Search code examples
pythonhtmlwebcgixmlrpclib

<type 'exceptions.NameError'>: global name '..' is not defined


I'm creating a python web interface. You complete the form, it sends it to a background python function, stuff happens.

I'm having trouble with one of these background functions, it's stumped me because it's not my first time creating a quick project like this, and I haven't encountered this error before.

I have three files: commands.py, header.py and product.py

product.py looks like this:

#!/usr/bin/python
import sys
import os
import cgitb
cgitb.enable()

sys.path.append(os.path.abspath("/var/www/cgi-bin/includes/"))
from header import *

page = get_page_name(sys.argv[0])

printDefaultPage(0,page)

and this is the route it takes through header.py:

def printDefaultPage(error,page): #this prints the page
        printHTTPHeaders() #print headers
        printHeader(page) #print header.inc
        printPage(page)
        printFooter() #print footer.inc

the above function prints out the headers, the head, the body of the page and the footer, all by calling other functions.

def printPage(page):
        if page == "product":
                print """<form method='post' id="register_form" name="client_frm"></br>
                <label for="id">Enter the subscription to use:</label>"""
                list_subscriptions()
                print """<a href="/cgi-bin/index.py"><input class="login" type="button" value="Back"></a>
                </form>"""

as you can see, printPage calls a function, list_subscriptions() which looks a bit like this:

def list_subscriptions():
        subscriptions=get_subscriptions()
        print """<select name = "sub">"""
        for s in subscriptions:
                id = s
                price = subscriptions.get(str(s)).get('price')
                scans = subscriptions.get(str(s)).get('scans_per_month')
                print """<option value ="""+str(id)+"""">"""+str(price)+""" - """+str(scans)+"""</option>"""
        print "</select>"

this in turn calls get_subscriptions, which exists in a file called commands, which is imported at the top:

#!/usr/bin/python
from commands import *

for completeness, here is get_subscriptions:

def get_subscriptions():
        subscriptions=proxy.get_subscriptions()
        return subscriptions

and on the other end:

def get_subscriptions(self):
        subscriptions=self.store.find(Subscription)
        subs = {}
        for sub in subscriptions:
                subs[str(sub.id)] = {}
                subs[str(sub.id)]['scans_per_month'] = sub.scans_per_month
                subs[str(sub.id)]['price'] = int(sub.price)
        return subs

get_subscriptions() returns a dictionary. The more baffling thing is when I run printPage("product") in the python command line, it works.

Python 2.7.6 (default, Mar 22 2014, 22:59:56) 
[GCC 4.8.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> from header import *
>>> printPage("product")
<form method='post' id="register_form" name="client_frm"></br>
                <label for="id">Enter the subscription to use:</label>
<select name = "sub">
<option value =1">19 - 20</option>
<option value =3">59 - 100</option>
<option value =2">39 - 50</option>
<option value =5">0 - 0</option>
<option value =4">0 - 1000</option>
</select>
<a href="/cgi-bin/index.py"><input class="login" type="button" value="Back"></a>
            </form>

Yet when I go to product.py in the browser expecting to see a nice select box instead I get this gross error:

 /var/www/cgi-bin/includes/header.py in list_subscriptions()
      7 
      8 def list_subscriptions():
=>    9         subscriptions=get_subscriptions()
     10         print """<select name = "sub">"""
     11         for s in subscriptions:
subscriptions undefined, get_subscriptions undefined

edit: the files are all chmod to 755, and are all owned by www-data.

edit: in the python command line printDefaultPage(0,"product") returns this:

>>> printDefaultPage(0,"product")
Content-Type: text/html
Vary: *
Title: Encriptor
Cache-Control: no-cache, no store, must-revalidate

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" + "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
        <html xmlns="http://www.w3.org/1999/xhtml">
        <head>
        <title>Encriptor Scanner</title>
        <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
        <meta name="copyright" content="encription.co.uk" />
        <meta name="language" content="en" />
        <link rel="shortcut icon" href="/css/images/favicon.ico" />
        <link href="/css/style.css" rel="stylesheet" type="text/css" />
<body>
 <div class="wrap" 
><div id="header">
        <p id="logname"></p>
    <img src="/images/raptor-logo.png" border="0" 
id="logo-main"
/></div>
<ul class="menu">
</ul>
<div id="content">
<form method='post' id="register_form" name="client_frm"></br>
            <label for="id">Enter the subscription to use:</label>
<select name = "sub">
<option value =1">19 - 20</option>
<option value =3">59 - 100</option>
<option value =2">39 - 50</option>
<option value =5">0 - 0</option>
<option value =4">0 - 1000</option>
</select>
<a href="/cgi-bin/index.py"><input class="login" type="button" value="Back"></a>
        </form>
</div>
        <p id="spacer">&nbsp;</p>
        </div>
        </div>  
        </br>
        <center><div id="footerlog">Encryptor v3 <span id="dot"$bull;</span>Copyright &copy; 2015 <a href="https://www.encription.co.uk" target="_ne$
        </div></center>
        </body>
        </html>

which is as expected, however is not working in the browser.


Solution

  • I've found the issue! Giving up on this I started working on another page, however soon the same problems began to surface.

    commands.py is already used by python, meaning I was importing the wrong module all along. Oopsie!