looking for nodes based on user input in python3 - python-3.x

Question

Code works well when I hardcode the nodes (e.g. node1), but not when I use user input - it always returns 0 instead of counting the numbers which are "node3". Here is the page I am using http://py4e-data.dr-chuck.net/comments_678016.xml - node1 = comments, node2 = comment, node3= count. Any suggestions?
import urllib.request, urllib.parse, urllib.error
import xml.etree.ElementTree as ET
import ssl
ctx = ssl.create_default_context()
ctx.check_hostname = False
ctx.verify_mode = ssl.CERT_NONE
url = input ("Which url?\n")
node1 = input ("Enter node1- ")
node2 = input ("Enter node2- ")
node3 = input ("Enter node3- ")
count = 0
try:
html = urllib.request.urlopen(url, context=ctx).read()
tree = ET.fromstring(html)
x = tree.findall(node1/node2)
for item in x:
c = int(item.find(node3).text)
count = count + c
print(count)
except:
print("Please only input complete urls")

Answer 1

Putting aside the user input angle, if you want "to sum up all numbers under "count"", change your xpath expression to
x = tree.findall('.//comment/count')
and then either do it the long way (which I personally prefer):
total = 0
for count in x:
total += int(count.text)
or use list comprehensions:
sum([int(count.text) for count in x])
In either case, the output is
2348

Answer 2

Found out what the mistake was - needed to concatenate strings:
x = tree.findall(node1 + "/" + node2)
import urllib.request, urllib.parse, urllib.error
import xml.etree.ElementTree as ET
import ssl
ctx = ssl.create_default_context()
ctx.check_hostname = False
ctx.verify_mode = ssl.CERT_NONE
url = input("Which url?\n")
node1 = input("Enter node1- ")
node2 = input("Enter node2- ")
node3 = input("Enter node3- ")
count = 0
try:
html = urllib.request.urlopen(url, context=ctx).read()
tree = ET.fromstring(html)
x = tree.findall(node1 + "/" + node2)
for item in x:
count += int(item.find(node3).text)
print(count)
except:
print("Please only input complete urls")