--- Log opened Wed Mar 24 00:00:06 2021
00:03 -!- jonatack_ [~jon@37.165.85.219] has quit [Quit: jonatack_]
00:22 -!- pigeons [~pigeons@androzani.sysevolve.com] has quit [Ping timeout: 258 seconds]
00:23 -!- pigeons [~pigeons@androzani.sysevolve.com] has joined #joinmarket
00:23 -!- pigeons is now known as Guest54955
00:49 -!- jungly [~jungly@host-212-171-255-34.pool212171.interbusiness.it] has joined #joinmarket
01:15 -!- jungly [~jungly@host-212-171-255-34.pool212171.interbusiness.it] has quit [Ping timeout: 240 seconds]
01:21 -!- jungly [~jungly@host-212-171-255-34.pool212171.interbusiness.it] has joined #joinmarket
04:05 -!- jonatack [~jon@37.164.189.52] has joined #joinmarket
05:41 < belcher> waxwing the pr https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/837 blocks me doing stuff, could you make the one edit so we can merge it? or if you dont have time tell me and ill fork it and do it
05:44 < waxwing> belcher, oh sure i'll do it in an hour or so
05:46 < belcher> ty
06:05 -!- undeath [~undeath@hashcat/team/undeath] has joined #joinmarket
06:15 < waxwing> belcher, done
06:16 < belcher> ty
06:16 < waxwing> these clean up commits can be a bit annoying due to need to rebase everything. but now was as good a time as any.
07:21 -!- jonatack [~jon@37.164.189.52] has quit [Read error: Connection reset by peer]
08:52 < belcher> someone linked this http://bitcoinkpis.com/privacy
08:52 < belcher> site which graphs the volume and other stats of the 3 coinjoin implementations (joinmarket, wasabi, samourai)
08:59 < waxwing> data: "proprietary". ok :)
09:00 < waxwing> i mean it's interesting. they should for sure open source which algo/calcs they use, really.
09:00 < waxwing> i'd be curious to see if they're doing on-chain scan like `snicker-finder.py` does (or maybe an even better one in algo terms, that would be interesting)
09:00 < waxwing> if they just use IRC data it's going to be a bit meh
09:17 -!- bsm1175321 [~mcelrath@50-192-147-253-static.hfc.comcastbusiness.net] has joined #joinmarket
09:47 -!- keithhub[m] [keithhubma@gateway/shell/matrix.org/x-emxcmjxhtobpjsew] has quit [Ping timeout: 260 seconds]
09:50 -!- keithhub[m] [keithhubma@gateway/shell/matrix.org/x-nyvvwyqwoojzbczx] has joined #joinmarket
09:53 -!- jonatack [~jon@37.172.241.248] has joined #joinmarket
10:00 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has quit [Ping timeout: 264 seconds]
10:00 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has joined #joinmarket
10:37 -!- Kukks [023a2e14@2.58.46.20] has joined #joinmarket
10:39 -!- Kukks is now known as MrKukks
11:10 -!- jungly [~jungly@host-212-171-255-34.pool212171.interbusiness.it] has quit [Ping timeout: 260 seconds]
11:20 -!- lukedashjr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket
11:23 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 264 seconds]
11:25 -!- lukedashjr is now known as luke-jr
11:41 -!- MrKukks [023a2e14@2.58.46.20] has quit [Quit: Connection closed]
12:44 -!- jungly [~jungly@host-212-171-255-34.pool212171.interbusiness.it] has joined #joinmarket
14:31 < belcher> looks like wallet tool `signmessage` doesnt work? perhaps another omission from the python-bitcointx update, tracking it down now
14:35 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has quit [Ping timeout: 260 seconds]
14:36 -!- jungly [~jungly@host-212-171-255-34.pool212171.interbusiness.it] has quit [Ping timeout: 268 seconds]
14:37 < waxwing> belcher, thanks for taking a look. i'm just looking now, and the situation is rather confusing: ecdsa_raw_sign,verify are still using coincurve (as indeed is the podle calculation, due to the need for key addition which python-bitcointx didn't expose)
14:38 < waxwing> but the BitcoinMessage object from python-bitcointx *is* used for constructing the message to be signed.
14:38 < belcher> yes i was just looking at how the IRC message signing works
14:38 < waxwing> i do think i missed testing this during that upgrade, except in one specific sense: the IRC message signatures
14:38 < waxwing> yeah
14:38 < belcher> i need it for using a fidelity bond TXO to sign a message, so most likely it will end up using coincurve as well
14:39 < belcher> well its a standard saying that software needs to be used to continue working, obviously if nobody complained then very few people use `signmessage`
14:39 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has joined #joinmarket
14:40 < waxwing> that part is obviously pretty important :) but ofc it only needs to be self-consistent
14:40 < waxwing> sorry i got disconnected right in the middle of that
14:40 < waxwing> i see on the log, yes that is clearly a valid point that has occurred to me in the past, i don't know anyone using it.
14:40 < waxwing> and so i'm not even sure if it was compatibly functioning before that PR
14:40 < waxwing> all i know is that about two years ago when i did test it, it was compatible with Core
14:41 < waxwing> on a very surface level i don't *think* anything should have changed? wasn't it using that same coincurve api function sign_recoverable before?
14:42 < belcher> i think its crashing just because of one extra parameter passed to the function
14:44 < waxwing> oh it's crashing, well that's certainly a simpler question
14:45 < belcher> any idea what `formsg` does? https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/5eee9b3a823cd4304178a05ba61445f11aa3cfa4/jmbitcoin/jmbitcoin/secp256k1_main.py#L182
14:45 < belcher> this `sign_recoverable` function
14:46 -!- k3tan [~pi@gateway/tor-sasl/k3tan] has joined #joinmarket
14:47 < belcher> ill just search inside python-bitcointx
14:47 < waxwing> yeah looks like i just removed the rawmsg argument because we're not using that any more
14:47 < belcher> core's message signing makes use of pubkey recovery iirc, so this code should support that? i was planning to use that to make the fidelity bond announcements shorter
14:48 < waxwing> i.e. all used of ecdsa_sign are now for message signing, not for signing where you pass a hash, and it does the BitcoinMessage thing
14:48 < waxwing> so it looks like i changed that in the other places but not the wallet signmessage code, oops
14:49 < waxwing> pubkey recovery: think we didn't do that yet, let me double check
14:50 < belcher> iirc core's verifymessage takes a bitcoin address, a message, and signature, so it must use pubkey recovery because its not possible to go from a regular bitcoin address to a pubkey
14:50 < waxwing> it was ages and you probably don't remember at the time but we jus t didn't get round to it
14:50 < waxwing> we talked about it quite a bit at the time here
14:50 < waxwing> (at the time i was doing multiple servers so we needed the snooping defence)
14:50 < waxwing> so we are still sending pubkeys in every public message because of that
14:50 < belcher> yeah
14:51 < belcher> i wonder how `signmessage` could be compatible... ill try it myself now with core
14:51 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has quit [Ping timeout: 260 seconds]
14:51 < waxwing> pretty sure i tested it both ways, but .. this is worth checking yeah.
14:53 < waxwing> hmm, this looks to have been the original commit: https://github.com/JoinMarket-Org/joinmarket-clientserver/commit/28912acb5f0b7b2575f9fc82b574b7ed51523e1a#diff-326acdd19b880cb124ebd72bd49b94eb9b14167dfac0e1dad049059ec3c18e94R390-R393
14:53 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has joined #joinmarket
14:53 < belcher> so bitcoin core's error message is "Address does not refer to key (code -3)" which implies that it doesnt work on any address, but has to be an address in your own wallet(?)
14:53 < belcher> otherwise it doesnt know the public key
14:54 < belcher> electrum has a sign/verify, sec
14:54 < waxwing> belcher, i encountered something similar recently; i was surprised that core didn't seem to support it for bech32
14:54 < waxwing> i.e. verify against address by just checking the pubkey possibilities
14:54 < waxwing> and i made the same point, i.e. it's weird because electrum can do it
14:55 < belcher> lets be honest signmessage/verifymessage are memes anyway
14:55 < belcher> but ok no problem, fidelity bond announcements will just contain pubkeys as well
14:56 < belcher> electrum says error wrong signature
14:57 < belcher> sign/verifymessage is only useful in cases like proving craig wright wrong, its very rarely useful otherwise
14:59 < waxwing> belcher, can you check if it works pre that PR? so 0.6.3 last release i think.
15:00 < belcher> yep ill try
15:01 < waxwing> i want to chase up when that 31+ thing disappeared
15:03 < waxwing> it was in the 'replace secp256k1-py with coincurve' here: https://github.com/JoinMarket-Org/joinmarket-clientserver/commit/abad5977fe890456511ea894d8d68258e81cbfc2#diff-326acdd19b880cb124ebd72bd49b94eb9b14167dfac0e1dad049059ec3c18e94
15:04 -!- k3tan [~pi@gateway/tor-sasl/k3tan] has quit [Remote host closed the connection]
15:04 < waxwing> so i'll check the difference between what the old and 'new' libs did in their sign_recoverable methods
15:04 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has quit [Ping timeout: 272 seconds]
15:04 -!- k3tan [~pi@gateway/tor-sasl/k3tan] has joined #joinmarket
15:06 < belcher> ill try it with 0.6.2 to be sure
15:07 < belcher> it also doesnt work, with the same error message "Address does not refer to key (code -3)"
15:08 < belcher> i can easily try with any tag using git reset
15:08 < waxwing> the first commit i linked above is where the code was added
15:09 < waxwing> seems like checkout that commit is the right thing to do; as i say, was tested at the time, but .. i think with secp256k1-py so therefore i guess it just wouldn't work with a JM install today(?)
15:09 < belcher> that commit is dated june 2017
15:09 < belcher> 0.6.2 is march 2020
15:10 < belcher> well ill try that commit
15:11 < belcher> "ModuleNotFoundError: No module named 'btc'" <--- yep looks like its too old to work easily
15:11 < waxwing> yeah because of secp256k1-py
15:11 < waxwing> yeah duh of course way different time
15:12 < belcher> i tried with v0.5.5 dated august 2019, so definitely before python-bitcointx, and it also produces a signature that doesnt verify
15:12 < waxwing> so it's like: after testing and fiddling around i made it work in 2017 with secp256k1-py but when james hilliard did the replacement, the signmessage function probably stopped working then.
15:12 < belcher> in fact its exactly the same signature as produced by 0.6.2
15:12 < belcher> aha
15:12 < waxwing> no, the delta is coincurve not python-bitcointx .. or, uh, at least, the *first* delta
15:13 < belcher> aha
15:13 < waxwing> and that was 2018 November
15:14 < waxwing> i'm still kinda surprised though, if it flat out didn't work since then and nobody complained. i'm probably forgetting some small details here and there.
15:14 < waxwing> anyway perhaps best is just to get it working as is :)
15:15 < waxwing> of course the subtlety is 'didn't work' means 'not compatible', not necessarily doesn't work at all
15:15 < waxwing> and i think the pbtx merge is what actually broke it as a syntax error, as you've pointed out
15:25 -!- jonatack [~jon@37.172.241.248] has quit [Ping timeout: 268 seconds]
15:25 < waxwing> it looks like it might be a case of just reading the last byte of what coincurve returns here: https://github.com/ofek/coincurve/blob/c3b98904be16aecf4c7f43135e05dd62ce180d48/coincurve/ecdsa.py#L48
15:32 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has joined #joinmarket
15:32 -!- bsm1175321 [~mcelrath@50-192-147-253-static.hfc.comcastbusiness.net] has quit [Remote host closed the connection]
15:44 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has quit [Ping timeout: 272 seconds]
15:50 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has joined #joinmarket
15:57 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has quit [Ping timeout: 272 seconds]
16:01 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has joined #joinmarket
16:08 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has quit [Ping timeout: 260 seconds]
16:24 -!- fiatjaf [~fiatjaf@2804:7f2:2a8c:f814:ea40:f2ff:fe85:d2dc] has joined #joinmarket
16:47 < waxwing> belcher, i can figure this out tomorrow (specifically: figure out the right calls in coincurve to make it compatible), if that sounds good?
16:47 < waxwing> (something to note here, we won't want to change the existing code for the bots irc stuff so it'll be another layer)
16:56 < belcher> waxwing as far as im concerned its no rush, because the fidelity bond stuff wont really use it
16:56 < waxwing> ok thanks
17:07 -!- undeath [~undeath@hashcat/team/undeath] has quit [Quit: WeeChat 3.0.1]
18:00 -!- mryandao [~mryandao@gateway/tor-sasl/mryandao] has quit [Remote host closed the connection]
18:00 -!- mryandao [~mryandao@gateway/tor-sasl/mryandao] has joined #joinmarket
19:06 -!- mryandao_ [~mryandao@gateway/tor-sasl/mryandao] has joined #joinmarket
19:06 -!- mryandao [~mryandao@gateway/tor-sasl/mryandao] has quit [Ping timeout: 268 seconds]
20:04 -!- Saloframes_ [~Saloframe@71-218-119-44.hlrn.qwest.net] has quit [Quit: Leaving]
20:06 -!- Saloframes [~Saloframe@71-218-119-44.hlrn.qwest.net] has joined #joinmarket
20:12 -!- Cory [~Cory@unaffiliated/cory] has quit [Ping timeout: 260 seconds]
20:17 -!- Pasha [~Cory@unaffiliated/cory] has joined #joinmarket
20:19 -!- belcher_ [~belcher@unaffiliated/belcher] has joined #joinmarket
20:20 -!- Pasha is now known as Cory
20:23 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 272 seconds]
20:29 -!- belcher_ is now known as belcher
22:15 -!- Saloframes_ [~Saloframe@71-218-119-44.hlrn.qwest.net] has joined #joinmarket
22:18 -!- Saloframes [~Saloframe@71-218-119-44.hlrn.qwest.net] has quit [Ping timeout: 245 seconds]
23:13 -!- jungly [~jungly@host-80-182-81-104.pool80182.interbusiness.it] has joined #joinmarket
23:29 -!- jungly [~jungly@host-80-182-81-104.pool80182.interbusiness.it] has quit [Ping timeout: 245 seconds]
23:32 -!- waxwing [~waxwing@unaffiliated/waxwing] has quit [Quit: ZNC 1.7.4+deb0+bionic0 - https://znc.in]
23:33 -!- waxwing [~waxwing@193.29.57.116] has joined #joinmarket
23:33 -!- waxwing [~waxwing@193.29.57.116] has quit [Changing host]
23:33 -!- waxwing [~waxwing@unaffiliated/waxwing] has joined #joinmarket
23:52 -!- jungly [~jungly@host-80-182-81-104.retail.telecomitalia.it] has joined #joinmarket
--- Log closed Thu Mar 25 00:00:07 2021