clear-skies 1.19.22__py3-none-any.whl → 2.0.23__py3-none-any.whl

clearskies/column.py

@@ -0,0 +1,1221 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any, Callable, Self, overload
+
+from clearskies import configs, configurable, decorators
+from clearskies.autodoc.schema import String as AutoDocString
+from clearskies.di import InjectableProperties, inject
+from clearskies.query.condition import ParsedCondition
+from clearskies.validator import Validator
+
+if TYPE_CHECKING:
+    from clearskies import Model, Schema, typing
+    from clearskies.autodoc.schema import Schema as AutoDocSchema
+    from clearskies.query.condition import Condition
+
+
+class Column(configurable.Configurable, InjectableProperties):
+    """
+    Columns are used to build schemes and enable a variety of levels of automation with.
+
+    Columns are used to define your schemas in clearskies, especially via models.  The column definitions are then used by endpoints
+    and other aspects of the clearskies framework to automate things like input validation, front-end/backend-transformations, and more.
+    """
+
+    """
+    The column class gets the full DI container, because it does a lot of object building itself
+    """
+    di = inject.Di()
+
+    """
+    A default value to set for this column.
+
+    The default is only used when creating a record for the first time, and only if
+    a value for this column has not been set.
+
+    ```python
+    import clearskies
+
+    class Widget(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        name = clearskies.columns.String(default="Jane Doe")
+
+    cli = clearskies.contexts.Cli(
+        clearskies.endpoints.Callable(
+            lambda widgets: widgets.create(no_data=True),
+            model_class=Widget,
+            readable_column_names=["id", "name"]
+        ),
+        classes=[Widget],
+    )
+
+    if __name__ == "__main__":
+        cli()
+    ```
+
+    Which when invoked returns:
+
+    ```json
+    {
+        "status": "success",
+        "error": "",
+        "data": {
+            "id": "03806afa-b189-4729-a43c-9da5aa17bf14",
+            "name": "Jane Doe"
+        },
+        "pagination": {},
+        "input_errors": {}
+    }
+    ```
+    """
+    default = configs.string.String(default=None)
+
+    """
+    A value to set for this column during a save operation.
+
+    Unlike the default value, a setable value is always set during a save, even on update.  It will
+    even override other values, so it is intended to be used in cases where the value is always controlled
+    programmatically.
+
+    ```python
+    import clearskies
+    import datetime
+
+    class Pet(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        name = clearskies.columns.String(setable="Spot")
+        date_of_birth = clearskies.columns.Date()
+        age = clearskies.columns.Integer(
+            setable=lambda data, model, now:
+                (now-dateparser.parse(model.latest("date_of_birth", data))).total_seconds()/(86400*365),
+        )
+        created = clearskies.columns.Created()
+
+    cli = clearskies.contexts.Cli(
+        clearskies.endpoints.Callable(
+            lambda pets: pets.create({"date_of_birth": "2020-05-03"}),
+            model_class=Pet,
+            readable_column_names=["id", "name", "date_of_birth", "age"]
+        ),
+        classes=[Pet],
+    )
+
+    if __name__ == "__main__":
+        cli()
+    ```
+
+    Note the use of `model.latest()` above.  This function returns either the column information from the data array
+    or, if not present, the latest column value from the model itself.  This makes it more flexible as it works
+    well with both update and create operations.
+
+    If you then execute this it will return something like:
+
+    ```json
+    {
+        "status": "success",
+        "error": "",
+        "data": {
+            "id": "ec4993f4-124a-44a2-8313-816d2ad51aae",
+            "name": "Spot",
+            "date_of_birth": "2020-05-03",
+            "age": 5,
+            "created": "2025-05-03T20:23:32+00:00"
+        },
+        "pagination": {},
+        "input_errors": {}
+    }
+    ```
+
+    e.g., `date_of_birth` is `age` years behind the current time (as recorded in the `created` timestamp).
+
+    """
+    setable = configs.string_or_callable.StringOrCallable(default=None)
+
+    """
+    Whether or not this column can be converted to JSON and included in an API response.
+
+    If this is set to False for a column and you attempt to set that column as a readable_column in an endpoint,
+    clearskies will throw an exception.
+    """
+    is_readable = configs.boolean.Boolean(default=True)
+
+    """
+    Whether or not this column can be set via an API call.
+
+    If this is set to False for a column and you attempt to set the column as a writeable column in an endpoint,
+    clearskies will throw an exception.
+    """
+    is_writeable = configs.boolean.Boolean(default=True)
+
+    """
+    Whether or not it is possible to search by this column
+
+    If this is set to False for a column and you attempt to set the column as a searchable column in an endpoint,
+    clearskies will throw an exception.
+    """
+    is_searchable = configs.boolean.Boolean(default=True)
+
+    """
+    Whether or not this column is temporary.  A temporary column is not persisted to the backend.
+
+    Temporary columns are useful when you want the developer or end user to set a value, but you use that value to
+    trigger additional behavior, rather than actually recording it.  Temporary columns often team up with actions
+    or are used to calculate other values.  For instance, in our setable example above, we had both an age and
+    a date of birth column, with the date of birth calculated from the age.  This obviously results in two columns
+    with similar data.  One could be marked as temporary and it will be available during the save operation, but
+    it will be skipped when saving data to the backend:
+
+    ```python
+    import clearskies
+
+    class Pet(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        name = clearskies.columns.String()
+        date_of_birth = clearskies.columns.Date(is_temporary=True)
+        age = clearskies.columns.Integer(
+            setable=lambda data, model, now:
+                (now-dateparser.parse(model.latest("date_of_birth", data))).total_seconds()/(86400*365),
+        )
+        created = clearskies.columns.Created()
+
+    cli = clearskies.contexts.Cli(
+        clearskies.endpoints.Callable(
+            lambda pets: pets.create({"name": "Spot", "date_of_birth": "2020-05-03"}),
+            model_class=Pet,
+            readable_column_names=["id", "age", "date_of_birth"],
+        ),
+        classes=[Pet],
+    )
+
+    if __name__ == "__main__":
+        cli()
+    ```
+
+    Which will return:
+
+    ```json
+    {
+        "status": "success",
+        "error": "",
+        "data": {
+            "id": "ee532cfa-91cf-4747-b798-3c6dcd79326e",
+            "age": 5,
+            "date_of_birth": null
+        },
+        "pagination": {},
+        "input_errors": {}
+    }
+    ```
+
+    e.g. the date_of_birth column is empty.  To be clear though, it's not just empty - clearskies made no attempt to set it.
+    If you were using an SQL database, you would not have to put a `date_of_birth` column in your table.
+
+    """
+    is_temporary = configs.boolean.Boolean(default=False)
+
+    """
+    Validators to use when checking the input for this column during write operations from the API.
+
+    Keep in mind that the validators are only checked when the column is exposed via a supporting endpoint.
+    You can still set whatever values you want when saving the model directly, e.g. `model.save(...)`
+
+    In the below example, we require a name that is at least 5 characters long, and the date of birth must
+    be in the past.  Note that date of birth is not required, so the end-user can create a record without
+    a date of birth.  In general, only the `Required` validator will reject a non-existent input.
+
+    ```python
+    import clearskies
+
+    class Pet(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        name = clearskies.columns.String(validators=[
+            clearskies.validators.Required(),
+            clearskies.validators.MinimumLength(5),
+        ])
+        date_of_birth = clearskies.columns.Date(validators=[
+            clearskies.validators.InThePast()
+        ])
+        created = clearskies.columns.Created()
+
+    wsgi = clearskies.contexts.WsgiRef(
+        clearskies.endpoints.Create(
+            model_class=Pet,
+            writeable_column_names=["name", "date_of_birth"],
+            readable_column_names=["id", "name", "date_of_birth", "created"],
+        ),
+    )
+    wsgi()
+    ```
+
+    You can then see the result of calling the endpoint with various kinds of invalid data:
+
+    ```bash
+    $ curl http://localhost:8080 -d '{"date_of_birth": "asdf"}'
+    {
+        "status": "input_errors",
+        "error": "",
+        "data": [],
+        "pagination": {},
+        "input_errors": {
+            "name": "'name' is required.",
+            "date_of_birth": "given value did not appear to be a valid date"
+        }
+    }
+
+    $ curl http://localhost:8080 -d '{"name":"asdf"}' | jq
+    {
+        "status": "input_errors",
+        "error": "",
+        "data": [],
+        "pagination": {},
+        "input_errors": {
+            "name": "'name' must be at least 5 characters long."
+        }
+    }
+
+    $ curl http://localhost:8080 -d '{"name":"Longer", "date_of_birth": "2050-01-01"}' | jq
+    {
+        "status": "input_errors",
+        "error": "",
+        "data": [],
+        "pagination": {},
+        "input_errors": {
+            "date_of_birth": "'date_of_birth' must be in the past"
+        }
+    }
+
+    $ curl http://localhost:8080 -d '{"name":"Long Enough"}' | jq
+    {
+        "status": "success",
+        "error": "",
+        "data": {
+            "id": "ace16b93-db91-49b3-a8f7-5dc6568d25f6",
+            "name": "Long Enough",
+            "date_of_birth": null,
+            "created": "2025-05-03T19:32:33+00:00"
+        },
+        "pagination": {},
+        "input_errors": {}
+    }
+    ```
+    """
+    validators = configs.validators.Validators(default=[])
+
+    """
+    Actions to take during the pre-save step of the save process if the column has changed during the active save operation.
+
+    Pre-save happens before the data is persisted to the backend.  Actions/callables in
+    this step must return a dictionary.  The data in the dictionary will be included in the save operation.
+    Since the save hasn't completed, any data in the model itself reflects the model before the save
+    operation started.  Actions in the pre-save step must **NOT** make any changes directly, but should **ONLY**
+    return modified data for the save operation.  In addition, they must be idempotent - they should always return
+    the same value when called with the same data.  This is because clearskies can call them more than once.  If
+    a pre-save hook changes the save data, then clearskies will call all the pre-save hooks again in case this
+    new data needs to trigger further changes.  Stateful changes should be reserved for the post_save or save_finished stages.
+
+    Callables and actions can request any dependencies provided by the DI system.  In addition, they can request
+    two named parameters:
+
+     1. `model` - the model involved in the save operation
+     2. `data` - the new data being saved
+
+    The key here is that the defined actions will be invoked regardless of how the save happens.  Whether the
+    model.save() function is called directly or the model is creatd/modified via an endpoint, your business logic
+    will always be executed.  This makes for easy reusability and consistency throughout your application.
+
+    Here's an example where we want to record a timestamp anytime an order status becomes a particular value:
+
+    ```python
+    import clearskies
+
+    class Order(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        status = clearskies.columns.Select(
+            ["Open", "On Hold", "Fulfilled"],
+            on_change_pre_save=[
+                lambda data, utcnow: {"fulfilled_at": utcnow} if data["status"] == "Fulfilled" else {},
+            ],
+        )
+        fulfilled_at = clearskies.columns.Datetime()
+
+    wsgi = clearskies.contexts.WsgiRef(
+        clearskies.endpoints.Create(
+            model_class=Order,
+            writeable_column_names=["status"],
+            readable_column_names=["id", "status", "fulfilled_at"],
+        ),
+    )
+    wsgi()
+    ```
+
+    You can then see the difference depending on what you set the status to:
+
+    ```bash
+    $ curl http://localhost:8080 -d '{"status":"Open"}' | jq
+    {
+        "status": "success",
+        "error": "",
+        "data": {
+            "id": "a732545f-51b3-4fd0-a6cf-576cf1b2872f",
+            "status": "Open",
+            "fulfilled_at": null
+        },
+        "pagination": {},
+        "input_errors": {}
+    }
+
+    $ curl http://localhost:8080 -d '{"status":"Fulfilled"}' | jq
+    {
+        "status": "success",
+        "error": "",
+        "data": {
+            "id": "c288bf43-2246-48e4-b168-f40cbf5376df",
+            "status": "Fulfilled",
+            "fulfilled_at": "2025-05-04T02:32:56+00:00"
+        },
+        "pagination": {},
+        "input_errors": {}
+    }
+
+    ```
+
+    """
+    on_change_pre_save = configs.actions.Actions(default=[])
+
+    """
+    Actions to take during the post-save step of the process if the column has changed during the active save.
+
+    Post-save happens after the data is persisted to the backend but before the full save process has finished.
+    Since the data has been persisted to the backend, any data returned by the callables/actions will be ignored.
+    If you need to make data changes you'll have to execute a separate save operation.
+    Since the save hasn't finished, the model is not yet updated with the new data, and
+    any data you fetch out of the model will refelect the data in the model before the save started.
+
+    Callables and actions can request any dependencies provided by the DI system.  In addition, they can request
+    three named parameters:
+
+     1. `model` - the model involved in the save operation
+     2. `data` - the new data being saved
+     3. `id` - the id of the record being saved
+
+    Here's an example of using a post-save action to record a simple audit trail when the order status changes:
+
+    ```python
+    import clearskies
+
+    class Order(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        status = clearskies.columns.Select(
+            ["Open", "On Hold", "Fulfilled"],
+            on_change_post_save=[
+                lambda model, data, order_histories: order_histories.create({
+                    "order_id": model.latest("id", data),
+                    "event": "Order status changed to " + data["status"]
+                }),
+            ],
+        )
+
+    class OrderHistory(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        event = clearskies.columns.String()
+        order_id = clearskies.columns.BelongsToId(Order)
+
+        # include microseconds in the created_at time so that we can sort our example by created_at
+        # and they come out in order (since, for our test program, they will all be created in the same second).
+        created_at = clearskies.columns.Created(date_format="%Y-%m-%d %H:%M:%S.%f")
+
+    def test_post_save(orders: Order, order_histories: OrderHistory):
+        my_order = orders.create({"status": "Open"})
+        my_order.status = "On Hold"
+        my_order.save()
+        my_order.save({"status": "Open"})
+        my_order.save({"status": "Fulfilled"})
+        return order_histories.where(OrderHistory.order_id.equals(my_order.id)).sort_by("created_at", "asc")
+
+    cli = clearskies.contexts.Cli(
+        clearskies.endpoints.Callable(
+            test_post_save,
+            model_class=OrderHistory,
+            return_records=True,
+            readable_column_names=["id", "event", "created_at"],
+        ),
+        classes=[Order, OrderHistory],
+    )
+    cli()
+    ```
+
+    Note that in our `on_change_post_save` lambda function, we use `model.latest("id", data)`.  We can't just use
+    `data["id"]` because `data` is a dictionary containing the information present in the save.  During the create
+    operation `data["id"]` will be populated, but during the subsequent edit operations it won't be - only the status
+    column is changing.  `model.latest("id", data)` is basically just short hand for: `data.get("id", model.id)`.
+    On the other hand, we can just use `data["status"]` because the `on_change` hook is attached to the status field,
+    so it will only fire when status is being changed, which means that the `status` key is guaranteed to be in
+    the dictionary when the lambda is executed.
+
+    Finally, the post-save action has a named parameter called `id`, so in this specific case we could use:
+
+    ```python
+    lambda data, id, order_histories: order_histories.create("order_id": id, "event": data["status"])
+    ```
+
+    When we execute the above script it will return something like:
+
+    ```json
+    {
+        "status": "success",
+        "error": "",
+        "data": [
+            {
+            "id": "c550d714-839b-4f25-a9e1-bd7e977185ff",
+            "event": "Order status changed to Open",
+            "created_at": "2025-05-04T14:09:42.960119+00:00"
+            },
+            {
+            "id": "f393d7b0-da21-4117-a7a4-0359fab802bb",
+            "event": "Order status changed to On Hold",
+            "created_at": "2025-05-04T14:09:42.960275+00:00"
+            },
+            {
+            "id": "5b528a10-4a08-47ae-938c-fc7067603f8e",
+            "event": "Order status changed to Open",
+            "created_at": "2025-05-04T14:09:42.960395+00:00"
+            },
+            {
+            "id": "91f77a88-1c38-49f7-aa1e-7f97bd9f962f",
+            "event": "Order status changed to Fulfilled",
+            "created_at": "2025-05-04T14:09:42.960514+00:00"
+            }
+        ],
+        "pagination": {},
+        "input_errors": {}
+    }
+    ```
+
+    """
+    on_change_post_save = configs.actions.Actions(default=[])
+
+    """
+    Actions to take during the save-finished step of the save process if the column has changed in the save.
+
+    Save-finished happens after the save process has completely finished and the model is updated with
+    the final data.  Any data returned by these actions will be ignored, since the save has already finished.
+    If you need to make data changes you'll have to execute a separate save operation.
+
+    Callables and actions can request any dependencies provided by the DI system.  In addition, they can request
+    the following parameter:
+
+     1. `model` - the model involved in the save operation
+
+    Unlike pre_save and post_save, `data` is not provided because this data has already been merged into the
+    model.  If you need some context from the completed save operation, use methods like `was_changed` and `previous_value`.
+    """
+    on_change_save_finished = configs.actions.Actions(default=[])
+
+    """
+    Use in conjunction with `created_by_source_type` to have this column automatically populated by data from an HTTP request.
+
+    So, for instance, setting `created_by_source_type` to `authorization_data` and setting this to `email`
+    will result in the email value from the authorization data being persisted into this column when the
+    record is saved.
+
+    NOTE: this is sometimes best set as a column override on an endpoint, rather than directly
+    on the model itself.  The reason is because the authorization data and header information is typically
+    only available during an HTTP request, so if you set this on the model level, you'll get an error
+    if you try to make saves to the model in a context where authorization data and/or headers don't exist.
+
+    See created_by_source_type for usage examples.
+    """
+    created_by_source_key = configs.string.String(default="")
+
+    """
+    Use in conjunction with `created_by_source_key` to have this column automatically populated by data from ann HTTP request.
+
+    So, for instance, setting this to `authorization_data` and setting `created_by_source_key` to `email`
+    will result in the email value from the authorization data being persisted into this column when the
+    record is saved.
+
+    NOTE: this is sometimes best set as a column override on an endpoint, rather than directly
+    on the model itself.  The reason is because the authorization data and header information is typically
+    only available during an HTTP request, so if you set this on the model level, you'll get an error
+    if you try to make saves to the model in a context where authorization data and/or headers don't exist.
+
+    Here's an example:
+
+    ```python
+    class User(clearskies.Model):
+        id_column_name = "id"
+        backend = clearskies.backends.MemoryBackend()
+
+        id = clearskies.columns.Uuid()
+        name = clearskies.columns.String()
+        account_id = clearskies.columns.String(
+            created_by_source_type="routing_data",
+            created_by_source_key="account_id",
+        )
+
+    wsgi = clearskies.contexts.WsgiRef(
+        clearskies.endpoints.Create(
+            User,
+            readable_column_names=["id", "account_id", "name"],
+            writeable_column_names=["name"],
+            url="/:account_id",
+        ),
+    )
+    wsgi()
+    ```
+
+    Note that `created_by_source_type` is `routing_data` and `created_by_source_key` is `account_id`.
+    This means that the endpoint that creates this record must have a routing parameter named `account_id`.
+    Naturally, our endpoint has a url of `/:account_id`, and so the parameter provided by the uesr gets
+    reflected into the save.
+
+    ```bash
+    $ curl http://localhost:8080/1-2-3-4 -d '{"name":"Bob"}' | jq
+    {
+        "status": "success",
+        "error": "",
+        "data": {
+            "id": "250ed725-d940-4823-aa9d-890be800404a",
+            "account_id": "1-2-3-4",
+            "name": "Bob"
+        },
+        "pagination": {},
+        "input_errors": {}
+    }
+    ```
+
+    """
+    created_by_source_type = configs.select.Select(
+        ["authorization_data", "http_header", "routing_data", ""], default=""
+    )
+
+    """
+    If True, and the key requested via created_by_source_key doesn't exist in the designated source, an error will be raised.
+    """
+    created_by_source_strict = configs.boolean.Boolean(default=True)
+
+    """ The model class this column is associated with. """
+    model_class = configs.Schema()
+
+    """ The name of this column. """
+    name = configs.string.String()
+
+    """
+    Simple flag to denote if the column is unique or not.
+
+    This is an internal cache.  Use column.is_unique instead.
+    """
+    _is_unique = False
+
+    """
+    Specify if this column has additional functionality to solve the n+1 problem.
+
+    Relationship columns may fetch data from additional tables when outputting results, but by default they
+    end up making an additional query for every record (in order to grab related data).  This is called the
+    n+1 problem - a query may fetch 10 records, and then make 10 individual additional queries to select
+    related data for each record (which obviously hampers performance).  The solution to this (when using
+    sql-like backends) is to add additional joins to the original query so that the data can all be fetched
+    at once.  Columns that are subject to this issue can set this flag to True and then  define the
+    `configure_n_plus_one` method to add the necessary joins.  This method will be called as needed.
+    """
+    wants_n_plus_one = False
+
+    """
+    Simple flag to denote if the column is required or not.
+
+    This is an internal cache.  Use column.is_required instead.
+    """
+    _is_required = False
+
+    """
+    The list of allowed search operators for this column.
+
+    All the various search methods reference this list.  The idea is that a column can just fill out this list
+    instead of having to override all the methods.
+    """
+    _allowed_search_operators = ["<=>", "!=", "<=", ">=", ">", "<", "=", "in", "is not null", "is null", "like"]
+
+    """
+    The class to use when documenting this column
+    """
+    auto_doc_class: type[AutoDocSchema] = AutoDocString
+
+    @decorators.parameters_to_properties
+    def __init__(
+        self,
+        default: str | None = None,
+        setable: str | Callable[..., str] | None = None,
+        is_readable: bool = True,
+        is_writeable: bool = True,
+        is_searchable: bool = True,
+        is_temporary: bool = False,
+        validators: typing.validator | list[typing.validator] = [],
+        on_change_pre_save: typing.action | list[typing.action] = [],
+        on_change_post_save: typing.action | list[typing.action] = [],
+        on_change_save_finished: typing.action | list[typing.action] = [],
+        created_by_source_type: str = "",
+        created_by_source_key: str = "",
+        created_by_source_strict: bool = True,
+    ):
+        pass
+
+    def get_model_columns(self):
+        """Return the columns or the model this column is attached to."""
+        return self.model_class.get_columns()
+
+    def finalize_configuration(self, model_class: type[Schema], name: str) -> None:
+        """
+        Finalize and check the configuration.
+
+        This is an external trigger called by the model class when the model class is ready.
+        The reason it exists here instead of in the constructor is because some columns are tightly
+        connected to the model class, and can't validate configuration until they know what the model is.
+        Therefore, we need the model involved, and the only way for a property to know what class it is
+        in is if the parent class checks in (which is what happens here).
+        """
+        self.model_class = model_class
+        self.name = name
+        self.finalize_and_validate_configuration()
+
+    def from_backend(self, value):
+        """
+        Take the backend representation and returns a python representation.
+
+        For instance, for an SQL date field, this will return a Python DateTime object
+        """
+        return str(value)
+
+    def to_backend(self, data: dict[str, Any]) -> dict[str, Any]:
+        """
+        Make any changes needed to save the data to the backend.
+
+        This typically means formatting changes - converting DateTime objects to database
+        date strings, etc...
+        """
+        if self.name not in data:
+            return data
+
+        return {**data, self.name: str(data[self.name])}
+
+    @overload
+    def __get__(self, instance: None, cls: type) -> Self:
+        pass
+
+    @overload
+    def __get__(self, instance: Model, cls: type):
+        pass
+
+    def __get__(self, instance, cls):
+        if instance is None:
+            # Normally this gets filled in when the model is initialized.  However, the condition builders (self.equals, etc...)
+            # can be called from the class directly, before the model is initialized and everything is populated.  This
+            # can cause trouble, but by filling in the model class we can give enough information for them to get the
+            # job done.  They have a special flow for this, we just have to provide the model class (and the __get__
+            # function is always called, so this fixes it).
+            self.model_class = cls
+            return self
+
+        # this makes sure we're initialized
+        if "name" not in self._config:  # type: ignore
+            instance.get_columns()
+
+        if self.name not in instance._data:
+            return None  # type: ignore
+
+        if self.name not in instance._transformed_data:
+            instance._transformed_data[self.name] = self.from_backend(instance._data[self.name])
+
+        return instance._transformed_data[self.name]
+
+    def __set__(self, instance: Model, value) -> None:
+        # this makes sure we're initialized
+        if "name" not in self._config:  # type: ignore
+            instance.get_columns()
+
+        instance._next_data[self.name] = value
+
+    def finalize_and_validate_configuration(self):
+        super().finalize_and_validate_configuration()
+
+        if self.setable is not None and self.created_by_source_type:
+            raise ValueError(
+                "You attempted to set both 'setable' and 'created_by_source_type', but these configurations are mutually exclusive.  You can only set one for a given column"
+            )
+
+        if (self.created_by_source_type and not self.created_by_source_key) or (
+            not self.created_by_source_type and self.created_by_source_key
+        ):
+            raise ValueError(
+                "You only set one of 'created_by_source_type' and 'created_by_source_key'.  You have to either set both of them (which enables the 'created_by' feature of the column) or you must set neither of them."
+            )
+
+    @property
+    def is_unique(self) -> bool:
+        """Return True/False to denote if this column should always have unique values."""
+        if self._is_unique is None:
+            self._is_unique = any([validator.is_unique for validator in self.validators])
+        return self._is_unique
+
+    @property
+    def is_required(self):
+        """Return True/False to denote if this column should is required."""
+        if self._is_required is None:
+            self._is_required = any([validator.is_required for validator in self.validators])
+        return self._is_required
+
+    def additional_write_columns(self, is_create=False) -> dict[str, Self]:
+        """
+        Return any additional columns that should be included in write operations.
+
+        Some column types, and some validation requirements, necessitate the presence of additional
+        columns in the save operation.  This function adds those in so they can be included in the
+        API call.
+        """
+        additional_write_columns: dict[str, Self] = {}
+        for validator in self.validators:
+            if not isinstance(validator, Validator):
+                continue
+            additional_write_columns = {
+                **additional_write_columns,
+                **validator.additional_write_columns(is_create=is_create),  # type: ignore
+            }
+        return additional_write_columns
+
+    def to_json(self, model: Model) -> dict[str, Any]:
+        """Grabs the column out of the model and converts it into a representation that can be turned into JSON."""
+        return {self.name: self.__get__(model, model.__class__)}
+
+    def input_errors(self, model: Model, data: dict[str, Any]) -> dict[str, Any]:
+        """
+        Check the given dictionary of data for any possible input errors.
+
+        This accepts all the data being saved, and not just the value for this column.  The reason is because
+        some input valdiation flows require more than one piece of data.  For instance, a user may be asked
+        to type a specific piece of input more than once to minimize the chance of typos, or a user may
+        have to provide their password when changing security-related columns.
+
+        This also returns a dictionary, rather than an error message, so that a column can also return an error
+        message for more than one column at a time if needed.
+
+        If there are no input errors then this should simply return an empty dictionary.
+
+        This method calls `self.input_error_for_value` and then also calls all the validators attached to the
+        column so, if you're building your own column and have some specific input validation you need to do,
+        you probably want to extend `input_error_for_value` as that is the one intended checks for a column type.
+
+        Note: this is not called when you directly invoke the `save`/`create` method of a model.  This is only
+        used by handlers when processing user input (e.g. API calls).
+        """
+        if self.name in data and data[self.name]:
+            error = self.input_error_for_value(data[self.name])
+            if error:
+                return {self.name: error}
+
+        for validator in self.validators:
+            if hasattr(validator, "injectable_properties"):
+                validator.injectable_properties(self.di)
+
+            error = validator(model, self.name, data)
+            if error:
+                return {self.name: error}
+
+        return {}
+
+    def check_search_value(
+        self, value: str, operator: str | None = None, relationship_reference: str | None = None
+    ) -> str:
+        """
+        Check if the given value is an allowed value.
+
+        This is called by the search operation in the various API-related handlers to validate a search value.
+
+        Generally, this just defers to self.input_error_for_value, but it is a separate method in case you
+        need to change your input validation logic specifically when checking a search value.
+        """
+        return self.input_error_for_value(value, operator=operator)
+
+    def input_error_for_value(self, value: str, operator: str | None = None) -> str:
+        """
+        Check if the given value is an allowed value.
+
+        This method is intended for checks that are specific to the column type (e.g. this is where an
+        email column checks that the value is an actual email, or a datetime column checks for a valid
+        datetime).  The `input_errors` method does a bit more, so in general it's easier to extend this one.
+
+        This method is passed in the value to check.  It should return a string.  If the data is valid,
+        then return an empty string.  Otherwise return a human-readable error message.
+
+        At times an operator will be passed in.  This is used when the user is searching instead of saving.
+        In this case, the check can vary depending on the operator.  For instance, if it's a wildcard search
+        then an email field only has to verify the type is a string (since the user may have only entered
+        the beginning of an email address), but if it's an exact search then you would expect the value to be
+        an actual email.
+
+        Note: this is not called when you directly invoke the `save`/`create` method of a model.  This is only
+        used by handlers when processing user input (e.g. API calls).
+        """
+        return ""
+
+    def pre_save(self, data: dict[str, Any], model: Model) -> dict[str, Any]:
+        """
+        Make any necessary changes to the data before starting the save process.
+
+        The difference between this and to_backend is that to_backend only affects
+        the data as it is going into the database, while this affects the data that will get persisted
+        in the object as well.  So for instance, for a "created" field, pre_save may fill in the current
+        date with a Python DateTime object when the record is being saved, and then to_backend may
+        turn that into an SQL-compatible date string.
+
+        Note: this is called during the `pre_save` step in the lifecycle of the save process.  See the
+        model class for more details.
+        """
+        if not model and self.created_by_source_type:
+            data[self.name] = self._extract_value_from_source_type()
+        if self.setable:
+            if callable(self.setable):
+                input_output = self.di.build("input_output", cache=True)
+                data[self.name] = self.di.call_function(
+                    self.setable, data=data, model=model, **input_output.get_context_for_callables()
+                )
+            else:
+                data[self.name] = self.setable
+        if not model and self.default and self.name not in data:
+            data[self.name] = self.default
+        if self.on_change_pre_save and model.is_changing(self.name, data):
+            data = self.execute_actions_with_data(self.on_change_pre_save, model, data)
+        return data
+
+    def post_save(self, data: dict[str, Any], model: Model, id: int | str) -> None:
+        """
+        Make any changes needed after persisting data to the backend.
+
+        This lives in the `post_save` hook of the save lifecycle.  See the model class for more details.
+        `data` is the data dictionary being saved.  `model` is obviously the model object that initiated the
+        save.
+
+        This happens after the backend is updated but before the model is updated.  Therefore, You can tell the
+        difference between a create operation and an update operation by checking if the model exists: `if model`.
+        For a create operation, the model will be empty (it evaluates to False).  The opposite is true for an update
+        operation.
+
+        Any return value will be ignored.  If you need to make additional changes in the backend, you
+        have to execute a new save operation.
+        """
+        if self.on_change_post_save and model.is_changing(self.name, data):
+            self.execute_actions_with_data(
+                self.on_change_post_save,
+                model,
+                data,
+                id=id,
+                context="on_change_post_save",
+                require_dict_return_value=False,
+            )
+
+    def save_finished(self, model: Model) -> None:
+        """
+        Make any necessary changes needed after a save has completely finished.
+
+        This is typically used for actions set by the developer.   Column-specific behavior usually lives in
+        `pre_save` or `post_save`.  See the model class for more details about the various lifecycle hooks during
+        a save.
+        """
+        if self.on_change_save_finished and model.was_changed(self.name):
+            self.execute_actions(self.on_change_save_finished, model)
+
+    def pre_delete(self, model):
+        """Make any changes needed to the data before starting the delete process."""
+        pass
+
+    def post_delete(self, model):
+        """Make any changes needed to the data before finishing the delete process."""
+        pass
+
+    def _extract_value_from_source_type(self) -> Any:
+        """For columns with `created_by_source_type` set, this fetches the appropriate value from the request."""
+        input_output = self.di.build("input_output", cache=True)
+        source_type = self.created_by_source_type
+        if source_type == "authorization_data":
+            data = input_output.authorization_data
+        elif source_type == "http_header":
+            data = input_output.request_headers
+        elif source_type == "routing_data":
+            data = input_output.routing_data
+
+        if self.created_by_source_key not in data and self.created_by_source_strict:
+            raise ValueError(
+                f"Column '{self.name}' is configured to load the key named '{self.created_by_source_key}' from "
+                + f"the {self.created_by_source_type}', but this key was not present in the request."
+            )
+
+        return data.get(self.created_by_source_key, "N/A")
+
+    def execute_actions_with_data(
+        self,
+        actions: list[typing.action],
+        model: Model,
+        data: dict[str, Any],
+        id: int | str | None = None,
+        context: str = "on_change_pre_save",
+        require_dict_return_value: bool = True,
+    ) -> dict[str, Any]:
+        """Execute a given set of actions and expects data to be both provided and returned."""
+        input_output = self.di.build("input_output", cache=True)
+        for index, action in enumerate(actions):
+            new_data = self.di.call_function(
+                action,
+                **{
+                    **input_output.get_context_for_callables(),
+                    **{
+                        "model": model,
+                        "data": data,
+                        "id": id,
+                    },
+                },
+            )
+            if not isinstance(new_data, dict):
+                if require_dict_return_value:
+                    raise ValueError(
+                        f"Return error for action #{index + 1} in 'on_change_pre_save' for column '{self.name}' in model '{self.model_class.__name__}': this action must return a dictionary but returned an object of type '{new_data.__class__.__name__}' instead"
+                    )
+                else:
+                    return new_data
+            data = {
+                **data,
+                **new_data,
+            }
+        return data
+
+    def execute_actions(
+        self,
+        actions: list[typing.action],
+        model: Model,
+    ) -> None:
+        """Execute a given set of actions."""
+        input_output = self.di.build("input_output", cache=True)
+        for action in actions:
+            self.di.call_function(action, model=model, **input_output.get_context_for_callables())
+
+    def values_match(self, value_1, value_2):
+        """
+        Compare two values to see if they are the same.
+
+        This is mainly used to compare incoming data with old data to determine if a column has changed.
+
+        Note that these checks shouldn't make any assumptions about whether or not data has gone through the
+        to_backend/from_backend functions.  For instance, a datetime field may find one value has a date
+        that is formatted as a string, and the other as a DateTime object.  Plan appropriately.
+        """
+        return value_1 == value_2
+
+    def add_search(self, model: Model, value: str, operator: str = "", relationship_reference: str = "") -> Model:
+        return model.where(self.condition(operator, value))
+
+    def build_condition(self, value: str, operator: str = "", column_prefix: str = ""):
+        """
+        Build condition for the read (and related) handlers to turn user input into a condition.
+
+        Note that this may look like it is vulnerable to SQLi, but it isn't.  These conditions aren't passed directly
+        into a query.  Rather, they are parsed by the condition parser before being sent into the backend.
+        The condition parser can safely reconstruct the original pieces, and the backend can then use the data
+        safely (and remember, the backend may not be an SQL anyway)
+
+        As a result, this is perfectly safe for any user input, assuming normal system flow.
+
+        That being said, this should probably be replaced by self.condition()...
+        """
+        if not operator:
+            operator = "="
+        if operator.lower() == "like":
+            return f"{column_prefix}{self.name} LIKE '%{value}%'"
+        return f"{column_prefix}{self.name}{operator}{value}"
+
+    def is_allowed_operator(
+        self,
+        operator: str,
+        relationship_reference: str = "",
+    ):
+        """Process user data to decide if the end-user is specifying an allowed operator."""
+        return operator.lower() in self._allowed_search_operators
+
+    def n_plus_one_add_joins(self, model: Model, column_names: list[str] = []) -> Model:
+        """Add any additional joins to solve the N+1 problem."""
+        return model
+
+    def n_plus_one_join_table_alias_prefix(self):
+        """
+        Create a table alias to use with joins for n+1 solutions.
+
+        When joining tables in for n+1 solutions, you can't just do a SELECT * on the new table, because that
+        often results in duplicate column names.  A solution that generally works across the board is to select
+        specific columns from the joined table and alias them, adding a common prefix.  Then, the data from the
+        joined table can be reconstructed automatically by finding all columns with that prefix (and then removing
+        the prefix).  This function returns that prefix for that alias.
+
+        Now, technically this isn't function isn't used at all by the base class, so this definition is fairly
+        pointless.  It isn't marked as an abstract method because most model columns don't need it either.
+        Rather, this function is here mostly for documentation so it's easier to understand how to implement
+        support for n+1 solutions when needed.  See the belongs_to column for a full implementation reference.
+        """
+        return "join_table_" + self.name
+
+    def add_join(self, model: Model) -> Model:
+        return model
+
+    def where_for_request(
+        self,
+        model: Model,
+        routing_data: dict[str, str],
+        authorization_data: dict[str, Any],
+        input_output,
+    ) -> Model:
+        """
+        Create a hook to automatically apply filtering whenever the column makes an appearance in a get/update/list/search handler.
+
+        This hook is called by all the handlers that execute queries, so if your column needs to automatically
+        do some filtering whenever the model shows up in an API endpoint, this is the place for it.
+        """
+        return model
+
+    def name_for_building_condition(self) -> str:
+        if self._config and "name" in self._config:
+            return self.name
+
+        if not self._config or not self._config.get("model_class"):
+            raise ValueError(
+                f"A condition builder was called but the model class isn't set.  This means that the __get__ method for column class {self.__class__.__name__} forgot to set `self.model_class = cls`"
+            )
+
+        for attribute_name in dir(self.model_class):
+            if id(getattr(self.model_class, attribute_name)) != id(self):
+                continue
+            self.name = attribute_name
+            break
+
+        return self.name
+
+    def equals(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if "=" not in self._allowed_search_operators:
+            raise ValueError(f"An 'equals search' is not allowed for '{self.model_class.__name__}.{name}'.")
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, "=", [value])
+
+    def spaceship(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if "<=>" not in self._allowed_search_operators:
+            raise ValueError(f"A 'spaceship' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, "<=>", [value])
+
+    def not_equals(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if "!=" not in self._allowed_search_operators:
+            raise ValueError(f"A 'not equals' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, "!=", [value])
+
+    def less_than_equals(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if "<=" not in self._allowed_search_operators:
+            raise ValueError(f"A 'less than or equals' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, "<=", [value])
+
+    def greater_than_equals(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if ">=" not in self._allowed_search_operators:
+            raise ValueError(
+                f"A 'greater than' or equals search is not allowed for '{self.model_class.__name__}.{name}'."
+            )
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, ">=", [value])
+
+    def less_than(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if "<" not in self._allowed_search_operators:
+            raise ValueError(f"A 'less than' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, "<", [value])
+
+    def greater_than(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if ">" not in self._allowed_search_operators:
+            raise ValueError(f"A 'greater than' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, ">", [value])
+
+    def is_in(self, values) -> Condition:
+        name = self.name_for_building_condition()
+        if "in" not in self._allowed_search_operators:
+            raise ValueError(f"An 'in' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        if not isinstance(values, list):
+            raise TypeError("You must pass a list in to column.is_in")
+        final_values = []
+        for value in values:
+            final_values.append(self.to_backend({name: value}).get(name))
+        return ParsedCondition(name, "in", final_values)  # type: ignore
+
+    def is_not_null(self) -> Condition:
+        name = self.name_for_building_condition()
+        if "is not null" not in self._allowed_search_operators:
+            raise ValueError(f"An 'is not null' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        return ParsedCondition(name, "is not null", [])
+
+    def is_null(self) -> Condition:
+        name = self.name_for_building_condition()
+        if "is null" not in self._allowed_search_operators:
+            raise ValueError(f"An 'is null' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        return ParsedCondition(name, "is null", [])
+
+    def like(self, value) -> Condition:
+        name = self.name_for_building_condition()
+        if "like" not in self._allowed_search_operators:
+            raise ValueError(f"A 'like' search is not allowed for '{self.model_class.__name__}.{name}'.")
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, "like", [value])
+
+    def condition(self, operator: str, value) -> Condition:
+        name = self.name_for_building_condition()
+        operator = operator.lower()
+        if operator not in self._allowed_search_operators:
+            raise ValueError(f"The operator '{operator}' is not allowed for '{self.model_class.__name__}.{name}'.")
+
+        # the search methods are more or less identical except:
+        if operator == "in":
+            return self.is_in(value)
+
+        value = self.to_backend({name: value}).get(name)
+        return ParsedCondition(name, operator, [value])
+
+    def is_allowed_search_operator(self, operator: str, relationship_reference: str = "") -> bool:
+        return operator in self._allowed_search_operators
+
+    def allowed_search_operators(self, relationship_reference: str = ""):
+        return self._allowed_search_operators
+
+    def join_table_alias(self) -> str:
+        raise NotImplementedError("Ooops, I don't support joins")
+
+    def documentation(self, name=None, example=None, value=None) -> list[AutoDocSchema]:
+        return [self.auto_doc_class(name if name is not None else self.name, example=example, value=value)]