Saving, backups & export
The moment your edits become bytes on disk — and everything that makes that moment reversible.
Reviewing before you save
Three ways to see what's about to happen, in increasing detail:
- Highlighted fields in the metadata panel — every value that differs from disk
- The change count in the toolbar — how many pending changes exist across every loaded file, not just the selected one
- View Changes — the full summary: which files, which fields, and what each is going from and to
The summary is worth opening before any large batch. It's the last point at which a mistake costs nothing.
Saving
Press Ctrl+S (Cmd+S on macOS) or click Save. In order, EveryMark:
- Backs up the originals — every file about to be written is copied to your backup folder first, with a record of what is changing.
- Validates every pending change — against the rules for each field type. Anything invalid is reported rather than written.
- Writes in parallel — files are processed simultaneously, with a progress indicator for the batch.
- Reports per file — successes clear their highlights; failures are named individually.
When the save completes, highlights disappear and the pending count returns to zero. Cached metadata for those files is invalidated automatically, so what you see afterwards is re-read from disk rather than remembered.
If a file can't be written — a permissions problem, a read-only volume, a format that won't accept a particular tag — that file is reported and the rest of the batch still completes. You aren't left guessing which half of four hundred files went through.
Discarding
Click Discard and confirm. Every pending change across every loaded file is dropped, highlights clear, and nothing on disk was ever touched. There is no cost to discarding — pending changes exist only in memory until a save.
Backups
Automatic backup is on by default. Before each save, the original file is copied into your backup folder along with metadata about the change: what was modified, when, and why.
| Detail | Behaviour |
|---|---|
| Default location | A backups/ folder inside the application's user-data directory. |
| Changing it | Settings → Backup location. Point it at a drive with room, especially if you work with RAW files. |
| Turning it off | Settings → Automatic backup. Reasonable if you have filesystem snapshots or your own versioning; otherwise leave it on. |
| Restoring | Backups are ordinary file copies — you can restore one by copying it back over the edited file yourself. A restore interface inside the app is in progress. |
A backup is a full copy of the original file, not a diff. Editing ten thousand RAW files means ten thousand RAW-sized copies. If space gets tight, prune the backup folder periodically or point it at a larger volume.
Exporting metadata
Export writes out what your files currently contain — it doesn't modify anything.
- Select the files you want to export.
- Open Advanced Actions in the metadata toolbar and choose Export.
- Pick CSV or JSON, and decide whether to include empty fields.
- Choose a location in the native save dialog.
CSV opens directly in a spreadsheet, one row per file. JSON is pretty-printed and keeps nested structures intact, which suits scripting. Both cover image, video, and audio fields, and both work for a single file or a whole selection.
What it's good for
- Auditing a collection — sort the CSV by date to find every file still sitting at 1970
- Client manifests — a delivery record of exactly what was handed over, with credits and copyright
- Checking your work — export before and after a batch and compare the two files
- Insurance — a readable record of your metadata that doesn't depend on the files surviving