S3 serving

[amlwwalker]

would be amazing it this could be pointed at an s3 bucket and serve contents (and recursive contents) to a browser. taking account of permisisons for the current user......

I saw this was in your road map, any idea when that might be implemented?

[andreaskoch]

I am currently working on the final documentation for version 1.0. And I am planning to release v1.0 around Christmas because I will have some time for programming 😀

Serving Amazon s3 support is definitely something which will be built after 1.0 is released.

I have two options for S3 support in mind:

1. Pointing allmark to an existing bucket and serving that.
2. Implementing a render-mode which creates a local folder structure with the converted HTML files - which could then be completely uploaded to S3 and published as a website.

The first option should be relatively easy to implement because I only have to implement a s3-version of the dataaccess / repository.

The render mode will be a bit more challenging because I would have to switch certain parts of the templates with templates that are adapted to work in "offline"-mode. And I would have to either disable all dynamic features (search, suggestions and live-reload) or implement alternatives that work offline.

My road map for version 2.0.0 currently contains the following items:

• Serving S3 buckets directly
• Render Mode + S3 Upload
• Self-Update for the allmark binary
• Package Management for themes
• A better default theme

And since I don't have any preference for one or the other yet you can pick the one I am going to start with in January ... or even better - help me build it 😏