gistfile1.txt

#!/bin/bash

# Requirements:
#  - Parameter checking: The script accepts at least one parameter (the base folder).
#    An optional second parameter "scan" activates the scan mode, which only shows missing
#    or incorrect DateTimeOriginal tags, without making any changes.
#
#  - Extract date from folder names: The script extracts the date from the name
#    of the folder. Supported formats are YYYY-MM-DD and YYYY-MM, where for the latter
#    the day 01 is automatically added.
#
#  - Scan mode: In scan mode, the script scans all files in a folder
#    (and all subfolders) and lists those where the DateTimeOriginal tag
#    is missing or set to an invalid date.
#
#  - Correction mode: In correction mode, the script sets the DateTimeOriginal tag for all
#    files without a valid date to the date extracted from the folder name. Invalid
#    tags are also corrected. For 3gp and MP4 files, this is only done if errors were found in scan mode.
#    For other files, it is done regardless.
#
#  - Efficient processing: To maximize efficiency, ExifTool should be operated in "batch" mode
#    by calling it once per folder instead of separate calls for each file. This reduces the startup time
#    of ExifTool and improves overall performance.
#
#  - Supported file types: Processing of files with the extensions .jpg, .jpeg, .png,
#    .mp4, and .avi.
#
#  - Recursive processing: Recursively traverse all subfolders of the specified base folder,
#    to perform the above operations at each level of the folder structure.
#
#  - User feedback: Informative outputs show the progress of processing, including
#    warnings when no valid data can be found for a folder. The processing
#    is done in alphabetical order of folder names.

if [ $# -lt 1 ] || [ $# -gt 2 ]; then
    echo "Usage: $0 <base folder> [scan]"
    exit 1
fi

base_dir="$1"
scan_only=false

if [ "$2" == "scan" ]; then
    scan_only=true
fi

# Function to extract the date from the folder name
extract_date_from_folder_name() {
    local folder_name="$1"
    # Extension of the Regex to recognize the format YYYY-MM and automatically add the day 01
    if [[ $folder_name =~ ([0-9]{4})-([0-9]{2})-([0-9]{2}) ]]; then
        echo "${BASH_REMATCH[1]}-${BASH_REMATCH[2]}-${BASH_REMATCH[3]}"
        return 0
    elif [[ $folder_name =~ ([0-9]{4})-([0-9]{2}) ]]; then
        echo "${BASH_REMATCH[1]}-${BASH_REMATCH[2]}-01"
        return 0
    else
        echo ""
        return 1
    fi
}

# Function to scan or correct EXIF data
process_exif_data() {
    local folder_path="$1"
    local date_from_folder="$2"
    echo "Processing folder: $folder_path"

    local files=()
    while IFS= read -r -d '' file; do
        files+=("$file")
    done < <(find "$folder_path" -maxdepth 1 -type f \( -iname "*.jpg" -o -iname "*.jpeg" -o -iname "*.png" -o -iname "*.mp4" -o -iname "*.3gp" \) -print0)

    if [ "${#files[@]}" -gt 0 ]; then
        if [ "$scan_only" == "true" ]; then
            echo "Scanning for missing or invalid DateTimeOriginal tags..."
            exiftool -api largefilesupport=1 -if 'not $DateTimeOriginal or $DateTimeOriginal =~ /^0000/' -filename -DateTimeOriginal "${files[@]}"
        else
            echo "Correcting EXIF data based on the date $date_from_folder..."
            exiftool -api largefilesupport=1 -d "%Y-%m-%d %H:%M:%S" "-DateTimeOriginal=$date_from_folder 00:00:00" -if 'not $DateTimeOriginal or $DateTimeOriginal =~ /^0000/' "${files[@]}" -overwrite_original
        fi
    else
        echo "No supported files found in $folder_path"
    fi
}

# Recursive function to traverse the folder structure
process_folder() {
    local current_folder="$1"
    local inherited_date="$2"

    local current_folder_name=$(basename "$current_folder")
    local date_from_current_folder=$(extract_date_from_folder_name "$current_folder_name")
    local effective_date=${date_from_current
    
    if [ ! -z "$effective_date" ]; then
        process_exif_data "$current_folder" "$effective_date"
    else
        echo "Warning: No valid date found for $current_folder. Skipping..."
    fi

    find "$current_folder" -mindepth 1 -maxdepth 1 -type d | sort | while read subfolder; do
        process_folder "$subfolder" "$effective_date"
    done
}

process_folder "$base_dir" ""