1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 """Simplexml module provides xmpppy library with all needed tools to handle XML nodes and XML streams.
18 I'm personally using it in many other separate projects. It is designed to be as standalone as possible."""
19
20 import xml.parsers.expat
21
23 """Returns provided string with symbols & < > " replaced by their respective XML entities."""
24 return txt.replace("&", "&").replace("<", "<").replace(">", ">").replace('"', """)
25
26 ENCODING='utf-8'
28 """Converts object "what" to unicode string using it's own __str__ method if accessible or unicode method otherwise."""
29 if type(what) == type(u''): return what
30 try: r=what.__str__()
31 except AttributeError: r=str(what)
32 if type(r)<>type(u''): return unicode(r,ENCODING)
33 return r
34
36 """ Node class describes syntax of separate XML Node. It have a constructor that permits node creation
37 from set of "namespace name", attributes and payload of text strings and other nodes.
38 It does not natively support building node from text string and uses NodeBuilder class for that purpose.
39 After creation node can be mangled in many ways so it can be completely changed.
40 Also node can be serialised into string in one of two modes: default (where the textual representation
41 of node describes it exactly) and "fancy" - with whitespace added to make indentation and thus make
42 result more readable by human.
43
44 Node class have attribute FORCE_NODE_RECREATION that is defaults to False thus enabling fast node
45 replication from the some other node. The drawback of the fast way is that new node shares some
46 info with the "original" node that is changing the one node may influence the other. Though it is
47 rarely needed (in xmpppy it is never needed at all since I'm usually never using original node after
48 replication (and using replication only to move upwards on the classes tree).
49 """
50 FORCE_NODE_RECREATION=0
51 - def __init__(self, tag=None, attrs={}, payload=[], parent=None, node=None):
52 """ Takes "tag" argument as the name of node (prepended by namespace, if needed and separated from it
53 by a space), attrs dictionary as the set of arguments, payload list as the set of textual strings
54 and child nodes that this node carries within itself and "parent" argument that is another node
55 that this one will be the child of. Also the __init__ can be provided with "node" argument that is
56 either a text string containing exactly one node or another Node instance to begin with. If both
57 "node" and other arguments is provided then the node initially created as replica of "node"
58 provided and then modified to be compliant with other arguments."""
59 if node:
60 if self.FORCE_NODE_RECREATION and type(node)==type(self): node=str(node)
61 if type(node)<>type(self): node=NodeBuilder(node,self)
62 else:
63 self.name,self.namespace,self.attrs,self.data,self.kids,self.parent = node.name,node.namespace,{},[],[],node.parent
64 for key in node.attrs.keys(): self.attrs[key]=node.attrs[key]
65 for data in node.data: self.data.append(data)
66 for kid in node.kids: self.kids.append(kid)
67 else: self.name,self.namespace,self.attrs,self.data,self.kids,self.parent = 'tag','',{},[],[],None
68
69 if tag: self.namespace, self.name = ([self.namespace]+tag.split())[-2:]
70 if parent: self.parent = parent
71 if self.parent and not self.namespace: self.namespace=self.parent.namespace
72 for attr in attrs.keys():
73 self.attrs[attr]=attrs[attr]
74 if type(payload) in (type(''),type(u'')): payload=[payload]
75 for i in payload:
76 if type(i)==type(self): self.addChild(node=i)
77 else: self.data.append(ustr(i))
78
80 """ Method used to dump node into textual representation.
81 if "fancy" argument is set to True produces indented output for readability."""
82 s = (fancy-1) * 2 * ' ' + "<" + self.name
83 if self.namespace:
84 if not self.parent or self.parent.namespace!=self.namespace:
85 s = s + ' xmlns="%s"'%self.namespace
86 for key in self.attrs.keys():
87 val = ustr(self.attrs[key])
88 s = s + ' %s="%s"' % ( key, XMLescape(val) )
89 s = s + ">"
90 cnt = 0
91 if self.kids:
92 if fancy: s = s + "\n"
93 for a in self.kids:
94 if not fancy and (len(self.data)-1)>=cnt: s=s+XMLescape(self.data[cnt])
95 elif (len(self.data)-1)>=cnt: s=s+XMLescape(self.data[cnt].strip())
96 s = s + a.__str__(fancy and fancy+1)
97 cnt=cnt+1
98 if not fancy and (len(self.data)-1) >= cnt: s = s + XMLescape(self.data[cnt])
99 elif (len(self.data)-1) >= cnt: s = s + XMLescape(self.data[cnt].strip())
100 if not self.kids and s[-1:]=='>':
101 s=s[:-1]+' />'
102 if fancy: s = s + "\n"
103 else:
104 if fancy and not self.data: s = s + (fancy-1) * 2 * ' '
105 s = s + "</" + self.name + ">"
106 if fancy: s = s + "\n"
107 return s
108 - def addChild(self, name=None, attrs={}, payload=[], namespace=None, node=None):
109 """ If "node" argument is provided, adds it as child node. Else creates new node from
110 the other arguments' values and adds it as well."""
111 if namespace: name=namespace+' '+name
112 if node:
113 newnode=node
114 node.parent = self
115 else: newnode=Node(tag=name, parent=self, attrs=attrs, payload=payload)
116 self.kids.append(newnode)
117 return newnode
119 """ Adds some CDATA to node. """
120 self.data.append(ustr(data))
122 """ Removes all CDATA from the node. """
123 self.data=[]
125 """ Deletes an attribute "key" """
126 del self.attrs[key]
128 """ Deletes the "node" from the node's childs list, if "node" is an instance.
129 Else deletes the first node that have specified name and (optionally) attributes. """
130 if type(node)<>type(self): node=self.getTag(node,attrs)
131 self.kids.remove(node)
132 return node
134 """ Returns all node's attributes as dictionary. """
135 return self.attrs
137 """ Returns value of specified attribute. """
138 try: return self.attrs[key]
139 except: return None
141 """ Returns all node's child nodes as list. """
142 return self.kids
144 """ Returns all node CDATA as string (concatenated). """
145 return ''.join(self.data)
147 """ Returns the name of node """
148 return self.name
150 """ Returns the namespace of node """
151 return self.namespace
153 """ Returns the parent of node (if present). """
154 return self.parent
156 """ Return the payload of node i.e. list of child nodes and CDATA entries.
157 F.e. for "<node>text1<nodea/><nodeb/> text2</node>" will be returned list:
158 ['text1', <nodea instance>, <nodeb instance>, ' text2']. """
159 ret=[]
160 for i in range(len(self.kids)+len(self.data)+1):
161 try:
162 if self.data[i]: ret.append(self.data[i])
163 except IndexError: pass
164 try: ret.append(self.kids[i])
165 except IndexError: pass
166 return ret
167 - def getTag(self, name, attrs={}, namespace=None):
168 """ Filters all child nodes using specified arguments as filter.
169 Returns the first found or None if not found. """
170 return self.getTags(name, attrs, namespace, one=1)
172 """ Returns attribute value of the child with specified name (or None if no such attribute)."""
173 try: return self.getTag(tag).attrs[attr]
174 except: return None
176 """ Returns cocatenated CDATA of the child with specified name."""
177 try: return self.getTag(tag).getData()
178 except: return None
192 """ Sets attribute "key" with the value "val". """
193 self.attrs[key]=val
195 """ Sets node's CDATA to provided string. Resets all previous CDATA!"""
196 self.data=[ustr(data)]
198 """ Changes the node name. """
199 self.name = val
201 """ Changes the node namespace. """
202 self.namespace=namespace
204 """ Sets node's parent to "node". WARNING: do not checks if the parent already present
205 and not removes the node from the list of childs of previous parent. """
206 self.parent = node
208 """ Sets node payload according to the list specified. WARNING: completely replaces all node's
209 previous content. If you wish just to add child or CDATA - use addData or addChild methods. """
210 if type(payload) in (type(''),type(u'')): payload=[payload]
211 if add: self.kids+=payload
212 else: self.kids=payload
213 - def setTag(self, name, attrs={}, namespace=None):
214 """ Same as getTag but if the node with specified namespace/attributes not found, creates such
215 node and returns it. """
216 node=self.getTags(name, attrs, namespace=namespace, one=1)
217 if node: return node
218 else: return self.addChild(name, attrs, namespace=namespace)
220 """ Creates new node (if not already present) with name "tag"
221 and sets it's attribute "attr" to value "val". """
222 try: self.getTag(tag).attrs[attr]=val
223 except: self.addChild(tag,attrs={attr:val})
225 """ Creates new node (if not already present) with name "tag" and (optionally) attributes "attrs"
226 and sets it's CDATA to string "val". """
227 try: self.getTag(tag,attrs).setData(ustr(val))
228 except: self.addChild(tag,attrs,payload=[ustr(val)])
230 """ Checks if node have attribute "key"."""
231 return self.attrs.has_key(key)
233 """ Returns node's attribute "item" value. """
234 return self.getAttr(item)
236 """ Sets node's attribute "item" value. """
237 return self.setAttr(item,val)
239 """ Deletes node's attribute "item". """
240 return self.delAttr(item,val)
242 """ Reduce memory usage caused by T/NT classes - use memory only when needed. """
243 if attr=='T':
244 self.T=T(self)
245 return self.T
246 if attr=='NT':
247 self.NT=NT(self)
248 return self.NT
249 raise AttributeError
250
252 """ Auxiliary class used to quick access to node's child nodes. """
253 - def __init__(self,node): self.__dict__['node']=node
259
261 """ Auxiliary class used to quick create node's child nodes. """
264 if isinstance(val,Node): self.node.addChild(attr,node=val)
265 else: return self.node.addChild(attr,payload=[val])
266
267 DBG_NODEBUILDER = 'nodebuilder'
269 """ Builds a Node class minidom from data parsed to it. This class used for two purposes:
270 1. Creation an XML Node from a textual representation. F.e. reading a config file. See an XML2Node method.
271 2. Handling an incoming XML stream. This is done by mangling
272 the __dispatch_depth parameter and redefining the dispatch method.
273 You do not need to use this class directly if you do not designing your own XML handler."""
274 - def __init__(self,data=None,initial_node=None):
275 """ Takes two optional parameters: "data" and "initial_node".
276 By default class initialised with empty Node class instance.
277 Though, if "initial_node" is provided it used as "starting point".
278 You can think about it as of "node upgrade".
279 "data" (if provided) feeded to parser immidiatedly after instance init.
280 """
281 self.DEBUG(DBG_NODEBUILDER, "Preparing to handle incoming XML stream.", 'start')
282 self._parser = xml.parsers.expat.ParserCreate(namespace_separator=' ')
283 self._parser.StartElementHandler = self.starttag
284 self._parser.EndElementHandler = self.endtag
285 self._parser.CharacterDataHandler = self.handle_data
286 self._parser.StartNamespaceDeclHandler = self.handle_namespace_start
287 self.Parse = self._parser.Parse
288
289 self.__depth = 0
290 self._dispatch_depth = 1
291 self._document_attrs = None
292 self._mini_dom=initial_node
293 self.last_is_data = 1
294 self._ptr=None
295 self.namespaces={"http://www.w3.org/XML/1998/namespace":'xml:'}
296 self.xmlns="http://www.w3.org/XML/1998/namespace"
297
298 if data: self._parser.Parse(data,1)
299
301 """ Method used to allow class instance to be garbage-collected. """
302 self._parser.StartElementHandler = None
303 self._parser.EndElementHandler = None
304 self._parser.CharacterDataHandler = None
305 self._parser.StartNamespaceDeclHandler = None
306
308 """XML Parser callback. Used internally"""
309 attlist=attrs.keys()
310 for attr in attlist:
311 sp=attr.rfind(" ")
312 if sp==-1: continue
313 ns=attr[:sp]
314 attrs[self.namespaces[ns]+attr[sp+1:]]=attrs[attr]
315 del attrs[attr]
316 self.__depth += 1
317 self.DEBUG(DBG_NODEBUILDER, "DEPTH -> %i , tag -> %s, attrs -> %s" % (self.__depth, tag, `attrs`), 'down')
318 if self.__depth == self._dispatch_depth:
319 if not self._mini_dom : self._mini_dom = Node(tag=tag, attrs=attrs)
320 else: Node.__init__(self._mini_dom,tag=tag, attrs=attrs)
321 self._ptr = self._mini_dom
322 elif self.__depth > self._dispatch_depth:
323 self._ptr.kids.append(Node(tag=tag,parent=self._ptr,attrs=attrs))
324 self._ptr = self._ptr.kids[-1]
325 if self.__depth == 1:
326 self._document_attrs = attrs
327 ns, name = (['']+tag.split())[-2:]
328 self.stream_header_received(ns, name, attrs)
329 if not self.last_is_data and self._ptr.parent: self._ptr.parent.data.append('')
330 self.last_is_data = 0
331
333 """XML Parser callback. Used internally"""
334 self.DEBUG(DBG_NODEBUILDER, "DEPTH -> %i , tag -> %s" % (self.__depth, tag), 'up')
335 if self.__depth == self._dispatch_depth:
336 self.dispatch(self._mini_dom)
337 elif self.__depth > self._dispatch_depth:
338 self._ptr = self._ptr.parent
339 else:
340 self.DEBUG(DBG_NODEBUILDER, "Got higher than dispatch level. Stream terminated?", 'stop')
341 self.__depth -= 1
342 self.last_is_data = 0
343 if self.__depth == 0: self.stream_footer_received()
344
346 """XML Parser callback. Used internally"""
347 self.DEBUG(DBG_NODEBUILDER, data, 'data')
348 if not self._ptr: return
349 if self.last_is_data:
350 self._ptr.data[-1] += data
351 else:
352 self._ptr.data.append(data)
353 self.last_is_data = 1
354
356 """XML Parser callback. Used internally"""
357 if prefix: self.namespaces[uri]=prefix+':'
358 else: self.xmlns=uri
359 - def DEBUG(self, level, text, comment=None):
360 """ Gets all NodeBuilder walking events. Can be used for debugging if redefined."""
362 """ Returns just built Node. """
363 return self._mini_dom
365 """ Gets called when the NodeBuilder reaches some level of depth on it's way up with the built
366 node as argument. Can be redefined to convert incoming XML stanzas to program events. """
368 """ Method called when stream just opened. """
371
373 """ Converts supplied textual string into XML node. Handy f.e. for reading configuration file.
374 Raises xml.parser.expat.parsererror if provided string is not well-formed XML. """
375 return NodeBuilder(xml).getDom()
376
378 """ Converts supplied textual string into XML node. Survives if xml data is cutted half way round.
379 I.e. "<html>some text <br>some more text". Will raise xml.parser.expat.parsererror on misplaced
380 tags though. F.e. "<b>some text <br>some more text</b>" will not work."""
381 return NodeBuilder(xml).getDom()
382