def test_unescape(self):
from clld.lib.bibtex import unescape, u_unescape
self.assertEqual(unescape(binary_type("\\ss \xef".encode('latin1'))), 'ß\xef')
self.assertEqual(unescape("\\ss "), 'ß')
self.assertEqual(u_unescape('?[\\u123] ?[\\u1234]'), '{ \u04d2')
s = '\u2013'
self.assertEqual(s, unescape(s))
self.assertEqual(unescape('?[\\u65533]'), '\ufffd')
def test_unescape(self):
from clld.lib.bibtex import unescape, u_unescape
self.assertEqual(unescape(binary_type("\\ss \xef".encode('latin1'))),
'ß\xef')
self.assertEqual(unescape("\\ss "), 'ß')
self.assertEqual(u_unescape('?[\\u123] ?[\\u1234]'), '{ \u04d2')
s = '\u2013'
self.assertEqual(s, unescape(s))
self.assertEqual(unescape('?[\\u65533]'), '\ufffd')
def unescape(string):
"""transform latex escape sequences of type \`\ae into unicode
"""
def _delatex(s):
try:
t = str(s)
result = t.decode('latex+latin1')
except UnicodeEncodeError: # pragma: no cover
result = string
u_result = unicode(result)
return u_result
res = u_unescape(_delatex(stripctrlchars(unicode(string).strip())))
for symbol in sorted(SYMBOLS.keys(), key=lambda s: len(s)):
res = res.replace(symbol, SYMBOLS[symbol])
if '\\' not in res:
res = res.replace('{', '')
res = res.replace('}', '')
res = res.replace('\\\\&{} ', '& ')
return res
