XMLParser._to_bool() - Code Metrics - Inspection of "xml and binary improvement/Cleanup" - FreeOpcUa/python-opcua - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#296)

by Olivier

created 2016-09-01 10:29 UTC

XMLParser._to_bool() A

↳ Parent: Project

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	0
CRAP Score	6

Importance

Changes

Metric	Value
cc	2
dl	0
loc	5
ccs	0
cts	0
cp	0
crap	6
rs	9.4285
c	0
b	0
f	0

"""
parse xml file from opcua-spec
"""
import logging
import re
import sys

import xml.etree.ElementTree as ET


class NodeData(object):

    def __init__(self):
        self.nodetype = None
        self.nodeid = None
        self.browsename = None
        self.displayname = None
        self.symname = None  # FIXME: this param is never used, why?
        self.parent = None
        self.parentlink = None
        self.desc = ""
        self.typedef = None
        self.refs = []
        self.nodeclass = None
        self.eventnotifier = 0

        # variable
        self.datatype = None
        self.rank = -1  # check default value
        self.value = None
        self.valuetype = None
        self.dimensions = None
        self.accesslevel = None
        self.useraccesslevel = None
        self.minsample = None

        # referencetype
        self.inversename = ""
        self.abstract = False 
        self.symmetric = False 

        # datatype
        self.definition = []


class RefStruct(object):

    def __init__(self):
        self.reftype = None
        self.forward = True
        self.target = None


class ExtObj(object):

    def __init__(self):
        self.typeid = None
        self.objname = None
        self.bodytype = None
        self.body = {}


class XMLParser(object):

    def __init__(self, xmlpath, server):
        self.server = server  # POC
        self.logger = logging.getLogger(__name__)
        self._retag = re.compile(r"(\{.*\})(.*)")
        self.path = xmlpath
        self.aliases = {}

        self.tree = ET.parse(xmlpath)
        self.root = self.tree.getroot()
        self.it = None

        self.namespaces = {}
        self._re_nodeid = re.compile(r"^ns=(?P<ns>\d+[^;]*);i=(?P<i>\d+)")

    def __iter__(self):
        nodes = []
        for child in self.root:
            name = self._retag.match(child.tag).groups()[1]
            if name == "Aliases":
                for el in child:
                    self.aliases[el.attrib["Alias"]] = self._get_node_id(el.text)
            elif name == 'NamespaceUris':
                for ns_index, ns_element in enumerate(child):
                    ns_uri = ns_element.text
                    ns_server_index = self.server.register_namespace(ns_uri)
                    self.namespaces[ns_index + 1] = (ns_server_index, ns_uri)
            else:
                node = self._parse_node(name, child)
                nodes.append(node)

        # The ordering of nodes currently only works if namespaces are
        # defined in XML.
        # Also, it is recommended not to use node ids without namespace prefix!
        if self.namespaces:
            nodes = self._sort_nodes_by_parentid(nodes)

        self.it = iter(nodes)
        return self

    def __next__(self):
        while True:
            if sys.version_info[0] < 3:
                child = self.it.next()
            else:
                child = self.it.__next__()
            return child

    def next(self):  # support for python2
        return self.__next__()

    def _sort_nodes_by_parentid(self, nodes):
        """
        Sort the list of nodes according theire parent node in order to respect
        the depency between nodes.

        :param nodes: list of NodeDataObjects
        :returns: list of sorted nodes
        """
        _nodes = list(nodes)
        # list of node ids that are already sorted / inserted
        sorted_nodes_ids = []
        # list of sorted nodes (i.e. XML Elements)
        sorted_nodes = []
        # list of namespace indexes that are relevant for this import
        # we can only respect ordering nodes for namespaces indexes that
        # are defined in the xml file itself. Thus we assume that all other
        # references namespaces are already known to the server and should
        # not create any dependency problems (like "NodeNotFound")
        relevant_namespaces = [str(i[0]) for i in self.namespaces.values()]
        while len(_nodes) > 0:
            pop_nodes = []
            for node in _nodes:
                insert = None
                # Get the node and parent node namespace and id parts
                node_ns, node_id = self._split_node_id(node.nodeid)
                parent_ns, parent_id = self._split_node_id(node.parent)

                # Insert nodes that
                #   (1) have no parent / parent_ns is None (e.g. namespace 0)
                #   (2) ns is not in list of relevant namespaces
                if (parent_ns is None or node_ns not in relevant_namespaces or
                    parent_id is None):
                    insert = 0
                else:
                    # Check if the nodes parent is already in the list of
                    # inserted nodes
                    if node.parent in sorted_nodes_ids:
                        insert = -1

                if insert == 0:
                    sorted_nodes.insert(insert, node)
                    sorted_nodes_ids.insert(insert, node.nodeid)
                    pop_nodes.append(node)
                elif insert == -1:
                    sorted_nodes.append(node)
                    sorted_nodes_ids.append(node.nodeid)
                    pop_nodes.append(node)

            # Remove inserted nodes from the list
            for node in pop_nodes:
                _nodes.pop(_nodes.index(node))
        return sorted_nodes

    def _split_node_id(self, value):
        """
        Split the fq node id into namespace and id part.

        :returns: (namespace, id)
        """
        if not value:
            return (None, value)
        r_match = self._re_nodeid.search(value)
        if r_match:
            return r_match.groups()

        return (None, value)

    def _get_node_id(self, value):
        """
        Check if the nodeid given in the xml model file must be converted
        to a already existing namespace id based on the files namespace uri

        :returns: NodeId (str)
        """
        result = value

        node_ns, node_id = self._split_node_id(value)
        if node_ns:
            ns_server = self.namespaces.get(int(node_ns), None)
            if ns_server:
                result = "ns={};i={}".format(ns_server[0], node_id)
        return result

    def _parse_node(self, name, child):
        """
        Parse a XML node and create a NodeData object.
        """
        obj = NodeData()
        obj.nodetype = name
        for key, val in child.attrib.items():
            self._set_attr(key, val, obj)
        self.logger.info("\n     Parsing node: %s %s", obj.nodeid, obj.browsename)
        obj.displayname = obj.browsename  # give a default value to display name
        for el in child:
            self._parse_tag(el, obj)
        return obj

    def _set_attr(self, key, val, obj):
        if key == "NodeId":
            obj.nodeid = self._get_node_id(val)
        elif key == "BrowseName":
            obj.browsename = val
        elif key == "SymbolicName":
            obj.symname = val
        elif key == "ParentNodeId":
            obj.parent = self._get_node_id(val)
        elif key == "DataType":
            obj.datatype = val
        elif key == "IsAbstract":
            obj.abstract = self._to_bool(val)
        elif key == "Executable":
            obj.executable = self._to_bool(val)
        elif key == "EventNotifier":
            obj.eventnotifier = 1 if val == "1" else 0
        elif key == "ValueRank":
            obj.rank = int(val)
        elif key == "ArrayDimensions":
            obj.dimensions = [int(i) for i in val.split(",")]
        elif key == "MinimumSamplingInterval":
            obj.minsample = int(val)
        elif key == "AccessLevel":
            obj.accesslevel = int(val)
        elif key == "UserAccessLevel":
            obj.useraccesslevel = int(val)
        elif key == "Symmetric":
            obj.symmetric = self._to_bool(val)
        else:
            self.logger.info("Attribute not implemented: %s:%s", key, val)

    def _to_bool(self, val):
        if val in ("False, false"):
            return False
        else:
            return True
    def _parse_tag(self, el, obj):
        tag = self._retag.match(el.tag).groups()[1]

        if tag == "DisplayName":
            obj.displayname = el.text
        elif tag == "Description":
            obj.desc = el.text
        elif tag == "References":
            self._parse_refs(el, obj)
        elif tag == "Value":
            self._parse_value(el, obj)
        elif tag == "InverseName":
            obj.inversename = el.text
        elif tag == "Definition":
            for field in el:
                obj.definition.append(field)
        else:
            self.logger.info("Not implemented tag: %s", el)

    def _parse_value(self, el, obj):
        self.logger.info("Parsing value")
        for val in el:
            self.logger.info("tag %s", val.tag)
            ntag = self._retag.match(val.tag).groups()[1]
            obj.valuetype = ntag
            if ntag in ("Int8", "UInt8", "Int16", "UInt16", "Int32", "UInt32", "Int64", "UInt64"):
                obj.value = int(val.text)
            elif ntag in ("Float", "Double"):
                obj.value = float(val.text)
            elif ntag in ("Boolean"):
                if val.text in ("True", "true", "1", "on", "On"):
                    obj.value = bool(1)
                else:
                    obj.value = bool(0)
            elif ntag in ("ByteString", "String"):
                mytext = val.text
                if mytext is None:  # support importing null strings
                    mytext = ""
                mytext = mytext.replace('\n', '').replace('\r', '')
                # obj.value.append('b"{}"'.format(mytext))
                obj.value = mytext
            elif ntag in ("Guid"):
                self._parse_value(val, obj)
            elif ntag == "ListOfExtensionObject":
                obj.valuetype = "ExtensionObject"
                obj.value = self._parse_list_of_extension_object(el)
            elif ntag == "ListOfLocalizedText":
                obj.valuetype = "LocalizedText"
                obj.value = self._parse_list_of_localized_text(el)
            else:
                self.logger.info("Value type not implemented: '%s', %s", ntag)

    def _get_text(self, el):
        txt = ""
        for text in el.itertext():
            txt += text
        return txt

    def _parse_list_of_localized_text(self, el):
        value = []
        for localized_text_list in el:
            for localized_text in localized_text_list:
                ntag = self._retag.match(localized_text.tag).groups()[1]
                for child in localized_text:
                    ntag = self._retag.match(child.tag).groups()[1]
                    if ntag == 'Text':
                        value.append(self._get_text(child))
        return value

    def _parse_list_of_extension_object(self, el):
        '''
        Parse a uax:ListOfExtensionObject Value
        
        Return an array with a value of each uax:ExtensionObject/*/* (each element is convert to a netry in a dict.
               also the valuetype is returned. The valuetype is  uax:ExtensionObject/*/tag()
        '''
        value = []
        for extension_object_list in el:
            for extension_object in extension_object_list:
                extension_object.find('Body')
                for extension_object_part in extension_object:
                    ext = ExtObj()
                    ntag = self._retag.match(extension_object_part.tag).groups()[1]
                    if ntag == 'TypeId':
                        ntag = self._retag.match(extension_object_part.find('*').tag).groups()[1]
                        ext.typeid = self._get_text(extension_object_part)
                    elif ntag == 'Body':
                        ntag = self._retag.match(extension_object_part.find('*').tag).groups()[1]
                        ext.objname = ntag
                        for body_item in extension_object_part.findall('*/*'):
                            ntag = self._retag.match(body_item.tag).groups()[1]

                            child = body_item.find('*')
                            if child is not None:
                                ext.body[ntag] = self._get_text(child)
                            else:
                                ext.body[ntag] = self._get_text(body_item)
                        value.append(ext)
        return value

    def _parse_refs(self, el, obj):
        for ref in el:
            if ref.attrib["ReferenceType"] == "HasTypeDefinition":
                obj.typedef = self._get_node_id(ref.text)
            elif "IsForward" in ref.attrib and ref.attrib["IsForward"] in ("false", "False"):
                # if obj.parent:
                    # sys.stderr.write("Parent is already set with: "+ obj.parent + " " + ref.text + "\n")
                obj.parent = self._get_node_id(ref.text)
                obj.parentlink = ref.attrib["ReferenceType"]
            else:
                struct = RefStruct()
                if "IsForward" in ref.attrib:
                    struct.forward = ref.attrib["IsForward"]
                struct.target = self._get_node_id(ref.text)
                struct.reftype = ref.attrib["ReferenceType"]
                obj.refs.append(struct)


1		"""
2		parse xml file from opcua-spec
3		"""
4	1	import logging
5	1	import re
6	1	import sys
7
8	1	import xml.etree.ElementTree as ET
9
10
11	1	class NodeData(object):
12
13	1	def __init__(self):
14	1	self.nodetype = None
15	1	self.nodeid = None
16	1	self.browsename = None
17	1	self.displayname = None
18	1	self.symname = None # FIXME: this param is never used, why?
19	1	self.parent = None
20	1	self.parentlink = None
21	1	self.desc = ""
22	1	self.typedef = None
23	1	self.refs = []
24	1	self.nodeclass = None
25	1	self.eventnotifier = 0
26
27		# variable
28	1	self.datatype = None
29	1	self.rank = -1 # check default value
30	1	self.value = None
31	1	self.valuetype = None
32	1	self.dimensions = None
33	1	self.accesslevel = None
34	1	self.useraccesslevel = None
35	1	self.minsample = None
36
37		# referencetype
38	1	self.inversename = ""
39	1	self.abstract = False
40	1	self.symmetric = False
41
42		# datatype
43	1	self.definition = []
44
45
46	1	class RefStruct(object):
47
48	1	def __init__(self):
49		self.reftype = None
50		self.forward = True
51		self.target = None
52
53
54	1	class ExtObj(object):
55
56	1	def __init__(self):
57	1	self.typeid = None
58	1	self.objname = None
59	1	self.bodytype = None
60	1	self.body = {}
61
62	1
63	1	class XMLParser(object):
64	1
65		def __init__(self, xmlpath, server):
66	1	self.server = server # POC
67	1	self.logger = logging.getLogger(__name__)
68	1	self._retag = re.compile(r"(\{.\})(.)")
69		self.path = xmlpath
70	1	self.aliases = {}
71	1
72	1	self.tree = ET.parse(xmlpath)
73		self.root = self.tree.getroot()
74		self.it = None
75	1
76	1	self.namespaces = {}
77	1	self._re_nodeid = re.compile(r"^ns=(?P<ns>\d+[^;]*);i=(?P<i>\d+)")
78
79		def __iter__(self):
80		nodes = []
81	1	for child in self.root:
82	1	name = self._retag.match(child.tag).groups()[1]
83		if name == "Aliases":
84	1	for el in child:
85		self.aliases[el.attrib["Alias"]] = self._get_node_id(el.text)
86		elif name == 'NamespaceUris':
87	1	for ns_index, ns_element in enumerate(child):
88	1	ns_uri = ns_element.text
89	1	ns_server_index = self.server.register_namespace(ns_uri)
90	1	self.namespaces[ns_index + 1] = (ns_server_index, ns_uri)
91	1	else:
92	1	node = self._parse_node(name, child)
93	1	nodes.append(node)
94	1
95	1	# The ordering of nodes currently only works if namespaces are
96		# defined in XML.
97	1	# Also, it is recommended not to use node ids without namespace prefix!
98	1	if self.namespaces:
99	1	nodes = self._sort_nodes_by_parentid(nodes)
100	1
101	1	self.it = iter(nodes)
102	1	return self
103
104	1	def __next__(self):
105		while True:
106	1	if sys.version_info[0] < 3:
107	1	child = self.it.next()
108		else:
109		child = self.it.__next__()
110		return child
111
112		def next(self): # support for python2
113		return self.__next__()
114
115		def _sort_nodes_by_parentid(self, nodes):
116		"""
117		Sort the list of nodes according theire parent node in order to respect
118		the depency between nodes.
119
120		:param nodes: list of NodeDataObjects
121		:returns: list of sorted nodes
122		"""
123		_nodes = list(nodes)
124		# list of node ids that are already sorted / inserted
125		sorted_nodes_ids = []
126		# list of sorted nodes (i.e. XML Elements)
127		sorted_nodes = []
128	1	# list of namespace indexes that are relevant for this import
129	1	# we can only respect ordering nodes for namespaces indexes that
130		# are defined in the xml file itself. Thus we assume that all other
131	1	# references namespaces are already known to the server and should
132		# not create any dependency problems (like "NodeNotFound")
133	1	relevant_namespaces = [str(i[0]) for i in self.namespaces.values()]
134	1	while len(_nodes) > 0:
135	1	pop_nodes = []
136	1	for node in _nodes:
137	1	insert = None
138	1	# Get the node and parent node namespace and id parts
139		node_ns, node_id = self._split_node_id(node.nodeid)
140		parent_ns, parent_id = self._split_node_id(node.parent)
141
142		# Insert nodes that
143		# (1) have no parent / parent_ns is None (e.g. namespace 0)
144		# (2) ns is not in list of relevant namespaces
145		if (parent_ns is None or node_ns not in relevant_namespaces or
146		parent_id is None):
147	1	insert = 0
148	1	else:
149	1	# Check if the nodes parent is already in the list of
150	1	# inserted nodes
151	1	if node.parent in sorted_nodes_ids:
152	1	insert = -1
153	1
154		if insert == 0:
155	1	sorted_nodes.insert(insert, node)
156		sorted_nodes_ids.insert(insert, node.nodeid)
157		pop_nodes.append(node)
158		elif insert == -1:
159		sorted_nodes.append(node)
160	1	sorted_nodes_ids.append(node.nodeid)
161	1	pop_nodes.append(node)
162	1
163		# Remove inserted nodes from the list
164	1	for node in pop_nodes:
165		_nodes.pop(_nodes.index(node))
166	1	return sorted_nodes
167
168		def _split_node_id(self, value):
169		"""
170		Split the fq node id into namespace and id part.
171
172		:returns: (namespace, id)
173		"""
174	1	if not value:
175	1	return (None, value)
176	1	r_match = self._re_nodeid.search(value)
177	1	if r_match:
178	1	return r_match.groups()
179
180		return (None, value)
181	1
182	1	def _get_node_id(self, value):
183		"""
184		Check if the nodeid given in the xml model file must be converted
185		to a already existing namespace id based on the files namespace uri
186
187		:returns: NodeId (str)
188		"""
189		result = value
190
191		node_ns, node_id = self._split_node_id(value)
192		if node_ns:
193		ns_server = self.namespaces.get(int(node_ns), None)
194		if ns_server:
195		result = "ns={};i={}".format(ns_server[0], node_id)
196		return result
197
198		def _parse_node(self, name, child):
199		"""
200		Parse a XML node and create a NodeData object.
201		"""
202		obj = NodeData()
203		obj.nodetype = name
204		for key, val in child.attrib.items():
205		self._set_attr(key, val, obj)
206		self.logger.info("\n Parsing node: %s %s", obj.nodeid, obj.browsename)
207		obj.displayname = obj.browsename # give a default value to display name
208		for el in child:
209		self._parse_tag(el, obj)
210		return obj
211
212		def _set_attr(self, key, val, obj):
213		if key == "NodeId":
214		obj.nodeid = self._get_node_id(val)
215		elif key == "BrowseName":
216		obj.browsename = val
217		elif key == "SymbolicName":
218		obj.symname = val
219		elif key == "ParentNodeId":
220		obj.parent = self._get_node_id(val)
221		elif key == "DataType":
222		obj.datatype = val
223		elif key == "IsAbstract":
224		obj.abstract = self._to_bool(val)
225		elif key == "Executable":
226		obj.executable = self._to_bool(val)
227		elif key == "EventNotifier":
228		obj.eventnotifier = 1 if val == "1" else 0
229		elif key == "ValueRank":
230		obj.rank = int(val)
231		elif key == "ArrayDimensions":
232		obj.dimensions = [int(i) for i in val.split(",")]
233		elif key == "MinimumSamplingInterval":
234		obj.minsample = int(val)
235		elif key == "AccessLevel":
236		obj.accesslevel = int(val)
237		elif key == "UserAccessLevel":
238		obj.useraccesslevel = int(val)
239		elif key == "Symmetric":
240		obj.symmetric = self._to_bool(val)
241		else:
242		self.logger.info("Attribute not implemented: %s:%s", key, val)
243
244		def _to_bool(self, val):
245		if val in ("False, false"):
246		return False
247		else:
248		return True
249		def _parse_tag(self, el, obj):
250		tag = self._retag.match(el.tag).groups()[1]
251
252		if tag == "DisplayName":
253		obj.displayname = el.text
254		elif tag == "Description":
255		obj.desc = el.text
256		elif tag == "References":
257		self._parse_refs(el, obj)
258		elif tag == "Value":
259		self._parse_value(el, obj)
260		elif tag == "InverseName":
261		obj.inversename = el.text
262		elif tag == "Definition":
263		for field in el:
264		obj.definition.append(field)
265		else:
266		self.logger.info("Not implemented tag: %s", el)
267
268		def _parse_value(self, el, obj):
269		self.logger.info("Parsing value")
270		for val in el:
271		self.logger.info("tag %s", val.tag)
272		ntag = self._retag.match(val.tag).groups()[1]
273		obj.valuetype = ntag
274		if ntag in ("Int8", "UInt8", "Int16", "UInt16", "Int32", "UInt32", "Int64", "UInt64"):
275		obj.value = int(val.text)
276		elif ntag in ("Float", "Double"):
277		obj.value = float(val.text)
278		elif ntag in ("Boolean"):
279		if val.text in ("True", "true", "1", "on", "On"):
280		obj.value = bool(1)
281		else:
282		obj.value = bool(0)
283		elif ntag in ("ByteString", "String"):
284		mytext = val.text
285		if mytext is None: # support importing null strings
286		mytext = ""
287		mytext = mytext.replace('\n', '').replace('\r', '')
288		# obj.value.append('b"{}"'.format(mytext))
289		obj.value = mytext
290		elif ntag in ("Guid"):
291		self._parse_value(val, obj)
292		elif ntag == "ListOfExtensionObject":
293		obj.valuetype = "ExtensionObject"
294		obj.value = self._parse_list_of_extension_object(el)
295		elif ntag == "ListOfLocalizedText":
296		obj.valuetype = "LocalizedText"
297		obj.value = self._parse_list_of_localized_text(el)
298		else:
299		self.logger.info("Value type not implemented: '%s', %s", ntag)
300
301		def _get_text(self, el):
302		txt = ""
303		for text in el.itertext():
304		txt += text
305		return txt
306
307		def _parse_list_of_localized_text(self, el):
308		value = []
309		for localized_text_list in el:
310		for localized_text in localized_text_list:
311		ntag = self._retag.match(localized_text.tag).groups()[1]
312		for child in localized_text:
313		ntag = self._retag.match(child.tag).groups()[1]
314		if ntag == 'Text':
315		value.append(self._get_text(child))
316		return value
317
318		def _parse_list_of_extension_object(self, el):
319		'''
320		Parse a uax:ListOfExtensionObject Value
321
322		Return an array with a value of each uax:ExtensionObject// (each element is convert to a netry in a dict.
323		also the valuetype is returned. The valuetype is uax:ExtensionObject/*/tag()
324		'''
325		value = []
326		for extension_object_list in el:
327		for extension_object in extension_object_list:
328		extension_object.find('Body')
329		for extension_object_part in extension_object:
330		ext = ExtObj()
331		ntag = self._retag.match(extension_object_part.tag).groups()[1]
332		if ntag == 'TypeId':
333		ntag = self._retag.match(extension_object_part.find('*').tag).groups()[1]
334		ext.typeid = self._get_text(extension_object_part)
335		elif ntag == 'Body':
336		ntag = self._retag.match(extension_object_part.find('*').tag).groups()[1]
337		ext.objname = ntag
338		for body_item in extension_object_part.findall('/'):
339		ntag = self._retag.match(body_item.tag).groups()[1]
340
341		child = body_item.find('*')
342		if child is not None:
343		ext.body[ntag] = self._get_text(child)
344		else:
345		ext.body[ntag] = self._get_text(body_item)
346		value.append(ext)
347		return value
348
349		def _parse_refs(self, el, obj):
350		for ref in el:
351		if ref.attrib["ReferenceType"] == "HasTypeDefinition":
352		obj.typedef = self._get_node_id(ref.text)
353		elif "IsForward" in ref.attrib and ref.attrib["IsForward"] in ("false", "False"):
354		# if obj.parent:
355		# sys.stderr.write("Parent is already set with: "+ obj.parent + " " + ref.text + "\n")
356		obj.parent = self._get_node_id(ref.text)
357		obj.parentlink = ref.attrib["ReferenceType"]
358		else:
359		struct = RefStruct()
360		if "IsForward" in ref.attrib:
361		struct.forward = ref.attrib["IsForward"]
362		struct.target = self._get_node_id(ref.text)
363		struct.reftype = ref.attrib["ReferenceType"]
364		obj.refs.append(struct)
365

FreeOpcUa / python-opcua

Pull Request — master (#296)

XMLParser._to_bool() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like