--- Log opened Wed Jul 24 00:00:19 2019
00:28 -!- elichai2 [uid212594@gateway/web/irccloud.com/x-jzwnvoevrxrrshiu] has quit [Quit: Connection closed for inactivity]
01:02 < stevenroose> andytoshi: ACK'ed
01:02 < stevenroose> andytoshi: really ACK'ed
01:39 < stevenroose> andytoshi: I mentioned on Rocket: in elements you can't use bitcoin::PrivateKey because it has the network attached (we could perhaps fix his with bitcoin_constants), so I prefer to be consistent and use both secp type over using secp::PrivKey and bitcoin::PubKey
03:08 -!- dpc [dpcmatrixo@gateway/shell/matrix.org/x-gpvkxclfnybxbuns] has quit [Ping timeout: 252 seconds]
03:48 -!- dpc [dpcmatrixo@gateway/shell/matrix.org/x-gtkifijnsrxaxppo] has joined #rust-bitcoin
04:09 -!- TamasBlummer [~Thunderbi@p200300DD67126447B9BF97D5482085BB.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
04:10 -!- TamasBlummer [~Thunderbi@p200300DD67126447B9BF97D5482085BB.dip0.t-ipconnect.de] has joined #rust-bitcoin
05:40 -!- DeanWeen [~dean@gateway/tor-sasl/deanguss] has quit [Ping timeout: 260 seconds]
06:34 -!- ghost43_ is now known as ghost43
06:56 -!- elichai2 [uid212594@gateway/web/irccloud.com/x-neblnlgotdayqnjg] has joined #rust-bitcoin
07:12 < stevenroose> andytoshi: https://github.com/rust-bitcoin/rust-bitcoin/pull/265 has approvals !!
07:55 -!- michaelsdunn1 [~michaelsd@unaffiliated/michaelsdunn1] has joined #rust-bitcoin
08:45 < BlueMatt> ariard: wait huh?
08:45 < BlueMatt> ariard: have you looked at the progress in https://github.com/rust-bitcoin/rust-lightning/pull/244 ?
08:46 < BlueMatt> I thought that approach was reasonable, though the pr still had bugs
08:46 < BlueMatt> and wasnt too complex
09:45 -!- belcher [~belcher@unaffiliated/belcher] has quit [Read error: Connection reset by peer]
09:46 < andytoshi> stevenroose: ok well secp256k1::PublicKey does not round-trip so it can't be used in a blockchain context
09:46 < andytoshi> you can add a new elements::PublicKey if you'd like (which may make sense anyway, given CT) ... though you'll then also need to add an elements::Script and elements::script::Builder (which we should also do anyway) to get .push_key back..
09:46 -!- belcher [~belcher@unaffiliated/belcher] has joined #rust-bitcoin
10:41 < TamasBlummer> https://github.com/rust-bitcoin/rust-bitcoin/pull/295 should be trivial to approve
10:42 < TamasBlummer> https://github.com/rust-bitcoin/rust-bitcoin/pull/281 is adding BIP158 support, touches no other functionality passes all tests in Bitcoin Core, needs second approval.
10:42 < TamasBlummer> thank you
10:47 < andytoshi> nice, i'm impressed we compile with forbid(unsafe)
10:48 < TamasBlummer> should I squash https://github.com/rust-bitcoin/rust-bitcoin/pull/281 or squash at merge into master
10:48 < TamasBlummer> ?
10:50 < andytoshi> usually best to squash before merging, if you intend to squash
10:52 < TamasBlummer> there is one approval on this already, I think I lose it if force push, right?
10:53 < andytoshi> probably :P
10:53 < andytoshi> we can just merge without squashing at all
10:53 < TamasBlummer> I see I lost it already.
10:53 < andytoshi> oh ok
10:53 < andytoshi> then go for it
10:56 < TamasBlummer> squashed
11:10 < ariard> BlueMatt: yes IIRC it doesn't save the my_current_commitment_point from the other guy to get back our to_remote output on remote commitment tx
11:10 < ariard> so we need to pass the point to ChannelMonitor AND raise an error for staling the channel
11:12 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 248 seconds]
11:15 < BlueMatt> ariard: hmmm maybe I'm misremembering, but cant you just pass it into the channe, and if the channel needs an update it can pass it to its internal ChannelMonitor and then update the state by returning it?
11:17 -!- belcher [~belcher@unaffiliated/belcher] has joined #rust-bitcoin
11:27 < ariard> BlueMatt: yes that's where I want to do something clean, I want to "update the state by returning it" and also return an error
11:27 < ariard> so rn I've something like ChannelError::Stale{ err, update }
11:37 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:b9fc:cc49:3c09:54f] has joined #rust-bitcoin
11:41 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:b9fc:cc49:3c09:54f] has quit [Client Quit]
11:41 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:b9fc:cc49:3c09:54f] has joined #rust-bitcoin
11:41 < TamasBlummer> stevenroose: applied your style changes to https://github.com/rust-bitcoin/rust-bitcoin/pull/295, please re-approve
11:42 < BlueMatt> ariard: right, that makes sense
11:43 < TamasBlummer> stevenroose: I meant https://github.com/rust-bitcoin/rust-bitcoin/pull/281
11:45 < TamasBlummer> BlueMatt: https://github.com/rust-bitcoin/rust-lightning/pull/357 should be trivial if you are not worried of some performance hot to get rid of this only unsafe language use
11:45 < TamasBlummer> *hit
11:46 < BlueMatt> TamasBlummer: hmmmm, the perf hit is probably fine, but derive(Default) kinda sucks for things we never want to be default :/
11:47 < TamasBlummer> initializing with 0s as before is the same
11:47 < BlueMatt> right, but it would be nicer to just push_back on each loop iteration
11:47 < BlueMatt> oh, its reverse hmmmm
11:48 < BlueMatt> I'm not toooo worried about perf there (its only us sending funds, iirc)
11:48 < BlueMatt> so could also just insert at front on each loop iteration
11:48 < BlueMatt> I think maybe that over ::default()?
11:48 < TamasBlummer> default is only used in allocating the vector
11:49 < BlueMatt> right, but memmove() 5 times isnt so expensive either
11:49 < BlueMatt> and that makes for even less chance of it being busted
11:49 < BlueMatt> also, its during send, so the next step is encryption, a few memmoves isnt gonna hurt us
11:51 < TamasBlummer> one might optimize later if that really matters. getting rid of unsafe is the goal here
11:52 < BlueMatt> right, the above suggestion is a de-optimization to make it even cleaner
11:52 < BlueMatt> but otherwise ackackackackack
11:52 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:b9fc:cc49:3c09:54f] has quit []
12:02 < TamasBlummer> do not know how to implment your comment on th ePR
12:06 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:b9fc:cc49:3c09:54f] has joined #rust-bitcoin
12:10 < BlueMatt> TamasBlummer: ah ok, nbd, will do in 1 sec
12:12 < BlueMatt> TamasBlummer: https://github.com/rust-bitcoin/rust-lightning/pull/358 look good to you?
12:33 < BlueMatt> ariard: lmk when you have something else you want me to review, gonna take a look at 355 now
12:34 < elichai2> I must say trying to write binary trees that point back to the parents isn't fun in rust
12:35 < andytoshi> just use rawptrs
12:36 < elichai2> I moved to do it in c++ with unique pointers for the kids and raw ones for the parents heh
12:36 < elichai2> But it should be somehow possible with RefCells maybe
12:37 -!- CubicEarth [~CubicEart@c-67-168-1-172.hsd1.wa.comcast.net] has quit [Ping timeout: 246 seconds]
12:39 < andytoshi> lol, you moved to C++ because the language wouldn't enforce the invariants of its types on you?
12:39 < andytoshi> you can do it with Rc and Refcell but it'll be complicated. just use rawptrs
12:42 < elichai2> no haha I moved to C++ because if I need to do it with raw pointers anyway i'd rather learn more c++ at the same time :) and make it compatible with bitcoin core's code
12:44 < andytoshi> heh ok, i won't dissuade you
12:44 < andytoshi> but tbh i have zero interest in learning C++
12:44 < elichai2> andytoshi: what's the next step in the GMP PR? figuring out the windows build system? or we'll just skip it for a while? (if anyone here knows compilation flags for windows and how they work would be glad for some help)
12:45 < andytoshi> i dunno, i think we're having trouble in rust-bitcoin as well with windows?
12:45 < andytoshi> like, is this somethnig travis needs t ofix?
12:45 < elichai2> oh with the bitcoinconsensus?
12:45 < elichai2> or with the failing travis?
12:45 < andytoshi> with failing travis
12:45 < andytoshi> maybe i'm thinking of bitcoinconsensus
12:46 < elichai2> I would try and restart this build: https://travis-ci.org/rust-bitcoin/rust-secp256k1/builds/562661003 maybe it's fixed
12:46 < BlueMatt> TamasBlummer: gonna take 358 (which still has your name on the commit, cause most of it was still you), unless you object
12:47 < BlueMatt> ariard: when you get a chance, I think I finally worked out the kinks in https://github.com/rust-bitcoin/rust-lightning/pull/356 so it should be good now
12:51 < andytoshi> update bitcoin_hashes: https://github.com/rust-bitcoin/rust-bitcoin/pull/296
12:51 < BlueMatt> oops, I had missed your comment TamasBlummer. oh well.
12:51 < andytoshi> TamasBlummer: can you change your PR either to incorporate 296 or to remove the bitcoin_hashes update?
12:51 < BlueMatt> andytoshi: thats...a big jump?
12:52 < andytoshi> BlueMatt: yes, rust-bitcoin has been blocked for a long time on 295
12:52 < BlueMatt> ohoh, right, yes
12:52 < andytoshi> err 265
12:52 < andytoshi> 289 and 293 also need acks
12:53 < andytoshi> oh 289 needs to be redone pretty much
12:53 < andytoshi> stevenroose: ^
13:21 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:b9fc:cc49:3c09:54f] has quit []
13:24 < elichai2> I see that rerunning the travis worked. that's great :)
14:18 < ariard> BlueMatt: looks great but don't burnt that much cpu on it!
14:18 < BlueMatt> hmm?
14:18 < BlueMatt> oh you mean 356, yea, I'm burning it in right now
14:18 < BlueMatt> finally been running for a while and no crashes :)
14:35 -!- michaelsdunn1 [~michaelsd@unaffiliated/michaelsdunn1] has quit [Remote host closed the connection]
16:04 < real_or_random> BlueMatt: can you have a brief look at https://github.com/rust-bitcoin/rust-secp256k1/pull/131 ? it does look good for us but it would be good to hear your opinion on this given it's about the cc crate
16:17 -!- michaelsdunn1 [~michaelsd@unaffiliated/michaelsdunn1] has joined #rust-bitcoin
16:47 -!- michaelsdunn1 [~michaelsd@unaffiliated/michaelsdunn1] has quit [Remote host closed the connection]
18:32 -!- TamasBlummer1 [~Thunderbi@p200300DD6712644089830B1D8358BB90.dip0.t-ipconnect.de] has joined #rust-bitcoin
18:34 -!- TamasBlummer [~Thunderbi@p200300DD67126447B9BF97D5482085BB.dip0.t-ipconnect.de] has quit [Ping timeout: 264 seconds]
18:34 -!- TamasBlummer1 is now known as TamasBlummer
18:45 -!- DeanWeen [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin
19:25 -!- elichai2 [uid212594@gateway/web/irccloud.com/x-neblnlgotdayqnjg] has quit [Quit: Connection closed for inactivity]
20:01 < TamasBlummer> andytoshi: incorporated 296 into 281, please review
20:02 < TamasBlummer> stevenroose: ^
20:21 -!- DeanWeen [~dean@gateway/tor-sasl/deanguss] has quit [Remote host closed the connection]
20:53 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin
23:23 < ariard> BlueMatt: okay data loss protect should work (#349)
23:23 < ariard> but not sure about the case if Bob send us bad per_commitment_secret, we can't now if we are fallen-behind so best would be also to stale the channel?
--- Log closed Thu Jul 25 00:00:20 2019