EduCoder_Study_RS/study_path_rs/initial_learning.py

# 导入需要的工具包
from py2neo import Graph, Node, Relationship, NodeMatcher, Subgraph
import pandas as pd
import numpy as np
import os
import sys
sys.path.append(os.getcwd())
import json
from config import neo4j_url,neo4j_username,neo4j_password
from pprint import pprint as pp

# 连接Neo4j数据库
graph = Graph(neo4j_url, auth=(neo4j_username,neo4j_password) )

def get_initial_path(course_id, platform):
  
  course_query = "match(c:`教学课堂`) where c.course_id='{}' return c.course_id".format(course_id)
  # 查询course_id信息，获取知识森林中课堂信息
  course_result = graph.run(course_query).data()
  
  # 判断course_id是否存在，如果存在，进行如下查询操作，返回查询结果；如不存在，返回空值
  if course_result:
    
    # 知识森林查询实训相关节点
    shixun_query = '''match(c:`教学课堂`) where c.course_id="{}" match(c)-[:`课堂课程`]->(s:`实践课程`)
    match(s)-[:`课程章节`]->(st:`章节`) match(st)-[:`章节知识点`]->(kn:`知识点`) 
    match(kn)-[:`知识点实训`]->(shixun:`实训`) return st.stage_id as stage_id,st.stage_name as stage_name,
    st.stage_sort	as stage_sort,kn.knowledge_id as knowledge_id,kn.knowledge as knowledge,
    shixun.shixun_id as shixun_id,shixun.shixun_name as shixun_name,shixun.link as shixun_link,
    count(shixun.shixun_id) as shixun_num'''.format(course_id)
    
    # 获取实训相关信息，若不存在，为统一处理，返回空DataFrame
    shixun = graph.run(shixun_query).data()

    if shixun:
      # 将查询结果转化为dataframe格式
      shixun_data = pd.DataFrame(shixun) 
    else:
      # 实训结果不存在，赋予一个相同结构的空值
      shixun_data = pd.DataFrame(columns=["stage_id","stage_name","stage_sort","knowledge_id",
                                          "knowledge","shixun_id","shixun_name","shixun_link","shixun_num"])

    # 知识森林查询教学视频相关节点
    video_query = '''match(c:`教学课堂`) where c.course_id="{}" match(c)-[:`课堂课程`]->(s:`实践课程`) 
    match(s)-[:`课程章节`]->(st:`章节`) match(st)-[:`章节知识点`]->(kn:`知识点`) 
    match(kn)-[:`知识点视频`]->(video:`教学视频`) return st.stage_id as stage_id, st.stage_name as stage_name,
    st.stage_sort	as stage_sort,kn.knowledge_id as knowledge_id,kn.knowledge as knowledge,
    video.video_item_id as video_id,video.video_name as video_name,video.link as video_link,
    count(video.video_item_id) as video_num'''.format(course_id)

    # 获取教学视频相关信息，若不存在，为统一处理，返回空DataFrame  
    video = graph.run(video_query).data()

    # 判断获取的教学视频是否为空  
    if video:
      # 将教学视频的查询结果转化为dataframe格式
      video_data = pd.DataFrame(video) 
    else:
      # 教学视频为空，返回相同数据结构的空值
      video_data = pd.DataFrame(columns=["stage_id","stage_name","stage_sort","knowledge_id",
                                         "knowledge","video_id","video_name","video_link","video_num"])

    
    #  知识森林查询教学课件相关节点
    att_query = '''match(c:`教学课堂`) where c.course_id="{}" match(c)-[:`课堂课程`]->(s:`实践课程`) 
    match(s)-[:`课程章节`]->(st:`章节`) match(st)-[:`章节知识点`]->(kn:`知识点`) 
    match(kn)-[:`知识点课件`]->(att:`课件`) return st.stage_id as stage_id,st.stage_name as stage_name,
    st.stage_sort as stage_sort,kn.knowledge_id as knowledge_id,kn.knowledge as knowledge ,
    att.attachment_id as attachment_id,att.filename as filename,att.link as attachment_link,
    count(att.attachment_id) as attachment_num'''.format(course_id)
    
    #获取教学课件相关信息，若不存在，为统一处理，返回空DataFrame
    att = graph.run(att_query).data()

    if att:
      # 将教学课件的查询结果转换成dataframe格式
      att_data = pd.DataFrame(att)
    else:
      # 教学课件为空，返回相同数据结构的空值
      att_data = pd.DataFrame(columns=["stage_id","stage_name","stage_sort","knowledge_id","knowledge",
                                       "attachment_id","filename","attachment_link"])

    #将三种教学资源合并  
    merged_data = pd.merge(shixun_data,video_data,on=['stage_id','stage_name',"stage_sort",'knowledge_id',
                                                      'knowledge'],how='outer')
    merged_data = pd.merge(merged_data,att_data,on=['stage_id','stage_name',"stage_sort",'knowledge_id',
                                                    'knowledge'],how='outer')
    # 默认为头歌地址，添加ilearning地址                                        
    if platform == '1':
      merged_data['attachment_link'] = merged_data['attachment_link'].str.replace('www.', 'ilearning.')
      merged_data['shixun_link'] = merged_data['shixun_link'].str.replace('www.', 'ilearning.')
      merged_data['video_link'] = merged_data['video_link'].str.replace('www.', 'ilearning.')

    #对于合并后的结果，对教学资源的数量进行填充
    merged_data['shixun_num'].fillna(0,inplace=True)
    merged_data['video_num'].fillna(0,inplace=True)
    merged_data['attachment_num'].fillna(0,inplace=True)
    
    # 分组获取实训字典
    shixun = (merged_data.groupby(['stage_id','stage_name',"stage_sort",'knowledge_id','knowledge',
                                   'shixun_num','video_num','attachment_num'])
    .apply(lambda x:x[['shixun_id','shixun_name', 'shixun_link']].to_dict('r'))
    .reset_index(name='shixun'))
  

    # 分组获取教学视频字典
    video = (merged_data.groupby(['stage_id','stage_name',"stage_sort",'knowledge_id','knowledge',
                                  'shixun_num','video_num','attachment_num'])
    .apply(lambda x:x[['video_id','video_name', 'video_link']].to_dict('r'))
    .reset_index(name='video'))

    # 分组获取教学课件字典
    att = (merged_data.groupby(['stage_id','stage_name',"stage_sort",'knowledge_id','knowledge',
                                'shixun_num','video_num','attachment_num'])
    .apply(lambda x:x[['attachment_id','filename', 'attachment_link']].to_dict('r'))
    .reset_index(name='attachment'))

    # 对分好组的教学资源字典按照章节知识点进行合并
    shixun_video = shixun.merge(video,how='outer',on=['stage_id','stage_name',"stage_sort",'knowledge_id',
                                                'knowledge','shixun_num','video_num','attachment_num'])
    res=shixun_video.merge(att,how='outer',on=['stage_id','stage_name',"stage_sort",'knowledge_id','knowledge',
                                         'shixun_num','video_num','attachment_num'])

    # 按照知识点进行分组，并命名为知识点
    result = (res.groupby(['stage_id','stage_name',"stage_sort"])
              .apply(lambda x :x[['knowledge_id','knowledge','shixun_num','video_num','attachment_num',
                                  'shixun','video','attachment']].to_dict('r'))
    .reset_index(name='knowledge'))

    # 将最终结果转换成JSON格式
    result = json.loads(result.to_json(orient='records',force_ascii=False))
    return result
  else:
    # 若course_id为空，或者course_id不存在，返回空字典
    return {}


if __name__ == '__main__':
  # 读取数据
  course_id = "19043"
  recommend_results = get_initial_path(course_id,'1')
  pp(recommend_results)