s460929/wmt-2020-pl-en
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Fix invalid tokenizer issued by #13, bump version to 2.1.1

Browse Source
This commit is contained in:
SuHun Han 2017-04-03 23:05:19 +09:00
parent 5e77f76ead
commit d991f5e3e3
3 changed files with 9 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
googletrans/__init__.py
View File

@ -1,6 +1,6 @@
"""Free Google Translate API for Python. Translates totally free of charge.""" """Free Google Translate API for Python. Translates totally free of charge."""
__all__ = 'Translator', __all__ = 'Translator',
__version_info__ = 2, 1, 0 __version_info__ = 2, 1, 1
__version__ = '.'.join(str(v) for v in __version_info__) __version__ = '.'.join(str(v) for v in __version_info__)

1
googletrans/gtoken.py
View File

@ -150,6 +150,7 @@ class TokenAcquirer(object):
# append calculated value if l is less than 2048 # append calculated value if l is less than 2048
elif l < 2048: elif l < 2048:
e.append(l >> 6 | 192) e.append(l >> 6 | 192)
e.append(l)
# append calculated value if l matches special condition # append calculated value if l matches special condition
elif (l & 64512) == 55296 and g + 1 < size and \ elif (l & 64512) == 55296 and g + 1 < size and \
ord(text[g + 1]) & 64512 == 56320: ord(text[g + 1]) & 64512 == 56320:

7
tests/test_client.py
View File

@ -27,6 +27,13 @@ def test_unicode(translator):
assert result.text == u'こんにちは。' assert result.text == u'こんにちは。'
def test_special_chars(translator):
text = u"Copyright © Google"
result = translator.translate(text, src='en', dest='fr')
assert result.text == text
def test_multiple_sentences(translator): def test_multiple_sentences(translator):
text = u"""Architecturally, the school has a Catholic character. text = u"""Architecturally, the school has a Catholic character.
Atop the Main Building's gold dome is a golden statue of the Virgin Mary. Atop the Main Building's gold dome is a golden statue of the Virgin Mary.