mirror of
https://github.com/ilri/csv-metadata-quality.git
synced 2024-06-09 13:25:08 +02:00
43 lines
996 B
Python
43 lines
996 B
Python
|
import pandas as pd
|
||
|
|
|
||
|
|
def issn(field):
|
||
|
|
"""Check if an ISSN is valid.
|
||
|
|
|
||
|
|
Prints the ISSN if invalid.
|
||
|
|
|
||
|
|
stdnum's is_valid() function never raises an exception.
|
||
|
|
|
||
|
|
See: https://arthurdejong.org/python-stdnum/doc/1.11/index.html#stdnum.module.is_valid
|
||
|
|
"""
|
||
|
|
|
||
|
|
from stdnum import issn
|
||
|
|
|
||
|
|
# Skip fields with missing values
|
||
|
|
if pd.isna(field):
|
||
|
|
return
|
||
|
|
|
||
|
|
# Try to split multi-value field on "||" separator
|
||
|
|
for value in field.split('||'):
|
||
|
|
|
||
|
|
if not issn.is_valid(value):
|
||
|
|
print(f'Invalid ISSN: {value}')
|
||
|
|
|
||
|
|
|
||
|
|
def isbn(field):
|
||
|
|
"""Check if an ISBN is valid.
|
||
|
|
|
||
|
|
Prints the ISBN if invalid.
|
||
|
|
|
||
|
|
stdnum's is_valid() function never raises an exception.
|
||
|
|
|
||
|
|
See: https://arthurdejong.org/python-stdnum/doc/1.11/index.html#stdnum.module.is_valid
|
||
|
|
"""
|
||
|
|
|
||
|
|
from stdnum import isbn
|
||
|
|
|
||
|
|
# Try to split multi-value field on "||" separator
|
||
|
|
for value in field.split('||'):
|
||
|
|
|
||
|
|
if not isbn.is_valid(value):
|
||
|
|
print(f'Invalid ISBN: {value}')
|