gistfile1.txt

#!/usr/bin/python3

import re
import datetime

def title(s,d):
    if datetime.datetime.strptime(d,'%Y%m%d') < datetime.datetime(2003, 1, 1, 0, 0):
        return 'do_not_file','do_not_file'

    ret = re.sub('[^0-9a-zA-Z]+', ' ', s)

    wordlist = (
        'mri research',
        'research',
        'converted',
        'mclean',
        'development',
        'investigators',
        'investigator',
        'run this one',
        'archived protocols',
        'archivedprotocols',
    )
    changed = False
    for word in wordlist:
        rpl = re.compile(re.escape(word), re.IGNORECASE)
        old = ret
        ret = rpl.sub('', ret)
        if old != ret:
            changed = True


    ret = re.sub('\s+',' ', ret)
    ret = re.sub('^\s','',ret)
    ret = re.sub('\s$','',ret)

    if ret == '':
        return 'do_not_file', 'do_not_file'

    special = ('EAOLSON', 'Silveri', 'Forester', 'Frederic', 'Ongur')
    if not changed and ret not in special:
        return 'do_not_file', 'do_not_file'
    
    if s.startswith('3P') or s.startswith('3T') or 'clinical' in s.lower():
        return 'do_not_file', 'do_not_file'


    return (ret, re.sub(' ','_',ret).lower())

# print ('Most recently used\toriginal StudyDescription\tProject')
with open("studynames-mostrecentdate.txt") as f:
    for line in f:
        d,raw = line.split('\t')
        raw = raw.rstrip()
        ret = title(raw,d)
        l= [ret[1], ret[0], raw]
        if not ret[0] == 'do_not_file':
            print(*l, sep='\t')