 240 seconds]
Kane` has quit [Ping timeout: 264 seconds]
<ansuz>
I can hook you up to my node in New Jersey
<ansuz>
that's the closest I have
nicolagreco has joined #ipfs
<ansuz>
PMing
<whyrusleeping>
sweet, thanks
edubai__ has quit [Quit: Connection closed for inactivity]
mildred has joined #ipfs
shea256 has joined #ipfs
<noffle>
whyrusleeping: o/
<whyrusleeping>
noffle: 'eyo
<whyrusleeping>
sup?
Matoro has quit [Ping timeout: 250 seconds]
disgusting_wall has joined #ipfs
<richardlitt>
morning.
<noffle>
whyrusleeping: hey! I'd like to have an IpfsNode, in a particular case, do a purely offline Resolve. kind of like how IpfsNode.SetupOfflineRouting, but non-mutating. it'd be nice to do one-time offline resolves, or maybe if the IpfsNode itself had online and offline resolvers/routers built in? seeking advice on how to approach this
<noffle>
generalized problem: sometimes you want to use the network to resolve a path, other times you want to use your local repo
<noffle>
richardlitt: \o
<whyrusleeping>
core.NewNode(ctx, nil)
<richardlitt>
<3 all around
<whyrusleeping>
ah but i see
<whyrusleeping>
my general tactic has been to construct a version of the needed part without the dht or exchange service
Matoro has joined #ipfs
<noffle>
so, as a new node?
<AdrianG>
whats the incentive to store content?
<whyrusleeping>
well, yeah. but what are you trying to do?
<whyrusleeping>
maybe we add an 'offline dagservice' to the IpfsNode?
<whyrusleeping>
and using the `--local` flag will use that one instead of the networked one?
<noffle>
whyrusleeping: I think the best shim for the 'ipfs mount tries to do a dht lookup for its own pubkey but often fails because ipfs init was just run' is to fallback to doing an offline resolve of its pubkey dagnode
<whyrusleeping>
or maybe just defer the actual resolving until we need it
<whyrusleeping>
lazy load it
<noffle>
not sure if that'll help the flakiness
<whyrusleeping>
hrm... yeah
<whyrusleeping>
because we still need to resolve it at some point
<noffle>
might make fail % lower
<whyrusleeping>
eh, thats just hiding the issue then
<whyrusleeping>
don't want that
<noffle>
I suppose 'ipfs mount' should also fail if you've never 'ipfs name publish'd too
shea256 has quit []
<whyrusleeping>
at least the ipns portion
<whyrusleeping>
but
<whyrusleeping>
no
<whyrusleeping>
thats wrong
<whyrusleeping>
we set an empty directory at first
<richardlitt>
Is it wrong that I have this bookmarked as an IPFS example
<lidel>
richardlitt, not sure if there is more to it :)
<richardlitt>
lidel: interesting. What does that mean for me, as a new person?
<richardlitt>
Why is an http2ipfs gateway important?
pfraze has quit [Remote host closed the connection]
<lidel>
richardlitt, it enables you to load content via IPFS instead of HTTP, for example if you open a resource from a public gateway (eg. https://ipfs.io/ipfs/QmTAsnXoWmLZQEpvyZscrReFzqxP3pvULfGVgpJuayrp1w) and have addon enabled, the request will not hit public gateway, but data will be loaded from IPFS swarm, in a decentralized and distributed way
devbug has quit [Ping timeout: 256 seconds]
<dignifiedquire>
richardlitt: do you mean the part about dist?
<richardlitt>
Does that make it faster?
<richardlitt>
dignifiedquire: yeah. I need to describe it better. Maybe I should take some stuff from my blog post draft?
<richardlitt>
lgierth: you still awake?
votz has left #ipfs ["Leaving"]
m0ns00nfup has quit [Quit: undefined]
<lidel>
richardlitt, depends. may be slower if content was already in the gateway. may be faster if it was not, and you fetch from fast peer close to you. generally there is a lot to improve performance-wise in go-ipfs, but we are getting there.
<dignifiedquire>
richardlitt: yeah I suggest doing short version of the blog post
<richardlitt>
lidel: Hmm. What other benefits are there, then?
<richardlitt>
dignifiedquire: Ok. I'll write something now.
<lidel>
richardlitt, support for fs:/ipfs/<hash> in the browser
mildred has quit [Ping timeout: 276 seconds]
<lidel>
richardlitt, as of now, the addon just redirects to local gateway and that is all. in future version we will have some diagnostic information about running daemon (number of peers etc) and DNSlink support.
<richardlitt>
lidel: Cool! Ok, I think I get it all now
<richardlitt>
That's really awesome.
<lidel>
richardlitt, long term goal is to bring IPFS to the browser, perhaps at some point we could run javascript-based node within the addon itself, but for now we use gateway from go-ipfs :)
<richardlitt>
:)
<richardlitt>
lidel: awesome. I'm trying to write this up for the Weekly roundup
<richardlitt>
How does this sound: * [**ifps-firefox-addon**](//github.com/lidel/ipfs-firefox-addon/releases/tag/v1.4.2) The Firefox addon that provides transparent access to IPFS resources via local HTTP2IPFS gateway has been updated to release 1.4.2. This is the first time we've mentioned this in this newsletter; go check it out. The addon enables you to load content
<richardlitt>
via IPFS instead of HTTP. For example, if you open a resource from a public gateway (eg. https://ipfs.io/ipfs/QmTAsnXoWmLZQEpvyZscrReFzqxP3pvULfGVgpJuayrp1w) and have addon enabled, the request will not hit public gateway, but data will be loaded from IPFS swarm, in a decentralized and distributed way. It also now supports the IPFS protocol scheme, meaning
<richardlitt>
that can look up a hash by going to `fs:/ipfs/<hash>` directly.