You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
83 lines
2.6 KiB
83 lines
2.6 KiB
"""Tests to ensure that the lxml tree builder generates good trees."""
|
|
|
|
import re
|
|
import warnings
|
|
|
|
try:
|
|
import lxml.etree
|
|
LXML_PRESENT = True
|
|
LXML_VERSION = lxml.etree.LXML_VERSION
|
|
except ImportError, e:
|
|
LXML_PRESENT = False
|
|
LXML_VERSION = (0,)
|
|
|
|
if LXML_PRESENT:
|
|
from bs4.builder import LXMLTreeBuilder, LXMLTreeBuilderForXML
|
|
|
|
from bs4 import (
|
|
BeautifulSoup,
|
|
BeautifulStoneSoup,
|
|
)
|
|
from bs4.element import Comment, Doctype, SoupStrainer
|
|
from bs4.testing import skipIf
|
|
from bs4.tests import test_htmlparser
|
|
from bs4.testing import (
|
|
HTMLTreeBuilderSmokeTest,
|
|
XMLTreeBuilderSmokeTest,
|
|
SoupTest,
|
|
skipIf,
|
|
)
|
|
|
|
@skipIf(
|
|
not LXML_PRESENT,
|
|
"lxml seems not to be present, not testing its tree builder.")
|
|
class LXMLTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest):
|
|
"""See ``HTMLTreeBuilderSmokeTest``."""
|
|
|
|
@property
|
|
def default_builder(self):
|
|
return LXMLTreeBuilder()
|
|
|
|
def test_out_of_range_entity(self):
|
|
self.assertSoupEquals(
|
|
"<p>foo�bar</p>", "<p>foobar</p>")
|
|
self.assertSoupEquals(
|
|
"<p>foo�bar</p>", "<p>foobar</p>")
|
|
self.assertSoupEquals(
|
|
"<p>foo�bar</p>", "<p>foobar</p>")
|
|
|
|
def test_entities_in_foreign_document_encoding(self):
|
|
# We can't implement this case correctly because by the time we
|
|
# hear about markup like "“", it's been (incorrectly) converted into
|
|
# a string like u'\x93'
|
|
pass
|
|
|
|
# In lxml < 2.3.5, an empty doctype causes a segfault. Skip this
|
|
# test if an old version of lxml is installed.
|
|
|
|
@skipIf(
|
|
not LXML_PRESENT or LXML_VERSION < (2,3,5,0),
|
|
"Skipping doctype test for old version of lxml to avoid segfault.")
|
|
def test_empty_doctype(self):
|
|
soup = self.soup("<!DOCTYPE>")
|
|
doctype = soup.contents[0]
|
|
self.assertEqual("", doctype.strip())
|
|
|
|
def test_beautifulstonesoup_is_xml_parser(self):
|
|
# Make sure that the deprecated BSS class uses an xml builder
|
|
# if one is installed.
|
|
with warnings.catch_warnings(record=True) as w:
|
|
soup = BeautifulStoneSoup("<b />")
|
|
self.assertEqual(u"<b/>", unicode(soup.b))
|
|
self.assertTrue("BeautifulStoneSoup class is deprecated" in str(w[0].message))
|
|
|
|
@skipIf(
|
|
not LXML_PRESENT,
|
|
"lxml seems not to be present, not testing its XML tree builder.")
|
|
class LXMLXMLTreeBuilderSmokeTest(SoupTest, XMLTreeBuilderSmokeTest):
|
|
"""See ``HTMLTreeBuilderSmokeTest``."""
|
|
|
|
@property
|
|
def default_builder(self):
|
|
return LXMLTreeBuilderForXML()
|