def test_extract_css_complex_css(self, samples_dir):
# Make sure we get proper external stylesheets.
html_input = samples_dir.join("sample2.html").read_text('utf-8')
result, css = extract_css(html_input, 'sample.html')
assert len(css) == 210
assert css.startswith('@page { size: 21cm')
return
def test_extract_css_utf8_unicode(self):
# we can handle umlauts in unicode-strings.
result, css = extract_css(
u"<html><body>ä</body></html>", 'sample.html')
assert css is None
assert result == u'<html><body>ä</body></html>'
return
def test_extract_css_complex_html(self, samples_dir):
# Make sure we have styles purged and replaced by a link
html_input = samples_dir.join("sample2.html").read_text('utf-8')
result, css = extract_css(html_input, 'sample.html')
assert '<style' not in result
link = '<link href="sample.css" rel="stylesheet" type="text/css"/>'
assert link in result
return
def test_extract_css_no_empty_comments(self):
# Make sure there are no empty comments in CSS
html_input_path = os.path.join(
os.path.dirname(__file__), 'input', 'sample2.html')
html_input = open(html_input_path, 'rb').read()
result, css = extract_css(html_input, 'sample.html')
assert '/*' not in result
return
def test_extract_css_complex_css(self):
# Make sure we get proper external stylesheets.
html_input_path = os.path.join(
os.path.dirname(__file__), 'input', 'sample2.html')
html_input = open(html_input_path, 'rb').read()
result, css = extract_css(html_input, 'sample.html')
assert len(css) == 156
assert css.startswith('@page { size: 21cm')
return
def test_extract_css_prettify(self):
# We can get prettified HTML (although it might be broken)
result, css = extract_css(
"<span>text<span>no</span>gap</span>", "sample.html",
prettify_html=True
)
assert result == (
"<span>\n text\n <span>\n no\n </span>\n gap\n</span>"
)
def test_extract_css_complex_html(self):
# Make sure we have styles purged and replaced by a link
html_input_path = os.path.join(
os.path.dirname(__file__), 'input', 'sample2.html')
html_input = open(html_input_path, 'rb').read()
result, css = extract_css(html_input, 'sample.html')
assert '<style' not in result
link = '<link href="sample.css" rel="stylesheet" type="text/css"/>'
assert link in result
return
def test_extract_css_contains_all_styles_from_input(self, samples_dir):
# Extracted CSS contains all styles from input HTML
content = samples_dir.join("sample2.html").read_text('utf-8')
html, css = extract_css(content, "sample.html")
assert css == (
"@page { size: 21cm 29.7cm; margin: 2cm }\n"
"p { margin-bottom: 0.21cm }\n"
"span.c2 { font-family: DejaVu Sans Mono, sans-serif }\n"
"span.c3 { font-family: DejaVu Sans Mönö, sans-serif }\n"
"p.c1 { margin-bottom: 0cm }\n \n "
)
def process(self, path, metadata):
ext = os.path.splitext(path)[1]
if ext not in self.supported_extensions:
return path, metadata
basename = os.path.basename(path)
src_path = os.path.join(copy_to_secure_location(path), basename)
remove_file_dir(path)
new_html, css = extract_css(
open(src_path, 'rb').read().decode('utf-8'),
basename,
prettify_html=self.options['css_cleaner_prettify_html'])
css, errors = cleanup_css(
css, minified=self.options['css_cleaner_minified'])
css_file = os.path.splitext(src_path)[0] + '.css'
if css is not None:
with open(css_file, 'wb') as fd:
fd.write(css.encode('utf-8'))
with open(src_path, 'wb') as fd:
fd.write(new_html.encode('utf-8'))
return src_path, metadata
def process(self, path, metadata):
ext = os.path.splitext(path)[1]
if ext not in self.supported_extensions:
return path, metadata
basename = os.path.basename(path)
src_path = os.path.join(
copy_to_secure_location(path), basename)
remove_file_dir(path)
new_html, css = extract_css(
open(src_path, 'rb').read().decode('utf-8'), basename,
prettify_html=self.options['css_cleaner_prettify_html'])
css, errors = cleanup_css(
css, minified=self.options['css_cleaner_minified'])
css_file = os.path.splitext(src_path)[0] + '.css'
if css is not None:
with open(css_file, 'wb') as fd:
fd.write(css.encode('utf-8'))
with open(src_path, 'wb') as fd:
fd.write(new_html.encode('utf-8'))
return src_path, metadata
def test_extract_css_puts_links_into_html(self, samples_dir):
# the returned HTML part has the styles replaced with a link:
content = samples_dir.join("sample2.html").read_text('utf-8')
html, css = extract_css(content, "sample.html")
assert html == (
'<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"\n'
' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">\n'
'\n'
'<html xmlns="http://www.w3.org/1999/xhtml">\n'
'<head>\n'
'<meta content="HTML Tidy for Linux/x86 (vers 6 November 2007)'
', see www.w3.org" name="generator"/>\n'
'<meta content="text/html; charset=utf-8" '
'http-equiv="CONTENT-TYPE"/>\n'
'<title></title>\n'
'<meta content="OpenOffice.org 2.4 (Linux)" name="GENERATOR"/>\n'
'<meta content="Uli Fouquet" name="AUTHOR"/>\n'
'<meta content="20110517;485000" name="CREATED"/>\n'
'<meta content="Uli Fouquet" name="CHANGEDBY"/>\n'
'<meta content="20110517;524000" name="CHANGED"/>\n'
'<link href="sample.css" rel="stylesheet" type="text/css"/>\n'
'\n'
'</head>\n'
'<body dir="ltr" lang="de-DE" xml:lang="de-DE">\n'
'<p class="c1">Some text</p>\n'
'<p class="c1"><br/></p>\n'
'<p class="c1">with <b>bold</b> and <i>italic</i> fonts.</p>\n'
'<p class="c1"><br/></p>\n'
'<p class="c1">Also a <span class="c2">complete different\n'
' font</span> is here. With umlaut: ä</p>\n'
'<p class="c1">Finally, some\n'
' <span class="c2">seam</span><span>less text.</span>\n'
'</p>\n'
'<p class="c1"><br/></p>\n'
'</body>\n'
'</html>\n'
)
def test_extract_css_utf8(self):
# we do not stumble over umlauts.
result, css = extract_css(
"<html><body>äö</body></html>", 'sample.html')
assert css is None
assert result == '<html><body>äö</body></html>'
def test_extract_css_no_empty_comments(self, samples_dir):
# Make sure there are no empty comments in CSS
html_input = samples_dir.join("sample2.html").read_text('utf-8')
result, css = extract_css(html_input, 'sample.html')
assert '/*' not in result
return
def test_extract_css_utf8_unicode(self):
result, css = extract_css(
"<html><body>ä</body></html>", 'sample.html')
assert css is None
assert result == '<html>\n <body>\n ä\n </body>\n</html>'
return
def test_extract_css_utf8(self):
result, css = extract_css(
"<html><body>äö</body></html>", 'sample.html')
assert css is None
assert result == '<html>\n <body>\n äö\n </body>\n</html>'
def test_extract_css_nested_styles(self):
# Trash in, trash out...
result, css = extract_css(
"<html><style>a<style>b</style></style></html>", 'sample.html')
assert css == 'a\nb'
def test_extract_css_empty_styles2(self):
# Also trashy docs can be handled
result, css = extract_css(
"<html><style /></html>", 'sample.html')
assert css is None
assert result == "<html>\n</html>"
def test_extract_css_empty_styles1(self):
# Also trashy docs can be handled
result, css = extract_css(
"<style></style>", 'sample.html')
assert css is None
assert result == ""
def test_extract_css_simple(self):
result, css = extract_css(
"<style>a, b</style>", 'sample.html')
link = '<link href="sample.css" rel="stylesheet" type="text/css"/>\n'
assert css == 'a, b'
assert result == link
def test_extract_css_trash(self):
# Also trashy docs can be handled
result, css = extract_css("", 'sample.html')
assert css is None
assert result == ""
def test_extract_css_no_prettify_by_default(self):
# by default we do not get prettified html
result, css = extract_css(
"<span>text<span>no</span>gap</span>", "sample.html"
)
assert result == "<span>text<span>no</span>gap</span>"