Add typing and checking it via mypy

.github/workflows/python.yml

@@ -5,7 +5,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [ '3.8', '3.9', '3.10', '3.11', '3.12' ]
+        python-version: [ '3.9', '3.10', '3.11', '3.12' ]
     steps:
     - name: Checkout
       uses: actions/checkout@v4

Makefile

@@ -1,11 +1,14 @@
 format:
 	black --target-version py39 --line-length 132 --exclude tests/testdata/ *.py epidoc/ tests/
 
-check: format-check unittest
+check: format-check type-check unittest
 
 format-check:
 	black --check --target-version py39 --line-length 132 --exclude tests/testdata/ *.py epidoc/ tests/
 
+type-check:
+	mypy epidoc
+
 unittest:
 	python -m unittest discover
 

epidoc/api.py

@@ -8,16 +8,16 @@
 class EpiDoc:
 
     title = None
-    idno = {}
+    idno: dict[str, str] = {}
     material = None
-    origin_dates = []
-    origin_place = {}
-    provenances = {}
-    terms = []
-    languages = {}
+    origin_dates: list[str] = []
+    origin_place: dict[str, str] = {}
+    provenances: dict[str, str] = {}
+    terms: list[str] = []
+    languages: dict[str, str] = {}
     commentary = None
     edition_language = None
-    edition_foreign_languages = {}
+    edition_foreign_languages: dict[str, int] = {}
 
     @classmethod
     def create(

epidoc/body.py

@@ -1,23 +1,25 @@
+from bs4 import Tag
+
 from .normalize import _normalize
 
 
 class _Edition:
     @staticmethod
-    def _edition(body):
+    def _edition(body: Tag):
         return body.find("div", type="edition")  # Note: limit to xml:space="preserve"?
 
     @staticmethod
-    def language(body):
+    def language(body: Tag):
         edition = _Edition._edition(body)
         if edition:
             return _normalize(edition.attrs.get("xml:lang"))
 
     @staticmethod
-    def foreign_languages(body):
+    def foreign_languages(body: Tag) -> dict[str, int]:
         edition = _Edition._edition(body)
         if not edition:
             return {}
-        result = {}
+        result: dict[str, int] = {}
         for elem in edition.find_all("foreign"):
             lang = _normalize(elem.attrs.get("xml:lang"))
             if not lang:

epidoc/header.py

@@ -1,28 +1,32 @@
+from typing import Any
+
+from bs4 import Tag
+
 from .normalize import _normalize, _normalized_get_text, _normalized_attrs
 
 
 class _History:
     @staticmethod
-    def origin_dates(history):
+    def origin_dates(history: Tag):
         result = []
-        for elem in history.origin.findAll("origdate"):
+        for elem in history.origin.findAll("origdate"):  # type: ignore
             date = _normalized_attrs(elem)
             date["text"] = _normalized_get_text(elem)
             result.append(date)
         return result
 
     @staticmethod
-    def origin_place(history):
-        origin_place = history.origin.origplace
+    def origin_place(history: Tag):
+        origin_place = history.origin.origplace  # type: ignore
         if not origin_place:
             return {}
         result = _normalized_attrs(origin_place)
         result["text"] = origin_place.getText().strip()
         return result
 
     @staticmethod
-    def provenances(history):
-        result = {}
+    def provenances(history: Tag) -> dict[str, Any]:
+        result: dict[str, Any] = {}
         for elem in history.findAll("provenance"):
             typ = _normalize(elem.attrs.get("type"))
             if typ is None:
@@ -31,7 +35,7 @@ def provenances(history):
         return result
 
     @staticmethod
-    def _provenance(provenance):
+    def _provenance(provenance: Tag):
         result = []
         # Note: For some it's provenance.p.placename
         for elem in provenance.findAll("placename"):
@@ -45,20 +49,20 @@ def _provenance(provenance):
 
 class _ProfileDesc:
     @staticmethod
-    def keyword_terms(profile_desc):
-        result = []
+    def keyword_terms(profile_desc: Tag) -> list[dict[str, Any]]:
+        result: list[dict[str, Any]] = []
         textclass = profile_desc.textclass
         if textclass is None:
             return result
-        for elem in textclass.keywords.findAll("term"):
+        for elem in textclass.keywords.findAll("term"):  # type: ignore
             term = _normalized_attrs(elem)
             term["text"] = _normalized_get_text(elem)
             result.append(term)
         return result
 
     @staticmethod
-    def lang_usage(profile_desc):
-        result = {}
+    def lang_usage(profile_desc: Tag) -> dict[str, str]:
+        result: dict[str, str] = {}
         lang_usage = profile_desc.langusage
         if lang_usage is None:
             return result

epidoc/normalize.py

@@ -1,6 +1,13 @@
-def _normalize(v):
+from typing import TypeVar, Any
+
+from bs4 import Tag
+
+T = TypeVar("T")
+
+
+def _normalize(v: T) -> T:
     if isinstance(v, str):
-        return v.lower().strip()
+        return v.lower().strip()  # type: ignore
     return v
 
 
@@ -11,7 +18,7 @@ def _normalized_get_text(raw):
     return parsed if parsed else None
 
 
-def _normalized_attrs(raw):
+def _normalized_attrs(raw: Tag) -> dict[str, Any]:
     parsed = {}
     for name, value in raw.attrs.items():
         parsed[_normalize(name)] = _normalize(value)

requirements.txt

@@ -1,5 +1,7 @@
-beautifulsoup4==4.10.0
+beautifulsoup4==4.12.3
 black==24.3.0  # Code formatter
 lxml==5.2.1
+mypy==1.10.0  # Type checker
 setuptools==69.2.0  # Only required for publish it as a package
+types-beautifulsoup4==4.12.0.20240229  # Stubs for beautifulsoup4
 wheel==0.43.0  # Only required for publish it as a package

setup.py

@@ -4,8 +4,8 @@
     long_description = f.read()
 
 requires = [
-    "beautifulsoup4>=4.9.0,<5",
-    "lxml>=4.5,<5",
+    "beautifulsoup4>=4.12,<5",
+    "lxml>=5.2,<6",
 ]
 
 setuptools.setup(
@@ -20,7 +20,7 @@
     long_description_content_type="text/markdown",
     name="epidoc",
     packages=["epidoc"],
-    python_requires=">=3.6",
+    python_requires=">=3.9",
     url="https://github.com/Xennis/epidoc-parser",
     version="0.0.1",
 )