def _get_encoding(self, infer=False):
enc = self._declared_encoding()
if enc and not encoding_exists(enc):
enc = None
if not enc and infer:
enc = self._body_inferred_encoding()
if not enc:
enc = self._DEFAULT_ENCODING
return resolve_encoding(enc)
def _get_encoding(self, infer=False):
enc = self._declared_encoding()
if enc and not encoding_exists(enc):
enc = None
if not enc and infer:
enc = self._body_inferred_encoding()
if not enc:
enc = self._DEFAULT_ENCODING
return resolve_encoding(enc)
def test_process_response_force_recalculate_encoding(self):
headers = {"Content-Type": "text/html", "Content-Encoding": "gzip"}
f = StringIO()
plainbody = """<html><head><title>Some page</title><meta http-equiv="Content-Type" content="text/html; charset=gb2312">"""
zf = GzipFile(fileobj=f, mode="wb")
zf.write(plainbody)
zf.close()
response = HtmlResponse("http;//www.example.com/page.html", headers=headers, body=f.getvalue())
request = Request("http://www.example.com/")
newresponse = self.mw.process_response(request, response, self.spider)
assert isinstance(newresponse, HtmlResponse)
self.assertEqual(newresponse.body, plainbody)
self.assertEqual(newresponse.encoding, resolve_encoding("gb2312"))
def test_process_response_encoding_inside_body(self):
headers = {
'Content-Type': 'text/html',
'Content-Encoding': 'gzip',
}
f = StringIO()
plainbody = """<html><head><title>Some page</title><meta http-equiv="Content-Type" content="text/html; charset=gb2312">"""
zf = GzipFile(fileobj=f, mode='wb')
zf.write(plainbody)
zf.close()
response = Response("http;//www.example.com/", headers=headers, body=f.getvalue())
request = Request("http://www.example.com/")
newresponse = self.mw.process_response(request, response, self.spider)
assert isinstance(newresponse, HtmlResponse)
self.assertEqual(newresponse.body, plainbody)
self.assertEqual(newresponse.encoding, resolve_encoding('gb2312'))
def test_process_response_force_recalculate_encoding(self):
headers = {
'Content-Type': 'text/html',
'Content-Encoding': 'gzip',
}
f = StringIO()
plainbody = """<html><head><title>Some page</title><meta http-equiv="Content-Type" content="text/html; charset=gb2312">"""
zf = GzipFile(fileobj=f, mode='wb')
zf.write(plainbody)
zf.close()
response = HtmlResponse("http;//www.example.com/page.html",
headers=headers,
body=f.getvalue())
request = Request("http://www.example.com/")
newresponse = self.mw.process_response(request, response, self.spider)
assert isinstance(newresponse, HtmlResponse)
self.assertEqual(newresponse.body, plainbody)
self.assertEqual(newresponse.encoding, resolve_encoding('gb2312'))
def _assert_response_encoding(self, response, encoding):
self.assertEqual(response.encoding, resolve_encoding(encoding))
def _assert_response_encoding(self, response, encoding):
self.assertEqual(response.encoding, resolve_encoding(encoding))
def test_resolve_encoding(self):
self.assertEqual(resolve_encoding('latin1', self._ENCODING_ALIASES),
'latin1')
self.assertEqual(resolve_encoding('foo', self._ENCODING_ALIASES),
'cp1252')
def test_resolve_encoding(self):
self.assertEqual(resolve_encoding('latin1', self._ENCODING_ALIASES),
'latin1')
self.assertEqual(resolve_encoding('foo', self._ENCODING_ALIASES),
'cp1252')