diff --git a/README.md b/README.md index 8183bbb..89d3e30 100644 --- a/README.md +++ b/README.md @@ -334,7 +334,7 @@ async def delete(): async def joins(): # Tho join two models use select_related - + # Django style book = await Book.objects.select_related("author").get(title="The Hobbit") # Python style @@ -348,7 +348,7 @@ async def joins(): # By default you also get a second side of the relation # constructed as lowercase source model name +'s' (books in this case) # you can also provide custom name with parameter related_name - + # Django style author = await Author.objects.select_related("books").all(name="J.R.R. Tolkien") # Python style @@ -601,7 +601,7 @@ metadata.drop_all(engine) * `prefetch_related(related: Union[List, str]) -> QuerySet` * `limit(limit_count: int) -> QuerySet` * `offset(offset: int) -> QuerySet` -* `count() -> int` +* `count(distinct: bool = True) -> int` * `exists() -> bool` * `max(columns: List[str]) -> Any` * `min(columns: List[str]) -> Any` diff --git a/docs/api/query-set/query-set.md b/docs/api/query-set/query-set.md index b7aa53e..36a6721 100644 --- a/docs/api/query-set/query-set.md +++ b/docs/api/query-set/query-set.md @@ -499,11 +499,19 @@ Returns a bool value to confirm if there are rows matching the given criteria #### count ```python - | async count() -> int + | async count(distinct: bool = True) -> int ``` Returns number of rows matching the given criteria (applied with `filter` and `exclude` if set before). +If `distinct` is `True` (the default), this will return the number of primary rows selected. If `False`, +the count will be the total number of rows returned +(including extra rows for `one-to-many` or `many-to-many` left `select_related` table joins). +`False` is the legacy (buggy) behavior for workflows that depend on it. + +**Arguments**: + +- `distinct` (`bool`): flag if the primary table rows should be distinct or not **Returns**: @@ -865,4 +873,3 @@ Bulk operations do not send signals. - `objects` (`List[Model]`): list of ormar models - `columns` (`List[str]`): list of columns to update - diff --git a/docs/api/relations/queryset-proxy.md b/docs/api/relations/queryset-proxy.md index 009c466..f36f8f4 100644 --- a/docs/api/relations/queryset-proxy.md +++ b/docs/api/relations/queryset-proxy.md @@ -150,14 +150,22 @@ Actual call delegated to QuerySet. #### count ```python - | async count() -> int + | async count(distinct: bool = True) -> int ``` Returns number of rows matching the given criteria (applied with `filter` and `exclude` if set before). +If `distinct` is `True` (the default), this will return the number of primary rows selected. If `False`, +the count will be the total number of rows returned +(including extra rows for `one-to-many` or `many-to-many` left `select_related` table joins). +`False` is the legacy (buggy) behavior for workflows that depend on it. Actual call delegated to QuerySet. +**Arguments**: + +- `distinct` (`bool`): flag if the primary table rows should be distinct or not + **Returns**: `int`: number of rows @@ -260,11 +268,11 @@ List of related models is cleared before the call. **Arguments**: -- `kwargs`: +- `kwargs`: **Returns**: -`_asyncio.Future`: +`_asyncio.Future`: #### get\_or\_none @@ -773,4 +781,3 @@ Actual call delegated to QuerySet. **Returns**: `QuerysetProxy`: QuerysetProxy - diff --git a/docs/index.md b/docs/index.md index 9ecee7d..740c158 100644 --- a/docs/index.md +++ b/docs/index.md @@ -22,7 +22,7 @@ ### Overview The `ormar` package is an async mini ORM for Python, with support for **Postgres, -MySQL**, and **SQLite**. +MySQL**, and **SQLite**. The main benefits of using `ormar` are: @@ -31,7 +31,7 @@ The main benefits of using `ormar` are: The goal was to create a simple ORM that can be **used directly (as request and response models) with [`fastapi`][fastapi]** that bases it's data validation on pydantic. -Ormar - apart from the obvious "ORM" in name - gets its name from _ormar_ in Swedish which means _snakes_, and _ormar(e)_ in Croatian which means _cabinet_. +Ormar - apart from the obvious "ORM" in name - gets its name from _ormar_ in Swedish which means _snakes_, and _ormar(e)_ in Croatian which means _cabinet_. And what's a better name for python ORM than snakes cabinet :) @@ -82,7 +82,7 @@ Ormar is built with: `ormar` is built as open-source software and will remain completely free (MIT license). -As I write open-source code to solve everyday problems in my work or to promote and build strong python +As I write open-source code to solve everyday problems in my work or to promote and build strong python community you can say thank you and buy me a coffee or sponsor me with a monthly amount to help ensure my work remains free and maintained. @@ -99,7 +99,7 @@ community you can say thank you and buy me a coffee or sponsor me with a monthly padding: 10px; line-height: 0px; height: 40px; -"> +"> @@ -141,30 +141,30 @@ engine = sqlalchemy.create_engine(DATABASE_URL) metadata.create_all(engine) ``` -For a sample configuration of alembic and more information regarding migrations and +For a sample configuration of alembic and more information regarding migrations and database creation visit [migrations][migrations] documentation section. ### Package versions **ormar is still under development:** We recommend pinning any dependencies (with i.e. `ormar~=0.9.1`) -`ormar` also follows the release numeration that breaking changes bump the major number, +`ormar` also follows the release numeration that breaking changes bump the major number, while other changes and fixes bump minor number, so with the latter you should be safe to update, yet always read the [releases][releases] docs before. `example: (0.5.2 -> 0.6.0 - breaking, 0.5.2 -> 0.5.3 - non breaking)`. ### Asynchronous Python -Note that `ormar` is an asynchronous ORM, which means that you have to `await` the calls to +Note that `ormar` is an asynchronous ORM, which means that you have to `await` the calls to the methods, that are scheduled for execution in an event loop. Python has a builtin module [`asyncio`][asyncio] that allows you to do just that. -Note that most "normal" python interpreters do not allow execution of `await` -outside of a function (because you actually schedule this function for delayed execution +Note that most "normal" python interpreters do not allow execution of `await` +outside of a function (because you actually schedule this function for delayed execution and don't get the result immediately). In a modern web framework (like `fastapi`), the framework will handle this for you, but if -you plan to do this on your own you need to perform this manually like described in the +you plan to do this on your own you need to perform this manually like described in the quick start below. ### Quick Start @@ -343,7 +343,7 @@ async def delete(): async def joins(): # Tho join two models use select_related - + # Django style book = await Book.objects.select_related("author").get(title="The Hobbit") # Python style @@ -357,7 +357,7 @@ async def joins(): # By default you also get a second side of the relation # constructed as lowercase source model name +'s' (books in this case) # you can also provide custom name with parameter related_name - + # Django style author = await Author.objects.select_related("books").all(name="J.R.R. Tolkien") # Python style @@ -610,7 +610,7 @@ metadata.drop_all(engine) * `prefetch_related(related: Union[List, str]) -> QuerySet` * `limit(limit_count: int) -> QuerySet` * `offset(offset: int) -> QuerySet` -* `count() -> int` +* `count(distinct: bool = True) -> int` * `exists() -> bool` * `max(columns: List[str]) -> Any` * `min(columns: List[str]) -> Any` @@ -670,10 +670,10 @@ All fields are required unless one of the following is set: * `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. -Autoincrement is set by default on int primary keys. + * `primary key` with `autoincrement` - When a column is set to primary key and autoincrement is set on this column. +Autoincrement is set by default on int primary keys. * `pydantic_only` - Field is available only as normal pydantic field, not stored in the database. - + ### Available signals Signals allow to trigger your function for a given event on a given Model. diff --git a/docs/queries/aggregations.md b/docs/queries/aggregations.md index 53426de..61c3714 100644 --- a/docs/queries/aggregations.md +++ b/docs/queries/aggregations.md @@ -3,7 +3,7 @@ Currently 6 aggregation functions are supported. -* `count() -> int` +* `count(distinct: bool = True) -> int` * `exists() -> bool` * `sum(columns) -> Any` * `avg(columns) -> Any` @@ -12,19 +12,23 @@ Currently 6 aggregation functions are supported. * `QuerysetProxy` - * `QuerysetProxy.count()` method + * `QuerysetProxy.count(distinct=True)` method * `QuerysetProxy.exists()` method * `QuerysetProxy.sum(columns)` method * `QuerysetProxy.avg(columns)` method * `QuerysetProxy.min(column)` method * `QuerysetProxy.max(columns)` method - + ## count -`count() -> int` +`count(distinct: bool = True) -> int` -Returns number of rows matching the given criteria (i.e. applied with `filter` and `exclude`) +Returns number of rows matching the given criteria (i.e. applied with `filter` and `exclude`). +If `distinct` is `True` (the default), this will return the number of primary rows selected. If `False`, +the count will be the total number of rows returned +(including extra rows for `one-to-many` or `many-to-many` left `select_related` table joins). +`False` is the legacy (buggy) behavior for workflows that depend on it. ```python class Book(ormar.Model): @@ -84,7 +88,7 @@ Returns sum value of columns for rows matching the given criteria (applied with You can pass one or many column names including related columns. -As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, +As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, you can have `sum(col1, col2)` and later add 2 returned sums in python) You cannot `sum` non numeric columns. @@ -138,7 +142,7 @@ Returns avg value of columns for rows matching the given criteria (applied with You can pass one or many column names including related columns. -As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, +As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, you can have `sum(col1, col2)` and later add 2 returned sums in python) You cannot `avg` non numeric columns. @@ -193,7 +197,7 @@ Returns min value of columns for rows matching the given criteria (applied with You can pass one or many column names including related columns. -As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, +As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, you can have `sum(col1, col2)` and later add 2 returned sums in python) If you aggregate on one column, the single value is directly returned as a result @@ -241,7 +245,7 @@ Returns min value of columns for rows matching the given criteria (applied with You can pass one or many column names including related columns. -As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, +As of now each column passed is aggregated separately (so `sum(col1+col2)` is not possible, you can have `sum(col1, col2)` and later add 2 returned sums in python) If you aggregate on one column, the single value is directly returned as a result @@ -292,7 +296,7 @@ select related etc related models directly from parent model. Works exactly the same as [count](./#count) function above but allows you to select columns from related objects from other side of the relation. -!!!tip +!!!tip To read more about `QuerysetProxy` visit [querysetproxy][querysetproxy] section ### exists @@ -320,6 +324,5 @@ objects from other side of the relation. Works exactly the same as [max](./#max) function above but allows you to select maximum of columns from related objects from other side of the relation. -!!!tip +!!!tip To read more about `QuerysetProxy` visit [querysetproxy][querysetproxy] section - diff --git a/docs/queries/index.md b/docs/queries/index.md index e2c5a0a..9389b97 100644 --- a/docs/queries/index.md +++ b/docs/queries/index.md @@ -8,7 +8,7 @@ and it's options. Most of the methods are also available through many to many relations and on reverse foreign key relations through `QuerysetProxy` interface. -!!!info +!!!info To see which relations are supported and how to construct relations visit [relations][relations]. @@ -34,12 +34,12 @@ To read more about any specific section or function please refer to the details * `Model.upsert()` method * `Model.save_related()` method - + * `QuerysetProxy` * `QuerysetProxy.create(**kwargs)` method * `QuerysetProxy.get_or_create(**kwargs)` method * `QuerysetProxy.update_or_create(**kwargs)` method - + !!!tip To read more about any or all of those functions visit [create](./create.md) section. @@ -62,7 +62,7 @@ To read more about any specific section or function please refer to the details * `QuerysetProxy.get_or_create(**kwargs)` method * `QuerysetProxy.first()` method * `QuerysetProxy.all(**kwargs)` method - + !!!tip To read more about any or all of those functions visit [read](./read.md) section. @@ -96,7 +96,7 @@ Instead of ormar models return raw data in form list of dictionaries or tuples. * `QuerysetProxy` * `QuerysetProxy.update_or_create(**kwargs)` method - + !!!tip To read more about any or all of those functions visit [update](./update.md) section. @@ -112,7 +112,7 @@ Instead of ormar models return raw data in form list of dictionaries or tuples. * `QuerysetProxy` * `QuerysetProxy.remove()` method * `QuerysetProxy.clear()` method - + !!!tip To read more about any or all of those functions visit [delete](./delete.md) section. @@ -129,7 +129,7 @@ Instead of ormar models return raw data in form list of dictionaries or tuples. * `QuerysetProxy` * `QuerysetProxy.select_related(related: Union[List, str])` method * `QuerysetProxy.prefetch_related(related: Union[List, str])` method - + !!!tip To read more about any or all of those functions visit [joins and subqueries](./joins-and-subqueries.md) section. @@ -152,7 +152,7 @@ Instead of ormar models return raw data in form list of dictionaries or tuples. * `QuerysetProxy.get_or_none(**kwargs)` method * `QuerysetProxy.get_or_create(**kwargs)` method * `QuerysetProxy.all(**kwargs)` method - + !!!tip To read more about any or all of those functions visit [filtering and sorting](./filter-and-sort.md) section. @@ -165,7 +165,7 @@ Instead of ormar models return raw data in form list of dictionaries or tuples. * `QuerysetProxy` * `QuerysetProxy.fields(columns: Union[List, str, set, dict])` method * `QuerysetProxy.exclude_fields(columns: Union[List, str, set, dict])` method - + !!!tip To read more about any or all of those functions visit [selecting columns](./select-columns.md) section. @@ -182,22 +182,22 @@ Instead of ormar models return raw data in form list of dictionaries or tuples. * `QuerysetProxy.paginate(page: int)` method * `QuerysetProxy.limit(limit_count: int)` method * `QuerysetProxy.offset(offset: int)` method - + !!!tip To read more about any or all of those functions visit [pagination](./pagination-and-rows-number.md) section. ### [Aggregated functions](./aggregations.md) -* `count() -> int` +* `count(distinct: bool = True) -> int` * `exists() -> bool` * `QuerysetProxy` - * `QuerysetProxy.count()` method + * `QuerysetProxy.count(distinct=True)` method * `QuerysetProxy.exists()` method - + !!!tip To read more about any or all of those functions visit [aggregations](./aggregations.md) section. - + [relations]: ../relations/index.md diff --git a/docs/relations/queryset-proxy.md b/docs/relations/queryset-proxy.md index fc22d75..0150913 100644 --- a/docs/relations/queryset-proxy.md +++ b/docs/relations/queryset-proxy.md @@ -6,35 +6,35 @@ But at the same time it exposes subset of QuerySet API, so you can filter, creat !!!note By default exposed QuerySet is already filtered to return only `Models` related to parent `Model`. - + So if you issue `post.categories.all()` you will get all categories related to that post, not all in table. !!!note - Note that when accessing QuerySet API methods through QuerysetProxy you don't + Note that when accessing QuerySet API methods through QuerysetProxy you don't need to use `objects` attribute like in normal queries. - + So note that it's `post.categories.all()` and **not** `post.categories.objects.all()`. - + To learn more about available QuerySet methods visit [queries][queries] !!!warning Querying related models from ManyToMany cleans list of related models loaded on parent model: - + Example: `post.categories.first()` will set post.categories to list of 1 related model -> the one returned by first() - - Example 2: if post has 4 categories so `len(post.categories) == 4` calling `post.categories.limit(2).all()` + + Example 2: if post has 4 categories so `len(post.categories) == 4` calling `post.categories.limit(2).all()` -> will load only 2 children and now `assert len(post.categories) == 2` - + This happens for all QuerysetProxy methods returning data: `get`, `all` and `first` and in `get_or_create` if model already exists. - - Note that value returned by `create` or created in `get_or_create` and `update_or_create` + + Note that value returned by `create` or created in `get_or_create` and `update_or_create` if model does not exist will be added to relation list (not clearing it). ## Read data from database ### get -`get(**kwargs): -> Model` +`get(**kwargs): -> Model` To grab just one of related models filtered by name you can use `get(**kwargs)` method. @@ -67,7 +67,7 @@ Tries to get a row meeting the criteria and if NoMatch exception is raised it cr `all(**kwargs) -> List[Optional["Model"]]` -To get a list of related models use `all()` method. +To get a list of related models use `all()` method. Note that you can filter the queryset, select related, exclude fields etc. like in normal query. @@ -88,7 +88,7 @@ assert news_posts[0].author == guido ### create -`create(**kwargs): -> Model` +`create(**kwargs): -> Model` Create related `Model` directly from parent `Model`. @@ -105,7 +105,7 @@ assert len(await post.categories.all()) == 2 Read more in queries documentation [create][create] For `ManyToMany` relations there is an additional functionality of passing parameters -that will be used to create a through model if you declared additional fields on explicitly +that will be used to create a through model if you declared additional fields on explicitly provided Through model. Given sample like this: @@ -274,7 +274,7 @@ With exclude_fields() you can select subset of model columns that will be exclud ### count -`count() -> int` +`count(distinct: bool = True) -> int` Returns number of rows matching the given criteria (i.e. applied with filter and exclude) @@ -309,4 +309,4 @@ Returns a bool value to confirm if there are rows matching the given criteria (a [exists]: ../queries/aggregations.md#exists [fields]: ../queries/select-columns.md#fields [exclude_fields]: ../queries/select-columns.md#exclude_fields -[order_by]: ../queries/filter-and-sort.md#order_by \ No newline at end of file +[order_by]: ../queries/filter-and-sort.md#order_by diff --git a/ormar/protocols/queryset_protocol.py b/ormar/protocols/queryset_protocol.py index 397f58b..fe58ca1 100644 --- a/ormar/protocols/queryset_protocol.py +++ b/ormar/protocols/queryset_protocol.py @@ -26,7 +26,7 @@ class QuerySetProtocol(Protocol): # pragma: nocover async def exists(self) -> bool: ... - async def count(self) -> int: + async def count(self, distinct: bool = True) -> int: ... async def clear(self) -> int: diff --git a/ormar/queryset/queryset.py b/ormar/queryset/queryset.py index beae8d7..0f3af12 100644 --- a/ormar/queryset/queryset.py +++ b/ormar/queryset/queryset.py @@ -678,16 +678,25 @@ class QuerySet(Generic[T]): expr = sqlalchemy.exists(expr).select() return await self.database.fetch_val(expr) - async def count(self) -> int: + async def count(self, distinct: bool = True) -> int: """ Returns number of rows matching the given criteria (applied with `filter` and `exclude` if set before). + If `distinct` is `True` (the default), this will return the number of primary rows selected. If `False`, + the count will be the total number of rows returned + (including extra rows for `one-to-many` or `many-to-many` left `select_related` table joins). + `False` is the legacy (buggy) behavior for workflows that depend on it. + + :param distinct: flag if the primary table rows should be distinct or not :return: number of rows :rtype: int """ expr = self.build_select_expression().alias("subquery_for_count") expr = sqlalchemy.func.count().select().select_from(expr) + if distinct: + expr_distinct = expr.group_by(self.model_meta.pkname).alias("subquery_for_group") + expr = sqlalchemy.func.count().select().select_from(expr_distinct) return await self.database.fetch_val(expr) async def _query_aggr_function(self, func_name: str, columns: List) -> Any: diff --git a/ormar/relations/querysetproxy.py b/ormar/relations/querysetproxy.py index 8170cf3..791d950 100644 --- a/ormar/relations/querysetproxy.py +++ b/ormar/relations/querysetproxy.py @@ -193,17 +193,22 @@ class QuerysetProxy(Generic[T]): """ return await self.queryset.exists() - async def count(self) -> int: + async def count(self, distinct: bool = True) -> int: """ Returns number of rows matching the given criteria (applied with `filter` and `exclude` if set before). + If `distinct` is `True` (the default), this will return the number of primary rows selected. If `False`, + the count will be the total number of rows returned + (including extra rows for `one-to-many` or `many-to-many` left `select_related` table joins). + `False` is the legacy (buggy) behavior for workflows that depend on it. Actual call delegated to QuerySet. + :param distinct: flag if the primary table rows should be distinct or not :return: number of rows :rtype: int """ - return await self.queryset.count() + return await self.queryset.count(distinct=distinct) async def max(self, columns: Union[str, List[str]]) -> Any: # noqa: A003 """ diff --git a/tests/test_queries/test_aggr_functions.py b/tests/test_queries/test_aggr_functions.py index 92c10a1..8cc4ae0 100644 --- a/tests/test_queries/test_aggr_functions.py +++ b/tests/test_queries/test_aggr_functions.py @@ -175,3 +175,15 @@ async def test_queryset_method(): assert await author.books.max(["year", "title"]) == dict( year=1930, title="Book 3" ) + +@pytest.mark.asyncio +async def test_count_method(): + async with database: + await sample_data() + + count = await Author.objects.select_related("books").count() + assert count == 1 + + # The legacy functionality + count = await Author.objects.select_related("books").count(distinct=False) + assert count == 3