# Prepare a directory with the files you wish to sell and cd to it
$ mkdir ~/ForSale && cd ~/ForSale
# Initialize the `_filebazaar.yaml` config file
$ filebazaar init
# Start filebazaar!
$ filebazaar
```
## Configuration
FileBazaar's configuration options can be managed using the `_filebazaar.yaml` file or via envirnoment variables. All config options are optional and have sane defaults, except for `token_secret` which is required. See [`lib/config.js`](#) for more details.
Below is an example `_filebazaar.yaml` file:
```yaml
---
### Server settings
port: 9678
host: 127.0.0.1
env: production
url: http://my-public-url.com/
### Lightning Kite
kite_url: http://localhost:9112
kite_token: API_TOKEN_CONFIGURED_IN_KITE
### FileBazaar settings
# The directory containing the files for sale
# defaults to the directory containing the _filebazaar.yaml file
directory: /home/shesek/ForSale
# The default file price, can be overridden for individual files (see below)
You can use `$ filebazaar init [directory]` to initialize a new `_filebazaar.yaml` config file. A random `token_secret` will be added for you. If no `[directory]` is specified, the file will be created in the working directory.
To start FileBazaar, run `$ filebazaar [path]`. You can either specify the path to the files directory or to the `_filebazaar.yaml` file. If none is specified, defaults to the working directory.
## File Preview
FileBazaar can currently generate previews for the following file types:
- Images: a preview image will be generated by pixelating the left-half and adding watermark text using [node-canvas](https://github.com/Automattic/node-canvas) and [graphicsmagick](http://www.graphicsmagick.org) (see [example image](https://i.imgur.com/OmrUysL.png)).
- Videos & audio: a preview will be generated by slicing off the first 30 seconds using [ffmpeg](http://ffmpeg.org).
- Documents: a preview image of the first page of the document will be generated using [unoconv](https://github.com/dagwieers/unoconv) (supports pdf, doc, docx, odt, and many others).
In addition, EXIF metadata will be extracted using [exiftool](https://www.sno.phy.queensu.ca/~phil/exiftool/) and displayed for all file types.