R语言-股票数据库(2)-股票日K线信息-未复权

######################################################################################################
#################################批量下载深市股票信息到csv文件########################################
######################################################################################################
#把下载好的数据按照市场分开，分别保存到txt文件即可，这里不再赘述。
#保存好的数据要稍微处理一下，日期的格式调整为yyyymmdd，write.table会把数据框的“列名”打印出来，列名也是要去掉的。
#有了股票代码和上市日期数据就可以自动化下载数据了，最好深市沪市分开进行，存在不同的文件夹下，R代码如下
#下载股票数据
library(RCurl)
#http://quotes.money.163.com/service/chddata.html?code=0600030&start=20030106&end=20140920&fields=TCLOSE;HIGH;LOW;TOPEN;LCLOSE;CHG;PCHG;TURNOVER;VOTURNOVER;VATURNOVER;TCAP;MCAP
url1    <- "http://quotes.money.163.com/service/chddata.html?code="
market  <- "1" # 1:深市，0:沪市
code    <- "000003"
url2    <- "&start="
start   <- "19900101"
url3    <- "&end="
end     <- "20170711"         #用当前最新日期
url4    <- "&fields=TCLOSE;HIGH;LOW;TOPEN;LCLOSE;CHG;PCHG;TURNOVER;VOTURNOVER;VATURNOVER;TCAP;MCAP"

#文件的存放路径
file.path   <- "E:/.../数据源/SZ/"

#股票代码+发行日期，格式：CODE制表符yyyymmdd
security <- readLines("E:/.../数据源/深市股票上市日.txt")
code <- vector(length = length(security))
listingdate <- vector(length = length(security))
security.tab <- data.frame(code, listingdate, stringsAsFactors=FALSE)

for (i in 1 : length(security))
{
security.tab[i,] = do.call(cbind,(strsplit(security[i],"\t")))          #do.call(cbind,[list]) 列表转矩阵
}

for (i in 1 : dim(security.tab)[1])
{
code <- security.tab$code[i]
start <- security.tab$listingdate[i]
cat(i,"\t---",code,"\n")
url <- paste(url1,market,code,url2,start,url3,end,url4,sep="")
file <- paste(file.path,code,".csv",sep="")
download.file(url,destfile=file, method="auto",quiet=T)
}

######################################################################################################
#################################批量导入深市股票信息到数据库#########################################
######################################################################################################
#创建表
create table stock_info_sz
(
data_dt varchar(10),
stock_id varchar(10),
stock_name  varchar(100),
shoupanjia  decimal(18,2),
zuigaojia   decimal(18,2),
zuidijia    decimal(18,2),
kaipanjia   decimal(18,2),
qianshoupan   decimal(18,2),
zhangdiee    decimal(18,2),
zhangdiefu   decimal(18,6),
huanshoulv   decimal(18,6),
chengjiaoliang decimal(18,0),
chengjiaojine  decimal(18,2),
zongshizhi     decimal(18,0),
liutongzhi     decimal(18,0)
)

#在成功获得数据，并保存到若干csv、txt文件之后，最后一步就是导入数据库MySQL。借助R包RMySQL，可以实现R和MySQL的连接，在R环境中直接操作MySQL数据库。
# 载入RMySQL包
library(RMySQL)
# 建立数据库连接
conn <- dbConnect(MySQL(),dbname = "gushi",username = "root")
# 启动非严格模式
dbSendQuery(conn, "SET @@sql_mode=ANSI;")
# win7环境下如果汉字乱码，就运行这条命令
dbSendQuery(conn, "SET NAMES GBK")

security <- readLines("E:/.../数据源/深市股票上市日.txt")
code <- vector(length = length(security))
listingdate <- vector(length = length(security))
security.tab <- data.frame(code, listingdate, stringsAsFactors=FALSE)

for (i in 1 : length(security))
{
security.tab[i,] = do.call(cbind,(strsplit(security[i],"\t")))          #do.call(cbind,[list]) 列表转矩阵
}

#文件的存放路径
file.path   <- "E:/.../数据源/SZ/"
for (i in 1 : dim(security.tab)[1])
{
code <- security.tab$code[i]
cat(i,"\t---",code,"\n")
file <- paste(file.path,code,".csv",sep="")
#拼接SQL
start <- "load data infile '"
end <- "' into table stock_info_sz character set GBK fields terminated by ',' lines terminated by '\n'"
sql <- paste(start,file,end,sep="")
res1 <-dbSendQuery(conn, sql)
}
#清除文件头数据
sql <- "delete from stock_info_sz where data_dt='日期'"
res1 <-dbSendQuery(conn, sql)
#去掉股票代码前的'
sql <- "update stock_info_sz set stock_id=replace(stock_id,'\\'','')"
res1 <-dbSendQuery(conn, sql)