如何在Python中实现通用树?这有内置的数据结构吗?


当前回答

我将根树实现为字典{child:parent}。比如根节点为0,树可能是这样的:

tree={1:0, 2:0, 3:1, 4:2, 5:3}

这种结构使得沿着一条路径从任意节点向上到根结点非常容易,这与我正在处理的问题有关。

其他回答

我已经在我的网站https://web.archive.org/web/20120723175438/www.quesucede.com/page/show/id/python_3_tree_implementation上发布了一个Python 3树的实现

代码如下:

import uuid

def sanitize_id(id):
    return id.strip().replace(" ", "")

(_ADD, _DELETE, _INSERT) = range(3)
(_ROOT, _DEPTH, _WIDTH) = range(3)

class Node:

    def __init__(self, name, identifier=None, expanded=True):
        self.__identifier = (str(uuid.uuid1()) if identifier is None else
                sanitize_id(str(identifier)))
        self.name = name
        self.expanded = expanded
        self.__bpointer = None
        self.__fpointer = []

    @property
    def identifier(self):
        return self.__identifier

    @property
    def bpointer(self):
        return self.__bpointer

    @bpointer.setter
    def bpointer(self, value):
        if value is not None:
            self.__bpointer = sanitize_id(value)

    @property
    def fpointer(self):
        return self.__fpointer

    def update_fpointer(self, identifier, mode=_ADD):
        if mode is _ADD:
            self.__fpointer.append(sanitize_id(identifier))
        elif mode is _DELETE:
            self.__fpointer.remove(sanitize_id(identifier))
        elif mode is _INSERT:
            self.__fpointer = [sanitize_id(identifier)]

class Tree:

    def __init__(self):
        self.nodes = []

    def get_index(self, position):
        for index, node in enumerate(self.nodes):
            if node.identifier == position:
                break
        return index

    def create_node(self, name, identifier=None, parent=None):

        node = Node(name, identifier)
        self.nodes.append(node)
        self.__update_fpointer(parent, node.identifier, _ADD)
        node.bpointer = parent
        return node

    def show(self, position, level=_ROOT):
        queue = self[position].fpointer
        if level == _ROOT:
            print("{0} [{1}]".format(self[position].name,
                                     self[position].identifier))
        else:
            print("\t"*level, "{0} [{1}]".format(self[position].name,
                                                 self[position].identifier))
        if self[position].expanded:
            level += 1
            for element in queue:
                self.show(element, level)  # recursive call

    def expand_tree(self, position, mode=_DEPTH):
        # Python generator. Loosly based on an algorithm from 'Essential LISP' by
        # John R. Anderson, Albert T. Corbett, and Brian J. Reiser, page 239-241
        yield position
        queue = self[position].fpointer
        while queue:
            yield queue[0]
            expansion = self[queue[0]].fpointer
            if mode is _DEPTH:
                queue = expansion + queue[1:]  # depth-first
            elif mode is _WIDTH:
                queue = queue[1:] + expansion  # width-first

    def is_branch(self, position):
        return self[position].fpointer

    def __update_fpointer(self, position, identifier, mode):
        if position is None:
            return
        else:
            self[position].update_fpointer(identifier, mode)

    def __update_bpointer(self, position, identifier):
        self[position].bpointer = identifier

    def __getitem__(self, key):
        return self.nodes[self.get_index(key)]

    def __setitem__(self, key, item):
        self.nodes[self.get_index(key)] = item

    def __len__(self):
        return len(self.nodes)

    def __contains__(self, identifier):
        return [node.identifier for node in self.nodes
                if node.identifier is identifier]

if __name__ == "__main__":

    tree = Tree()
    tree.create_node("Harry", "harry")  # root node
    tree.create_node("Jane", "jane", parent = "harry")
    tree.create_node("Bill", "bill", parent = "harry")
    tree.create_node("Joe", "joe", parent = "jane")
    tree.create_node("Diane", "diane", parent = "jane")
    tree.create_node("George", "george", parent = "diane")
    tree.create_node("Mary", "mary", parent = "diane")
    tree.create_node("Jill", "jill", parent = "george")
    tree.create_node("Carol", "carol", parent = "jill")
    tree.create_node("Grace", "grace", parent = "bill")
    tree.create_node("Mark", "mark", parent = "jane")

    print("="*80)
    tree.show("harry")
    print("="*80)
    for node in tree.expand_tree("harry", mode=_WIDTH):
        print(node)
    print("="*80)

Greg Hewgill的回答很好,但如果你每层需要更多的节点,你可以使用列表|字典来创建它们:然后使用方法按名称或顺序(如id)访问它们。

class node(object):
    def __init__(self):
        self.name=None
        self.node=[]
        self.otherInfo = None
        self.prev=None
    def nex(self,child):
        "Gets a node by number"
        return self.node[child]
    def prev(self):
        return self.prev
    def goto(self,data):
        "Gets the node by name"
        for child in range(0,len(self.node)):
            if(self.node[child].name==data):
                return self.node[child]
    def add(self):
        node1=node()
        self.node.append(node1)
        node1.prev=self
        return node1

现在只需创建一个根并建立它: 例:

tree=node()  #create a node
tree.name="root" #name it root
tree.otherInfo="blue" #or what ever 
tree=tree.add() #add a node to the root
tree.name="node1" #name it

    root
   /
child1

tree=tree.add()
tree.name="grandchild1"

       root
      /
   child1
   /
grandchild1

tree=tree.prev()
tree=tree.add()
tree.name="gchild2"

          root
           /
        child1
        /    \
grandchild1 gchild2

tree=tree.prev()
tree=tree.prev()
tree=tree.add()
tree=tree.name="child2"

              root
             /   \
        child1  child2
       /     \
grandchild1 gchild2


tree=tree.prev()
tree=tree.goto("child1") or tree=tree.nex(0)
tree.name="changed"

              root
              /   \
         changed   child2
        /      \
  grandchild1  gchild2

这应该足够让你开始思考如何让它工作了

嗨,你可以试试itertree(我是作者)。

该包与任何树包的方向相同,但关注点略有不同。在巨大的树(>100000个项目)上的性能要好得多,它处理迭代器具有有效的过滤机制。

>>>from itertree import *
>>>root=iTree('root')

>>># add some children:
>>>root.append(iTree('Africa',data={'surface':30200000,'inhabitants':1257000000}))
>>>root.append(iTree('Asia', data={'surface': 44600000, 'inhabitants': 4000000000}))
>>>root.append(iTree('America', data={'surface': 42549000, 'inhabitants': 1009000000}))
>>>root.append(iTree('Australia&Oceania', data={'surface': 8600000, 'inhabitants': 36000000}))
>>>root.append(iTree('Europe', data={'surface': 10523000 , 'inhabitants': 746000000}))
>>># you might use __iadd__ operator for adding too:
>>>root+=iTree('Antarktika', data={'surface': 14000000, 'inhabitants': 1100})

>>># for building next level we select per index:
>>>root[0]+=iTree('Ghana',data={'surface':238537,'inhabitants':30950000})
>>>root[0]+=iTree('Niger', data={'surface': 1267000, 'inhabitants': 23300000})
>>>root[1]+=iTree('China', data={'surface': 9596961, 'inhabitants': 1411780000})
>>>root[1]+=iTree('India', data={'surface': 3287263, 'inhabitants': 1380004000})
>>>root[2]+=iTree('Canada', data={'type': 'country', 'surface': 9984670, 'inhabitants': 38008005})    
>>>root[2]+=iTree('Mexico', data={'surface': 1972550, 'inhabitants': 127600000 })
>>># extend multiple items:
>>>root[3].extend([iTree('Australia', data={'surface': 7688287, 'inhabitants': 25700000 }), iTree('New Zealand', data={'surface': 269652, 'inhabitants': 4900000 })])
>>>root[4]+=iTree('France', data={'surface': 632733, 'inhabitants': 67400000 }))
>>># select parent per TagIdx - remember in itertree you might put items with same tag multiple times:
>>>root[TagIdx('Europe'0)]+=iTree('Finland', data={'surface': 338465, 'inhabitants': 5536146 })

创建的树可以被渲染:

>>>root.render()
iTree('root')
     └──iTree('Africa', data=iTData({'surface': 30200000, 'inhabitants': 1257000000}))
         └──iTree('Ghana', data=iTData({'surface': 238537, 'inhabitants': 30950000}))
         └──iTree('Niger', data=iTData({'surface': 1267000, 'inhabitants': 23300000}))
     └──iTree('Asia', data=iTData({'surface': 44600000, 'inhabitants': 4000000000}))
         └──iTree('China', data=iTData({'surface': 9596961,  'inhabitants': 1411780000}))
         └──iTree('India', data=iTData({'surface': 3287263, 'inhabitants': 1380004000}))
     └──iTree('America', data=iTData({'surface': 42549000, 'inhabitants': 1009000000}))
         └──iTree('Canada', data=iTData({'surface': 9984670, 'inhabitants': 38008005}))
         └──iTree('Mexico', data=iTData({'surface': 1972550, 'inhabitants': 127600000}))
     └──iTree('Australia&Oceania', data=iTData({'surface': 8600000, 'inhabitants': 36000000}))
         └──iTree('Australia', data=iTData({'surface': 7688287, 'inhabitants': 25700000}))
         └──iTree('New Zealand', data=iTData({'surface': 269652, 'inhabitants': 4900000}))
     └──iTree('Europe', data=iTData({'surface': 10523000, 'inhabitants': 746000000}))
         └──iTree('France', data=iTData({'surface': 632733, 'inhabitants': 67400000}))
         └──iTree('Finland', data=iTData({'surface': 338465, 'inhabitants': 5536146}))
     └──iTree('Antarktika', data=iTData({'surface': 14000000, 'inhabitants': 1100}))

过滤可以这样做:

>>>item_filter = Filter.iTFilterData(data_key='inhabitants', data_value=iTInterval(0, 20000000))
>>>iterator=root.iter_all(item_filter=item_filter)
>>>for i in iterator:
>>>    print(i)
iTree("'New Zealand'", data=iTData({'surface': 269652, 'inhabitants': 4900000}), subtree=[])
iTree("'Finland'", data=iTData({'surface': 338465, 'inhabitants': 5536146}), subtree=[])
iTree("'Antarktika'", data=iTData({'surface': 14000000, 'inhabitants': 1100}), subtree=[])

我使用嵌套字典实现了树。这很容易做到,而且对我来说,它在相当大的数据集上很有效。我在下面发布了一个示例,你可以在谷歌代码中看到更多

  def addBallotToTree(self, tree, ballotIndex, ballot=""):
    """Add one ballot to the tree.

    The root of the tree is a dictionary that has as keys the indicies of all 
    continuing and winning candidates.  For each candidate, the value is also
    a dictionary, and the keys of that dictionary include "n" and "bi".
    tree[c]["n"] is the number of ballots that rank candidate c first.
    tree[c]["bi"] is a list of ballot indices where the ballots rank c first.

    If candidate c is a winning candidate, then that portion of the tree is
    expanded to indicate the breakdown of the subsequently ranked candidates.
    In this situation, additional keys are added to the tree[c] dictionary
    corresponding to subsequently ranked candidates.
    tree[c]["n"] is the number of ballots that rank candidate c first.
    tree[c]["bi"] is a list of ballot indices where the ballots rank c first.
    tree[c][d]["n"] is the number of ballots that rank c first and d second.
    tree[c][d]["bi"] is a list of the corresponding ballot indices.

    Where the second ranked candidates is also a winner, then the tree is 
    expanded to the next level.  

    Losing candidates are ignored and treated as if they do not appear on the 
    ballots.  For example, tree[c][d]["n"] is the total number of ballots
    where candidate c is the first non-losing candidate, c is a winner, and
    d is the next non-losing candidate.  This will include the following
    ballots, where x represents a losing candidate:
    [c d]
    [x c d]
    [c x d]
    [x c x x d]

    During the count, the tree is dynamically updated as candidates change
    their status.  The parameter "tree" to this method may be the root of the
    tree or may be a sub-tree.
    """

    if ballot == "":
      # Add the complete ballot to the tree
      weight, ballot = self.b.getWeightedBallot(ballotIndex)
    else:
      # When ballot is not "", we are adding a truncated ballot to the tree,
      # because a higher-ranked candidate is a winner.
      weight = self.b.getWeight(ballotIndex)

    # Get the top choice among candidates still in the running
    # Note that we can't use Ballots.getTopChoiceFromWeightedBallot since
    # we are looking for the top choice over a truncated ballot.
    for c in ballot:
      if c in self.continuing | self.winners:
        break # c is the top choice so stop
    else:
      c = None # no candidates left on this ballot

    if c is None:
      # This will happen if the ballot contains only winning and losing
      # candidates.  The ballot index will not need to be transferred
      # again so it can be thrown away.
      return

    # Create space if necessary.
    if not tree.has_key(c):
      tree[c] = {}
      tree[c]["n"] = 0
      tree[c]["bi"] = []

    tree[c]["n"] += weight

    if c in self.winners:
      # Because candidate is a winner, a portion of the ballot goes to
      # the next candidate.  Pass on a truncated ballot so that the same
      # candidate doesn't get counted twice.
      i = ballot.index(c)
      ballot2 = ballot[i+1:]
      self.addBallotToTree(tree[c], ballotIndex, ballot2)
    else:
      # Candidate is in continuing so we stop here.
      tree[c]["bi"].append(ballotIndex)

并没有内置树,但是可以通过从List继承Node类型并编写遍历方法来轻松地构造一个树。如果你这样做,我发现平分法很有用。

您还可以浏览PyPi上的许多实现。

如果我没记错的话,Python标准库不包含树数据结构,原因和。net基类库不包含树数据结构是一样的:内存的局部性降低了,导致缓存丢失更多。在现代处理器上,将大量内存放入缓存通常会更快,而“指针丰富”的数据结构会抵消这种好处。