原文地址：https://mp.weixin.qq.com/s/ifVrdv_yF6dQ4kQ3qha0Hw
library("rvest")
library("dplyr")
library("ggplot2")
library("grid")
library("showtext")
library("Cairo")font.add("myfont","msyh.ttc")
找一份网上的最新统计数据
url<-"http://bbs.thmz.com/thread-2715502-1-1.html"
table<-read_html(url,encoding="gbk")%>%html_nodes("td.t_f>table")%>%html_table(header =TRUE,trim =TRUE,fill=TRUE)%>%as.data.frame()
城市数据
mydata<-data.frame(label=c("上海","北京","杭州","广州","深圳","苏州","成都","天津","南京","武汉","宁波","重庆","西安","无锡","厦门","青岛","长沙","南通","常州","福州"))
整理数据
mydata<-merge(mydata,table[,c("城市","门店数量")],by.x="label",by.y="城市",all.x=TRUE)mydata[mydata$label=="常州","门店数量"]=27
names(mydata)[2]<-"value"
mydata<-arrange(mydata,-value)%>%transform(id=1:20,class=c(1,6,rep(1:6,3)))mydata$label<-as.character(mydata$label)
#标签拆成单字换行，竖排布局：
label<-strsplit(mydata$label,"")
for(iin1:length(label)){mydata$label_ff[[i]]<-paste0(label[[i]],collapse="\n")}mydata[1:2,"label_ff"]<-c("上海","北京")mydata[16:20,"label_ff"]<-c("青岛","厦门","常州","福州","南通")
#计算标签的旋转角度：
mydata$angle=c(rev(9*(1:10-1)+4.5),-(9*(1:10-1)+4.5))mydata$angle[16:20]<-rev(9*(1:5-1)+4.5)
图形可视化过程：
p<-ggplot(mydata)+geom_col(aes(x=id,y=value,fill=factor(class)))+geom_hline(yintercept =c(25,50,100,200,500),linetype=2,size=.25)+geom_text(aes(x=id,y=value+12,label=label_ff,angle=angle),family="myfont",size=3.5,lineheight=1)+#坐标轴放大一倍占位：scale_x_continuous(limits=c(0,40),expand=c(0,0))+#Y延伸到负值突出圆心的空白scale_y_continuous(limits=c(-100,600))+scale_fill_manual(values=c("#00643E","#207A57","#3D8C6D","#59A284","#76B69B","#95CBB3"),guide=FALSE)+coord_polar(start=-14.245)+theme_void();p