curl::curl_download(
  "https://github.com/duckdb/duckdb-data/releases/download/v1.0/lineitemsf1.snappy.parquet",
  "lineitemsf1.snappy.parquet"
)

library(arrow, warn.conflicts = FALSE)
library(dplyr, warn.conflicts = FALSE)

open_dataset("lineitemsf1.snappy.parquet") |>
  filter(
    l_shipdate >= "1994-01-01",
    l_shipdate < "1995-01-01",
    l_discount >= 0.05,
    l_discount < 0.07,
    l_quantity < 24
  ) |>
  summarise(revenue = sum(l_extendedprice * l_discount, na.rm = TRUE)) |>
  collect()

library(duckdb)

con <- dbConnect(duckdb(), ":memory:")

query <- "
FROM
  'lineitemsf1.snappy.parquet'
SELECT
  SUM(l_extendedprice * l_discount) AS revenue
WHERE
  l_shipdate >= '1994-01-01'
  AND l_shipdate < '1995-01-01'
  AND l_discount >= 0.05
  AND l_discount < 0.07
  AND l_quantity < 24
"

dbGetQuery(con, query)

library(duckdb)
library(dplyr, warn.conflicts = FALSE)
library(dbplyr, warn.conflicts = FALSE)

con <- dbConnect(duckdb(), ":memory:")

tbl(con, "lineitemsf1.snappy.parquet") |>
  filter(
    l_shipdate >= "1994-01-01",
    l_shipdate < "1995-01-01",
    l_discount >= 0.05,
    l_discount < 0.07,
    l_quantity < 24
  ) |>
  summarise(revenue = sum(l_extendedprice * l_discount, na.rm = TRUE)) |>
  collect()

library(rpolars)

scan_parquet("lineitemsf1.snappy.parquet")$filter(
  (pl$col("l_shipdate") >= "1994-01-01") &
    (pl$col("l_shipdate") < "1995-01-01") &
    (pl$col("l_discount") >= 0.05) &
    (pl$col("l_discount") < 0.07) &
    (pl$col("l_quantity") < 24)
)$select(
  (pl$col("l_extendedprice") * pl$col("l_discount"))$sum()$alias("revenue")
)$collect() |>
  as.data.frame()