如何从数组的字符串表示形式重构数组？

Question

I've got a huge CSV that was generated by a Python script. Some cells include arrays of data, while others include single-item arrays. Some examples:

cell01 == ['"July, 2002"', 'CUREE Publication No. CEA-01.', 'Project No. 3126', 'Prepared for Consortium of Universities for Research in Earthquake Engineering.']
cell02 == ['[Memorandum from Ralph J. Johnson on Andy Place].']
cell03 == ["Financial statements for the years ended March 31, 1991 and 1990 and independent auditors' report"]

Ideally, I'd like to parse all this data into structures that look like the following:

cell01_parsed[0] == '"July, 2002"'
cell01_parsed[1] == 'CUREE Publication No. CEA-01.'
cell01_parsed[2] == 'Project No. 3126'
cell01_parsed[3] == 'Prepared for Consortium of Universities for Research in Earthquake Engineering.'

cell02_parsed == '[Memorandum from Ralph J. Johnson on Andy Place].'

cell03_parsed == 'Financial statements for the years ended March 31, 1991 and 1990 and independent auditors\' report'

However, when I use a csv.reader() or csv.DictReader() , these lines are parsed as strings, not arrays. What would be an easy way to do this? I can't use split(',') since some of the strings have commas in the middle of items.

Answer 1

You could try to split your strings by regex (find out one that fits your data) like so:

import re
test_str = '"July, 2002", CUREE Publication No. CEA-01.' 
re.compile(',(?!.+\")').split(test_str)