Update libs
This commit is contained in:
Ruud
@@ -1,20 +1,3 @@
|
||||
import warnings
|
||||
import re
|
||||
|
||||
import _base
|
||||
from html5lib.constants import DataLossWarning
|
||||
import html5lib.constants as constants
|
||||
import etree as etree_builders
|
||||
from html5lib import ihatexml
|
||||
|
||||
try:
|
||||
import lxml.etree as etree
|
||||
except ImportError:
|
||||
pass
|
||||
|
||||
fullTree = True
|
||||
tag_regexp = re.compile("{([^}]*)}(.*)")
|
||||
|
||||
"""Module for supporting the lxml.etree library. The idea here is to use as much
|
||||
of the native library as possible, without using fragile hacks like custom element
|
||||
names that break between releases. The downside of this is that we cannot represent
|
||||
@@ -26,12 +9,34 @@ Docypes with no name
|
||||
When any of these things occur, we emit a DataLossWarning
|
||||
"""
|
||||
|
||||
from __future__ import absolute_import, division, unicode_literals
|
||||
|
||||
import warnings
|
||||
import re
|
||||
import sys
|
||||
|
||||
from . import _base
|
||||
from ..constants import DataLossWarning
|
||||
from .. import constants
|
||||
from . import etree as etree_builders
|
||||
from .. import ihatexml
|
||||
|
||||
import lxml.etree as etree
|
||||
|
||||
|
||||
fullTree = True
|
||||
tag_regexp = re.compile("{([^}]*)}(.*)")
|
||||
|
||||
comment_type = etree.Comment("asd").tag
|
||||
|
||||
|
||||
class DocumentType(object):
|
||||
def __init__(self, name, publicId, systemId):
|
||||
self.name = name
|
||||
self.name = name
|
||||
self.publicId = publicId
|
||||
self.systemId = systemId
|
||||
|
||||
|
||||
class Document(object):
|
||||
def __init__(self):
|
||||
self._elementTree = None
|
||||
@@ -42,118 +47,126 @@ class Document(object):
|
||||
|
||||
def _getChildNodes(self):
|
||||
return self._childNodes
|
||||
|
||||
|
||||
childNodes = property(_getChildNodes)
|
||||
|
||||
|
||||
def testSerializer(element):
|
||||
rv = []
|
||||
finalText = None
|
||||
filter = ihatexml.InfosetFilter()
|
||||
infosetFilter = ihatexml.InfosetFilter()
|
||||
|
||||
def serializeElement(element, indent=0):
|
||||
if not hasattr(element, "tag"):
|
||||
if hasattr(element, "getroot"):
|
||||
#Full tree case
|
||||
if hasattr(element, "getroot"):
|
||||
# Full tree case
|
||||
rv.append("#document")
|
||||
if element.docinfo.internalDTD:
|
||||
if not (element.docinfo.public_id or
|
||||
if not (element.docinfo.public_id or
|
||||
element.docinfo.system_url):
|
||||
dtd_str = "<!DOCTYPE %s>"%element.docinfo.root_name
|
||||
dtd_str = "<!DOCTYPE %s>" % element.docinfo.root_name
|
||||
else:
|
||||
dtd_str = """<!DOCTYPE %s "%s" "%s">"""%(
|
||||
element.docinfo.root_name,
|
||||
dtd_str = """<!DOCTYPE %s "%s" "%s">""" % (
|
||||
element.docinfo.root_name,
|
||||
element.docinfo.public_id,
|
||||
element.docinfo.system_url)
|
||||
rv.append("|%s%s"%(' '*(indent+2), dtd_str))
|
||||
rv.append("|%s%s" % (' ' * (indent + 2), dtd_str))
|
||||
next_element = element.getroot()
|
||||
while next_element.getprevious() is not None:
|
||||
next_element = next_element.getprevious()
|
||||
while next_element is not None:
|
||||
serializeElement(next_element, indent+2)
|
||||
serializeElement(next_element, indent + 2)
|
||||
next_element = next_element.getnext()
|
||||
elif isinstance(element, basestring):
|
||||
#Text in a fragment
|
||||
rv.append("|%s\"%s\""%(' '*indent, element))
|
||||
elif isinstance(element, str) or isinstance(element, bytes):
|
||||
# Text in a fragment
|
||||
assert isinstance(element, str) or sys.version_info.major == 2
|
||||
rv.append("|%s\"%s\"" % (' ' * indent, element))
|
||||
else:
|
||||
#Fragment case
|
||||
# Fragment case
|
||||
rv.append("#document-fragment")
|
||||
for next_element in element:
|
||||
serializeElement(next_element, indent+2)
|
||||
elif type(element.tag) == type(etree.Comment):
|
||||
rv.append("|%s<!-- %s -->"%(' '*indent, element.text))
|
||||
serializeElement(next_element, indent + 2)
|
||||
elif element.tag == comment_type:
|
||||
rv.append("|%s<!-- %s -->" % (' ' * indent, element.text))
|
||||
if hasattr(element, "tail") and element.tail:
|
||||
rv.append("|%s\"%s\"" % (' ' * indent, element.tail))
|
||||
else:
|
||||
assert isinstance(element, etree._Element)
|
||||
nsmatch = etree_builders.tag_regexp.match(element.tag)
|
||||
if nsmatch is not None:
|
||||
ns = nsmatch.group(1)
|
||||
tag = nsmatch.group(2)
|
||||
prefix = constants.prefixes[ns]
|
||||
rv.append("|%s<%s %s>"%(' '*indent, prefix,
|
||||
filter.fromXmlName(tag)))
|
||||
rv.append("|%s<%s %s>" % (' ' * indent, prefix,
|
||||
infosetFilter.fromXmlName(tag)))
|
||||
else:
|
||||
rv.append("|%s<%s>"%(' '*indent,
|
||||
filter.fromXmlName(element.tag)))
|
||||
rv.append("|%s<%s>" % (' ' * indent,
|
||||
infosetFilter.fromXmlName(element.tag)))
|
||||
|
||||
if hasattr(element, "attrib"):
|
||||
attributes = []
|
||||
for name, value in element.attrib.iteritems():
|
||||
for name, value in element.attrib.items():
|
||||
nsmatch = tag_regexp.match(name)
|
||||
if nsmatch is not None:
|
||||
ns, name = nsmatch.groups()
|
||||
name = filter.fromXmlName(name)
|
||||
name = infosetFilter.fromXmlName(name)
|
||||
prefix = constants.prefixes[ns]
|
||||
attr_string = "%s %s"%(prefix, name)
|
||||
attr_string = "%s %s" % (prefix, name)
|
||||
else:
|
||||
attr_string = filter.fromXmlName(name)
|
||||
attr_string = infosetFilter.fromXmlName(name)
|
||||
attributes.append((attr_string, value))
|
||||
|
||||
for name, value in sorted(attributes):
|
||||
rv.append('|%s%s="%s"' % (' '*(indent+2), name, value))
|
||||
rv.append('|%s%s="%s"' % (' ' * (indent + 2), name, value))
|
||||
|
||||
if element.text:
|
||||
rv.append("|%s\"%s\"" %(' '*(indent+2), element.text))
|
||||
rv.append("|%s\"%s\"" % (' ' * (indent + 2), element.text))
|
||||
indent += 2
|
||||
for child in element.getchildren():
|
||||
for child in element:
|
||||
serializeElement(child, indent)
|
||||
if hasattr(element, "tail") and element.tail:
|
||||
rv.append("|%s\"%s\"" %(' '*(indent-2), element.tail))
|
||||
if hasattr(element, "tail") and element.tail:
|
||||
rv.append("|%s\"%s\"" % (' ' * (indent - 2), element.tail))
|
||||
serializeElement(element, 0)
|
||||
|
||||
if finalText is not None:
|
||||
rv.append("|%s\"%s\""%(' '*2, finalText))
|
||||
rv.append("|%s\"%s\"" % (' ' * 2, finalText))
|
||||
|
||||
return "\n".join(rv)
|
||||
|
||||
|
||||
def tostring(element):
|
||||
"""Serialize an element and its child nodes to a string"""
|
||||
rv = []
|
||||
finalText = None
|
||||
|
||||
def serializeElement(element):
|
||||
if not hasattr(element, "tag"):
|
||||
if element.docinfo.internalDTD:
|
||||
if element.docinfo.doctype:
|
||||
dtd_str = element.docinfo.doctype
|
||||
else:
|
||||
dtd_str = "<!DOCTYPE %s>"%element.docinfo.root_name
|
||||
dtd_str = "<!DOCTYPE %s>" % element.docinfo.root_name
|
||||
rv.append(dtd_str)
|
||||
serializeElement(element.getroot())
|
||||
|
||||
elif type(element.tag) == type(etree.Comment):
|
||||
rv.append("<!--%s-->"%(element.text,))
|
||||
|
||||
|
||||
elif element.tag == comment_type:
|
||||
rv.append("<!--%s-->" % (element.text,))
|
||||
|
||||
else:
|
||||
#This is assumed to be an ordinary element
|
||||
# This is assumed to be an ordinary element
|
||||
if not element.attrib:
|
||||
rv.append("<%s>"%(element.tag,))
|
||||
rv.append("<%s>" % (element.tag,))
|
||||
else:
|
||||
attr = " ".join(["%s=\"%s\""%(name, value)
|
||||
for name, value in element.attrib.iteritems()])
|
||||
rv.append("<%s %s>"%(element.tag, attr))
|
||||
attr = " ".join(["%s=\"%s\"" % (name, value)
|
||||
for name, value in element.attrib.items()])
|
||||
rv.append("<%s %s>" % (element.tag, attr))
|
||||
if element.text:
|
||||
rv.append(element.text)
|
||||
|
||||
for child in element.getchildren():
|
||||
for child in element:
|
||||
serializeElement(child)
|
||||
|
||||
rv.append("</%s>"%(element.tag,))
|
||||
rv.append("</%s>" % (element.tag,))
|
||||
|
||||
if hasattr(element, "tail") and element.tail:
|
||||
rv.append(element.tail)
|
||||
@@ -161,56 +174,57 @@ def tostring(element):
|
||||
serializeElement(element)
|
||||
|
||||
if finalText is not None:
|
||||
rv.append("%s\""%(' '*2, finalText))
|
||||
rv.append("%s\"" % (' ' * 2, finalText))
|
||||
|
||||
return "".join(rv)
|
||||
|
||||
|
||||
|
||||
class TreeBuilder(_base.TreeBuilder):
|
||||
documentClass = Document
|
||||
doctypeClass = DocumentType
|
||||
elementClass = None
|
||||
commentClass = None
|
||||
fragmentClass = Document
|
||||
fragmentClass = Document
|
||||
implementation = etree
|
||||
|
||||
def __init__(self, namespaceHTMLElements, fullTree = False):
|
||||
def __init__(self, namespaceHTMLElements, fullTree=False):
|
||||
builder = etree_builders.getETreeModule(etree, fullTree=fullTree)
|
||||
filter = self.filter = ihatexml.InfosetFilter()
|
||||
infosetFilter = self.infosetFilter = ihatexml.InfosetFilter()
|
||||
self.namespaceHTMLElements = namespaceHTMLElements
|
||||
|
||||
class Attributes(dict):
|
||||
def __init__(self, element, value={}):
|
||||
self._element = element
|
||||
dict.__init__(self, value)
|
||||
for key, value in self.iteritems():
|
||||
for key, value in self.items():
|
||||
if isinstance(key, tuple):
|
||||
name = "{%s}%s"%(key[2], filter.coerceAttribute(key[1]))
|
||||
name = "{%s}%s" % (key[2], infosetFilter.coerceAttribute(key[1]))
|
||||
else:
|
||||
name = filter.coerceAttribute(key)
|
||||
name = infosetFilter.coerceAttribute(key)
|
||||
self._element._element.attrib[name] = value
|
||||
|
||||
def __setitem__(self, key, value):
|
||||
dict.__setitem__(self, key, value)
|
||||
if isinstance(key, tuple):
|
||||
name = "{%s}%s"%(key[2], filter.coerceAttribute(key[1]))
|
||||
name = "{%s}%s" % (key[2], infosetFilter.coerceAttribute(key[1]))
|
||||
else:
|
||||
name = filter.coerceAttribute(key)
|
||||
name = infosetFilter.coerceAttribute(key)
|
||||
self._element._element.attrib[name] = value
|
||||
|
||||
class Element(builder.Element):
|
||||
def __init__(self, name, namespace):
|
||||
name = filter.coerceElement(name)
|
||||
name = infosetFilter.coerceElement(name)
|
||||
builder.Element.__init__(self, name, namespace=namespace)
|
||||
self._attributes = Attributes(self)
|
||||
|
||||
def _setName(self, name):
|
||||
self._name = filter.coerceElement(name)
|
||||
self._name = infosetFilter.coerceElement(name)
|
||||
self._element.tag = self._getETreeTag(
|
||||
self._name, self._namespace)
|
||||
|
||||
|
||||
def _getName(self):
|
||||
return filter.fromXmlName(self._name)
|
||||
|
||||
return infosetFilter.fromXmlName(self._name)
|
||||
|
||||
name = property(_getName, _setName)
|
||||
|
||||
def _getAttributes(self):
|
||||
@@ -218,24 +232,23 @@ class TreeBuilder(_base.TreeBuilder):
|
||||
|
||||
def _setAttributes(self, attributes):
|
||||
self._attributes = Attributes(self, attributes)
|
||||
|
||||
|
||||
attributes = property(_getAttributes, _setAttributes)
|
||||
|
||||
def insertText(self, data, insertBefore=None):
|
||||
data = filter.coerceCharacters(data)
|
||||
data = infosetFilter.coerceCharacters(data)
|
||||
builder.Element.insertText(self, data, insertBefore)
|
||||
|
||||
def appendChild(self, child):
|
||||
builder.Element.appendChild(self, child)
|
||||
|
||||
|
||||
class Comment(builder.Comment):
|
||||
def __init__(self, data):
|
||||
data = filter.coerceComment(data)
|
||||
data = infosetFilter.coerceComment(data)
|
||||
builder.Comment.__init__(self, data)
|
||||
|
||||
def _setData(self, data):
|
||||
data = filter.coerceComment(data)
|
||||
data = infosetFilter.coerceComment(data)
|
||||
self._element.text = data
|
||||
|
||||
def _getData(self):
|
||||
@@ -245,9 +258,9 @@ class TreeBuilder(_base.TreeBuilder):
|
||||
|
||||
self.elementClass = Element
|
||||
self.commentClass = builder.Comment
|
||||
#self.fragmentClass = builder.DocumentFragment
|
||||
# self.fragmentClass = builder.DocumentFragment
|
||||
_base.TreeBuilder.__init__(self, namespaceHTMLElements)
|
||||
|
||||
|
||||
def reset(self):
|
||||
_base.TreeBuilder.reset(self)
|
||||
self.insertComment = self.insertCommentInitial
|
||||
@@ -262,13 +275,13 @@ class TreeBuilder(_base.TreeBuilder):
|
||||
return self.document._elementTree
|
||||
else:
|
||||
return self.document._elementTree.getroot()
|
||||
|
||||
|
||||
def getFragment(self):
|
||||
fragment = []
|
||||
element = self.openElements[0]._element
|
||||
if element.text:
|
||||
fragment.append(element.text)
|
||||
fragment.extend(element.getchildren())
|
||||
fragment.extend(list(element))
|
||||
if element.tail:
|
||||
fragment.append(element.tail)
|
||||
return fragment
|
||||
@@ -278,59 +291,79 @@ class TreeBuilder(_base.TreeBuilder):
|
||||
publicId = token["publicId"]
|
||||
systemId = token["systemId"]
|
||||
|
||||
if not name or ihatexml.nonXmlNameBMPRegexp.search(name) or name[0] == '"':
|
||||
warnings.warn("lxml cannot represent null or non-xml doctype", DataLossWarning)
|
||||
if not name:
|
||||
warnings.warn("lxml cannot represent empty doctype", DataLossWarning)
|
||||
self.doctype = None
|
||||
else:
|
||||
coercedName = self.infosetFilter.coerceElement(name)
|
||||
if coercedName != name:
|
||||
warnings.warn("lxml cannot represent non-xml doctype", DataLossWarning)
|
||||
|
||||
doctype = self.doctypeClass(coercedName, publicId, systemId)
|
||||
self.doctype = doctype
|
||||
|
||||
doctype = self.doctypeClass(name, publicId, systemId)
|
||||
self.doctype = doctype
|
||||
|
||||
def insertCommentInitial(self, data, parent=None):
|
||||
self.initial_comments.append(data)
|
||||
|
||||
|
||||
def insertCommentMain(self, data, parent=None):
|
||||
if (parent == self.document and
|
||||
self.document._elementTree.getroot()[-1].tag == comment_type):
|
||||
warnings.warn("lxml cannot represent adjacent comments beyond the root elements", DataLossWarning)
|
||||
super(TreeBuilder, self).insertComment(data, parent)
|
||||
|
||||
def insertRoot(self, token):
|
||||
"""Create the document root"""
|
||||
#Because of the way libxml2 works, it doesn't seem to be possible to
|
||||
#alter information like the doctype after the tree has been parsed.
|
||||
#Therefore we need to use the built-in parser to create our iniial
|
||||
#tree, after which we can add elements like normal
|
||||
# Because of the way libxml2 works, it doesn't seem to be possible to
|
||||
# alter information like the doctype after the tree has been parsed.
|
||||
# Therefore we need to use the built-in parser to create our iniial
|
||||
# tree, after which we can add elements like normal
|
||||
docStr = ""
|
||||
if self.doctype and self.doctype.name and not self.doctype.name.startswith('"'):
|
||||
docStr += "<!DOCTYPE %s"%self.doctype.name
|
||||
if (self.doctype.publicId is not None or
|
||||
self.doctype.systemId is not None):
|
||||
docStr += ' PUBLIC "%s" "%s"'%(self.doctype.publicId or "",
|
||||
self.doctype.systemId or "")
|
||||
if self.doctype:
|
||||
assert self.doctype.name
|
||||
docStr += "<!DOCTYPE %s" % self.doctype.name
|
||||
if (self.doctype.publicId is not None or
|
||||
self.doctype.systemId is not None):
|
||||
docStr += (' PUBLIC "%s" ' %
|
||||
(self.infosetFilter.coercePubid(self.doctype.publicId or "")))
|
||||
if self.doctype.systemId:
|
||||
sysid = self.doctype.systemId
|
||||
if sysid.find("'") >= 0 and sysid.find('"') >= 0:
|
||||
warnings.warn("DOCTYPE system cannot contain single and double quotes", DataLossWarning)
|
||||
sysid = sysid.replace("'", 'U00027')
|
||||
if sysid.find("'") >= 0:
|
||||
docStr += '"%s"' % sysid
|
||||
else:
|
||||
docStr += "'%s'" % sysid
|
||||
else:
|
||||
docStr += "''"
|
||||
docStr += ">"
|
||||
if self.doctype.name != token["name"]:
|
||||
warnings.warn("lxml cannot represent doctype with a different name to the root element", DataLossWarning)
|
||||
docStr += "<THIS_SHOULD_NEVER_APPEAR_PUBLICLY/>"
|
||||
|
||||
try:
|
||||
root = etree.fromstring(docStr)
|
||||
except etree.XMLSyntaxError:
|
||||
print docStr
|
||||
raise
|
||||
|
||||
#Append the initial comments:
|
||||
root = etree.fromstring(docStr)
|
||||
|
||||
# Append the initial comments:
|
||||
for comment_token in self.initial_comments:
|
||||
root.addprevious(etree.Comment(comment_token["data"]))
|
||||
|
||||
#Create the root document and add the ElementTree to it
|
||||
|
||||
# Create the root document and add the ElementTree to it
|
||||
self.document = self.documentClass()
|
||||
self.document._elementTree = root.getroottree()
|
||||
|
||||
|
||||
# Give the root element the right name
|
||||
name = token["name"]
|
||||
namespace = token.get("namespace", self.defaultNamespace)
|
||||
if namespace is None:
|
||||
etree_tag = name
|
||||
else:
|
||||
etree_tag = "{%s}%s"%(namespace, name)
|
||||
etree_tag = "{%s}%s" % (namespace, name)
|
||||
root.tag = etree_tag
|
||||
|
||||
#Add the root element to the internal child/open data structures
|
||||
|
||||
# Add the root element to the internal child/open data structures
|
||||
root_element = self.elementClass(name, namespace)
|
||||
root_element._element = root
|
||||
self.document._childNodes.append(root_element)
|
||||
self.openElements.append(root_element)
|
||||
|
||||
#Reset to the default insert comment function
|
||||
self.insertComment = super(TreeBuilder, self).insertComment
|
||||
|
||||
# Reset to the default insert comment function
|
||||
self.insertComment = self.insertCommentMain
|
||||
|
||||
Reference in New Issue
Block a user