freedak/Scrapling
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: duplicate ID segments in full-path selector generation (#241)

Browse Source
This commit is contained in:
Karim shoair
2026-04-13 12:04:20 +02:00
committed by GitHub
parent c2e66ab753 3c23436448
commit 65c54f3b91
2 changed files with 51 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scrapling/core/mixins.py
+6 -1
View File
@@ -26,7 +26,12 @@ class SelectorsGeneration:
if target.parent: if target.parent:
if target.attrib.get("id"): if target.attrib.get("id"):
# id is enough # id is enough
part = f"#{target.attrib['id']}" if css else f"[@id='{target.attrib['id']}']" if css:
part = f"#{target.attrib['id']}"
elif full_path:
part = f"*[@id='{target.attrib['id']}']"
else:
part = f"[@id='{target.attrib['id']}']"
selectorPath.append(part) selectorPath.append(part)
if not full_path: if not full_path:
return " > ".join(reversed(selectorPath)) if css else "//*" + "/".join(reversed(selectorPath)) return " > ".join(reversed(selectorPath)) if css else "//*" + "/".join(reversed(selectorPath))
tests/parser/test_general.py
+44
View File
@@ -321,6 +321,50 @@ def test_selectors_generation(page):
_traverse(page) _traverse(page)
def test_full_path_selector_no_duplicate_ids():
"""Test that full path selectors don't duplicate id segments (regression test)"""
html = '<html><body><div id="main"><p id="target">Hello</p></div></body></html>'
page = Selector(html)
target = page.css("#target").first
# CSS full path should not duplicate id selectors
css_full = target.generate_full_css_selector
assert css_full.count("#target") == 1, f"Duplicate #target in CSS full path: {css_full}"
assert css_full.count("#main") == 1, f"Duplicate #main in CSS full path: {css_full}"
# XPath full path should not duplicate id selectors
xpath_full = target.generate_full_xpath_selector
assert xpath_full.count("@id='target'") == 1, f"Duplicate @id='target' in XPath full path: {xpath_full}"
assert xpath_full.count("@id='main'") == 1, f"Duplicate @id='main' in XPath full path: {xpath_full}"
# The generated CSS selector should actually select the correct element
result = page.css(css_full)
assert len(result) == 1
assert result.first.text == "Hello"
# The generated XPath selector should also select the correct element
result = page.xpath(xpath_full)
assert len(result) == 1, f"XPath '{xpath_full}' selected {len(result)} elements, expected 1"
assert result.first.text == "Hello"
def test_full_path_selector_mixed_id_and_no_id():
"""Test full path selectors with a mix of elements with and without ids"""
html = '<html><body><div id="wrapper"><section><p>Text</p></section></div></body></html>'
page = Selector(html)
target = page.css("p").first
css_full = target.generate_full_css_selector
# p has no id, so it should appear as a tag name; div has id
assert "#wrapper" in css_full
assert css_full.count("#wrapper") == 1
# Verify the selector works
result = page.css(css_full)
assert len(result) == 1
assert result.first.text == "Text"
# Miscellaneous Tests # Miscellaneous Tests
def test_getting_all_text(page): def test_getting_all_text(page):
"""Test getting all text from the page""" """Test getting all text from the page"""