--- Log opened Sat May 02 00:00:10 2020
00:21 -!- jonatack_ [~jon@37.165.194.161] has quit [Quit: jonatack_]
00:23 -!- jonatack [~jon@37.165.194.161] has joined #rust-bitcoin
00:32 -!- kallewoof_ [~quassel@240d:1a:759:6000:a7b1:451a:8874:e1ac] has joined #rust-bitcoin
00:32 -!- kallewoof [~quassel@240d:1a:759:6000:a7b1:451a:8874:e1ac] has quit [Read error: Connection reset by peer]
00:50 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has joined #rust-bitcoin
03:55 -!- jonatack_ [~jon@37.170.239.229] has joined #rust-bitcoin
03:58 -!- jonatack [~jon@37.165.194.161] has quit [Ping timeout: 272 seconds]
04:01 -!- kallewoof_ [~quassel@240d:1a:759:6000:a7b1:451a:8874:e1ac] has left #rust-bitcoin ["http://quassel-irc.org - Chat comfortably. Anywhere."]
04:03 -!- kallewoof [~quassel@240d:1a:759:6000:a7b1:451a:8874:e1ac] has joined #rust-bitcoin
04:18 -!- Dean_Guss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin
04:19 -!- DeanWeen [~dean@gateway/tor-sasl/deanguss] has quit [Ping timeout: 240 seconds]
04:49 -!- jonatack_ [~jon@37.170.239.229] has quit [Ping timeout: 260 seconds]
04:51 -!- jonatack_ [~jon@134.19.179.235] has joined #rust-bitcoin
07:32 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has quit [Read error: Connection reset by peer]
07:33 -!- mauz555_ [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has joined #rust-bitcoin
07:35 -!- mauz555_ [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has quit [Read error: Connection reset by peer]
07:35 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has joined #rust-bitcoin
07:55 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has quit [Remote host closed the connection]
08:48 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has joined #rust-bitcoin
12:39 < BlueMatt> ariard: which irc convo was this in reference to https://github.com/rust-bitcoin/rust-lightning/pull/610#issuecomment-622222052 ?
12:39 < BlueMatt> I dont see a reason to wait on the witness thing, its no material difference in diff?
12:40 < BlueMatt> like, we already have that info right at the callsite, so moving it into the callee is super trivial
12:59 -!- jonatack_ [~jon@134.19.179.235] has quit [Ping timeout: 272 seconds]
13:34 -!- jonatack [~jon@2a01:e0a:53c:a200:bb54:3be5:c3d0:9ce5] has joined #rust-bitcoin
14:11 < ariard> BlueMatt: I'm glad to do it directly in a follow-up PR, but as we miss the MINIMALIF bug during last review round
14:11 < ariard> I would prefer keep diff as it is, instead of silently breaking other stuff
14:11 < ariard> BlueMatt: 05/01, grep "MINIMALIF"
14:13 < ariard> BlueMatt: lightning-sync-block is pretty coool! Do you envision to keep in rust-lightning/ or maybe it should in its own repo like rust-chain-provider/ ?
14:13 < ariard> I mean it can be reused for other stuff that a LN stack right?
14:38 < BlueMatt> ariard: i mean its a -2 +2 LoC diff to add it
14:39 < BlueMatt> I dont think its reasonable to take a PR that has gone through only initial review and suggest that no more changes should happen except in follow-up, at least for code thats new in the pr.
14:40 < BlueMatt> as for lightning-block-sync, I dunno, I haven't really thought about it tooo much. I wanted to get something down on paper first in terms of refactoring out the chainsync stuff from rust-lightning-bitcoinrpc so that it can be reused
14:41 < BlueMatt> I'm happy to pull it outsomewhere else...I'd like to make it more robust - I think for lightning a requirement for a chain sync thing is multiple redundant block sources, as well as some "backup block sources", eg that maybe you never query unless something indicates something bad is afoot
14:42 < BlueMatt> ie maybe you fetch blocks from your local node or an electrum server, and fetch headers from headers-over-dns or headers-over-radio, and if you find that you're missing blocks that you have headers for, you go ask like centralized website for those blocks.
14:42 < BlueMatt> this way you maintain privacy unless something bad happens, at which point you fall back to something non-private.
15:01 < ariard> BlueMatt: I would be okay if it was a -2 +2 but in fact I need to pack more stuff in InputDescriptors
15:02 < ariard> BlueMatt: but the fact is, I reintroduced a bug which was would be easier to spot if diff was smaller, don't feel good about this :/
15:02 < BlueMatt> so write a test :p
15:03 < ariard> BlueMatt: lol it's the test were are pending on core
15:03 < BlueMatt> right....
15:03 < ariard> neeh will add commits on top... will try to make them small
15:03 < BlueMatt> I'm super confused though
15:03 < BlueMatt> https://github.com/rust-bitcoin/rust-lightning/pull/610/files#diff-9564e57e0af8450570741fff759a918bR582-R586
15:03 < BlueMatt> we already have all the data required to build the redeemscript at the callsite
15:03 < BlueMatt> because we're literally building the redeemscript two lines up
15:03 < BlueMatt> so why cant we pass it through super, super trivially
15:04 < BlueMatt> grrr, of course github linking doesnt fucking work
15:04 < ariard> HTLC redeem script you need remote party keys and hash/timelock
15:04 < BlueMatt> channelmonitor.rs:588
15:05 < BlueMatt> same with channelmonitor:608
15:05 < BlueMatt> I dont get it?
15:05 < BlueMatt> for both of sign_justice_transaction and sign_remote_htlc_transaction it should be +2 - 2
15:06 < BlueMatt> even if it isnt for others
15:06 < BlueMatt> which, fine, we can do those later
15:07 < ariard> hmmmm, right now I'm reading again we already have remote base key
15:07 < ariard> and revocation key is here us
15:07 < ariard> well okay maybe not +2/-2 but like ~20 LoC
15:08 < ariard> I was confused I tried to do first then realize I miss more stuff
15:08 < BlueMatt> well, ok, whatever, but it should be super trivial
15:08 < ariard> yeah okay
15:09 < ariard> wrt to lightning-block-sync, yeah definetely multiple source of blocks providers is something you want
15:09 < ariard> it's really a hybrid model, you may even ask headers from watchtowers or something like this
15:09 < BlueMatt> right, but not just multiple sources, but I think we need a 'heirarchy'
15:10 < ariard> right, don't burn you privacy, unless something is really wrong
15:10 < ariard> and you want same for tx broadcast
15:10 < BlueMatt> and, yea, i think the big thing is supporting doing ~something with headers-only sources
15:10 < BlueMatt> yea, i havent gotten far enough to think about tx broadcast yet
15:10 < ariard> Yes, I would favor on the long-term put this in its own repo, to avoid rust-lightning being huge with all batteries
15:11 < ariard> but for now I think it's fine
15:12 < ariard> what we need is a middleware to set reconciliate between headers sources
15:12 < BlueMatt> yea, thats fine, I was also wondering what to do with it and murmel...like, it would be cool to *also* have a p2p spv client etc, but I super dont have time to go dig into murmel or refactor it.
15:12 < BlueMatt> huh?! no, we dont need to do set reconiliation with headers sources.
15:13 < BlueMatt> its a linear set of headers, and we need to find the fork point, this is trivial
15:13 < ariard> urggh sorry wasn't using the good work, but a way to say "I'm seeing 2 tips with more N blocks diff something is wrong"
15:13 < ariard> *word
15:13 < BlueMatt> oh you just mean like a warning system?
15:13 < ariard> yes, a warning system or watchdog
15:14 < BlueMatt> the MicroSPVClient thing in the PR already knows how to pick the best chain, but its also very deliberate in that it does not store anything in normal operation
15:14 < BlueMatt> so no full header trees
15:14 < ariard> I agree with a p2p spv client, specially with a robust addrman manager, like one not super easy to trick
15:15 < BlueMatt> yea, well thats a whole other can of worms
15:15 < BlueMatt> and a huge project
15:15 < ariard> wrt to murmel, I will make a post in the slack saying what it would be nice and how you can ties all pieces toghether
15:15 < BlueMatt> would be cool to drive murmel forward more, and then this would just be a part of murmel (eg 'well i think we're missing headers, lets make more connections/fetch via http') but thats a *ton* of work.
15:16 < BlueMatt> and I dont know how complete murmel is
15:16 < ariard> I agree, but I'm not willingly to commit on this :p
15:16 < ariard> I've done a quick review a while ago, like a year
15:17 < BlueMatt> hah, yea.
15:17 < ariard> BlueMatt: if you're bored this weekend, our paper with gleb is ready for a 1st review round :)
15:17 < ariard> and we speak in depth about LN and light clients, more than in origianl blog post
15:17 < BlueMatt> lol "bored" would not describe this weekend for me
15:18 < BlueMatt> I took two days off to write this spv thing, so now I have to catch up on 10 other things :/
15:18 < ariard> it was a joke, I know
15:18 < ariard> no worries we have other good reviewers, but just for fyi
15:18 < BlueMatt> all good, i figured
15:19 < ariard> ->bbl, have-to-grab-stuff-before-everything-close-at-7pm-because-quarantine
15:19 < BlueMatt> oof. gl.
16:16 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:9c09:7e12:30c0:810] has quit [Remote host closed the connection]
16:19 -!- Dean_Guss [~dean@gateway/tor-sasl/deanguss] has quit [Ping timeout: 240 seconds]
16:20 -!- Dean_Guss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin
--- Log closed Sun May 03 00:00:11 2020