Merge pull request #346 from collerek/add_index

Add multi columns non-unique index and sql_nullable setting. Important performance booster for dict().

README.md

@@ -657,7 +657,8 @@ The following keyword arguments are supported on all field types.
 
 All fields are required unless one of the following is set:
 
-* `nullable` - Creates a nullable column. Sets the default to `None`.
+* `nullable` - Creates a nullable column. Sets the default to `False`. Read the fields common parameters for details.
+* `sql_nullable` - Used to set different setting for pydantic and the database. Sets the default to `nullable` value. Read the fields common parameters for details.
 * `default` - Set a default value for the field. **Not available for relation fields**
 * `server_default` - Set a default value for the field on server side (like sqlalchemy's `func.now()`). **Not available for relation fields**
 * `primary key` with `autoincrement` - When a column is set to primary key and autoincrement is set on this column.

docs/fields/common-parameters.md

@@ -22,18 +22,41 @@ Used both in sql and pydantic (changes pk field to optional for autoincrement).
 
 ## nullable
 
-`nullable`: `bool` = `not primary_key` -> defaults to False for primary key column, and True for all other. 
+`nullable`: `bool` = `False` -> defaults to False for all fields except relation fields.
+
+Automatically changed to True if user provide one of the following:
+
+* `default` value or function is provided
+* `server_default` value or function is provided
+* `autoincrement` is set on `Integer` `primary_key` field
+* **[DEPRECATED]**`pydantic_only=True` is set 
 
 Specifies if field is optional or required, used both with sql and pydantic.
 
+By default, used for both `pydantic` and `sqlalchemy` as those are the most common settings:
+
+* `nullable=False` - means database column is not null and field is required in pydantic
+* `nullable=True` - means database column is null and field is optional in pydantic
+
+If you want to set different setting for pydantic and the database see `sql_nullable` below.
+
 !!!note
     By default all `ForeignKeys` are also nullable, meaning the related `Model` is not required.
     
     If you change the `ForeignKey` column to `nullable=False`, it becomes required.
     
 
-!!!info
+## sql_nullable
-    If you want to know more about how you can preload related models during queries and how the relations work read the [queries][queries] and [relations][relations] sections. 
+
+`sql_nullable`: `bool` = `nullable` -> defaults to the value of nullable (described above). 
+
+Specifies if field is not null or allows nulls in the database only. 
+
+Use this setting in combination with `nullable` only if you want to set different options on pydantic model and in the database. 
+
+A sample usage might be i.e. making field not null in the database, but allow this field to be nullable in pydantic (i.e. with `server_default` value).
+That will prevent the updates of the field to null (as with `server_default` set you cannot insert null values already as the default value would be used)
+
 
 
 ## default

docs/index.md

@@ -660,7 +660,8 @@ The following keyword arguments are supported on all field types.
 
 All fields are required unless one of the following is set:
 
-  * `nullable` - Creates a nullable column. Sets the default to `None`.
+  * `nullable` - Creates a nullable column. Sets the default to `False`. Read the fields common parameters for details.
+  * `sql_nullable` - Used to set different setting for pydantic and the database. Sets the default to `nullable` value. Read the fields common parameters for details.
   * `default` - Set a default value for the field. **Not available for relation fields**
   * `server_default` - Set a default value for the field on server side (like sqlalchemy's `func.now()`). **Not available for relation fields**
   * `primary key` with `autoincrement` - When a column is set to primary key and autoincrement is set on this column. 

docs/models/index.md

@@ -357,11 +357,16 @@ You can overwrite this parameter by providing `Meta` class `tablename` argument.
 
 On a model level you can also set model-wise constraints on sql columns.
 
-Right now only `UniqueColumns` constraint is present. 
+Right now only `IndexColumns` and `UniqueColumns` constraints are supported. 
+
+!!!note
+        Note that both constraints should be used only if you want to set a name on constraint or want to set the index on multiple columns, otherwise `index` and `unique` properties on ormar fields are preferred.
 
 !!!tip
     To read more about columns constraints like `primary_key`, `unique`, `ForeignKey` etc. visit [fields][fields].
 
+#### UniqueColumns
+
 You can set this parameter by providing `Meta` class `constraints` argument.
 
 ```Python hl_lines="14-17"
@@ -373,6 +378,20 @@ You can set this parameter by providing `Meta` class `constraints` argument.
         To set one column as unique use [`unique`](../fields/common-parameters.md#unique) common parameter. 
         Of course you can set many columns as unique with this param but each of them will be checked separately.
 
+#### IndexColumns
+
+You can set this parameter by providing `Meta` class `constraints` argument.
+
+```Python hl_lines="14-17"
+--8<-- "../docs_src/models/docs017.py"
+```
+
+!!!note
+        Note that constraints are meant for combination of columns that should be in the index. 
+        To set one column index use [`unique`](../fields/common-parameters.md#index) common parameter. 
+        Of course, you can set many columns as indexes with this param but each of them will be a separate index.
+
+
 ### Pydantic configuration
 
 As each `ormar.Model` is also a `pydantic` model, you might want to tweak the settings of the pydantic configuration.

docs/releases.md

@@ -1,3 +1,21 @@
+# 0.10.19
+
+## ✨ Features
+
+* Add support for multi-column non-unique `IndexColumns` in `Meta.constraints` [#307](https://github.com/collerek/ormar/issues/307)
+* Add `sql_nullable` field attribute that allows to set different nullable setting for pydantic model and for underlying sql column [#308](https://github.com/collerek/ormar/issues/308)
+
+## 🐛 Fixes
+
+* Enable caching of relation map to increase performance [#337](https://github.com/collerek/ormar/issues/337)
+* Clarify and fix documentation in regard of nullable fields [#339](https://github.com/collerek/ormar/issues/339)
+
+## 💬 Other
+
+* Bump supported `databases` version to `<=5.2`.
+
+
+
 # 0.10.18
 
 ## 🐛 Fixes

docs_src/models/docs017.py

@@ -0,0 +1,21 @@
+import databases
+import sqlalchemy
+
+import ormar
+
+database = databases.Database("sqlite:///db.sqlite")
+metadata = sqlalchemy.MetaData()
+
+
+class Course(ormar.Model):
+    class Meta:
+        database = database
+        metadata = metadata
+        # define your constraints in Meta class of the model
+        # it's a list that can contain multiple constraints
+        # hera a combination of name and column will have a compound index in the db
+        constraints = [ormar.IndexColumns("name", "completed")]
+
+    id: int = ormar.Integer(primary_key=True)
+    name: str = ormar.String(max_length=100)
+    completed: bool = ormar.Boolean(default=False)

ormar/__init__.py

@@ -51,6 +51,7 @@ from ormar.fields import (
     Float,
     ForeignKey,
     ForeignKeyField,
+    IndexColumns,
     Integer,
     JSON,
     LargeBinary,
@@ -77,7 +78,7 @@ class UndefinedType:  # pragma no cover
 
 Undefined = UndefinedType()
 
-__version__ = "0.10.18"
+__version__ = "0.10.19"
 __all__ = [
     "Integer",
     "BigInteger",
@@ -102,6 +103,7 @@ __all__ = [
     "Undefined",
     "UUID",
     "UniqueColumns",
+    "IndexColumns",
     "QuerySetProtocol",
     "RelationProtocol",
     "ModelMeta",

ormar/fields/__init__.py

@@ -5,7 +5,8 @@ as well as relation Fields (ForeignKey, ManyToMany).
 Also a definition for custom CHAR based sqlalchemy UUID field
 """
 from ormar.fields.base import BaseField
-from ormar.fields.foreign_key import ForeignKey, ForeignKeyField, UniqueColumns
+from ormar.fields.constraints import IndexColumns, UniqueColumns
+from ormar.fields.foreign_key import ForeignKey, ForeignKeyField
 from ormar.fields.many_to_many import ManyToMany, ManyToManyField
 from ormar.fields.model_fields import (
     BigInteger,
@@ -36,6 +37,7 @@ __all__ = [
     "DateTime",
     "String",
     "JSON",
+    "IndexColumns",
     "Integer",
     "Text",
     "Float",
@@ -45,7 +47,6 @@ __all__ = [
     "ManyToMany",
     "ManyToManyField",
     "BaseField",
-    "UniqueColumns",
     "ForeignKeyField",
     "ThroughField",
     "Through",
@@ -54,4 +55,5 @@ __all__ = [
     "DECODERS_MAP",
     "ENCODERS_MAP",
     "LargeBinary",
+    "UniqueColumns",
 ]

ormar/fields/base.py

@@ -43,6 +43,7 @@ class BaseField(FieldInfo):
         self.primary_key: bool = kwargs.pop("primary_key", False)
         self.autoincrement: bool = kwargs.pop("autoincrement", False)
         self.nullable: bool = kwargs.pop("nullable", False)
+        self.sql_nullable: bool = kwargs.pop("sql_nullable", False)
         self.index: bool = kwargs.pop("index", False)
         self.unique: bool = kwargs.pop("unique", False)
         self.pydantic_only: bool = kwargs.pop("pydantic_only", False)
@@ -265,7 +266,7 @@ class BaseField(FieldInfo):
                 self.column_type,
                 *self.construct_constraints(),
                 primary_key=self.primary_key,
-                nullable=self.nullable and not self.primary_key,
+                nullable=self.sql_nullable,
                 index=self.index,
                 unique=self.unique,
                 default=self.ormar_default,

ormar/fields/constraints.py

@@ -0,0 +1,22 @@
+from typing import Any
+
+from sqlalchemy import Index, UniqueConstraint
+
+
+class UniqueColumns(UniqueConstraint):
+    """
+    Subclass of sqlalchemy.UniqueConstraint.
+    Used to avoid importing anything from sqlalchemy by user.
+    """
+
+
+class IndexColumns(Index):
+    def __init__(self, *args: Any, name: str = None) -> None:
+        if not name:
+            name = "TEMPORARY_NAME"
+        super().__init__(name, *args)
+
+    """
+    Subclass of sqlalchemy.Index.
+    Used to avoid importing anything from sqlalchemy by user.
+    """

ormar/fields/foreign_key.py

@@ -18,7 +18,6 @@ from typing import (
 import sqlalchemy
 from pydantic import BaseModel, create_model
 from pydantic.typing import ForwardRef, evaluate_forwardref
-from sqlalchemy import UniqueConstraint
 
 import ormar  # noqa I101
 from ormar.exceptions import ModelDefinitionError, RelationshipInstanceError
@@ -160,13 +159,6 @@ def validate_not_allowed_fields(kwargs: Dict) -> None:
         )
 
 
-class UniqueColumns(UniqueConstraint):
-    """
-    Subclass of sqlalchemy.UniqueConstraint.
-    Used to avoid importing anything from sqlalchemy by user.
-    """
-
-
 @dataclass
 class ForeignKeyConstraint:
     """
@@ -242,6 +234,9 @@ def ForeignKey(  # noqa CFQ002
     skip_reverse = kwargs.pop("skip_reverse", False)
     skip_field = kwargs.pop("skip_field", False)
 
+    sql_nullable = kwargs.pop("sql_nullable", None)
+    sql_nullable = nullable if sql_nullable is None else sql_nullable
+
     validate_not_allowed_fields(kwargs)
 
     if to.__class__ == ForwardRef:
@@ -263,6 +258,7 @@ def ForeignKey(  # noqa CFQ002
         alias=name,
         name=kwargs.pop("real_name", None),
         nullable=nullable,
+        sql_nullable=sql_nullable,
         constraints=constraints,
         unique=unique,
         column_type=column_type,

ormar/fields/model_fields.py

@@ -75,6 +75,7 @@ class ModelFieldFactory:
         default = kwargs.pop("default", None)
         server_default = kwargs.pop("server_default", None)
         nullable = kwargs.pop("nullable", None)
+        sql_nullable = kwargs.pop("sql_nullable", None)
         pydantic_only = kwargs.pop("pydantic_only", False)
 
         primary_key = kwargs.pop("primary_key", False)
@@ -86,6 +87,13 @@ class ModelFieldFactory:
 
         overwrite_pydantic_type = kwargs.pop("overwrite_pydantic_type", None)
 
+        nullable = is_field_nullable(
+            nullable, default, server_default, pydantic_only
+        ) or is_auto_primary_key(primary_key, autoincrement)
+        sql_nullable = (
+            nullable if sql_nullable is None else (sql_nullable and not primary_key)
+        )
+
         namespace = dict(
             __type__=cls._type,
             __pydantic_type__=overwrite_pydantic_type
@@ -97,8 +105,8 @@ class ModelFieldFactory:
             primary_key=primary_key,
             default=default,
             server_default=server_default,
-            nullable=is_field_nullable(nullable, default, server_default, pydantic_only)
+            nullable=nullable,
-            or is_auto_primary_key(primary_key, autoincrement),
+            sql_nullable=sql_nullable,
             index=kwargs.pop("index", False),
             unique=kwargs.pop("unique", False),
             pydantic_only=pydantic_only,

ormar/models/helpers/models.py

@@ -75,6 +75,8 @@ def populate_default_options_values(
         if field.__type__ == bytes
     }
 
+    new_model.__relation_map__ = None
+
 
 class Connection(sqlite3.Connection):
     def __init__(self, *args: Any, **kwargs: Any) -> None:  # pragma: no cover

ormar/models/helpers/sqlalchemy.py

@@ -285,24 +285,40 @@ def populate_meta_sqlalchemy_table_if_required(meta: "ModelMeta") -> None:
 
     :param meta: Meta class of the Model without sqlalchemy table constructed
     :type meta: Model class Meta
-    :return: class with populated Meta.table
-    :rtype: Model class
     """
     if not hasattr(meta, "table") and check_for_null_type_columns_from_forward_refs(
         meta
     ):
-        for constraint in meta.constraints:
+        set_constraint_names(meta=meta)
-            if isinstance(constraint, sqlalchemy.UniqueConstraint):
-                constraint.name = (
-                    f"uc_{meta.tablename}_"
-                    f'{"_".join([str(col) for col in constraint._pending_colargs])}'
-                )
         table = sqlalchemy.Table(
             meta.tablename, meta.metadata, *meta.columns, *meta.constraints
         )
         meta.table = table
 
 
+def set_constraint_names(meta: "ModelMeta") -> None:
+    """
+    Populates the names on IndexColumn and UniqueColumns constraints.
+
+    :param meta: Meta class of the Model without sqlalchemy table constructed
+    :type meta: Model class Meta
+    """
+    for constraint in meta.constraints:
+        if isinstance(constraint, sqlalchemy.UniqueConstraint) and not constraint.name:
+            constraint.name = (
+                f"uc_{meta.tablename}_"
+                f'{"_".join([str(col) for col in constraint._pending_colargs])}'
+            )
+        elif (
+            isinstance(constraint, sqlalchemy.Index)
+            and constraint.name == "TEMPORARY_NAME"
+        ):
+            constraint.name = (
+                f"ix_{meta.tablename}_"
+                f'{"_".join([col for col in constraint._pending_colargs])}'
+            )
+
+
 def update_column_definition(
     model: Union[Type["Model"], Type["NewBaseModel"]], field: "ForeignKeyField"
 ) -> None:

ormar/models/metaclass.py

@@ -17,6 +17,7 @@ import sqlalchemy
 from sqlalchemy.sql.schema import ColumnCollectionConstraint
 
 import ormar  # noqa I100
+import ormar.fields.constraints
 from ormar import ModelDefinitionError  # noqa I100
 from ormar.exceptions import ModelError
 from ormar.fields import BaseField
@@ -219,7 +220,8 @@ def update_attrs_from_base_meta(  # noqa: CCR001
                     parent_value=parent_value,
                 )
                 parent_value = [
-                    ormar.UniqueColumns(*x._pending_colargs) for x in parent_value
+                    ormar.fields.constraints.UniqueColumns(*x._pending_colargs)
+                    for x in parent_value
                 ]
             if isinstance(current_value, list):
                 current_value.extend(parent_value)

ormar/models/mixins/relation_mixin.py

@@ -1,5 +1,6 @@
 from typing import (
     Callable,
+    Dict,
     List,
     Optional,
     Set,
@@ -20,6 +21,7 @@ class RelationMixin:
         from ormar import ModelMeta
 
         Meta: ModelMeta
+        __relation_map__: Optional[List[str]]
         _related_names: Optional[Set]
         _through_names: Optional[Set]
         _related_fields: Optional[List]
@@ -120,7 +122,11 @@ class RelationMixin:
 
     @classmethod
     def _iterate_related_models(  # noqa: CCR001
-        cls, node_list: NodeList = None, source_relation: str = None
+        cls,
+        node_list: NodeList = None,
+        parsed_map: Dict = None,
+        source_relation: str = None,
+        recurrent: bool = False,
     ) -> List[str]:
         """
         Iterates related models recursively to extract relation strings of
@@ -130,12 +136,15 @@ class RelationMixin:
         :rtype: List[str]
         """
         if not node_list:
+            if cls.__relation_map__:
+                return cls.__relation_map__
             node_list = NodeList()
+            parsed_map = dict()
             current_node = node_list.add(node_class=cls)
         else:
             current_node = node_list[-1]
-        relations = cls.extract_related_names()
+        relations = sorted(cls.extract_related_names())
-        processed_relations = []
+        processed_relations: List[str] = []
         for relation in relations:
             if not current_node.visited(relation):
                 target_model = cls.Meta.model_fields[relation].to
@@ -144,12 +153,23 @@ class RelationMixin:
                     relation_name=relation,
                     parent_node=current_node,
                 )
+                relation_key = f"{cls.get_name()}_{relation}"
+                parsed_map = cast(Dict, parsed_map)
+                deep_relations = parsed_map.get(relation_key)
+                if not deep_relations:
                     deep_relations = target_model._iterate_related_models(
-                    source_relation=relation, node_list=node_list
+                        source_relation=relation,
+                        node_list=node_list,
+                        recurrent=True,
+                        parsed_map=parsed_map,
                     )
+                    parsed_map[relation_key] = deep_relations
                 processed_relations.extend(deep_relations)
 
-        return cls._get_final_relations(processed_relations, source_relation)
+        result = cls._get_final_relations(processed_relations, source_relation)
+        if not recurrent:
+            cls.__relation_map__ = result
+        return result
 
     @staticmethod
     def _get_final_relations(

ormar/models/newbasemodel.py

@@ -76,6 +76,7 @@ class NewBaseModel(pydantic.BaseModel, ModelTableProxy, metaclass=ModelMetaclass
         __tablename__: str
         __metadata__: sqlalchemy.MetaData
         __database__: databases.Database
+        __relation_map__: Optional[List[str]]
         _orm_relationship_manager: AliasManager
         _orm: RelationsManager
         _orm_id: int

requirements.txt

@@ -1,6 +1,5 @@
-databases[sqlite]>=0.3.2,<0.5.3
+databases>=0.3.2,<0.5.3
-databases[postgresql]>=0.3.2,<0.5.3
+
-databases[mysql]>=0.3.2,<0.5.3
 pydantic >=1.6.1,!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<=1.8.2
 sqlalchemy>=1.3.18,<=1.4.23
 typing_extensions>=3.7,<3.10.0.3

tests/test_exclude_include_dict/test_complex_relation_tree_performance.py

@@ -0,0 +1,420 @@
+from datetime import datetime
+from typing import List, Optional, Union
+
+import databases
+import pydantic
+import pytest
+import sqlalchemy
+
+import ormar as orm
+
+from tests.settings import DATABASE_URL
+
+database = databases.Database(DATABASE_URL, force_rollback=True)
+metadata = sqlalchemy.MetaData()
+
+
+class MainMeta(orm.ModelMeta):
+    database = database
+    metadata = metadata
+
+
+class ChagenlogRelease(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "changelog_release"
+
+
+class CommitIssue(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "commit_issues"
+
+
+class CommitLabel(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "commit_label"
+
+
+class MergeRequestCommit(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "merge_request_commits"
+
+
+class MergeRequestIssue(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "merge_request_issues"
+
+
+class MergeRequestLabel(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "merge_request_labels"
+
+
+class ProjectLabel(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "project_label"
+
+
+class PushCommit(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "push_commit"
+
+
+class PushLabel(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "push_label"
+
+
+class TagCommit(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "tag_commits"
+
+
+class TagIssue(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "tag_issue"
+
+
+class TagLabel(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+
+    class Meta(MainMeta):
+        tablename = "tag_label"
+
+
+class UserProject(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    access_level: int = orm.Integer(default=0)
+
+    class Meta(MainMeta):
+        tablename = "user_project"
+
+
+class Label(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    title: str = orm.String(max_length=100)
+    description: str = orm.Text(default="")
+    type: str = orm.String(max_length=100, default="")
+
+    class Meta(MainMeta):
+        tablename = "labels"
+
+
+class Project(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    name: str = orm.String(max_length=100)
+    description: str = orm.Text(default="")
+    git_url: str = orm.String(max_length=500, default="")
+    labels: Optional[Union[List[Label], Label]] = orm.ManyToMany(
+        Label, through=ProjectLabel, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    changelog_jira_tag: str = orm.String(max_length=100, default="")
+    change_type_jira_tag: str = orm.String(max_length=100, default="")
+    jira_prefix: str = orm.String(max_length=10, default="SAN")
+    type: str = orm.String(max_length=10, default="cs")
+    target_branch_name: str = orm.String(max_length=100, default="master")
+    header: str = orm.String(max_length=250, default="")
+    jira_url: str = orm.String(max_length=500,)
+    changelog_file: str = orm.String(max_length=250, default="")
+    version_file: str = orm.String(max_length=250, default="")
+
+    class Meta(MainMeta):
+        tablename = "projects"
+
+
+class Issue(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    summary: str = orm.Text(default="")
+    description: str = orm.Text(default="")
+    changelog: str = orm.Text(default="")
+    link: str = orm.String(max_length=500)
+    issue_type: str = orm.String(max_length=100)
+    key: str = orm.String(max_length=100)
+    change_type: str = orm.String(max_length=100, default="")
+    data: pydantic.Json = orm.JSON(default={})
+
+    class Meta(MainMeta):
+        tablename = "issues"
+
+
+class User(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    username: str = orm.String(max_length=100, unique=True)
+    name: str = orm.String(max_length=200, default="")
+
+    class Meta(MainMeta):
+        tablename = "users"
+
+
+class Branch(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    name: str = orm.String(max_length=200)
+    description: str = orm.Text(default="")
+    automatic_tags: bool = orm.Boolean(default=False)
+    is_it_locked: bool = orm.Boolean(default=True)
+    prefix_tag: str = orm.String(max_length=50, default="")
+    postfix_tag: str = orm.String(max_length=50, default="")
+    project: Project = orm.ForeignKey(Project, ondelete="CASCADE", onupdate="CASCADE")
+
+    class Meta(MainMeta):
+        tablename = "branches"
+
+
+class Changelog(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    content: str = orm.Text(default="")
+    version: str = orm.Text(default="")
+    past_changelog: int = orm.Integer(default=0)
+    label: Label = orm.ForeignKey(
+        Label, nullable=True, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    project: Project = orm.ForeignKey(Project, ondelete="CASCADE", onupdate="CASCADE")
+    created_date: datetime = orm.DateTime(default=datetime.utcnow())
+
+    class Meta(MainMeta):
+        tablename = "changelogs"
+
+
+class Commit(orm.Model):
+    id: str = orm.String(max_length=500, primary_key=True)
+    short_id: str = orm.String(max_length=500)
+    title: str = orm.String(max_length=500)
+    message: str = orm.Text(default="")
+    url = orm.String(max_length=500, default="")
+    author_name = orm.String(max_length=500, default="")
+    labels: Optional[Union[List[Label], Label]] = orm.ManyToMany(
+        Label, through=CommitLabel, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    issues: Optional[Union[List[Issue], Issue]] = orm.ManyToMany(
+        Issue, through=CommitIssue, ondelete="CASCADE", onupdate="CASCADE"
+    )
+
+    class Meta(MainMeta):
+        tablename = "commits"
+
+
+class MergeRequest(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    idd: int = orm.Integer(default=0)
+    title: str = orm.String(max_length=500)
+    state: str = orm.String(max_length=100)
+    merge_status: str = orm.String(max_length=100)
+    description: str = orm.Text(default="")
+    source: Branch = orm.ForeignKey(Branch, related_name="source")
+    target: Branch = orm.ForeignKey(Branch, related_name="target")
+    labels: Optional[Union[List[Label], Label]] = orm.ManyToMany(
+        Label, through=MergeRequestLabel, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    commits: Optional[Union[List[Commit], Commit]] = orm.ManyToMany(
+        Commit, through=MergeRequestCommit, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    issues: Optional[Union[List[Issue], Issue]] = orm.ManyToMany(
+        Issue, through=MergeRequestIssue, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    project: Project = orm.ForeignKey(Project, ondelete="CASCADE", onupdate="CASCADE")
+
+    class Meta(MainMeta):
+        tablename = "merge_requests"
+
+
+class Push(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    branch: Branch = orm.ForeignKey(
+        Branch, nullable=True, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    has_locking_changes: bool = orm.Boolean(default=False)
+    sha: str = orm.String(max_length=200)
+    labels: Optional[Union[List[Label], Label]] = orm.ManyToMany(
+        Label, through=PushLabel, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    commits: Optional[Union[List[Commit], Commit]] = orm.ManyToMany(
+        Commit,
+        through=PushCommit,
+        through_relation_name="push",
+        through_reverse_relation_name="commit_id",
+        ondelete="CASCADE",
+        onupdate="CASCADE",
+    )
+    author: User = orm.ForeignKey(User, ondelete="CASCADE", onupdate="CASCADE")
+    project: Project = orm.ForeignKey(Project, ondelete="CASCADE", onupdate="CASCADE")
+
+    class Meta(MainMeta):
+        tablename = "pushes"
+
+
+class Tag(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    name: str = orm.String(max_length=200)
+    ref: str = orm.String(max_length=200)
+    project: Project = orm.ForeignKey(Project, ondelete="CASCADE", onupdate="CASCADE")
+    title: str = orm.String(max_length=200, default="")
+    description: str = orm.Text(default="")
+    commits: Optional[Union[List[Commit], Commit]] = orm.ManyToMany(
+        Commit,
+        through=TagCommit,
+        through_relation_name="tag",
+        through_reverse_relation_name="commit_id",
+        ondelete="CASCADE",
+        onupdate="CASCADE",
+    )
+    issues: Optional[Union[List[Issue], Issue]] = orm.ManyToMany(
+        Issue, through=TagIssue, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    labels: Optional[Union[List[Label], Label]] = orm.ManyToMany(
+        Label, through=TagLabel, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    user: User = orm.ForeignKey(
+        User, nullable=True, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    branch: Branch = orm.ForeignKey(
+        Branch, nullable=True, ondelete="CASCADE", onupdate="CASCADE"
+    )
+
+    class Meta(MainMeta):
+        tablename = "tags"
+
+
+class Release(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    title: str = orm.String(max_length=200, default="")
+    description: str = orm.Text(default="")
+    tag: Tag = orm.ForeignKey(Tag, ondelete="CASCADE", onupdate="CASCADE")
+    changelogs: List[Changelog] = orm.ManyToMany(
+        Changelog, through=ChagenlogRelease, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    data: pydantic.Json = orm.JSON(default={})
+
+    class Meta(MainMeta):
+        tablename = "releases"
+
+
+class Webhook(orm.Model):
+    id: int = orm.Integer(name="id", primary_key=True)
+    object_kind = orm.String(max_length=100)
+    project: Project = orm.ForeignKey(Project, ondelete="CASCADE", onupdate="CASCADE")
+    merge_request: MergeRequest = orm.ForeignKey(
+        MergeRequest, nullable=True, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    tag: Tag = orm.ForeignKey(
+        Tag, nullable=True, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    push: Push = orm.ForeignKey(
+        Push, nullable=True, ondelete="CASCADE", onupdate="CASCADE"
+    )
+    created_at: datetime = orm.DateTime(default=datetime.now())
+    data: pydantic.Json = orm.JSON(default={})
+    status: int = orm.Integer(default=200)
+    error: str = orm.Text(default="")
+
+
+@pytest.fixture(autouse=True, scope="module")
+def create_test_database():
+    engine = sqlalchemy.create_engine(DATABASE_URL)
+    metadata.drop_all(engine)
+    metadata.create_all(engine)
+    yield
+    metadata.drop_all(engine)
+
+
+@pytest.mark.asyncio
+async def test_very_complex_relation_map():
+    async with database:
+        tags = [
+            {"id": 18, "name": "name-18", "ref": "ref-18"},
+            {"id": 17, "name": "name-17", "ref": "ref-17"},
+            {"id": 12, "name": "name-12", "ref": "ref-12"},
+        ]
+        payload = [
+            {
+                "id": 9,
+                "title": "prueba-2321",
+                "description": "\n<!--- start changelog ver.v.1.3.0.0 -->\n### [v.1.3.0.0] - 2021-08-19\n#### Resolved Issues\n\n#### Task\n\n- Probar flujo de changelog Automatic  Jira: [SAN-86](https://htech.atlassian.net/browse/SAN-86)\n\n    Description: Se probara el flujo de changelog automatic.  \n\n    Changelog: Se agrega función para extraer texto del campo changelog del dashboard de Sanval y ponerlo directamente en el changelog.md del repositorio. \n\n\n            \n<!--- end changelog ver.v.1.3.0.0 -->\n",
+                "data": {},
+            },
+            {
+                "id": 8,
+                "title": "prueba-123-prod",
+                "description": "\n<!--- start changelog ver.v.1.2.0.0 -->\n### [v.1.2.0.0] - 2021-08-19\n#### Resolved Issues\n\n#### Task\n\n- Probar flujo de changelog Automatic  Jira: [SAN-86](https://htech.atlassian.net/browse/SAN-86)\n\n    Description: Se probara el flujo de changelog automatic.  \n\n    Changelog: Se agrega función para extraer texto del campo changelog del dashboard de Sanval y ponerlo directamente en el changelog.md del repositorio. \n\n\n            \n<!--- end changelog ver.v.1.2.0.0 -->\n",
+                "data": {},
+            },
+            {
+                "id": 6,
+                "title": "prueba-3-2",
+                "description": "\n<!--- start changelog ver.v.1.1.0.0 -->\n### [v.1.1.0.0] - 2021-07-29\n#### Resolved Issues\n\n#### Task\n\n- Probar flujo de changelog Automatic  Jira: [SAN-86](https://htech.atlassian.net/browse/SAN-86)\n\n    Description: Se probara el flujo de changelog automatic.  \n\n    Changelog: Se agrega función para extraer texto del campo changelog del dashboard de Sanval y ponerlo directamente en el changelog.md del repositorio. \n\n\n            \n<!--- end changelog ver.v.1.1.0.0 -->\n",
+                "data": {},
+            },
+        ]
+        saved_tags = []
+        for tag in tags:
+            saved_tags.append(await Tag(**tag).save())
+
+        for ind, pay in enumerate(payload):
+            await Release(**pay, tag=saved_tags[ind]).save()
+
+        releases = await Release.objects.order_by(Release.id.desc()).all()
+        dicts = [release.dict() for release in releases]
+
+        result = [
+            {
+                "id": 9,
+                "title": "prueba-2321",
+                "description": "\n<!--- start changelog ver.v.1.3.0.0 -->\n### [v.1.3.0.0] - 2021-08-19\n#### Resolved Issues\n\n#### Task\n\n- Probar flujo de changelog Automatic  Jira: [SAN-86](https://htech.atlassian.net/browse/SAN-86)\n\n    Description: Se probara el flujo de changelog automatic.  \n\n    Changelog: Se agrega función para extraer texto del campo changelog del dashboard de Sanval y ponerlo directamente en el changelog.md del repositorio. \n\n\n            \n<!--- end changelog ver.v.1.3.0.0 -->\n",
+                "data": {},
+                "tag": {
+                    "id": 18,
+                    "taglabel": None,
+                    "tagcommit": None,
+                    "tagissue": None,
+                },
+                "changelogs": [],
+            },
+            {
+                "id": 8,
+                "title": "prueba-123-prod",
+                "description": "\n<!--- start changelog ver.v.1.2.0.0 -->\n### [v.1.2.0.0] - 2021-08-19\n#### Resolved Issues\n\n#### Task\n\n- Probar flujo de changelog Automatic  Jira: [SAN-86](https://htech.atlassian.net/browse/SAN-86)\n\n    Description: Se probara el flujo de changelog automatic.  \n\n    Changelog: Se agrega función para extraer texto del campo changelog del dashboard de Sanval y ponerlo directamente en el changelog.md del repositorio. \n\n\n            \n<!--- end changelog ver.v.1.2.0.0 -->\n",
+                "data": {},
+                "tag": {
+                    "id": 17,
+                    "taglabel": None,
+                    "tagcommit": None,
+                    "tagissue": None,
+                },
+                "changelogs": [],
+            },
+            {
+                "id": 6,
+                "title": "prueba-3-2",
+                "description": "\n<!--- start changelog ver.v.1.1.0.0 -->\n### [v.1.1.0.0] - 2021-07-29\n#### Resolved Issues\n\n#### Task\n\n- Probar flujo de changelog Automatic  Jira: [SAN-86](https://htech.atlassian.net/browse/SAN-86)\n\n    Description: Se probara el flujo de changelog automatic.  \n\n    Changelog: Se agrega función para extraer texto del campo changelog del dashboard de Sanval y ponerlo directamente en el changelog.md del repositorio. \n\n\n            \n<!--- end changelog ver.v.1.1.0.0 -->\n",
+                "data": {},
+                "tag": {
+                    "id": 12,
+                    "taglabel": None,
+                    "tagcommit": None,
+                    "tagissue": None,
+                },
+                "changelogs": [],
+            },
+        ]
+
+        assert dicts == result

tests/test_inheritance_and_pydantic_generation/test_inheritance_concrete.py

@@ -8,6 +8,7 @@ import sqlalchemy as sa
 from sqlalchemy import create_engine
 
 import ormar
+import ormar.fields.constraints
 from ormar import ModelDefinitionError, property_field
 from ormar.exceptions import ModelError
 from tests.settings import DATABASE_URL
@@ -45,7 +46,9 @@ class DateFieldsModel(ormar.Model):
         abstract = True
         metadata = metadata
         database = db
-        constraints = [ormar.UniqueColumns("creation_date", "modification_date")]
+        constraints = [
+            ormar.fields.constraints.UniqueColumns("creation_date", "modification_date")
+        ]
 
     created_date: datetime.datetime = ormar.DateTime(
         default=datetime.datetime.now, name="creation_date"
@@ -58,7 +61,7 @@ class DateFieldsModel(ormar.Model):
 class Category(DateFieldsModel, AuditModel):
     class Meta(ormar.ModelMeta):
         tablename = "categories"
-        constraints = [ormar.UniqueColumns("name", "code")]
+        constraints = [ormar.fields.constraints.UniqueColumns("name", "code")]
 
     id: int = ormar.Integer(primary_key=True)
     name: str = ormar.String(max_length=50, unique=True, index=True)

tests/test_meta_constraints/test_index_constraints.py

@@ -0,0 +1,68 @@
+import asyncpg  # type: ignore
+import databases
+import pytest
+import sqlalchemy
+
+import ormar.fields.constraints
+from tests.settings import DATABASE_URL
+
+database = databases.Database(DATABASE_URL, force_rollback=True)
+metadata = sqlalchemy.MetaData()
+
+
+class Product(ormar.Model):
+    class Meta:
+        tablename = "products"
+        metadata = metadata
+        database = database
+        constraints = [
+            ormar.fields.constraints.IndexColumns("company", "name", name="my_index"),
+            ormar.fields.constraints.IndexColumns("location", "company_type"),
+        ]
+
+    id: int = ormar.Integer(primary_key=True)
+    name: str = ormar.String(max_length=100)
+    company: str = ormar.String(max_length=200)
+    location: str = ormar.String(max_length=200)
+    company_type: str = ormar.String(max_length=200)
+
+
+@pytest.fixture(autouse=True, scope="module")
+def create_test_database():
+    engine = sqlalchemy.create_engine(DATABASE_URL)
+    metadata.drop_all(engine)
+    metadata.create_all(engine)
+    yield
+    metadata.drop_all(engine)
+
+
+def test_table_structure():
+    assert len(Product.Meta.table.indexes) > 0
+    indexes = sorted(
+        list(Product.Meta.table.indexes), key=lambda x: x.name, reverse=True
+    )
+    test_index = indexes[0]
+    assert test_index.name == "my_index"
+    assert [col.name for col in test_index.columns] == ["company", "name"]
+
+    test_index = indexes[1]
+    assert test_index.name == "ix_products_location_company_type"
+    assert [col.name for col in test_index.columns] == ["location", "company_type"]
+
+
+@pytest.mark.asyncio
+async def test_index_is_not_unique():
+    async with database:
+        async with database.transaction(force_rollback=True):
+            await Product.objects.create(
+                name="Cookies", company="Nestle", location="A", company_type="B"
+            )
+            await Product.objects.create(
+                name="Mars", company="Mars", location="B", company_type="Z"
+            )
+            await Product.objects.create(
+                name="Mars", company="Nestle", location="C", company_type="X"
+            )
+            await Product.objects.create(
+                name="Mars", company="Mars", location="D", company_type="Y"
+            )

tests/test_meta_constraints/test_unique_constraints.py

@@ -1,4 +1,3 @@
-import asyncio
 import sqlite3
 
 import asyncpg  # type: ignore
@@ -7,7 +6,7 @@ import pymysql
 import pytest
 import sqlalchemy
 
-import ormar
+import ormar.fields.constraints
 from tests.settings import DATABASE_URL
 
 database = databases.Database(DATABASE_URL, force_rollback=True)
@@ -19,22 +18,15 @@ class Product(ormar.Model):
         tablename = "products"
         metadata = metadata
         database = database
-        constraints = [ormar.UniqueColumns("name", "company")]
+        constraints = [ormar.fields.constraints.UniqueColumns("name", "company")]
 
     id: int = ormar.Integer(primary_key=True)
     name: str = ormar.String(max_length=100)
     company: str = ormar.String(max_length=200)
 
 
-@pytest.fixture(scope="module")
-def event_loop():
-    loop = asyncio.get_event_loop()
-    yield loop
-    loop.close()
-
-
 @pytest.fixture(autouse=True, scope="module")
-async def create_test_database():
+def create_test_database():
     engine = sqlalchemy.create_engine(DATABASE_URL)
     metadata.drop_all(engine)
     metadata.create_all(engine)

tests/test_model_definition/test_overwriting_sql_nullable.py

@@ -0,0 +1,62 @@
+import sqlite3
+from typing import Optional
+
+import asyncpg
+import databases
+import pymysql
+import sqlalchemy
+from sqlalchemy import create_engine, text
+
+import ormar
+import pytest
+
+from tests.settings import DATABASE_URL
+
+db = databases.Database(DATABASE_URL, force_rollback=True)
+metadata = sqlalchemy.MetaData()
+
+
+class BaseMeta(ormar.ModelMeta):
+    metadata = metadata
+    database = db
+
+
+class PrimaryModel(ormar.Model):
+    class Meta(BaseMeta):
+        tablename = "primary_models"
+
+    id: int = ormar.Integer(primary_key=True)
+    name: str = ormar.String(max_length=255, index=True)
+    some_text: Optional[str] = ormar.Text(nullable=True, sql_nullable=False)
+    some_other_text: Optional[str] = ormar.String(
+        max_length=255, nullable=True, sql_nullable=False, server_default=text("''")
+    )
+
+
+@pytest.fixture(autouse=True, scope="module")
+def create_test_database():
+    engine = create_engine(DATABASE_URL)
+    metadata.create_all(engine)
+    yield
+    metadata.drop_all(engine)
+
+
+@pytest.mark.asyncio
+async def test_create_models():
+    async with db:
+        primary = await PrimaryModel(
+            name="Foo", some_text="Bar", some_other_text="Baz"
+        ).save()
+        assert primary.id == 1
+
+        primary2 = await PrimaryModel(name="Foo2", some_text="Bar2").save()
+        assert primary2.id == 2
+
+        with pytest.raises(
+            (
+                sqlite3.IntegrityError,
+                pymysql.IntegrityError,
+                asyncpg.exceptions.NotNullViolationError,
+            )
+        ):
+            await PrimaryModel(name="Foo3").save()