re-interpreted max_leaf_position #1493
Conversation
|
do we need to understand a little better what is going on here? i see this works but i dont know why |
witness_at_checkpoint_depth calls (1, 2) max_leaf_position with the supplied checkpoint depth, and the semantics of checkpoint_depth == zero has changed from meaning "the latest version of the tree" to "the top checkpoint". (Previously it was 1-indexed with zero having special meaning. Now it's zero-indexed with None being the special case of "don't look at the checkpoints at all, use the latest tree". witness_at_checkpoint_depth keeps the move from 1-indexed to 0-indexed, but instead of accepting an option, simply no longer supports the 'use latest tree' option..I'm guessing that option is a footgun in some manner, although I don't believe we were ever shot with it) |
resolves #1491
caused by a mistake i made as part of #1482. really good the test caught it or it could have wreaked havoc upon the shardtrees