File Recovery Hero — exFAT recovery lab v1, 2026-08-31

A synthetic 32 MiB image with three deleted files whose contiguous data is
intact (JPG, MP4, WAV), plus a fully overwritten JPG negative control.
No customer data. Directory entries were constructed deliberately, not by a
camera recording session. Not a representative benchmark or recovery promise.

1. Extract the ZIP into a new folder. Use Python 3.8 or later.
2. Inside the extracted folder: python3 verify.py
3. Open only frh-exfat-lab-v1.img as the source in recovery software.
   Do not write it onto a physical drive. Do not repair the image.
4. Export recovered files into a separate, empty folder.
   reference/ contains the originals: never copy them into the export folder!
5. python3 verify.py --recovered /path/to/export-folder
   On Windows, the command may be python instead of python3.

The script only reads: no installation, mounting or network requests.
It checks complete SHA-256 equality regardless of recovered file names.
Expected: three exact matches; no exact match for GONE.JPG.
Finding an old file name alone does not prove successful recovery.

results.json records the publisher's macOS source-core test at the specified
Git revision using an internal harness. It is not a test of the public app UI
or Windows. JPG, MP4 and WAV match byte for byte and decode completely.
JPG nevertheless receives .UNVERIFIED because of an internal end-of-file read.
This warning is disclosed, not counted as a fault-free result.
GONE.JPG is detected but not exported.

build-fixture.py documents generation. Rebuild on macOS with hdiutil and
FFmpeg: python3 build-fixture.py --out new-folder --ffmpeg /path/to/ffmpeg
It creates new files only. Rebuilding reproduces the scenario, not necessarily
identical image bytes across different macOS or FFmpeg versions.

Report and stable source URL:
https://file-recovery-hero.com/en/recovery-lab/
