CODEWITHSUNDEEP
rdplyr
Pravellika
Pravellika

Reputation: 182

Loop through dataframe and execute SQL query

I have a dataframe which has 6 rows with SQL queries in each line. I want to execute every query and assign the output to a dataframe. currently I have 6 line of codes as below:

df6 <- tbl(conn,sql(df[6,])) Is there a way I can loop through and assign the same in one statement using dplyr mutate function? I tried below:

i <- 1:nrow(df)
df %>% mutate(paste0('df',i) <- tbl(conn,sql(df[i,])))

This throws the following error:

Error: Column q1 must be a 1d atomic vector or a list.

Any help is appreciated.

Upvotes: 0

Views: 725

Answers (2)

Pravellika
Pravellika

Reputation: 182

With the help from Shinobi_Atobe's function I could develop it into what I wanted as below:

list2env(setNames(lapply(df$sql_query,function(x) {tbl(conn, sql(x))}), paste0('df', 1:nrow(df))), envir=.GlobalEnv)

This executes sql query from every row in the dataframe and stores the output in individual dataframes named as df1,df2...

Upvotes: 0

Shinobi_Atobe
Shinobi_Atobe

Reputation: 1963

Your question seems a little unclear, but assuming you have a dataframe with a column titled 'sql_code', you can just apply a function to each element in that column 

lapply(df$sql_code, function(x) tbl(conn, sql(x)))

you will end up with a list that you can put together with bind_cols() (if the data frames are of the same dimensions)

Upvotes: 1

Related Questions