def test_expat_xmldecl():
e = parse.tree(
b"<?xml version='1.0' encoding='utf-8' standalone='yes'?><a/>",
parse.Expat(),
parse.NS(html),
parse.Node(),
validate=True)
assert not isinstance(e[0], xml.XML)
e = parse.tree(b"<a/>",
parse.Expat(xmldecl=True),
parse.NS(html),
parse.Node(),
validate=True)
assert not isinstance(e[0], xml.XML)
e = parse.tree(b"<?xml version='1.0'?><a/>",
parse.Expat(xmldecl=True),
parse.NS(html),
parse.Node(),
validate=True)
assert isinstance(e[0], xml.XML)
assert e[0].content == 'version="1.0"'
e = parse.tree(b"<?xml version='1.0' encoding='utf-8'?><a/>",
parse.Expat(xmldecl=True),
parse.NS(html),
parse.Node(),
validate=True)
assert isinstance(e[0], xml.XML)
assert e[0].content == 'version="1.0" encoding="utf-8"'
e = parse.tree(
b"<?xml version='1.0' encoding='utf-8' standalone='yes'?><a/>",
parse.Expat(xmldecl=True),
parse.NS(html),
parse.Node(),
validate=True)
assert isinstance(e[0], xml.XML)
assert e[0].content == 'version="1.0" encoding="utf-8" standalone="yes"'
def xsc2txt(instream, outstream, title, width):
e = parse.tree(parse.Stream(instream), parse.SGMLOP(), parse.NS(doc),
parse.Node(pool=xsc.docpool()))
if title is None:
title = xsc.Null
else:
title = doc.title(title)
e = html.html(html.body(doc.section(title, e)))
e = e.conv()
outstream.write(html.astext(e, width=width))
def test_parseurls():
# Check proper URL handling when parsing ``URLAttr`` or ``StyleAttr`` attributes
node = parse.tree(
b'<a href="4.html" style="background-image: url(3.gif);"/>',
parse.Expat(),
parse.NS(html),
parse.Node(base="root:1/2.html"),
validate=True)
assert str(node[0]["style"]) == "background-image: url(root:1/3.gif)"
assert node[0]["style"].urls() == [url.URL("root:1/3.gif")]
assert str(node[0]["href"]) == "root:1/4.html"
assert node[0]["href"].forInput(
root="gurk/hurz.html") == url.URL("gurk/1/4.html")
def test_parsingmethods():
t = "abc\U00012345\u3042xyz"
s = f'<?xml version="1.0" encoding="utf-8"?><a title="{t}">{t}</a>'
b = s.encode("utf-8")
def check(*pipeline):
node = parse.tree(*pipeline, validate=True)
node = node.walknodes(a)[0]
assert str(node) == t
assert str(node["title"]) == t
prefixes = {None: a.xmlns}
pool = xsc.Pool(a)
check(b, parse.Expat(), parse.NS(a.xmlns), parse.Node(pool))
check(s, parse.Encoder(encoding="utf-8"), parse.Expat(), parse.NS(a.xmlns),
parse.Node(pool))
check(parse.Iter(b), parse.Expat(), parse.NS(a.xmlns),
parse.Node(pool)) # parse byte by byte
check(parse.Stream(io.BytesIO(b), bufsize=1), parse.Expat(),
parse.NS(a.xmlns), parse.Node(pool))
check(parse.ETree(cElementTree.fromstring(b), defaultxmlns=a.xmlns),
parse.Node(pool))
def makexnd(stream, encoding=None, shareattrs="dupes", model="simple"): # :obj:`stream` can be a stream, an :class:`URL` or ``str``/``bytes`` encoding = None if isinstance(stream, str): encoding = "utf-8" stream = stream.encode(encoding) node = parse.tree(stream, parse.Expat(encoding=encoding), parse.NS(tld), parse.Node()) # get and convert the taglib object xnd = misc.first(node.walknodes(tld.taglib)).asxnd(model=model) if shareattrs=="dupes": xnd.shareattrs(False) elif shareattrs=="all": xnd.shareattrs(True) return xnd
def test_nsparse():
# A prepopulated prefix mapping and xmlns attributes should work together
xml = b"""
<x:a>
<x:a xmlns:x='http://www.w3.org/1999/xhtml'>
<x:a xmlns:x='http://xmlns.livinglogic.de/xist/ns/doc'>gurk</x:a>
</x:a>
</x:a>
"""
check = doc.a(html.a(doc.a("gurk")))
node = parse.tree(xml,
parse.Expat(),
parse.NS(x=doc),
parse.Node(),
validate=True)
node = node.walknodes(
xsc.Element)[0].compacted() # get rid of the Frag and whitespace
assert node == check
def getdoc(thing, format):
if thing.__doc__ is None:
return xsc.Null
# Remove indentation
lines = textwrap.dedent(thing.__doc__).split("\n")
# remove empty lines
while lines and not lines[0]:
del lines[0]
while lines and not lines[-1]:
del lines[-1]
text = "\n".join(lines)
modulename = _getmodulename(thing)
if inspect.ismethod(thing):
base = f"METHOD-DOCSTRING({modulename}.{thing.__class__.__name__}.{thing.__qualname__})"
elif isinstance(thing, property):
base = f"PROPERTY-DOCSTRING({modulename}.{thing})"
elif inspect.isfunction(thing):
base = f"FUNCTION-DOCSTRING({modulename}.{thing.__qualname__})"
elif inspect.isclass(thing):
base = f"CLASS-DOCSTRING({modulename}.{thing.__qualname__})"
elif inspect.ismodule(thing):
base = f"MODULE-DOCSTRING({modulename})"
else:
base = "DOCSTRING"
lformat = format.lower()
if lformat == "plaintext":
return xsc.Text(text)
elif lformat == "restructuredtext":
from ll.xist.ns import rest, doc
return rest.fromstring(text, base=base).conv(target=doc)
elif lformat == "xist":
from ll.xist.ns import doc
node = parse.tree(parse.String(text), parse.SGMLOP(), parse.NS(doc),
parse.Node(pool=xsc.docpool(), base=base))
if not node[
p]: # optimization: one paragraph docstrings don't need a <p> element.
node = p(node)
if inspect.ismethod(thing):
# Use the original method instead of the decorator
realthing = thing
while hasattr(realthing, "__wrapped__"):
realthing = realthing.__wrapped__
for ref in node.walknodes(pyref):
if "module" not in ref.attrs:
ref["module"] = _getmodulename(realthing)
if "class_" not in ref.attrs:
ref["class_"] = thing.__self__.__class__.__name__
if "method" not in ref.attrs:
ref["method"] = thing.__name__
elif inspect.isfunction(thing):
# Use the original method instead of the decorator
while hasattr(thing, "__wrapped__"):
thing = thing.__wrapped__
for ref in node.walknodes(pyref):
if "module" not in ref.attrs:
ref["module"] = _getmodulename(thing)
elif inspect.isclass(thing):
for ref in node.walknodes(pyref):
if "module" not in ref.attrs:
ref["module"] = _getmodulename(thing)
if "class_" not in ref.attrs:
ref["class_"] = thing.__name__
elif inspect.ismodule(thing):
for ref in node.walknodes(pyref):
if "module" not in ref.attrs:
ref["module"] = thing.__name__
return node
else:
raise ValueError(f"unsupported __docformat__ {format!r}")
def test_parse_tidy_empty():
e = parse.tree(b"", parse.Tidy(), parse.NS(), parse.Node(), validate=True)
assert not e
def test_frag():
e = parse.tree(b"das ist <b>klaus</b>. das ist <b>erich</b>",
parse.SGMLOP(), parse.NS(html), parse.Node())
assert "".join(map(str, e.walknodes(e // html.b))) == "klauserich"
def test_comments():
d = b'<html><head><style type="text/css">/*nix*/ p{/*nix*/ color: red;}</style></head><body><p>gurk</p></body></html>'
node = parse.tree(d, parse.Expat(), parse.NS(html), parse.Node())
css.applystylesheets(node)
assert str(node.walknodes(html.p)[0].attrs.style) == "color: red;"