CODEWITHSUNDEEP
rrodbc
jenswirf
jenswirf

Reputation: 7307

How to insert a dataframe into a SQL Server table?

I'm trying to upload a dataframe to a SQL Server table, I tried breaking it down to a simple SQL query string..

library(RODBC)
con <- odbcDriverConnect("driver=SQL Server; server=database")

df <- data.frame(a=1:10, b=10:1, c=11:20)

values <- paste("(",df$a,",", df$b,",",df$c,")", sep="", collapse=",")

cmd <- paste("insert into MyTable values ", values)

result <- sqlQuery(con, cmd, as.is=TRUE)

..which seems to work but does not scale very well. Is there an easier way?

Upvotes: 26

Views: 75121

Answers (3)

Shreyas K
Shreyas K

Reputation: 116

This worked for me and I found it to be simpler. 

library(sqldf)
library(odbc)
con <- dbConnect(odbc(),
                 Driver = "SQL Server",
                 Server = "ServerName",
                 Database = "DBName",
                 UID = "UserName",
                 PWD = "Password")
dbWriteTable(conn = con, 
             name = "TableName", 
             value = x)  ## x is any data frame

Upvotes: 9

agstudy
agstudy

Reputation: 121568

Since insert INTO is limited to 1000 rows, you can dbBulkCopy from rsqlserver package.

dbBulkCopy is a DBI extension that interfaces the Microsoft SQL Server popular command-line utility named bcp to quickly bulk copying large files into table. For example:

url = "Server=localhost;Database=TEST_RSQLSERVER;Trusted_Connection=True;"
conn <- dbConnect('SqlServer',url=url)
## I assume the table already exist
dbBulkCopy(conn,name='T_BULKCOPY',value=df,overwrite=TRUE)
dbDisconnect(conn)

Upvotes: 6

IRTFM
IRTFM

Reputation: 263301

[edited] Perhaps pasting the names(df) would solve the scaling problem:

   values <- paste( " df[  , c(", 
                     paste( names(df),collapse=",") ,
                                   ")] ", collapse="" ) 
      values
      #[1] " df[  , c( a,b,c )] "

You say your code is "working".. I would also have thought one would use sqlSave rather than sqlQuery if one wanted to "upload".

I would have guessed this would be more likely to do what you described:

 sqlSave(con, df, tablename = "MyTable")

Upvotes: 21

Related Questions