Python Regex (?x) fail

gistfile1.py

from nltk import regexp_tokenize

regex1 = r'''(?x)
  (?:H|S)\. ?(?:(?:J|R)\. )?(?:Con\. )?(?:Res\. )?\d+     # Bills
| ([A-Z]\.)+                                              # Abbreviations (U.S.A., etc.)
| ([A-Z]+\&[A-Z]+)                                        # Internal ampersands (AT&T, etc.)
| (Mr\.|Dr\.|Mrs\.|Ms\.)                                  # Mr., Mrs., etc.
| \d*\.\d+                                                # Numbers with decimal points.
| \d\d?:\d\d                                              # Times.
| \$?[,0-9]+                                              # Numbers with thousands separators.
| (((a|A)|(p|P))\.(m|M)\.)                                # a.m., p.m., A.M., P.M.
| \w+((-|')\w+)*                                          # Words with optional internal hyphens.
| \$?\d+(\.\d+)?%?                                        # Currency and percentages.
| \.\.\.                                                  # Ellipsis
| [][.,;"'?():-_`]
'''

regex2 = r'''(?x)(?:H|S)\. ?(?:(?:J|R)\. )?(?:Con\. )?(?:Res\. )?\d+|([A-Z]\.)+|([A-Z]+\&[A-Z]+)|(Mr\.|Dr\.|Mrs\.|Ms\.)|\d*\.\d+|\d\d?:\d\d|\$?[,0-9]+|(((a|A)|(p|P))\.(m|M)\.)|\w+((-|')\w+)*|\$?\d+(\.\d+)?%?|\.\.\.|[][.,;"'?():-_`]'''

regex3 = r'''(?:H|S)\. ?(?:(?:J|R)\. )?(?:Con\. )?(?:Res\. )?\d+|([A-Z]\.)+|([A-Z]+\&[A-Z]+)|(Mr\.|Dr\.|Mrs\.|Ms\.)|\d*\.\d+|\d\d?:\d\d|\$?[,0-9]+|(((a|A)|(p|P))\.(m|M)\.)|\w+((-|')\w+)*|\$?\d+(\.\d+)?%?|\.\.\.|[][.,;"'?():-_`]'''

print regexp_tokenize('blah H.R. 2354 blah', regex1)
# prints ['blah', 'H.R.', '2354', 'blah']

print regexp_tokenize('blah H.R. 2354 blah', regex2)
# prints ['blah', 'H.R.', '2354', 'blah']

print regexp_tokenize('blah H.R. 2354 blah', regex3)
# prints ['blah', 'H.R. 2354', 'blah']

'''
Turns out you have to escape your whitespace in verbose mode...
'''

regex = r'''(?x)
  (?:H|S)\.\ ?(?:(?:J|R)\.\ )?(?:Con\.\ )?(?:Res\.\ )?\d+ # Bills
| ([A-Z]\.)+                                              # Abbreviations (U.S.A., etc.)
| ([A-Z]+\&[A-Z]+)                                        # Internal ampersands (AT&T, etc.)
| (Mr\.|Dr\.|Mrs\.|Ms\.)                                  # Mr., Mrs., etc.
| \d*\.\d+                                                # Numbers with decimal points.
| \d\d?:\d\d                                              # Times.
| \$?[,0-9]+                                              # Numbers with thousands separators.
| (((a|A)|(p|P))\.(m|M)\.)                                # a.m., p.m., A.M., P.M.
| \w+((-|')\w+)*                                          # Words with optional internal hyphens.
| \$?\d+(\.\d+)?%?                                        # Currency and percentages.
| \.\.\.                                                  # Ellipsis
| [][.,;"'?():-_`]
  '''

print regexp_tokenize('blah H.R. 2354 blah', regex)
# prints ['blah', 'H.R. 2354', 'blah']