This function takes a character vector of variable names and checks them against the official AEME variable names. If an input variable does not match any official name, the function attempts to find the closest match using both exact keyword matching and fuzzy string matching (Levenshtein distance). The function returns a character vector of the same length as the input, where each element is either the original variable name (if it was valid) or the best-matching official AEME variable name. If no suitable match is found for an invalid variable, a warning is issued.
Arguments
- x
Character vector of variable names to check.
- key_filter
Optional string to filter the AEME variable names by a specific keyword. If provided, only AEME variable names containing this keyword will be considered for matching. This can help improve matching accuracy by limiting the pool of candidate variable names to those relevant to a particular context (e.g., "met" for meteorological variables).
Value
Character vector of variable names, with invalid names replaced by the closest official AEME variable name where possible.
Examples
guess_aeme_vars(c("temp", "oxy", "ph", "chla", "tp", "tn"))
#> ✔ Variable 'temp' matched to 'HYD_temp'.
#> ✔ Variable 'oxy' matched to 'CHM_oxy'.
#> ✔ Variable 'ph' matched to 'CHM_ph'.
#> ✔ Variable 'chla' matched to 'PHY_tchla'.
#> ✔ Variable 'tp' matched to 'PHS_tp'.
#> ✔ Variable 'tn' matched to 'NIT_tn'.
#> [1] "HYD_temp" "CHM_oxy" "CHM_ph" "PHY_tchla" "PHS_tp" "NIT_tn"
guess_aeme_vars(c("temp", "swr", "lwr", "wind", "precip"), key_filter = "MET")
#> ✔ Variable 'temp' matched to 'MET_tmpair'.
#> ✔ Variable 'swr' matched to 'MET_radswd'.
#> ✔ Variable 'lwr' matched to 'MET_radlwd'.
#> ✔ Variable 'wind' matched to 'MET_wndspd'.
#> ✔ Variable 'precip' matched to 'MET_pprain'.
#> [1] "MET_tmpair" "MET_radswd" "MET_radlwd" "MET_wndspd" "MET_pprain"