< bitcoin-git> [bitcoin] laanwj closed pull request #15791: net: Hardcoded seeds update for 0.18 (master...2019_04_hardcoded_seeds_update) https://github.com/bitcoin/bitcoin/pull/15791
< bitcoin-git> [bitcoin] sipa opened pull request #15826: Pure python EC (master...201904_pythonec) https://github.com/bitcoin/bitcoin/pull/15826
< jonasschnelli> sipa: since you commented on #15615, would be good to get a ack/nack from your side
< gribble> https://github.com/bitcoin/bitcoin/issues/15615 | Add log output during initial header sync by jonasschnelli · Pull Request #15615 · bitcoin/bitcoin · GitHub
< jonasschnelli> Same for #15512
< gribble> https://github.com/bitcoin/bitcoin/issues/15512 | Add ChaCha20 encryption option (XOR) by jonasschnelli · Pull Request #15512 · bitcoin/bitcoin · GitHub
< sipa> jonasschnelli: will have a look tomorrow +and remind me if i forget)
< jonasschnelli> sipa: sure. Thanks. No hurry and sorry for the bother
< * wumpus> wishes we had called 'signrawtransactionwithwallet' just 'signwithwallet'
< * gmaxwell> wishes someone had suggested that
< wumpus> yes, I don't mean it as criticism, just reflection after some use
< wumpus> it doesn't matter for programmatic use; I guess aliases on bitcoin-cli could be a thing
< gmaxwell> bash aliases are a thing too.. if we break the interface again we could remember this and rename it.
< wumpus> I hope that won't be necessary any time soon! anyhow, might add something like 'keep RPC command names short, put documentation and further descrption in the help; rationale: short names tend to be easier to remember and save typing' to the developer notes
< gmaxwell> +1 I think get got signrawtransactionwithwallet because already a bunch of rpcs are really long
< Sentineo> hm maybe if the rpc interface could narrow down the search when you input less chars than the full length, like 'sh ip ro' in cisco's CLI instead of show ip route
< gmaxwell> makes adding any new command dangerous
< echeveria> seems dangerous in a system handling money.
< gmaxwell> you can use bash programmable completion so that tab works with bitcoin-cli if you really want...
< echeveria> a helpful script grabs your input and slams out your BTC into the abyss.
< echeveria> that's what you wanted, right, right?
< gmaxwell> of course the bright bulbs w/ fedora first thing they did with programmable completion was use it to make your system make network requests (and wait for them to time out) for 'missing commands' that you tried to complete.
< gmaxwell> presumably leaking god knows how many private keys to their servers...
< Sentineo> did not think of that, lol. And yeah, true.
< fanquake> jonatack Just fyi, you shouldn't have to manually add commits and authors to the 0.18 release notes, wumpus/someone will have a script to update them at the next rc.
< wumpus> right, it makes coordination really difficult if everyone updates those lists
< wumpus> (this will result in people and PRs added multiple times)
< wumpus> shouldn't be necessary as they're (initially) generated using scripts
< wumpus> gmaxwell: oh no; !!of course!! someone has to add network requests
< wumpus> gmaxwell: I do know that ubuntu does it in case of completion of filesnames for scp (which is reasonably useful, but also potentially a leak), but contacting the servers of the distribution for list of commands seems a bad idea
< gmaxwell> I think they changed it so that it doesn't leak the command at least anymore, but it does cause obvious delays still.. first thing I do on any fedora system is uninstall packagekit-command-not-found
< wumpus> reminds me of GUI input fields leaking their contents for completion/spellcheck in real time, very good for passphrases
< gmaxwell> there was some bitcoin wallet that was spellcheck leaking its key field recently
< Sentineo> lol nice
< bitcoin-git> [bitcoin] MarcoFalke pushed 4 commits to 0.18: https://github.com/bitcoin/bitcoin/compare/fa3993bfe856...e57462c6ba60
< bitcoin-git> bitcoin/0.18 7fcbe7d Pieter Wuille: Keep full pubkeys in FlatSigningProvider::origins
< bitcoin-git> bitcoin/0.18 802dcd3 Pieter Wuille: Import all origin info in importmulti; even for non-importing pubkeys
< bitcoin-git> bitcoin/0.18 235550d Pieter Wuille: Take non-importing keys into account for spendability warning in descripto...
< bitcoin-git> [bitcoin] MarcoFalke merged pull request #15803: [0.18] Backport 15749: importmulti only imports origin info for PKH outputs (0.18...201904_backport_15749) https://github.com/bitcoin/bitcoin/pull/15803
< fanquake> Are there any old wallets that definitely wouldn't open with 0.18 / current master?
< fanquake> Feel like there was an issue recently opened, but cannot find it.
< bitcoin-git> [bitcoin] fanquake opened pull request #15829: qt: update request payment button text and tab description (master...squash_rebase_14484) https://github.com/bitcoin/bitcoin/pull/15829
< bitcoin-git> [bitcoin] fanquake closed pull request #14484: changed request payment button text and tab description (master...master) https://github.com/bitcoin/bitcoin/pull/14484
< bitcoin-git> [bitcoin] merland opened pull request #15830: Improvements to test/README.md (master...patch-1) https://github.com/bitcoin/bitcoin/pull/15830
< bitcoin-git> [bitcoin] MarcoFalke opened pull request #15831: test: Add test that addmultisigaddress fails for watchonly addresses (master...1904-testMulti) https://github.com/bitcoin/bitcoin/pull/15831
< bitcoin-git> [bitcoin] MarcoFalke pushed 4 commits to master: https://github.com/bitcoin/bitcoin/compare/598323911e93...0c9de67f343c
< bitcoin-git> bitcoin/master e502c3c practicalswift: tests: Reduce noise level in test_bitcoin output
< bitcoin-git> bitcoin/master 5fd73c8 practicalswift: Avoid repeated log messages in tests by connecting to signal handlers (Thr...
< bitcoin-git> bitcoin/master 0aef39d practicalswift: Silence "Test case [...] did not check any assertions" warnings when runni...
< bitcoin-git> [bitcoin] MarcoFalke merged pull request #15352: tests: Reduce noise level in test_bitcoin output (master...test_bitcoin-output-spam) https://github.com/bitcoin/bitcoin/pull/15352
< MarcoFalke> #proposedmeetingtopic net: Hardcoded seeds update for 0.18 (#15791, #7398) (wumpus)
< gribble> https://github.com/bitcoin/bitcoin/issues/15791 | net: Hardcoded seeds update for 0.18 by laanwj · Pull Request #15791 · bitcoin/bitcoin · GitHub
< gribble> https://github.com/bitcoin/bitcoin/issues/7398 | Improve seed generation script for clarity by jgeewax · Pull Request #7398 · bitcoin/bitcoin · GitHub
< sipa> MarcoFalke: what is the minimal python version available for functional tests?
< MarcoFalke> 3.5
< sipa> dang. secrets is in 3.6
< sipa> ok
< MarcoFalke> hah, I didn't know our tests are secret
< sipa> they're not
< sipa> but it has a convenient way to generating a 256 bit random number
< sipa> surprisingly, our tests break when the ecdsa signing is made deterministic
< sipa> i suspect it's actually constructing a "forked" chain when two blocks at the same level are actually completely identical
< MarcoFalke> jup
< sipa> this is feature_block.py
< MarcoFalke> [08:35] <fanquake> Are there any old wallets that definitely wouldn't open with 0.18 / current master?
< MarcoFalke> All wallets should open
< gmaxwell> sipa: can you use PID or something that will inherently be unique in those tests?
< sipa> gmaxwell: no; a counter would work fine though :)
< bitcoin-git> [bitcoin] r8921039 opened pull request #15833: remove out-of-date comment on pay-to-witness support (master...fix_comment_ExtractDestinations_does_support_pay_to_witness) https://github.com/bitcoin/bitcoin/pull/15833
< fanquake> MarcoFalke I thought so, was just curious after reading 15821