intermediate model to store the result of structure matching

emmet-core/emmet/core/electrode.py

@@ -203,3 +203,55 @@ def from_composition_and_entries(
         d["num_steps"] = d.pop("nsteps", None)
         d["last_updated"] = datetime.utcnow()
         return cls(task_id=task_id, framework=Composition(d["framework_formula"]), **d)
+
+
+class StructureGroupDoc(BaseModel):
+    """
+    Document model for the intermediate structure matching database used to build the insertion electrode documents.
+    """
+
+    task_id: str = Field(
+        None,
+        "The combined task_id of the grouped document is given by the numerically smallest task id followed by '_Li' or whichever working atom is considered the working ion during grouping.",
+    )
+
+    structure_matched: bool = Field(
+        None,
+        description="True if the structures in this group has been matched to each other.  This is False for groups that contain all the left over structures with the same framework.",
+    )
+
+    has_distinct_compositions: bool = Field(
+        None,
+        description="True if multiple working ion fractions are available in the group, which means a voltage step exits.",
+    )
+
+    grouped_task_ids: List[str] = Field(
+        None,
+        description="The ids of the materials that have been grouped by the structure matcher.",
+    )
+
+    entry_data: Dict = Field(
+        None,
+        description="Dictionary keyed by the task_id, contains the 'composition' and 'volume' of each material.",
+    )
+
+    framework_formula: str = Field(
+        None, description="The formula of the host framework."
+    )
+
+    working_ion: WorkingIon = Field(None, description="The working ion")
+
+    chemsys: str = Field(
+        None,
+        description="The chemsys this group belongs to.  Always includes the working ion",
+    )
+
+    last_updated: datetime = Field(
+        None,
+        description="Timestamp for the most recent calculation for this Material document",
+    )
+
+    # Make sure that the datetime field is properly formatted
+    @validator("last_updated", pre=True)
+    def last_updated_dict_ok(cls, v):
+        return MontyDecoder().process_decoded(v)