fix: relax lxml version constraint und use fixed date parsers list (#184)

* #183 relax lxml version constraint

* formatting corrected for black on python 3.13

* fixed 'mypy -p htmldate' complaint

---------

Co-authored-by: Andrei <andrei@riskfront.ai>

Author: zhemaituk

htmldate/extractors.py

@@ -12,7 +12,6 @@
 
 # coverage for date parsing
 from dateparser import DateDataParser  # type: ignore  # third-party, slow
-from dateparser_data.settings import default_parsers
 
 from dateutil.parser import parse as dateutil_parse
 
@@ -34,9 +33,8 @@
     settings={
         "NORMALIZE": True,  # False may be faster
         "PARSERS": [
-            p
-            for p in default_parsers
-            if p not in ("no-spaces-time", "relative-time", "timestamp")
+            "custom-formats",
+            "absolute-time",
         ],
         "PREFER_DATES_FROM": "past",
         "PREFER_LOCALE_DATE_ORDER": True,

htmldate/utils.py

@@ -46,6 +46,7 @@
 
 class Extractor:
     "Defines a class to store all extraction options."
+
     __slots__ = ["extensive", "format", "max", "min", "original"]
 
     # consider dataclasses for Python 3.7+

pyproject.toml

@@ -53,7 +53,7 @@ dependencies = [
     "dateparser >= 1.1.2",  # 1.1.3+ slower
     # see tests on Github Actions
     "lxml == 4.9.2 ; platform_system == 'Darwin' and python_version <= '3.8'",
-    "lxml >= 5.3.0, < 6 ; platform_system != 'Darwin' or python_version > '3.8'",
+    "lxml >= 5.3.0 ; platform_system != 'Darwin' or python_version > '3.8'",
     "python-dateutil >= 2.9.0.post0",
     "urllib3 >= 1.26, < 3",
 ]