--- Log opened Mon May 03 00:00:44 2021
00:25 -!- b10c [~b10c@static.55.136.76.144.clients.your-server.de] has joined #rust-bitcoin
01:50 < dr_orlovsky> andytoshi: can I pls ask you to fix https://github.com/rust-bitcoin/bitcoin_hashes/pull/123 and publish bitcoin_hashes release so we can unblock progress on 0.26.1?
03:29 -!- Albertha84Borer [~Albertha8@static.57.1.216.95.clients.your-server.de] has joined #rust-bitcoin
05:18 -!- fiatjaf [~fiatjaf@2804:7f2:298d:1e20:ea40:f2ff:fe85:d2dc] has joined #rust-bitcoin
05:38 < dr_orlovsky> elichai2: you may be interested: I had a good progress on simplifying derivation macros and creating helper library that allows to write attribute requirements for syn. Some details and how it can be related to rust-bitcoin: https://github.com/rust-bitcoin/rust-bitcoin/issues/510#issuecomment-831231971
05:42 -!- Albertha84Borer [~Albertha8@static.57.1.216.95.clients.your-server.de] has quit [Ping timeout: 265 seconds]
06:21 -!- ghost43 [~daer@gateway/tor-sasl/daer] has quit [Remote host closed the connection]
06:21 -!- ghost43 [~daer@gateway/tor-sasl/daer] has joined #rust-bitcoin
08:14 -!- gribble [~gribble@unaffiliated/nanotube/bot/gribble] has quit [Read error: Connection reset by peer]
08:20 -!- valwal [~valwal@96.224.58.144] has joined #rust-bitcoin
08:22 -!- gribble [~gribble@unaffiliated/nanotube/bot/gribble] has joined #rust-bitcoin
09:54 -!- jonatack [~jon@88.127.52.83] has quit [Ping timeout: 252 seconds]
09:56 -!- jonatack [jon@gateway/vpn/airvpn/jonatack] has joined #rust-bitcoin
12:34 -!- grimler32 [~grimler67@pool-108-49-215-20.bstnma.fios.verizon.net] has joined #rust-bitcoin
12:40 < ariard> valwal: https://github.com/rust-bitcoin/rust-lightning/pull/845#discussion_r625319209 lmk you're thinking here, otherwise i think #845 is quite ready
12:48 < grimler32> Can Anyone help me with a xk memo replica or backwards compatable D.W.G?
12:53 < andytoshi> dr_orlovsky: lol oops i forgot, yeah will do this in 5 mins
12:53 -!- Netsplit *.net <-> *.split quits: junderw, evalr, justinmoon_, CubicEarth, belcher, ariard
12:55 -!- Netsplit over, joins: belcher, justinmoon_, CubicEarth, ariard, junderw, evalr
13:09 < dr_orlovsky> andytoshi: I understand :) the same happens for me all the time
13:13 < andytoshi> ok pushed
13:23 < dr_orlovsky> thanks. LGTM. Who has the approval powers to rust_bitcoin?
14:01 < dr_orlovsky> stevenroose, elichai2, whoever else has the merge power on bitcoin_hashes, can you pls review https://github.com/rust-bitcoin/bitcoin_hashes/pull/123 ?
14:02 < dr_orlovsky> its trivial minor version bump to get the release and unblock route for bitcoin 0.26.1
15:06 < andytoshi> ok tagged and released
15:09 < andytoshi> good to go for 0.26.1 then?
15:59 < ariard> BlueMatt: https://github.com/rust-bitcoin/rust-lightning/pull/845#discussion_r625372926 because the first htlc is now considered as a dust as we bump acceptor's limit from 26 sat to 330
15:59 < BlueMatt> ariard: yep I have a patch, give me 2 minutes to put it in git :)
15:59 < ariard> though i agree we should replace the fuzz check by a new capture logger specific to dust fallback
16:05 < BlueMatt> ariard: see top commit at https://git.bitcoin.ninja/index.cgi?p=rust-lightning;a=shortlog;h=refs/heads/2021-05-845-tweak
16:05 < BlueMatt> do you wanna take it on 845 or should I just open as a new pr?
16:05 < ariard> BlueMatt: looking on it
16:07 < ariard> BlueMatt: ah surprised that didn't break more of the coverage by bumping HTLC value...well the other way around is to add a new logger capture for dust
16:07 < ariard> because currently all HTLC in full_stack are non-dust ones? i'm fine with either way
16:08 < BlueMatt> no, the first few htlcs are dust
16:08 < BlueMatt> they just dont get broadcast on-chain iirc
16:08 < BlueMatt> but those seeds are still in my fuzz directories
16:08 < BlueMatt> so they'll still get coverage
16:08 < BlueMatt> this way we also hit the broadcasting of the final commitment transaction
16:09 < ariard> right but we can hit the broadcast of the final commitment transaction by capturing in channelmonitors L1815
16:10 < BlueMatt> L1815 in...what?
16:10 < ariard> in channelmonitors.rs
16:10 < BlueMatt> right, point being the previous seeds are still floating around, so those will still be used
16:11 < BlueMatt> but good to *also* have a seed that is non-dust
16:11 < BlueMatt> i guess we could add *both*, but I dont really wanna have a new test that tests even more seeds
16:13 < BlueMatt> i believe that that last htlc is the only htlc which we forward onwards as a non-dust htlc
16:13 < BlueMatt> so its the only non-dust outbound/received htlc
16:13 < ariard> BlueMatt: right, better to preserve existing seeds, taking your commit on top of #845
16:13 < BlueMatt> but i may be wrong, it just looked like the highest number
16:15 < ariard> hmmm testing locally
16:17 < ariard> BlueMatt: well patch works well, so 845 should be ready
16:22 < BlueMatt> awesome. you happy with https://github.com/rust-bitcoin/rust-lightning/pull/901#discussion_r625410605 ?
16:24 < ariard> BlueMatt: yep so much compilation magic! re-acking
16:25 < BlueMatt> cool. wanna ack https://github.com/rust-bitcoin/rust-lightning/pull/909 ? :)
16:27 < ariard> BlueMatt: well i wanna doing it a hour ago, was waiting on 845/901 :)
16:27 < BlueMatt> heh
16:27 < BlueMatt> yea, well both I'll merge once ci passes
16:28 < ariard> done!
16:42 < BlueMatt> huh. so rust-bitcoin is failing to build
16:45 < BlueMatt> andytoshi: did you upload a new bitcoin-0.26 that overloaded the previous somehow?
16:46 < BlueMatt> oh, no, i think the patch for our secp just went stale again
16:46 < BlueMatt> ughhhhh
16:47 < BlueMatt> can we land https://github.com/rust-bitcoin/rust-secp256k1/pull/282 ? its causing our CI to fail every time a new version is uploaded because the patch in our cargo.toml breaks
16:52 < BlueMatt> ohhhhhhhh, no, andytoshi bitcoin_hashes broke
16:53 < BlueMatt> rust-bitcoin fails to build with the latest version of bitcoin_hashes, with error[E0433]: failed to resolve: maybe a missing crate `core`?
16:53 < BlueMatt> in 54 | hash_newtype!(FilterHeader, sha256d::Hash, 32, doc="Filter header, as defined in BIP-157");
17:06 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 252 seconds]
17:12 -!- jeremyrubin [~jr@024-176-247-182.res.spectrum.com] has joined #rust-bitcoin
17:19 -!- belcher [~belcher@unaffiliated/belcher] has joined #rust-bitcoin
17:20 < andytoshi> sigh
17:20 < BlueMatt> andytoshi: I...really dont know how to fix this
17:20 < BlueMatt> the issue is rust-bitcoin doesn't have an "extern crate core"
17:20 < BlueMatt> which is needed to have a ::core
17:21 < BlueMatt> but you cant re-export core in no-std, even as a "pub use"
17:21 < BlueMatt> so, like...what can you do?
17:21 < andytoshi> can rust-bitcoin add an 'extern crate core'?
17:21 < BlueMatt> that would fix it
17:21 < Thomas[m]4> <BlueMatt "but you cant re-export core in n"> I am curious, why not?
17:22 < andytoshi> what was the previous situation? weren't we re-exporting core?
17:22 < BlueMatt> but its...kinda awkward that the public interface for bitcoin_hashes would *require* it if you use the macros
17:22 < BlueMatt> the previous situation only built for std
17:22 < andytoshi> ah lol
17:22 < BlueMatt> only in no-std are you not allowed to re-export
17:22 < BlueMatt> Thomas[m]4: "error: extern crate `core` is private, and cannot be re-exported (error E0365), consider declaring with `pub`"
17:22 < andytoshi> BlueMatt: agreed, this is a pretty awkward requirement ... hmmm
17:23 < andytoshi> but i think that may be the situation that we're in
17:23 < BlueMatt> andytoshi: if we re-add the pub extern crate part then we only support building on std for use of those macros
17:23 < BlueMatt> which is maybe ok
17:23 < BlueMatt> but that means we need to fix it For Real come no_std support in rust-bitcoin
17:23 < BlueMatt> soooo....i dunno
17:24 < andytoshi> that error sounds like you just need to add an extra pub?
17:24 < BlueMatt> pub...where?
17:24 < BlueMatt> if you set no_std you seem to get an implicit extern crate core
17:24 < BlueMatt> without pub
17:25 < andytoshi> ah i see
17:25 < Thomas[m]4> Can you do a: `pub mod export { pub mod core { pub use ::core; }  }`?
17:25 < BlueMatt> lets see!
17:25 < andytoshi> lol!
17:25 < Thomas[m]4> and then the macro would need: `$crate::export::core` instead
17:25 < BlueMatt> oh, wait, I lied
17:26 < BlueMatt> I'm a moron, the fix is eaier
17:40 < BlueMatt> nope, none of that worked. same error as before
17:40 < BlueMatt> "extern crate core is private".....
17:41 < Thomas[m]4> I guess you need `fmt` from core?
17:41 < BlueMatt> ah, true, could just pull in the bits we need
17:41 < BlueMatt> let me try that
17:44 < BlueMatt> ah! that worked.
17:44 < BlueMatt> ok, let me clean up the commit message, but https://github.com/rust-bitcoin/bitcoin_hashes/pull/125
17:47 < Thomas[m]4> Neat, probably want to add a `doc(hidden)` to make sure this doesn't show up in docs.rs given that it is really just a workaround.
17:47 < BlueMatt> right
17:48 < BlueMatt> ok, take a gander at https://github.com/rust-bitcoin/bitcoin_hashes/pull/125, andytoshi
17:51 < Thomas[m]4> How did the test of the `embedded` crate not catch that?
17:51 < Thomas[m]4> I was gonna suggest to add a test that uses the macros from an external crate but `embedded` is literally that.
17:53 < BlueMatt> it doesnt use rust-bitcoin
17:53 < BlueMatt> it just uses bitcoin_hashes
17:53 < BlueMatt> ie, it didnt call the macros that blew up
17:53 < BlueMatt> oh, it does, huh
17:54 < BlueMatt> oh, no, sorry, the issue was you needed to call those macros from std
17:54 < Thomas[m]4> yep that is what I thought :D
17:54 < BlueMatt> not no_std
17:56 < andytoshi> lol
17:56 < andytoshi> arguably we should add a test for this
17:56 < andytoshi> but even more arguably, this is a bug in rust and we can't test every fucking feature combination as a dep
17:56 < BlueMatt> yea....
17:56 < BlueMatt> the features stuff is such a mess.
17:57 < andytoshi> BlueMatt: can you add a minor version bump to the PR
17:57 < BlueMatt> yea
17:58 < BlueMatt> andytoshi: done
18:00 < andytoshi> kk one sec
18:07 -!- jeremyrubin [~jr@024-176-247-182.res.spectrum.com] has quit [Read error: Connection reset by peer]
18:09 -!- jeremyrubin [~jr@024-176-247-182.res.spectrum.com] has joined #rust-bitcoin
18:11 < andytoshi> ok tagged and released
18:12 < BlueMatt> thanks
18:12 < BlueMatt> now back to cutting a rust-lightning release
18:16 < dr_orlovsky> andytoshi: thank you! So with 0.26.1 we have a milestone listing all PRs: https://github.com/rust-bitcoin/rust-bitcoin/milestone/11
18:16 < dr_orlovsky> 15 of them are already merged, 7 are pending. Out of these 7, 5 already approved by two reviewers and are awaiting your merge, andytoshi
18:17 < dr_orlovsky> the last 2 are the one with error derives which was dependent on upstream releases, which I just updated: https://github.com/rust-bitcoin/rust-bitcoin/pull/559
18:18 < dr_orlovsky> and the second one is proposal of `CONTRIBUTING.md` file which has 1 ACK and passed three reviewing rounds and likely ready (but need your opinion on that): https://github.com/rust-bitcoin/rust-bitcoin/pull/587
18:18 < dr_orlovsky> also I hope milestone list will help you prepare CHANGELOG (or I can do it, if you busy)
18:20 < dr_orlovsky> sgeisler, stevenroose, sanket1729: for https://github.com/rust-bitcoin/rust-bitcoin/pull/559 we will need a second ACK, so will much appreciate your participation
18:23 < andytoshi> nice, thanks! my brain is a bit fried tonight, i will try to get through these all tomorrow
18:35 < dr_orlovsky> sure, thank you!
18:39 < dr_orlovsky> also, FYI to have everything related to 0.26.1 in one place here, I just recognized that we can use bech32m from bech32 0.8 crate without API break in 0.26.1 and did a PR for that: https://github.com/rust-bitcoin/rust-bitcoin/pull/601
18:41 < sanket1729> dr_orlovsky: ACKed it and other PRs. Looks like there is some unrelated fuzzer failure
18:42 < dr_orlovsky> and nightly still fails with core dependency: error[E0603]: crate `core` is private
18:43 < sanket1729> Ah, yeah. IDK why, but I always assume fuzzer failure if it fails after 10 mins or so.
18:43 < dr_orlovsky> preparing PR to fix that
18:43 < dr_orlovsky> yes, with fuzzier also something weird, I had to remove it as a feature here: https://github.com/rust-bitcoin/rust-bitcoin/pull/559/files#diff-2e9d962a08321605940b5a657135052fbcef87b5e360662bb527c96d9a615542R17
18:44 < dr_orlovsky> I assume the feature was removed from bitcoin_hashes
18:48 < dr_orlovsky> my fix: https://github.com/rust-bitcoin/rust-bitcoin/pull/602
18:50 < dr_orlovsky> going to bed for now, will check all CI status tomorrow morning and if this PR will succeed, will ask to merge it first so I can re-base other newly updated PRs with failing CI on top
18:51 -!- CubicEarth [~CubicEart@c-67-168-1-172.hsd1.wa.comcast.net] has quit [Ping timeout: 252 seconds]
18:54 -!- CubicEarth [~CubicEart@c-67-168-1-172.hsd1.wa.comcast.net] has joined #rust-bitcoin
19:04 < dr_orlovsky> ok, CI seems to be fixed, rebased PRs, so everything is ready for the merge & review (3 PRs are pending review, including the CI fix) - see https://github.com/rust-bitcoin/rust-bitcoin/milestone/11
19:32 < BlueMatt> sgeisler: we tagged new releases for rust-lightning, also bumping lightning-invoice's version. Not sure how you want to deal with uploads going forward - I'm happy to just do it with the rest of rust-lightning stuff, but wanted to check with you.
19:33 < BlueMatt> this would be the first version where it depends on rust-lightning and also has c bindings and also has now been updated for using invoicefeatures/supporting feature flag requirements/etc.
19:38 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has quit [Ping timeout: 240 seconds]
19:53 -!- DeanGuss [~dean@gateway/tor-sasl/deanguss] has joined #rust-bitcoin
22:14 -!- jeremyrubin [~jr@024-176-247-182.res.spectrum.com] has quit [Ping timeout: 252 seconds]
23:27 -!- jeremyrubin [~jr@024-176-247-182.res.spectrum.com] has joined #rust-bitcoin
--- Log closed Tue May 04 00:00:45 2021