python/jc
1
0
Fork 0
mirror of https://github.com/kellyjonbrazil/jc.git synced 2025-06-17 00:07:37 +02:00
Code Issues Releases Activity

add ParseError test to streaming csv parser

Browse Source
This commit is contained in:
Kelly Brazil
2021-10-26 07:36:44 -07:00
parent 3ab9b43a2e
commit b280c4fc18
2 changed files with 11 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
tests/fixtures/generic/csv-insurance-streaming.json vendored
View File

File diff suppressed because one or more lines are too long

29
tests/test_csv_s.py
View File

@ -2,6 +2,7 @@ import os
import json
import unittest
import jc.parsers.csv_s
from jc.exceptions import ParseError
THIS_DIR = os.path.dirname(os.path.abspath(__file__))
@ -38,9 +39,6 @@ class MyTests(unittest.TestCase):
with open(os.path.join(THIS_DIR, os.pardir, 'tests/fixtures/generic/csv-homes.csv'), 'r', encoding='utf-8') as f:
self.generic_csv_homes = f.read()
with open(os.path.join(THIS_DIR, os.pardir, 'tests/fixtures/generic/csv-insurance.csv'), 'r', encoding='utf-8') as f:
self.generic_csv_insurance = f.read()
with open(os.path.join(THIS_DIR, os.pardir, 'tests/fixtures/generic/csv-10k-sales-records.csv'), 'r', encoding='utf-8') as f:
self.generic_csv_10k_sales_records = f.read()
@ -69,9 +67,6 @@ class MyTests(unittest.TestCase):
with open(os.path.join(THIS_DIR, os.pardir, 'tests/fixtures/generic/csv-homes-streaming.json'), 'r', encoding='utf-8') as f:
self.generic_csv_homes_streaming_json = json.loads(f.read())
with open(os.path.join(THIS_DIR, os.pardir, 'tests/fixtures/generic/csv-insurance-streaming.json'), 'r', encoding='utf-8') as f:
self.generic_csv_insurance_streaming_json = json.loads(f.read())
with open(os.path.join(THIS_DIR, os.pardir, 'tests/fixtures/generic/csv-10k-sales-records-streaming.json'), 'r', encoding='utf-8') as f:
self.generic_csv_10k_sales_records_streaming_json = json.loads(f.read())
@ -81,12 +76,16 @@ class MyTests(unittest.TestCase):
"""
self.assertEqual(list(jc.parsers.csv_s.parse('', quiet=True)), [])
# no test for unparsable rows yet - CSV library fails right away if it can't decode the data
# def test_csv_unparsable(self):
# data = 'unparsable data'
# g = jc.parsers.csv_s.parse(data.splitlines(), quiet=True)
# with self.assertRaises(ParseError):
# list(g)
def test_csv_unparsable(self):
"""
Test CSV streaming parser with '\r' newlines. This will raise ParseError due to a Python bug
that does not correctly iterate on that line ending with sys.stdin. Simulating this failure since
standard str.splitlines() does work correctly.
"""
data = r'unparsable\rdata'
g = jc.parsers.csv_s.parse(data.splitlines(), quiet=True)
with self.assertRaises(ParseError):
list(g)
def test_csv_s_biostats(self):
"""
@ -136,12 +135,6 @@ class MyTests(unittest.TestCase):
"""
self.assertEqual(list(jc.parsers.csv_s.parse(self.generic_csv_homes.splitlines(), quiet=True)), self.generic_csv_homes_streaming_json)
def test_csv_s_insurance(self):
"""
Test 'insurance.csv' file
"""
self.assertEqual(list(jc.parsers.csv_s.parse(self.generic_csv_insurance.splitlines(), quiet=True)), self.generic_csv_insurance_streaming_json)
def test_csv_s_10k_records(self):
"""
Test '10k-sales-records.csv' file