How to keep an additional value when using connect-by in Oracle?

View v_documents stores only the latest flow_id, but each document can have multiple older flows as well. View v_flow_ids has two columns - flow_id and parent_id, and using this view it is possible to find all the successive flow_ids for one particular document, starting with the latest flow_id.

However, when doing this for multiple documents at once, I lose track of which flow_id belongs to which document:

select ff.flow_id, max(ff.repo_id) as repo_id --d.id would also be necessary here
from (select f.flow_id
      from v_flow_ids f  
      start with f.flow_id in (select d.flow_id
                               from v_documents d
                               where d.id in (28211,14631,28212,14632,28192))
      connect by prior f.parent_id = f.flow_id) t
left join v_flow_additional_info ff on t.flow_id = ff.flow_id
group by ff.flow_id, ff.external_id

Is it possible to somehow also get the document id in the final result? So that each flow_id is assigned its corresponding v_documents.id?

JOIN v_documents to the other queries and use CONNECT_BY_ROOT, something like:

SELECT ff.flow_id,
       max(ff.repo_id) as repo_id,
       t.id
FROM   (
         SELECT f.flow_id,
                CONNECT_BY_ROOT(d.id) AS id
         FROM   v_flow_ids f 
                LEFT OUTER JOIN v_documents d
                ON f.flow_id = d.flow_id
         START WITH d.id IN (28211,14631,28212,14632,28192)
         CONNECT BY PRIOR f.parent_id = f.flow_id
       ) t
       LEFT OUTER JOIN v_flow_additional_info ff
       ON t.flow_id = ff.flow_id
GROUP BY
       ff.flow_id,
       ff.external_id,
       t.id

However, without some sample data and an idea of what your expected output is, it is impossible to test.