< gmaxwell>
Luke-Jr: we know why that miner was problematic, they disabled script validation.
< Luke-Jr>
gmaxwell: was that confirmed?
< Luke-Jr>
afaik it was just speculated..
< gmaxwell>
I confirmed it.
< Luke-Jr>
interesting.
< phantomcircuit>
gmaxwell, do you think it's likely that we will require anything in the coinbase scriptSig area beyond the block height for a soft fork ever again?
< phantomcircuit>
(yes i know predicting the future and what not)
< gmaxwell>
phantomcircuit: I would think it somewhat likely.
< phantomcircuit>
gmaxwell, was afraid you'd say that
< * phantomcircuit>
grubles and goes back to things
< CodeShark>
why is qa/pull-tester/rpc-tests.py always getting modified when I build?
< CodeShark>
oh, the permissions seem to have changed
< CodeShark>
old mode 100644
< CodeShark>
new mode 100755
< CodeShark>
the fix: git config core.filemode false
< GitHub76>
[bitcoin] randy-waterhouse opened pull request #6760: Changed qa/pull-tester/run-bitcoind-for-test.sh.in to non-executable mode. (master...master) https://github.com/bitcoin/bitcoin/pull/6760
< jgarzik>
CS = coding style? clang works as CS checker. Using that is The Plan - reformat the source code according to the already-chosen clang style at some flag day. Then checks from that point can be automated.
< CodeShark>
speaking of which, how are we on the C++11/C++14 stuff?
< CodeShark>
I guess backports make that very hard now
< jonasschnelli>
github merge script does use git@github.com: instead of https://github.com, wouldn't the later be our preferred way. I assume git@github.com uses plaintext communication. Sure. commits are signed, but still https might be the preferred way?
< jgarzik>
jonasschnelli, git@github is ssh
< jgarzik>
jonasschnelli, git:// is plaintext
< jonasschnelli>
jgarzik: Ah. Right.. Thanks. I missed the @. nm
< btcdrak>
jgarzik: yes, a codestyle checker - it doesnt need to be on nazi mode, just catch things like whitespace, UTF8 BOM, and basic style. Having it as part of the pull request CI (which includes Travis) makes it much easier for contributors to get it right first and makes less work for reviewers).
< CodeShark>
jonasschnelli: which one's your pride and joy right now? I'll have a look
< CodeShark>
PR that is
< jonasschnelli>
CodeShark: I keep my +1 credit because non of my PR are merge-or-review-ready state. I still like to get the bip32 patch into 0.12. Will come back to your offer soon. :)
< jonasschnelli>
*are in
< jgarzik>
btcdrak, having a "fail unless clang-CS compliant" or "auto format tree to clang-CS style" policy fixes a lot of those issues automatically
< jonasschnelli>
clang auto-format was rejected because clang-format wasn't deterministic enough (event with same version of clang-format) IIRC