Miyako

Note: This repository is archived and https://miyako.tkesgar.com is no longer available.

Miyako is an app to download images from Facebook. It works by making API requests to Facebook Graph API to get the image URLs. For certain node types like albums, Miyako can also compress all of the images into a single ZIP file for one-click download.

I create this to make downloading various reference photos from Internet easier. Downloading all of them one-by-one is a complete waste of time.

Also, I want to learn React and Redux.

Supported URLs

Currently only photo and album node type are supported.
API requests are subject to permission scope. The most basic scope only allows access on public nodes (Pages, for example).
Miyako works by extracting the node ID from the provided URL via a bunch of regexes. Since Facebook URLs are subject to change, please submit an issue if you find a URL that Miyako cannot parse.

Deployment

You will need these things to get started:

Node.js
npm
Facebook App ID with Facebook Login turned on. You might also need to configure the Site URL settings and Login Callback.

After cloning this repository, run these steps bellow:

npm install to install the dependencies.
Copy app/_config.sample.js into app/config.js and modify it as neccessary.
npm run build to compile the source into the public folder.

You can now serve the public folder with your favourite HTTP server with pushState support, for example pushserve or pushstate-server.

Contributing

Contributions are greatly appreciated. Feel free to check things to do below and submit a pull request. You can also find TODO comments in the source code, or check the issues page.

Todo

Features
- Add more URL support
  - Permalinks (from Facebook embed)
  - Videos
  - Facebook redirection URLs (URLs that change in-between navigations)
  - Singular links (must get page ID and prepend it to the post ID)
- Change the filename of downloaded files.
  - We cannot do this with direct download from image URLs provided by Facebook API. As a workaround, we have to load the image into a <canvas> and obtain the blob data.
- Modify the downloaded images using <canvas> to add some information (for example the author or image owner).
- Better user interface and UX flow
Maintenance
- Testing
- Documentation
- Better architecture (let's just say I'm still new to React 🐧)

Miyako who?

From Hidamari Sketch.

License

Licensed under MIT License.

tkesgar / miyako