ds

import csv  
import json  
  
def loaddata():  
    file = open(r'thesis.csv', 'r')  # 读入论文信息文件  
    fileopen = csv.reader(file)  
    header = next(fileopen)  
  
    #存储第一作者的信息  
    first_author = []  
  
    #存储最终的学者信息  
    scholar = []  
  
    for row in fileopen:  
        if row[0] not in first_author:  # 第一作者统计  
            first_author.append(row[0])  
  
    # 由于所有的第一、第二作者加起来太多，我们只统计第一作者  
    scholar = first_author  
  
    print("data loaded")  
    return scholar  
  
loaddata函数用于初始化数据，载入csv文件，将所有的筛选出来的第一作者赋给scholar数组  
  
def updatelink(scholar):  
    file1 = open(r'thesis.csv', 'r')  # 再次打开论文信息文件  
    fileopen1 = csv.reader(file1)  
    header1 = next(fileopen1)  
  
    # 第一作者之间互相合作的次数统计的二维矩阵  
    link = [[0 for j in range(len(scholar))] for i in range(len(scholar))]  
  
    for row in fileopen1:  
        if row[0] in scholar:  
            cowoker = row[2].split(';')  
  
            for woker in cowoker:  
                if woker in scholar:  
                    link[scholar.index(row[0])][scholar.index(woker)] += 1  
  
    for i in range(len(scholar)):  
        link[i][i] = 0  
  
    print("link updated")  
    return link  
  
updatelink函数用于生成一个二维数组link[][]来存储每两个学者之间的合作信息  
  
def authorinfo(scholar,link):  
    data = []  
  
    # 用来统计每个第一作者和其他人一共合作的次数  
    times = [0] * len(scholar)  
  
    for i in range(len(scholar)):  
        times[i] += sum(link[i])  
        for j in range(len(scholar)):  
            times[i] += link[j][i]  
  
    for i in range(len(scholar)):  # 统计每两个人之间的合作次数  
        for j in range(i + 1, len(scholar)):  
            temp = link[i][j] + link[j][i]  
            link[i][j], link[j][i] = temp, temp  
  
    # 用来生成每个学者、每个节点的信息，适应echart所需要的格式  
    for i in range(len(scholar)):  
        if times[i] != 0:  
            person = {}  
            person["name"] = scholar[i]  
            person["symbolSize"] = times[i] * 5  
            person["category"] = i % 5  
            person["draggable"] = "true"  
            person["value"] = times[i]  
            data.append(person)  
  
    print("author")  
    return data  
  
  
authorinfo函数用于将学者的信息进行整理，获得echart所需要的节点信息，这里我们会进一步筛选掉一部分学者  
  
def linksinfo(scholar,link):  
    links = []  
  
    for i in range(len(scholar)):  # 生成每条边的信息  
        for j in range(i + 1, len(scholar)):  
            if link[i][j] != 0:  
                links.append({"source": scholar[i], "target": scholar[j], "value": link[i][j]})  
  
    print("links")  
    return links  
  
linksinfo函数用于生成边的信息。  
  
if __name__=='__main__':  
    scholar = loaddata()  
    link = updatelink(scholar)  
    data = authorinfo(scholar,link)  
    links = linksinfo(scholar,link)  
  
    output = {}  
    output['data'] = data  # 生成json格式的文件  
    output['links'] = links  
  
    with open("C:\wamp64\www\project\data.json", "w") as f:  # 将数据写入json文件中  
        json.dump(output, f, sort_keys=True, indent=4, separators=(',', ':'))  
  
    print("finished")