X-Git-Url: https://git.phdru.name/?p=extfs.d.git;a=blobdiff_plain;f=xml;h=6cf2276f9c3fbcdcd0adcf0dce5fb658ea126644;hp=5ed82db406ae9a061a733b046c0f050102adc9df;hb=af76dda2d3c72d5712ae4559c38da2829398a27b;hpb=7333a3485fd782d8f7b3850e88f649d75afd505d diff --git a/xml b/xml index 5ed82db..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 @@ -37,13 +37,13 @@ The VFS was inspired by a FUSE xmlfs: https://github.com/halhen/xmlfs """ -__version__ = "0.5.1" +__version__ = "0.6.1" __author__ = "Oleg Broytman " __copyright__ = "Copyright (C) 2013 PhiloSoft Design" __license__ = "GPL" -default_implementation = None # Can be None for default choice, - # 'lxml', 'elementtree' or 'minidom' +force_implementation = None # Can be None for default choice, + # 'lxml', 'elementtree' or 'minidom' use_minidom = True use_elementtree = False @@ -115,16 +115,50 @@ locale.setlocale(locale.LC_ALL, '') class XmlVfs(object): + """Abstract base class""" + + supports_namespaces = False + def __init__(self): self.parse() def list(self): self._list(self.getroot()) - def has_ns(self, node): - return False - - def get_child_node(self, node, i): + def _list(self, node, path=''): + n = len(self.getchildren(node)) + if n: + width = int(math.log10(n)) + 1 + template = "%%0%dd" % width + else: + template = "%d" + n = 0 + for element in self.getchildren(node): + if not self.istag(element): + continue + n += 1 + tag = self.getlocalname(self.gettag(element)) + if path: + subpath = '%s/%s %s' % (path, template % n, tag) + else: + subpath = '%s %s' % (template % n, tag) + subpath_encoded = subpath.encode(default_encoding, "replace") + print "dr-xr-xr-x 1 user group 0 Jan 1 00:00 %s" % subpath_encoded + if self.getattrs(element): + attr_text = self.attrs2text(element) + print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/attributes" % ( + len(attr_text), subpath_encoded) + if self.supports_namespaces and self.has_ns(element): + ns_text = self.ns2text(element) + print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/namespaces" % ( + len(ns_text), subpath_encoded) + text = self.collect_text(element) + if text: + print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/text" % ( + 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): @@ -133,21 +167,26 @@ class XmlVfs(object): return element xml_error('There are less than %d nodes' % i) + def attrs2text(self, node): + attr_accumulator = [] + for name, value in self.getattrs(node): + name = self.getlocalname(name).encode(default_encoding, "replace") + value = value.encode(default_encoding, "replace") + attr_accumulator.append("%s=%s" % (name, value)) + return '\n'.join(attr_accumulator) + + def has_ns(self, node): + return False + class MiniDOMXmlVfs(XmlVfs): def parse(self): self.document = xml.dom.minidom.parse(sys.argv[2]) - def hasattrs(self, node): - return bool(node.attributes) - - def attrs2text(self, node): + def getattrs(self, node): attrs = node.attributes attrs = [attrs.item(i) for i in range(attrs.length)] - return '\n'.join(["%s=%s" % - (a.name.encode(default_encoding, "replace"), - a.value.encode(default_encoding, "replace")) - for a in attrs]) + return [(a.name, a.value) for a in attrs] def collect_text(self, node): text_accumulator = [] @@ -163,51 +202,26 @@ class MiniDOMXmlVfs(XmlVfs): if text: text_accumulator.append(text) return '\n'.join(text_accumulator).encode(default_encoding, "replace") - def _list(self, node, path=''): - childNodes = node.childNodes - n = 0 - for element in childNodes: - if element.localName: - n += 1 - if n: - width = int(math.log10(n)) + 1 - template = "%%0%dd" % width - else: - template = "%d" - n = 0 - for element in childNodes: - if element.localName: - n += 1 - if path: - subpath = '%s/%s %s' % (path, template % n, element.localName) - else: - subpath = '%s %s' % (template % n, element.localName) - subpath_encoded = subpath.encode(default_encoding, "replace") - print "dr-xr-xr-x 1 user group 0 Jan 1 00:00 %s" % subpath_encoded - if self.hasattrs(element): - attr_text = self.attrs2text(element) - print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/attributes" % ( - len(attr_text), subpath_encoded) - text = self.collect_text(element) - if text: - print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/text" % ( - len(text), subpath_encoded) - self._list(element, subpath) - def getroot(self): return self.document def getchildren(self, node): return node.childNodes + def gettag(self, node): + return node.localName + def istag(self, node): return bool(node.localName) + def getlocalname(self, name): + return name + if use_elementtree or use_lxml: class CommonEtreeXmlVfs(XmlVfs): - def hasattrs(self, node): - return bool(node.attrib) + def getattrs(self, node): + return node.attrib.items() def collect_text(self, node): text_accumulator = [] @@ -223,12 +237,12 @@ 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) + def gettag(self, node): + return node.tag + def istag(self, node): return isinstance(node.tag, basestring) @@ -263,126 +277,65 @@ if use_elementtree: self.document = ET.parse(sys.argv[2], PIParser()) - def attrs2text(self, node): - attr_accumulator = [] - for name, value in node.attrib.items(): - name = name.encode(default_encoding, "replace") - value = value.encode(default_encoding, "replace") - if name.startswith('{'): - name = name.split('}', 1)[1] # Remove XML namespace - attr_accumulator.append("%s=%s" % (name, value)) - return '\n'.join(attr_accumulator) - - def _list(self, node, path=''): - n = len(node) - if n: - width = int(math.log10(n))+1 - template = "%%0%dd" % width - else: - template = "%d" - n = 0 - for element in node: - if not isinstance(element.tag, basestring): - continue - n += 1 - tag = element.tag - if tag.startswith('{'): - tag = tag.split('}', 1)[1] # Remove XML namespace - if path: - subpath = '%s/%s %s' % (path, template % n, tag) - else: - subpath = '%s %s' % (template % n, tag) - subpath_encoded = subpath.encode(default_encoding, "replace") - print "dr-xr-xr-x 1 user group 0 Jan 1 00:00 %s" % subpath_encoded - if self.hasattrs(element): - attr_text = self.attrs2text(element) - print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/attributes" % ( - len(attr_text), subpath_encoded) - text = self.collect_text(element) - if text: - print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/text" % ( - len(text), subpath_encoded) - self._list(element, subpath) + def getroot(self): + return self.document.getroot() + + def getlocalname(self, name): + if name.startswith('{'): + name = name.split('}', 1)[1] # Remove XML namespace + return name if use_lxml: class LxmlEtreeXmlVfs(CommonEtreeXmlVfs): + supports_namespaces = True + def parse(self): self.document = etree.parse(sys.argv[2]) - def attrs2text(self, node): - attr_accumulator = [] - for name, value in node.attrib.items(): - name = etree.QName(name).localname.encode(default_encoding, "replace") - value = value.encode(default_encoding, "replace") - attr_accumulator.append("%s=%s" % (name, value)) - return '\n'.join(attr_accumulator) + 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 list(self): - self._list(self.getroot()) - - def _list(self, node, path=''): - n = len(node) - if n: - width = int(math.log10(n))+1 - template = "%%0%dd" % width - else: - template = "%d" - n = 0 - for element in node: - if not isinstance(element.tag, basestring): - continue - n += 1 - tag = etree.QName(element.tag).localname - if path: - subpath = '%s/%s %s' % (path, template % n, tag) - else: - subpath = '%s %s' % (template % n, tag) - subpath_encoded = subpath.encode(default_encoding, "replace") - print "dr-xr-xr-x 1 user group 0 Jan 1 00:00 %s" % subpath_encoded - if self.hasattrs(element): - attr_text = self.attrs2text(element) - print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/attributes" % ( - len(attr_text), subpath_encoded) - if element.nsmap: - ns_text = self.ns2text(element) - print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/namespaces" % ( - len(ns_text), subpath_encoded) - text = self.collect_text(element) - if text: - print "-r--r--r-- 1 user group %d Jan 1 00:00 %s/text" % ( - len(text), subpath_encoded) - self._list(element, subpath) - - def getroot(self): - return [self.document.getroot()] - def build_xmlvfs(): - if default_implementation is None: + if force_implementation is None: if use_lxml: return LxmlEtreeXmlVfs() elif use_elementtree: return ElementTreeXmlVfs() else: return MiniDOMXmlVfs() - elif default_implementation == 'minidom': + elif force_implementation == 'minidom': return MiniDOMXmlVfs() - elif default_implementation == 'elementtree': + elif force_implementation == 'elementtree': return ElementTreeXmlVfs() - elif default_implementation == 'lxml': + elif force_implementation == 'lxml': return LxmlEtreeXmlVfs() + else: + raise ValueError('Unknown implementation "%s", expected "minidom", "elementtree" or "lxml"' % force_implementation) def mcxml_list(): @@ -403,23 +356,23 @@ 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: xml_error('Unknown file') if path_comp == 'attributes': - if xmlvfs.hasattrs(node): + if xmlvfs.getattrs(node): text = xmlvfs.attrs2text(node) else: 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)