A simple Newick tree parser

newick_parser.py

class Newick(object):


  def __init__(self, newick_file):
	
		f = open(newick_file, 'r')
		t = f.read().split(';')[0]
		t.replace('"', "'")
		t = t.splitlines()
		self.t = ''.join(t)
		self.node_counter = 0
		self.tree = {}
		self.parseNewick(self.t)
	
		
	def _listChildren(self, nt):

		children = []
		sibling = ''
		scount = 0
		openpars = 0
		closepars = 0
		quote = 0
	
		for s in nt:
			scount += 1
			
			if s is '(':
				openpars += 1
			
			if s is ')':
				closepars += 1
				
			if s is "'":
				quote += 1
			
			if s is ',':
				if quote % 2 == 0:
					if openpars == 0 and closepars == 0:
						children.append(sibling)
						sibling = ''
		
					elif openpars == closepars:
						children.append(sibling)
						sibling = ''
						openpars = 0
						closepars = 0
			
					elif openpars > closepars:
						sibling += s
				else:
					sibling += s
			else:
				sibling += s
			
			if scount == len(nt):
				children.append(sibling)
		
		return children
	

	def parseNewick(self, t, parent=None):
	
		self.node_counter += 1
		node_id = self.node_counter
	
		if '):' in t:
			fpstart = t.find('(')
			fpend = -1
			rcount = -1
			
			while fpend == -1:
				rcount += -1
				fpend = t.find(')', rcount)
				
			node = t[fpend + 1:]
			node_type = 'internal'
			
			nt = t[fpstart+1:fpend]
			children = self._listChildren(nt)
			for child in children:
				self.parseNewick(child, parent=node_id)
				
		else:
			node = t
			node_type = 'terminal'
			
			
		if ':' in node and len(node.split(':')) == 2:
			name, length = node.split(':')
		else:
			name = node
			length = ''
		self.tree[node_id] = {
			'name': name.replace("'", ""), 
			'length': length,
			'type': node_type,
			'parent': parent
		}