some code cleanup

nlp/algorithms/finder/time_finder.py

@@ -416,9 +416,6 @@
 _str_brackets = r'[(){}\[\]]'
 _regex_brackets = re.compile(_str_brackets)
 
-#_CANDIDATE_FIELDS = ['start', 'end', 'match_text', 'regex']
-#_Candidate = namedtuple('_Candidate', _CANDIDATE_FIELDS)
-
 
 ###############################################################################
 def enable_debug():
@@ -427,107 +424,6 @@ def enable_debug():
     _TRACE = True
 
 
-# ###############################################################################
-# def _has_overlap(a1, b1, a2, b2):
-#     """
-#     Determine if intervals [a1, b1) and [a2, b2) overlap at all.
-#     """
-
-#     assert a1 <= b1
-#     assert a2 <= b2
-
-#     if b2 <= a1:
-#         return False
-#     elif a2 >= b1:
-#         return False
-#     else:
-#         return True
-
-# ###############################################################################
-# def _remove_overlap(candidates):
-#     """
-#     Given a set of match candidates, resolve into nonoverlapping matches.
-#     Take the longest match at any given position.
-
-#     ASSUMES that the candidate list has been sorted by matching text length,
-#     from longest to shortest.
-#     """
-
-#     if _TRACE:
-#         print('called _remove_overlap...')
-
-#     results = []
-#     overlaps = []
-#     indices = [i for i in range(len(candidates))]
-
-#     i = 0
-#     while i < len(indices):
-
-#         if _TRACE:
-#             print('\tstarting indices: {0}'.format(indices))
-
-#         index_i = indices[i]
-#         start_i = candidates[index_i].start
-#         end_i   = candidates[index_i].end
-#         len_i   = end_i - start_i
-
-#         overlaps.append(i)
-#         candidate_index = index_i
-
-#         j = i+1
-#         while j < len(indices):
-#             index_j = indices[j]
-#             start_j = candidates[index_j].start
-#             end_j   = candidates[index_j].end
-#             len_j   = end_j - start_j
-
-#             # does candidate[j] overlap candidate[i] at all
-#             if _has_overlap(start_i, end_i, start_j, end_j):
-#                 if _TRACE:
-#                     print('\t\t{0} OVERLAPS {1}, lengths {2}, {3}'.
-#                           format(candidates[index_i].match_text,
-#                                  candidates[index_j].match_text,
-#                                  len_i, len_j))
-#                 overlaps.append(j)
-#                 # keep the longest match at any overlap region
-#                 if len_j > len_i:
-#                     start_i = start_j
-#                     end_i   = end_j
-#                     len_i   = len_j
-#                     candidate_index = index_j
-#             j += 1
-
-#         if _TRACE:
-#             print('\t\t\twinner: {0}'.
-#                   format(candidates[candidate_index].match_text))
-#             print('\t\t\tappending {0} to results'.
-#                   format(candidates[candidate_index].match_text))
-
-#         results.append(candidates[candidate_index])
-
-#         if _TRACE:
-#             print('\t\toverlaps: {0}'.format(overlaps))
-
-#         # remove all overlaps
-#         new_indices = []
-#         for k in range(len(indices)):
-#             if k not in overlaps:
-#                 new_indices.append(indices[k])
-#         indices = new_indices
-
-#         if _TRACE:
-#             print('\t\tindices after removing overlaps: {0}'.format(indices))
-
-#         if 0 == len(indices):
-#             break
-
-#         # start over
-#         i = 0
-#         overlaps = []
-
-#     return results
-
-
 ###############################################################################
 def _clean_sentence(sentence):
     """