Prodream/ormar

Fork 0

Files

collerek 72b0336b75 update docs with tip on order_by

2020-11-26 12:22:12 +01:00

24 KiB

Raw Blame History

Queries

QuerySet

Each Model is auto registered with a QuerySet that represents the underlaying query and it's options.

Most of the methods are also available through many to many relation interface.

!!!info To see which one are supported and how to construct relations visit relations.

Given the Models like this

--8<-- "../docs_src/queries/docs001.py"

we can demonstrate available methods to fetch and save the data into the database.

create

create(**kwargs): -> Model

Creates the model instance, saves it in a database and returns the updates model (with pk populated if not passed and autoincrement is set).

The allowed kwargs are Model fields names and proper value types.

malibu = await Album.objects.create(name="Malibu")
await Track.objects.create(album=malibu, title="The Bird", position=1)

The alternative is a split creation and persistence of the Model.

malibu = Album(name="Malibu")
await malibu.save()

!!!tip Check other Model methods in models

get

get(**kwargs): -> Model

Get's the first row from the db meeting the criteria set by kwargs.

If no criteria set it will return the first row in db.

Passing a criteria is actually calling filter(**kwargs) method described below.

track = await Track.objects.get(name='The Bird')
# note that above is equivalent to await Track.objects.filter(name='The Bird').get()
track2 = track = await Track.objects.get()
track == track2 # True since it's the only row in db in our example

!!!warning If no row meets the criteria NoMatch exception is raised.

If there are multiple rows meeting the criteria the `MultipleMatches` exception is raised.

get_or_create

get_or_create(**kwargs) -> Model

Combination of create and get methods.

Tries to get a row meeting the criteria and if NoMatch exception is raised it creates a new one with given kwargs.

album = await Album.objects.get_or_create(name='The Cat')
# object is created as it does not exist
album2 = await Album.objects.get_or_create(name='The Cat')
assert album == album2
# return True as the same db row is returned

!!!warning Despite being a equivalent row from database the album and album2 in example above are 2 different python objects! Updating one of them will not refresh the second one until you excplicitly load() the fresh data from db.

!!!note Note that if you want to create a new object you either have to pass pk column value or pk column has to be set as autoincrement

update

update(each: bool = False, **kwargs) -> int

QuerySet level update is used to update multiple records with the same value at once.

You either have to filter the QuerySet first or provide a each=True flag to update whole table.

If you do not provide this flag or a filter a QueryDefinitionError will be raised.

Return number of rows updated.

--8<-- "../docs_src/queries/docs002.py"

!!!warning Queryset needs to be filtered before updating to prevent accidental overwrite.

To update whole database table `each=True` needs to be provided as a safety switch

update_or_create

update_or_create(**kwargs) -> Model

Updates the model, or in case there is no match in database creates a new one.

--8<-- "../docs_src/queries/docs003.py"

!!!note Note that if you want to create a new object you either have to pass pk column value or pk column has to be set as autoincrement

bulk_create

bulk_create(objects: List["Model"]) -> None

Allows you to create multiple objects at once.

A valid list of Model objects needs to be passed.

--8<-- "../docs_src/queries/docs004.py"

bulk_update

bulk_update(objects: List["Model"], columns: List[str] = None) -> None

Allows to update multiple instance at once.

All Models passed need to have primary key column populated.

You can also select which fields to update by passing columns list as a list of string names.

# continuing the example from bulk_create
# update objects
for todo in todoes:
    todo.completed = False

# perform update of all objects at once
# objects need to have pk column set, otherwise exception is raised
await ToDo.objects.bulk_update(todoes)

completed = await ToDo.objects.filter(completed=False).all()
assert len(completed) == 3

delete

delete(each: bool = False, **kwargs) -> int

QuerySet level delete is used to delete multiple records at once.

You either have to filter the QuerySet first or provide a each=True flag to delete whole table.

If you do not provide this flag or a filter a QueryDefinitionError will be raised.

Return number of rows deleted.

--8<-- "../docs_src/queries/docs005.py"

all

all(self, **kwargs) -> List[Optional["Model"]]

Returns all rows from a database for given model for set filter options.

Passing kwargs is a shortcut and equals to calling filter(**kwrags).all().

If there are no rows meeting the criteria an empty list is returned.

tracks = await Track.objects.select_related("album").all(title='Sample')
# will return a list of all Tracks with title Sample

tracks = await Track.objects.all()
# will return a list of all Tracks in database

filter

filter(**kwargs) -> QuerySet

Allows you to filter by any Model attribute/field as well as to fetch instances, with a filter across an FK relationship.

track = Track.objects.filter(name="The Bird").get()
# will return a track with name equal to 'The Bird'
 
tracks = Track.objects.filter(album__name="Fantasies").all()
# will return all tracks where the columns album name = 'Fantasies'

You can use special filter suffix to change the filter operands:

exact - like album__name__exact='Malibu' (exact match)
iexact - like album__name__iexact='malibu' (exact match case insensitive)
contains - like album__name__conatins='Mal' (sql like)
icontains - like album__name__icontains='mal' (sql like case insensitive)
in - like album__name__in=['Malibu', 'Barclay'] (sql in)
gt - like position__gt=3 (sql >)
gte - like position__gte=3 (sql >=)
lt - like position__lt=3 (sql <)
lte - like position__lte=3 (sql <=)
startswith - like album__name__startswith='Mal' (exact start match)
istartswith - like album__name__istartswith='mal' (exact start match case insensitive)
endswith - like album__name__endswith='ibu' (exact end match)
iendswith - like album__name__iendswith='IBU' (exact end match case insensitive)

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

exclude

exclude(**kwargs) -> QuerySet

Works exactly the same as filter and all modifiers (suffixes) are the same, but returns a not condition.

So if you use filter(name='John') which equals to where name = 'John' in SQL, the exclude(name='John') equals to where name <> 'John'

Note that all conditions are joined so if you pass multiple values it becomes a union of conditions.

exclude(name='John', age>=35) will become where not (name='John' and age>=35)

notes = await Track.objects.exclude(position_gt=3).all()
# returns all tracks with position < 3

select_related

select_related(related: Union[List, str]) -> QuerySet

Allows to prefetch related models during the same query.

With select_related always only one query is run against the database, meaning that one (sometimes complicated) join is generated and later nested models are processed in python.

To fetch related model use ForeignKey names.

To chain related Models relation use double underscores between names.

!!!note If you are coming from django note that ormar select_related differs -> in django you can select_related only singe relation types, while in ormar you can select related across ForeignKey relation, reverse side of ForeignKey (so virtual auto generated keys) and ManyToMany fields (so all relations as of current version).

!!!tip To control which model fields to select use fields() and exclude_fields() QuerySet methods.

!!!tip To control order of models (both main or nested) use order_by() method.

album = await Album.objects.select_related("tracks").all()
# will return album will all columns tracks

You can provide a string or a list of strings

classes = await SchoolClass.objects.select_related(
["teachers__category", "students"]).all()
# will return classes with teachers and teachers categories
# as well as classes students

Exactly the same behavior is for Many2Many fields, where you put the names of Many2Many fields and the final Models are fetched for you.

!!!warning If you set ForeignKey field as not nullable (so required) during all queries the not nullable Models will be auto prefetched, even if you do not include them in select_related.

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

prefetch_related

prefetch_related(related: Union[List, str]) -> QuerySet

Allows to prefetch related models during query - but opposite to select_related each subsequent model is fetched in a separate database query.

With prefetch_related always one query per Model is run against the database, meaning that you will have multiple queries executed one after another.

To fetch related model use ForeignKey names.

To chain related Models relation use double underscores between names.

!!!tip To control which model fields to select use fields() and exclude_fields() QuerySet methods.

!!!tip To control order of models (both main or nested) use order_by() method.

album = await Album.objects.prefetch_related("tracks").all()
# will return album will all columns tracks

You can provide a string or a list of strings

classes = await SchoolClass.objects.prefetch_related(
["teachers__category", "students"]).all()
# will return classes with teachers and teachers categories
# as well as classes students

Exactly the same behavior is for Many2Many fields, where you put the names of Many2Many fields and the final Models are fetched for you.

!!!warning If you set ForeignKey field as not nullable (so required) during all queries the not nullable Models will be auto prefetched, even if you do not include them in select_related.

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

select_related vs prefetch_related

Which should you use -> select_related or prefetch_related?

Well, it really depends on your data. The best answer is try yourself and see which one performs faster/better in your system constraints.

What to keep in mind:

Performance

Number of queries: select_related always executes one query against the database, while prefetch_related executes multiple queries. Usually the query (I/O) operation is the slowest one but it does not have to be.

Number of rows: Imagine that you have 10 000 object in one table A and each of those objects have 3 children in table B, and subsequently each object in table B has 2 children in table C. Something like this:

                     Model C
                   /
           Model B - Model C
         / 
Model A  - Model B - Model C
       \           \ 
        \            Model C
         \
           Model B - Model C
                   \ 
                     Model C

That means that select_related will always return 60 000 rows (10 000 * 3 * 2) later compacted to 10 000 models.

How many rows will return prefetch_related?

Well, that depends, if each of models B and C is unique it will return 10 000 rows in first query, 30 000 rows (each of 3 children of A in table B are unique) in second query and 60 000 rows (each of 2 children of model B in table C are unique) in 3rd query.

In this case select_related seems like a better choice, not only it will run one query comparing to 3 of prefetch_related but will also return 60 000 rows comparing to 100 000 of prefetch_related (10+30+60k).

But what if each Model A has exactly the same 3 models B and each models C has exactly same models C? select_related will still return 60 000 rows, while prefetch_related will return 10 000 for model A, 3 rows for model B and 2 rows for Model C. So in total 10 006 rows. Now depending on the structure of models (i.e. if it has long Text() fields etc.) prefetch_related might be faster despite it needs to perform three separate queries instead of one.

Memory

ormar is a mini ORM meaning that it does not keep a registry of already loaded models.

That means that in select_related example above you will always have 10 000 Models A, 30 000 Models B (even if the unique number of rows in db is 3 - processing of select_related spawns new child models for each parent model). And 60 000 Models C.

If the same Model B is shared by rows 1, 10, 100 etc. and you update one of those, the rest of rows that share the same child will not be updated on the spot. If you persist your changes into the database the change will be available only after reload (either each child separately or the whole query again). That means that select_related will use more memory as each child is instantiated as a new object - obviously using it's own space.

!!!note This might change in future versions if we decide to introduce caching.

!!!warning By default all children (or event the same models loaded 2+ times) are completely independent, distinct python objects, despite that they represent the same row in db.

They will evaluate to True when compared, so in example above: 

```python
# will return True if child1 of both rows is the same child db row 
row1.child1 == row100.child1

# same here:
model1 = await Model.get(pk=1)
model2 = await Model.get(pk=1) # same pk = same row in db
# will return `True`
model1 == model2
``` 

but 

```python
# will return False (note that id is a python `builtin` function not ormar one).
id(row1.child1) == (ro100.child1)

# from above - will also return False
id(model1) == id(model2)
```

On the contrary - with prefetch_related each unique distinct child model is instantiated only once and the same child models is shared across all parent models. That means that in prefetch_related example above if there are 3 distinct models in table B and 2 in table C, there will be only 5 children nested models shared between all model A instances. That also means that if you update any attribute it will be updated on all parents as they share the same child object.

limit

limit(limit_count: int) -> QuerySet

You can limit the results to desired number of rows.

tracks = await Track.objects.limit(1).all()
# will return just one Track

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

offset

offset(offset: int) -> QuerySet

You can also offset the results by desired number of rows.

tracks = await Track.objects.offset(1).limit(1).all()
# will return just one Track, but this time the second one

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

count

count() -> int

Returns number of rows matching the given criteria (applied with filter and exclude)

# returns count of rows in db
no_of_books = await Book.objects.count()

exists

exists() -> bool

Returns a bool value to confirm if there are rows matching the given criteria (applied with filter and exclude)

# returns a boolean value if given row exists
has_sample = await Book.objects.filter(title='Sample').exists()

fields

fields(columns: Union[List, str, set, dict]) -> QuerySet

With fields() you can select subset of model columns to limit the data load.

!!!note Note that fields() and exclude_fields() works both for main models (on normal queries like get, all etc.) as well as select_related and prefetch_related models (with nested notation).

Given a sample data like following:

--8<-- "../docs_src/queries/docs006.py"

You can select specified fields by passing a str, List[str], Set[str] or dict with nested definition.

To include related models use notation {related_name}__{column}[__{optional_next} etc.].

all_cars = await Car.objects.select_related('manufacturer').fields(['id', 'name', 'manufacturer__name']).all()
for car in all_cars:
    # excluded columns will yield None
    assert all(getattr(car, x) is None for x in ['year', 'gearbox_type', 'gears', 'aircon_type'])
    # included column on related models will be available, pk column is always included
    # even if you do not include it in fields list
    assert car.manufacturer.name == 'Toyota'
    # also in the nested related models - you cannot exclude pk - it's always auto added
    assert car.manufacturer.founded is None

fields() can be called several times, building up the columns to select.

If you include related models into select_related() call but you won't specify columns for those models in fields

implies a list of all fields for those nested models.

all_cars = await Car.objects.select_related('manufacturer').fields('id').fields(
    ['name']).all()
# all fiels from company model are selected
assert all_cars[0].manufacturer.name == 'Toyota'
assert all_cars[0].manufacturer.founded == 1937

!!!warning Mandatory fields cannot be excluded as it will raise ValidationError, to exclude a field it has to be nullable.

You cannot exclude mandatory model columns - manufacturer__name in this example.

await Car.objects.select_related('manufacturer').fields(['id', 'name', 'manufacturer__founded']).all()
# will raise pydantic ValidationError as company.name is required

!!!tip Pk column cannot be excluded - it's always auto added even if not explicitly included.

You can also pass fields to include as dictionary or set.

To mark a field as included in a dictionary use it's name as key and ellipsis as value.

To traverse nested models use nested dictionaries.

To include fields at last level instead of nested dictionary a set can be used.

To include whole nested model specify model related field name and ellipsis.

Below you can see examples that are equivalent:

--8<-- "../docs_src/queries/docs009.py"

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

exclude_fields

exclude_fields(columns: Union[List, str, set, dict]) -> QuerySet

With exclude_fields() you can select subset of model columns that will be excluded to limit the data load.

It's the opposite of fields() method so check documentation above to see what options are available.

Especially check above how you can pass also nested dictionaries and sets as a mask to exclude fields from whole hierarchy.

Below you can find few simple examples:

--8<-- "../docs_src/queries/docs008.py"

!!!warning Mandatory fields cannot be excluded as it will raise ValidationError, to exclude a field it has to be nullable.

!!!tip Pk column cannot be excluded - it's always auto added even if explicitly excluded.

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

order_by

order_by(columns: Union[List, str]) -> QuerySet

With order_by() you can order the results from database based on your choice of fields.

You can provide a string with field name or list of strings with different fields.

Ordering in sql will be applied in order of names you provide in order_by.

!!!tip By default if you do not provide ordering ormar explicitly orders by all primary keys

!!!warning If you are sorting by nested models that causes that the result rows are unsorted by the main model ormar will combine those children rows into one main model.

Sample raw database rows result (sort by child model desc):
```
MODEL: 1 - Child Model - 3
MODEL: 2 - Child Model - 2
MODEL: 1 - Child Model - 1
```

will result in 2 rows of result:
```
MODEL: 1 - Child Models: [3, 1] # encountered first in result, all children rows combined
MODEL: 2 - Child Modles: [2]
```

The main model will never duplicate in the result

Given sample Models like following:

--8<-- "../docs_src/queries/docs007.py"

To order by main model field just provide a field name

toys = await Toy.objects.select_related("owner").order_by("name").all()
assert [x.name.replace("Toy ", "") for x in toys] == [
    str(x + 1) for x in range(6)
]
assert toys[0].owner == zeus
assert toys[1].owner == aphrodite

To sort on nested models separate field names with dunder '__'.

You can sort this way across all relation types -> ForeignKey, reverse virtual FK and ManyToMany fields.

toys = await Toy.objects.select_related("owner").order_by("owner__name").all()
assert toys[0].owner.name == toys[1].owner.name == "Aphrodite"
assert toys[2].owner.name == toys[3].owner.name == "Hermes"
assert toys[4].owner.name == toys[5].owner.name == "Zeus"

To sort in descending order provide a hyphen in front of the field name

owner = (
    await Owner.objects.select_related("toys")
        .order_by("-toys__name")
        .filter(name="Zeus")
        .get()
)
assert owner.toys[0].name == "Toy 4"
assert owner.toys[1].name == "Toy 1"

!!!note All methods that do not return the rows explicitly returns a QueySet instance so you can chain them together

So operations like `filter()`, `select_related()`, `limit()` and `offset()` etc. can be chained.

Something like `Track.object.select_related("album").filter(album__name="Malibu").offset(1).limit(1).all()`

24 KiB Raw Blame History

Queries

QuerySet

create

get

get_or_create

update

update_or_create

bulk_create

bulk_update

delete

all

filter

exclude

select_related

prefetch_related

select_related vs prefetch_related

Performance

Memory

limit

offset

count

exists

fields

exclude_fields

order_by

24 KiB

Raw Blame History