Skip to contents

Remove duplicated photos before or after photo name appended to end of photo file name by fpr_photo_rename

Usage

fpr_photo_remove_dupes(
  dir_target = NULL,
  dry_run = TRUE,
  min_replicates = 2,
  col_time = date_time_original,
  col_photo_name = source_file,
  col_model = model,
  col_iso = iso,
  remove_renamed = FALSE,
  ...
)

Arguments

dir_target

String. Full path name of directory to be scanned for duplicate photos with different names.

dry_run

Logical. Should photos selected by `remove_renamed` be removed or should a dataframe be produced to show the identical photos? Defaults to TRUE.

min_replicates

Integer. Minimum number of identical photos (based on identical `col_time`, `col_model` and `col_iso` values in photo metadata) that should be identified. Defaults to 2. Setting this to higher than 2 is intended for `dry_run` = TRUE as it is likely to remove photos used for more than one thing vs removing renamed photos.

col_time

The bare (unquoted) name of the column containing the exif parameter stamping when photo was created. Defaults to `date_time_original`.

col_photo_name

The bare (unquoted) name of the column containing the exif parameter with full path photo name. Defaults to `source_file`.

col_model

The bare (unquoted) name of the column containing the exif parameter stamping camera type that took the photo. Defaults to `model`.

col_iso

he bare (unquoted) name of the column containing the exif parameter stamping photo iso. Defaults to `iso`.

remove_renamed

Logical. Should renamed photos be removed (based on length of photo name)? Defaults to FALSE so that duplicate photos that have not been renamed are removed.

...

Not used.

Value

A dataframe if `dry_run` is TRUE (default) or removal of photos specified through `remove_renamed` param.