Setup simple tables or collections

MongoDB

# If data is in a format natively supported by MongoDB, we don't need to do anything.
# However to manually specify datatypes, do as below
from superduper import Schema, Document
from superduper_pillow import pil_image
from superduper.components.datatype import pickle_serializer

fields = {
    'serialized_content': pickle_serializer,
    'img_content': pil_image,
}

schema = Schema(identifier="my-schema", fields=fields)

# Add schema to system
db.apply(schema)

# Now assert `Document` instances, specifying this schema
db['documents'].insert_many([
    Document({
        'serialized_content': item,
        'img_content': img,
    }, schema='my-schema')
    for item, img in data
])        

SQL

# If data is in a format natively supported by MongoDB, we don't need to do anything.
# However to manually specify datatypes, do as below
from superduper import Schema
from superduper_pillow import pil_image_hybrid
from superduper.components.datatype import pickle_serializer

fields = {
    'serialized_content': pickle_serializer,
    'img_content': pil_image_hybrid,
}

schema = Schema(identifier="my-schema", fields=fields)
db.apply(schema)

# Now assert `Document` instances, specifying this schema
db['documents'].insert_many([
    Document({
        'serialized_content': item,
        'img_content': img,
    }, schema='my-schema')
    for item, img in data
])