The VFS represents tags as directories; the directories are numbered to
distinguish tags with the same name; numbering also helps to sort tags by their
-order in XML instead of sorting them by name. Attributes, text nodes and
-comments are represented as text files; attributes are shown in a file named
-"attributes", attributes are listed in the file as name=value lines (I
-deliberately ignore a small chance of newline characters in values); names and
-values are reencoded to the console encoding. Text nodes and comments are
-collected in a file named "text", stripped and reencoded. The filesystem is
+order in XML instead of sorting them by name and prevents name clash when tag
+names coincide with the names of special files used by XML VFS. Attributes,
+text nodes and comments are represented as text files; attributes are shown in
+a file named "attributes", attributes are listed in the file as name=value
+lines (I deliberately ignore a small chance of newline characters in values);
+names and values are reencoded to the console encoding. Text nodes and comments
+are collected in a file named "text", stripped and reencoded. The filesystem is
read-only.
Implementation based on minidom doesn't understand namespaces, it just shows
"""
-__version__ = "1.1.2"
+__version__ = "1.1.4"
__author__ = "Oleg Broytman <phd@phdru.name>"
__copyright__ = "Copyright (C) 2013-2015 PhiloSoft Design"
__license__ = "GPL"
def ns2text(self, node):
ns_accumulator = []
for name, value in self._get_local_ns(node).items():
- if name is None: name = ''
- name = name.encode(default_encoding, "replace")
+ if name:
+ name = name.encode(default_encoding, "replace")
+ else:
+ name = 'xmlns'
value = value.encode(default_encoding, "replace")
ns_accumulator.append("%s=%s" % (name, value))
return '\n'.join(ns_accumulator)