Revision 4625
Added by Aaron Marcuse-Kubitza about 12 years ago
bin/subtract | ||
---|---|---|
1 |
#!/usr/bin/env python |
|
2 |
# Subtracts map spreadsheet A1->C from A0->B to produce A->B. |
|
3 |
# Other A0 entries are subtracted as well: |
|
4 |
# - Empty entries without a comment documenting why they are empty, if there is |
|
5 |
# a matching entry in A1. |
|
6 |
# - Entries whose input and output maps to the same non-empty value in A1 |
|
7 |
# Multi-safe (supports an input appearing multiple times). Note that if there is |
|
8 |
# *any* mapping for an input in subtract_map, all mappings for that input in |
|
9 |
# from_map will be excluded. |
|
10 |
# Case- and punctuation-insensitive. |
|
11 |
|
|
12 |
import csv |
|
13 |
import operator |
|
14 |
import os.path |
|
15 |
import sys |
|
16 |
|
|
17 |
sys.path.append(os.path.dirname(__file__)+"/../lib") |
|
18 |
|
|
19 |
import maps |
|
20 |
import opts |
|
21 |
import util |
|
22 |
|
|
23 |
def main(): |
|
24 |
ignore = opts.env_flag('ignore') |
|
25 |
try: _prog_name, map_1_path = sys.argv[:2] |
|
26 |
except ValueError: |
|
27 |
raise SystemExit('Usage: env [ignore=1] '+sys.argv[0]+' <from_map ' |
|
28 |
'subtract_map [compare_col_num...] [| '+sys.argv[0] |
|
29 |
+' subtract_map_2]... >difference_map') |
|
30 |
compare_col_nums = map(int, sys.argv[2:]) # 0-based |
|
31 |
if compare_col_nums == []: |
|
32 |
compare_col_nums = None # list_subset() value for all columns |
|
33 |
|
|
34 |
def compare_on(row): |
|
35 |
return tuple(map(maps.simplify, util.list_subset(row, compare_col_nums, |
|
36 |
default=None))) |
|
37 |
|
|
38 |
headers = [None]*2 |
|
39 |
|
|
40 |
# Get map 1 |
|
41 |
input_cols = set() |
|
42 |
compare_cols = set() |
|
43 |
map_ = dict() |
|
44 |
stream = open(map_1_path, 'rb') |
|
45 |
reader = csv.reader(stream) |
|
46 |
headers[1] = reader.next() |
|
47 |
for row in reader: |
|
48 |
if row[0] != '': |
|
49 |
input_cols.add(maps.simplify(row[0])) |
|
50 |
compare_cols.add(compare_on(row)) |
|
51 |
if reduce(operator.and_, (v == '' for v in row[1:])): # all empty |
|
52 |
map_[row[0]] = row[1] |
|
53 |
stream.close() |
|
54 |
|
|
55 |
# Open map 0 |
|
56 |
reader = csv.reader(sys.stdin) |
|
57 |
headers[0] = reader.next() |
|
58 |
|
|
59 |
# Check col labels |
|
60 |
combinable = maps.combinable(*headers) |
|
61 |
if not combinable and not ignore: |
|
62 |
raise SystemExit('Map error: ' |
|
63 |
'Map 0 column 0 label doesn\'t contain map 1 column 0 label') |
|
64 |
|
|
65 |
# Subtract map 1 from map 0 |
|
66 |
writer = csv.writer(sys.stdout) |
|
67 |
writer.writerow(headers[0]) |
|
68 |
for row in reader: |
|
69 |
if not combinable or not ( |
|
70 |
(maps.is_nonexplicit_empty_mapping(row) |
|
71 |
and maps.simplify(row[0]) in input_cols) |
|
72 |
or compare_on(row) in compare_cols |
|
73 |
or util.have_same_value(map_, row[0], row[1]) # map to same place |
|
74 |
): |
|
75 |
# not combinable or not in map 1 |
|
76 |
writer.writerow(row) |
|
77 |
|
|
78 |
main() |
|
79 | 0 |
Also available in: Unified diff
Removed no longer used subtract (use filter_out_ci instead)