S3
How to publish your distributable Electron app artifacts to Amazon S3
The S3 target publishes your Make artifacts to an Amazon S3 bucket.
Installation
Usage
To use @electron-forge/publisher-s3
, add it to the publishers
array in your Forge configuration:
Configuration options are documented in PublisherS3Config
.
Authentication
It is recommended to follow the Amazon AWS guide and set either a shared credentials guide or the proper environment variables. However, if that is not possible, the publisher config allows the setting of the accessKeyId
and secretAccessKey
configuration options.
Key management
By default, the S3 publisher will upload its objects to the {prefix}/{platform}/{arch}/{name}
key, where:
{prefix}
is the value of theconfig.folder
option (defaults to the"name"
field in your package.json).{platform}
is the target platform for the artifact you are publishing.{arch}
is the target architecture for the artifact you are publishing.{name}
is the file name of the artifact you are publishing.
If you run the Publish command multiple times on the same platform for the same version (e.g. simultaneously publishing ia32
and x64
Windows artifacts), your uploads can get overwritten in the S3 bucket.
To avoid this problem, you can use the keyResolver
option to generate the S3 key programmatically.
Auto updating from S3
You can configure Electron's built-in autoUpdater
module to use the artifacts published by the S3 publisher. This is a two-step process:
First, you must configure @electron-forge/publisher-s3
to publish your files into an auto-updater compatible layout and use @electron-forge/maker-zip
+ @electron-forge/maker-squirrel
to build your application.
With Forge configured correctly, the second step is to configure the autoUpdater
module inside your app's main process. The simplest form is shown below but you might want to hook additional events to show UI to your user or ask them if they want to update your app right now.
Last updated