Redo non-tracked sequences

Sometimes, previously reviewed animal tracking sequences need to be redone. This occurs when animal speeds cannot be calculated due to an insufficient number of logged positions, when additional species need to be tracked, or when REM results indicate a potential issue.

Usage

redo_non_tracked(
  track_data,
  seq_done_in = NULL,
  redo_toofew_positions = NULL,
  redo_non_tracked = NULL,
  add_tracked = NULL,
  email = Sys.getenv("email")
)

Arguments

track_data: a camtrapdp datapackage
seq_done_in: a character list with viewed sequences
redo_toofew_positions: a boolean to hardcode redoing too few positions (default = NULL)
redo_non_tracked: a boolean to hardcode redoing non-tracked sequences (default = NULL)
add_tracked: a boolean to hardcode adding tracked sequences to seq_done_out (default = NULL)
email: Optional. The email address used for Google Sheets authentication. Defaults to the "email" system environment variable.

Value

A list containing the following exports:

seq_done_out: A character vector of sequence IDs to be tracked.
non_tracked_parsing_error: A dataframe of sequences with parsing errors.
non_tracked_toofew_poles: A dataframe of sequences with a calibration lacking poles.
non_tracked_toofew_positions: A dataframe of sequences with too few positions.
non_tracked_other_issues: A dataframe of sequences with other remarks.

Details

This function detects non-tracked sequences in seq_done_in and removes them from the list if redo_non_tracked == TRUE. The modified list is then exported as seq_done_out. If redo_non_tracked == NULL (the default), the user is prompted whether they want to perform the rerun.

Sequences with issues calculating animal speed are exported as separate dataframes. If there are sequences lacking positions, they are removed from seq_done_out if redo_toofew_positions == TRUE. If redo_toofew_positions == NULL (the default), the user is prompted whether to perform the rerun. Regardless of the redo_toofew_positions value, these cases are always exported as a separate dataframe.

Currently, the function flags the following issues:

Parsing errors
Calibration with too few poles
Sequences with too few positions

Other issues are also exported in the non_tracked_other_issues dataframe.

Author

Sander Devisscher

Examples

if (FALSE) { # \dontrun{
zip_dir <- tempdir()
files_dir <- paste0(zip_dir, "\\Files\\")

download_gdrive_if_missing(gfileID = "1U7cSwhUZRJlR_zpJvYyhsi24J2QV8U8q",
                           destfile = file.path(zip_dir, "gmu8.zip"),
                           update_always = TRUE,
                           email = Sys.getenv("email"))

unzip(paste0(zip_dir, "/gmu8.zip"),
      exdir = file.path(files_dir))

gmu8 <- camtrapdp::read_camtrapdp(paste0(files_dir, "/datapackage.json"))

calibration_deps <- gmu8 %>%
  camtrapdp::filter_observations(cameraSetupType == "calibration")

obs_list <- calibration_deps$data$observations
dep_list <- unique(obs_list$deploymentID)

data <- gmu8 |>
  camtrapdp::filter_deployments(deploymentID %in% dep_list)

track_data <- data %>%
  camtrapdp::filter_observations(scientificName %in% "Sus scrofa")

file_path <- "../fis-rem-tracking/GMU8/Input/tracking_seq_done.csv"

seq_done <- tryCatch({
  # Attempt to read the CSV file
  read.csv(file_path)
}, error = function(e) {
  # If an error occurs (e.g., file not found), create the dummy data frame
  data.frame(sequenceID = character())
})

seq_done <- c(seq_done$sequenceID)

redo_non_tracked <- redo_non_tracked(track_data = track_data,
                                     seq_done_in = seq_done)

seq_done_out <- redo_non_tracked$seq_done_out
parsing_issues <- redo_non_tracked$non_tracked_parsing_error
toofew_poles <- redo_non_tracked$non_tracked_toofew_poles
toofew_positions <- redo_non_tracked$non_tracked_toofew_positions
other_issues <- redo_non_tracked$non_tracked_other_issues
} # }