Hash Filenames

File: src/scripts/hash-filenames.sh · Rename files in a directory to random hash names.

Renames matching files in the target directory (defaults to the current directory) to random hashes. Pass -v/--verbose for per-file output and -m/--move to move the hashed files into hashed_00X batch folders. Target extensions are read from HASH_FILENAMES_FILE_EXTENSIONS in the project root .env (copy .env.example to .env); this key is required.

The original→hash pairs are persisted as JSON to a hash_filenames_mapping.txt file in the target directory. On a rerun the script reads that mapping and skips files already recorded there (idempotency otherwise relies on detecting already-hashed names). The mapping step needs jq: when jq is absent the script logs a warning and skips reading/writing the mapping rather than failing.

Pass -n/--dry-run to preview the run: it prints would rename/would move/ would remove lines and makes no filesystem change (no rename, no batch move, no mapping file). Before its first real mutation the script prompts for confirmation; pass -y/--yes (or -n/--dry-run) to bypass the prompt.

Parameters

Flag	Required	Default	Description
-d, --directory	no	current dir	Directory to process
-v, --verbose	no	off	Per-file output
-m, --move	no	off	Move hashed files into hashed_00X batches
-n, --dry-run	no	off	Print intended changes; make no filesystem change
-y, --yes	no	off	Skip the confirmation prompt before mutating
-h, --help	—	—	Print usage and exit

.env keys

HASH_FILENAMES_FILE_EXTENSIONS (required) — comma-separated, case-insensitive, leading dot optional.

Usage

# Show help
bash hash-filenames.sh -h

# Hash a directory, verbose, move into batches
bash hash-filenames.sh -d /path/to/dir -v -m