Re: Pandoc Lua script to filter specific markdown sub sections during PDF generation

[Albert Krewinkel <albert+pandoc-9EawChwDxG8hFhg+JK9F0w@public.gmane.org>]

Denis Maier writes:

> You mean like removing the sub section header? Or the whole subsection until the
> start of the next section?
> Both should be possible with a lua filter.  The first is trivial, the second
> more complex (at least for me ;-). But I doubt it wouldn't be possible.
> The tricky part is possibly that pandoc's AST does not reveal the hierarchy
> here:
>
> [Header 1 ("header-1",[],[]) [Str "Header",Space,Str "1"]
> ,Para [Str "Text"]
> ,Header 2 ("header-2",[],[]) [Str "Header",Space,Str "2"]
> ,Para [Str "Text"]
> ,Header 2 ("header-2-1",[],[]) [Str "Header",Space,Str "2"]
> ,Para [Str "Text"]
> ,Header 1 ("header-1-1",[],[]) [Str "Header",Space,Str "1"]
> ,Para [Str "Text"]]
>
> So, I imagine you have to walk over the AST and when you encounter a "Header 2"
> element, you delete this and everything else until you find the next "Head 1"
> element.

That's a good approach. Untested Lua:

    local keep_deleting = false
    function Block (b)
      if b.t == 'Header' and b.level >= 3 then
        keep_deleting = true
        return {}
      elseif b.t == 'Header' then
        keep_deleting = false
      elseif keep_deleting then
        return {}
      end
    end

The alternative is to let pandoc wrap all sections in divs and to delete
those that are undesired. See
https://gist.github.com/tarleb/a0f41adfa7b0e5a9be441e945f843299


--
Albert Krewinkel
GPG: 8eed e3e2 e8c5 6f18 81fe  e836 388d c0b2 1f63 1124