Python how to get variable from between multiple occurrences in string

Question

Say I had an input file (temp.tmpl) that looked like this:

PTF @
ARB @ C @ @ A @ @ C @
OSN @ B @ @ A @ 
SDA @ B @
CPN 3.23
SNL 3.26 

And in some other file (candidate.txt):

A 3.323 B 4.325 C 6.32 D 723 E 8 F 9 G 1.782
H 7
I 4
J 9
K 10

And I wanted to replace A, B , and C with their assigned values. The way this needs to be accomplished for my task is by finding the variables A, B, and C by looking for @ @...Then knowing this is obviously a varible. Then replacing them. This is what I have tried:

reader = open('candidate.txt', 'r')
out = open('output.txt', 'w')

dictionary = dict()
for line in reader.readlines():
    pairs = line.split()
    for variable, value in zip(pairs[::2],pairs[1::2]):
        dictionary[variable] = value

#Now to open the template file
template = open('temp.tmpl', 'r')
for line1 in template:
    if line1[1]:
        confirm = line1.split(' ')[0].lower()
        symbol = line1.split(' ')[1]

        if confirm == 'ptf':
            next(template)

        elif symbol in line1:

            start = line1.find(symbol)+len(symbol)
            end = line1[start:].find(symbol)
            variable = line1[start:start + end].strip()
            print variable

And I can not seem to figure out how to handle the lines with multiple sets of variables.
Thank you so much in advance.

Answer 1

Using an re? The question was altered, and here is my modified solution:

import re

# Create translation dictionary
codes = re.split(r'\s',open('candidate.txt').read())
trans = dict(zip(codes[::2], codes[1::2]))

outfh = open('out.txt','w')
infh  = open('data.txt')

# First line contains the symbol, but has a trailing space!
symbol = re.sub(r'PTF (.).*',r'\1', infh.readline()[:-1])

for line in infh:
    line = re.sub('\\'+ symbol + r' ([ABC]) ' + '\\' + symbol,
               lambda m: '%s %s %s' % (symbol,trans[m.groups()[0]],symbol),
               line)
    outfh.write(line) 

outfh.close()

The dict using two zips is a trick to create a dictionary from a [key,value,key,value,...] list

trans is a dictionary with the names and their respective values.
r'@ ([ABC]) @' captures either A or B or C within the @ signs The lambda function is passed a match object, on which we are calling the groups() method. This returns a tuple of the matching parentheses groups, in this case either A or B or C. We use this for the key to the dictionary trans, and hence replace it with the value.