Merge pull request #7 from ahida-development/pydantic2

Breaking change: Update to pydantic v2

CHANGELOG.md

@@ -4,6 +4,11 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
+### Changed
+- Updated to [Pydantic V2](https://docs.pydantic.dev/latest/): the new version has useful features such as multiple aliases for fields
+- In Pydantic V2, url fields are stored as a URL class and cannot be directly used as strings - use `str(model.url_field)` to use them as string.
+### Fixed
+- Updated some schema fields to reflect latest changes to OLS4 - still a bit of a moving target!
 
 ## [0.3.0] - 2023-06-05
 ### Added
@@ -99,4 +104,3 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 [0.1.0]: https://github.com/ahida-development/ols-py/compare/0.0.3...0.1.0
 [0.0.3]: https://github.com/ahida-development/ols-py/compare/0.0.2...0.0.3
 [0.0.2]: https://github.com/ahida-development/ols-py/tree/0.0.2
-

pyproject.toml

@@ -19,6 +19,7 @@ classifiers = [
   "Programming Language :: Python",
   "Programming Language :: Python :: 3",
   "Programming Language :: Python :: 3.10",
+  "Programming Language :: Python :: 3.11",
   "Topic :: Software Development :: Libraries :: Python Modules",
   "Typing :: Typed",
 ]
@@ -30,7 +31,7 @@ packages = [
 [tool.poetry.dependencies]
 python = ">=3.10.1, <4.0"
 requests = ">=2.0, <3.0"
-pydantic = "^1.10.2"
+pydantic = "^2.1.1"
 
 [tool.poetry.group.jupyterlab]
 optional = true

src/ols_py/client.py

@@ -94,7 +94,7 @@ def get_ontologies(
         :param page: Page number of results (starting at 0)
         :param size: Number of results per page (API default is 20)
         """
-        params = schemas.requests.PageParams(page=page, size=size).dict(
+        params = schemas.requests.PageParams(page=page, size=size).model_dump(
             exclude_none=True
         )
         ontology_list = self.get_with_schema(

src/ols_py/ols4_schemas/responses.py

@@ -2,28 +2,32 @@
 
 from typing import Optional
 
-from pydantic import BaseModel, Extra, Field
+from pydantic import AliasChoices, BaseModel, Field
 
 from ..schemas import responses
 from ..schemas.common import EntityType
 
 
 # Structure of search results has changed from OLS v3,
 #   some fields are now list[str] instead of str
-class SearchResultItem(BaseModel, extra=Extra.allow):
-    id: Optional[str]
-    annotations: Optional[list[str]]
-    annotations_trimmed: Optional[list[str]]
-    description: Optional[list[str]]
-    iri: Optional[str]
-    label: Optional[list[str]]
-    obo_id: Optional[list[str]]
-    ontology_name: Optional[str]
-    ontology_prefix: Optional[str]
-    subset: Optional[list[str]]
-    short_form: Optional[list[str]]
-    synonym: Optional[list[str]]
-    type: Optional[EntityType]
+class SearchResultItem(BaseModel, extra="allow"):
+    id: Optional[str] = None
+    annotations: Optional[list[str]] = None
+    annotations_trimmed: Optional[list[str]] = None
+    description: Optional[list[str]] = None
+    iri: Optional[str] = None
+    label: Optional[str] = None
+    obo_id: Optional[str] = None
+    ontology_name: Optional[str] = None
+    ontology_prefix: Optional[str] = None
+    subset: Optional[list[str]] = None
+    short_form: Optional[str] = None
+    # OLS4 has recently switched to "synonyms" for this field,
+    #   where OLS3 has "synonym"
+    synonyms: Optional[list[str]] = Field(
+        default=None, validation_alias=AliasChoices("synonyms", "synonym")
+    )
+    type: Optional[EntityType] = None
 
 
 class SearchResponseResponse(BaseModel):

src/ols_py/schemas/common.py

@@ -2,7 +2,8 @@
 
 from typing import TYPE_CHECKING, Literal
 
-from pydantic import constr
+from pydantic import StringConstraints
+from typing_extensions import Annotated
 
 EntityType = Literal["class", "property", "individual", "ontology"]
 
@@ -11,4 +12,4 @@
 if TYPE_CHECKING:
     AnnotationFieldName = str
 else:
-    AnnotationFieldName = constr(regex=r"^\w+_annotation$")
+    AnnotationFieldName = Annotated[str, StringConstraints(pattern=r"^\w+_annotation$")]

src/ols_py/schemas/requests.py

@@ -2,7 +2,7 @@
 
 from typing import Literal, Optional, TypedDict
 
-from pydantic import BaseModel, NonNegativeInt, PositiveInt, validator
+from pydantic import BaseModel, NonNegativeInt, PositiveInt, field_validator
 
 from .common import AnnotationFieldName, EntityType
 
@@ -13,9 +13,9 @@ class PageParams(BaseModel):
     resources
     """
 
-    size: Optional[PositiveInt]
+    size: Optional[PositiveInt] = None
     """Number of results per page"""
-    page: Optional[NonNegativeInt]
+    page: Optional[NonNegativeInt] = None
     """Which page to fetch (starting at 0)"""
 
 
@@ -35,7 +35,10 @@ class PageParams(BaseModel):
     "ontology_prefix",
     "short_form",
     "subset",
+    # TODO: OLS3 uses synonym, OLS4 uses synonyms. Best option for now
+    #   is to allow both.
     "synonym",
+    "synonyms",
     "type",
 ]
 SearchQueryFields = Literal[
@@ -62,47 +65,48 @@ class SearchParams(BaseModel):
 
     q: str
     """Query to search for"""
-    ontology: Optional[list[str]]
+    ontology: Optional[list[str]] = None
     """Ontologies to search, e.g. `["mondo", "upheno"]`"""
-    type: Optional[EntityType]
+    type: Optional[EntityType] = None
     """Type of term to search for, e.g. "class", "property" """
-    slim: Optional[list[str]]
-    fieldList: Optional[list[SearchReturnFields | AnnotationFieldName]]
+    slim: Optional[list[str]] = None
+    fieldList: Optional[list[SearchReturnFields | AnnotationFieldName]] = None
     """Which fields to return in the results"""
-    queryFields: Optional[list[SearchQueryFields | AnnotationFieldName]]
+    queryFields: Optional[list[SearchQueryFields | AnnotationFieldName]] = None
     """Which fields to search over"""
-    exact: Optional[bool]
-    groupField: Optional[bool]
+    exact: Optional[bool] = None
+    groupField: Optional[bool] = None
     """Group results by unique ID"""
-    obsoletes: Optional[bool]
+    obsoletes: Optional[bool] = None
     """Include obsoleted terms in the results"""
-    local: Optional[bool]
+    local: Optional[bool] = None
     """Only return terms in a defining ontology"""
-    childrenOf: Optional[list[str]]
+    childrenOf: Optional[list[str]] = None
     """Restrict results to children of these terms"""
-    allChildrenOf: Optional[list[str]]
+    allChildrenOf: Optional[list[str]] = None
     """
     Restrict results to children of these terms, plus other
     child-like relations e.g. "part of", "develops from"
     """
-    rows: Optional[int]
+    rows: Optional[int] = None
     """
     Number of results per page
     """
-    start: Optional[int]
+    start: Optional[int] = None
     """
     Index of first result
     """
 
-    @validator(
+    @field_validator(
         "ontology",
         "slim",
         "childrenOf",
         "allChildrenOf",
         "fieldList",
         "queryFields",
-        pre=True,
+        mode="before",
     )
+    @classmethod
     def _single_to_list(cls, v):
         """
         If a single string is passed but we expect

src/ols_py/schemas/responses.py

@@ -3,7 +3,7 @@
 from typing import Any, Optional
 
 import pydantic
-from pydantic import BaseModel, Extra, Field, HttpUrl
+from pydantic import BaseModel, ConfigDict, Field, HttpUrl
 
 from ols_py.schemas.common import EntityType
 
@@ -38,23 +38,21 @@ class Term(BaseModel):
     # Not specifying annotations for now, not sure if these
     #   are fixed
     annotation: dict[str, list[str]]
-    synonyms: Optional[list[str]]
+    synonyms: Optional[list[str]] = None
     ontology_name: str
     ontology_prefix: str
     ontology_iri: pydantic.AnyUrl
     is_obsolete: bool
-    term_replaced_by: Optional[Any]
+    term_replaced_by: Optional[Any] = None
     has_children: bool
     is_root: bool
     short_form: str
     # Higher level terms may not have an obo ID, e.g.
     # terms will be descendants of http://www.w3.org/2002/07/owl#Thing
-    obo_id: Optional[str]
-    in_subset: Optional[Any]
+    obo_id: Optional[str] = None
+    in_subset: Optional[Any] = None
     links: dict[str, Link] = Field(..., alias="_links")
-
-    class Config:
-        extra = "allow"
+    model_config = ConfigDict(extra="allow")
 
 
 class ApiInfoLinks(BaseModel):
@@ -84,9 +82,7 @@ class OntologyItem(BaseModel):
     status: str
     numberOfProperties: int
     numberOfTerms: int
-
-    class Config:
-        extra = "allow"
+    model_config = ConfigDict(extra="allow")
 
 
 class OntologyListEmbedded(BaseModel):
@@ -137,20 +133,20 @@ class TermInDefiningOntology(BaseModel):
     page: PageInfo
 
 
-class SearchResultItem(BaseModel, extra=Extra.allow):
-    id: Optional[str]
-    annotations: Optional[list[str]]
-    annotations_trimmed: Optional[list[str]]
-    description: Optional[list[str]]
-    iri: Optional[str]
-    label: Optional[str]
-    obo_id: Optional[str]
-    ontology_name: Optional[str]
-    ontology_prefix: Optional[str]
-    subset: Optional[list[str]]
-    short_form: Optional[str]
-    synonym: Optional[list[str]]
-    type: Optional[EntityType]
+class SearchResultItem(BaseModel, extra="allow"):
+    id: Optional[str] = None
+    annotations: Optional[list[str]] = None
+    annotations_trimmed: Optional[list[str]] = None
+    description: Optional[list[str]] = None
+    iri: Optional[str] = None
+    label: Optional[str] = None
+    obo_id: Optional[str] = None
+    ontology_name: Optional[str] = None
+    ontology_prefix: Optional[str] = None
+    subset: Optional[list[str]] = None
+    short_form: Optional[str] = None
+    synonym: Optional[list[str]] = None
+    type: Optional[EntityType] = None
 
 
 class SearchResponseResponse(BaseModel):

tests/test_client.py

@@ -84,7 +84,7 @@ def test_get_term(ebi_client):
         ontology_id="go", iri="http://purl.obolibrary.org/obo/GO_0043226"
     )
     assert term.ontology_name == "go"
-    assert term.iri == "http://purl.obolibrary.org/obo/GO_0043226"
+    assert str(term.iri) == "http://purl.obolibrary.org/obo/GO_0043226"
     assert term.label == "organelle"
 
 
@@ -175,11 +175,11 @@ def test_get_term_in_defining_ontology(ebi_client):
     iri = "http://purl.obolibrary.org/obo/MONDO_0018660"
     resp = ebi_client.get_term_in_defining_ontology(iri=iri)
     term = resp.embedded.terms[0]
-    assert term.iri == iri
+    assert str(term.iri) == iri
     assert term.ontology_name == "mondo"
     # Should also allow searching by OBO ID etc. by passing params
     obo_id = "MONDO:0018660"
     resp_from_obo = ebi_client.get_term_in_defining_ontology(params={"obo_id": obo_id})
     term_from_obo = resp_from_obo.embedded.terms[0]
-    assert term_from_obo.iri == iri
+    assert str(term_from_obo.iri) == iri
     assert term_from_obo.ontology_name == "mondo"

tests/test_ols4_client.py

@@ -33,15 +33,15 @@ def test_search_returns_synonyms(ols4_client):
             "ontology": "ncbitaxon",
             "rows": 10,
             "queryFields": ["label"],
-            "fieldList": ["iri", "label", "synonym"],
+            "fieldList": ["iri", "label", "obo_id", "synonyms"],
             # Use exact, we just want to make sure we get bos taurus so
             #   we can check its synonyms
             "exact": True,
             "childrenOf": ["http://purl.obolibrary.org/obo/NCBITaxon_9903"],
         },
     )
     first_result = resp.response.docs[0]
-    assert "cow" in first_result.synonym
+    assert "cow" in first_result.synonyms
 
 
 def test_get_term_in_defining_ontology(ols4_client):
@@ -51,7 +51,7 @@ def test_get_term_in_defining_ontology(ols4_client):
     iri = "http://purl.obolibrary.org/obo/MONDO_0018660"
     resp = ols4_client.get_term_in_defining_ontology(iri=iri)
     term = resp.embedded.terms[0]
-    assert term.iri == iri
+    assert str(term.iri) == iri
     assert term.ontology_name == "mondo"
     # OBO ID search should be working now
     obo_id = "MONDO:0018660"
@@ -64,4 +64,4 @@ def test_get_term_in_defining_ontology(ols4_client):
         params={"short_form": short_form}
     )
     assert resp_from_short_form.page.totalElements == 1
-    assert resp_from_short_form.embedded.terms[0].iri == iri
+    assert str(resp_from_short_form.embedded.terms[0].iri) == iri

tests/test_schemas.py

@@ -56,8 +56,11 @@ def test_search_fields():
     """
     return_fields = set(get_args(ols_py.schemas.requests.SearchReturnFields))
     result_item_fields = set(
-        ols_py.schemas.responses.SearchResultItem.schema()["properties"].keys()
+        ols_py.schemas.responses.SearchResultItem.model_json_schema()[
+            "properties"
+        ].keys()
     )
     # Result items have an extra id field
     assert result_item_fields - return_fields == {"id"}
-    assert return_fields - result_item_fields == set()
+    # TODO: currently have to allow for "synonyms" from OLS4
+    assert return_fields - result_item_fields == {"synonyms"}