SQLSoup multiple joins

Question

I want to use SQLSoup for querying a MySQL database.

The query I want to issue contains multiple joins. Unfortunately the tutorial examples involve table with completely different column names (no duplicate column names across tables).

I my case, all my tables have a pkey column named id.

I do the following:

s = SQLSoup(DATABASE_URI)
s.join(s.logical_interface, s.with_labels(s.physical_interface), s.physical_interface.id == s.logical_interface.physical_interface_id).first()

Which results in:

InvalidRequestError: Implicitly combining column logical_interface.physical_interface_id with column foo.physical_interface_id under attribute 'physical_interface_id'.  Please configure one or more attributes for these same-named columns explicitly

Therefore I do the following: s.join(s.with_labels(s.logical_interface), s.with_labels(s.physical_interface), s.physical_interface.id == s.logical_interface.physical_interface_id).first()

Which results in (see the complete query below):

OperationalError: (OperationalError) (1066, "Not unique table/alias: 'foo'") 'SELECT foo.logical_interface_id AS ... 

It seems that the same table alias foo is used for every tables, which obviously causes troubles.

Is there a correct way to do this ?

---

Complete query:

'SELECT foo.logical_interface_id AS foo_logical_interface_id, foo.logical_interface_physical_interface_id AS foo_logical_interface_physical_interface_id, foo.logical_interface_name AS foo_logical_interface_name, foo.logical_interface_description AS foo_logical_interface_description, foo.logical_interface_active AS foo_logical_interface_active, foo.logical_interface_ifindex AS foo_logical_interface_ifindex, foo.logical_interface_ifinoctets AS foo_logical_interface_ifinoctets, foo.logical_interface_ifoutoctets AS foo_logical_interface_ifoutoctets, foo.logical_interface_lastpolloctets AS foo_logical_interface_lastpolloctets, foo.logical_interface_ifsigqunerroreds AS foo_logical_interface_ifsigqunerroreds, foo.logical_interface_ifsigqcorrecteds AS foo_logical_interface_ifsigqcorrecteds, foo.logical_interface_ifsigquncorrectables AS foo_logical_interface_ifsigquncorrectables, foo.logical_interface_lastpollfec AS foo_logical_interface_lastpollfec, foo.physical_interface_id AS foo_physical_interface_id, foo.physical_interface_docsis_mac_id AS foo_physical_interface_docsis_mac_id, foo.physical_interface_name AS foo_physical_interface_name, foo.physical_interface_type AS foo_physical_interface_type, foo.physical_interface_module AS foo_physical_interface_module \nFROM (SELECT logical_interface.id AS logical_interface_id, logical_interface.physical_interface_id AS logical_interface_physical_interface_id, logical_interface.name AS logical_interface_name, logical_interface.description AS logical_interface_description, logical_interface.active AS logical_interface_active, logical_interface.ifindex AS logical_interface_ifindex, logical_interface.ifinoctets AS logical_interface_ifinoctets, logical_interface.ifoutoctets AS logical_interface_ifoutoctets, logical_interface.lastpolloctets AS logical_interface_lastpolloctets, logical_interface.ifsigqunerroreds AS logical_interface_ifsigqunerroreds, logical_interface.ifsigqcorrecteds AS logical_interface_ifsigqcorrecteds, logical_interface.ifsigquncorrectables AS logical_interface_ifsigquncorrectables, logical_interface.lastpollfec AS logical_interface_lastpollfec \nFROM logical_interface) AS foo INNER JOIN (SELECT physical_interface.id AS physical_interface_id, physical_interface.docsis_mac_id AS physical_interface_docsis_mac_id, physical_interface.name AS physical_interface_name, physical_interface.type AS physical_interface_type, physical_interface.module AS physical_interface_module \nFROM physical_interface) AS foo ON physical_interface.id = logical_interface.physical_interface_id \n LIMIT %s' (1,)

Answer 1

The answer is, "no".

The reason is that SQLSoup is not maintained and is obsolete (and I can say this because I am the only maintainer SQLSoup has ever had). Use automap, which I wrote as a modern replacement to SQLSoup.