A boilerplate to automate and ease the creation of Generative Tokens on fxhash. This project uses webpack and webpack-dev-server to improve the development and deployment experience.
Before diving into the development of your token, we recommend reading the Guide to mint a Generative Token to get some understanding of the process.
If you are looking for a simpler boilerplate, you can use the fxhash simple boilerplate instead.
- provide a local environment in which you can iterate and use modern features from the javascript ecosystem
- automate the creation of a .zip file ready to be uploaded on fxhash
You will need to have nodejs installed.
First, make sure that your node version is >= 14
Clone the repository on your machine and move to the directory
$ git clone https://github.com/fxhash/fxhash-webpack-boilerplate.git your_folder && cd your_folder
Install the packages required for the local environment
$ npm i
$ npm start
This last command will start a local http server with live reloading enabled so that you can iterate faster on your projects. Open http://localhost:8080 to see your project in the browser.
$ npm run build
Will bundle your js dependencies into a single minified bundle.js
file, move your files from the public/
to the dist/
folder, and link the bundle.js
with the index.html
.
Moreover, it will create a dist-zipped/project.zip
file which can be directly imported on fxhash.
Once the environment is started, you can edit the src/index.js
file to start building your artwork. The index.html
file is located in the public/
folder.
You can import libraries using npm
or by adding the library file in the public/
folder and link those using relative paths in the index.html
.
Any file in the public/
folder will be added to the final project.
fxhash requires you to use a javascript code snippet so that the platform can inject some code when tokens will be generated from your Generative Token. The code snippet is already in the index.html
file of this boilerplate, so you don't have to add it yourself.
During the development stages, the snippet will generate a random hash each time the page is refreshed. This way, it helps you reproduce the conditions in which your token will be executed on fxhash.
It creates 2 variables:
fxhash
: a random 64 characters hexadecimal string. This particular variable will be hardcoded with a static hash when someone mints a token from your GTfxrand()
: a PRNG function that generates deterministic PRN between 0 and 1. Simply use it instead of Math.random().
The index.js of this boilerplate quickly demonstrates how to use these.
This is how Generative Tokens work on fxhash:
- you upload your project to the platform (see next section)
- you mint your project
- when a collector will mint its unique token from your Generative Token, a random hash will be hard-coded in the fxhash code snippet
- the token will now have its own
index.html
file, with a static hash, ensuring its immutability
The Guide to mint a Generative Token give in-depth details about this process.
Once you are happy with the results, you can run the following command:
$ npm run build
This will create a dist-zipped/project.zip
file.
Go to https://fxhash.xyz/sandbox/ and upload the project.zip
file in there to see if it works properly.
If your token does not work properly, you can iterate easily by updating your files, running $ npm run build
again, and upload the zip file again.
Finally, you can mint your token using the same project.zip
file.
Theses rules must be followed to ensure that your token will be future-proof, accepted by fxhash, and behave in the intended way
- the zip file must be under 15 Mb
- any path to a resource must be relative (./path/to/file.ext)
- no external resources allowed, you must put all your resources in the
public/
folder (sub-folders are OK) - no network calls allowed (but calls to get resources from within your
public/
folder) - you must handle any viewport size (by implementing a response to the
resize
event of thewindow
) - you cannot use random number generation without a seed (the same input hash must always yield the same output). The
fxrand
function does a very good job in that regard.