fixes for submission of projects

Author: esoteric-ephemera

mp_api/client/contribs/client.py

@@ -16,7 +16,7 @@
 from math import isclose
 from pathlib import Path
 from tempfile import gettempdir
-from typing import TYPE_CHECKING, Literal, cast, overload
+from typing import TYPE_CHECKING, Literal, overload
 from urllib.parse import urlsplit
 
 import orjson
@@ -69,7 +69,7 @@
 
 if TYPE_CHECKING:
     from collections.abc import Generator, Iterable, Sequence
-    from typing import Any
+    from typing import Any, cast
 
     from mp_api.client.contribs.types import (
         AllIdMap,
@@ -861,7 +861,7 @@ def create_project(
                 "references": [{"label": "REF", "url": url}],
             }
         )
-        resp = self.projects.createProject(project=project.model_dump()).result()
+        resp = self.projects.createProject(project=project.to_draft()).result()
         owner = resp.get("owner")
         if owner:
             MPCC_LOGGER.info(f"Project `{name}` created with owner `{owner}`")
@@ -1432,12 +1432,13 @@ def _collect_ids_as_sets(
                 ret[project] = id_sets
 
             project_sets = ret[project]
-            cast(set[str], project_sets["ids"]).add(contrib["id"])
-            cast(set[str], project_sets["identifiers"]).add(contrib["identifier"])
+            if TYPE_CHECKING:
+                cast(set[str], project_sets["ids"]).add(contrib["id"])
+                cast(set[str], project_sets["identifiers"]).add(contrib["identifier"])
 
             if data_id_field:
                 data_value = contrib.get("data", {}).get(data_id_field)
-                if isinstance(data_value, str):
+                if TYPE_CHECKING and isinstance(data_value, str):
                     cast(set[str], project_sets[f"{data_id_field}_set"]).add(data_value)
 
             for component in components:
@@ -1448,7 +1449,9 @@ def _collect_ids_as_sets(
                 if component not in project_sets:
                     project_sets[component] = {"ids": set(), "md5s": set()}
 
-                component_sets = cast(ComponentIdSets, project_sets[component])
+                if TYPE_CHECKING:
+                    component_sets = cast(ComponentIdSets, project_sets[component])
+
                 for item in component_items:
                     if not isinstance(item, dict):
                         continue

mp_api/client/contribs/schemas.py

@@ -98,10 +98,17 @@ def _get_pydantic_from_dataframe(
     """
     columns_renamed = {}
     columns_to_unit = {}
+    character_replacements = str.maketrans(
+        {
+            "^": "**",
+            ":": "",
+            "#": "",
+        }
+    )
     for col in df.columns:
         base_name, unit = _get_unit(col)
         base_name = _to_camel_case(base_name)
-        columns_renamed[col] = base_name
+        columns_renamed[col] = base_name.translate(character_replacements)
         if unit:
             columns_to_unit[col] = unit
 
@@ -179,7 +186,19 @@ def flatten_other(cls, d: dict) -> dict[str, str | None]:
     @field_serializer("other", mode="plain")
     def unflatten_other(self, v: dict[str, str]) -> dict[str, Any]:
         """Unflatten column metadata."""
-        return unflatten_dict(v)
+        return unflatten_dict(v or {})
+
+    def to_draft(self) -> dict[str, Any]:
+        """Strip out fields that cannot be used in creating a project.
+
+        The API forbids including `is_public` and `is_approved` when
+        submitting a project, even if these fields are False.
+        """
+        return {
+            k: v
+            for k, v in self.model_dump().items()
+            if k not in {"is_approved", "is_public"}
+        }
 
 
 class ContribMeta(_DictLikeAccess):
@@ -331,6 +350,14 @@ def from_dataframe(
             for idx, entry in enumerate(sanitized)
         ]
 
+    def to_submission(self) -> dict[str, Any]:
+        """Pop null keys."""
+        return {
+            k: v
+            for k, v in self.model_dump(mode="json").items()
+            if v is not None and k != "id"
+        }
+
 
 class QueryResult(_DictLikeAccess):
     """Result of query_contributions."""