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)
65 return XmlNode(self.node.getparent(), self.namespaces)
67 def remove_elements(self, name):
69 Removes all occurences of an element from the tree. Start at
70 specified root_node if specified, otherwise start at tree's root.
73 if not element_name.startswith('//'):
74 element_name = '//' + element_name
75 elements = self.node.xpath('%s ' % name, namespaces=self.namespaces)
76 for element in elements:
77 parent = element.getparent()
78 parent.remove(element)
80 def remove(self, element):
81 if isinstance(element, XmlNode):
82 self.node.remove(element.node)
84 self.node.remove(element)
86 def get(self, key, *args):
87 return self.node.get(key, *args)
89 def items(self): return self.node.items()
91 def set(self, key, value):
92 self.node.set(key, value)
94 def set_text(self, text):
98 del self.node.attrib[key]
100 def iterchildren(self):
101 return self.node.iterchildren()
104 return etree.tostring(self.node, encoding='UTF-8', pretty_print=True)
111 def __init__(self, xml=None, namespaces=None):
113 self.namespaces = namespaces
114 self.default_namespace = None
116 if isinstance(xml, basestring):
118 if isinstance(xml, XmlNode):
120 self.namespaces = xml.namespaces
121 elif isinstance(xml, etree._ElementTree) or isinstance(xml, etree._Element):
122 self.parse_xml(etree.tostring(xml))
124 def parse_xml(self, xml):
126 parse rspec into etree
128 parser = etree.XMLParser(remove_blank_text=True)
130 tree = etree.parse(xml, parser)
132 # 'rspec' file doesnt exist. 'rspec' is proably an xml string
134 tree = etree.parse(StringIO(xml), parser)
136 raise InvalidXML(str(e))
137 root = tree.getroot()
138 self.namespaces = dict(root.nsmap)
140 if 'default' not in self.namespaces and None in self.namespaces:
141 # If the 'None' exist, then it's pointing to the default namespace. This makes
142 # it hard for us to write xpath queries for the default naemspace because lxml
143 # wont understand a None prefix. We will just associate the default namespeace
144 # with a key named 'default'.
145 self.namespaces['default'] = self.namespaces.pop(None)
148 self.namespaces['default'] = 'default'
150 self.root = XmlNode(root, self.namespaces)
152 for key in self.root.attrib.keys():
153 if key.endswith('schemaLocation'):
154 # schema location should be at the end of the list
155 schema_parts = self.root.attrib[key].split(' ')
156 self.schema = schema_parts[1]
157 namespace, schema = schema_parts[0], schema_parts[1]
160 def parse_dict(self, d, root_tag_name='xml', element = None):
162 if self.root is None:
163 self.parse_xml('<%s/>' % root_tag_name)
170 # handle repeating fields
171 for (key, value) in d.items():
172 if isinstance(value, list):
175 if isinstance(val, dict):
176 child_element = etree.SubElement(element, key)
177 self.parse_dict(val, key, child_element)
178 elif isinstance(val, basestring):
179 child_element = etree.SubElement(element, key).text = val
181 elif isinstance(value, int):
182 d[key] = unicode(d[key])
186 # element.attrib.update will explode if DateTimes are in the
189 # looks like iteritems won't stand side-effects
191 if not isinstance(d[k],StringTypes):
194 element.attrib.update(d)
196 def validate(self, schema):
198 Validate against rng schema
200 relaxng_doc = etree.parse(schema)
201 relaxng = etree.RelaxNG(relaxng_doc)
202 if not relaxng(self.root):
203 error = relaxng.error_log.last_error
204 message = "%s (line %s)" % (error.message, error.line)
205 raise InvalidXML(message)
208 def xpath(self, xpath, namespaces=None):
210 namespaces = self.namespaces
211 return self.root.xpath(xpath, namespaces=namespaces)
213 def set(self, key, value, node=None):
216 return node.set(key, value)
218 def remove_attribute(self, name, node=None):
221 node.remove_attribute(name)
223 def add_element(self, name, **kwds):
225 Wrapper around etree.SubElement(). Adds an element to
226 specified parent node. Adds element to root node is parent is
230 xmlnode = parent.add_element(name, *kwds)
233 def remove_elements(self, name, node = None):
235 Removes all occurences of an element from the tree. Start at
236 specified root_node if specified, otherwise start at tree's root.
241 node.remove_elements(name)
243 def attributes_list(self, elem):
244 # convert a list of attribute tags into list of tuples
245 # (tagnme, text_value)
249 opts.append((e.tag, str(e.text).strip()))
252 def get_element_attributes(self, elem=None, depth=0):
254 elem = self.root_node
255 if not hasattr(elem, 'attrib'):
256 # this is probably not an element node with attribute. could be just and an
257 # attribute, return it
259 attrs = dict(elem.attrib)
260 attrs['text'] = str(elem.text).strip()
261 attrs['parent'] = elem.getparent()
262 if isinstance(depth, int) and depth > 0:
263 for child_elem in list(elem):
264 key = str(child_elem.tag)
266 attrs[key] = [self.get_element_attributes(child_elem, depth-1)]
268 attrs[key].append(self.get_element_attributes(child_elem, depth-1))
270 attrs['child_nodes'] = list(elem)
273 def append(self, elem):
274 return self.root.append(elem)
276 def iterchildren(self):
277 return self.root.iterchildren()
279 def merge(self, in_xml):
286 return etree.tostring(self.root.node, encoding='UTF-8', pretty_print=True)
288 # XXX smbaker, for record.load_from_string
289 def todict(self, elem=None):
293 d.update(elem.attrib)
294 d['text'] = elem.text
295 for child in elem.iterchildren():
296 if child.tag not in d:
298 d[child.tag].append(self.todict(child))
300 if len(d)==1 and ("text" in d):
305 def save(self, filename):
306 f = open(filename, 'w')
307 f.write(self.toxml())
311 #if __name__ == '__main__':
312 # rspec = RSpec('/tmp/resources.rspec')