R:使用 RJDBC 将 csv 文件插入数据库

Question

由于RJDBC是我能够在Ubuntu上工作的唯一软件包，我尝试使用它来INSERT一个CSV文件进入数据库。

我可以做以下工作:

# Connecting to database
library(RJDBC)
drv <- JDBC('com.microsoft.sqlserver.jdbc.SQLServerDriver', 'drivers/sqljdbc42.jar', identifier.quote="'")
connection_string <- "jdbc:sqlserver://blablaserver;databaseName=testdatabase"
ch <- dbConnect(drv, connection_string, "username", "password")

# Inserting a row
dbSendQuery(ch, "INSERT INTO cpr_esben.CPR000_Startrecord (SORTFELT_10,OPGAVENR,PRODDTO,PRODDTOFORRIG,opretdato) VALUES ('TEST', 123, '2012-01-01', '2012-01-01', '2012-01-01')")

插入有效。接下来，我尝试使用相同的数据创建一个 CSV 文件的INSERT，该文件由默认的“选项卡”分隔，并且我正在 Windows 上工作。

# Creating csv
df <- data.frame(matrix(c('TEST', 123, '2012-01-01', '2012-01-01', '2012-01-01'), nrow = 1), stringsAsFactors = F)
colnames(df) <- c("SORTFELT_10","OPGAVENR","PRODDTO","PRODDTOFORRIG","opretdato")
class(df$SORTFELT_10) <- "character"
class(df$OPGAVENR) <- "character"
class(df$PRODDTO) <- "character"
class(df$PRODDTOFORRIG) <- "character"
class(df$opretdato) <- "character"
write.table(df, file = "test.csv", col.names = FALSE, quote = FALSE)

# Inserting CSV to database
dbSendQuery(ch, "INSERT cpr_esben.CPR000_Startrecord FROM 'test.csv'")

Unable to retrieve JDBC result set for INSERT cpr_esben.CPR000_Startrecord FROM 'test.csv' (Incorrect syntax near the keyword 'FROM'.)

当我尝试插入 csv 文件时，您对我做错了什么有什么建议吗？我没有收到关键字“FROM”附近的语法错误错误？

Answer 1

如果您根据数据创建报表会怎样？像这样的东西:

# Data from your example
df <- data.frame(matrix(c('TEST', 123, '2012-01-01', '2012-01-01', '2012-01-01'), nrow = 1), stringsAsFactors = F)
colnames(df) <- c("SORTFELT_10","OPGAVENR","PRODDTO","PRODDTOFORRIG","opretdato")
class(df$SORTFELT_10) <- "character"
class(df$OPGAVENR) <- "character"
class(df$PRODDTO) <- "character"
class(df$PRODDTOFORRIG) <- "character"
class(df$opretdato) <- "character"

# Formatting rows to insert into SQL statement
rows <- apply(df, 1, function(x){paste0('"', x, '"', collapse = ', ')})
rows <- paste0('(', rows, ')')

# SQL statement
statement <- paste0(
  "INSERT INTO cpr_esben.CPR000_Startrecord (", 
  paste0(colnames(df), collapse = ', '), 
  ')',
  ' VALUES ',
  paste0(rows, collapse = ', ')
)

dbSendQuery(ch, statement)

这应该适用于 df 中的任意数量的行