Skip to contents

Read a plain-header CSV meteorological file

read_plain_csv.Rd

Reads a standard CSV file in which the first row is a column header. The only preprocessing performed is normalising the timestamp column name: any column whose name matches "timestamp" (case-insensitive) is renamed to "TIMESTAMP" so that downstream metamet functions can locate it consistently.

Usage

read_plain_csv(fname, ...)

Arguments

fname

Path to the CSV file.

...

Additional arguments forwarded to data.table::fread, e.g. sep, na.strings.

Value

A data.table with column names exactly as written in the file header. No renaming is performed; the caller's dt_meta must use name_dt values that match the actual column names. Downstream metamet functions (metamet_wide_to_long, convert_time_char_to_posix) normalise the time column to TIMESTAMP automatically based on dt_meta.

Details

This reader is the appropriate choice for plain-CSV exports such as EIDC data packages, where there is no multi-row BADC-CSV header and no Campbell TOA5 instrument-info preamble.

See also

read_ceda_csv, read_obs_autodetect

Examples

if (FALSE) { # \dontrun{
dt <- read_plain_csv(
  pkg_extdata("UK-WHM/historical/eidc/whim_met_2002_2023.csv")
)
} # }