diff options
| author | Jules Laplace <julescarbon@gmail.com> | 2018-11-25 22:19:15 +0100 |
|---|---|---|
| committer | Jules Laplace <julescarbon@gmail.com> | 2018-11-25 22:19:15 +0100 |
| commit | ee3d0d98e19f1d8177d85af1866fd0ee431fe9ea (patch) | |
| tree | 41372528e78d4328bc2a47bbbabac7e809c58894 /scraper/s2-merge-csv.py | |
| parent | 255b8178af1e25a71fd23703d30c0d1f74911f47 (diff) | |
moving stuff
Diffstat (limited to 'scraper/s2-merge-csv.py')
| -rw-r--r-- | scraper/s2-merge-csv.py | 28 |
1 files changed, 28 insertions, 0 deletions
diff --git a/scraper/s2-merge-csv.py b/scraper/s2-merge-csv.py new file mode 100644 index 00000000..301f47ae --- /dev/null +++ b/scraper/s2-merge-csv.py @@ -0,0 +1,28 @@ +import os +import glob +import time +import simplejson as json +import click +import operator +from util import * +import random + +@click.command() +@click.option('--path', '-d', default='report/institutions_geocoded', help='Path to CSVs') +def s2_merge_csv(path): + print(path) + lookup = {} + for fn in glob.iglob('{}/*.csv'.format(path)): + rows = read_csv(fn, keys=False) + for row in rows: + key = row[0] + if key not in lookup: + lookup[key] = row + keys = sorted(lookup.keys()) + deduped = [] + for key in keys: + deduped.append(lookup[key]) + write_csv('{}.csv'.format(path), keys=None, rows=deduped) + +if __name__ == '__main__': + s2_merge_csv() |
