for k, df in dps.items():
df_ = df.reset_index().dropna()
yield k, df_
if __name__ == '__main__':
print('reading source files...')
data = pd.read_excel(source, skiprows=2, sheetname='KILM 16b')
print('creating concept file...')
cdf = extract_concepts(data)
path = os.path.join(out_dir, 'ddf--concepts.csv')
cdf.to_csv(path, index=False)
print('creating entities files...')
country = extract_entities_country(data)
path = os.path.join(out_dir, 'ddf--entities--country.csv')
country.to_csv(path, index=False)
print('creating datapoints...')
for k, df in extract_datapoints(data):
path = os.path.join(out_dir,
'ddf--datapoints--{}--by--country--year.csv'.format(k))
df.to_csv(path, index=False)
print('creating index file...')
create_index_file(out_dir)
print('Done.')
dps.columns = ['country', 'year', to_concept_id(conc)]
dps['country'] = dps['country'].map(to_concept_id)
return to_concept_id(conc), dps.dropna()
if __name__ == '__main__':
print('reading source files...')
data = pd.read_excel(source, skiprows=2, sheetname='KILM 07b')
print('creating concept files...')
cdf = extract_concepts(data)
path = os.path.join(out_dir, 'ddf--concepts.csv')
cdf.to_csv(path, index=False)
print('creating entities files...')
country = extract_entities_country(data)
path = os.path.join(out_dir, 'ddf--entities--country.csv')
country.to_csv(path, index=False)
print('creating datapoints files...')
k, datapoints = extract_datapoints(data)
path = os.path.join(out_dir,
'ddf--datapoints--{}--by--country--year.csv'.format(k))
datapoints.to_csv(path, index=False)
print('creating index file...')
create_index_file(out_dir)
print('Done.')