<Kubuxu>
whyrusleeping: one more comment, cid raw format is 0x73. Should it be same as in bin multicodec 0x55?
<Kubuxu>
or should we register 0x73
<mib_kd743naq>
to rephrase: while having the leaves ( or in fact all nodes aside from the root ) in cid format is the way forward, today I still have to have the file "root" node be an old style unixfs-ish mdag added via old style `block put`
<whyrusleeping>
its 0x73? i thought it was at 0x55...
jedahan has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<mib_kd743naq>
I would expect adding one of those variants, the other to "auto-hash and auto-provide" itself
<mib_kd743naq>
whyrusleeping: could you explain why this is undesirable?
jedahan has joined #ipfs
<whyrusleeping>
mib_kd743naq: eh... i was initially against it
jedahan has quit [Read error: Connection reset by peer]
<whyrusleeping>
but we're pushing for everything to be self describing
jedahan has joined #ipfs
<whyrusleeping>
and the usecases for dedupe at that level are very few
<whyrusleeping>
ask jbenet and daviddias for more elaboration
<mib_kd743naq>
right... ok given that ipfs.io/ipfs/z<v1><pbuf><mhash of pbuf> works I guess I will move all my tooling to `block put -f {raw,protobuf}` and move forward with that
<mib_kd743naq>
^^ that should keep working for a while right? :)
<mib_kd743naq>
well... aside from \x72 changing to \x55 from what I understand...
<mib_kd743naq>
for raw leaves
<whyrusleeping>
Yeah, i'm changing that now
<whyrusleeping>
good catch Kubuxu <3
<mib_kd743naq>
does this mean any blocks already on the network that have links to raw leaves via \x72 will stop working? ;)
<Kubuxu>
yes
<Kubuxu>
sorry we are unable to provide stability guarantees for non even pre RC versions.
<mib_kd743naq>
given it never shipped as a release it's fair game I suppose
<mib_kd743naq>
so I guess my strategy here will be a regular balanced tree, starting with the leaves, and stuffing them recursively into blocks switching to the next block once size is about to cross 1m
mib_kd743naq has quit [Ping timeout: 260 seconds]
zbyte64 has joined #ipfs
zbyte64 has quit [Ping timeout: 258 seconds]
taaem has joined #ipfs
cwahlers has quit [Ping timeout: 256 seconds]
m3s has quit [Remote host closed the connection]
cwahlers has joined #ipfs
Encrypt has quit [Quit: Quit]
cwahlers_ has joined #ipfs
cwahlers has quit [Ping timeout: 244 seconds]
brandonz has quit [Ping timeout: 260 seconds]
jokoon has joined #ipfs
maxlath has joined #ipfs
PseudoNoob has quit [Remote host closed the connection]
bastianilso__ has joined #ipfs
dignifiedquire has quit [Quit: Connection closed for inactivity]
bastianilso__ has quit [Quit: bastianilso__]
bastianilso__ has joined #ipfs
bastianilso__ has quit [Ping timeout: 240 seconds]
<Mateon1>
Is there any documentation on what the trickle-dag generator is?
jdloft has joined #ipfs
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
Boomerang has joined #ipfs
pfrazee has joined #ipfs
pfrazee_ has joined #ipfs
pfrazee has quit [Read error: Connection reset by peer]
corvinux has joined #ipfs
pfrazee_ has quit [Read error: Connection reset by peer]
pfrazee has joined #ipfs
Qwertie has quit [Ping timeout: 256 seconds]
anonymuse has joined #ipfs
maxlath has quit [Ping timeout: 268 seconds]
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
pfrazee has quit [Ping timeout: 244 seconds]
pfrazee has joined #ipfs
bastianilso__ has joined #ipfs
maxlath has joined #ipfs
PseudoNoob has joined #ipfs
bastianilso__ has quit [Quit: bastianilso__]
bastianilso__ has joined #ipfs
jokoon has joined #ipfs
<Kubuxu>
mib_kd743naq: there is some other limitation in bitswap iirc
<Kubuxu>
Mateon1: I don't think so, it would be great to have one.
BeautifulBash has joined #ipfs
<BeautifulBash>
Is it possible to use Jekyl over ipfs?
<BeautifulBash>
Jekyll*
bastianilso__ has quit [Quit: bastianilso__]
jedahan has joined #ipfs
jedahan has quit [Client Quit]
ianopolous has joined #ipfs
dignifiedquire has joined #ipfs
maxlath has quit [Ping timeout: 258 seconds]
seg_violation has quit [Ping timeout: 260 seconds]
seg_violation has joined #ipfs
<graffen>
BeautifulBash: Yup, https://graffen.dk is a simple jekyll site and published to IPFS at /ipns/graffen.dk
<BeautifulBash>
graffen, holy shit the future is here!! :D
<BeautifulBash>
Moving forward, do you think we will see no need for having a node be in control over our IPFS websites?
<graffen>
BeautifulBash: It is :) I've simply set up a small web server to proxy to the ipfs node running on the same machine and restrict queries to /ipns/graffen.dk/*. It's really simple
<graffen>
BeautifulBash: Well as soon as we get web server plugins that support IPFS directly there won't be need to run a separate node. The web server itself will just be a node I guess.
mib_kd743naq has joined #ipfs
<BeautifulBash>
graffen, this is such a big step forward for decentralisation... The only thing that could improve upon this is for everyone being able to access .bit domains without any tampering
<mib_kd743naq>
Kubuxu: wait... you mean even 1MB blocks ( total size, with all recordkeeping ) are too big?
<BeautifulBash>
that would be so perfect. No hosting, no information sensitive domain registration...
<Kubuxu>
mib_kd743naq: no, 1MiB is ok, but it isn't hard limit
<Kubuxu>
but if you would have to either: use `ipfs object patch add-link ...` or implement HAMT
<mib_kd743naq>
hm... I read through the entire ticket and browsed some of the code + HAMT on wikipedia
<mib_kd743naq>
I am failing to understand the most basic use case here - what does a "sharded unixfs directory" actually mean..?
<deltab>
if you have a directory with many thousands, maybe even millions of entries, it'll be slow to access and inefficient to update
jokoon has quit [Quit: Leaving]
<deltab>
breaking it into smaller shards improves that
<deltab>
still conceptually one directory, but stored in multiple objects
<deltab>
that's my understanding of it anyway
<Kubuxu>
yup
<mib_kd743naq>
ahhhhh basically to be able to (handwavy term) fuse multiple unixfs directory objects into a single set of links presented in order
<mib_kd743naq>
obvious in hindsight, thanks for the hint
TheWhisper_ has joined #ipfs
atrapado_ has joined #ipfs
anonymuse has quit [Remote host closed the connection]
anewuser has joined #ipfs
anonymuse has joined #ipfs
eternalfrost has joined #ipfs
pfrazee has quit [Remote host closed the connection]
maxlath has joined #ipfs
oleavr has quit [Ping timeout: 250 seconds]
lachenmayer has quit [Ping timeout: 250 seconds]
aristid has quit [Ping timeout: 250 seconds]
barnacs has quit [Ping timeout: 250 seconds]
xa0 has quit [Ping timeout: 250 seconds]
rawtaz has quit [Ping timeout: 250 seconds]
barnacs has joined #ipfs
strixy has quit [Ping timeout: 250 seconds]
patagonicus has quit [Ping timeout: 250 seconds]
oleavr has joined #ipfs
rawtaz has joined #ipfs
xa0 has joined #ipfs
xa0 has joined #ipfs
xa0 has quit [Changing host]
patagonicus has joined #ipfs
aristid has joined #ipfs
strixy has joined #ipfs
lachenmayer has joined #ipfs
<mib_kd743naq>
is there a milestone indicator somewhere on github for the release of ipfs with raw leaf support?
<mib_kd743naq>
( too many repos, I got kinda lost )
<lgierth>
mib_kd743naq: if there is one it's in the go-ipfs repo
TheWhisper_ has quit [Quit: Leaving]
captain_morgan has quit [Remote host closed the connection]
zbyte64 has joined #ipfs
<eternalfrost>
I'm having an issue with the python ipfs implementation, py-ipfs-api.
<eternalfrost>
The recursice directory add function does not add the directory structure, just the leaves
<eternalfrost>
For the example on Git, api.add('fake_dir', recursive=True) posts any files in the folder structure but the references between them are not included
<eternalfrost>
This can be confirmed with the refs command.
<eternalfrost>
The same call in the default go-ifps implementation works correctly.
kerozene has quit [Quit: brb]
zbyte64 has quit [Ping timeout: 248 seconds]
kerozene has joined #ipfs
Encrypt has quit [Quit: Quit]
anonymuse has quit [Remote host closed the connection]
Boomerang has quit [Remote host closed the connection]
pfrazee has joined #ipfs
dignifiedquire has quit [Quit: Connection closed for inactivity]
hpk has quit [Ping timeout: 260 seconds]
espadrine has quit [Ping timeout: 260 seconds]
hpk has joined #ipfs
BeautifulBash has quit [Remote host closed the connection]
ylp has quit [Ping timeout: 260 seconds]
ylp has joined #ipfs
zbyte64 has joined #ipfs
zbyte64 has quit [Ping timeout: 265 seconds]
<lgierth>
haad daviddias: how do you organize repos locally, is it $org/$repo or just flat? i'm wondering if the myrepos script can be applied to js-land too
<daviddias>
lgierth: I've it all flat
<lgierth>
ok -- in go-land we have the $org/$repo structure kinda dictated by how imports work
<lgierth>
how do you do operations on a group of repos?
<lgierth>
e.g. git pull all of it
corvinux has quit [Quit: Leaving]
<lgierth>
or update a certain dependency in all of them and bubbling that up through the dep tree
<lgierth>
i started ipfs/devtools and wanna collect tools and practices there
dignifiedquire has joined #ipfs
maxlath has quit [Quit: maxlath]
maxlath has joined #ipfs
corvinux has joined #ipfs
ygrek has joined #ipfs
corvinux has quit [Quit: Leaving]
Encrypt has joined #ipfs
ribasushi has quit [Ping timeout: 245 seconds]
bastianilso__ has joined #ipfs
<Mateon1>
lgierth: Every time I fix something, I clone the repo, and delete after fixing the thing. I only really work on small issues. That reminds me that I need to work a bit on orbit on Windows, still, but last I tried `npm install`ing it, it still depended on an unfixed version of go-ipfs-dep.
<lgierth>
yeah i was more thinking of the work that the maintainers usually have to do
<Mateon1>
I wish there was a way to show the dependency tree of a non-installed package in npm
<lgierth>
there probably is using the npm api
ribasushi has joined #ipfs
ulrichard has joined #ipfs
ylp has quit [Ping timeout: 260 seconds]
anonymuse has joined #ipfs
ylp has joined #ipfs
ivo_ has quit [Remote host closed the connection]
anonymuse has quit [Ping timeout: 256 seconds]
jholden has joined #ipfs
robattila256 has quit [Ping timeout: 265 seconds]
ulrichard has quit [Quit: Ex-Chat]
<pjz>
okay
zbyte64 has joined #ipfs
<pjz>
anyone awake?
ion has quit [Quit: rebooting server]
maxlath has quit [Ping timeout: 265 seconds]
fil_redpill has joined #ipfs
<fil_redpill>
oh yes that's going to be a huge win
soloojos has quit [Ping timeout: 245 seconds]
ion has joined #ipfs
ion has quit [Client Quit]
ion has joined #ipfs
taaem has quit [Read error: Connection reset by peer]
ivo_ has joined #ipfs
taaem has joined #ipfs
<pjz>
?
<pjz>
what is?
ribasushi has quit [Ping timeout: 260 seconds]
zbyte64 has quit [Remote host closed the connection]
tmg has joined #ipfs
Rspigler has joined #ipfs
zbyte64 has joined #ipfs
zbyte64 has quit [Ping timeout: 244 seconds]
jholden has quit [Ping timeout: 260 seconds]
<Kubuxu>
jbenet: any timebox when you could review pnet?
Encrypt has quit [Quit: Quit]
ivo_ has quit [Quit: Leaving]
zbyte64 has joined #ipfs
mildred2 has joined #ipfs
mildred1 has quit [Ping timeout: 265 seconds]
slothbag has joined #ipfs
captain_morgan has joined #ipfs
PseudoNoob has quit [Remote host closed the connection]
mildred2 has quit [Quit: WeeChat 1.5]
taaem has quit [Read error: Connection reset by peer]
ribasushi has joined #ipfs
taaem has joined #ipfs
Rspigler has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
HastaJun_ has quit [Ping timeout: 260 seconds]
atrapado_ has quit [Quit: Leaving]
bastianilso__ has quit [Ping timeout: 244 seconds]
bastianilso__ has joined #ipfs
taaem has quit [Remote host closed the connection]
taaem has joined #ipfs
taaem has quit [Read error: Connection reset by peer]
HastaJun_ has joined #ipfs
taaem has joined #ipfs
mguentner has quit [Read error: Connection reset by peer]
anewuser has quit [Quit: anewuser]
dbri3 has joined #ipfs
dbri has quit [Read error: Connection reset by peer]
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
pfrazee has quit [Remote host closed the connection]
pfrazee has joined #ipfs
bastianilso__ has quit [Quit: bastianilso__]
maxlath has joined #ipfs
pfrazee has quit [Ping timeout: 248 seconds]
reit has quit [Quit: Leaving]
mguentner has joined #ipfs
<deltab>
hi mguentner
<deltab>
a couple of days ago you were asking about adding fields to protobuf
<deltab>
it's designed to handle that gracefully, so there's no need to repurpose existing fields
anonymuse has joined #ipfs
<mguentner>
deltab: yeah. glad you remember :)
maxlath has quit [Quit: maxlath]
<mguentner>
so filestore stores the files for unixfs. and there will be a new unixfs (possibly ipld) that then uses these files and maps context (the respective filesystem) to them (like ro,x,mime, etc.)
kulelu88 has quit [Quit: Leaving]
ipfs has joined #ipfs
<lgierth>
no filestore has nothing to do with unixfs -- at least not more than any other datastore
<lgierth>
the part about ipld is correct
pfrazee has joined #ipfs
ylp has quit [Ping timeout: 260 seconds]
anonymuse has quit [Read error: Connection reset by peer]