def test_contents_encoding_1(self):
"""
Testing the logic of choosing the encoding and the process of
encoding contents.
"""
from aiida.tools.dbexporters.tcod import cif_encode_contents
self.assertEquals(cif_encode_contents('simple line')[1], None)
self.assertEquals(cif_encode_contents(' ;\n ;')[1], None)
self.assertEquals(cif_encode_contents(';\n'),
('=3B\n', 'quoted-printable'))
self.assertEquals(cif_encode_contents('line\n;line'),
('line\n=3Bline', 'quoted-printable'))
self.assertEquals(cif_encode_contents('tabbed\ttext'),
('tabbed=09text', 'quoted-printable'))
self.assertEquals(cif_encode_contents('angstrom Å'),
('angstrom =C3=85', 'quoted-printable'))
self.assertEquals(cif_encode_contents('.'),
('=2E', 'quoted-printable'))
self.assertEquals(cif_encode_contents('?'),
('=3F', 'quoted-printable'))
self.assertEquals(cif_encode_contents('.?'), ('.?', None))
# This one is particularly tricky: a long line is folded by the QP
# and the semicolon sign becomes the first character on a new line.
self.assertEquals(
cif_encode_contents("Å{};a".format("".join(
"a" for i in range(0, 69)))),
('=C3=85aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'
'aaaaaaaaaaaaaaaaaaaaaaaaaaaaa=\n=3Ba', 'quoted-printable'))
self.assertEquals(cif_encode_contents('angstrom ÅÅÅ'),
('YW5nc3Ryb20gw4XDhcOF', 'base64'))
self.assertEquals(
cif_encode_contents("".join("a" for i in range(0, 2048)))[1], None)
self.assertEquals(
cif_encode_contents("".join("a" for i in range(0, 2049)))[1],
'quoted-printable')
self.assertEquals(cif_encode_contents('datatest')[1], None)
self.assertEquals(cif_encode_contents('data_test')[1], 'base64')
def test_contents_encoding_1(self):
"""
Testing the logic of choosing the encoding and the process of
encoding contents.
"""
from aiida.tools.dbexporters.tcod import cif_encode_contents
self.assertEquals(cif_encode_contents(b'simple line')[1], None)
self.assertEquals(cif_encode_contents(b' ;\n ;')[1], None)
self.assertEquals(cif_encode_contents(b';\n'),
(b'=3B\n', 'quoted-printable'))
self.assertEquals(cif_encode_contents(b'line\n;line'),
(b'line\n=3Bline', 'quoted-printable'))
self.assertEquals(cif_encode_contents(b'tabbed\ttext'),
(b'tabbed=09text', 'quoted-printable'))
# Angstrom symbol 'Å' will be encoded as two bytes, thus encoding it
# for CIF will produce two quoted-printable entities, '=C3' and '=85',
# one for each byte.
self.assertEquals(cif_encode_contents(u'angstrom Å'.encode('utf-8')),
(b'angstrom =C3=85', 'quoted-printable'))
self.assertEquals(cif_encode_contents(b'.'),
(b'=2E', 'quoted-printable'))
self.assertEquals(cif_encode_contents(b'?'),
(b'=3F', 'quoted-printable'))
self.assertEquals(cif_encode_contents(b'.?'), (b'.?', None))
# This one is particularly tricky: a long line is folded by the QP
# and the semicolon sign becomes the first character on a new line.
self.assertEquals(
cif_encode_contents(u"Å{};a".format("".join(
"a" for i in range(0, 69))).encode('utf-8')),
(b'=C3=85aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'
b'aaaaaaaaaaaaaaaaaaaaaaaaaaaaa=\n=3Ba', 'quoted-printable'))
self.assertEquals(cif_encode_contents(u'angstrom ÅÅÅ'.encode('utf-8')),
(b'YW5nc3Ryb20gw4XDhcOF', 'base64'))
self.assertEquals(
cif_encode_contents("".join(
"a" for i in range(0, 2048)).encode('utf-8'))[1], None)
self.assertEquals(
cif_encode_contents("".join(
"a" for i in range(0, 2049)).encode('utf-8'))[1],
'quoted-printable')
self.assertEquals(cif_encode_contents(b'datatest')[1], None)
self.assertEquals(cif_encode_contents(b'data_test')[1], 'base64')