Created
December 9, 2010 15:44
-
-
Save sgk/734855 to your computer and use it in GitHub Desktop.
Simple to use XML parser
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import xml.sax.handler | |
import xml.sax | |
class Node(object): | |
def __init__(self, name=None, attrs=None): | |
self.name_ = name | |
self.attrs_ = attrs | |
self.dict_ = {} | |
self.str_ = '' | |
def add(self, name, attrs=None): | |
node = Node(name, attrs) | |
self.dict_[name] = self.dict_.get(name, []) + [node] | |
return node | |
def addstring(self, s): | |
self.str_ += s | |
def get(self, name): | |
try: | |
return self.dict_[name][0] | |
except KeyError, IndexError: | |
return None | |
def all(self, name): | |
return self.dict_.get(name, []) | |
def __str__(self): | |
return self.str_ | |
def __unicode__(self): | |
return self.str_ | |
def __getitem__(self, key): | |
return self.dict_[key][0] | |
def __len__(self): | |
return len(self.dict_) | |
def keys(self): | |
return self.dict_.keys() | |
def __getattr__(self, name): | |
try: | |
return self.dict_[name][0] | |
except (KeyError, IndexError): | |
return Node() | |
class Handler(xml.sax.handler.ContentHandler): | |
def __init__(self): | |
self.root_ = Node() | |
self.stack_ = [self.root_] | |
def startElement(self, name, attrs): | |
node = self.stack_[-1].add(name, attrs) | |
self.stack_.append(node) | |
def endElement(self, name): | |
self.stack_.pop() | |
def result(self): | |
return self.root_ | |
def characters(self, content): | |
self.stack_[-1].addstring(content) | |
def parse(fname_or_stream): | |
h = Handler() | |
xml.sax.parse(fname_or_stream, h) | |
return h.result() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment