新方法进行R语言分列整合---基于业务的统计分析

rm(list = ls())


library(tidyr)
library(dplyr)
library(lubridate)
library(rJava)
library(xlsxjars)
library(xlsx)
library(data.table)
library(readr)
library(readxl)


setwd('G:\\R\\data\\0412')


dir()


newdata <- read_excel('data.xlsx',sheet = 'Sheet4')


newdata1 <- read_excel('data.xlsx',sheet = '客户ID')


newdata1_1 <- newdata1[,c(1,3)]


newdata$n <- 0


for(i in 1:nrow(newdata)){
  
  newdata$n[i] <- i
}


newdata <- data.frame(newdata)


text1 <- strsplit(as.character((newdata$ITEM3)),',')


text2 <- mapply(cbind,newdata[,4],text1)


data1 <- rbind(data.frame(text2[1]),data.frame(text2[2]))


for (i in 3:nrow(newdata)) {
  
  data1 <- rbind(data1,data.frame(text2[i]))
  print(i)
  
}


binddata1 <- newdata[,c(1,4)]


str(binddata1)


str(data1)


str(newdata1_1)


data1$X1 <- as.numeric(data1$X1)


data1$X2 <- as.numeric(data1$X2)


data1_1 <- left_join(data1,binddata1,by=c('X1'='n'))




data1_2 <- left_join(data1_1,newdata1_1,by=c('X2'='ID'))




data2 <- data1_2 %>% group_by(DEMANDCODE) %>%
                    summarise(companynames = paste(NAME,collapse = ','))




result <- left_join(newdata,data2,by='DEMANDCODE')