To retrieve the section content with this approach I had to make 2 requests
get a list of all the sections of an article to identify its numeric id (needed for the 2nd request)
request of a particular section content based on the id.
The bad
The solution return the entire section and not just the first paragraphs as onebox appears to do.
I was thinking that using the wiki api might make the process more efficient at least, but it is not
additional processing is required to fix relative a tags
it also return some nonsense [edit] text
my conclusion is that is probably better to just go with the current approach and use Nokogiri for section parsing
Technically my Dev environment is working but it is limping due to virtualbox issues.
Hopefully the next virtualbox release will magically fix the bug I experience.
But luckily the onebox can be worked as a standalone, wasn’t too much trouble to get it working on windows.