fix: restrict generated m2m relationships

Only add m2m relationship based on real junctions:
a table has FKs to two other tables and the FK columns
are part of the PK columns.

Also add some stats to big_schema.sql

CHANGELOG.md

@@ -49,6 +49,8 @@ This project adheres to [Semantic Versioning](http://semver.org/).
  - #2155, Ignore `max-rows` on POST, PATCH, PUT and DELETE - @steve-chavez
  - #2239, Fix misleading disambiguation error where the content of the `relationship` key looks like valid syntax - @laurenceisla
  - #2254, Fix inferring a foreign key column as a primary key column on views - @steve-chavez
+ - #2070, Restrict generated many-to-many relationships - @steve-chavez
+   + Only adds many-to-many relationships when: a table has FKs to two other tables and these FK columns are part of the table's PK columns.
 
 ### Changed
 
@@ -59,6 +61,8 @@ This project adheres to [Semantic Versioning](http://semver.org/).
  - #2156, using PATCH/DELETE with `limit/offset` throws an error on views - @steve-chavez
  - #2155, `max-rows` is no longer applied on POST/PATCH/PUT/DELETE returned rows - @steve-chavez
    + This was misleading because the affected rows were not really affected by `max-rows`, only the returned rows were limited
+ - #2070, Restrict generated many-to-many relationships - @steve-chavez
+   + A primary key that contains the foreign key columns is now needed for generating many-to-many relationships.
 
 ## [9.0.0] - 2021-11-25
 

src/PostgREST/DbStructure.hs

@@ -29,6 +29,7 @@ module PostgREST.DbStructure
 
 import qualified Data.Aeson          as JSON
 import qualified Data.HashMap.Strict as M
+import qualified Data.Set            as S
 import qualified Hasql.Decoders      as HD
 import qualified Hasql.Encoders      as HE
 import qualified Hasql.Statement     as SQL
@@ -93,8 +94,8 @@ queryDbStructure schemas extraSearchPath prepared = do
   m2oRels <- SQL.statement mempty $ allM2ORels pgVer prepared
   procs   <- SQL.statement schemas $ allProcs pgVer prepared
 
-  let rels          = addO2MRels $ addM2MRels $ addViewM2ORels keyDeps m2oRels
+  let tabsWViewsPks = addViewPrimaryKeys tabs keyDeps
-      tabsWViewsPks = addViewPrimaryKeys tabs keyDeps
+      rels          = addO2MRels $ addM2MRels tabsWViewsPks $ addViewM2ORels keyDeps m2oRels
 
   return $ removeInternal schemas $ DbStructure {
       dbTables = tabsWViewsPks
@@ -370,13 +371,19 @@ addO2MRels :: [Relationship] -> [Relationship]
 addO2MRels rels = rels ++ [ Relationship ft t (O2M cons (swap <$> cols))
                           | Relationship t ft (M2O cons cols) <- rels ]
 
-addM2MRels :: [Relationship] -> [Relationship]
+-- | Adds a m2m relationship if a table has FKs to two other tables and the FK columns are part of the PK columns
-addM2MRels rels = rels ++ [ Relationship t ft
+addM2MRels :: TablesMap -> [Relationship] -> [Relationship]
-                              (M2M $ Junction jt1 cons1 cons2 (swap <$> cols) (swap <$> fcols))
+addM2MRels tbls rels = rels ++ catMaybes
-                          | Relationship jt1 t  (M2O cons1 cols)  <- rels
+  [ let
-                          , Relationship jt2 ft (M2O cons2 fcols) <- rels
+      jtCols = S.fromList $ (fst <$> cols) ++ (fst <$> fcols)
-                          , jt1 == jt2
+      pkCols = S.fromList $ maybe mempty tablePKCols $ M.lookup jt1 tbls
-                          , cons1 /= cons2]
+    in if S.isSubsetOf jtCols pkCols
+      then Just $ Relationship t ft (M2M $ Junction jt1 cons1 cons2 (swap <$> cols) (swap <$> fcols))
+      else Nothing
+  | Relationship jt1 t  (M2O cons1 cols)  <- rels
+  , Relationship jt2 ft (M2O cons2 fcols) <- rels
+  , jt1 == jt2
+  , cons1 /= cons2]
 
 addViewPrimaryKeys :: TablesMap -> [ViewKeyDependency] -> TablesMap
 addViewPrimaryKeys tabs keyDeps =

test/io/big_schema.sql

@@ -1,28 +1,84 @@
 /*
 This is a 2018 version of the apflora schema https://github.com/barbalex/apf2/tree/master/sql/apflora - latest version likely has differing contents
 
-We use it to test our metadata generation because it contains a good amount of views(281).
+We use it to test our metadata generation because it contains a good amount of db objects.
 
 Custom roles and privileges where removed.
 
 postgrest-with-postgresql-14 -f test/io/big_schema.sql psql
 
+Has 12 functions:
+
 select count(*)  from information_schema.routines where specific_schema = 'apflora';
  count
 -------
     12
 
+Has 45 tables:
+
 select count(*)  from information_schema.tables where table_schema = 'apflora' and table_type = 'BASE TABLE';
  count
 -------
     45
 (1 row)
 
+Has 281 views:
+
 select count(*)  from information_schema.views where table_schema = 'apflora';
  count
 -------
    281
 (1 row)
+
+Has 45 pkcols where none of them is a composite primary key:
+
+with pkcols as (
+  select kcu.table_schema,
+         kcu.table_name,
+         tco.constraint_name,
+         array_agg(kcu.column_name order by kcu.ordinal_position) as key_columns
+  from information_schema.table_constraints tco
+  join information_schema.key_column_usage kcu
+       on kcu.constraint_name = tco.constraint_name
+       and kcu.constraint_schema = tco.constraint_schema
+       and kcu.constraint_name = tco.constraint_name
+  where tco.constraint_type = 'PRIMARY KEY' and kcu.table_schema = 'apflora'
+  group by kcu.table_schema, kcu.table_name, tco.constraint_name
+)
+select count(*) from pkcols
+where array_length(key_columns, 1) > 1;
+ count
+-------
+     0
+(1 row)
+
+Has 50 foreign key relationships:
+
+with fk_rel as (
+  select ns1.nspname as table_schema,
+             tab.relname as table_name,
+             ns2.nspname as foreign_table_schema,
+             other.relname as foreign_table_name,
+             conname     as constraint_name,
+             column_info.cols as columns
+  from pg_constraint,
+  lateral (
+    select array_agg(row(cols.attname, refs.attname) order by cols.attnum) as cols
+    from ( select unnest(conkey) as col, unnest(confkey) as ref) k,
+    lateral (select * from pg_attribute where attrelid = conrelid and attnum = col) as cols,
+    lateral (select * from pg_attribute where attrelid = confrelid and attnum = ref) as refs) as column_info,
+  lateral (select * from pg_namespace where pg_namespace.oid = connamespace) as ns1,
+  lateral (select * from pg_class where pg_class.oid = conrelid) as tab,
+  lateral (select * from pg_class where pg_class.oid = confrelid) as other,
+  lateral (select * from pg_namespace where pg_namespace.oid = other.relnamespace) as ns2
+  where contype = 'f' and conparentid = 0
+)
+select count(*) from fk_rel;
+
+ count
+-------
+    50
+(1 row)
 */
 
 SET statement_timeout = 0;

test/spec/Feature/Query/QuerySpec.hs

@@ -434,12 +434,12 @@ spec actualPgVersion = do
 
     it "requesting data using many<->many relation defined by composite keys" $
       get "/users_tasks?user_id=eq.1&task_id=eq.1&select=user_id,files(filename,content)" `shouldRespondWith`
-        [json|[{"user_id":1,"files":[{"filename":"command.com","content":"#include <unix.h>"},{"filename":"autoexec.bat","content":"@ECHO OFF"},{"filename":"README.md","content":"# make $$$!"}]}]|]
+        [json|[{"user_id":1,"files":[{"filename":"autoexec.bat","content":"@ECHO OFF"},{"filename":"command.com","content":"#include <unix.h>"},{"filename":"README.md","content":"# make $$$!"}]}]|]
         { matchHeaders = [matchContentTypeJson] }
 
     it "requesting data using many<->many (composite keys) relation using hint" $
       get "/users_tasks?user_id=eq.1&task_id=eq.1&select=user_id,files!touched_files(filename,content)" `shouldRespondWith`
-        [json|[{"user_id":1,"files":[{"filename":"command.com","content":"#include <unix.h>"},{"filename":"autoexec.bat","content":"@ECHO OFF"},{"filename":"README.md","content":"# make $$$!"}]}]|]
+        [json|[{"user_id":1,"files":[{"filename":"autoexec.bat","content":"@ECHO OFF"},{"filename":"command.com","content":"#include <unix.h>"},{"filename":"README.md","content":"# make $$$!"}]}]|]
         { matchHeaders = [matchContentTypeJson] }
 
     it "requesting children with composite key" $
@@ -447,6 +447,12 @@ spec actualPgVersion = do
         [json|[{"user_id":2,"task_id":6,"comments":[{"content":"Needs to be delivered ASAP"}]}]|]
         { matchHeaders = [matchContentTypeJson] }
 
+    -- https://github.com/PostgREST/postgrest/issues/2070
+    it "one-to-many embeds without a disambiguation error due to wrongly generated many-to-many relationships" $
+      get "/plate?select=*,well(*)" `shouldRespondWith`
+        [json|[]|]
+        { matchHeaders = [matchContentTypeJson] }
+
     describe "computed columns" $ do
       it "computed column on table" $
         get "/items?id=eq.1&select=id,always_true" `shouldRespondWith`

test/spec/fixtures/privileges.sql

@@ -172,6 +172,8 @@ GRANT ALL ON TABLE
     , limited_delete_items_cpk
     , limited_delete_items_no_pk
     , limited_delete_items_view
+    , plate
+    , well
 TO postgrest_test_anonymous;
 
 GRANT INSERT ON TABLE insertonly TO postgrest_test_anonymous;

test/spec/fixtures/schema.sql

@@ -791,18 +791,13 @@ CREATE TABLE files (
 );
 
 CREATE TABLE touched_files (
-    user_id integer NOT NULL,
+  user_id integer NOT NULL,
-    task_id integer NOT NULL,
+  task_id integer NOT NULL,
-    project_id integer NOT NULL,
+  project_id integer NOT NULL,
-    filename text NOT NULL,
+  filename text NOT NULL,
-    CONSTRAINT fk_users_tasks
+  constraint fk_users_tasks foreign key (user_id, task_id) references users_tasks (user_id, task_id) on delete cascade on update cascade,
-	FOREIGN KEY (user_id, task_id)
+  constraint fk_upload foreign key (project_id, filename) references files (project_id,filename) on delete cascade on update cascade,
-	REFERENCES users_tasks (user_id, task_id)
+  primary key(user_id, task_id, project_id, filename)
-	ON DELETE CASCADE ON UPDATE CASCADE,
-    CONSTRAINT fk_upload
-	FOREIGN KEY (project_id, filename)
-	REFERENCES files (project_id,filename)
-	ON DELETE CASCADE ON UPDATE CASCADE
 );
 
 create table private.articles (
@@ -1215,7 +1210,8 @@ create table test.part (
 
 create table test.being_part (
   being int not null references test.being(being),
-  part int not null references test.part(part)
+  part int not null references test.part(part),
+  primary key(being, part)
 );
 
 create function test.single_out_param(num int, OUT num_plus_one int) AS $$
@@ -1927,10 +1923,11 @@ create table sites (
 alter table sites rename constraint sites_main_project_id_fkey to main_project;
 
 create table jobs (
-  job_id          uuid    primary key
+  job_id          uuid
 , name text
 , site_id         int     not null references sites (site_id)
 , big_project_id  int     not null references big_projects (big_project_id)
+, primary key(job_id, site_id, big_project_id)
 );
 
 create view main_jobs as
@@ -1955,12 +1952,13 @@ create table whatev_sites (
 );
 
 create table whatev_jobs (
-  job_id        uuid    primary key
+  job_id        uuid
 , name text
 , site_id_1     int not null references whatev_sites (id)
 , project_id_1  int not null references whatev_projects (id)
 , site_id_2     int not null references whatev_sites (id)
 , project_id_2  int not null references whatev_projects (id)
+, primary key(job_id, site_id_1, project_id_1, site_id_2, project_id_2)
 );
 
 -- circular reference
@@ -2294,7 +2292,8 @@ A test for partitioned tables$$;
         car_dealer_city varchar(64) not null,
         quantity int not null,
         foreign key (car_model_name, car_model_year) references test.car_models (name, year),
-        foreign key (car_dealer_name, car_dealer_city) references test.car_dealers (name, city)
+        foreign key (car_dealer_name, car_dealer_city) references test.car_dealers (name, city),
+        primary key (car_model_name, car_model_year, car_dealer_name, car_dealer_city, quantity)
       ) partition by range (quantity);
 
       create table test.car_models_car_dealers_10to20 partition of test.car_models_car_dealers
@@ -2521,3 +2520,36 @@ create function reset_limited_items(tbl_name text default '') returns void as $_
   $$::text,
   tbl_name, tbl_name);
 end; $_$ language plpgsql volatile;
+
+-- tables for ensuring we generate real junctions for many-to-many relationships
+create table plate (
+    plate_id int primary key
+);
+
+create table well (
+    well_id int primary key,
+    plate_id int not null,
+    parent_well_id int,
+   CONSTRAINT well_parent_well_id_fkey
+      FOREIGN KEY(parent_well_id)
+	  REFERENCES well(well_id),
+   CONSTRAINT well_plate_id_fkey
+      FOREIGN KEY(plate_id)
+	  REFERENCES plate(plate_id)
+);
+
+create table plate_plan_step (
+    plate_plan_step_id int primary key,
+    from_well_id int,
+    to_well_id int,
+    to_plate_id int,
+   CONSTRAINT plate_plan_step_from_well_id_fkey
+      FOREIGN KEY(from_well_id)
+	  REFERENCES well(well_id),
+   CONSTRAINT plate_plan_step_to_plate_id_fkey
+      FOREIGN KEY(to_plate_id)
+	  REFERENCES plate(plate_id),
+   CONSTRAINT plate_plan_step_to_well_id_fkey
+      FOREIGN KEY(to_well_id)
+	  REFERENCES well(well_id)
+);