converter from json to TypedDict and BaseModel

pyproject.toml

@@ -50,11 +50,12 @@ dev = [
   "numpydoc",
 
   # For schema generation.
-  "pydantic>=2.6",
+  "pydantic<3",
+  "datamodel-code-generator",
 ]
 
 [project.scripts]
-regenerate-schema = "event_model.documents.generate.__main__:main"
+regenerate-documents = "event_model.generate.__main__:generate"
 
 [project.urls]
 GitHub = "https://github.com/bluesky/event-model"

src/event_model/__init__.py

@@ -39,15 +39,23 @@
 from .documents.event_descriptor import (
     Configuration,
     DataKey,
-    Dtype,
     EventDescriptor,
     Limits,
     LimitsRange,
     PerObjectHint,
 )
 from .documents.event_page import EventPage, PartialEventPage
 from .documents.resource import PartialResource, Resource
-from .documents.run_start import Calculation, Hints, Projection, Projections, RunStart
+from .documents.run_start import (
+    CalculatedEventProjection,
+    Calculation,
+    ConfigurationProjection,
+    Hints,
+    LinkedEventProjection,
+    Projections,
+    RunStart,
+    StaticProjection,
+)
 from .documents.run_stop import RunStop
 from .documents.stream_datum import StreamDatum, StreamRange
 from .documents.stream_resource import StreamResource
@@ -67,7 +75,6 @@
     "PartialEvent",
     "Configuration",
     "DataKey",
-    "Dtype",
     "EventDescriptor",
     "Limits",
     "LimitsRange",
@@ -78,6 +85,10 @@
     "Resource",
     "Calculation",
     "Hints",
+    "LinkedEventProjection",
+    "StaticProjection",
+    "CalculatedEventProjection",
+    "ConfigurationProjection",
     "Projection",
     "Projections",
     "RunStart",
@@ -1821,9 +1832,8 @@ class MismatchedDataKeys(InvalidData):
     DocumentNames.resource: "schemas/resource.json",
     DocumentNames.stream_datum: "schemas/stream_datum.json",
     DocumentNames.stream_resource: "schemas/stream_resource.json",
-    # DEPRECATED:
-    DocumentNames.bulk_events: "schemas/bulk_events.json",
     DocumentNames.bulk_datum: "schemas/bulk_datum.json",
+    DocumentNames.bulk_events: "schemas/bulk_events.json",
 }
 schemas = {}
 for name, filename in SCHEMA_NAMES.items():

src/event_model/basemodels/__init__.py

@@ -0,0 +1,49 @@
+# generated in `event_model/generate`
+
+from typing import Tuple, Type, Union
+
+from .bulk_datum import *  # noqa: F403
+from .bulk_events import *  # noqa: F403
+from .datum import *  # noqa: F403
+from .datum_page import *  # noqa: F403
+from .event import *  # noqa: F403
+from .event_descriptor import *  # noqa: F403
+from .event_page import *  # noqa: F403
+from .partial_resource import *  # noqa: F403
+from .resource import *  # noqa: F403
+from .run_start import *  # noqa: F403
+from .run_stop import *  # noqa: F403
+from .stream_datum import *  # noqa: F403
+from .stream_resource import *  # noqa: F403
+
+BasemodelType = Union[
+    Type[BulkDatum],  # noqa: F405,
+    Type[BulkEvents],  # noqa: F405,
+    Type[Datum],  # noqa: F405,
+    Type[DatumPage],  # noqa: F405,
+    Type[Event],  # noqa: F405,
+    Type[EventDescriptor],  # noqa: F405,
+    Type[EventPage],  # noqa: F405,
+    Type[PartialResource],  # noqa: F405,
+    Type[RunStart],  # noqa: F405,
+    Type[RunStop],  # noqa: F405,
+    Type[StreamDatum],  # noqa: F405,
+    Type[StreamResource],  # noqa: F405,
+    Type[Resource],  # noqa: F405,
+]
+
+ALL_BASEMODELS: Tuple[BasemodelType, ...] = (
+    BulkDatum,  # noqa: F405
+    BulkEvents,  # noqa: F405
+    Datum,  # noqa: F405
+    DatumPage,  # noqa: F405
+    Event,  # noqa: F405
+    EventDescriptor,  # noqa: F405
+    EventPage,  # noqa: F405
+    PartialResource,  # noqa: F405
+    RunStart,  # noqa: F405
+    RunStop,  # noqa: F405
+    StreamDatum,  # noqa: F405
+    StreamResource,  # noqa: F405
+    Resource,  # noqa: F405
+)

src/event_model/basemodels/bulk_datum.py

@@ -0,0 +1,32 @@
+# ruff: noqa
+# type: ignore
+# generated by datamodel-codegen:
+#   filename:  bulk_datum.json
+
+from __future__ import annotations
+
+from typing import Any, Dict, List
+
+from pydantic import BaseModel, ConfigDict
+
+
+class BulkDatum(BaseModel):
+    """
+    Document to reference a quanta of externally-stored data
+    """
+
+    model_config = ConfigDict(
+        extra="forbid",
+    )
+    datum_kwarg_list: List[Dict[str, Any]]
+    """
+    Array of arguments to pass to the Handler to retrieve one quanta of data
+    """
+    resource: str
+    """
+    UID of the Resource to which all these Datum documents belong
+    """
+    datum_ids: List[str]
+    """
+    Globally unique identifiers for each Datum (akin to 'uid' for other Document types), typically formatted as '<resource>/<integer>'
+    """

src/event_model/basemodels/bulk_events.py

@@ -0,0 +1,52 @@
+# ruff: noqa
+# type: ignore
+# generated by datamodel-codegen:
+#   filename:  bulk_events.json
+
+from __future__ import annotations
+
+from typing import Any, Dict, List, Optional
+
+from pydantic import BaseModel, ConfigDict, Field, RootModel
+
+
+class BulkEvent(BaseModel):
+    """
+    Document to record a quanta of collected data
+    """
+
+    model_config = ConfigDict(
+        extra="forbid",
+    )
+    data: Dict[str, Any]
+    """
+    The actual measurement data
+    """
+    timestamps: Dict[str, Any]
+    """
+    The timestamps of the individual measurement data
+    """
+    filled: Optional[Dict[str, Any]] = None
+    """
+    Mapping the keys of externally-stored data to a boolean indicating whether that data has yet been loaded
+    """
+    descriptor: str
+    """
+    UID to point back to Descriptor for this event stream
+    """
+    seq_num: int
+    """
+    Sequence number to identify the location of this Event in the Event stream
+    """
+    time: float
+    """
+    The event time.  This maybe different than the timestamps on each of the data entries
+    """
+    uid: str
+    """
+    Globally unique identifier for this Event
+    """
+
+
+class BulkEvents(RootModel[Dict[str, List[BulkEvent]]]):
+    root: Dict[str, List[BulkEvent]] = Field(..., title="bulk_events")

src/event_model/basemodels/datum.py

@@ -0,0 +1,32 @@
+# ruff: noqa
+# type: ignore
+# generated by datamodel-codegen:
+#   filename:  datum.json
+
+from __future__ import annotations
+
+from typing import Any, Dict
+
+from pydantic import BaseModel, ConfigDict, Field
+
+
+class Datum(BaseModel):
+    """
+    Document to reference a quanta of externally-stored data
+    """
+
+    model_config = ConfigDict(
+        extra="forbid",
+    )
+    datum_id: str = Field(..., title="Datum Id")
+    """
+    Globally unique identifier for this Datum (akin to 'uid' for other Document types), typically formatted as '<resource>/<integer>'
+    """
+    datum_kwargs: Dict[str, Any] = Field(..., title="Datum Kwargs")
+    """
+    Arguments to pass to the Handler to retrieve one quanta of data
+    """
+    resource: str = Field(..., title="Resource")
+    """
+    The UID of the Resource to which this Datum belongs
+    """

src/event_model/basemodels/datum_page.py

@@ -0,0 +1,36 @@
+# ruff: noqa
+# type: ignore
+# generated by datamodel-codegen:
+#   filename:  datum_page.json
+
+from __future__ import annotations
+
+from typing import Dict, List
+
+from pydantic import BaseModel, ConfigDict, Field, RootModel
+
+
+class DataFrameForDatumPage(RootModel[List[str]]):
+    root: List[str] = Field(..., title="DataFrameForDatumPage")
+
+
+class DatumPage(BaseModel):
+    """
+    Page of documents to reference a quanta of externally-stored data
+    """
+
+    model_config = ConfigDict(
+        extra="forbid",
+    )
+    datum_id: DataFrameForDatumPage
+    """
+    Array unique identifiers for each Datum (akin to 'uid' for other Document types), typically formatted as '<resource>/<integer>'
+    """
+    datum_kwargs: Dict[str, List] = Field(..., title="Datum Kwargs")
+    """
+    Array of arguments to pass to the Handler to retrieve one quanta of data
+    """
+    resource: str = Field(..., title="Resource")
+    """
+    The UID of the Resource to which all Datums in the page belong
+    """

src/event_model/basemodels/event.py

@@ -0,0 +1,48 @@
+# ruff: noqa
+# type: ignore
+# generated by datamodel-codegen:
+#   filename:  event.json
+
+from __future__ import annotations
+
+from typing import Any, Dict, Optional, Union
+
+from pydantic import BaseModel, Field
+
+
+class PartialEvent(BaseModel):
+    data: Dict[str, Any] = Field(..., title="data")
+    """
+    The actual measurement data
+    """
+    filled: Optional[Dict[str, Union[bool, str]]] = Field(None, title="filled")
+    """
+    Mapping each of the keys of externally-stored data to the boolean False, indicating that the data has not been loaded, or to foreign keys (moved here from 'data' when the data was loaded)
+    """
+    time: float = Field(..., title="time")
+    """
+    The event time. This maybe different than the timestamps on each of the data entries.
+    """
+    timestamps: Dict[str, Any] = Field(..., title="timestamps")
+    """
+    The timestamps of the individual measurement data
+    """
+
+
+class Event(PartialEvent):
+    """
+    Document to record a quanta of collected data
+    """
+
+    descriptor: str = Field(..., title="descriptor")
+    """
+    UID of the EventDescriptor to which this Event belongs
+    """
+    seq_num: int = Field(..., title="seq_num")
+    """
+    Sequence number to identify the location of this Event in the Event stream
+    """
+    uid: str = Field(..., title="uid")
+    """
+    Globally unique identifier for this Event
+    """