< bitcoin-git>
[bitcoin] domob1812 closed pull request #14144: Refactoring: Clarify code using encrypted_batch in CWallet (master...encrypted-batch) https://github.com/bitcoin/bitcoin/pull/14144
< bitcoin-git>
[bitcoin] jnewbery opened pull request #17477: Remove the mempool's NotifyEntryAdded and NotifyEntryRemoved signals (master...2019-11-remove-mempool-signals2) https://github.com/bitcoin/bitcoin/pull/17477
< bitcoin-git>
[bitcoin] jnewbery opened pull request #17479: Return BlockValidationState from ProcessNewBlock if CheckBlock/AcceptBlock fails (master...2019-11-processnewblock-early-return) https://github.com/bitcoin/bitcoin/pull/17479
< bitcoin-git>
[bitcoin] theStack opened pull request #17480: test: add unit test for non-standard txs with too large scriptSig (master...20191114-test-check-for-non-standard-txs-with-too-large-scriptsig) https://github.com/bitcoin/bitcoin/pull/17480
< wumpus>
and field of the CChainParams structure is used uninitialized in a computation, and this can have effect on GUI warnings
< wumpus>
so I guess it's serious enough to forego 0.19.0 and do 0.19.0.1 immediately
< MarcoFalke>
I'd say so too
< achow101>
ack
< jnewbery>
I agree :(
< wumpus>
this does give some logistical issues, I don't think we've ever skipped a major release, but doing 0.19.0.1 with the 0.19.0 release notes and releasing it as if it is 0.19.0 should work, I guess
< MarcoFalke>
Even though it seems the compiler might have compiled it correctly with -O2, not worth to rely on that
< jnewbery>
there are a group of very regular attendees who you might recognize from PR comments: lightlike, jkczyz, zenogais, fjahr, pinheadmz, amiti, ariard, ...
< jnewbery>
and we've had a few guest hosts as well: harding, MarcoFalke, ...
< jnewbery>
I still think it's a great way to help new contributors. If you're interested in guest hosting some time, please message me
< jnewbery>
Also tell your friends who want to start contributing that reviewing/testing is a great way to help, and review club could be a fun way for them to start
< wumpus>
yes I think it's a great initative to get people involved inreview!
< jnewbery>
(to clarify: I'm talking about PR review club, not taproot review club, which is another great initiative!)
< jnewbery>
that's all I had
< wumpus>
thank you
< wumpus>
I'm trying to be involved once in a while too! but usually miss the meetings :(
< wumpus>
any other topics?
< jonatack>
i discussed it a fair amount in the latest stephan livera podcast too
< jonatack>
including links to the club website and twitter account
< wumpus>
great!
< jnewbery>
thanks jonatack. I should have included you in the regular attendees list!
< jonatack>
jnewbery: np!
< wumpus>
#endmeeting
< lightningbot>
Meeting ended Thu Nov 14 19:24:50 2019 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
< bitcoin-git>
bitcoin/0.19 890dc0a Wladimir J. van der Laan: doc: Re-add release notes of 0.19.0
< bitcoin-git>
[bitcoin] ryanofsky opened pull request #17482: util: Disallow network-qualified command line options (master...pr/wdqual) https://github.com/bitcoin/bitcoin/pull/17482
< jonasschnelli>
Sorry for missing the meeting again.
< jonasschnelli>
I just saw that the Bitcoin Core Code Signing Association Apple developer programm expires in 15 days
< jeremyrubin>
is ~everyone post DST now? does it make sense to move the meeting an hour up (or is this a settled matter of sticking to non-adjusted time)
< instagibbs>
sticking to iceland time seems to be the solution
< * jeremyrubin>
misses every meeting till spring
< bitcoin-git>
[bitcoin] MarcoFalke opened pull request #17483: scripted-diff: Set gitian arch back to amd64 (master...1911-gitianRevertToAmd64) https://github.com/bitcoin/bitcoin/pull/17483
< bitcoin-git>
[bitcoin] ariard opened pull request #17484: CWallet: add cached m_is_ibd to remove isInitialBlockDownload (master...2019-11-wallet-remove-isibd) https://github.com/bitcoin/bitcoin/pull/17484
< bitcoin-git>
[bitcoin] jnewbery opened pull request #17485: net processing: Don't reach into CBlockIndex to check for block mutation (master...2019-11-processnewblock-early-return2) https://github.com/bitcoin/bitcoin/pull/17485
< sipa>
weight is the consensus level thing that needs to be exact, and we only want integer arithmetic for
< sipa>
vsize is the more familiar not as exact thing but that doesn't risk blowing up people's fees by 4, because they're doing estimation per size and then applying it to weight
< sipa>
as for non-segwit transactions, vsize == size
< darosior>
sipa: Thank you for your answer. I don't understand the trailing `3` though. For more context I've been working in making core using weight units for every feerate computation (mempool, fee estimation, coin selection, ....) and had two last broken functional test I could not get rid of (0.001% fee approximation wrong). Turned out they were using
< darosior>
decoderawtransaction which uses this same formula but hardcoded. Using the BIP formula makes the fee estimation match.
< sipa>
darosior: for fee estimarion purposes rounding up is less harmful than rounding down
< darosior>
Ok, thanks
< darosior>
Then getting rid of that if we don't round up anymore for fee estimation seems ok ? :)