The big binary tree traverses all the descendant nodes under a node?

A large binary tree with uncertain tree hierarchy. Now we need to realize the following requirements:
1. Record hierarchical relationships. For example: to give you a node id:9, needs to find out the son or grandson under this node, or. Or all descendant nodes
2. I would also like to query the statistical requirements

.

for requirement 1, I am talking about the node id stored in redis, and the structure is that zset, key is node id,score is hierarchical relationship. If it is a son node, score is-1 and grandson is-2. The levels are infinite, and there may be many layers. The later the score is, the smaller it is.

but there is a problem with this storage, that is, it is not easy to query statistical requirements. For example, I need to query all the new nodes under a node for the day. For this kind of demand, at present, all I can think of is to get the id, of all the descendant nodes under the current node from redis, and then go to the database through id to find out the new nodes of the day through in Filter. This has a problem, that is, if there are too many descendant nodes, the efficiency of in query is very low, and the service will fail.

I wonder if anyone has a better plan. Neo4j, is currently being studied as a graphic database. I wonder if such a requirement can be achieved. Has a great god ever used it?

Python php java golang
Mar.23,2021

your requirements see no reason to use redis. You can simply use a relational database to implement a tree-level relationship and use a field to save the parent-child path relationship. For example, root 1, which has two nodes, 1-2 talks, 1-3, and 2, has a son node, 1-2-4 talk, and 1-3-5, so statistics can easily get all descendants of this node with a simple like. For example, query path like'1% colors, that is, all descendants of id=1, path like'1-2, that is, all descendants of id=2.

there's nothing wrong with the way you store binary trees. You said that you got the id, of all the descendant nodes from redis and then went to the database to use the new nodes of in Filter that day. Is that so? Select * from tab where id in (ids) and date = sysdate;, that is, if there are many descendant nodes, the ids will be very large. I don't know if that's what you want. But both id and score of your existing redis can be indexed to a relational database (id is equivalent to the primary key, distinct (score) up to 2), so there is no need to use redis. Then hash, the query conditions that often appear in the business, such as this date. If you add a lot of new data every day, you can even make a sub-table.

Previous: Windows server 2012, use nginx for load balancing, session cannot be shared

Next: 403 error occurred in win server2008+IIS7 installation django environment

css mysql arrays josn react html typescript webpack npm sass R objective-c .net sql-server jquery python-3.x angularjs django angular excel regex iphone ajax linux xml pandas vba spring database wordpress string wpf xcode windows bash postgresql oracle multithreading eclipse list firebase algorithm macos forms image scala visual-studio azure bootstrap spring-boot react-native python-2.7 docker performance function winforms matlab powershell apache dataframe api sqlite numpy rest shell selenium flutter dart maven loops qt swing android-studio csv express file class tensorflow sorting codeigniter perl
© 2021 CodesHelper
Advertising Contact us About us
Menu