Read data from a database

read_from_database takes a database connection and table name, applies a dplyr-based query, and returns either a lazy-table or a data frame of the query results.

Usage

read_from_database(
  db,
  table_name,
  query_fun = NULL,
  collect = TRUE,
  pull = NULL
)

Arguments

db

A database connection or the path to a database file (if sqlite or duckdb file extension).

table_name

A character string specifying the table to read from.

query_fun

A function taking a single argument (the data.frame-like object to be queried) and returning a data frame. Default is NULL, which does not apply a query fun. Most dplyr functions will be implemented here (see dbplyr::dbplyr for details). (e.g. query_fun = \(df) df |> dplyr::select(column1, column2)) Anything else beyond variable names needs to be prefaced with !! (e.g. ... |> dplyr::filter(month |> dplyr::between(!!select_months[1], !!select_months[2]))).

collect

A logical value indicating whether to use dplyr::collect() to fetch the data from the database. Will be overidden if pull is not NULL. Default is TRUE.

pull

A single character string to use with dplyr::pull() to pull a column after applying query_fun. Default is NULL, which does not pull any columns.

Value

query_fun applied to the table table_name in db where if:

• collect is TRUE (the default), returns a tibble (data.frame)

• collect is FALSE, returns a lazy table of the query

• Unless pull is a column name, returns a vector of the specified column