使用anytree从具有缩进的文件生成树

Question

我正在使用anytree从文件Tree.txt生成一棵树。每个缩进是3个空格（）。以下是我的文件的外观：

ROOT
   Node1
      Node2
         Node3
            Node4
   Node5
   Node6

以下是我到目前为止对生成代码的了解：

from anytree import Node, RenderTree, find_by_attr

def parse_tree():
    with open('Tree.txt', 'r') as file:
        file_contents = file.readlines()
        root = Node("ROOT")

        current_indent = -1 # store index for comparing later
        for (index, line) in enumerate(file_contents):

            leading_spaces = len(line) - len(line.lstrip(' '))
            indent = int(leading_spaces / 3) # indent = 3 spaces
            
            if indent > current_indent: # node must be 1 level down
                
                # get the previous Node (1 level up)
                previous_line = file_contents[index - 1].strip()
                current_line = line.strip()
                parent = find_by_attr(root, previous_line)
                Node(current_line, parent=parent) # start searching from top node (root) for `previous_line`
                current_indent = indent
            else: # node is on same level or higher
                print(f"What to do for {line.strip()}?") 
                # what should I do here?

        print(RenderTree(root)) # print the output

parse_tree()

但是，这会打印出：

What to do for Node5?
What to do for Node6?
Node('/ROOT')
└── Node('/ROOT/Node1')
    └── Node('/ROOT/Node1/Node2')
        └── Node('/ROOT/Node1/Node2/Node3')
            └── Node('/ROOT/Node1/Node2/Node3/Node4')

生成的树很好，但直到缺少Node4-Node5和Node6为止。这是意料之中的，因为我没有处理当前缩进小于前一缩进的情况（请参见^{）

当当前缩进小于上一缩进时，我应该怎么做？我知道我需要提高n级别，但我如何确定这是哪个级别

Answer 1

需要一种堆栈来跟踪节点的深度，然后可以使用初始空间的数量来确定将每个叶附加到哪个父节点。例如，三个空格意味着向下一级，因此需要附加到堆栈中索引0处的根节点。（你也可以说，通过使用一个堆栈，我们得到了树叶的绝对深度，而你试图解决它相对，如果我说得通的话）。我怀疑如果您使用find_by_attr，在有同名叶子的情况下，这可能不起作用

treestring = '''
ROOT
   Node1
      Node2
         Node3
            Node4
   Node5
   Node6'''.strip()

leaves = treestring.splitlines()

# initialize stack with the root node
stack = {0: Node(leaves.pop(0))}

for leaf in leaves:
    
    # determine the node's depth
    leading_spaces = len(leaf)-len(leaf.lstrip(' '))
    level = int(leading_spaces/3)
    
    # add the node to the stack, set as parent the node that's one level up
    stack[level] = Node(leaf.strip(), parent=stack[level-1])

tree = stack[0]

for pre, _, node in RenderTree(tree):
    print(f"{pre}{node.name}")

结果:

ROOT
├── Node1
│   └── Node2
│       └── Node3
│           └── Node4
├── Node5
└── Node6