2 from types import StringTypes
4 from StringIO import StringIO
5 from sfa.util.faults import InvalidXML
10 def filter_value(key, value):
12 if isinstance(value, str):
14 value = value.replace('*', '')
15 xpath = 'contains(%s, "%s")' % (key, value)
17 xpath = '%s="%s"' % (key, value)
25 for (key, value) in filter.items():
30 if isinstance(value, str):
31 filter_list.append(XpathFilter.filter_value(key, value))
32 elif isinstance(value, list):
33 stmt = ' or '.join([XpathFilter.filter_value(key, str(val)) for val in value])
34 filter_list.append(stmt)
36 xpath = ' and '.join(filter_list)
37 xpath = '[' + xpath + ']'
41 def __init__(self, node, namespaces):
44 self.namespaces = namespaces
45 self.attrib = node.attrib
48 def xpath(self, xpath, namespaces=None):
50 namespaces = self.namespaces
51 elems = self.node.xpath(xpath, namespaces=namespaces)
52 return [XmlNode(elem, namespaces) for elem in elems]
54 def add_element(self, tagname, **kwds):
55 element = etree.SubElement(self.node, tagname, **kwds)
56 return XmlNode(element, self.namespaces)
58 def append(self, elem):
59 if isinstance(elem, XmlNode):
60 self.node.append(elem.node)
62 self.node.append(elem)
64 def remove_elements(name):
66 Removes all occurences of an element from the tree. Start at
67 specified root_node if specified, otherwise start at tree's root.
70 if not element_name.startswith('//'):
71 element_name = '//' + element_name
72 elements = self.node.xpath('%s ' % name, namespaces=self.namespaces)
73 for element in elements:
74 parent = element.getparent()
75 parent.remove(element)
78 self.node.remove(element)
80 def get(self, key, *args):
81 return self.node.get(key, *args)
83 def items(self): return self.node.items()
85 def set(self, key, value):
86 self.node.set(key, value)
88 def set_text(self, text):
92 del self.node.attrib[key]
94 def iterchildren(self):
95 return self.node.iterchildren()
98 return etree.tostring(self.node, encoding='UTF-8', pretty_print=True)
105 def __init__(self, xml=None, namespaces=None):
107 self.namespaces = namespaces
108 self.default_namespace = None
110 if isinstance(xml, basestring):
112 if isinstance(xml, XmlNode):
114 self.namespaces = xml.namespaces
115 elif isinstance(xml, etree._ElementTree) or isinstance(xml, etree._Element):
116 self.parse_xml(etree.tostring(xml))
118 def parse_xml(self, xml):
120 parse rspec into etree
122 parser = etree.XMLParser(remove_blank_text=True)
124 tree = etree.parse(xml, parser)
126 # 'rspec' file doesnt exist. 'rspec' is proably an xml string
128 tree = etree.parse(StringIO(xml), parser)
130 raise InvalidXML(str(e))
131 root = tree.getroot()
132 self.namespaces = dict(root.nsmap)
134 if 'default' not in self.namespaces and None in self.namespaces:
135 # If the 'None' exist, then it's pointing to the default namespace. This makes
136 # it hard for us to write xpath queries for the default naemspace because lxml
137 # wont understand a None prefix. We will just associate the default namespeace
138 # with a key named 'default'.
139 self.namespaces['default'] = self.namespaces.pop(None)
142 self.namespaces['default'] = 'default'
144 self.root = XmlNode(root, self.namespaces)
146 for key in self.root.attrib.keys():
147 if key.endswith('schemaLocation'):
148 # schema location should be at the end of the list
149 schema_parts = self.root.attrib[key].split(' ')
150 self.schema = schema_parts[1]
151 namespace, schema = schema_parts[0], schema_parts[1]
154 def parse_dict(self, d, root_tag_name='xml', element = None):
156 if self.root is None:
157 self.parse_xml('<%s/>' % root_tag_name)
164 # handle repeating fields
165 for (key, value) in d.items():
166 if isinstance(value, list):
169 if isinstance(val, dict):
170 child_element = etree.SubElement(element, key)
171 self.parse_dict(val, key, child_element)
172 elif isinstance(val, basestring):
173 child_element = etree.SubElement(element, key).text = val
175 elif isinstance(value, int):
176 d[key] = unicode(d[key])
180 # element.attrib.update will explode if DateTimes are in the
183 # looks like iteritems won't stand side-effects
185 if not isinstance(d[k],StringTypes):
188 element.attrib.update(d)
190 def validate(self, schema):
192 Validate against rng schema
194 relaxng_doc = etree.parse(schema)
195 relaxng = etree.RelaxNG(relaxng_doc)
196 if not relaxng(self.root):
197 error = relaxng.error_log.last_error
198 message = "%s (line %s)" % (error.message, error.line)
199 raise InvalidXML(message)
202 def xpath(self, xpath, namespaces=None):
204 namespaces = self.namespaces
205 return self.root.xpath(xpath, namespaces=namespaces)
207 def set(self, key, value, node=None):
210 return node.set(key, value)
212 def remove_attribute(self, name, node=None):
215 node.remove_attribute(name)
218 def add_element(self, name, **kwds):
220 Wrapper around etree.SubElement(). Adds an element to
221 specified parent node. Adds element to root node is parent is
225 xmlnode = parent.add_element(name, *kwds)
228 def remove_elements(self, name, node = None):
230 Removes all occurences of an element from the tree. Start at
231 specified root_node if specified, otherwise start at tree's root.
236 node.remove_elements(name)
238 def attributes_list(self, elem):
239 # convert a list of attribute tags into list of tuples
240 # (tagnme, text_value)
244 opts.append((e.tag, str(e.text).strip()))
247 def get_element_attributes(self, elem=None, depth=0):
249 elem = self.root_node
250 if not hasattr(elem, 'attrib'):
251 # this is probably not an element node with attribute. could be just and an
252 # attribute, return it
254 attrs = dict(elem.attrib)
255 attrs['text'] = str(elem.text).strip()
256 attrs['parent'] = elem.getparent()
257 if isinstance(depth, int) and depth > 0:
258 for child_elem in list(elem):
259 key = str(child_elem.tag)
261 attrs[key] = [self.get_element_attributes(child_elem, depth-1)]
263 attrs[key].append(self.get_element_attributes(child_elem, depth-1))
265 attrs['child_nodes'] = list(elem)
268 def merge(self, in_xml):
275 return etree.tostring(self.root.node, encoding='UTF-8', pretty_print=True)
277 # XXX smbaker, for record.load_from_string
278 def todict(self, elem=None):
282 d.update(elem.attrib)
283 d['text'] = elem.text
284 for child in elem.iterchildren():
285 if child.tag not in d:
287 d[child.tag].append(self.todict(child))
289 if len(d)==1 and ("text" in d):
294 def save(self, filename):
295 f = open(filename, 'w')
296 f.write(self.toxml())
300 #if __name__ == '__main__':
301 # rspec = RSpec('/tmp/resources.rspec')