In listaddressgroupings push down the IsMine check to run on each input. #1872
Conversation
|
Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/41906a3ea8322716d6391e5afde90dc81f017651 for binaries and test log. |
|
Haven't tested, but code changes look ok. |
|
Sanity testing on a testnet-in-a-box "hell wallet", I get an extra empty array: Old: New: |
|
@gavinandresen Ah. Thanks. I'm pretty sure thats just from adding an empty group after encountering a transaction where it can't add any linked ins or outs. Pushing a revised patch that I expect will fix that (though I don't have a reproduction setup right now). It might be helpful to encrypt that wallet with some impossible key and post it. |
|
It is a testnet-in-a-box wallet, so it is not sensitive. Wallet and blockchain are at: |
|
Indeed, the empty group is gone with the latest update. |
|
ACK |
|
Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/f00b335d53380e10b528d3a4e3bf663624eeb61f for binaries and test log. |
| BOOST_FOREACH(CTxOut txout, pcoin->vout) | ||
| if (IsChange(txout)) | ||
| { | ||
| CWalletTx tx = mapWallet[pcoin->vin[0].prevout.hash]; |
There was a problem hiding this comment.
Any reason to refer to vin[0] specifically?
There was a problem hiding this comment.
Or even better, what is that 'tx' variable doing there at all?
|
ACK if you remove the CWalletTx tx = .... dead line of code that sipa pointed out. |
This avoids a potential crash when trying to read the scrippubkeys on transactions where the first input IsMine but some of the rest are not when running listaddressgroupings.
|
Yea, dunno how that made it into there.. cruft from an earlier version. Sorry about that. Fixed and tested. |
|
Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/a3fad2119bcf4638ed137fc5cddc2f34a3bd1dc2 for binaries and test log. |
|
This one is lingering around for quite some time, I count 2 ACKs, any reason not to merge it or does it need a rebase? |
In listaddressgroupings push down the IsMine check to run on each input.
In listaddressgroupings push down the IsMine check to run on each input.
…killed main.{h,cpp})
34ad9d5 [Cleanup] Fix cs_main in net_processing (random-zebra)
f798b5f Rename the remaining main.{h,cpp} to validation.{h,cpp} (random-zebra)
be294cb [Validation] Move Block spam filter to net_processing (random-zebra)
520ae4b Move network-msg-processing code out of main to its own file (random-zebra)
9b894ed [Validation] save mapBlockSource in ProcessMessage (not ProcessNewBlock) (random-zebra)
2f37add Remove orphan state wipe from UnloadBlockIndex. (Matt Corallo)
Pull request description:
Finally split main.{h,cpp} file in net_processing.{h,cpp}/validation.{h,cpp} as per upstream (bitcoin#9260) 🎉
Built on top of
- [x] bitcoin#1866
ACKs for top commit:
Fuzzbawls:
ACK 34ad9d5
furszy:
ACK 34ad9d5 with a minor nit.
Tree-SHA512: 18b1457759d8684344c271236c63e1d9c83ed6e651da3efd2a7d8ab5af577fdfa85fd0f08107a26f952c3ae5da0793ebb77a2c5a710c65161f5132178db931e3
This commit fixes a crash from an out-of-range vector element access attempt that stems from the same situation with Gridcoin as in the following Bitcoin PR: bitcoin/bitcoin#1872. Very strange our code did not already have this change.
This commit fixes a crash from an out-of-range vector element access attempt that stems from the same situation with Gridcoin as in the following Bitcoin PR: bitcoin/bitcoin#1872. Very strange our code did not already have this change.
This avoids a potential crash when trying to read the scrippubkeys on
transactions where the first input IsMine but some of the rest are not
when running listaddressgroupings.