Python pycurl.MAXREDIRS Examples

def url_check(self, url):
        '''下载地址检查'''

        url_info = {}
        proto = urlparse.urlparse(url)[0]
        if proto not in VALIDPROTOCOL:
            print 'Valid protocol should be http or ftp, but % s found < %s >!' % (proto, url)
        else:
            ss = StringIO()
            curl = pycurl.Curl()
            curl.setopt(pycurl.FOLLOWLOCATION, 1)
            curl.setopt(pycurl.MAXREDIRS, 5)
            curl.setopt(pycurl.CONNECTTIMEOUT, 30)
            curl.setopt(pycurl.TIMEOUT, 300)
            curl.setopt(pycurl.NOSIGNAL, 1)
            curl.setopt(pycurl.NOPROGRESS, 1)
            curl.setopt(pycurl.NOBODY, 1)
            curl.setopt(pycurl.HEADERFUNCTION, ss.write)
            curl.setopt(pycurl.URL, url)

        try:
            curl.perform()
        except:
            pass

        if curl.errstr() == '' and curl.getinfo(pycurl.RESPONSE_CODE) in STATUS_OK:
            url_info['url'] = curl.getinfo(pycurl.EFFECTIVE_URL)
            url_info['file'] = os.path.split(url_info['url'])[1]
            url_info['size'] = int(
                curl.getinfo(pycurl.CONTENT_LENGTH_DOWNLOAD))
            url_info['partible'] = (ss.getvalue().find('Accept - Ranges') != -1)

        return url_info 

def generate_curl(self, url=None, headers=None):
        """ 生成一个curl, 返回 curl 实例和用于获取结果的 buffer
        """
        curl = pycurl.Curl()
        buff = StringIO()

        curl.setopt(pycurl.COOKIEFILE, "cookie")
        curl.setopt(pycurl.COOKIEJAR, "cookie_jar")
        curl.setopt(pycurl.SHARE, self.http._share)
        curl.setopt(pycurl.WRITEFUNCTION, buff.write)
        curl.setopt(pycurl.FOLLOWLOCATION, 1)
        curl.setopt(pycurl.MAXREDIRS, 5)
        curl.setopt(pycurl.TIMEOUT, 3)
        curl.setopt(pycurl.CONNECTTIMEOUT, 3)

        if url:
            curl.setopt(pycurl.URL, url)

        if headers:
            self.set_curl_headers(curl, headers)

        return curl, buff 

def test_pycurl_insecure(self):
        curl = CurlStub(b"result")
        result = fetch("http://example.com/get-ca-cert", curl=curl,
                       insecure=True)
        self.assertEqual(result, b"result")
        self.assertEqual(curl.options,
                         {pycurl.URL: b"http://example.com/get-ca-cert",
                          pycurl.FOLLOWLOCATION: 1,
                          pycurl.MAXREDIRS: 5,
                          pycurl.CONNECTTIMEOUT: 30,
                          pycurl.LOW_SPEED_LIMIT: 1,
                          pycurl.LOW_SPEED_TIME: 600,
                          pycurl.NOSIGNAL: 1,
                          pycurl.WRITEFUNCTION: Any(),
                          pycurl.SSL_VERIFYPEER: False,
                          pycurl.DNS_CACHE_TIMEOUT: 0,
                          pycurl.ENCODING: b"gzip,deflate"}) 

def test_headers(self):
        curl = CurlStub(b"result")
        result = fetch("http://example.com",
                       headers={"a": "1", "b": "2"}, curl=curl)
        self.assertEqual(result, b"result")
        self.assertEqual(curl.options,
                         {pycurl.URL: b"http://example.com",
                          pycurl.FOLLOWLOCATION: 1,
                          pycurl.MAXREDIRS: 5,
                          pycurl.CONNECTTIMEOUT: 30,
                          pycurl.LOW_SPEED_LIMIT: 1,
                          pycurl.LOW_SPEED_TIME: 600,
                          pycurl.NOSIGNAL: 1,
                          pycurl.WRITEFUNCTION: Any(),
                          pycurl.HTTPHEADER: ["a: 1", "b: 2"],
                          pycurl.DNS_CACHE_TIMEOUT: 0,
                          pycurl.ENCODING: b"gzip,deflate"}) 

def test_cainfo(self):
        curl = CurlStub(b"result")
        result = fetch("https://example.com", cainfo="cainfo", curl=curl)
        self.assertEqual(result, b"result")
        self.assertEqual(curl.options,
                         {pycurl.URL: b"https://example.com",
                          pycurl.FOLLOWLOCATION: 1,
                          pycurl.MAXREDIRS: 5,
                          pycurl.CONNECTTIMEOUT: 30,
                          pycurl.LOW_SPEED_LIMIT: 1,
                          pycurl.LOW_SPEED_TIME: 600,
                          pycurl.NOSIGNAL: 1,
                          pycurl.WRITEFUNCTION: Any(),
                          pycurl.CAINFO: b"cainfo",
                          pycurl.DNS_CACHE_TIMEOUT: 0,
                          pycurl.ENCODING: b"gzip,deflate"}) 

def test_post_data(self):
        curl = CurlStub(b"result")
        result = fetch("http://example.com", post=True, data="data", curl=curl)
        self.assertEqual(result, b"result")
        self.assertEqual(curl.options[pycurl.READFUNCTION](), b"data")
        self.assertEqual(curl.options,
                         {pycurl.URL: b"http://example.com",
                          pycurl.FOLLOWLOCATION: 1,
                          pycurl.MAXREDIRS: 5,
                          pycurl.CONNECTTIMEOUT: 30,
                          pycurl.LOW_SPEED_LIMIT: 1,
                          pycurl.LOW_SPEED_TIME: 600,
                          pycurl.NOSIGNAL: 1,
                          pycurl.WRITEFUNCTION: Any(),
                          pycurl.POST: True,
                          pycurl.POSTFIELDSIZE: 4,
                          pycurl.READFUNCTION: Any(),
                          pycurl.DNS_CACHE_TIMEOUT: 0,
                          pycurl.ENCODING: b"gzip,deflate"}) 

def test_post(self):
        curl = CurlStub(b"result")
        result = fetch("http://example.com", post=True, curl=curl)
        self.assertEqual(result, b"result")
        self.assertEqual(curl.options,
                         {pycurl.URL: b"http://example.com",
                          pycurl.FOLLOWLOCATION: 1,
                          pycurl.MAXREDIRS: 5,
                          pycurl.CONNECTTIMEOUT: 30,
                          pycurl.LOW_SPEED_LIMIT: 1,
                          pycurl.LOW_SPEED_TIME: 600,
                          pycurl.NOSIGNAL: 1,
                          pycurl.WRITEFUNCTION: Any(),
                          pycurl.POST: True,
                          pycurl.DNS_CACHE_TIMEOUT: 0,
                          pycurl.ENCODING: b"gzip,deflate"}) 

def test_basic(self):
        curl = CurlStub(b"result")
        result = fetch("http://example.com", curl=curl)
        self.assertEqual(result, b"result")
        self.assertEqual(curl.options,
                         {pycurl.URL: b"http://example.com",
                          pycurl.FOLLOWLOCATION: 1,
                          pycurl.MAXREDIRS: 5,
                          pycurl.CONNECTTIMEOUT: 30,
                          pycurl.LOW_SPEED_LIMIT: 1,
                          pycurl.LOW_SPEED_TIME: 600,
                          pycurl.NOSIGNAL: 1,
                          pycurl.WRITEFUNCTION: Any(),
                          pycurl.DNS_CACHE_TIMEOUT: 0,
                          pycurl.ENCODING: b"gzip,deflate"}) 

def _set_common(self, c):
        c.setopt(pycurl.FOLLOWLOCATION, 1)
        c.setopt(pycurl.AUTOREFERER, 1)
        c.setopt(pycurl.ENCODING, self._accept_encoding)
        c.setopt(pycurl.MAXREDIRS, 255)
        c.setopt(pycurl.CONNECTTIMEOUT, 30)
        c.setopt(pycurl.TIMEOUT, 300)
        c.setopt(pycurl.NOSIGNAL, 1)
        if self._proxy:
            c.setopt(pycurl.PROXY, self._proxy)
        if self._url.scheme == 'https':
            c.setopt(pycurl.SSLVERSION, 3)
            c.setopt(pycurl.SSL_VERIFYPEER, 0) 

def test_create_curl(self):
        curls = []

        def pycurl_Curl():
            curl = CurlStub(b"result")
            curls.append(curl)
            return curl
        Curl = pycurl.Curl
        try:
            pycurl.Curl = pycurl_Curl
            result = fetch("http://example.com")
            curl = curls[0]
            self.assertEqual(result, b"result")
            self.assertEqual(curl.options,
                             {pycurl.URL: b"http://example.com",
                              pycurl.FOLLOWLOCATION: 1,
                              pycurl.MAXREDIRS: 5,
                              pycurl.CONNECTTIMEOUT: 30,
                              pycurl.LOW_SPEED_LIMIT: 1,
                              pycurl.LOW_SPEED_TIME: 600,
                              pycurl.NOSIGNAL: 1,
                              pycurl.WRITEFUNCTION: Any(),
                              pycurl.DNS_CACHE_TIMEOUT: 0,
                              pycurl.ENCODING: b"gzip,deflate"})
        finally:
            pycurl.Curl = Curl 

def secure_download(url, cainfo=""):
    import pycurl
    import io

    c = pycurl.Curl()
    c.setopt(pycurl.URL, url.encode("ascii"))

    # -k / --insecure
    # c.setopt(pycurl.SSL_VERIFYPEER, 0)

    # Verify certificate
    c.setopt(pycurl.SSL_VERIFYPEER, 1)

    # Verify CommonName or Subject Alternate Name
    c.setopt(pycurl.SSL_VERIFYHOST, 2)

    # --cacert
    if cainfo:
        c.setopt(pycurl.CAINFO, cainfo)

    res = io.StringIO()

    c.setopt(pycurl.WRITEFUNCTION, res.write)

    # follow up to 10 http location: headers
    c.setopt(pycurl.FOLLOWLOCATION, 1)
    c.setopt(pycurl.MAXREDIRS, 10)

    c.perform()
    c.close()
    data = res.getvalue()
    res.close()

    return data 

def test_get_html_http_no_callbacks(self):
        import pycurl
        import io

        # Test http with no callbacks
        mock_curl_instance = Mock()
        mock_stringio_instance = Mock()

        mock_stringio_instance.getvalue.return_value = "mock StringIO value"

        # Expected pycurl setopt calls
        expected_setopt_calls = [
            call(pycurl.URL, b"https://example.com/file.extension"),
            call(pycurl.WRITEFUNCTION, mock_stringio_instance.write),
            call(pycurl.FOLLOWLOCATION, 1),
            call(pycurl.MAXREDIRS, 10),
            call(
                pycurl.USERAGENT,
                "Fedora Upstream Release Monitoring "
                "(https://fedoraproject.org/wiki/Upstream_release_monitoring)",
            ),
            call(pycurl.CONNECTTIMEOUT, 10),
            call(pycurl.TIMEOUT, 30),
        ]

        with patch.object(pycurl, "Curl") as mock_curl_constructor:
            with patch.object(io, "StringIO") as mock_stringio_constructor:
                mock_curl_constructor.return_value = mock_curl_instance
                mock_stringio_constructor.return_value = mock_stringio_instance

                get_html_result = helpers.get_html("https://example.com/file.extension")

        # Check curl calls
        mock_curl_instance.setopt.assert_has_calls(expected_setopt_calls)
        mock_curl_instance.perform.assert_called_once_with()
        mock_curl_instance.close.assert_called_once_with()

        self.assertEqual(get_html_result, "mock StringIO value") 

def test_secure_download(self):
        import pycurl
        import io

        # Required mocks
        mock_curl_instance = Mock()
        mock_stringio_instance = Mock()

        mock_stringio_instance.getvalue.return_value = "mock StringIO value"

        # Expected pycurl setopt calls
        expected_setopt_calls = [
            call(pycurl.URL, b"https://example.com/file.extension"),
            call(pycurl.SSL_VERIFYPEER, 1),
            call(pycurl.SSL_VERIFYHOST, 2),
            call(pycurl.CAINFO, "/etc/certs/cabundle.pem"),
            call(pycurl.WRITEFUNCTION, mock_stringio_instance.write),
            call(pycurl.FOLLOWLOCATION, 1),
            call(pycurl.MAXREDIRS, 10),
        ]

        with patch.object(pycurl, "Curl") as mock_curl_constructor:
            with patch.object(io, "StringIO") as mock_stringio_constructor:
                mock_curl_constructor.return_value = mock_curl_instance
                mock_stringio_constructor.return_value = mock_stringio_instance

                secure_download_result = helpers.secure_download(
                    "https://example.com/file.extension", "/etc/certs/cabundle.pem"
                )

        # Check curl calls
        mock_curl_instance.setopt.assert_has_calls(expected_setopt_calls)
        mock_curl_instance.perform.assert_called_once_with()
        mock_curl_instance.close.assert_called_once_with()

        # Check StringIO call
        mock_stringio_instance.getvalue.assert_called_once_with()

        self.assertEqual(secure_download_result, "mock StringIO value") 

def test_secure_download_no_cainfo(self):
        import pycurl
        import io

        # Required mocks
        mock_curl_instance = Mock()
        mock_stringio_instance = Mock()

        mock_stringio_instance.getvalue.return_value = "mock StringIO value"

        mock_stringio_instance.getvalue.return_value = "mock StringIO value"

        expected_setopt_calls = [
            call(pycurl.URL, b"https://example.com/file.extension"),
            call(pycurl.SSL_VERIFYPEER, 1),
            call(pycurl.SSL_VERIFYHOST, 2),
            call(pycurl.WRITEFUNCTION, mock_stringio_instance.write),
            call(pycurl.FOLLOWLOCATION, 1),
            call(pycurl.MAXREDIRS, 10),
        ]

        with patch.object(pycurl, "Curl") as mock_curl_constructor:
            with patch.object(io, "StringIO") as mock_stringio_constructor:
                mock_curl_constructor.return_value = mock_curl_instance
                mock_stringio_constructor.return_value = mock_stringio_instance

                secure_download_result = helpers.secure_download(
                    "https://example.com/file.extension"
                )

        # Check curl calls
        mock_curl_instance.setopt.assert_has_calls(expected_setopt_calls)
        mock_curl_instance.perform.assert_called_once_with()
        mock_curl_instance.close.assert_called_once_with()

        # Check StringIO call
        mock_stringio_instance.getvalue.assert_called_once_with()

        self.assertEqual(secure_download_result, "mock StringIO value") 

def __init__(self, url):
        self.curl = pycurl.Curl()
        self.curl.setopt(pycurl.FOLLOWLOCATION, 1)
        self.curl.setopt(pycurl.MAXREDIRS, 5)
        self.curl.setopt(pycurl.CONNECTTIMEOUT, 30)
        self.curl.setopt(pycurl.TIMEOUT, 300)
        self.curl.setopt(pycurl.NOSIGNAL, 1)
        self.curl.setopt(pycurl.WRITEFUNCTION, self.write_cb)
        self.curl.setopt(pycurl.URL, url)
        self.curl.connection = self

        # 合计下载字节数
        self.total_downloaded = 0 

def __init__(self, base_url="", fakeheaders=[]):
        self.handle = pycurl.Curl()
        # These members might be set.
        self.set_url(base_url)
        self.verbosity = 0
        self.fakeheaders = fakeheaders
        # Nothing past here should be modified by the caller.
        self.payload = None
        self.payload_io = BytesIO()
        self.hrd = ""
        # Verify that we've got the right site; harmless on a non-SSL connect.
        self.set_option(pycurl.SSL_VERIFYHOST, 2)
        # Follow redirects in case it wants to take us to a CGI...
        self.set_option(pycurl.FOLLOWLOCATION, 1)
        self.set_option(pycurl.MAXREDIRS, 5)
        self.set_option(pycurl.NOSIGNAL, 1)
        # Setting this option with even a nonexistent file makes libcurl
        # handle cookie capture and playback automatically.
        self.set_option(pycurl.COOKIEFILE, "/dev/null")
        # Set timeouts to avoid hanging too long
        self.set_timeout(30)
        # Use password identification from .netrc automatically
        self.set_option(pycurl.NETRC, 1)
        self.set_option(pycurl.WRITEFUNCTION, self.payload_io.write)
        def header_callback(x):
            self.hdr += x.decode('ascii')
        self.set_option(pycurl.HEADERFUNCTION, header_callback) 

def __init__(self, url, auth, verify):
        self.url = url
        self.received_buffer = BytesIO()

        headers = ['Cache-Control: no-cache', 'Accept: text/event-stream']

        self.curl = pycurl.Curl()
        self.curl.setopt(pycurl.URL, url)
        self.curl.setopt(pycurl.ENCODING, 'gzip')
        self.curl.setopt(pycurl.CONNECTTIMEOUT, 10)
        self.curl.setopt(pycurl.WRITEDATA, self.received_buffer)

        # Marathon >= 1.7.x returns 30x responses for /v2/events responses
        # when they're coming from a non-leader. So we follow redirects.
        self.curl.setopt(pycurl.FOLLOWLOCATION, True)
        self.curl.setopt(pycurl.MAXREDIRS, 1)
        self.curl.setopt(pycurl.UNRESTRICTED_AUTH, True)

        # The below settings are to prevent the connection from hanging if the
        # connection breaks silently. Since marathon-lb only listens, silent
        # connection failure results in marathon-lb waiting infinitely.
        #
        # Minimum bytes/second below which it is considered "low speed". So
        # "low speed" here refers to 0 bytes/second.
        self.curl.setopt(pycurl.LOW_SPEED_LIMIT, 1)
        # How long (in seconds) it's allowed to go below the speed limit
        # before it times out
        self.curl.setopt(pycurl.LOW_SPEED_TIME, 300)

        if auth and type(auth) is DCOSAuth:
            auth.refresh_auth_header()
            headers.append('Authorization: %s' % auth.auth_header)
        elif auth:
            self.curl.setopt(pycurl.HTTPAUTH, pycurl.HTTPAUTH_BASIC)
            self.curl.setopt(pycurl.USERPWD, '%s:%s' % auth)
        if verify:
            self.curl.setopt(pycurl.CAINFO, verify)
        else:
            self.curl.setopt(pycurl.SSL_VERIFYHOST, 0)
            self.curl.setopt(pycurl.SSL_VERIFYPEER, 0)

        self.curl.setopt(pycurl.HTTPHEADER, headers)

        self.curlmulti = pycurl.CurlMulti()
        self.curlmulti.add_handle(self.curl)

        self.status_code = 0 

def run(self):
        c = pycurl.Curl()
        c.setopt(pycurl.URL, self.url)
        c.setopt(pycurl.FOLLOWLOCATION, True)
        c.setopt(pycurl.MAXREDIRS, 4)
        #c.setopt(pycurl.NOBODY, 1)

        c.setopt(c.VERBOSE, True)

        #c.setopt(pycurl.CONNECTTIMEOUT, 20)
        
        if self.useragent:
            c.setopt(pycurl.USERAGENT, self.useragent)
        
        # add cookies, if available
        if self.cookies:
            c.setopt(pycurl.COOKIE, self.cookies)
        
        #realurl = c.getinfo(pycurl.EFFECTIVE_URL)
        realurl = self.url
        print("realurl",realurl)
        self.filename = realurl.split("/")[-1].strip()
        c.setopt(pycurl.URL, realurl)
        c.setopt(pycurl.FOLLOWLOCATION, True)
        c.setopt(pycurl.NOPROGRESS, False)
        c.setopt(pycurl.XFERINFOFUNCTION, self.getProgress)
        
        c.setopt(pycurl.SSL_VERIFYPEER, False)
        c.setopt(pycurl.SSL_VERIFYHOST, False)
        
        # configure pycurl output file
        if self.path == False:
            self.path = os.getcwd()
        filepath = os.path.join(self.path, self.filename)
        
         
        if os.path.exists(filepath):## remove old file,restart download 
            os.system("rm -rf " + filepath)
        
        buffer = StringIO() 
        c.setopt(pycurl.WRITEDATA, buffer)
        
        self._dst_path = filepath

        # add cookies, if available
        if self.cookies:
            c.setopt(pycurl.COOKIE, self.cookies)
        
        self._stop = False
        self.progress["stopped"] = False 
        # download file
        try:
            c.perform()
        except pycurl.error, error:
            errno,errstr = error
            print("curl error: %s" % errstr)
            self._errors.append(errstr)
            self._stop = True
            self.progress["stopped"] = True
            self.stop() 

def get_html(url, callback=None, errback=None):
    if url.startswith("ftp://"):
        import urllib

        req = urllib.request.urlopen(url)  # nosec
        data = req.read()
        if callback:
            try:
                for cb in callback:
                    cb(data)
            except TypeError:
                callback(data)
        return data
    else:
        if callback:
            from twisted.web.client import getPage

            df = getPage(url)
            try:
                for cb in callback:
                    df.addCallback(cb)
            except TypeError:
                df.addCallback(callback)

            if errback:
                try:
                    for eb in errback:
                        df.addErrback(eb)
                except TypeError:
                    df.addErrback(errback)
        else:
            import io
            import pycurl

            res = io.StringIO()

            c = pycurl.Curl()
            c.setopt(pycurl.URL, url.encode("ascii"))

            c.setopt(pycurl.WRITEFUNCTION, res.write)
            c.setopt(pycurl.FOLLOWLOCATION, 1)
            c.setopt(pycurl.MAXREDIRS, 10)
            c.setopt(
                pycurl.USERAGENT,
                "Fedora Upstream Release Monitoring "
                "(https://fedoraproject.org/wiki/Upstream_release_monitoring)",
            )
            c.setopt(pycurl.CONNECTTIMEOUT, 10)
            c.setopt(pycurl.TIMEOUT, 30)

            c.perform()
            c.close()

            data = res.getvalue()
            res.close()
            return data