fix: resolve pyright errors in processing modules

- add Abstract type annotation and import in metadata.py
- add type annotations for list variables to fix unknown append types
- fix is_doi return type by wrapping with bool() in validation.py

All tests continue to pass and type checking is now clean
across all processing and validation modules.

doi2dataset/api/processors.py

@@ -6,12 +6,15 @@ including license processing and abstract extraction/cleaning.
 """
 
 import re
-from typing import Any
+from typing import TYPE_CHECKING, Any
 
 from rich.console import Console
 
 from ..core.models import Abstract, License
 
+if TYPE_CHECKING:
+    from .client import APIClient
+
 
 class LicenseProcessor:
     """
@@ -68,12 +71,12 @@ class AbstractProcessor:
     # Icons for console output - TODO: should be moved to a constants module
     ICONS = {"info": "ℹ️", "warning": "⚠️", "error": "❌"}
 
-    def __init__(self, api_client, console: Console | None = None):
+    def __init__(self, api_client: "APIClient", console: Console | None = None):
         """
         Initialize with an APIClient instance.
 
         Args:
-            api_client: The API client to use for requests.
+            api_client (APIClient): The API client to use for requests.
             console (Console | None): Rich console instance for output.
         """
         self.api_client = api_client

doi2dataset/processing/metadata.py

@@ -21,7 +21,7 @@ from ..core.metadata_fields import (
     ControlledVocabularyMetadataField,
     PrimitiveMetadataField,
 )
-from ..core.models import Person
+from ..core.models import Abstract, Person
 from ..processing.citation import CitationBuilder
 from ..processing.utils import NameProcessor, PIFinder, SubjectMapper
 
@@ -335,13 +335,13 @@ class MetadataProcessor:
 
         return return_dict
 
-    def _build_description(self, data: dict[str, Any], abstract) -> str:
+    def _build_description(self, data: dict[str, Any], abstract: Abstract) -> str:
         """
         Build the description field by combining a header and the abstract.
 
         Args:
             data (dict[str, Any]): The metadata.
-            abstract: The abstract object.
+            abstract (Abstract): The abstract object.
 
         Returns:
             str: The full description.
@@ -404,7 +404,7 @@ class MetadataProcessor:
             list[Person]: List of matching PIs for use as corresponding authors.
         """
         involved_pis: list[Person] = []
-        authors_in_publication = []
+        authors_in_publication: list[Person] = []
 
         # Build list of authors from publication
         for authorship in data.get("authorships", []):

doi2dataset/processing/utils.py

@@ -223,7 +223,7 @@ class SubjectMapper:
         Returns:
             list[str]: List of mapped controlled vocabulary terms.
         """
-        mapped = []
+        mapped: list[str] = []
         for subject in subjects:
             # Try exact match first
             if subject in cls.CONTROLLED_VOCAB:

doi2dataset/utils/validation.py

@@ -29,7 +29,7 @@ def validate_doi(doi: str) -> bool:
     Returns:
         bool: True if the DOI is valid, False otherwise.
     """
-    return is_doi(doi)
+    return bool(is_doi(doi))
 
 
 def normalize_doi(doi: str) -> str: