gistfile1.py

overlaps = []
for postword in post:
    matchcount = 0
    wordmatches = []
    for bncword in bnc:
        if(postword['word'] == bncword['word']):
            bnclogf = math.log(float(bncword['count']))
            wordmatches.append({'word': postword['word'],
                            'postlogf': postword['logf'],
                            'bnclogf': bnclogf,
                            'diff': float(postword['logf']) - bnclogf
                            })
    # find the max of duplicates
    if len(wordmatches) > 0:
        while len(wordmatches) > 1:
            if wordmatches[0]['bnclogf'] > wordmatches[1]['bnclogf']:
                del wordmatches[1]
            else:
                del wordmatches[0]
        overlaps.append(wordmatches[0])