fix.py

import sys
import re

count = 0    # count the entries and use it as an index value

with open(sys.argv[2], "w") as outfile:
   with open(sys.argv[1], "r") as infile:
      for line in infile:
         #line = line.replace('~', ' ')
         line = re.sub(r'[\{\}\^\\]', '', line)
         line = re.sub(r'\\\'', '', line)
         match = re.match(r'([ ]*)', line)
         if match is not None and len(match.groups()) > 0 and len(match.groups()[0]) == 4:
            # This is the start of a record
            line = '      - id: ' + line.strip().replace(':','').replace('-','') + '\n'
         outfile.write(line)


#          if line[0] == '-':
#             # begining of record
#             # Always seem to be `- AUTHOR:`
#             # Replace with `- bib:` and move author to the next line
#             outfile.write('  - bib: "'+str(count)+'"\n')
#             outfile.write('    '+line[1:].strip()+'\n')
#             count = count + 1
#          elif line.strip()[0] == '-':
#             # This line is indented and has a dash
#             # Assume it's a name
#             # Deal with bibtex inverted names
#             # Remove tildes
#             names = line.strip()[1:].split(',')
#             name = ''
#             if len(names) > 1:
#                # flip the order
#                name = ' '.join([names[1].strip(), names[0].strip()])
#             else:
#                name = names[0]
#             outfile.write('      - name: "'+name.strip()+'"\n')
#          else:
#             # This line should be fine as is
# #            if line.count(':') > 1:
# #               line = re.sub(r'([\s]+[A-Z]+:\s)([\w\s:\-\(\),]+\Z)', r'\1"\2"', line[:-1])
# #            else:
#             line = line[:-1]
#             outfile.write('  '+line+'\n')