class HttpCompressionTest(TestCase):
def setUp(self):
self.spider = Spider('foo')
self.mw = HttpCompressionMiddleware()
def _getresponse(self, coding):
if coding not in FORMAT:
raise ValueError()
samplefile, contentencoding = FORMAT[coding]
with open(join(SAMPLEDIR, samplefile), 'rb') as sample:
body = sample.read()
headers = {
'Server': 'Yaws/1.49 Yet Another Web Server',
'Date': 'Sun, 08 Mar 2009 00:41:03 GMT',
'Content-Length': len(body),
'Content-Type': 'text/html',
'Content-Encoding': contentencoding,
}
response = Response('http://scrapytest.org/', body=body, headers=headers)
response.request = Request('http://scrapytest.org', headers={'Accept-Encoding': 'gzip,deflate'})
return response
def test_process_request(self):
request = Request('http://scrapytest.org')
assert 'Accept-Encoding' not in request.headers
self.mw.process_request(request, self.spider)
self.assertEqual(request.headers.get('Accept-Encoding'), 'gzip,deflate')
def test_process_response_gzip(self):
response = self._getresponse('gzip')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'gzip')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_rawdeflate(self):
response = self._getresponse('rawdeflate')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'deflate')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_zlibdelate(self):
response = self._getresponse('zlibdeflate')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'deflate')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_plain(self):
response = Response('http://scrapytest.org', body='<!DOCTYPE...')
request = Request('http://scrapytest.org')
assert not response.headers.get('Content-Encoding')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is response
assert newresponse.body.startswith('<!DOCTYPE')
def test_multipleencodings(self):
response = self._getresponse('gzip')
response.headers['Content-Encoding'] = ['uuencode', 'gzip']
request = response.request
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
self.assertEqual(newresponse.headers.getlist('Content-Encoding'), ['uuencode'])
def test_process_response_encoding_inside_body(self):
headers = {
'Content-Type': 'text/html',
'Content-Encoding': 'gzip',
}
f = BytesIO()
plainbody = b"""<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 = BytesIO()
plainbody = b"""<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_gzipped_contenttype(self):
response = self._getresponse('gzip')
response.headers['Content-Type'] = 'application/gzip'
request = response.request
newresponse = self.mw.process_response(request, response, self.spider)
self.assertIs(newresponse, response)
self.assertEqual(response.headers['Content-Encoding'], 'gzip')
self.assertEqual(response.headers['Content-Type'], 'application/gzip')
def setUp(self):
self.spider = Spider('foo')
self.mw = HttpCompressionMiddleware()
class HttpCompressionTest(TestCase):
def setUp(self):
self.spider = BaseSpider('foo')
self.mw = HttpCompressionMiddleware()
def _getresponse(self, coding):
if coding not in FORMAT:
raise ValueError()
samplefile, contentencoding = FORMAT[coding]
with open(join(SAMPLEDIR, samplefile), 'rb') as sample:
body = sample.read()
headers = {
'Server': 'Yaws/1.49 Yet Another Web Server',
'Date': 'Sun, 08 Mar 2009 00:41:03 GMT',
'Content-Length': len(body),
'Content-Type': 'text/html',
'Content-Encoding': contentencoding,
}
response = Response('http://scrapytest.org/',
body=body,
headers=headers)
response.request = Request('http://scrapytest.org',
headers={'Accept-Encoding': 'gzip,deflate'})
return response
def test_process_request(self):
request = Request('http://scrapytest.org')
assert 'Accept-Encoding' not in request.headers
self.mw.process_request(request, self.spider)
self.assertEqual(request.headers.get('Accept-Encoding'),
'x-gzip,gzip,deflate')
def test_process_response_gzip(self):
response = self._getresponse('gzip')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'gzip')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_rawdeflate(self):
response = self._getresponse('rawdeflate')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'deflate')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_zlibdelate(self):
response = self._getresponse('zlibdeflate')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'deflate')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_plain(self):
response = Response('http://scrapytest.org', body='<!DOCTYPE...')
request = Request('http://scrapytest.org')
assert not response.headers.get('Content-Encoding')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is response
assert newresponse.body.startswith('<!DOCTYPE')
def test_multipleencodings(self):
response = self._getresponse('gzip')
response.headers['Content-Encoding'] = ['uuencode', 'gzip']
request = response.request
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
self.assertEqual(newresponse.headers.getlist('Content-Encoding'),
['uuencode'])
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 setUp(self):
self.spider = BaseSpider('foo')
self.mw = HttpCompressionMiddleware()
def process_response(self, request, response, spider):
response = _chuncked.process_response(request, response, spider)
return HttpCompressionMiddleware.process_response(self,request, response, spider)
class HttpCompressionTest(TestCase):
def setUp(self):
self.spider = BaseSpider('foo')
self.mw = HttpCompressionMiddleware()
def _getresponse(self, coding):
if coding not in FORMAT:
raise ValueError()
samplefile, contentencoding = FORMAT[coding]
with open(join(SAMPLEDIR, samplefile), 'rb') as sample:
body = sample.read()
headers = {
'Server': 'Yaws/1.49 Yet Another Web Server',
'Date': 'Sun, 08 Mar 2009 00:41:03 GMT',
'Content-Length': len(body),
'Content-Type': 'text/html',
'Content-Encoding': contentencoding,
}
response = Response('http://scrapytest.org/', body=body, headers=headers)
response.request = Request('http://scrapytest.org', headers={'Accept-Encoding': 'gzip,deflate'})
return response
def test_process_request(self):
request = Request('http://scrapytest.org')
assert 'Accept-Encoding' not in request.headers
self.mw.process_request(request, self.spider)
self.assertEqual(request.headers.get('Accept-Encoding'), 'gzip,deflate')
def test_process_response_gzip(self):
response = self._getresponse('gzip')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'gzip')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_rawdeflate(self):
response = self._getresponse('rawdeflate')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'deflate')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_zlibdelate(self):
response = self._getresponse('zlibdeflate')
request = response.request
self.assertEqual(response.headers['Content-Encoding'], 'deflate')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith('<!DOCTYPE')
assert 'Content-Encoding' not in newresponse.headers
def test_process_response_plain(self):
response = Response('http://scrapytest.org', body='<!DOCTYPE...')
request = Request('http://scrapytest.org')
assert not response.headers.get('Content-Encoding')
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is response
assert newresponse.body.startswith('<!DOCTYPE')
def test_multipleencodings(self):
response = self._getresponse('gzip')
response.headers['Content-Encoding'] = ['uuencode', 'gzip']
request = response.request
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
self.assertEqual(newresponse.headers.getlist('Content-Encoding'), ['uuencode'])
def setUp(self):
self.spider = BaseSpider("foo")
self.mw = HttpCompressionMiddleware()
class HttpCompressionTest(TestCase):
def setUp(self):
self.spider = BaseSpider("foo")
self.mw = HttpCompressionMiddleware()
def _getresponse(self, coding):
if coding not in FORMAT:
raise ValueError()
samplefile, contentencoding = FORMAT[coding]
with open(join(SAMPLEDIR, samplefile), "rb") as sample:
body = sample.read()
headers = {
"Server": "Yaws/1.49 Yet Another Web Server",
"Date": "Sun, 08 Mar 2009 00:41:03 GMT",
"Content-Length": len(body),
"Content-Type": "text/html",
"Content-Encoding": contentencoding,
}
response = Response("http://scrapytest.org/", body=body, headers=headers)
response.request = Request("http://scrapytest.org", headers={"Accept-Encoding": "gzip,deflate"})
return response
def test_process_request(self):
request = Request("http://scrapytest.org")
assert "Accept-Encoding" not in request.headers
self.mw.process_request(request, self.spider)
self.assertEqual(request.headers.get("Accept-Encoding"), "x-gzip,gzip,deflate")
def test_process_response_gzip(self):
response = self._getresponse("gzip")
request = response.request
self.assertEqual(response.headers["Content-Encoding"], "gzip")
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith("<!DOCTYPE")
assert "Content-Encoding" not in newresponse.headers
def test_process_response_rawdeflate(self):
response = self._getresponse("rawdeflate")
request = response.request
self.assertEqual(response.headers["Content-Encoding"], "deflate")
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith("<!DOCTYPE")
assert "Content-Encoding" not in newresponse.headers
def test_process_response_zlibdelate(self):
response = self._getresponse("zlibdeflate")
request = response.request
self.assertEqual(response.headers["Content-Encoding"], "deflate")
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
assert newresponse.body.startswith("<!DOCTYPE")
assert "Content-Encoding" not in newresponse.headers
def test_process_response_plain(self):
response = Response("http://scrapytest.org", body="<!DOCTYPE...")
request = Request("http://scrapytest.org")
assert not response.headers.get("Content-Encoding")
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is response
assert newresponse.body.startswith("<!DOCTYPE")
def test_multipleencodings(self):
response = self._getresponse("gzip")
response.headers["Content-Encoding"] = ["uuencode", "gzip"]
request = response.request
newresponse = self.mw.process_response(request, response, self.spider)
assert newresponse is not response
self.assertEqual(newresponse.headers.getlist("Content-Encoding"), ["uuencode"])
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 process_response(self, request, response, spider):
response = _chuncked.process_response(request, response, spider)
return HttpCompressionMiddleware.process_response(
self, request, response, spider)
