Repository navigation
Batch Data Generator
The Batch Data Generator creates files or folders in bulk. It is useful for testing, placeholders, benchmarks, or filling out a template folder structure. Open it from the Batch Data Generator button in the main toolbar.
At the top, choose what to create:
- Dummy Files: files of a defined size.
- Folders: empty folders. (Size and allocation options do not apply here.)
Set the number of items to create, and optionally tick Hidden so the created items get the hidden attribute.
| Mode | Behaviour |
|---|---|
| Fixed size | Every file is the size you enter, in the chosen unit (KB, MB, GB, or TB). |
| Random size | Each file gets a random size between a minimum and a maximum, set by the sliders or by typing the numbers. The unit applies to both. |
| Type | Meaning |
|---|---|
| Sparse | Logical files with minimal disk usage: created instantly, using almost no real space, while still reporting the chosen size. |
| Fully Allocated | Physical files that really use the space on disk. |
Before a fully-allocated run, ANE checks free space and warns you if it would not fit. If a drive fills up mid-run, ANE stops, removes the unfinished file, and tells you what happened. See Warnings and Safety.
Choose how the created items are named:
- Customized name: build names from patterns (see Name Patterns Reference). For files, you also choose a format (extension).
- Mirror of content: place a twin next to every existing file, using your chosen size options. You decide whether the twin keeps the original format (with a twin suffix) or takes a new format from the list. A Scope setting controls how deep the mirroring reaches.
Pick a ready-made pattern from the presets list, each shows a live preview of the first name. Presets include hour patterns that run from 01-00 to 24-00.
Turn on Date sequence to create one item per date across a start-to-end range, with a step in days. The {date:…} token in your pattern then counts up day by day, and the number of items comes from the range.
- Preview lists exactly what would be created, including the total size, before anything is generated.
- A progress bar and log track the run; Clear empties the log.
- Stop takes effect within a moment, even in the middle of a very large file, and the unfinished file is removed automatically.
- All settings are locked while a run is in progress, and unlock again when it finishes.
- Open the Batch Data Generator and stay on Dummy Files.
- How many:
100· Size: Fixed50MB · Allocation: Fully Allocated. - Name pattern: choose a preset such as
File_{n:000}. - Click Preview to confirm the total size (about 5 GB), then Generate.
- Name Patterns Reference: every naming token explained.
- Warnings and Safety: disk-space handling and the Stop button.
- Examples and Use Cases: more generation scenarios.
© 2025–2026 Avarwand. All rights reserved.
Getting started
Using ANE
- Features Overview
- Batch Renaming
- Live Search
- Selection File Edit
- Batch Data Generator
- Name Patterns Reference
Help
About