WIP

blurry/schema_validation.py

@@ -1,48 +1,167 @@
 import importlib
 from collections.abc import MutableMapping
+from datetime import datetime
 from pathlib import Path
+from typing import Literal
 
-from pydantic.v1 import ValidationError
+from pydantic import BaseModel, Field, ValidationError
 from rich.console import Console
 
 from blurry.settings import SETTINGS
 
 
+Boolean = Literal["True"] | Literal["False"]
+
+
+class Thing(BaseModel):
+    additionalType: str | None = None
+    alternateName: str | None = None
+    description: str | None = None
+    image: "ImageObject | None" = None
+    name: str | None = None
+
+
+class Person(Thing):
+    type_: str = Field(default="Person", alias="@type", frozen=True)
+    additionalName: str | None = None
+    affiliation: "Organization | None" = None
+    email: str | None = None
+    familyName: str | None = None
+    givenName: str | None = None
+    honorificPrefix: str | None = None
+    honorificSuffix: str | None = None
+    telephone: str | None = None
+    worksFor: "Organization | None" = None
+
+
+class Organization(Thing):
+    type_: str = Field(default="Organization", alias="@type", frozen=True)
+
+
+class CreativeWork(Thing):
+    type_: str = Field(default="CreativeWork", alias="@type", frozen=True)
+    abstract: str | None = None
+    author: Person | Organization | None = None
+    creator: Person | Organization | None = None
+    dateCreated: datetime | None = None
+    dateModified: datetime | None = None
+    datePublished: datetime | None = None
+    headline: str | None = None
+
+
+class MediaObject(CreativeWork):
+    type_: str = Field(default="MediaObject", alias="@type", frozen=True)
+    associatedArticle: "NewsArticle | None" = None
+    bitrate: str | None = None
+    contentSize: str | None = None
+    contentUrl: str | None = None
+    url: str | None = None
+    embedUrl: str | None = None
+
+
+class ImageObject(MediaObject):
+    type_: str = Field(default="ImageObject", alias="@type", frozen=True)
+    caption: MediaObject | str | None = None
+    embeddedTextCaption: str | None = None
+    # exifData: PropertyValue | Text | None
+    representativeOfPage: Boolean | None = None
+
+
+class WebSite(CreativeWork):
+    type_: str = Field(default="WebSite", alias="@type", frozen=True)
+    issn: str | None = None
+
+
+class WebPage(CreativeWork):
+    type_: str = Field(default="WebPage", alias="@type", frozen=True)
+    # breadcrumb: BreadcrumbList | Text | None
+    lastReviewed: datetime | None = None
+    # mainContentOfPage: WebPageElement | None
+    primaryImageOfPage: ImageObject | None = None
+    relatedLink: str | None = None
+    reviewedBy: Organization | Person | None = None
+    significantLink: str | None = None
+    # speakable: SpeakableSpecification | None
+    # specialty: Specialty | None
+
+
+class Article(CreativeWork):
+    type_: str = Field(default="Article", alias="@type", frozen=True)
+    author: Person | Organization  # pyright: ignore to narrow the type
+    datePublished: datetime  # pyright: ignore to narrow the type
+    headline: str  # pyright: ignore to narrow the type
+    image: ImageObject | str | list[ImageObject | str]  # pyright: ignore to permit Google's recommendation
+
+
+class NewsArticle(Article):
+    type_: str = Field(default="NewsArticle", alias="@type", frozen=True)
+    dateline: str | None = None
+    printColumn: str | None = None
+    printEdition: str | None = None
+    printPage: str | None = None
+    printSection: str | None = None
+
+
+class SocialMediaPosting(Article):
+    type_: str = Field(default="SocialMediaPosting", alias="@type", frozen=True)
+    sharedContent: CreativeWork | None = None
+
+
+class BlogPosting(SocialMediaPosting):
+    type_: str = Field(default="BlogPosting", alias="@type", frozen=True)
+
+
+# class Review(BaseModel):
+#     author: Person | Organization
+#     itemReviewed: (
+#         Book
+#         | Course
+#         | CreativeWorkSeason
+#         | CreativeWorkSeries
+#         | Episode
+#         | Event
+#         | Game
+#         | HowTo
+#         | LocalBusiness
+#         | MediaObject
+#         | Movie
+#         | MusicPlaylist
+#         | MusicRecording
+#         | Organization
+#         | Product
+#         | Recipe
+#         | SoftwareApplication
+#     )
+
+
 def validate_front_matter_as_schema(
     path: Path, schema_variables: MutableMapping, console: Console
 ):
     """
-    Validates schema data using pydantic_schemaorg, disallowing extra fields
+    Validates schema data using partial Schema.org types based on Google's support for them:
+    https://developers.google.com/search/docs/appearance/structured-data/search-gallery
     """
-    schema_type = schema_variables["@type"]
+    schematype = schema_variables["@type"]
 
-    if mapped_schema_type := SETTINGS["TEMPLATE_SCHEMA_TYPES"].get(schema_type):
-        schema_type = mapped_schema_type
+    if mapped_schematype_ := SETTINGS["TEMPLATE_SCHEMA_TYPES"].get(schematype):
+        schematype = mapped_schematype_
 
-    # Import pydantic_schemaorg model
+    # Get the schema model class from this module
     try:
-        pydantic_schemaorg_model_module = importlib.import_module(
-            f"pydantic2_schemaorg.{schema_type}"
-        )
+        module = importlib.import_module("blurry.schema_validation")
     except ModuleNotFoundError:
-        console.print(
-            f"{path}: Could not find Schema type for {schema_type}. Skipping."
-        )
+        console.print(f"{path}: Could not import module.")
         return
 
-    SchemaModel = getattr(pydantic_schemaorg_model_module, schema_type)
-
-    # Create new Pydantic model that forbids extra fields
-    class SchemaModelWithoutExtraFields(SchemaModel, extra="forbid"):  # type: ignore
-        pass
+    SchemaModel = getattr(module, schematype)
 
     # Validate model and print errors
     try:
-        SchemaModelWithoutExtraFields(**schema_variables)
+        SchemaModel(**schema_variables)
     except ValidationError as e:
         for error in e.errors():
             msg = error["msg"]
             loc = error["loc"]
-            console.print(
-                f"{path}: {schema_type} schema validation error: {msg}: {loc}"
-            )
+            if len(loc) == 1:
+                loc = loc[0]
+            console.print(f"{path}: {schematype} schema validation error: {msg}: {loc}")

pyproject.toml

@@ -34,10 +34,10 @@ rich = "^13.9.4"
 selectolax = "^0.3.27"
 typer = "^0.15.1"
 htmlmin2 = "^0.1.13"
-pydantic2-schemaorg = "0.3.0"
 dpath = "^2.2.0"
 jinja2-simple-tags = "^0.6.1"
 jinjax = "^0.48"
+pydantic = "^2.10.5"
 
 [tool.poetry.scripts]
 blurry = 'blurry:main'

tests/test_schema_validation.py

@@ -48,6 +48,30 @@ def test_validate_front_matter_as_schema_with_extra_value():
     test_console = Console()
     test_console.print = MagicMock()
     validate_front_matter_as_schema(path, front_matter_with_extra_value, test_console)
+    assert not test_console.print.called
+
+
+MARKDOWN_WITH_WRONG_TYPE_IN_TOML_FRONT_MATTER = """
++++
+"@type" = "WebPage"
+name = "Introduction"
+abstract = "A Python-powered static site generator with a focus on page speed and SEO."
+datePublished = 2023-04-09
+dateModified = true
++++
+
+# Blurry: A Python-powered static site generator
+""".strip()
+
+
+def test_validate_front_matter_as_schema_with_wrong_type():
+    _, front_matter_with_extra_value = get_data(
+        MARKDOWN_WITH_WRONG_TYPE_IN_TOML_FRONT_MATTER
+    )
+    path = Path("pages/intro.md")
+    test_console = Console()
+    test_console.print = MagicMock()
+    validate_front_matter_as_schema(path, front_matter_with_extra_value, test_console)
     test_console.print.assert_called_with(
-        "pages/intro.md: WebPage schema validation error: extra fields not permitted: ('extra_value',)"
+        "pages/intro.md: WebPage schema validation error: Input should be a valid datetime: dateModified"
     )