src_dbi {dbplyr} | R Documentation |
src_dbi()
is a general dplyr backend that connects to any
DBI driver. src_memdb()
connects to a temporary in-memory SQLite
database, that's useful for testing and experimenting.
You can generate a tbl()
directly from the DBI connection, or
go via src_dbi()
.
src_dbi(con, auto_disconnect = FALSE) ## S3 method for class 'src_dbi' tbl(src, from, ...)
con |
An object that inherits from DBI::DBIConnection, typically generated by DBI::dbConnect |
auto_disconnect |
Should the connection be automatically closed when
the src is deleted? Set to |
src |
Either a |
from |
Either a string (giving a table name) or literal |
... |
Needed for compatibility with generic; currently ignored. |
All data manipulation on SQL tbls are lazy: they will not actually
run the query or retrieve the data unless you ask for it: they all return
a new tbl_dbi
object. Use compute()
to run the query and save the
results in a temporary in the database, or use collect()
to retrieve the
results to R. You can see the query with show_query()
.
For best performance, the database should have an index on the variables
that you are grouping by. Use explain()
to check that the database is using
the indexes that you expect.
There is one exception: do()
is not lazy since it must pull the data
into R.
An S3 object with class src_dbi
, src_sql
, src
.
# Basic connection using DBI ------------------------------------------- library(dplyr) con <- DBI::dbConnect(RSQLite::SQLite(), ":memory:") src <- src_dbi(con, auto_disconnect = TRUE) # Add some data copy_to(src, mtcars) src DBI::dbListTables(con) # To retrieve a single table from a source, use `tbl()` src %>% tbl("mtcars") # You can also use pass raw SQL if you want a more sophisticated query src %>% tbl(sql("SELECT * FROM mtcars WHERE cyl = 8")) # Alternatively, you can use the `src_sqlite()` helper src2 <- src_sqlite(":memory:", create = TRUE) # If you just want a temporary in-memory database, use src_memdb() src3 <- src_memdb() # To show off the full features of dplyr's database integration, # we'll use the Lahman database. lahman_sqlite() takes care of # creating the database. if (has_lahman("sqlite")) { lahman_p <- lahman_sqlite() batting <- lahman_p %>% tbl("Batting") batting # Basic data manipulation verbs work in the same way as with a tibble batting %>% filter(yearID > 2005, G > 130) batting %>% select(playerID:lgID) batting %>% arrange(playerID, desc(yearID)) batting %>% summarise(G = mean(G), n = n()) # There are a few exceptions. For example, databases give integer results # when dividing one integer by another. Multiply by 1 to fix the problem batting %>% select(playerID:lgID, AB, R, G) %>% mutate( R_per_game1 = R / G, R_per_game2 = R * 1.0 / G ) # All operations are lazy: they don't do anything until you request the # data, either by `print()`ing it (which shows the first ten rows), # or by `collect()`ing the results locally. system.time(recent <- filter(batting, yearID > 2010)) system.time(collect(recent)) # You can see the query that dplyr creates with show_query() batting %>% filter(G > 0) %>% group_by(playerID) %>% summarise(n = n()) %>% show_query() }