implement design

pyproject.toml

@@ -36,7 +36,8 @@ dependencies = [
     "tqdm>=4.66, <5",
     "fastwarc>=0.14, <1",
     "chardet>=5.2, <6",
-    "dill>=0.3, <1"
+    "dill>=0.3, <1",
+    "deprecated>=1.2.14, <2"
 ]
 
 [project.urls]
@@ -53,6 +54,7 @@ dev = [
     "types-python-dateutil>=2.8, <3",
     "types-requests>=2.28, <3",
     "types-colorama>=0.4, <1",
+    "types-deprecated>=1.2.9, <2"
 ]
 
 [tool.mypy]

src/fundus/parser/base_parser.py

@@ -24,6 +24,7 @@
 
 import lxml.html
 import more_itertools
+from deprecated import deprecated
 from lxml.etree import XPath
 
 from fundus.logging import create_logger
@@ -289,7 +290,14 @@ def __init__(self):
             mapping[validation_date] = _ParserCache(versioned_parser)
         self._parser_mapping = mapping
 
+    @deprecated(
+        version="0.4.1",
+        reason=f"Calling <class: ParserProxy> is legacy behaviour. Use <method: parse> of <class: ParserProxy> instead",
+    )
     def __call__(self, crawl_date: Optional[Union[datetime, date]] = None) -> BaseParser:
+        return self._get_parser(crawl_date)
+
+    def _get_parser(self, crawl_date: Optional[Union[datetime, date]] = None) -> BaseParser:
         if crawl_date is None:
             return self._get_latest_cache()()
 
@@ -343,3 +351,11 @@ def _get_latest_cache(self) -> _ParserCache:
     @property
     def latest_version(self) -> Type[BaseParser]:
         return self._get_latest_cache().factory
+
+    def parse(
+        self,
+        html: str,
+        crawl_date: Union[datetime, date],
+        error_handling: Literal["suppress", "catch", "raise"] = "raise",
+    ) -> Dict[str, Any]:
+        return self._get_parser(crawl_date).parse(html, error_handling)

src/fundus/publishers/base_objects.py

@@ -162,7 +162,7 @@ def search(
         unique_attributes = set(attributes)
         spec: Publisher
         for publisher in cls:
-            if unique_attributes.issubset(publisher.parser().attributes().names) and (
+            if unique_attributes.issubset(publisher.parser.latest_version.attributes().names) and (
                 publisher.supports(source_types) if source_types else True
             ):
                 matched.append(publisher)

src/fundus/scraping/scraper.py

@@ -1,3 +1,4 @@
+import warnings
 from typing import Dict, Iterator, List, Literal, Optional, Type
 
 import more_itertools
@@ -34,7 +35,7 @@ def scrape(
                 parser = self.parser_mapping[html.source_info.publisher]
 
                 try:
-                    extraction = parser(html.crawl_date).parse(html.content, error_handling)
+                    extraction = parser.parse(html.content, html.crawl_date, error_handling)
 
                 except Exception as error:
                     if error_handling == "raise":