X-Git-Url: https://git.phdru.name/?p=extfs.d.git;a=blobdiff_plain;f=xml;h=6cf2276f9c3fbcdcd0adcf0dce5fb658ea126644;hp=e78e47e4bc20d13403eb85f7cf8b43591ec18fdf;hb=af76dda2d3c72d5712ae4559c38da2829398a27b;hpb=db63b272c4064cfc541f89f70aa8354ee6920905 diff --git a/xml b/xml index e78e47e..6cf2276 100755 --- a/xml +++ b/xml @@ -26,7 +26,7 @@ read-only. Implementation based on minidom doesn't understand namespaces, it just shows them among other attributes. ElementTree-based implementation doesn't show namespaces at all. Implementation based on lxml.etree shows namespaces in a -separate file "namespaces"; every child tag includes its parent's namespaces. +separate file "namespaces". It is useful to have a top-down view on an XML structure but it's especially convenient to extract text values from tags. One can get, for example, a @@ -37,7 +37,7 @@ The VFS was inspired by a FUSE xmlfs: https://github.com/halhen/xmlfs """ -__version__ = "0.6.0" +__version__ = "0.6.1" __author__ = "Oleg Broytman " __copyright__ = "Copyright (C) 2013 PhiloSoft Design" __license__ = "GPL" @@ -158,6 +158,15 @@ class XmlVfs(object): len(text), subpath_encoded) self._list(element, subpath) + def get_tag_node(self, node, i): + n = 0 + for element in self.getchildren(node): + if self.istag(element): + n += 1 + if n == i: + return element + xml_error('There are less than %d nodes' % i) + def attrs2text(self, node): attr_accumulator = [] for name, value in self.getattrs(node): @@ -169,15 +178,6 @@ class XmlVfs(object): def has_ns(self, node): return False - def get_child_node(self, node, i): - n = 0 - for element in self.getchildren(node): - if self.istag(element): - n += 1 - if n == i: - return element - xml_error('There are less than %d nodes' % i) - class MiniDOMXmlVfs(XmlVfs): def parse(self): @@ -237,9 +237,6 @@ if use_elementtree or use_lxml: if text: text_accumulator.append(text) return '\n'.join(text_accumulator).encode(default_encoding, "replace") - def getroot(self): - return self.document.getroot() - def getchildren(self, node): return list(node) @@ -280,6 +277,9 @@ if use_elementtree: self.document = ET.parse(sys.argv[2], PIParser()) + def getroot(self): + return self.document.getroot() + def getlocalname(self, name): if name.startswith('{'): name = name.split('}', 1)[1] # Remove XML namespace @@ -293,23 +293,32 @@ if use_lxml: def parse(self): self.document = etree.parse(sys.argv[2]) + def getroot(self): + return [self.document.getroot()] + + def getlocalname(self, name): + return etree.QName(name).localname + + def _get_local_ns(self, node): + this_nsmap = node.nsmap + parent = node.getparent() + if parent is not None: + parents_nsmap = parent.nsmap + for key in parents_nsmap: + del this_nsmap[key] + return this_nsmap + def has_ns(self, node): - return bool(node.nsmap) + return bool(self._get_local_ns(node)) def ns2text(self, node): ns_accumulator = [] - for name, value in node.nsmap.items(): + for name, value in self._get_local_ns(node).items(): name = name.encode(default_encoding, "replace") value = value.encode(default_encoding, "replace") ns_accumulator.append("%s=%s" % (name, value)) return '\n'.join(ns_accumulator) - def getroot(self): - return [self.document.getroot()] - - def getlocalname(self, name): - return etree.QName(name).localname - def build_xmlvfs(): if force_implementation is None: @@ -347,7 +356,7 @@ def mcxml_copyout(): for path_comp in xml_filename.split('/'): if ' ' in path_comp: i = int(path_comp.split(' ', 1)[0]) - node = xmlvfs.get_child_node(node, i) + node = xmlvfs.get_tag_node(node, i) elif path_comp in ('attributes', 'namespaces', 'text'): break else: @@ -360,10 +369,10 @@ def mcxml_copyout(): xml_error('There are no attributes') elif path_comp == 'namespaces': - if xmlvfs.has_ns(node): + if xmlvfs.supports_namespaces and xmlvfs.has_ns(element): text = xmlvfs.ns2text(node) else: - xml_error('There are no attributes') + xml_error('There are no namespaces') elif path_comp == 'text': text = xmlvfs.collect_text(node)