<arthuredelstein>
I'm wondering if a workaround might be to do everything through blob: URIs.
<lgierth>
arthuredelstein: i wonder if you can even put 1 MB in a blob URI :)
<lgierth>
you also can't stream through blob URIs
<lgierth>
and there's no verification of the hashes
WindPower has quit [Ping timeout: 265 seconds]
kaotisk has quit [Ping timeout: 240 seconds]
<arthuredelstein>
What I mean is, the extension would download ipfs: resources and pass them to content page via blob URIs.
shizy has quit [Ping timeout: 265 seconds]
<arthuredelstein>
Use a content script to pass ipfs URIs to the extension and then receive blob resource in return
<arthuredelstein>
*resources
<arthuredelstein>
the extension would still verify hashes etc
<lgierth>
ah i see
<lgierth>
it would probably work, but the size/streaming problem remains
<lgierth>
so it's unfortunately just a hack
<lgierth>
right now we capture custom search engine queries that start in ipfs:// and redirect them to the http gateway
<arthuredelstein>
I don't think there is such a size limit for blobs. At least I think they can be pretty big.
<lgierth>
just need gigabytes of ram eh? :D
<arthuredelstein>
Why? You just need the size of the file you have downloaded, which you do anyway in http: browsing :)
<arthuredelstein>
a blob is just an opaque reference to some file-like object
<lgierth>
i think i confuse what you're saying - i was thinking you'd just put the raw content (base64-shielded) as the blob URI's content
<arthuredelstein>
that sounds more like a data URI to me. I'm thinking of a blob URI which is just a hash
<lgierth>
ah, yeah i figure that could work e.g. for <img> tags in http pages, given you add a content script?
<arthuredelstein>
I think so. You could have a content script traverse the dom, send ipfs: URIs to the extension, get blob URIs back an inject these back into the DOM by replacing the ipfs URIs.
<lgierth>
hackityhack
<lgierth>
we wanna be able to have completely http://-less websites though
<arthuredelstein>
Of course.
<lgierth>
so one way or the other we'll need proper programmable protocol handlers, because everything else is just a hack with varying degrees of limitations
kaotisk has joined #ipfs
ygrek has joined #ipfs
k33k has left #ipfs ["leaving"]
k33k has joined #ipfs
<walle303>
I'm generating a search index for the more up-to-date english wikipedia now, but here's the current version that I just managed to get up ipfs/QmTg2VcPJgLMdLFwTNQovpPQqcU2mKXKXu6mKNPvfQnqc5
<walle303>
The IPNS link at the bottom doesn't work yet because it's not ready yet, don't pin anything yet
k33k has quit [Quit: leaving]
<walle303>
If something is broken you should probably tell me
k33k has joined #ipfs
<arthuredelstein>
lgierth: You're probably right about that. I just thought it would be fun to have something working in chrome/firefox even before they implement proper programmable protocol handlers
<lgierth>
hehe agreed, it is :)
<arthuredelstein>
I wonder if there are open bugs for firefox/chrome to implement custom protocol handlers in the webextensions api?
<arthuredelstein>
I see some things but it's not clear what the status is.
<lgierth>
yes in fact, one sec
<lgierth>
there's no progress apart from open issues
<lgierth>
somebody needs to go and implement a webextension api with their experiments toolkit
<lgierth>
and then it'll take a few interations and months
<Poeticode>
or if there's any static site generator + cms you guys know for sure works with ipfs. I'm just now starting on a site, and figuring out what technologies I want to use
Encrypt has quit [Quit: Quit]
fredthomsen has quit [Quit: Leaving.]
fredthomsen has joined #ipfs
<lgierth>
Poeticode: we use hugo for the project websites
<lgierth>
Poeticode: anything that can produce relative links should work
<Poeticode>
thanks lgierth, I get a lot of search results with Hugo + IPFS :D gonna give it a whirl
discopatrick has quit [Quit: Connection closed for inactivity]
Fess has quit [Ping timeout: 265 seconds]
newhouse has joined #ipfs
fleeky has quit [Read error: Connection reset by peer]
fleeky has joined #ipfs
voker57 has quit [Quit: No Ping reply in 180 seconds.]
voker57 has joined #ipfs
voker57 has joined #ipfs
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
clickjack has joined #ipfs
clickjac_ has quit [Ping timeout: 264 seconds]
fleeky has quit [Ping timeout: 240 seconds]
randomstrangerb has joined #ipfs
randomstrangerb has quit [Remote host closed the connection]
fleeky has joined #ipfs
fredthomsen has quit [Quit: Leaving.]
muravey has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
randomstrangerb has joined #ipfs
Encrypt has joined #ipfs
muravey has joined #ipfs
k333k has joined #ipfs
k33k has quit [Ping timeout: 255 seconds]
k3333k has joined #ipfs
silotis has quit [Remote host closed the connection]
silotis has joined #ipfs
Xiti has quit [Quit: Xiti]
k333k has quit [Ping timeout: 250 seconds]
kaotisk has quit [Ping timeout: 260 seconds]
Xiti has joined #ipfs
kaotisk has joined #ipfs
k3333k has quit [Ping timeout: 265 seconds]
ligi has joined #ipfs
ligi has quit [Changing host]
ligi has joined #ipfs
ilyaigpetrov has quit [Quit: Connection closed for inactivity]
randomstrangerb has quit [Ping timeout: 260 seconds]
randomstrangerb has joined #ipfs
ligi has quit [Ping timeout: 240 seconds]
A124 has quit [Ping timeout: 248 seconds]
<cipres>
DokterBob: thank you for putting ipfs-search.com back up!
<lgierth>
Poeticode: also e.g. github.com/ipfs/website, libp2p/website, ipld/website, ipfs/docs
Ecran has joined #ipfs
zhaba_ has quit [Ping timeout: 240 seconds]
zautomata has joined #ipfs
Encrypt has quit [Quit: Quit]
biodrone has joined #ipfs
zautomata has quit [Ping timeout: 272 seconds]
zautomata has joined #ipfs
rehypothecation has quit [Remote host closed the connection]
<mokos>
DokterBob: Do you rank pages ? Why did you shut it down originally ?
Adbray has joined #ipfs
ligi has joined #ipfs
fredthomsen has joined #ipfs
ligi has quit [Ping timeout: 240 seconds]
M0xf5f5f5[m] has joined #ipfs
fredthomsen has quit [Quit: Leaving.]
chriscool2 has joined #ipfs
muravey has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
randomstrangerb has quit [Ping timeout: 264 seconds]
randomstrangerb has joined #ipfs
<lord|>
is there a good way, for example, to have one node serve 33.3% of a file's blocks, another node serves another 33.3% of the file's blocks, and the last node serves the last 33%
John__ has joined #ipfs
droman has joined #ipfs
<lgierth>
lord|: the ipfs-cluster team are working on it, but not yet
<Powersource>
theeere we go, thanks whoever turned their daemon on
<lgierth>
Powersource: i just transferred it to another host
<lgierth>
not sure what was up with the original source host
<lgierth>
maybe the connection between it and your node was hanging or so
kaotisk has quit [Ping timeout: 248 seconds]
chriscool2 has quit [Quit: Leaving.]
<Powersource>
and now I realize i've already watched this one
nibble has quit [Ping timeout: 255 seconds]
ligi has quit [Ping timeout: 272 seconds]
maxzor_ has joined #ipfs
mrdrdz[m] has joined #ipfs
randomstrangerb has quit [Ping timeout: 252 seconds]
randomstrangerb has joined #ipfs
John__ has quit [Quit: Leaving]
<lgierth>
:')
<Powersource>
rust-ipfs is coming up right, if I've understood correctly? Or does rust-libp2p still need significant work?
Jesin has quit [Quit: Leaving]
Dodgy-Debaser has joined #ipfs
<lgierth>
rust-libp2p is progressing but a complete libp2p implementation is a ton of work
<lgierth>
the concrete plan for now is only rust-libp2p, but rust-ipfs might follow in the future