def tokens(cls, selector):
tokens = List()
prev = None
for op, tag in ElementPath.xpath_tokenizer(selector):
if op:
token = Op(cls.unquote_attribute(op) if prev == "=" else op)
else:
token = Tag(cls.unquote_attribute(tag) if prev == "=" else tag)
tokens.append(token)
prev = token
return tokens
def tokens(cls, selector):
tokens = List()
prev = None
for op, tag in ElementPath.xpath_tokenizer(selector):
if op:
token = Op(cls.unquote_attribute(op) if prev == '=' else op)
else:
token = Tag(cls.unquote_attribute(tag) if prev == '=' else tag)
tokens.append(token)
prev = token
return tokens
def xpath_tokenizer(p):
"""
Test the XPath tokenizer.
>>> # tests from the xml specification
>>> xpath_tokenizer("*")
['*']
>>> xpath_tokenizer("text()")
['text', '()']
>>> xpath_tokenizer("@name")
['@', 'name']
>>> xpath_tokenizer("@*")
['@', '*']
>>> xpath_tokenizer("para[1]")
['para', '[', '1', ']']
>>> xpath_tokenizer("para[last()]")
['para', '[', 'last', '()', ']']
>>> xpath_tokenizer("*/para")
['*', '/', 'para']
>>> xpath_tokenizer("/doc/chapter[5]/section[2]")
['/', 'doc', '/', 'chapter', '[', '5', ']', '/', 'section', '[', '2', ']']
>>> xpath_tokenizer("chapter//para")
['chapter', '//', 'para']
>>> xpath_tokenizer("//para")
['//', 'para']
>>> xpath_tokenizer("//olist/item")
['//', 'olist', '/', 'item']
>>> xpath_tokenizer(".")
['.']
>>> xpath_tokenizer(".//para")
['.', '//', 'para']
>>> xpath_tokenizer("..")
['..']
>>> xpath_tokenizer("../@lang")
['..', '/', '@', 'lang']
>>> xpath_tokenizer("chapter[title]")
['chapter', '[', 'title', ']']
>>> xpath_tokenizer("employee[@secretary and @assistant]")
['employee', '[', '@', 'secretary', '', 'and', '', '@', 'assistant', ']']
>>> # additional tests
>>> xpath_tokenizer("{http://spam}egg")
['{http://spam}egg']
>>> xpath_tokenizer("./spam.egg")
['.', '/', 'spam.egg']
>>> xpath_tokenizer(".//{http://spam}egg")
['.', '//', '{http://spam}egg']
"""
out = []
for op, tag in ElementPath.xpath_tokenizer(p):
out.append(op or tag)
return out
def _structure_parse_root_path(self, _root_path):
"""Extract information from the root path to get sufficient information
for specifying the destination and creating new nodes. \n
* If many XPaths in the string, it uses the first one
* Parses the root node name
* Parses the row parent node path
* Parses the row node name"""
if _root_path == "":
raise Exception("_parse_root_path: Root path cannot be empty")
# Parse XPath tokens
_root_xpath_tokens = list(_elementpath.xpath_tokenizer(self.rows_xpath))
# Use only first path for destination
# TODO: This part is never used, why?
try:
_first_splitter = _root_xpath_tokens.index(('', '|'))
print("_parse_root_path: Multiple XPaths, using first.")
_split_xpath = "".join((_a[0] + _a[1] for _a in _root_xpath_tokens[0:_first_splitter]))
except ValueError:
_split_xpath = self.rows_xpath
# Get the root node name
if _root_xpath_tokens[0][0] == "/":
_root_node_name = _root_xpath_tokens[1][1]
else:
raise Exception(
"_parse_root_path: It is necessary to have an absolute (\"/node\") top node name in the XPath")
print("_root_node_name=" + str(_root_node_name))
# Get the row node name
_row_node_name_idx = find_previous_match(_root_xpath_tokens, len(_root_xpath_tokens) - 1, ("/", ""))
if _row_node_name_idx == -1:
raise Exception("_parse_root_path: Cannot find start of condition.\nXPath = " + _root_path)
else:
_row_node_name = _root_xpath_tokens[_row_node_name_idx + 1][1]
if _row_node_name_idx == 0:
raise Exception("_parse_root_path: The row node cannot be the root node.\nXPath = " + _root_path)
print("_structure_row_node_name=" + str(_row_node_name))
# Move on backward, what's left is the path to the parent of the row node, we need it to be able
# to create rows, as the path may not return a node to find a parent from
_row_node_parent_xpath = "".join((_a[0] + _a[1] for _a in _root_xpath_tokens[0:_row_node_name_idx]))
print("_row_node_parent_xpath=" + str(_row_node_parent_xpath))
return _root_node_name, _row_node_name, _row_node_parent_xpath
def tokens(cls, selector):
"""
>>> xpath_tokenizer('resource-lists')
['resource-lists']
>>> xpath_tokenizer('list[@name="friends"]')
['list', '[', '@', 'name', '=', 'friends', ']']
We cannot properly tokenize an URI like this :(
>>> uri_ugly = 'external[@anchor="http://xcap.example.org/resource-lists/users/sip:[email protected]/index/~~/resource-lists/list[@name="mkting"]"]'
>>> len(xpath_tokenizer(uri_ugly)) # expected 7
10
To feed such URI to this function, replace quote \" with "
>>> uri_nice = 'external[@anchor="http://xcap.example.org/resource-lists/users/sip:[email protected]/index/~~/resource-lists/list[@name="mkting"]"]'
>>> len(xpath_tokenizer(uri_nice)) # expected 7
7
"""
def unquote_attr_value(s):
# XXX currently equivalent but differently encoded URIs won't be considered equal (", etc.)
if len(s) > 1 and s[0] == s[-1] and s[0] in '"\'':
return s[1:-1]
raise NodeParsingError
tokens = List()
prev = None
for op, tag in ElementPath.xpath_tokenizer(selector):
if prev == '=':
unq = unquote_attr_value
else:
unq = lambda x:x
if op:
x = Op(unq(op))
else:
x = Tag(unq(tag))
tokens.append(x)
prev = x
return tokens
def tokens(cls, selector):
"""
>>> xpath_tokenizer('resource-lists')
['resource-lists']
>>> xpath_tokenizer('list[@name="friends"]')
['list', '[', '@', 'name', '=', 'friends', ']']
We cannot properly tokenize an URI like this :(
>>> uri_ugly = 'external[@anchor="http://xcap.example.org/resource-lists/users/sip:[email protected]/index/~~/resource-lists/list[@name="mkting"]"]'
>>> len(xpath_tokenizer(uri_ugly)) # expected 7
10
To feed such URI to this function, replace quote \" with "
>>> uri_nice = 'external[@anchor="http://xcap.example.org/resource-lists/users/sip:[email protected]/index/~~/resource-lists/list[@name="mkting"]"]'
>>> len(xpath_tokenizer(uri_nice)) # expected 7
7
"""
def unquote_attr_value(s):
# XXX currently equivalent but differently encoded URIs won't be considered equal (", etc.)
if len(s) > 1 and s[0] == s[-1] and s[0] in '"\'':
return s[1:-1]
raise NodeParsingError
tokens = List()
prev = None
for op, tag in ElementPath.xpath_tokenizer(selector):
if prev == '=':
unq = unquote_attr_value
else:
unq = lambda x: x
if op:
x = Op(unq(op))
else:
x = Tag(unq(tag))
tokens.append(x)
prev = x
return tokens
def _structure_create_xpath_nodes(self, _node, _xpath):
"""Used an xpath to create nodes that match the path and its conditions(names, attributes and so forth)"""
print("_create_xpath_nodes: " + str(_xpath))
_curr_node = _node
# Break up the string in its tokens
_tokens = list(_elementpath.xpath_tokenizer(_xpath))
print(str(_tokens))
# Iterate through tokens, until we have gone through the entire XPath.
_token_idx = 0
# But first, move past any root reference and conditions.
# TODO: This is a bit ugly, perhaps.
if len(_tokens) > 1 and (_tokens[0][0] == "/") and (_tokens[1][1] == _node.tag):
print("_create_xpath_nodes: Ignoring root node path and condition.")
_token_idx += 1
while _token_idx < len(_tokens) and _tokens[_token_idx][0] != "/":
_token_idx += 1
while _token_idx < len(_tokens):
print("_tokens[" + str(_token_idx) + "][0]:" + str(_tokens[_token_idx][0]))
# Is this a new level?
if _tokens[_token_idx][0] == "/":
# Then the next is the name of the node
_token_idx += 1
if _tokens[_token_idx][0] == "":
_next_name = _tokens[_token_idx][1]
# Is the next token a condition?
if _token_idx + 1 < len(_tokens) and _tokens[_token_idx + 1][0] == "[":
# It was, move on
_token_idx += 1
# Find ending of condition
_end_cond_idx = self.find_next_match(_tokens, _token_idx, ("]", ""))
if _end_cond_idx == -1:
raise Exception("_create_xpath_nodes: Cannot find end of condition.\nXPath = " + _xpath)
# Create relative path
_check_path = "".join((_a[0] + _a[1] for _a in _tokens[_token_idx - 1:_end_cond_idx + 1]))
# Then check if it exists
print("_check_path:" + str(_check_path))
_found_nodes = _curr_node.xpath(_check_path)
# Node found, move on
if _found_nodes and len(_found_nodes) == 1:
_token_idx += 1
_curr_node = _found_nodes[0]
else:
# If not found create node
print("add node: " + str(_next_name))
_curr_node = SubElement(_curr_node, _next_name)
# If they can be discerned(@id=value), add attributes
if _tokens[_token_idx + 3][0] == "=" and _tokens[_token_idx + 1][0] == "@":
print("set attribute to satisfy this path: " + str(_check_path))
_curr_node.set(_tokens[_token_idx + 2][1], str(_tokens[_token_idx + 4][0]).strip("\"'"))
_token_idx += 5
else:
if isinstance(_curr_node, str):
raise Exception(
"_structure_create_xpath_nodes - Internal error: Node is a string, missing node ref, "
"was _add_node_ref used? \nData :" + _curr_node + " | " + _next_name + " | " + _xpath +
" | " + str(_tokens))
_found_nodes = _curr_node.xpath(_next_name)
# Node found, move on
if len(_found_nodes) == 1:
_curr_node = _found_nodes[0]
else:
print("Create new node: " + str(_next_name))
_curr_node = SubElement(_curr_node, _next_name)
_token_idx += 1
return _curr_node