From e269a2fd0aafcd83af7cf1e65bba65c0e5a2c18b Mon Sep 17 00:00:00 2001 From: Geoffrey Sneddon Date: Fri, 1 Nov 2013 18:35:13 +0000 Subject: [PATCH] Remove the isstring hack in lxml treewalker. This fixes #115 as a result, as we're no longer type-sniffing strings. --- html5lib/treewalkers/lxmletree.py | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/html5lib/treewalkers/lxmletree.py b/html5lib/treewalkers/lxmletree.py index 375cc2e8..bc934ac0 100644 --- a/html5lib/treewalkers/lxmletree.py +++ b/html5lib/treewalkers/lxmletree.py @@ -87,10 +87,6 @@ def __init__(self, fragment_root, obj): self.tail = ensure_str(self.obj.tail) else: self.tail = None - self.isstring = isinstance(obj, str) or isinstance(obj, bytes) - # Support for bytes here is Py2 - if self.isstring: - self.obj = ensure_str(self.obj) def __getattr__(self, name): return getattr(self.obj, name) @@ -143,7 +139,7 @@ def getNodeDetails(self, node): elif isinstance(node, Doctype): return _base.DOCTYPE, node.name, node.public_id, node.system_id - elif isinstance(node, FragmentWrapper) and node.isstring: + elif isinstance(node, FragmentWrapper) and not hasattr(node, "tag"): return _base.TEXT, node.obj elif node.tag == etree.Comment: