Modern pythonic solution to http://code.google.com/edu/languages/google-python-class/exercises/baby-names.html

babynames.py

#!/usr/bin/python
import sys
import re

def extract_names(path):
    year = None
    ranks = {}
    def add_rank(name, rank):
        # add name to ranks if its not already present or if its smaller
        if name not in ranks or rank < ranks[name]:
            ranks[name] = rank
    with open(path) as f:
        for line in f:
            m = re.search(r'Popularity in (\d{4})', line)
            if m:
                year = m.group(1)
                continue
            m = re.search(
                r'<td>(\d+)</td><td>([^<]+)</td><td>([^<]+)</td>', line)
            if m:
                rank, male_name, female_name = m.groups()
                rank = int(rank)
                add_rank(male_name, rank)
                add_rank(female_name, rank)
    return [year] + sorted("%s %s" % t for t in ranks.items())
    
def main(args):
    summarize = '--summaryfile' in args
    if summarize:
        args.remove('--summaryfile')
    for arg in args:
        summary = '\n'.join(extract_names(arg))
        if summarize:
            with open(arg + '.summary', 'w') as f:
                f.write(summary + '\n')
        else:
            print summary

if __name__ == '__main__':
    main(sys.argv[1:])

Should probably not continue to search for year once its been found.