def test_multibyte_conservative_truncation():
"""Ensure truncating a multibyte url slug won't massively shorten it."""
# this string has a comma as the 6th char which will be converted to an
# underscore, so if truncation amount isn't restricted, it would result in
# a 46-char slug instead of the full 100.
original = 'パイソンは、汎用のプログラミング言語である'
assert len(convert_to_url_slug(original, 100)) == 100
def test_multibyte_whole_character_truncation():
"""Ensure truncation happens at the edge of a multibyte character."""
# each of these characters url-encodes to 3 bytes = 9 characters each, so
# only the first character should be included for all lengths from 9 - 17
original = 'コード'
for limit in range(9, 18):
assert convert_to_url_slug(original, limit) == '%E3%82%B3'
def __init__(self, group: Group, page_name: str, markdown: str, user: User):
"""Create a new wiki page."""
self.group = group
self.page_name = page_name
self.slug = convert_to_url_slug(page_name)
# prevent possible conflict with url for creating a new page
if self.slug == "new_page":
raise ValueError("Invalid page name")
if self.file_path.exists():
raise ValueError("Wiki page already exists")
# create the directory for the group if it doesn't already exist
self.file_path.parent.mkdir(mode=0o755, exist_ok=True)
self.edit(markdown, user, "Create page")
def add_anchors_to_headings(html: str) -> str:
"""Replace all heading elements with ones with ids that link to themselves."""
soup = BeautifulSoup(html, features="html5lib")
headings = soup.find_all(["h1", "h2", "h3", "h4", "h5", "h6"])
for heading in headings:
# generate an anchor from the string contents of the heading
anchor = convert_to_url_slug("".join(heading.strings))
# create a link to that anchor, and put the heading's contents inside it
link = soup.new_tag("a", href=f"#{anchor}")
link.contents = heading.contents
# put that link in a replacement same-level heading with the anchor as id
new_heading = soup.new_tag(heading.name, id=anchor)
new_heading.append(link)
heading.replace_with(new_heading)
# html5lib adds <html> and <body> tags around the fragment, strip them back out
return "".join([str(tag) for tag in soup.body.children])
def url_slug(self) -> str:
"""Return the url slug for this topic."""
return convert_to_url_slug(self.title)
def test_multibyte_url_slug():
"""Ensure converting/truncating a slug with encoded characters works."""
original = 'Python ist eine üblicherweise höhere Programmiersprache'
expected = 'python_ist_eine_%C3%BCblicherweise'
assert convert_to_url_slug(original, 45) == expected
def test_url_slug_truncation():
"""Ensure a simple url slug truncates as expected."""
original = "Here's another string to truncate."
assert convert_to_url_slug(original, 15) == 'heres_another'
def test_url_slug_with_punctuation():
"""Ensure url slug conversion with punctuation works as expected."""
original = "Here's a string. It has (some) punctuation!"
expected = 'heres_a_string_it_has_some_punctuation'
assert convert_to_url_slug(original) == expected
def test_simple_url_slug_conversion():
"""Ensure that a simple url slug conversion works as expected."""
assert convert_to_url_slug("A Simple Test") == 'a_simple_test'
def test_url_slug_with_apostrophes():
"""Ensure url slugs don't replace apostrophes with underscores."""
original = "Here's what we don’t want as underscores"
expected = "heres_what_we_dont_want_as_underscores"
assert convert_to_url_slug(original) == expected
