Question

While going through one of the problems in Python Challenge, I am trying to solve it as follows:

Read the input in a text file with characters as follows:

DQheAbsaMLjTmAOKmNsLziVMenFxQdATQIjItwtyCHyeMwQTNxbbLXWZnGmDqHhXnLHfEyvzxMhSXzd
BEBaxeaPgQPttvqRvxHPEOUtIsttPDeeuGFgmDkKQcEYjuSuiGROGfYpzkQgvcCDBKrcYwHFlvPzDMEk
MyuPxvGtgSvWgrybKOnbEGhqHUXHhnyjFwSfTfaiWtAOMBZEScsOSumwPssjCPlLbLsPIGffDLpZzMKz
jarrjufhgxdrzywWosrblPRasvRUpZLaUbtDHGZQtvZOvHeVSTBHpitDllUljVvWrwvhpnVzeWVYhMPs
kMVcdeHzFZxTWocGvaKhhcnozRSbWsIEhpeNfJaRjLwWCvKfTLhuVsJczIYFPCyrOJxOPkXhVuCqCUgE
luwLBCmqPwDvUPuBRrJZhfEXHXSBvljqJVVfEGRUWRSHPeKUJCpMpIsrV.......

What I need is to go through this text file and pick all lower case letters that are enclosed by only three upper-case letters on each side.

The python script that I wrote to do the above is as follows:

import re

pattern = re.compile("[a-z][A-Z]{3}([a-z])[A-Z]{3}[a-z]")
f = open('/Users/Dev/Sometext.txt','r')
for line in f:
    result = pattern.search(line)
    if result:
       print result.groups()

 f.close()

The above given script, instead of returning the capture(list of lower case characters), returns all the text blocks that meets the regular expression criteria, like

aXCSdFGHj
vCDFeTYHa
nHJUiKJHo
.........
.........

Can somebody tell me what exactly I am doing wrong here? And instead of looping through the entire file, is there an alternate way to run the regular expression search on the entire file?

Thanks

Answer 1

Change result.groups() to result.group(1) and you will get just the single letter match.

A second problem with your code is that it will not find multiple results on one line. So instead of using re.search you’ll need re.findall or re.finditer. findall will return strings or tuples of strings, whereas finditer returns match objects.

Here’s where I approached the same problem:

import urllib
import re    

pat = re.compile('[a-z][A-Z]{3}([a-z])[A-Z]{3}[a-z]')
print ''.join(pat.findall(urllib.urlopen(
    "http://www.pythonchallenge.com/pc/def/equality.html").read())) 

Note that re.findall and re.finditer return non-overlapping results. So when using the above pattern with re.findall searching against string 'aBBBcDDDeFFFg', your only match will be 'c', but not 'e'. Fortunately, this Python Challenge problem contains no such such examples.