| 1 |
4586
|
aaronmk
|
#!/usr/bin/env python
|
| 2 |
|
|
# Canonicalizes a spreadsheet column to a vocabulary.
|
| 3 |
4649
|
aaronmk
|
# The column header is also canonicalized. CSVs without a header are supported.
|
| 4 |
4586
|
aaronmk
|
# Unrecognized names are left untouched, permitting successive runs on different
|
| 5 |
|
|
# vocabularies.
|
| 6 |
7449
|
aaronmk
|
# Raises an error if two input terms map to the same simplified string.
|
| 7 |
4586
|
aaronmk
|
# Case- and punctuation-insensitive.
|
| 8 |
|
|
|
| 9 |
|
|
import csv
|
| 10 |
|
|
import re
|
| 11 |
|
|
import sys
|
| 12 |
|
|
|
| 13 |
7449
|
aaronmk
|
class OnceOnlyDict(dict):
|
| 14 |
|
|
'''A dict that only allows each key to be assigned once (no overwriting).'''
|
| 15 |
|
|
|
| 16 |
|
|
def __setitem__(self, key, value):
|
| 17 |
|
|
if key in self: raise KeyError(key)
|
| 18 |
|
|
dict.__setitem__(self, key, value)
|
| 19 |
|
|
|
| 20 |
4586
|
aaronmk
|
def simplify(str_): return re.sub(r'[\W_]+', r'', str_.lower())
|
| 21 |
|
|
|
| 22 |
|
|
def main():
|
| 23 |
|
|
try: _prog_name, col_num, vocab_path = sys.argv
|
| 24 |
|
|
except ValueError: raise SystemExit('Usage: '+sys.argv[0]
|
| 25 |
|
|
+' <in col# vocab [| '+sys.argv[0]+' col# vocab_2]... >out')
|
| 26 |
|
|
col_num = int(col_num)
|
| 27 |
|
|
|
| 28 |
|
|
# Get vocab
|
| 29 |
7449
|
aaronmk
|
dict_ = OnceOnlyDict()
|
| 30 |
4586
|
aaronmk
|
stream = open(vocab_path, 'rb')
|
| 31 |
|
|
reader = csv.reader(stream)
|
| 32 |
4626
|
aaronmk
|
for row in reader: dict_[simplify(row[0])] = row[0]
|
| 33 |
4586
|
aaronmk
|
stream.close()
|
| 34 |
|
|
|
| 35 |
|
|
# Canonicalize input
|
| 36 |
|
|
reader = csv.reader(sys.stdin)
|
| 37 |
|
|
writer = csv.writer(sys.stdout)
|
| 38 |
|
|
for row in reader:
|
| 39 |
|
|
term = simplify(row[col_num])
|
| 40 |
|
|
try: row[col_num] = dict_[term]
|
| 41 |
|
|
except KeyError: pass
|
| 42 |
|
|
writer.writerow(row)
|
| 43 |
|
|
|
| 44 |
|
|
main()
|
