Is it possible to insert (add) a row to a SQLite db table using dplyr package?

Question

I am new to the database connection capabilities of dplyr package, but I am very interested in using it for an SQLite connection. I followed this tutorial and created an SQL

Answer 1

In this newsgroup. Hadley explained the purpose of the function dplyr::copy_to(). It is intended to create temporary test tables. The email exchange ends by suggesting to use RMySQL::dbWriteTable() to append data to an existing table. The same applies to SQLite databases, as explained in the accepted answer above.

To append a data frame dtf which has the same column names as an existing database table, I used:

library(RMySQL)
DB <- dbConnect(MySQL(), user="usename", host="localhost",
                   password="***", dbname="dbname")
dbWriteTable(DB, "tablename", dtf, append=TRUE, row.names = FALSE)

Answer 2

The main reason I use dplyr to write into a database is that I do not want to switch between languages mid code. For your question the best solution I think is to use the db_insert_into() function of dplyr as illustrated by StatSandwich

Answer 3

You can perform SQL operations on a database/table created via dplyr, but you have to revert to RSQLite/DBI calls and change how you made the database/table:

library(dplyr)

my_db <- src_sqlite("my_db.sqlite3", create=TRUE) 
copy_to(my_db, iris, "my_table", temporary=FALSE) # need to set temporary to FALSE

# grab the db connection from the object created by src_sqlite
# and issue the INSERT That way

res <- dbSendQuery(my_db$con, 
                   'INSERT INTO my_table VALUES (9.9, 9.9, 9.9, 9.9, "new")')

Answer 4

No, you can do this all within dplyr.

require(dplyr)

my_db <- src_sqlite( "my_db.sqlite3", create = TRUE)                 # create src
copy_to( my_db, iris, "my_table", temporary = FALSE)                 # create table
newdf = iris                                                         # create new data
db_insert_into( con = my_db$con, table = "my_table", values = newdf) # insert into