Backports 0.15 pr2 #2597
Backports 0.15 pr2 #2597
Conversation
|
Looks good 👍 |
PastaPastaPasta
force-pushed
the
backports-0.15-pr2
branch
from
952ca76 to
da1b0dc
Compare
|
Needs rebase |
PastaPastaPasta
force-pushed
the
backports-0.15-pr2
branch
from
da1b0dc to
d780aa2
Compare
|
Rebased |
a87d02a use EXIT_ codes instead of magic numbers (Marko Bencun)
…seTransaction eaea2bb Removed redundant parameter from mempool.PrioritiseTransaction (gubatron)
bc8fd12 Remove harmless read of unusued priority estimates (Alex Morcos)
c578408 Add exclude option to rpc-tests.py (John Newbery)
3f95a80 Fix docstrings in qa tests (John Newbery)
3333ad0 qa: Set correct path for binaries in rpc tests (MarcoFalke)
ef9f495 Trivial: fix comments referencing AppInit2 (Marko Bencun)
dc222f8 Trivial: Rephrase the definition of difficulty in the code. (Karl-Johan Alm)
30aedcb BIP32 extra test vector (Pieter Wuille)
864890a [qa] Make import-rescan.py watchonly check reliable (Russell Yanofsky) Tree-SHA512: ea0e2b1d4fc8f35174c3d575fb751b428daf6ad3aa944fad4e3ddcc9195e4f17051473acabc54203b1d27cca64cf911b737ab92e986c40ef384410652e2dbea1
PastaPastaPasta
force-pushed
the
backports-0.15-pr2
branch
from
d780aa2 to
062413c
Compare
|
FYI: the "merge" commit above is me trying Github GUI conflict resolution tool. |
codablock
left a comment
codablock
left a comment
There was a problem hiding this comment.
utACK, but I'm not sure I like the conflict resolution via merge conflict, as it creates funny commit histories. I always prefer to do a rebase. At the same time, I assume we want to avoid too much repeated reviews after rebases touch all commits and invalidate old reviews?
|
I'm squash&merging this, so history won't be funny in the main ( |
|
Hmm...Preserving history (non-squashed merges) when doing backports is of high value IMHO as it allows fine grained blaming/annotating to figure which change was introduced by which PR and especially why it was introduced. |
|
Yep, it makes sense for huge non-trivial backports (like you made the last time :P) but the last two by Pasta were rather small and/or trivial ones imo. |
|
I mean, changes in some critical parts like networking (e.g. compact blocks) or chain verification should probably be merged with no squashing, changes in comments, cmd-line params etc are fine to be merged as one singe commit IMO. |
|
Imo history should be preserved for trivial backports too. Makes it a lot easier to check the history and see if a certain pr was already backported |
4 participants
Will be rebased immediately after pr1 is merged