< bitcoin-git>
[bitcoin] ariard opened pull request #17719: Document better -keypool as a look-ahead safety mechanism (master...2019-12-improve-keypool-doc) https://github.com/bitcoin/bitcoin/pull/17719
< bitcoin-git>
[bitcoin] theStack opened pull request #17720: test: add unit test for non-standard "scriptsig-not-pushonly" txs (master...20191211-test-check-for-non-standard-txs-with-non-push-scriptsig) https://github.com/bitcoin/bitcoin/pull/17720
< jonatack>
sipa: gleb: wrt to #16702, how long does the asmap buildmap.py script typically take to run? It's been at the "[INFO] Loading" step for me for some time.
< gribble>
https://github.com/bitcoin/bitcoin/issues/16702 | p2p: supplying and using asmap to improve IP bucketing in addrman by naumenkogs . Pull Request #16702 . bitcoin/bitcoin . GitHub
< wumpus>
why is so much of addrman's implementation in the header instead of the cpp?
< wumpus>
(as it's included through net.h in *many* places, this is bad for compile times, probably)
< jonatack>
gleb: confirmed, https://github.com/sipa/asmap/demo.map works as asmap file: "2019-12-11T15:01:16Z Opened asmap file (932999 bytes) from disk."
< belcher>
i think scantxoutset may have been broken in bitcoin core 0.19 ? regardless of what parameters i run it with (e.g. "scantxoutset abort", "scantxoutset status") it just returns the help doc, according to the release notes the only change was https://github.com/bitcoin/bitcoin/pull/16285/files but i dont see anything that wouldve broken it, it works fine in 0.18
< belcher>
im on regtest, in case its important
< harding>
I can confirm `scantxoutset abort` returns the help doc on latest master. Waiting for 0.18.1 to start now to attempt to reproduce there.
< harding>
It looks like it's expecting a second parameter (even though that doesn't make sense with "abort").
< jonatack>
Same for me as well
< harding>
Can also confirm that `scantxoutset abort` returns the expected result on 0.18.1.
< belcher>
im about to try on mainnet as well
< jonatack>
on mainnet
< harding>
Oh, should've mentioned my tests were on mainnet.
< achow101>
same, giving it an empty array second arg works as expected
< belcher>
confirmed on mainnet, `scantxoutset abort` gives the help doc
< achow101>
it's because of the RPCArg stuff which doesn't really let you do optional second arg depending on what the first arg is
< harding>
Simplest solution is just to add some text to the help saying that an empty second arg can be passed. scantxoutset is marked as experimental, which was supposed to make it easy for devs to make API changes, so I think it's probably ok that the 16240 change is not backwards compatible.
< achow101>
I don't think we have any RPCs that are like that
< harding>
achow101: back before named args, we had several that required you to pass an empty string in position n in order to pass an argument to position n+1.
< harding>
(I think we still have some like that if using positional arguments.)
< achow101>
that's just because of positional args
< achow101>
I don't think we have any that change whether the next arg is optional or required depending on the value of the first arg
< achow101>
besides scantxoutset
< luke-jr>
that sounds like "optional, but it's an error to omit it if you have another value"
< luke-jr>
I think maybe some of the complex RPCs have it too - you can't pass a fee estimation type without a confirm target or vice versa
< bitcoin-git>
[bitcoin] MarcoFalke merged pull request #17050: tests: Add fuzzing harnesses for functions parsing scripts, numbers, JSON and HD keypaths (bip32) (master...fuzzers) https://github.com/bitcoin/bitcoin/pull/17050
< bitcoin-git>
[bitcoin] MarcoFalke opened pull request #17726: ci: Use python 3.7 on Windows Github Actions (master...1912-ciGitHubWindows) https://github.com/bitcoin/bitcoin/pull/17726
< bitcoin-git>
[bitcoin] sipsorcery closed pull request #17709: WIP: msvc add an ignore for warning C4834 related to nodiscard attribute. (master...msvc_ignore_c4834) https://github.com/bitcoin/bitcoin/pull/17709
< fanquake>
sipsorcery if you're going to rebuild the windows qt binaries, I've for a qt related PR you might want to wait for.
< fanquake>
Will put it up tonight
< sipsorcery>
fanquake: 2 hours too late :), just uploaded new binaries
< sipsorcery>
can easily redo
< fanquake>
sipsorcery hah all good
< bitcoin-git>
[bitcoin] sipsorcery opened pull request #17727: WIP: Update msvc build to ignore warning and bump Qt version to 5.12.6. (master...msvc_qtupdate) https://github.com/bitcoin/bitcoin/pull/17727
< bitcoin-git>
[bitcoin] sipsorcery closed pull request #17710: msvc: Change appveyor build image to previous visual studio 2019. (master...appveyor_prev2019) https://github.com/bitcoin/bitcoin/pull/17710
< bitcoin-git>
[bitcoin] achow101 opened pull request #17728: rpc: require second argument only for scantxoutset start action (master...fix-scantxoutset-args) https://github.com/bitcoin/bitcoin/pull/17728