eBookReaderSwitch/thirdparty/curl/tests/curl_test_data.py

62 lines
1.9 KiB
Python
Executable File

#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# Project ___| | | | _ \| |
# / __| | | | |_) | |
# | (__| |_| | _ <| |___
# \___|\___/|_| \_\_____|
#
# Copyright (C) 2017, Daniel Stenberg, <daniel@haxx.se>, et al.
#
# This software is licensed as described in the file COPYING, which
# you should have received as part of this distribution. The terms
# are also available at https://curl.haxx.se/docs/copyright.html.
#
# You may opt to use, copy, modify, merge, publish, distribute and/or sell
# copies of the Software, and permit persons to whom the Software is
# furnished to do so, under the terms of the COPYING file.
#
# This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
# KIND, either express or implied.
#
"""Module for extracting test data from the test data folder"""
from __future__ import (absolute_import, division, print_function,
unicode_literals)
import os
import re
import logging
log = logging.getLogger(__name__)
REPLY_DATA = re.compile("<reply>\s*<data>(.*?)</data>", re.MULTILINE | re.DOTALL)
class TestData(object):
def __init__(self, data_folder):
self.data_folder = data_folder
def get_test_data(self, test_number):
# Create the test file name
filename = os.path.join(self.data_folder,
"test{0}".format(test_number))
log.debug("Parsing file %s", filename)
with open(filename, "rb") as f:
contents = f.read().decode("utf-8")
m = REPLY_DATA.search(contents)
if not m:
raise Exception("Couldn't find a <reply><data> section")
# Left-strip the data so we don't get a newline before our data.
return m.group(1).lstrip()
if __name__ == '__main__':
td = TestData("./data")
data = td.get_test_data(1)
print(data)