--- Log opened Fri Aug 06 00:00:31 2021 00:12 < JMBridge> [darkscience/joincoins] Friendly reminder, day after tomorrow the darkscience irc v2 will be brown out downtime, update your joinmarket.cfg to osv3 if not done yet. if you have darkirc6tqgpnwd3blln3yfv5ckl47eg7llfxkmtovrv7c7iwohhb6ad.onion in your config nothing needs to be changed. references: https://darkscience.net/news/onion-v2-deprecation-timeline/ 00:37 -!- NorrinRadd [~username@102.67.16.112] has joined #joinmarket 00:40 -!- NorrinRadd [~username@102.67.16.112] has quit [Client Quit] 01:00 -!- NorrinRadd [~username@102.67.16.112] has joined #joinmarket 01:05 -!- NorrinRadd [~username@102.67.16.112] has quit [Ping timeout: 240 seconds] 01:58 < waxwing> thanks joincoins 02:00 < waxwing> undeath i realised last night that my #969 depends on identifying "is this a standard wallet scriptpubkey", which you address in #957 with `_is_native_keypair`. 02:00 -!- BUSY [~BUSY@user/busy] has quit [Ping timeout: 272 seconds] 02:00 < waxwing> i have two nits on that name: native can confuse with native as in segwit, and second a keypair is not custom or not, only a script (or path) is. 02:00 < waxwing> but that's just naming. 02:01 < waxwing> otherwise if we can merge some version of #957, i can use the same code in #969 02:01 < waxwing> (currently it tries to judge with wallet.TYPE which is just plain wrong) 02:06 -!- NorrinRadd [~username@102.67.16.112] has joined #joinmarket 02:11 -!- NorrinRadd [~username@102.67.16.112] has quit [Ping timeout: 258 seconds] 02:21 -!- NorrinRadd [~username@102.67.16.112] has joined #joinmarket 02:24 -!- undeath [~undeath@user/undeath] has joined #joinmarket 02:26 -!- NorrinRadd [~username@102.67.16.112] has quit [Ping timeout: 272 seconds] 02:29 -!- NorrinRadd [~username@102.67.16.112] has joined #joinmarket 02:31 < waxwing> undeath, ping, i think you weren't here when i wrote the above ^ 02:32 < undeath> pong. yes, i wasn't 02:34 < undeath> yes. i had trouble coming up with a good name for that method 02:35 -!- NorrinRadd [~username@102.67.16.112] has quit [Quit: My MacBook has gone to sleep. ZZZzzz…] 02:35 < waxwing> is_standard_script or is_wallet_script ? ('standard' has a special meaning in bitcoin consensus but i don't see any potential confusion there) 02:35 < waxwing> is_wallet_script is kinda dumb i guess, maybe not that one 02:36 < waxwing> is_standard_wallet_script :) 02:36 < undeath> standard seems confusing here 02:36 < undeath> oh yes 02:36 < undeath> that last one seems ok 02:36 < waxwing> you could even use default 02:36 < undeath> and make it non-protected if you need it elsewhere in the code 02:36 < waxwing> is_default_wallet_script 02:36 < waxwing> oh good point, yes, non _ 02:37 < undeath> that is a bit confusing given the naming of other methods and the path argument it takes 02:37 < waxwing> i feel like script is more fundamental than path 02:37 < undeath> but a docstring can take care of that 02:38 < waxwing> i feel like it should be a class method? 02:39 < undeath> on a related note, i think the implementation for the ImportWalletMixin is not correct 02:39 < undeath> that wouldn't work for imported keys 02:40 < undeath> for FB wallets the engine will be different but the "standard" type could still be the same 02:40 < waxwing> i see. 02:41 < undeath> though with the ability removed to import other key types that's a non-issue? 02:43 < waxwing> talking about this right? https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/957/files#diff-b48a1717de3d71415880b8a18d7e461225bd8ddfdc09010254439897bc19fd6fR1768 02:43 < undeath> yes 02:46 < undeath> ah, no 02:46 < undeath> that works 02:46 < undeath> there's a special engine for timelocked addresses that does not replace the normal one 02:46 < undeath> perfect 02:47 < waxwing> _TIMELOCK_ENGINE right 02:47 < undeath> yep 02:49 < waxwing> so i guess you'll just rename that function and we're basically all good? is #957 something we can merge do you think? 02:49 < waxwing> i'm just taking one more look through it 02:49 < undeath> have pushed a commit to rename the method 02:50 < undeath> will squash once you're done with your review 02:50 < undeath> #957 should be good to merge 02:50 < undeath> after squashing 02:51 < waxwing> i'm just curious, why 'obj'? i guess the IDE complains about object or something? 02:51 < undeath> no, that was just lazyness 02:51 < undeath> i changed the docstring, then realized the datetime_ method is needed and recreated the "old" docstring 02:52 < undeath> only saw in the diff that it causes a change 02:52 < undeath> i'll revert that 02:53 < undeath> done 02:57 < waxwing> you can change var names away from 'native' too if you like but i don't care too much about that 02:57 < undeath> oh, right 03:02 < undeath> that required more commit than it should but i think we're good now 03:10 < waxwing> yeah i think it's fine. as per comment it passes, can you squash after rebase now (it's a trival rebase iirc) undeath 03:11 < waxwing> or before, not after :) 03:26 < undeath> done 03:42 < undeath> just saw your review comments. will fix 03:47 < waxwing> i've tested manually making sure that we get the behaviour intended re: blocking ioauth, even though of course it doesn't work in the scenario we mentiond. i'm happy to merge at this point. OK with you undeath ? 03:47 < undeath> just pushed a commit to fix the two issues you commented on 03:47 < undeath> good to merge now 03:48 < waxwing> "scenario we mentioned" = it can incorrectly try to spend TLU and fail, but that's known and fixed in the other PR 03:51 -!- JMBridged [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket 03:52 < undeath> ugh wait 03:52 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Ping timeout: 244 seconds] 03:52 -!- JMBridged is now known as JMBridge 03:54 < undeath> yes, ok. everything as intended 04:00 -!- NorrinRadd [~username@102.67.16.112] has joined #joinmarket 04:31 -!- NorrinRadd [~username@102.67.16.112] has quit [Quit: My MacBook has gone to sleep. ZZZzzz…] 04:39 -!- k3tan [~pi@user/k3tan] has quit [Ping timeout: 272 seconds] 04:41 -!- k3tan [~pi@user/k3tan] has joined #joinmarket 05:23 -!- undeath [~undeath@user/undeath] has quit [Quit: WeeChat 3.1] 07:52 -!- jonatack [~jonatack@user/jonatack] has quit [Quit: Client closed] 07:52 -!- jonatack [~jonatack@user/jonatack] has joined #joinmarket 08:13 < dc81> what happens if you have a second fidelity utxo in your wallet? 08:25 < waxwing> dc81, it only chooses the most valuable 08:25 < waxwing> i think it mentions this in the fidelity_bond.md guide 09:33 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Quit: Bridge selfdestructed.] 09:35 -!- undeath [~undeath@user/undeath] has joined #joinmarket 09:53 -!- jonatack [~jonatack@user/jonatack] has quit [Quit: Client closed] 10:08 < undeath> i'm having a kind of deja vu with #970. wasn't there a pr for this already? 10:50 < belcher> new fidelity bond locking up 10btc https://blockstream.info/tx/a574046d33532b5cd3b64acec76a4757db3f40c39f05e2f4303b2c66063fa7b3 only until march 2022 10:50 < belcher> unfortunately they created a change address 10:51 < belcher> though if it goes straight back into more coinjoins maybe it doesnt matter 11:18 -!- jonatack [~jonatack@user/jonatack] has joined #joinmarket 11:24 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket 12:01 < waxwing> if we get #965 and #969 merged i suggest making 0.9.1 early next week. 12:01 < undeath> yeah, lots of fixes 12:02 < waxwing> it's basically just a release fixing the bugs in spending timelocked. btw i think we need a bit more text in the fidelity_bonds guide to help people with this. belcher 12:03 < belcher> what kind of text 12:03 < belcher> iv got a todo to write something about what the guy said in the telegram group 12:03 < waxwing> the psbt bug will remain but i think that's ok as long as it's documented that it doesn't work with that. it can be patched later. 12:03 < belcher> about only one utxo being announced or used 12:04 < waxwing> we need to tell them all the conditions on what is possible. sendpayment with direct send always works, inclusion in taker is possible (but arguable if recommended). inclusion in maker is not possible it gets auto-frozen when you start. 12:04 < waxwing> spend in psbt is not possible. 12:04 < waxwing> and any recommendations we have about spending into and out of them, maybe that's already in doc. 12:04 < waxwing> (note that even the first, direct_send, was bugged in the release if it was not a sweep) 12:05 < waxwing> (which is perhaps the biggest reason for a quick bugfix release) 12:05 < waxwing> we already have the warning that you can't export so that part's ok 12:05 < belcher> sounds good to me 12:10 < waxwing> undeath, re: podle error string construction in test, i guess i'll just go with your "add to script_map" suggestion because it's the smallest change, and no script should ever reach that code path that isn't in the wallet already, even if it's exotic. 12:11 < undeath> but you'd have to parse the script for that 12:11 < waxwing> how do you mean? 12:12 < undeath> you need to construct the script's path to put something in the script map 12:12 < waxwing> yeah but we already have dummy paths in that test 12:12 < undeath> oh, it's only for the test? 12:12 < waxwing> yes 12:12 < undeath> well, yes, whatever I guess 12:13 < waxwing> what i mean is, the insertion of this script is just for the test 12:13 < undeath> yeah 12:13 < undeath> the tests are hacky as hell anyway 12:13 < waxwing> because the test is on a DummyWallet which now inherits from LegacyWallet because it has hardcoded legacy utxos 12:13 < waxwing> indeed 12:21 < waxwing> ok so that works, i can inject a dummy path for that in _script_map, and i'll also remove the change in output.py. smaller delta. 12:22 < undeath> same thing in taker.py though where you're also catching AssertionError 12:23 < undeath> but there you can just use is_known_script() 12:27 < waxwing> yeah that's what i meant. but anyway let's see :) 13:46 < undeath> waxwing: i think you forgot to commit jmclient/output.py 13:52 <+JoinMarketRelay> [hackint/kristapsk_] Is there any reason we show xpub for fidelity bonds (m/49'/1'/0'/2 and m/49'/1'/0'/3)? I don't see it useful, as they aren't currently importable to any non-CJ wallet anyway. Or I'm missing something? 13:54 < waxwing> undeath, yes that was a brainfart: i did checkout, forgetting that i was reverting a change. will add. 14:04 < undeath> tests are failing because of an assert with the hacked path 14:04 < undeath> otherwise the code looks good to me 15:59 -!- jonatack [~jonatack@user/jonatack] has quit [Quit: Client closed] 16:03 -!- undeath [~undeath@user/undeath] has quit [Quit: WeeChat 3.1] 16:07 -!- jonatack [~jonatack@user/jonatack] has joined #joinmarket 16:28 -!- openoms_ [~quassel@gateway/tor-sasl/openoms] has joined #joinmarket 16:28 -!- openoms [~quassel@gateway/tor-sasl/openoms] has quit [Ping timeout: 244 seconds] 16:58 -!- BUSY [~BUSY@user/busy] has joined #joinmarket 17:32 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Remote host closed the connection] 17:32 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket 18:40 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Ping timeout: 244 seconds] 18:44 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket 18:44 -!- busybusybusy [~BUSY@user/busy] has joined #joinmarket 18:45 -!- BUSY [~BUSY@user/busy] has quit [Killed (NickServ (GHOST command used by busybusybusy))] 18:45 -!- busybusybusy is now known as BUSY 19:44 -!- belcher_ [~belcher@user/belcher] has joined #joinmarket 19:48 -!- belcher [~belcher@user/belcher] has quit [Ping timeout: 258 seconds] 20:16 -!- NorrinRadd [~username@154.6.20.108] has joined #joinmarket 20:16 -!- jonatack [~jonatack@user/jonatack] has quit [Ping timeout: 246 seconds] 21:10 -!- ufotofu [~ufotofu@user/ufotofu] has joined #joinmarket --- Log closed Sat Aug 07 00:00:31 2021