With an encrypted wallet the GUI was prompting for a passphrase every time the user requested a new address. This is unnecessary, increases the exposure to keyboard sniffers, and discourages using fresh addresses for every transaction.
Instead only prompt for a passphrase when the keypool runs out, also call the new address function with the flag that prevents reuse.
So, the design choices here were to reuse or prompt when the keypool ran dry. Reuse has the advantage that the behavior would never change even when the pool empties, but it would reuse a previously assigned address, which is really surprising. There would also be an option to reuse after prompting if the wallet didn't unlock.
Codeshark, said prompt on empty. I agree. If someone wants to reuse when they don't have the key— the list is there.
I tested this by starting up with an encrypted wallet and hitting new address until the keypool ran out and it promoted me. I dismissed the dialog and retried a few times before finally letting it unlock and confirming it worked as expected.
Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/0eb506a01a32cc78609ee9071b0d5302488496ec for binaries and test log. This test script verifies pulls every time they are updated. It, however, dies sometimes and fails to test properly. If you are waiting on a test, please check timestamps to verify that the test.log is moving at http://jenkins.bluematt.me/pull-tester/current/ Contact BlueMatt on freenode if something looks broken.
I took a quick look at GetKeyFromPool() and AFAIK we want to get rid of that default address/kay anyway, right? That pull could also remove the fAllowReuse parameter from that function, as it's false after this pull everywhere in the code.
Code looks good and does what it sais, no need to enter the passphrase when requesting a new receiving address.
AFAIK GetKeyFromPool() is now true at least once after paymentrequest pull was merged. Can you guys have a look into if this makes sense.
With an encrypted wallet the GUI was prompting for a passphrase every time
the user requested a new address. This is unnecessary, increases the
exposure to keyboard sniffers, and discourages using fresh addresses for
every transaction.
Instead only prompt for a passphrase when the keypool runs out, also call
the new address function with the flag that prevents reuse.
Thanks to AlexNagy on IRC for pointing this out and who wouldn't take any
lip from a curmudgeonly developer and insisted on what he knew to be true.
With the GUI password fix this was always false.
fAllowReuse gone and rebased post paymentrequests so I could remove the flag there too.
Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/71ac5052d83fcba21a09e5e2b7ad66faea6bd42a for binaries and test log. This test script verifies pulls every time they are updated. It, however, dies sometimes and fails to test properly. If you are waiting on a test, please check timestamps to verify that the test.log is moving at http://jenkins.bluematt.me/pull-tester/current/ Contact BlueMatt on freenode if something looks broken.
ACK
ACK core changes.