emmet-api 0.87.0.dev1__tar.gz → 0.87.0.dev2__tar.gz

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: emmet-api
-Version: 0.87.0.dev1
+Version: 0.87.0.dev2
 Summary: Emmet API Server
 Author-email: The Materials Project <feedback@materialsproject.org>
 License: Modified BSD

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/emmet/api/core/documentation.py

@@ -133,7 +133,7 @@ tags_meta = [
             methodology discussed by Zimmerman *et al.* in [10.3389/fmats.2017.00034](https://doi.org/10.3389/fmats.2017.00034) \
             and [10.1039/C9RA07755C](https://doi.org/10.1039/C9RA07755C). \
             Note that similarity data has not been calculated for all materials in the Materials \
-            Project database. See the `imilarityDoc` schema for a full list of fields returned by this route.",
+            Project database. See the `SimilarityDoc` schema for a full list of fields returned by this route.",
     },
     {
         "name": "Materials XAS",

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/emmet/api/routes/materials/materials/query_operators.py

@@ -1,5 +1,5 @@
 from itertools import permutations
-from typing import Literal
+from typing import Literal, Any
 
 from fastapi import Body, HTTPException, Query
 from emmet.api.query_operator import QueryOperator
@@ -15,6 +15,9 @@ from emmet.api.routes.materials.materials.utils import (
 )
 from emmet.core.symmetry import CrystalSystem
 from emmet.core.vasp.calc_types import RunType
+from emmet.core.vasp.material import BlessedCalcs
+
+BLESSED_CALC_RUN_TYPES = sorted(BlessedCalcs.model_fields)
 
 
 class FormulaQuery(QueryOperator):
@@ -212,7 +215,7 @@ class BlessedCalcsQuery(QueryOperator):
 
     def query(
         self,
-        run_type: RunType = Query(
+        run_type: Literal[*BLESSED_CALC_RUN_TYPES] | RunType = Query(  # type: ignore[valid-type]
             ..., description="Calculation run type of blessed task data"
         ),
         energy_min: float | None = Query(
@@ -222,26 +225,50 @@ class BlessedCalcsQuery(QueryOperator):
             None, description="Maximum total uncorrected DFT energy in eV/atom"
         ),
     ) -> STORE_PARAMS:
-        crit = {f"entries.{run_type}.energy": {}}  # type: dict
 
-        if energy_min:
-            crit[f"entries.{run_type}.energy"].update({"$gte": energy_min})
+        parsed_run_type: str | None = None
+        if isinstance(run_type, RunType):
+            aliases = {
+                RunType.PBE: "GGA",
+                RunType.PBE_U: "GGA_U",
+                RunType.GGA: "GGA",
+                RunType.GGA_U: "GGA_U",
+                RunType.SCAN: "SCAN",
+                RunType.r2SCAN: "R2SCAN",
+                RunType.HSE06: "HSE",
+            }
+            parsed_run_type = aliases.get(run_type)
+
+        elif run_type in BLESSED_CALC_RUN_TYPES:
+            parsed_run_type = run_type
+
+        if parsed_run_type is None:
+            raise ValueError(
+                f"Unsupported {run_type=}, please choose one of "
+                f"{', '.join(BLESSED_CALC_RUN_TYPES)}"
+            )
+
+        crit: dict[str, Any] = {f"entries.{parsed_run_type}.energy": {}}
+
+        if energy_min is not None:
+            crit[f"entries.{parsed_run_type}.energy"].update({"$gte": energy_min})
 
-        if energy_max:
-            crit[f"entries.{run_type}.energy"].update({"$lte": energy_max})
+        if energy_max is not None:
+            crit[f"entries.{parsed_run_type}.energy"].update({"$lte": energy_max})
 
-        if not crit[f"entries.{run_type}.energy"]:
-            return {"criteria": {f"entries.{run_type}": {"$exists": True}}}
+        if not crit[f"entries.{parsed_run_type}.energy"]:
+            return {"criteria": {f"entries.{parsed_run_type}": {"$ne": None}}}
 
         return {"criteria": crit}
 
     def post_process(self, docs, query):
+        """Return only the blessed entry."""
         run_type = list(query["criteria"].keys())[0].split(".")[1]
 
         return_data = [
             {
                 "material_id": doc["material_id"],
-                "blessed_entry": doc["entries"][run_type],
+                "entries": {run_type: doc["entries"][run_type]},
             }
             for doc in docs
         ]

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/emmet/api/routes/materials/similarity/query_operators.py

@@ -2,7 +2,7 @@
 
 from fastapi import HTTPException, Query
 
-from emmet.core.similarity import SimilarityMethod
+from emmet.core.similarity import SimilarityMethod, _vector_from_hex_and_norm
 
 from emmet.api.query_operator import QueryOperator
 from emmet.api.utils import STORE_PARAMS
@@ -16,15 +16,27 @@ SIM_METHOD_TO_FEAT_VEC_LENGTH = {
 class SimilarityFeatureVectorQuery(QueryOperator):
     """Generate a feature-vector-based query.
 
-    TODO: Add an `embedding` kwarg to select between
-    multiple embedding methods used to gauge
-    similarity (e.g., ML-based metrics.)
+    The feature_vector_hex field is limited to 3_000 characters
+    to prevent malicious submission of overly-long string data.
+
+    In the worst case, a random unit vector of floats with
+    length 130 has roughly a hex string length of 2,100 characters.
+
+    Note that we pass unit vectors through only.
+    Passing vectors with non-unit norm would increase the lenght of
+    `feature_vector_hex` uncontrollably.
     """
 
     def query(
         self,
-        feature_vector: list[float] = Query(
-            ..., description="A row vector of floats representing a structure."
+        feature_vector_hex: str = Query(
+            ...,
+            description="A compressed, hex representation of a row unit vector of floats.",
+            max_length=3_000,
+        ),
+        feature_vector_norm: float = Query(
+            ...,
+            description="The norm of the feature vector",
         ),
         method: str | SimilarityMethod | None = Query(
             None,
@@ -37,6 +49,9 @@ class SimilarityFeatureVectorQuery(QueryOperator):
     ) -> STORE_PARAMS:
         """Identify similar materials."""
 
+        feature_vector = _vector_from_hex_and_norm(
+            feature_vector_hex, feature_vector_norm
+        )
         if method is None:
             try:
                 method = next(
@@ -65,8 +80,10 @@ class SimilarityFeatureVectorQuery(QueryOperator):
         ):
             raise HTTPException(
                 status_code=400,
-                detail=f"Invalid feature vector for method {method.value}: "
-                f"should be a list of {ref_fv_len} floats.",
+                detail=(
+                    f"Invalid feature vector for method {method.value}: "  # type: ignore[union-attr]
+                    f"should be a list of {ref_fv_len} floats.",
+                ),
             )
 
         index_name = "similarity_feature_vector"
@@ -74,7 +91,7 @@ class SimilarityFeatureVectorQuery(QueryOperator):
         # and I was not forward thinking in naming it.
         # TODO: homogenize once we have other data built out
         if method != SimilarityMethod.CRYSTALNN:
-            index_name += f"_{method.value.lower()}"
+            index_name += f"_{method.value.lower()}"  # type: ignore[union-attr]
 
         pipeline = [
             {
@@ -100,7 +117,7 @@ class SimilarityFeatureVectorQuery(QueryOperator):
         }
 
     def post_process(self, docs, query):
-        self.total_doc = docs[0]["meta"]["count"]["total"]
+        self.total_doc = len(docs)
         return docs
 
     def meta(self):

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/emmet/api/routes/materials/summary/resources.py

@@ -84,6 +84,7 @@ def summary_resource(summary_store):
 
 
 def summary_stats_resource(summary_store):
+    """NOTE: Currently unused, API endpoint no longer available."""
     resource = AggregationResource(
         summary_store,
         SummaryStats,

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/emmet_api.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: emmet-api
-Version: 0.87.0.dev1
+Version: 0.87.0.dev2
 Summary: Emmet API Server
 Author-email: The Materials Project <feedback@materialsproject.org>
 License: Modified BSD

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/material_resources.py

@@ -156,9 +156,17 @@ from emmet.api.routes.materials.eos.resources import eos_resource
 materials_resources.extend([eos_resource(eos_store)])
 
 # Similarity
-from emmet.api.routes.materials.similarity.resources import similarity_resource
+from emmet.api.routes.materials.similarity.resources import (
+    similarity_resource,
+    similarity_feature_vector_resource,
+)
 
-materials_resources.extend([similarity_resource(similarity_store)])
+materials_resources.extend(
+    [
+        similarity_resource(similarity_store),
+        similarity_feature_vector_resource(similarity_store),
+    ]
+)
 
 # XAS
 from emmet.api.routes.materials.xas.resources import xas_resource
@@ -242,12 +250,9 @@ materials_resources.extend([provenance_resource(provenance_store)])
 # Summary
 from emmet.api.routes.materials.summary.resources import (
     summary_resource,
-    summary_stats_resource,
 )
 
-materials_resources.extend(
-    [summary_stats_resource(summary_store), summary_resource(summary_store)]
-)
+materials_resources.extend([summary_resource(summary_store)])
 
 # Electronic Structure
 from emmet.api.routes.materials.electronic_structure.resources import (

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/requirements/deployment.txt

@@ -40,7 +40,7 @@ ddtrace==4.0.0
     # via emmet-api (emmet/emmet-api/pyproject.toml)
 dnspython==2.8.0
     # via pymongo
-emmet-core==0.86.1
+emmet-core==0.86.2rc0
     # via emmet-api (emmet/emmet-api/pyproject.toml)
 envier==0.6.1
     # via ddtrace

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/materials/materials/test_query_operators.py

@@ -4,6 +4,7 @@ from pymatgen.core.structure import Structure
 
 from emmet.api.core.settings import MAPISettings
 from emmet.api.routes.materials.materials.query_operators import (
+    BlessedCalcsQuery,
     ChemsysQuery,
     DeprecationQuery,
     ElementsQuery,
@@ -15,6 +16,55 @@ from emmet.api.routes.materials.materials.query_operators import (
     SymmetryQuery,
 )
 from emmet.core.symmetry import CrystalSystem
+from emmet.core.vasp.calc_types import RunType
+
+
+def test_blessed_calcs_query():
+
+    op = BlessedCalcsQuery()
+    for mapped, possible_user_input in {
+        "GGA": ["GGA", RunType.GGA, RunType.PBE],
+        "GGA_U": ["GGA_U", RunType.PBE_U],
+        "R2SCAN": ["R2SCAN", RunType.r2SCAN],
+        "SCAN": ["SCAN", RunType.SCAN],
+        "HSE": ["HSE", RunType.HSE06],
+    }.items():
+
+        assert all(
+            op.query(run_type=user_inp, energy_min=None, energy_max=None)
+            == {"criteria": {f"entries.{mapped}": {"$ne": None}}}
+            for user_inp in possible_user_input
+        )
+
+        assert all(
+            op.query(
+                run_type=user_inp,
+                energy_min=0.0,
+                energy_max=None,
+            )
+            == {"criteria": {f"entries.{mapped}.energy": {"$gte": 0.0}}}
+            for user_inp in possible_user_input
+        )
+
+        assert all(
+            op.query(
+                run_type=user_inp,
+                energy_min=None,
+                energy_max=5.0,
+            )
+            == {"criteria": {f"entries.{mapped}.energy": {"$lte": 5.0}}}
+            for user_inp in possible_user_input
+        )
+
+        assert all(
+            op.query(
+                run_type=user_inp,
+                energy_min=0.0,
+                energy_max=5.0,
+            )
+            == {"criteria": {f"entries.{mapped}.energy": {"$gte": 0.0, "$lte": 5.0}}}
+            for user_inp in possible_user_input
+        )
 
 
 def test_formula_query():

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/materials/similarity/test_query_operators.py

@@ -1,13 +1,17 @@
 """Test crystalline similarity vector search."""
 
+import numpy as np
 import pytest
 
-from emmet.core.similarity import SimilarityMethod
+from emmet.core.similarity import (
+    SimilarityMethod,
+    _vector_to_hex_and_norm,
+    _vector_from_hex_and_norm,
+)
 from emmet.api.routes.materials.similarity.query_operators import (
     SimilarityFeatureVectorQuery,
     SIM_METHOD_TO_FEAT_VEC_LENGTH,
 )
-import numpy as np
 
 
 @pytest.mark.parametrize(
@@ -19,6 +23,8 @@ def test_similarity_structure_search(test_dir, method):
     fv = np.random.rand(
         SIM_METHOD_TO_FEAT_VEC_LENGTH[SimilarityMethod.CRYSTALNN]
     ).tolist()
+    fv_hex, fv_norm = _vector_to_hex_and_norm(fv)
+
     limit = 10
     q = {
         "pipeline": [
@@ -26,7 +32,7 @@ def test_similarity_structure_search(test_dir, method):
                 "$vectorSearch": {
                     "index": "similarity_feature_vector",
                     "path": "feature_vector",
-                    "queryVector": fv,
+                    "queryVector": _vector_from_hex_and_norm(fv_hex, fv_norm),
                     "numCandidates": limit,
                     "limit": limit,
                 }
@@ -42,9 +48,14 @@ def test_similarity_structure_search(test_dir, method):
         ]
     }
 
-    manual_q = op.query(feature_vector=fv, method=method, _limit=limit)
+    manual_q = op.query(
+        feature_vector_hex=fv_hex,
+        feature_vector_norm=fv_norm,
+        method=method,
+        _limit=limit,
+    )
     assert manual_q == q
 
     doc = [{"meta": {"count": {"total": limit}}}]
     assert op.post_process(doc, q) == doc
-    assert op.meta() == {"total_doc": limit}
+    assert "total_doc" in op.meta()

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/materials/synthesis/test_adaptor.py

@@ -19,7 +19,7 @@ def test_string2comp():
 
 def test_convert_recipe():
     with zopen(
-        os.path.join(MAPISettings().TEST_FILES, "synth_doc_adaptor.json.gz")
+        os.path.join(MAPISettings().TEST_FILES, "synth_doc_adaptor.json.gz"), "rt"
     ) as file:
         synth_doc = load(file)
 

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/materials/synthesis/test_adaptor_synpro.py

@@ -112,7 +112,8 @@ def test_get_material_formula():
 
 def test_convert_one():
     with zopen(
-        os.path.join(MAPISettings().TEST_FILES, "synth_doc_adaptor_synpro.json.gz")
+        os.path.join(MAPISettings().TEST_FILES, "synth_doc_adaptor_synpro.json.gz"),
+        "rt",
     ) as file:
         synth_doc = load(file)
 

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/materials/synthesis/test_utils.py

@@ -20,7 +20,9 @@ def test_make_ellipsis():
 
 
 def test_mask_paragraphs():
-    with zopen(os.path.join(MAPISettings().TEST_FILES, "synth_doc.json.gz")) as file:
+    with zopen(
+        os.path.join(MAPISettings().TEST_FILES, "synth_doc.json.gz"), "rt"
+    ) as file:
         synth_doc = load(file)
 
     doc = SynthesisSearchResultModel(**synth_doc)
@@ -30,7 +32,9 @@ def test_mask_paragraphs():
 
 
 def test_mask_highlights():
-    with zopen(os.path.join(MAPISettings().TEST_FILES, "synth_doc.json.gz")) as file:
+    with zopen(
+        os.path.join(MAPISettings().TEST_FILES, "synth_doc.json.gz"), "rt"
+    ) as file:
         synth_doc = load(file)
 
     doc = SynthesisSearchResultModel(**synth_doc)

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/materials/tasks/test_query_operators.py

@@ -30,7 +30,7 @@ def test_entries_query():
     assert op.query(task_ids=" mp-149, mp-13") == q
 
     with zopen(
-        os.path.join(MAPISettings().TEST_FILES, "tasks_Li_Fe_V.json.gz")
+        os.path.join(MAPISettings().TEST_FILES, "tasks_Li_Fe_V.json.gz"), "rt"
     ) as file:
         tasks = json.load(file)
     docs = op.post_process(tasks, q)
@@ -45,7 +45,7 @@ def test_trajectory_query():
     assert op.query(task_ids=" mp-149, mp-13") == q
 
     with zopen(
-        os.path.join(MAPISettings().TEST_FILES, "tasks_Li_Fe_V.json.gz")
+        os.path.join(MAPISettings().TEST_FILES, "tasks_Li_Fe_V.json.gz"), "rt"
     ) as file:
         tasks = json.load(file)
     docs = op.post_process(tasks, q)

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/materials/tasks/test_utils.py

@@ -11,7 +11,8 @@ from emmet.api.routes.materials.tasks.utils import (
 
 def test_calcs_reversed_to_trajectory():
     with zopen(
-        os.path.join(MAPISettings().TEST_FILES, "calcs_reversed_mp_1031016.json.gz")
+        os.path.join(MAPISettings().TEST_FILES, "calcs_reversed_mp_1031016.json.gz"),
+        "rt",
     ) as file:
         calcs_reversed = load(file)
         trajectories = calcs_reversed_to_trajectory(calcs_reversed)
@@ -27,7 +28,9 @@ def test_calcs_reversed_to_trajectory():
 
 
 def test_task_to_entry():
-    with zopen(os.path.join(MAPISettings().TEST_FILES, "test_task.json.gz")) as file:
+    with zopen(
+        os.path.join(MAPISettings().TEST_FILES, "test_task.json.gz"), "rt"
+    ) as file:
         doc = load(file)
 
     entry_dict = task_to_entry(doc)

{emmet_api-0.87.0.dev1 → emmet_api-0.87.0.dev2}/tests/molecules/tasks/test_utils.py

@@ -8,7 +8,8 @@ from emmet.api.routes.molecules.tasks.utils import calcs_reversed_to_trajectory
 
 def test_calcs_reversed_to_trajectory():
     with zopen(
-        os.path.join(MAPISettings().TEST_FILES, "calcs_reversed_mpcule_36097.json.gz")
+        os.path.join(MAPISettings().TEST_FILES, "calcs_reversed_mpcule_36097.json.gz"),
+        "rt",
     ) as file:
         calcs_reversed = load(file)
         trajectories = calcs_reversed_to_trajectory(calcs_reversed)