I don't have a problem with that. What would it require?

We'll need to convert the sources so they match the structure that Antora expects - a module-aware directory structure, nav.adoc instead of includes, and other stuff described in https://pagure.io/fedora-docs/template.

This might pose a bit of a challenge for anyone wishing to contribute because Antora currently needs docker to build and run a local preview. A workaround would be to maintain a second index file that uses normal ASCIIDoc include statements that would be used for preview. In that case a preview is as simple as running asciidoctor, but we'd need to keep both indexes in sync. @rkratky has a script that can convert the asciidoctor index file into the Antora one.

@eischmann Can you please give @fedora-docs (https://pagure.io/group/fedora-docs) permissions to at least allow us to work with issues? Currently I can't even assign this to myself :).

Can you share some context on why docker is a blocker here @pbokoc ?

@bex I wouldn't call it a blocker, it's just more difficult to set up in a reasonable way (=install docker, enable the service, add your user account to the docker group, and whatever else I'm forgetting right now) as opposed to just installing asciidoctor and running it. It's not a huge issue if it's properly documented, but the extra setup time discourages drive-by contributions, while at the same time this project is perfect for those.

@bex I wouldn't call it a blocker, it's just more difficult to set up in a reasonable way (=install docker, enable the service, add your user account to the docker group, and whatever else I'm forgetting right now) as opposed to just installing asciidoctor and running it. It's not a huge issue if it's properly documented, but the extra setup time discourages drive-by contributions, while at the same time this project is perfect for those.

I did a quick test on a fresh install of F28. I had to do these three steps:

sudo dnf install -y docker
sudo systemctl start docker # optionally enable
build.sh # calls with sudo so we are good

@bex Alright, fair enough :)

@eischmann Can you please add @fedora-docs to Contributors?

@bex Alright, fair enough :)
@eischmann Can you please add @fedora-docs to Contributors?

Added with commit rights.

@bex @pbokoc
I'm new to this whole docker and container concept and didn't find it difficult.

I have been trying to adopt en-US part for Antora - https://pagure.io/fork/ojn/ambassadors/fedora-handbook/
it seems to me that the best way would be to have each language in a separate repo for this to work.

I'm new to Antora thing, but something I noticed that since it's based on Asciidoctor.js and this scripts were using the regular Ruby version some methods of structuring the contents do not play well with it.
I don't know if xref https://docs.antora.org/antora/1.0/asciidoc/page-to-page-xref/ would cause any issues for the AsciidoctorRuby builds. There should be a good way to organize this with ifdef declarations, that's something to look into...

I'm new to this whole docker and container concept and didn't find it difficult.

Woot! This is a lot of work by @asamalik - he did an amazing job!

I have been trying to adopt en-US part for Antora - https://pagure.io/fork/ojn/ambassadors/fedora-handbook/
it seems to me that the best way would be to have each language in a separate repo for this to work.

I suspect separate branches may be best here. If we aren't going to use Zanata to do the translations, we'll just need to code an exception for the build for this repo.

I'm new to Antora thing, but something I noticed that since it's based on Asciidoctor.js and this scripts were using the regular Ruby version some methods of structuring the contents do not play well with it.

Antora has a specific structure it uses as a site building tool. I believe it is reasonable for us to recreate the print build using a method similar to @asamalik 's work around for single-html when it was briefly needed. I leave the details to @asamalik though I am happy to help.

The workaround for making a single-html version that @bex has mentioned is nothing more than creating a page that includes all other pages in it. I've generated it with a script that did something like this (I don't have that script anymore, sorry):

1. Take the nav.adoc (the menu definition) and change all xrefs to includes
2. Apply the "leveloffset" or "indent" (sorry I don't remember the exact name) to each include based on its original offset in the menu
3. Put the resulting file somewhere with the other .adoc files, and Antora will build it for you. You don't need to include it in the menu in order for it to build, but you might want to reference it somewhere so people can find it.