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

chore: migrating to ruff and updating pre-commit hooks

Browse Source
This commit is contained in:
Karim shoair
2025-04-13 17:32:00 +02:00
parent f34b42ea33
commit 0c8dd63f87
35 changed files with 2324 additions and 1182 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scrapling/core/mixins.py
+20 -16
View File
@@ -1,32 +1,33 @@
class SelectorsGeneration:
"""Selectors generation functions
Trying to generate selectors like Firefox or maybe cleaner ones!? Ehm
Inspiration: https://searchfox.org/mozilla-central/source/devtools/shared/inspector/css-logic.js#591"""
def __general_selection(self, selection: str = 'css', full_path=False) -> str:
def __general_selection(self, selection: str = "css", full_path=False) -> str:
"""Generate a selector for the current element.
:return: A string of the generated selector.
"""
selectorPath = []
target = self
css = selection.lower() == 'css'
css = selection.lower() == "css"
while target is not None:
if target.parent:
if target.attrib.get('id'):
if target.attrib.get("id"):
# id is enough
part = (
f'#{target.attrib["id"]}' if css
f"#{target.attrib['id']}"
if css
else f"[@id='{target.attrib['id']}']"
)
selectorPath.append(part)
if not full_path:
return (
" > ".join(reversed(selectorPath)) if css
else '//*' + "/".join(reversed(selectorPath))
" > ".join(reversed(selectorPath))
if css
else "//*" + "/".join(reversed(selectorPath))
)
else:
part = f'{target.tag}'
part = f"{target.tag}"
# We won't use classes anymore because I some websites share exact classes between elements
# classes = target.attrib.get('class', '').split()
# if classes and css:
@@ -41,23 +42,26 @@ class SelectorsGeneration:
if counter[target.tag] > 1:
part += (
f":nth-of-type({counter[target.tag]})" if css
f":nth-of-type({counter[target.tag]})"
if css
else f"[{counter[target.tag]}]"
)
selectorPath.append(part)
target = target.parent
if target is None or target.tag == 'html':
if target is None or target.tag == "html":
return (
" > ".join(reversed(selectorPath)) if css
else '//' + "/".join(reversed(selectorPath))
" > ".join(reversed(selectorPath))
if css
else "//" + "/".join(reversed(selectorPath))
)
else:
break
return (
" > ".join(reversed(selectorPath)) if css
else '//' + "/".join(reversed(selectorPath))
" > ".join(reversed(selectorPath))
if css
else "//" + "/".join(reversed(selectorPath))
)
@property
@@ -79,11 +83,11 @@ class SelectorsGeneration:
"""Generate a XPath selector for the current element
:return: A string of the generated selector.
"""
return self.__general_selection('xpath')
return self.__general_selection("xpath")
@property
def generate_full_xpath_selector(self) -> str:
"""Generate a complete XPath selector for the current element
:return: A string of the generated selector.
"""
return self.__general_selection('xpath', full_path=True)
return self.__general_selection("xpath", full_path=True)