B tree is a highly balanced binary search tree. To perform an insertion operation, you must first obtain the position of the inserted node. Follow the node to be larger than the left subtree and smaller than the right subtree, and split the node when necessary.
<code>BreeInsertion(T, k)r root[T]if n[r] = 2t - 1<br/> s = AllocateNode()<br/> root[T] = s<br/> leaf[s] = FALSE<br/> n[s] <- 0<br/> c1[s] <- r<br/> BtreeSplitChild(s, 1, r)<br/> BtreeInsertNonFull(s, k)else BtreeInsertNonFull(r, k)BtreeInsertNonFull(x, k)i = n[x]if leaf[x]<br/> while i ≥ 1 and k < keyi[x]<br/> keyi+1 [x] = keyi[x]<br/> i = i - 1<br/> keyi+1[x] = k<br/> n[x] = n[x] + 1else while i ≥ 1 and k < keyi[x]<br/> i = i - 1<br/> i = i + 1<br/> if n[ci[x]] == 2t - 1<br/> BtreeSplitChild(x, i, ci[x])<br/> if k &rt; keyi[x]<br/> i = i + 1<br/> BtreeInsertNonFull(ci[x], k)BtreeSplitChild(x, i)BtreeSplitChild(x, i, y)z = AllocateNode()leaf[z] = leaf[y]n[z] = t - 1for j = 1 to t - 1<br/> keyj[z] = keyj+t[y]if not leaf [y]<br/> for j = 1 to t<br/> cj[z] = cj + t[y]n[y] = t - 1for j = n[x] + 1 to i + 1<br/> cj+1[x] = cj[x]ci+1[x] = zfor j = n[x] to i<br/> keyj+1[x] = keyj[x]keyi[x] = keyt[y]n[x] = n[x] + 1</code>
<code>class BTreeNode:<br/> def __init__(self, leaf=False):<br/> self.leaf = leaf<br/> self.keys = []<br/> self.child = []<br/> <br/>class BTree:<br/> def __init__(self, t):<br/> self.root = BTreeNode(True)<br/> self.t = t<br/> <br/> def insert(self, k):<br/> root = self.root<br/> if len(root.keys) == (2 * self.t) - 1:<br/> temp = BTreeNode()<br/> self.root = temp<br/> temp.child.insert(0, root)<br/> self.split_child(temp, 0)<br/> self.insert_non_full(temp, k)<br/> else:<br/> self.insert_non_full(root, k)<br/> <br/> def insert_non_full(self, x, k):<br/> i = len(x.keys) - 1<br/> if x.leaf:<br/> x.keys.append((None, None))<br/> while i >= 0 and k[0] < x.keys[i][0]:<br/> x.keys[i + 1] = x.keys[i]<br/> i -= 1<br/> x.keys[i + 1] = k<br/> else:<br/> while i >= 0 and k[0] < x.keys[i][0]:<br/> i -= 1<br/> i += 1<br/> if len(x.child[i].keys) == (2 * self.t) - 1:<br/> self.split_child(x, i)<br/> if k[0] > x.keys[i][0]:<br/> i += 1<br/> self.insert_non_full(x.child[i], k)<br/> <br/> def split_child(self, x, i):<br/> t = self.t<br/> y = x.child[i]<br/> z = BTreeNode(y.leaf)<br/> x.child.insert(i + 1, z)<br/> x.keys.insert(i, y.keys[t - 1])<br/> z.keys = y.keys[t: (2 * t) - 1]<br/> y.keys = y.keys[0: t - 1]<br/> if not y.leaf:<br/> z.child = y.child[t: 2 * t]<br/> y.child = y.child[0: t - 1]<br/> <br/> def print_tree(self, x, l=0):<br/> print("Level ", l, " ", len(x.keys), end=":")<br/> for i in x.keys:<br/> print(i, end=" ")<br/> print()<br/> l += 1<br/> if len(x.child) > 0:<br/> for i in x.child:<br/> self.print_tree(i, l)<br/> <br/>def main():<br/> B = BTree(3)<br/> <br/> for i in range(10):<br/> B.insert((i, 2 * i))<br/> <br/> B.print_tree(B.root)<br/> <br/>if __name__ == '__main__':<br/> main()</code>
The above is the detailed content of Principle diagram of Python implementation of B-tree insertion algorithm. For more information, please follow other related articles on the PHP Chinese website!