Merge pull request #46 from valentinogagliardi/maintenance/body-html

Simplify traversal; increase coverage

Author: valentinogagliardi

pyproject.toml

@@ -18,6 +18,8 @@ classifiers = [
   "Intended Audience :: Developers",
   "License :: OSI Approved :: MIT License",
   "Programming Language :: Python :: 3 :: Only",
+  "Programming Language :: Python :: 3.8",
+  "Programming Language :: Python :: 3.9",
   "Programming Language :: Python :: 3.10",
   "Programming Language :: Python :: 3.11",
   "Programming Language :: Python :: 3.12",

src/unbrowsed/queries.py

@@ -14,6 +14,8 @@
 from unbrowsed.types import AriaRoles
 from unbrowsed.resolvers import RoleResolver
 
+SELECTOR = "*:not(html):not(body)"
+
 
 class QueryResult:
     """Wrapper class for query result."""
@@ -170,29 +172,22 @@ def query_by_text(dom: Parser, text: str, exact=True) -> Optional[QueryResult]:
     search_text = TextMatch(text, exact=exact)
     matches = []
 
-    for element in dom.css("*"):
+    for element in dom.css(SELECTOR):
         element_text = element.text(deep=True, strip=True)
 
         if search_text.matches(element_text):
             matches.append(element)
 
     if len(matches) > 1:
-        exclusions = ("html", "body")
-        filtered_matches = [m for m in matches if m.tag not in exclusions]
-
-        if filtered_matches:
-            matches = filtered_matches
-
-        if len(matches) > 1:
-            for i, parent in enumerate(matches):
-                for j, child in enumerate(matches):
-                    if i != j and is_parent_of(parent, child):
-                        return QueryResult(matches[i])
+        for i, parent in enumerate(matches):
+            for j, child in enumerate(matches):
+                if i != j and is_parent_of(parent, child):
+                    return QueryResult(matches[i])
 
-            raise MultipleElementsFoundError(
-                f"Found {len(matches)} elements with text '{text}'. "
-                f"Use query_all_by_text if multiple matches are expected."
-            )
+        raise MultipleElementsFoundError(
+            f"Found {len(matches)} elements with text '{text}'. "
+            f"Use query_all_by_text if multiple matches are expected."
+        )
 
     if not matches:
         return None
@@ -276,7 +271,7 @@ def query_by_role(
     )
     matches = []
 
-    for element in dom.css("*"):
+    for element in dom.css(SELECTOR):
         if not role_matcher.matches(element):
             continue
 
@@ -288,13 +283,6 @@ def query_by_role(
 
         matches.append(element)
 
-    if len(matches) > 1:
-        exclusions = ("html", "body")
-        filtered_matches = [m for m in matches if m.tag not in exclusions]
-
-        if filtered_matches:
-            matches = filtered_matches
-
         if len(matches) > 1:
             for i, parent in enumerate(matches):
                 for j, child in enumerate(matches):

src/unbrowsed/resolvers.py

@@ -21,9 +21,7 @@ def resolve(node: LexborNode) -> Optional[str]:
                     text = element.text(deep=True, strip=True)
                     if text:
                         name_texts.append(text)
-
-            if name_texts:
-                return " ".join(name_texts)
+            return " ".join(name_texts)
 
         if aria_label := node.attributes.get("aria-label"):
             aria_label = aria_label.strip()
@@ -63,9 +61,7 @@ def resolve(node: LexborNode) -> Optional[str]:
                     return f"{alt_text} {node_text}"
                 return alt_text
 
-            content = node.text(deep=True, strip=True)
-            if content:
-                return content
+            return node.text(deep=True, strip=True)
 
         if title := node.attributes.get("title"):
             if title.strip():
@@ -176,11 +172,7 @@ def matches(self, node: LexborNode) -> bool:
 
     @staticmethod
     def get_implicit_role(node: LexborNode):
-        if not hasattr(node, "tag"):
-            return
         tag = node.tag
-        if not tag:
-            return
         handler = RoleResolver.get_implicit_role_mapping().get(
             tag  # type: ignore
         )
@@ -199,10 +191,7 @@ def get_td_role(node: LexborNode) -> str:
         while ancestor and ancestor.tag != "table":
             ancestor = ancestor.parent
 
-        if not ancestor:
-            return ""
-
-        table_role = ancestor.attributes.get("role")
+        table_role = ancestor.attributes.get("role")  # type: ignore
 
         if not table_role:
             return "cell"

tests/test_get_by_label_text.py

@@ -21,7 +21,8 @@ def labels_html():
 def test_get_by_label_text(labels_html):
     dom = parse_html(labels_html)
 
-    assert get_by_label_text(dom, "Email")
+    get_by_label_text(dom, "Email")
+    get_by_label_text(dom, "Password")
 
     with pytest.raises(NoElementsFoundError) as exc:
         get_by_label_text(dom, "email")
@@ -30,8 +31,6 @@ def test_get_by_label_text(labels_html):
         "Use query_by_label_text if expecting no matches." == str(exc.value)
     )
 
-    assert get_by_label_text(dom, "Password")
-
     html = """
         <form>
             <label for="email">Email Address</label>
@@ -40,15 +39,15 @@ def test_get_by_label_text(labels_html):
         """
     dom = parse_html(html)
 
-    assert get_by_label_text(dom, "Email Address")
+    get_by_label_text(dom, "Email Address")
 
 
 def test_get_by_label_text_exact(labels_html):
     dom = parse_html(labels_html)
 
-    assert get_by_label_text(dom, "email", exact=False)
-    assert get_by_label_text(dom, "password", exact=False)
-    assert get_by_label_text(dom, "passw", exact=False)
+    get_by_label_text(dom, "email", exact=False)
+    get_by_label_text(dom, "password", exact=False)
+    get_by_label_text(dom, "passw", exact=False)
 
 
 def test_get_by_label_text_no_match():
@@ -95,4 +94,14 @@ def test_get_by_label_text_nested_control():
     """
     dom = parse_html(html)
 
-    assert get_by_label_text(dom, "Username")
+    get_by_label_text(dom, "Username")
+
+
+def test_no_for_attr():
+    html = """
+    <label>Password</label>
+    <input id="password" type="password">
+    """
+    dom = parse_html(html)
+    with pytest.raises(NoElementsFoundError):
+        get_by_label_text(dom, "Password")

tests/test_get_by_role.py

@@ -712,6 +712,17 @@ def test_gey_by_role_select():
     dom = parse_html(html)
     get_by_role(dom, "combobox")
 
+    html = """
+    <label for="shakes">Which shakes would you like to order?</label>
+    <select id="shakes" name="shakes" size="1">
+      <option>Vanilla Shake</option>
+      <option>Strawberry Shake</option>
+      <option>Chocolate Shake</option>
+    </select>
+    """
+    dom = parse_html(html)
+    get_by_role(dom, "combobox")
+
     html = """
     <label for="shakes">Which shakes would you like to order?</label>
     <select id="shakes" name="shakes" size="2">

tests/test_resolvers.py

@@ -382,3 +382,14 @@ def test_role_matcher_input_types():
 
     unknown = dom.css_first('input[type="unknown"]')
     assert not RoleResolver.get_implicit_role(unknown)
+
+
+def test_input_without_label():
+    html = """
+        <input id="username" type="text">
+        <label>Username</label>
+    """
+    parser = LexborHTMLParser(html)
+    input_element = parser.css_first("input")
+
+    assert AccessibleNameResolver.resolve(input_element) is None