--- Log opened Fri Jul 26 00:00:22 2019 00:33 < stevenroose> andytoshi: go ahead and merge it :) 01:42 -!- ccdle12 [~ccdle12@cpc139350-aztw33-2-0-cust310.18-1.cable.virginm.net] has joined #rust-bitcoin 03:27 -!- gribble [~gribble@unaffiliated/nanotube/bot/gribble] has quit [Remote host closed the connection] 03:40 -!- gribble [~gribble@unaffiliated/nanotube/bot/gribble] has joined #rust-bitcoin 05:13 < stevenroose> sgeisler, BlueMatt: any chance I could get a review on https://github.com/rust-bitcoin/rust-bitcoin/pull/255 ? (refactor of the Address type) 05:13 < stevenroose> Let's try release rust-bitcoin v0.19.0 today! :) 05:15 < stevenroose> Also in need for a second ACK: https://github.com/rust-bitcoin/rust-bitcoin/pull/271 (adds JSON serializaton for all bip32 types and changed the OutPoint JSON to a string ":") 05:17 < stevenroose> https://github.com/rust-bitcoin/rust-bitcoin/pull/289 can be ACK'ed as well (exporting rust-bitcoin deps as bitcoin::hashes and bitcoin::secp256k1). It doesn't export the bitcoin_bech32 crate, but that dep is dropped in #255. 05:18 < sgeisler> stevenroose: currently reviewing #255 05:21 < stevenroose> Thanks! 05:28 < stevenroose> Let me try finish the compact blocks PR as well, even though it could go in as a .1 release later. 06:13 -!- ccdle12 [~ccdle12@cpc139350-aztw33-2-0-cust310.18-1.cable.virginm.net] has quit [] 06:19 < stevenroose> The bip39 crate doesn't have any unit tests that verify its implementation against the unit test vectors. Turns out their implementation is wrong. Quite glad I spent the last 2 days creating a C binding to libwally's bip39 implementation: https://github.com/infincia/bip39-rs/issues/21 06:23 < andytoshi> stevenroose: 265 and 271 both need rebase (and i'd kinda like both in for 0.19.0, though i'm not gonna delay the release for them) 06:23 < andytoshi> i guess they can both go in 0.19.1.. 06:29 < stevenroose> refactored #255 06:30 < stevenroose> (I assume that's what you meant with 265?) 06:31 < andytoshi> heh yeah i think so 06:31 < andytoshi> i gotta get a better monitor.. 06:31 < andytoshi> or maybe put my glasses on 06:33 < stevenroose> #271 rebased as wel 06:34 -!- elichai2 [uid212594@gateway/web/irccloud.com/x-ilwtsqaqfyvnibqo] has joined #rust-bitcoin 06:35 < andytoshi> oh i meant 289, not 265 06:35 < stevenroose> rebased #289 as well 06:35 < andytoshi> thanks :) 06:36 < elichai2> andytoshi: feeling any better today? 06:36 < stevenroose> BlueMatt: some fuzzing error: https://api.travis-ci.org/v3/job/555135910/log.txt 06:38 < stevenroose> Less priority, but the compact blocks PR is also ready for review: https://github.com/rust-bitcoin/rust-bitcoin/pull/249 06:38 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:70c3:8adb:1887:bce0] has joined #rust-bitcoin 06:40 < andytoshi> elichai2: mostly, yep, thanks 06:41 < elichai2> Good :) I realized something yesterday. `cc` build only static libraries, meaning that the `check_gmp` basically just checks that it can find the right headers, because AFAIK static libs don't pass the linker 06:42 < andytoshi> yeah, i saw that. frustrating. 06:42 < elichai2> this might be good enough, but then we need to gate it to linux/mac only(as targets not hosts), otherwise it will break cross compiling by finding the headers but the library doesn't really exist for the target 06:44 < andytoshi> gross 06:44 < elichai2> yeah... 06:44 < elichai2> it's weird how autotools still has a big advantage over "modern" stuff 06:57 < TamasBlummer> stevenroose: approved 249 after comparing with bip. feels right. the tests are sparse, but lets go ahead with this as it does not break anything existing 07:07 < stevenroose> TamasBlummer: thanks. I know the tests are sparse. Not easy to generate them and frankly there isn't much logic to tests, it's mostly data structures. 07:11 < andytoshi> is somebody using CB/blocked on CB in rust-bitcoin 07:12 < andytoshi> it would be cool if we could get fairly quick turnaround on actual usage 07:13 < TamasBlummer> it is only interesting for a project with mempool, which I do not have. 07:17 < stevenroose> andytoshi: I don't think so, so it's definitely not needed for 0.19. The reason I started the impl many months ago was because I noticed that Elements was exposing some CB RPCs that were totally stateless for the blocksigner to use. Not sure if they are even still in use or if the blocksigner ever used them. I just thought it was silly to have an HTTP roundtrip for something local code could do. 07:18 < andytoshi> yeah, the blocksigner will use this 07:19 < andytoshi> but it's not an immediate priority 07:19 < andytoshi> and we never got around to implementing it 07:58 -!- michaelsdunn1 [~michaelsd@unaffiliated/michaelsdunn1] has joined #rust-bitcoin 09:35 < stevenroose> andytoshi: I spotted the bug you found in address, I'm a bit puzzled the tests didnt catch that, there are tests with hardcoded script pubkeys. 09:35 < stevenroose> Ah, damn that, they only have version 0 vectors of course :| 09:37 < andytoshi> :/ 09:37 < andytoshi> frustrating 09:38 < andytoshi> i don't know if taproot programs are supposed to be 32 or 33 bytes, or i'd suggest making some taproot vectors 09:50 < stevenroose> So somehow the bip173 test vectors weren't all included n the Address unit tests. I think because Addres didn't used to support segwit addresses othe rthan version 0. So I redid the test vectors and included all of them. They all pass, so I think I got it right this time. 09:51 < stevenroose> It's very strange to me though that v0 is 0x00 and v1-v16 is 0x51.. 09:55 < elichai2> andytoshi: i'm pretty sure it wasn't defined yet if witness v1 program will have a compressed key or a X only key 10:10 < dongcarl> elichai2: https://github.com/rust-bitcoin/rust-bech32/blob/master/README.md 10:10 < andytoshi> stevenroose: awesome thanks 10:11 < andytoshi> and yes, Script is very strange in many ways 10:11 < elichai2> dongcarl: 👍 Thanks! 10:12 < elichai2> stevenroose: that actually confused me a lot too 10:19 < stevenroose> K, I'm heading out and going offline. All my PRs should be up to date and ready for review/mergable if ACK'd. :) 10:19 < stevenroose> But please don't skip review to get a release out :) The Address bug was a bit nasty.. Anyway, cheers! 10:20 < andytoshi> heh, well, we would've caught it as soon as it was used at least 10:24 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has quit [Remote host closed the connection] 10:25 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin 10:26 < dongcarl> andytoshi: What can I review to help with the release? 10:26 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has quit [Remote host closed the connection] 10:26 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin 10:27 < andytoshi> dongcarl: 255 (refactor address); 271 (serde for outpoint); 289 (export some dependencies) 10:27 < andytoshi> then i think we should be good to go 10:27 < dongcarl> andytoshi: roger, 10:27 < andytoshi> dongcarl: the motivation for #255 is that we can't support non-v0 segwit addresses unless we do our own bech32 support 10:27 < andytoshi> unless we want to go through rust-bitcoin-bech32 which has not been working for us 10:28 < andytoshi> we shouldn't have merged v0-only segwit support; i'm not sure why i didnt' think to look at that originally :/ 10:28 < andytoshi> the whole point of bip 173 is that people don't have to upgrade their software to deal with new address versions 10:29 < dongcarl> andytoshi: I'm a bit new to this issue, but are you saying that we can reduce the patchset for #255 if we fixed rust-bitcoin-bech32 to not be v0-only? 10:29 < andytoshi> maabye 10:30 < andytoshi> i think we also need access to some private fields of WitnessProgram 10:30 < andytoshi> and the result would still not be very clean because we'd have an address enum variant that was just wrapping a bech32 struct 10:31 < dongcarl> andytoshi: Okay, sounds like I should take a closer look, will get back to you 10:31 < andytoshi> also, i think the bulk of #255 is the new Address::from_script and is_standard logic 10:31 < andytoshi> and tests 10:31 < andytoshi> even if we could retain bitcoin-bech32, i'm doubtful we'd get a ton of reduction in the diff 10:32 < dongcarl> andytoshi: Cool, let me experiment a bit, and come back with numbers/patches 10:34 < andytoshi> cool. not a huge rush .. unfortunately my sleep schedule is a bit offset this week so i'm not overlapping with stevenroose as much as i usually do 10:34 < andytoshi> and it's friday evening for him, so maybe he's not back til monday :) 10:35 < dongcarl> got it 10:35 < dongcarl> scheduling-wise I'm carving out friday for rust-bitcoin 10:35 < dongcarl> s/friday/fridays/ 10:41 < andytoshi> awesome 11:12 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has quit [Remote host closed the connection] 11:12 -!- michaelsdunn1 [~michaelsd@unaffiliated/michaelsdunn1] has quit [] 11:14 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin 11:17 -!- reallll [~belcher@unaffiliated/belcher] has joined #rust-bitcoin 11:20 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 268 seconds] 12:01 -!- mauz555 [~mauz555@2a01:e35:8ab1:dea0:70c3:8adb:1887:bce0] has quit [] 16:10 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has quit [Ping timeout: 260 seconds] 16:17 < ariard> BlueMatt: was working on my covenant talk, back on pushing changes tonight! 16:17 < ariard> excited to see codecov in action 17:50 < BlueMatt> stevenroose: thats not an error? 17:50 < BlueMatt> stevenroose: the thing you linked me to ends with Done. Your build exited with 0. 17:53 -!- andytoshi [~apoelstra@unaffiliated/andytoshi] has quit [Ping timeout: 258 seconds] 18:28 -!- TamasBlummer1 [~Thunderbi@p200300DD6712644829EE84C905B174A1.dip0.t-ipconnect.de] has joined #rust-bitcoin 18:30 -!- TamasBlummer [~Thunderbi@p200300DD67126497CD76CC54AE545468.dip0.t-ipconnect.de] has quit [Ping timeout: 264 seconds] 18:30 -!- TamasBlummer1 is now known as TamasBlummer 19:21 -!- elichai2 [uid212594@gateway/web/irccloud.com/x-ilwtsqaqfyvnibqo] has quit [Quit: Connection closed for inactivity] 19:21 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin 20:50 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has quit [Ping timeout: 260 seconds] 21:08 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin 23:57 -!- reallll is now known as belcher --- Log closed Sat Jul 27 00:00:22 2019