PST 文件中使用的这个“BTree”变体是什么？

Question

I've been working on parsing PST files with the help of https://www.five-ten-sg.com/libpst/rn01re06.html and https://blogs.msdn.microsoft.com/openspecification/2010/11/30/ms-pst-how-to-navigate-the-node-btree/ where they refer to it as a BTree, but it doesn't seem to follow the rules I can find for regular BTrees.

I've mapped it so I can show it visually

Basically its made up of nodes that contain children (up to 20) and those are the same structure and so on until the leaf nodes, which are different, but the internal nodes aren't used for any actual data storing and seem to (sometimes) share the keys of the the first node in their children.

I'm asking because I'm trying to find out common ways of inserting, deleting, searching this data but regular BTree algorithms like on https://www.geeksforgeeks.org/b-tree-set-1-introduction-2/ aren't working since this data doesn't seem to be structured in exactly the same way.

I've modified the search algorithm to traverse the current children in reverse until I find a child with an ID less than or equal to the one I'm searching for, and then look at that nodes children, which is working great but I'd rather not reinvent the wheel here and miss things if I don't have to.

Answer 1

Here is a different display which I think shows more clearly that this is a standard B-Tree. The B-Tree index (Node or Block index) in the intermediate node is the index of the lowest leaf under the node; a Node or Block index can hence be found easily. (Disclaimer - this image was generated from a tool I developed .)