8 from xml.dom import minidom
9 from types import StringTypes, ListType
13 def __init__(self, xml = None, xsd = None, NSURL = None):
15 Class to manipulate RSpecs. Reads and parses rspec xml into python dicts
16 and reads python dicts and writes rspec xml
18 self.xsd = # Schema. Can be local or remote file.
19 self.NSURL = # If schema is remote, Name Space URL to query (full path minus filename)
20 self.rootNode = # root of the DOM
21 self.dict = # dict of the RSpec.
22 self.schemaDict = {} # dict of the Schema
35 self.dict = self.toDict()
37 self._parseXSD(self.NSURL + self.xsd)
40 def _getText(self, nodelist):
43 if node.nodeType == node.TEXT_NODE:
47 # The rspec is comprised of 2 parts, and 1 reference:
48 # attributes/elements describe individual resources
49 # complexTypes are used to describe a set of attributes/elements
50 # complexTypes can include a reference to other complexTypes.
53 def _getName(self, node):
54 '''Gets name of node. If tag has no name, then return tag's localName'''
56 if not node.nodeName.startswith("#"):
59 elif node.attributes.has_key("name"):
60 name = node.attributes.get("name").value
64 # Attribute. {name : nameofattribute, {items: values})
65 def _attributeDict(self, attributeDom):
66 '''Traverse single attribute node. Create a dict {attributename : {name: value,}]}'''
67 node = {} # parsed dict
68 for attr in attributeDom.attributes.keys():
69 node[attr] = attributeDom.attributes.get(attr).value
73 def appendToDictOrCreate(self, dict, key, value):
74 if (dict.has_key(key)):
75 dict[key].append(value)
80 def toDict(self, nodeDom=None, parentdict=None, siblingdict={}, parent=None):
82 convert an XML to a nested dict:
83 * Non-terminal nodes (elements with string children and attributes) are simple dictionaries
84 * Terminal nodes (the rest) are nested dictionaries
90 curNodeName = nodeDom.localName
92 if (nodeDom.hasChildNodes()):
94 for child in nodeDom.childNodes[:-1]:
96 siblingdict = self.appendToDictOrCreate(siblingdict, curNodeName, child.nodeValue)
98 childdict = self.toDict(child, None, childdict, curNodeName)
100 child = nodeDom.childNodes[-1]
101 if (child.nodeValue):
102 siblingdict = self.appendToDictOrCreate(siblingdict, curNodeName, child.nodeValue)
104 siblingdict = self.toDict(child, siblingdict, childdict, curNodeName)
106 # Keep the attributes separate from text nodes
108 for attribute in nodeDom.attributes.keys():
109 attrdict = self.appendToDictOrCreate(attrdict, attribute, nodeDom.getAttribute(attribute))
111 self.appendToDictOrCreate(siblingdict, curNodeName, attrdict)
113 self.appendToDictOrCreate(siblingdict, curNodeName, [])
115 if (parentdict is not None):
116 parentdict = self.appendToDictOrCreate(parentdict, parent, siblingdict)
123 # def toDict(self, nodeDom = None):
125 # convert this rspec to a dict and return it.
129 # nodeDom = self.rootNode
131 # elementName = nodeDom.nodeName
132 # if elementName and not elementName.startswith("#"):
133 # # attributes have tags and values. get {tag: value}, else {type: value}
134 # node[elementName] = self._attributeDict(nodeDom)
135 # # resolve the child nodes.
136 # if nodeDom.hasChildNodes():
137 # for child in nodeDom.childNodes:
138 # childName = self._getName(child)
139 # # skip null children
142 # # initialize the possible array of children
143 # if not node[elementName].has_key(childName):
144 # node[elementName][childName] = []
145 # # if child node has text child nodes
146 # # append the children to the array as strings
147 # if child.hasChildNodes() and isinstance(child.childNodes[0], minidom.Text):
148 # for nextchild in child.childNodes:
149 # node[elementName][childName].append(nextchild.data)
150 # # convert element child node to dict
152 # childdict = self.toDict(child)
153 # for value in childdict.values():
154 # node[elementName][childName].append(value)
155 # #node[childName].append(self.toDict(child))
161 convert this rspec to an xml string and return it.
163 return self.rootNode.toxml()
166 def toprettyxml(self):
168 print this rspec in xml in a pretty format.
170 return self.rootNode.toprettyxml()
173 def parseFile(self, filename):
175 read a local xml file and store it as a dom object.
177 dom = minidom.parse(filename)
178 self.rootNode = dom.childNodes[0]
181 def parseString(self, xml):
183 read an xml string and store it as a dom object.
185 xml = xml.replace('\n', '').replace('\t', '').strip()
186 dom = minidom.parseString(xml)
187 self.rootNode = dom.childNodes[0]
190 def _httpGetXSD(self, xsdURI):
191 # split the URI into relevant parts
192 host = xsdURI.split("/")[2]
193 if xsdURI.startswith("https"):
194 conn = httplib.HTTPSConnection(host,
195 httplib.HTTPSConnection.default_port)
196 elif xsdURI.startswith("http"):
197 conn = httplib.HTTPConnection(host,
198 httplib.HTTPConnection.default_port)
199 conn.request("GET", xsdURI)
200 # If we can't download the schema, raise an exception
201 r1 = conn.getresponse()
204 return r1.read().replace('\n', '').replace('\t', '').strip()
207 def _parseXSD(self, xsdURI):
209 Download XSD from URL, or if file, read local xsd file and set schemaDict
211 # Since the schema definiton is a global namespace shared by and agreed upon by
212 # others, this should probably be a URL. Check for URL, download xsd, parse, or
213 # if local file, use local file.
215 if xsdURI.startswith("http"):
217 schemaDom = minidom.parseString(self._httpGetXSD(xsdURI))
219 # logging.debug("%s: web file not found" % xsdURI)
220 # logging.debug("Using local file %s" % self.xsd")
222 print "Can't find %s on the web. Continuing." % xsdURI
224 if os.path.exists(xsdURI):
225 # logging.debug("using local copy.")
226 print "Using local %s" % xsdURI
227 schemaDom = minidom.parse(xsdURI)
229 raise Exception("Can't find xsd locally")
230 self.schemaDict = self.toDict(schemaDom.childNodes[0])
233 def dict2dom(self, rdict, include_doc = False):
235 convert a dict object into a dom object.
238 def elementNode(tagname, rd):
239 element = minidom.Element(tagname)
240 for key in rd.keys():
241 if isinstance(rd[key], StringTypes) or isinstance(rd[key], int):
242 element.setAttribute(key, str(rd[key]))
243 elif isinstance(rd[key], dict):
244 child = elementNode(key, rd[key])
245 element.appendChild(child)
246 elif isinstance(rd[key], list):
248 if isinstance(item, dict):
249 child = elementNode(key, item)
250 element.appendChild(child)
251 elif isinstance(item, StringTypes) or isinstance(item, int):
252 child = minidom.Element(key)
253 text = minidom.Text()
255 child.appendChild(text)
256 element.appendChild(child)
259 # Minidom does not allow documents to have more then one
260 # child, but elements may have many children. Because of
261 # this, the document's root node will be the first key/value
262 # pair in the dictionary.
263 node = elementNode(rdict.keys()[0], rdict.values()[0])
265 rootNode = minidom.Document()
266 rootNode.appendChild(node)
272 def parseDict(self, rdict, include_doc = True):
274 Convert a dictionary into a dom object and store it.
276 self.rootNode = self.dict2dom(rdict, include_doc)
279 def getDictsByTagName(self, tagname, dom = None):
281 Search the dom for all elements with the specified tagname
282 and return them as a list of dicts
287 doms = dom.getElementsByTagName(tagname)
288 dictlist = [self.toDict(d) for d in doms]
289 for item in dictlist:
290 for value in item.values():
294 def getDictByTagNameValue(self, tagname, value, dom = None):
296 Search the dom for the first element with the specified tagname
297 and value and return it as a dict.
302 dicts = self.getDictsByTagName(tagname, dom)
305 if rdict.has_key('name') and rdict['name'] in [value]:
311 def filter(self, tagname, attribute, blacklist = [], whitelist = [], dom = None):
313 Removes all elements where:
314 1. tagname matches the element tag
315 2. attribute matches the element attribte
316 3. attribute value is in valuelist
323 if dom.localName in [tagname] and dom.attributes.has_key(attribute):
324 if whitelist and dom.attributes.get(attribute).value not in whitelist:
325 dom.parentNode.removeChild(dom)
326 if blacklist and dom.attributes.get(attribute).value in blacklist:
327 dom.parentNode.removeChild(dom)
329 if dom.hasChildNodes():
330 for child in dom.childNodes:
331 self.filter(tagname, attribute, blacklist, whitelist, child)
334 def validateDicts(self):
344 def pprint(self, r = None, depth = 0):
346 Pretty print the dict
349 if r == None: r = self.dict
351 for tab in range(0,depth): line += " "
352 # check if it's nested
355 print line + "%s:" % i
356 self.pprint(r[i], depth + 1)
357 elif type(r) in (tuple, list):
358 for j in r: self.pprint(j, depth + 1)
359 # not nested so just print.
361 print line + "%s" % r
365 class RecordSpec(Rspec):
368 def parseDict(self, rdict, include_doc = False):
370 Convert a dictionary into a dom object and store it.
372 self.rootNode = self.dict2dom(rdict, include_doc)
374 def dict2dom(self, rdict, include_doc = False):
376 if not len(rdict.keys()) == 1:
377 record_dict = {self.root_tag : rdict}
378 return Rspec.dict2dom(self, record_dict, include_doc)