def main(input_path_1, input_path_2, match_input_path, output_path_1,
output_path_2):
"""Main entry point."""
dialect_1 = get_dialect(input_path_1)
dialect_2 = get_dialect(input_path_2)
data1, rows1, columns1, corner1 = read_csv(input_path_1, dialect_1)
data2, rows2, columns2, corner2 = read_csv(input_path_2, dialect_2)
try:
match_results = json.loads(match_input_path)
except Exception:
with open(match_input_path) as f:
match_results = json.load(f)
mode1, mode2 = match_results.get('mode', 'cc').lower()
if mode1 == 'r':
data1 = data1.transpose()
rows1, columns1 = columns1, rows1
if mode2 == 'r':
data2 = data2.transpose()
rows2, columns2 = columns2, rows2
index1, index2 = zip(*(tuple(p['index'] for p in assign['pair'][:2])
for assign in match_results.get('assignments', [])))
index1 = list(index1)
index2 = list(index2)
new_cols1 = columns1[index1]
new_data1 = data1[:, index1]
new_cols2 = columns2[index2]
new_data2 = data2[:, index2]
write_csv(new_data1, rows1, new_cols1, corner1, output_path_1, csv.excel)
write_csv(new_data2, rows2, new_cols2, corner2, output_path_2, csv.excel)
"""Remove columns with missing/invalid data."""
import csv
import math
# include<lib/csv.py>
# ifdef<LINTING>
from included_files import get_dialect, read_csv, write_csv
from girder_worker_environment import input_path_1, input_path_2
# endif
dialect_1 = get_dialect(input_path_1)
dialect_2 = get_dialect(input_path_2)
data1, rows1, columns1, corner = read_csv(input_path_1, dialect_1)
data2, rows2, columns2, corner = read_csv(input_path_2, dialect_2)
invalid_indexes = set()
n_columns1 = len(columns1)
n_columns2 = len(columns2)
for column_index in range(max(n_columns1, n_columns2)):
is_invalid = False
if column_index < n_columns1:
for row_index in range(len(rows1)):
is_invalid = (math.isinf(data1[row_index][column_index])
or math.isnan(data1[row_index][column_index]))
"""Extract the columns from one or two data sets."""
# include<lib/csv.py>
import json
# ifdef<LINTING>
from included_files import get_dialect, ColumnExtractor, RowExtractor
from girder_worker_environment import \
extract_columns, input_path_1, input_path_2, single_mode
# endif
feature_extractor = ColumnExtractor if extract_columns else RowExtractor
if single_mode:
dialect1 = get_dialect(input_path_1)
list1 = None
with open(input_path_1, 'rU') as input1:
list1 = list(feature_extractor(input1, dialect1))
values = ({'id': a, 'description': a} for a in set(list1))
extract_result = list(sorted(values, key=(lambda x: x['id'])))
else:
dialect1 = get_dialect(input_path_1)
dialect2 = get_dialect(input_path_2)
list1 = None
with open(input_path_1, 'rU') as input1:
list1 = list(feature_extractor(input1, dialect1))
"""Remove headers from columns & rows."""
import csv
from six import PY2
from tempfile import NamedTemporaryFile
# include<lib/csv.py>
# ifdef<LINTING>
from included_files import get_dialect
from girder_worker_environment import input_path
# endif
dialect = get_dialect(input_path)
with open(input_path, 'rU') as input_f:
reader = csv.reader(input_f, dialect)
output_f = None
if PY2:
output_f = NamedTemporaryFile(mode='wb', delete=False)
else:
output_f = NamedTemporaryFile(mode='w', newline='', delete=False)
output_path = output_f.name
with output_f:
writer = csv.writer(output_f, dialect=csv.excel)
# Unconditionally skip the first row and
# the first value in every subsequent row.