Re: Pandoc filters to produces multiple document reading lines of a CSV

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

Hi Gianluca,

Gianluca Carbone <gianlucarbone96-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:

> I want to obtain multiple documents in output (markdown documents)
> which uses a specifc template already defined and working by me.
>
> The input file, must be a CSV file with an header for example:
> // CSV file \\
> name, surname, date, location
> Josh, White, 2022-10-12, Rome
> Kate, Brown, 2021-09-30, London
>
> I want in output in this case 2 documents:
>
>  1. One with content filled by the values of first record of CSV
>  2. One with content filled by the values of the second record of CSV
>
> How can I do something like this ?

Not sure I understand this correctly, but I assume you have a template
that contains things like

    ${name} ${surname}, born ${date} in ${location}

and you'd like to use that with a CSV table, using the header label as
variable name? If that's correct, then the following custom writer
should help:

``` lua
local stringify = pandoc.utils.stringify

function Writer (doc, opts)
  assert(doc.blocks[1] and doc.blocks[1].t == 'Table', 'expected table')
  -- table does not have colspans, footer, etc. Convert to simple table
  local tbl = pandoc.utils.to_simple_table(doc.blocks[1])
  local result = pandoc.List{}
  for _, row in ipairs(tbl.rows) do
    local meta = pandoc.Meta(doc.meta)   -- clone
    for i, cell in ipairs(tbl.header) do
      meta[stringify(cell)] = row[i]
    end
    local rowdoc = pandoc.Pandoc({}, meta)
    result:insert(pandoc.write(rowdoc, 'markdown', opts))
  end
  return table.concat(result, '\f')
end
```

Save the code to a file `meta-from-table.lua` and call pandoc with

    pandoc --to=meta-from-table.lua ...

where `...` are your other options, including the template. The output
is built from all rows, with a line separator character between each
record.

Hope that helps,

Albert

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