def _on_connect(self):
self._remove_timeout()
if self.request.request_timeout:
self._timeout = self.io_loop.add_timeout(
self.start_time + self.request.request_timeout,
stack_context.wrap(self._on_timeout))
if (self.request.method not in self._SUPPORTED_METHODS and
not self.request.allow_nonstandard_methods):
raise KeyError("unknown method %s" % self.request.method)
for key in ('network_interface',
'proxy_host', 'proxy_port',
'proxy_username', 'proxy_password'):
if getattr(self.request, key, None):
raise NotImplementedError('%s not supported' % key)
if "Connection" not in self.request.headers:
self.request.headers["Connection"] = "close"
if "Host" not in self.request.headers:
if '@' in self.parsed.netloc:
self.request.headers["Host"] = self.parsed.netloc.rpartition('@')[-1]
else:
self.request.headers["Host"] = self.parsed.netloc
username, password = None, None
if self.parsed.username is not None:
username, password = self.parsed.username, self.parsed.password
elif self.request.auth_username is not None:
username = self.request.auth_username
password = self.request.auth_password or ''
if username is not None:
auth = utf8(username) + b":" + utf8(password)
self.request.headers["Authorization"] = (b"Basic " +
base64.b64encode(auth))
if self.request.user_agent:
self.request.headers["User-Agent"] = self.request.user_agent
if not self.request.allow_nonstandard_methods:
if self.request.method in ("POST", "PATCH", "PUT"):
assert self.request.body is not None
else:
assert self.request.body is None
if self.request.body is not None:
self.request.headers["Content-Length"] = str(len(
self.request.body))
if (self.request.method == "POST" and
"Content-Type" not in self.request.headers):
self.request.headers["Content-Type"] = "application/x-www-form-urlencoded"
if self.request.use_gzip:
self.request.headers["Accept-Encoding"] = "gzip"
req_path = ((self.parsed.path or '/') +
(('?' + self.parsed.query) if self.parsed.query else ''))
request_lines = [utf8("%s %s HTTP/1.1" % (self.request.method,
req_path))]
for k, v in self.request.headers.get_all():
line = utf8(k) + b": " + utf8(v)
if b'\n' in line:
raise ValueError('Newline in header: ' + repr(line))
request_lines.append(line)
self.stream.write(b"\r\n".join(request_lines) + b"\r\n\r\n")
if self.request.body is not None:
self.stream.write(self.request.body)
self.stream.read_until_regex(b"\r?\n\r?\n", self._on_headers)
def parse_image(self, url, filename = None):
"""处理img标签的src并映射到本地文件"""
url = escape.utf8(url)
image_guid = hashlib.sha1(url).hexdigest()
x = url.split('.')
ext = 'jpg'
if len(x) > 1:
ext = x[-1]
if len(ext) > 4:
ext = ext[0:3]
ext = re.sub('[^a-zA-Z]', '', ext)
ext = ext.lower()
if ext not in ['jpg', 'jpeg', 'gif', 'png', 'bmp']:
ext = 'jpg'
y = url.split('/')
h = hashlib.sha1(str(y[2])).hexdigest()
hash_dir = os.path.join(h[0:1], h[1:2])
filename = image_guid + '.' + ext
img_dir = os.path.join(self.work_dir, 'data', 'images', hash_dir)
fullname = os.path.join(img_dir, filename)
if not os.path.exists(img_dir):
os.makedirs(img_dir)
localimage = 'images/%s/%s' % (hash_dir, filename)
return localimage, fullname
def parse_image(self, url, filename=None):
"""处理img标签的src并映射到本地文件"""
url = escape.utf8(url)
image_guid = hashlib.sha1(url).hexdigest()
x = url.split('.')
ext = 'jpg'
if len(x) > 1:
ext = x[-1]
if len(ext) > 4:
ext = ext[0:3]
ext = re.sub('[^a-zA-Z]', '', ext)
ext = ext.lower()
if ext not in ['jpg', 'jpeg', 'gif', 'png', 'bmp']:
ext = 'jpg'
y = url.split('/')
h = hashlib.sha1(str(y[2])).hexdigest()
hash_dir = os.path.join(h[0:1], h[1:2])
filename = image_guid + '.' + ext
img_dir = os.path.join(krconfig.work_dir, 'data', 'images', hash_dir)
fullname = os.path.join(img_dir, filename)
if not os.path.exists(img_dir):
os.makedirs(img_dir)
localimage = 'images/%s/%s' % (hash_dir, filename)
return localimage, fullname
def environ(request):
"""Converts a `tornado.httpserver.HTTPRequest` to a WSGI environment.
"""
hostport = request.host.split(":")
if len(hostport) == 2:
host = hostport[0]
port = int(hostport[1])
else:
host = request.host
port = 443 if request.protocol == "https" else 80
environ = {
"REQUEST_METHOD": request.method,
"SCRIPT_NAME": "",
"PATH_INFO": to_wsgi_str(escape.url_unescape(request.path, encoding=None)),
"QUERY_STRING": request.query,
"REMOTE_ADDR": request.remote_ip,
"SERVER_NAME": host,
"SERVER_PORT": str(port),
"SERVER_PROTOCOL": request.version,
"wsgi.version": (1, 0),
"wsgi.url_scheme": request.protocol,
"wsgi.input": BytesIO(escape.utf8(request.body)),
"wsgi.errors": sys.stderr,
"wsgi.multithread": False,
"wsgi.multiprocess": True,
"wsgi.run_once": False,
}
if "Content-Type" in request.headers:
environ["CONTENT_TYPE"] = request.headers.pop("Content-Type")
if "Content-Length" in request.headers:
environ["CONTENT_LENGTH"] = request.headers.pop("Content-Length")
for key, value in request.headers.items():
environ["HTTP_" + key.replace("-", "_").upper()] = value
return environ
def resolve(self, host, port, family=0):
# getHostByName doesn't accept IP addresses, so if the input
# looks like an IP address just return it immediately.
if twisted.internet.abstract.isIPAddress(host):
resolved = host
resolved_family = socket.AF_INET
elif twisted.internet.abstract.isIPv6Address(host):
resolved = host
resolved_family = socket.AF_INET6
else:
deferred = self.resolver.getHostByName(utf8(host))
resolved = yield gen.Task(deferred.addCallback)
if twisted.internet.abstract.isIPAddress(resolved):
resolved_family = socket.AF_INET
elif twisted.internet.abstract.isIPv6Address(resolved):
resolved_family = socket.AF_INET6
else:
resolved_family = socket.AF_UNSPEC
if family != socket.AF_UNSPEC and family != resolved_family:
raise Exception('Requested socket family %d but got %d' %
(family, resolved_family))
result = [
(resolved_family, (resolved, port)),
]
raise gen.Return(result)
def down_image(self, url, referer=None, filename=None):
"""download image"""
print "download: %s" % url,
url = escape.utf8(url)
image_guid = hashlib.sha1(url).hexdigest()
convert_image = self.get_config('reader', 'convert_image')
x = url.split('.')
ext = 'jpg'
if len(x) > 1:
ext = x[-1]
if len(ext) > 4:
ext = ext[0:3]
ext = re.sub('[^a-zA-Z]','', ext)
ext = ext.lower()
if ext not in ['jpg', 'jpeg', 'gif','png','bmp']:
ext = 'jpg'
y = url.split('/')
h = hashlib.sha1(str(y[2])).hexdigest()
hash_dir = os.path.join(h[0:1], h[1:2])
filename = image_guid + '.' + ext
img_dir = os.path.join(self.work_dir, 'data', 'images', hash_dir)
fullname = os.path.join(img_dir, filename)
localimage = 'images/%s/%s' % (hash_dir, filename)
if os.path.isfile(fullname) is False:
if not os.path.exists(img_dir):
os.makedirs( img_dir )
try:
req = urllib2.Request(url)
req.add_header('User-Agent', self.user_agent)
req.add_header('Accept-Language', 'zh-cn,zh;q=0.7,nd;q=0.3')
if referer is not None:
req.add_header('Referer', referer)
response = urllib2.urlopen(req)
localFile = open(fullname, 'wb')
localFile.write(response.read())
response.close()
localFile.close()
if convert_image :
os.system('convert -colorspace gray \'%s[0]\' \
%s'% (fullname, fullname)
)
print "done."
except Exception, e:
print 'fail: %s' % e
localimage = False
finally:
def down_image(self, url, referer=None, filename=None):
"""download image"""
print "download: %s" % url,
url = escape.utf8(url)
image_guid = hashlib.sha1(url).hexdigest()
convert_image = self.get_config('reader', 'convert_image')
x = url.split('.')
ext = 'jpg'
if len(x) > 1:
ext = x[-1]
if len(ext) > 4:
ext = ext[0:3]
ext = re.sub('[^a-zA-Z]', '', ext)
ext = ext.lower()
if ext not in ['jpg', 'jpeg', 'gif', 'png', 'bmp']:
ext = 'jpg'
y = url.split('/')
h = hashlib.sha1(str(y[2])).hexdigest()
hash_dir = os.path.join(h[0:1], h[1:2])
filename = image_guid + '.' + ext
img_dir = os.path.join(self.work_dir, 'data', 'images', hash_dir)
fullname = os.path.join(img_dir, filename)
localimage = 'images/%s/%s' % (hash_dir, filename)
if os.path.isfile(fullname) is False:
if not os.path.exists(img_dir):
os.makedirs(img_dir)
try:
req = urllib2.Request(url)
req.add_header('User-Agent', self.user_agent)
req.add_header('Accept-Language', 'zh-cn,zh;q=0.7,nd;q=0.3')
if referer is not None:
req.add_header('Referer', referer)
response = urllib2.urlopen(req)
localFile = open(fullname, 'wb')
localFile.write(response.read())
response.close()
localFile.close()
if convert_image:
os.system('convert -colorspace gray \'%s[0]\' \
%s' % (fullname, fullname))
print "done."
except Exception, e:
print 'fail: %s' % e
localimage = False
finally:
def generate(self, writer):
value = self.value
# Compress lots of white space to a single character. If the whitespace
# breaks a line, have it continue to break a line, but just with a
# single \n character
if writer.compress_whitespace and "<pre>" not in value:
value = re.sub(r"([\t ]+)", " ", value)
value = re.sub(r"(\s*\n\s*)", "\n", value)
if value:
writer.write_line('_append(%r)' % escape.utf8(value), self.line)
def __call__(self, request):
data = {}
response = []
def start_response(status, response_headers, exc_info=None):
data["status"] = status
data["headers"] = response_headers
return response.append
app_response = self.wsgi_application(
WSGIContainer.environ(request), start_response)
response.extend(app_response)
body = b"".join(response)
if hasattr(app_response, "close"):
app_response.close()
if not data:
raise Exception("WSGI app did not call start_response")
status_code = int(data["status"].split()[0])
headers = data["headers"]
header_set = set(k.lower() for (k, v) in headers)
body = escape.utf8(body)
if status_code != 304:
if "content-length" not in header_set:
headers.append(("Content-Length", str(len(body))))
if "content-type" not in header_set:
headers.append(("Content-Type", "text/html; charset=UTF-8"))
if "server" not in header_set:
headers.append(("Server", "TornadoServer/%s" % tornado.version))
parts = [escape.utf8("HTTP/1.1 " + data["status"] + "\r\n")]
for key, value in headers:
parts.append(escape.utf8(key) + b": " + escape.utf8(value) + b"\r\n")
parts.append(b"\r\n")
parts.append(body)
request.write(b"".join(parts))
request.finish()
self._log(status_code, request)
def parse_body_arguments(content_type, body, arguments, files):
"""Parses a form request body.
Supports ``application/x-www-form-urlencoded`` and
``multipart/form-data``. The ``content_type`` parameter should be
a string and ``body`` should be a byte string. The ``arguments``
and ``files`` parameters are dictionaries that will be updated
with the parsed contents.
"""
if content_type.startswith("application/x-www-form-urlencoded"):
uri_arguments = parse_qs_bytes(native_str(body), keep_blank_values=True)
for name, values in uri_arguments.items():
if values:
arguments.setdefault(name, []).extend(values)
elif content_type.startswith("multipart/form-data"):
fields = content_type.split(";")
for field in fields:
k, sep, v = field.strip().partition("=")
if k == "boundary" and v:
parse_multipart_form_data(utf8(v), body, arguments, files)
break
else:
gen_log.warning("Invalid multipart/form-data")
def __init__(self, url, method="GET", headers=None, body=None,
auth_username=None, auth_password=None,
connect_timeout=None, request_timeout=None,
if_modified_since=None, follow_redirects=None,
max_redirects=None, user_agent=None, use_gzip=None,
network_interface=None, streaming_callback=None,
header_callback=None, prepare_curl_callback=None,
proxy_host=None, proxy_port=None, proxy_username=None,
proxy_password=None, allow_nonstandard_methods=None,
validate_cert=None, ca_certs=None,
allow_ipv6=None,
client_key=None, client_cert=None):
r"""All parameters except ``url`` are optional.
:arg string url: URL to fetch
:arg string method: HTTP method, e.g. "GET" or "POST"
:arg headers: Additional HTTP headers to pass on the request
:type headers: `~tornado.httputil.HTTPHeaders` or `dict`
:arg string auth_username: Username for HTTP "Basic" authentication
:arg string auth_password: Password for HTTP "Basic" authentication
:arg float connect_timeout: Timeout for initial connection in seconds
:arg float request_timeout: Timeout for entire request in seconds
:arg if_modified_since: Timestamp for ``If-Modified-Since`` header
:type if_modified_since: `datetime` or `float`
:arg bool follow_redirects: Should redirects be followed automatically
or return the 3xx response?
:arg int max_redirects: Limit for ``follow_redirects``
:arg string user_agent: String to send as ``User-Agent`` header
:arg bool use_gzip: Request gzip encoding from the server
:arg string network_interface: Network interface to use for request
:arg callable streaming_callback: If set, ``streaming_callback`` will
be run with each chunk of data as it is received, and
``HTTPResponse.body`` and ``HTTPResponse.buffer`` will be empty in
the final response.
:arg callable header_callback: If set, ``header_callback`` will
be run with each header line as it is received (including the
first line, e.g. ``HTTP/1.0 200 OK\r\n``, and a final line
containing only ``\r\n``. All lines include the trailing newline
characters). ``HTTPResponse.headers`` will be empty in the final
response. This is most useful in conjunction with
``streaming_callback``, because it's the only way to get access to
header data while the request is in progress.
:arg callable prepare_curl_callback: If set, will be called with
a ``pycurl.Curl`` object to allow the application to make additional
``setopt`` calls.
:arg string proxy_host: HTTP proxy hostname. To use proxies,
``proxy_host`` and ``proxy_port`` must be set; ``proxy_username`` and
``proxy_pass`` are optional. Proxies are currently only supported
with ``curl_httpclient``.
:arg int proxy_port: HTTP proxy port
:arg string proxy_username: HTTP proxy username
:arg string proxy_password: HTTP proxy password
:arg bool allow_nonstandard_methods: Allow unknown values for ``method``
argument?
:arg bool validate_cert: For HTTPS requests, validate the server's
certificate?
:arg string ca_certs: filename of CA certificates in PEM format,
or None to use defaults. Note that in ``curl_httpclient``, if
any request uses a custom ``ca_certs`` file, they all must (they
don't have to all use the same ``ca_certs``, but it's not possible
to mix requests with ``ca_certs`` and requests that use the defaults.
:arg bool allow_ipv6: Use IPv6 when available? Default is false in
``simple_httpclient`` and true in ``curl_httpclient``
:arg string client_key: Filename for client SSL key, if any
:arg string client_cert: Filename for client SSL certificate, if any
"""
if headers is None:
headers = httputil.HTTPHeaders()
if if_modified_since:
headers["If-Modified-Since"] = httputil.format_timestamp(
if_modified_since)
self.proxy_host = proxy_host
self.proxy_port = proxy_port
self.proxy_username = proxy_username
self.proxy_password = proxy_password
self.url = url
self.method = method
self.headers = headers
self.body = utf8(body)
self.auth_username = auth_username
self.auth_password = auth_password
self.connect_timeout = connect_timeout
self.request_timeout = request_timeout
self.follow_redirects = follow_redirects
self.max_redirects = max_redirects
self.user_agent = user_agent
self.use_gzip = use_gzip
self.network_interface = network_interface
self.streaming_callback = stack_context.wrap(streaming_callback)
self.header_callback = stack_context.wrap(header_callback)
self.prepare_curl_callback = stack_context.wrap(prepare_curl_callback)
self.allow_nonstandard_methods = allow_nonstandard_methods
self.validate_cert = validate_cert
self.ca_certs = ca_certs
self.allow_ipv6 = allow_ipv6
self.client_key = client_key
self.client_cert = client_cert
self.start_time = time.time()
def _curl_setup_request(curl, request, buffer, headers):
curl.setopt(pycurl.URL, native_str(request.url))
# libcurl's magic "Expect: 100-continue" behavior causes delays
# with servers that don't support it (which include, among others,
# Google's OpenID endpoint). Additionally, this behavior has
# a bug in conjunction with the curl_multi_socket_action API
# (https://sourceforge.net/tracker/?func=detail&atid=100976&aid=3039744&group_id=976),
# which increases the delays. It's more trouble than it's worth,
# so just turn off the feature (yes, setting Expect: to an empty
# value is the official way to disable this)
if "Expect" not in request.headers:
request.headers["Expect"] = ""
# libcurl adds Pragma: no-cache by default; disable that too
if "Pragma" not in request.headers:
request.headers["Pragma"] = ""
# Request headers may be either a regular dict or HTTPHeaders object
if isinstance(request.headers, httputil.HTTPHeaders):
curl.setopt(pycurl.HTTPHEADER,
[native_str("%s: %s" % i) for i in request.headers.get_all()])
else:
curl.setopt(pycurl.HTTPHEADER,
[native_str("%s: %s" % i) for i in request.headers.items()])
if request.header_callback:
curl.setopt(pycurl.HEADERFUNCTION, request.header_callback)
else:
curl.setopt(pycurl.HEADERFUNCTION,
lambda line: _curl_header_callback(headers, line))
if request.streaming_callback:
write_function = request.streaming_callback
else:
write_function = buffer.write
if bytes_type is str: # py2
curl.setopt(pycurl.WRITEFUNCTION, write_function)
else: # py3
# Upstream pycurl doesn't support py3, but ubuntu 12.10 includes
# a fork/port. That version has a bug in which it passes unicode
# strings instead of bytes to the WRITEFUNCTION. This means that
# if you use a WRITEFUNCTION (which tornado always does), you cannot
# download arbitrary binary data. This needs to be fixed in the
# ported pycurl package, but in the meantime this lambda will
# make it work for downloading (utf8) text.
curl.setopt(pycurl.WRITEFUNCTION, lambda s: write_function(utf8(s)))
curl.setopt(pycurl.FOLLOWLOCATION, request.follow_redirects)
curl.setopt(pycurl.MAXREDIRS, request.max_redirects)
curl.setopt(pycurl.CONNECTTIMEOUT_MS, int(1000 * request.connect_timeout))
curl.setopt(pycurl.TIMEOUT_MS, int(1000 * request.request_timeout))
if request.user_agent:
curl.setopt(pycurl.USERAGENT, native_str(request.user_agent))
else:
curl.setopt(pycurl.USERAGENT, "Mozilla/5.0 (compatible; pycurl)")
if request.network_interface:
curl.setopt(pycurl.INTERFACE, request.network_interface)
if request.use_gzip:
curl.setopt(pycurl.ENCODING, "gzip,deflate")
else:
curl.setopt(pycurl.ENCODING, "none")
if request.proxy_host and request.proxy_port:
curl.setopt(pycurl.PROXY, request.proxy_host)
curl.setopt(pycurl.PROXYPORT, request.proxy_port)
if request.proxy_username:
credentials = '%s:%s' % (request.proxy_username,
request.proxy_password)
curl.setopt(pycurl.PROXYUSERPWD, credentials)
else:
curl.setopt(pycurl.PROXY, '')
if request.validate_cert:
curl.setopt(pycurl.SSL_VERIFYPEER, 1)
curl.setopt(pycurl.SSL_VERIFYHOST, 2)
else:
curl.setopt(pycurl.SSL_VERIFYPEER, 0)
curl.setopt(pycurl.SSL_VERIFYHOST, 0)
if request.ca_certs is not None:
curl.setopt(pycurl.CAINFO, request.ca_certs)
else:
# There is no way to restore pycurl.CAINFO to its default value
# (Using unsetopt makes it reject all certificates).
# I don't see any way to read the default value from python so it
# can be restored later. We'll have to just leave CAINFO untouched
# if no ca_certs file was specified, and require that if any
# request uses a custom ca_certs file, they all must.
pass
if request.allow_ipv6 is False:
# Curl behaves reasonably when DNS resolution gives an ipv6 address
# that we can't reach, so allow ipv6 unless the user asks to disable.
# (but see version check in _process_queue above)
curl.setopt(pycurl.IPRESOLVE, pycurl.IPRESOLVE_V4)
# Set the request method through curl's irritating interface which makes
# up names for almost every single method
curl_options = {
"GET": pycurl.HTTPGET,
"POST": pycurl.POST,
"PUT": pycurl.UPLOAD,
"HEAD": pycurl.NOBODY,
}
custom_methods = set(["DELETE"])
for o in curl_options.values():
curl.setopt(o, False)
if request.method in curl_options:
curl.unsetopt(pycurl.CUSTOMREQUEST)
curl.setopt(curl_options[request.method], True)
elif request.allow_nonstandard_methods or request.method in custom_methods:
curl.setopt(pycurl.CUSTOMREQUEST, request.method)
else:
raise KeyError('unknown method ' + request.method)
# Handle curl's cryptic options for every individual HTTP method
if request.method in ("POST", "PUT"):
request_buffer = BytesIO(utf8(request.body))
curl.setopt(pycurl.READFUNCTION, request_buffer.read)
if request.method == "POST":
def ioctl(cmd):
if cmd == curl.IOCMD_RESTARTREAD:
request_buffer.seek(0)
curl.setopt(pycurl.IOCTLFUNCTION, ioctl)
curl.setopt(pycurl.POSTFIELDSIZE, len(request.body))
else:
curl.setopt(pycurl.INFILESIZE, len(request.body))
if request.auth_username is not None:
userpwd = "%s:%s" % (request.auth_username, request.auth_password or '')
curl.setopt(pycurl.HTTPAUTH, pycurl.HTTPAUTH_BASIC)
curl.setopt(pycurl.USERPWD, native_str(userpwd))
gen_log.debug("%s %s (username: %r)", request.method, request.url,
request.auth_username)
else:
curl.unsetopt(pycurl.USERPWD)
gen_log.debug("%s %s", request.method, request.url)
if request.client_cert is not None:
curl.setopt(pycurl.SSLCERT, request.client_cert)
if request.client_key is not None:
curl.setopt(pycurl.SSLKEY, request.client_key)
if threading.activeCount() > 1:
# libcurl/pycurl is not thread-safe by default. When multiple threads
# are used, signals should be disabled. This has the side effect
# of disabling DNS timeouts in some environments (when libcurl is
# not linked against ares), so we don't do it when there is only one
# thread. Applications that use many short-lived threads may need
# to set NOSIGNAL manually in a prepare_curl_callback since
# there may not be any other threads running at the time we call
# threading.activeCount.
curl.setopt(pycurl.NOSIGNAL, 1)
if request.prepare_curl_callback is not None:
request.prepare_curl_callback(curl)