Github user jackylk commented on the issue:
https://github.com/apache/carbondata/pull/1749
If the table data is generated without enabling carbon hive metastore
option, the table data includes schema file, so this table folder can be used
to create another external table.
If the table data is generated with enabling carbon hive metastore option,
the table data does not include schema file. When creating external table using
such table folder, we can generate the schema file by either deserializing the
carbonSchema string in hive metastore or read the header of data file.
This is not implemented in this PR.
---
