Bringing arrow-adbc to R via DBI, adbi aims to provide DBI-compliant database access.
You can install the development version of adbi from GitHub with:
# install.packages("devtools")
devtools::install_github("r-dbi/adbi")
The data.frame
API of DBI is supported.
library(DBI)
# Create an SQLite connection using the adbcsqlite backend
con <- dbConnect(adbi::adbi("adbcsqlite"), uri = ":memory:")
# Write a table
dbWriteTable(con, "swiss", datasets::swiss)
# Query it
dbGetQuery(con, "SELECT * from swiss WHERE Agriculture < 40")
#> Fertility Agriculture Examination Education Catholic Infant.Mortality
#> 1 80.2 17.0 15 12 9.96 22.2
#> 2 92.5 39.7 5 5 93.40 20.2
#> 3 85.8 36.5 12 7 33.77 20.3
#> 4 76.1 35.3 9 7 90.57 26.6
#> 5 71.7 34.0 17 8 3.30 20.0
#> 6 55.7 19.4 26 28 12.11 20.2
#> 7 54.3 15.2 31 20 2.15 10.8
#> 8 58.3 26.8 25 19 18.46 20.9
#> 9 70.4 38.4 26 12 5.62 20.3
#> 10 65.7 7.7 29 11 13.79 20.5
#> 11 72.7 16.7 22 13 11.22 18.9
#> 12 64.4 17.6 35 32 16.92 23.0
#> 13 77.6 37.6 15 7 4.97 20.0
#> 14 67.6 18.7 25 7 8.65 19.5
#> 15 35.0 1.2 37 53 42.34 18.0
#> 16 42.8 27.7 22 29 58.33 19.3
# Prepared statements
res <- dbSendQuery(con, "SELECT * from swiss WHERE Agriculture < ?")
dbBind(res, 30)
dbFetch(res)
#> Fertility Agriculture Examination Education Catholic Infant.Mortality
#> 1 80.2 17.0 15 12 9.96 22.2
#> 2 55.7 19.4 26 28 12.11 20.2
#> 3 54.3 15.2 31 20 2.15 10.8
#> 4 58.3 26.8 25 19 18.46 20.9
#> 5 65.7 7.7 29 11 13.79 20.5
#> 6 72.7 16.7 22 13 11.22 18.9
#> 7 64.4 17.6 35 32 16.92 23.0
#> 8 67.6 18.7 25 7 8.65 19.5
#> 9 35.0 1.2 37 53 42.34 18.0
#> 10 42.8 27.7 22 29 58.33 19.3
dbBind(res, 20)
dbFetch(res)
#> Fertility Agriculture Examination Education Catholic Infant.Mortality
#> 1 80.2 17.0 15 12 9.96 22.2
#> 2 55.7 19.4 26 28 12.11 20.2
#> 3 54.3 15.2 31 20 2.15 10.8
#> 4 65.7 7.7 29 11 13.79 20.5
#> 5 72.7 16.7 22 13 11.22 18.9
#> 6 64.4 17.6 35 32 16.92 23.0
#> 7 67.6 18.7 25 7 8.65 19.5
#> 8 35.0 1.2 37 53 42.34 18.0
# Cleanup
dbClearResult(res)
More interestingly, the recent arrow-extension API of DBI is supported as well.
# Queries
dbGetQueryArrow(con, "SELECT * from swiss WHERE Agriculture < 40")
#> <nanoarrow_array_stream struct<Fertility: double, Agriculture: double, Examination: int64, Education: int64, Catholic: double, Infant.Mortality: double>>
#> $ get_schema:function ()
#> $ get_next :function (schema = x$get_schema(), validate = TRUE)
#> $ release :function ()
# Prepared statements
res <- dbSendQueryArrow(con, "SELECT * from swiss WHERE Agriculture < ?")
dbBind(res, 30)
ret <- dbFetchArrow(res)
ret$length
#> NULL
dbBind(res, 20)
# Chunked fetches
while (!dbHasCompleted(res)) {
ret <- dbFetchArrowChunk(res)
message("fetched ", ret$length, " rows")
}
#> fetched 8 rows
#> fetched 0 rows
# Cleanup
dbClearResult(res)
dbDisconnect(con)