1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268
|
#! /usr/bin/env python3
# $Id: test_error_reporting.py 9068 2022-06-13 12:05:08Z milde $
# Author: Günter Milde <milde@users.sf.net>
# Copyright: This module has been placed in the public domain.
"""
Test `EnvironmentError` reporting.
In some locales, the `errstr` argument of IOError and OSError contains
non-ASCII chars.
In Python 2, converting an exception instance to `str` or `unicode`
might fail, with non-ASCII chars in arguments and the default encoding
and errors ('ascii', 'strict').
Therefore, Docutils must not use string interpolation with exception
instances like, e.g., ::
try:
something
except IOError as error:
print('Found %s' % error)
unless the minimal required Python version has this problem fixed.
"""
from io import StringIO, BytesIO
import os
import sys
import unittest
import warnings
from docutils import frontend, utils
import docutils.parsers.rst
warnings.filterwarnings('ignore', category=DeprecationWarning,
module='.*error_reporting')
from docutils.utils.error_reporting import SafeString, ErrorString, ErrorOutput # noqa: E402, E501
class SafeStringTests(unittest.TestCase):
# test data:
bs = b'\xc3\xbc' # str(bs) returns repr(bs)
us = u'\xfc' # bytes(us) fails (requires encoding argument)
be = Exception(bs)
ue = Exception(us) # bytes(ue) fails
# wrapped test data:
wbs = SafeString(bs)
wus = SafeString(us)
wbe = SafeString(be)
wue = SafeString(ue)
def test_7bit(self):
# wrapping (not required with 7-bit chars) must not change the
# result of conversions:
bs7 = b'foo'
us7 = u'foo'
be7 = Exception(bs7)
ue7 = Exception(us7)
self.assertEqual(str(bs7), str(SafeString(bs7)))
self.assertEqual(str(us7), str(SafeString(us7)))
self.assertEqual(str(be7), str(SafeString(be7)))
self.assertEqual(str(ue7), str(SafeString(ue7)))
def test_ustr(self):
"""Test conversion to a unicode-string."""
# unicode(self.bs) fails
self.assertEqual(str, type(str(self.wbs)))
self.assertEqual(str(self.us), str(self.wus))
# unicode(self.be) fails
self.assertEqual(str, type(str(self.wbe)))
self.assertEqual(str, type(str(self.ue)))
self.assertEqual(str, type(str(self.wue)))
self.assertEqual(self.us, str(self.wue))
def test_str(self):
"""Test conversion to a string
(bytes in Python 2, unicode in Python 3).
"""
self.assertEqual(str(self.bs), str(self.wbs))
self.assertEqual(str(self.be), str(self.wbe))
self.assertEqual(str(self.us), str(self.wus))
self.assertEqual(str(self.ue), str(self.wue))
class ErrorStringTests(unittest.TestCase):
bs = b'\xc3\xbc' # unicode(bs) fails, str(bs) in Python 3 return repr()
us = u'\xfc' # bytes(us) fails; str(us) fails in Python 2
def test_str(self):
self.assertEqual('Exception: spam',
str(ErrorString(Exception('spam'))))
self.assertEqual('IndexError: '+str(self.bs),
str(ErrorString(IndexError(self.bs))))
self.assertEqual('ImportError: %s' % SafeString(self.us),
str(ErrorString(ImportError(self.us))))
def test_unicode(self):
self.assertEqual(u'Exception: spam',
str(ErrorString(Exception(u'spam'))))
self.assertEqual(u'IndexError: '+self.us,
str(ErrorString(IndexError(self.us))))
self.assertEqual(u'ImportError: %s' % SafeString(self.bs),
str(ErrorString(ImportError(self.bs))))
# ErrorOutput tests
# -----------------
# Stub: Buffer with 'strict' auto-conversion of input to byte string:
class BBuf(BytesIO):
def write(self, data):
if isinstance(data, str):
data.encode('ascii', 'strict')
super(BBuf, self).write(data)
# Stub: Buffer expecting unicode string:
class UBuf(StringIO):
def write(self, data):
# emulate Python 3 handling of stdout, stderr
if isinstance(data, bytes):
raise TypeError('must be unicode, not bytes')
super(UBuf, self).write(data)
class ErrorOutputTests(unittest.TestCase):
def test_defaults(self):
e = ErrorOutput()
self.assertEqual(e.stream, sys.stderr)
def test_bbuf(self):
buf = BBuf() # buffer storing byte string
e = ErrorOutput(buf, encoding='ascii')
# write byte-string as-is
e.write(b'b\xfc')
self.assertEqual(buf.getvalue(), b'b\xfc')
# encode unicode data with backslashescape fallback replacement:
e.write(u' u\xfc')
self.assertEqual(buf.getvalue(), b'b\xfc u\\xfc')
# handle Exceptions with Unicode string args
# unicode(Exception(u'e\xfc')) # fails in Python < 2.6
e.write(AttributeError(u' e\xfc'))
self.assertEqual(buf.getvalue(), b'b\xfc u\\xfc e\\xfc')
# encode with `encoding` attribute
e.encoding = 'utf-8'
e.write(u' u\xfc')
self.assertEqual(buf.getvalue(), b'b\xfc u\\xfc e\\xfc u\xc3\xbc')
def test_ubuf(self):
buf = UBuf() # buffer only accepting unicode string
# decode of binary strings
e = ErrorOutput(buf, encoding='ascii')
e.write(b'b\xfc')
self.assertEqual(buf.getvalue(), u'b\ufffd') # REPLACEMENT CHARACTER
# write Unicode string and Exceptions with Unicode args
e.write(u' u\xfc')
self.assertEqual(buf.getvalue(), u'b\ufffd u\xfc')
e.write(AttributeError(u' e\xfc'))
self.assertEqual(buf.getvalue(), u'b\ufffd u\xfc e\xfc')
# decode with `encoding` attribute
e.encoding = 'latin1'
e.write(b' b\xfc')
self.assertEqual(buf.getvalue(), u'b\ufffd u\xfc e\xfc b\xfc')
class SafeStringTests_locale(unittest.TestCase):
"""
Test docutils.SafeString with 'problematic' locales.
The error message in `EnvironmentError` instances comes from the OS
and in some locales (e.g. ru_RU), contains high bit chars.
"""
# test data:
bs = b'\xc3\xbc'
us = u'\xfc'
try:
open(b'\xc3\xbc')
except IOError as e: # in Python 3 the name for the exception instance
bioe = e # is local to the except clause
try:
open(u'\xfc')
except IOError as e:
uioe = e
except UnicodeEncodeError:
try:
open(u'\xfc'.encode(sys.getfilesystemencoding(), 'replace'))
except IOError as e:
uioe = e
try:
os.chdir(b'\xc3\xbc')
except OSError as e:
bose = e
try:
os.chdir(u'\xfc')
except OSError as e:
uose = e
except UnicodeEncodeError:
try:
os.chdir(u'\xfc'.encode(sys.getfilesystemencoding(), 'replace'))
except OSError as e:
uose = e
# wrapped test data:
wbioe = SafeString(bioe)
wuioe = SafeString(uioe)
wbose = SafeString(bose)
wuose = SafeString(uose)
def test_ustr(self):
"""Test conversion to a unicode-string."""
# unicode(bioe) fails with e.g. 'ru_RU.utf8' locale
self.assertEqual(str, type(str(self.wbioe)))
self.assertEqual(str, type(str(self.wuioe)))
self.assertEqual(str, type(str(self.wbose)))
self.assertEqual(str, type(str(self.wuose)))
def test_str(self):
"""Test conversion to a string
(bytes in Python 2, unicode in Python 3).
"""
self.assertEqual(str(self.bioe), str(self.wbioe))
self.assertEqual(str(self.uioe), str(self.wuioe))
self.assertEqual(str(self.bose), str(self.wbose))
self.assertEqual(str(self.uose), str(self.wuose))
class ErrorReportingTests(unittest.TestCase):
"""
Test cases where error reporting can go wrong.
Do not test the exact output (as this varies with the locale), just
ensure that the correct exception is thrown.
"""
# These tests fail with a 'problematic locale',
# Docutils revision < 7035, and Python 2:
parser = docutils.parsers.rst.Parser()
"""Parser shared by all ParserTestCases."""
settings = frontend.get_default_settings(parser)
settings.report_level = 1
settings.halt_level = 1
settings.warning_stream = ''
document = utils.new_document('test data', settings)
def test_include(self):
source = '.. include:: bogus.txt'
self.assertRaises(utils.SystemMessage,
self.parser.parse, source, self.document)
def test_raw_file(self):
source = ('.. raw:: html\n'
' :file: bogus.html\n')
self.assertRaises(utils.SystemMessage,
self.parser.parse, source, self.document)
def test_csv_table(self):
source = ('.. csv-table:: external file\n'
' :file: bogus.csv\n')
self.assertRaises(utils.SystemMessage,
self.parser.parse, source, self.document)
if __name__ == '__main__':
unittest.main()
|