Adding WonderfulSoup.force_scrape_one

minet/scrape/__init__.py

@@ -5,7 +5,7 @@
 # Module exposing utilities related to minet's scraping DSL.
 #
 from minet.scrape.classes.definition import scrape, DefinitionScraper, validate
-from minet.scrape.soup import WonderfulSoup, SelectionError
+from minet.scrape.soup import WonderfulSoup, SelectionError, ExtractionError
 from minet.scrape.regex import (
     extract_encodings_from_xml,
     extract_canonical_link,
@@ -19,6 +19,7 @@
     "validate",
     "WonderfulSoup",
     "SelectionError",
+    "ExtractionError",
     "extract_encodings_from_xml",
     "extract_canonical_link",
     "extract_javascript_relocation",

minet/scrape/soup.py

@@ -19,6 +19,10 @@ class SelectionError(Exception):
     pass
 
 
+class ExtractionError(Exception):
+    pass
+
+
 def extract(elem: "MinetTag", target: Optional[str]) -> Optional[str]:
     if target is None or target == "text":
         return elem.get_text()
@@ -52,15 +56,23 @@ def select(self, css: str, *args, **kwargs) -> List["MinetTag"]:
 
         return cast(List["MinetTag"], super().select(css, *args, **kwargs))
 
-    def scrape_one(
-        self, css: str, target: Optional[str] = None, strict: bool = False
-    ) -> Optional[str]:
+    def force_scrape_one(self, css: str, target: Optional[str] = None) -> str:
         elem = self.select_one(css)
 
         if elem is None:
-            if strict:
-                raise SelectionError(css)
+            raise SelectionError(css)
+
+        value = extract(elem, target)
 
+        if value is None:
+            raise ExtractionError(target)
+
+        return value
+
+    def scrape_one(self, css: str, target: Optional[str] = None) -> Optional[str]:
+        elem = self.select_one(css)
+
+        if elem is None:
             return None
 
         return extract(elem, target)
@@ -130,10 +142,10 @@ def __init__(
         super().__init__(
             markup,
             features,
-            element_classes=WONDERFUL_ELEMENT_CLASSES,
+            element_classes=WONDERFUL_ELEMENT_CLASSES,  # type: ignore
             parse_only=parse_only,
             multi_valued_attributes=None,
-        )  # type: ignore
+        )
 
 
 @contextmanager

test/soup_test.py

@@ -1,6 +1,6 @@
 import pytest
 
-from minet.scrape.soup import WonderfulSoup, SelectionError
+from minet.scrape.soup import WonderfulSoup, SelectionError, ExtractionError
 
 HTML = """
 <div>
@@ -50,7 +50,10 @@ def test_scrape_one(self):
         assert soup.scrape_one("link") is None
 
         with pytest.raises(SelectionError):
-            assert soup.scrape_one("link", strict=True)
+            assert soup.force_scrape_one("link")
+
+        with pytest.raises(ExtractionError):
+            assert soup.force_scrape_one("h1", "blabla")
 
     def test_get(self):
         soup = WonderfulSoup('<div class=" a  b  ">Ok</div>')