Niclas,
I've been looking at asciidoctor for a while.
There's built-in support for source code includes:
include::{rootDir}/path/to/SomeFile.java[tags=whatever,indent=0]
And in the source file:
// tag::whatever[]
...
// end::whatever[]
PDF output using java only is pretty well supported through docbook and fop.
The asciidoctor HTML backend does not support chunked output with
cross-references. Which we use a lot. The only way to get chunked output
with cross-references is through docbook.
The docbook XSLs are published on maven central as a zip so we can
automate their usage too.
We would have to write/maintain the necessary plumbing code but it is
feasible to produce all the documentation without anything installed but
java.
I have a prototype build lying on disk, I'll review it and see if I can
come up with something satisfying.
Cheers
Niclas Hedhman a écrit :
> I should probably add that there is perhaps some challenge to get the
> Source Code inclusions to work. Not sure if AsciiDoctor will support those
> Python scripts we have...
>
> On Sat, Apr 16, 2016 at 1:54 PM, Niclas Hedhman <[email protected]> wrote:
>
>> Gang,
>> Perhaps my prayers has been heard by the Almighty (just kidding), but
>> there seems to be a toolchain for AsciiDoc that makes our docs a bit neater.
>>
>> http://asciidoctor.org/
>>
>> If someone wants to take a shot at integrating this to the build system,
>> and plainly remove the old one (do in a separate branch so we can iron it
>> out), I would be eternally grateful, buy you beers until you fall off the
>> chair, whatever...
>>
>>
>> Cheers
>> --
>> Niclas Hedhman, Software Developer
>> http://zest.apache.org - New Energy for Java
>>
>
>
>
