Question

I need to filter my query with categories table which has many2many relation with another table. Is it possible to filter query with many2many relation?

Table res_partner has many2many field category_id relating to table res_partner_category.res_partner, or let’s just say partners can have many categories. What I need is to filter res_partners table where it has category named ‘business’ or ‘retail’. If it doesn’t have any of these categories, it should not be shown.

Also there is another field in res_partner which is category_value_ids and has one2many relation with res_partners_category_value:

res_partner has following fields with relations:

• category_id to res_partner_category (many2many)
• category_value_ids to res_partner_category_value (one2many)
• name (char)

res_partner_category has following fields with relations:

• partner_ids to res_partner (many2many)
• name (char)

res_partner_category_value has following fields with relations:

• category_group_id to res_partner_category (many2one)
• category_id to res_partner_category (many2one)
• object_id tores_partner (many2one)

But if I try to use res_partner_category_value table in SQL query I get error that I can’t use it in query.

So for example, if there are 4 partners with these categories:

• first: categ1, categ2, business
• second: retail
• third: retail, business
• fourth: categ1, categ2

The query should return first, second and third partners.
One person told me it’s not possible to filter like this with many2many relation. So I wonder is it really not possible or just complicated?

EDIT:
I found one more table called res_partner_category_rel. I didn’t see it, because in Openerp administration interface, where you can see all objects of database, that table is not shown. You can only see it directly through database.
So I was confused by this “missing” table:

res_partner_category_rel:

• partner_id (many2one)
• category_id (many2one)

Answer 1

Setup

This is the test case you should have provided:

CREATE TABLE partner (
  partner_id serial PRIMARY KEY
, partner    text
);
INSERT INTO partner (partner) VALUES 
  ('partner1')
, ('partner2')
, ('partner3')
, ('partner4')
;

CREATE TABLE category (
  category_id serial PRIMARY KEY
, category    text
);
INSERT INTO category (category) VALUES 
  ('categ1')
, ('categ2')
, ('business')
, ('retail')
;

CREATE TABLE partner_category (
  partner_id  int REFERENCES partner(partner_id)
, category_id int REFERENCES category(category_id)
, CONSTRAINT cat_pk PRIMARY KEY (partner_id, category_id)
);
INSERT INTO partner_category (partner_id, category_id) VALUES 
  (1,1), (1,2), (1,3)
, (2,4)
, (3,3), (3,4)
, (4,1), (4,2);

Solution

One way:

SELECT p.*
FROM   partner p
WHERE  EXISTS (SELECT FROM partner_category pc WHERE pc.partner_id = p.partner_id AND pc.category_id = 3)
OR     EXISTS (SELECT FROM partner_category pc WHERE pc.partner_id = p.partner_id AND pc.category_id = 4)
ORDER  BY p.partner_id;

Another:

SELECT p.*
FROM        (SELECT partner_id FROM partner_category WHERE category_id = 3) pc1
FULL   JOIN (SELECT partner_id FROM partner_category WHERE category_id = 4) pc2 USING (partner_id)
JOIN   partner p USING (partner_id)
ORDER  BY p.partner_id;

fiddle
_{Old sqlfiddle}

The second one assumes unique (partner_id, category_id) in partner_category.