wsgiref —- WSGI Utilities and Reference Implementation


The Web Server Gateway Interface (WSGI) is a standard interface between webserver software and web applications written in Python. Having a standardinterface makes it easy to use an application that supports WSGI with a numberof different web servers.

Only authors of web servers and programming frameworks need to know every detailand corner case of the WSGI design. You don't need to understand every detailof WSGI just to install a WSGI application or to write a web application usingan existing framework.

wsgiref is a reference implementation of the WSGI specification that canbe used to add WSGI support to a web server or framework. It provides utilitiesfor manipulating WSGI environment variables and response headers, base classesfor implementing WSGI servers, a demo HTTP server that serves WSGI applications,and a validation tool that checks WSGI servers and applications for conformanceto the WSGI specification (PEP 3333).

See wsgi.readthedocs.io for more information about WSGI, and linksto tutorials and other resources.

wsgiref.util — WSGI environment utilities

This module provides a variety of utility functions for working with WSGIenvironments. A WSGI environment is a dictionary containing HTTP requestvariables as described in PEP 3333. All of the functions taking an _environ_parameter expect a WSGI-compliant dictionary to be supplied; please seePEP 3333 for a detailed specification.

  • wsgiref.util.guessscheme(_environ)
  • Return a guess for whether wsgi.urlscheme should be "http" or "https", bychecking for a HTTPS environment variable in the _environ dictionary. Thereturn value is a string.

This function is useful when creating a gateway that wraps CGI or a CGI-likeprotocol such as FastCGI. Typically, servers providing such protocols willinclude a HTTPS variable with a value of "1" "yes", or "on" when a requestis received via SSL. So, this function returns "https" if such a value isfound, and "http" otherwise.

  • wsgiref.util.requesturi(_environ, include_query=True)
  • Return the full request URI, optionally including the query string, using thealgorithm found in the "URL Reconstruction" section of PEP 3333. Ifinclude_query is false, the query string is not included in the resulting URI.
  • wsgiref.util.applicationuri(_environ)
  • Similar to request_uri(), except that the PATH_INFO andQUERY_STRING variables are ignored. The result is the base URI of theapplication object addressed by the request.
  • wsgiref.util.shiftpath_info(_environ)
  • Shift a single name from PATHINFO to SCRIPT_NAME and return the name.The _environ dictionary is modified in-place; use a copy if you need to keepthe original PATH_INFO or SCRIPT_NAME intact.

If there are no remaining path segments in PATH_INFO, None is returned.

Typically, this routine is used to process each portion of a request URI path,for example to treat the path as a series of dictionary keys. This routinemodifies the passed-in environment to make it suitable for invoking another WSGIapplication that is located at the target URI. For example, if there is a WSGIapplication at /foo, and the request URI path is /foo/bar/baz, and theWSGI application at /foo calls shift_path_info(), it will receive thestring "bar", and the environment will be updated to be suitable for passing toa WSGI application at /foo/bar. That is, SCRIPT_NAME will change from/foo to /foo/bar, and PATH_INFO will change from /bar/baz to/baz.

When PATH_INFO is just a "/", this routine returns an empty string andappends a trailing slash to SCRIPT_NAME, even though empty path segments arenormally ignored, and SCRIPT_NAME doesn't normally end in a slash. This isintentional behavior, to ensure that an application can tell the differencebetween URIs ending in /x from ones ending in /x/ when using thisroutine to do object traversal.

  • wsgiref.util.setuptesting_defaults(_environ)
  • Update environ with trivial defaults for testing purposes.

This routine adds various parameters required for WSGI, including HTTP_HOST,SERVER_NAME, SERVER_PORT, REQUEST_METHOD, SCRIPT_NAME,PATH_INFO, and all of the PEP 3333-defined wsgi.* variables. Itonly supplies default values, and does not replace any existing settings forthese variables.

This routine is intended to make it easier for unit tests of WSGI servers andapplications to set up dummy environments. It should NOT be used by actual WSGIservers or applications, since the data is fake!

用法示例:

  1. from wsgiref.util import setup_testing_defaults
  2. from wsgiref.simple_server import make_server
  3.  
  4. # A relatively simple WSGI application. It's going to print out the
  5. # environment dictionary after being updated by setup_testing_defaults
  6. def simple_app(environ, start_response):
  7. setup_testing_defaults(environ)
  8.  
  9. status = '200 OK'
  10. headers = [('Content-type', 'text/plain; charset=utf-8')]
  11.  
  12. start_response(status, headers)
  13.  
  14. ret = [("%s: %s\n" % (key, value)).encode("utf-8")
  15. for key, value in environ.items()]
  16. return ret
  17.  
  18. with make_server('', 8000, simple_app) as httpd:
  19. print("Serving on port 8000...")
  20. httpd.serve_forever()

In addition to the environment functions above, the wsgiref.util modulealso provides these miscellaneous utilities:

  • wsgiref.util.ishop_by_hop(_header_name)
  • Return True if 'header_name' is an HTTP/1.1 "Hop-by-Hop" header, as defined byRFC 2616.
  • class wsgiref.util.FileWrapper(filelike, blksize=8192)
  • A wrapper to convert a file-like object to an iterator. The resulting objectssupport both getitem() and iter() iteration styles, forcompatibility with Python 2.1 and Jython. As the object is iterated over, theoptional blksize parameter will be repeatedly passed to the _filelike_object's read() method to obtain bytestrings to yield. When read()returns an empty bytestring, iteration is ended and is not resumable.

If filelike has a close() method, the returned object will also have aclose() method, and it will invoke the filelike object's close()method when called.

用法示例:

  1. from io import StringIO
  2. from wsgiref.util import FileWrapper
  3.  
  4. # We're using a StringIO-buffer for as the file-like object
  5. filelike = StringIO("This is an example file-like object"*10)
  6. wrapper = FileWrapper(filelike, blksize=5)
  7.  
  8. for chunk in wrapper:
  9. print(chunk)

wsgiref.headers — WSGI response header tools

This module provides a single class, Headers, for convenientmanipulation of WSGI response headers using a mapping-like interface.

  • class wsgiref.headers.Headers([headers])
  • Create a mapping-like object wrapping headers, which must be a list of headername/value tuples as described in PEP 3333. The default value of headers isan empty list.

Headers objects support typical mapping operations includinggetitem(), get(), setitem(), setdefault(),delitem() and contains(). For each ofthese methods, the key is the header name (treated case-insensitively), and thevalue is the first value associated with that header name. Setting a headerdeletes any existing values for that header, then adds a new value at the end ofthe wrapped header list. Headers' existing order is generally maintained, withnew headers added to the end of the wrapped list.

Unlike a dictionary, Headers objects do not raise an error when you tryto get or delete a key that isn't in the wrapped header list. Getting anonexistent header just returns None, and deleting a nonexistent header doesnothing.

Headers objects also support keys(), values(), anditems() methods. The lists returned by keys() and items() caninclude the same key more than once if there is a multi-valued header. Thelen() of a Headers object is the same as the length of itsitems(), which is the same as the length of the wrapped header list. Infact, the items() method just returns a copy of the wrapped header list.

Calling bytes() on a Headers object returns a formatted bytestringsuitable for transmission as HTTP response headers. Each header is placed on aline with its value, separated by a colon and a space. Each line is terminatedby a carriage return and line feed, and the bytestring is terminated with ablank line.

In addition to their mapping interface and formatting features, Headersobjects also have the following methods for querying and adding multi-valuedheaders, and for adding headers with MIME parameters:

  • getall(_name)
  • Return a list of all the values for the named header.

The returned list will be sorted in the order they appeared in the originalheader list or were added to this instance, and may contain duplicates. Anyfields deleted and re-inserted are always appended to the header list. If nofields exist with the given name, returns an empty list.

  • addheader(_name, value, **_params)
  • Add a (possibly multi-valued) header, with optional MIME parameters specifiedvia keyword arguments.

name is the header field to add. Keyword arguments can be used to set MIMEparameters for the header field. Each parameter must be a string or None.Underscores in parameter names are converted to dashes, since dashes are illegalin Python identifiers, but many MIME parameter names include dashes. If theparameter value is a string, it is added to the header value parameters in theform name="value". If it is None, only the parameter name is added.(This is used for MIME parameters without a value.) Example usage:

  1. h.add_header('content-disposition', 'attachment', filename='bud.gif')

The above will add a header that looks like this:

  1. Content-Disposition: attachment; filename="bud.gif"

在 3.5 版更改: headers parameter is optional.

wsgiref.simple_server — a simple WSGI HTTP server

This module implements a simple HTTP server (based on http.server)that serves WSGI applications. Each server instance serves a single WSGIapplication on a given host and port. If you want to serve multipleapplications on a single host and port, you should create a WSGI applicationthat parses PATH_INFO to select which application to invoke for eachrequest. (E.g., using the shift_path_info() function fromwsgiref.util.)

  • wsgiref.simpleserver.make_server(_host, port, app, server_class=WSGIServer, handler_class=WSGIRequestHandler)
  • Create a new WSGI server listening on host and port, accepting connectionsfor app. The return value is an instance of the supplied server_class, andwill process requests using the specified handler_class. app must be a WSGIapplication object, as defined by PEP 3333.

用法示例:

  1. from wsgiref.simple_server import make_server, demo_app
  2.  
  3. with make_server('', 8000, demo_app) as httpd:
  4. print("Serving HTTP on port 8000...")
  5.  
  6. # Respond to requests until process is killed
  7. httpd.serve_forever()
  8.  
  9. # Alternative: serve one request, then exit
  10. httpd.handle_request()
  • wsgiref.simpleserver.demo_app(_environ, start_response)
  • This function is a small but complete WSGI application that returns a text pagecontaining the message "Hello world!" and a list of the key/value pairs providedin the environ parameter. It's useful for verifying that a WSGI server (suchas wsgiref.simple_server) is able to run a simple WSGI applicationcorrectly.
  • class wsgiref.simpleserver.WSGIServer(_server_address, RequestHandlerClass)
  • Create a WSGIServer instance. server_address should be a(host,port) tuple, and RequestHandlerClass should be the subclass ofhttp.server.BaseHTTPRequestHandler that will be used to processrequests.

You do not normally need to call this constructor, as the make_server()function can handle all the details for you.

WSGIServer is a subclass of http.server.HTTPServer, so allof its methods (such as serve_forever() and handle_request()) areavailable. WSGIServer also provides these WSGI-specific methods:

  • setapp(_application)
  • Sets the callable application as the WSGI application that will receiverequests.

  • get_app()

  • Returns the currently-set application callable.

Normally, however, you do not need to use these additional methods, asset_app() is normally called by make_server(), and theget_app() exists mainly for the benefit of request handler instances.

  • class wsgiref.simpleserver.WSGIRequestHandler(_request, client_address, server)
  • Create an HTTP handler for the given request (i.e. a socket), client_address(a (host,port) tuple), and server (WSGIServer instance).

You do not need to create instances of this class directly; they areautomatically created as needed by WSGIServer objects. You can,however, subclass this class and supply it as a handler_class to themake_server() function. Some possibly relevant methods for overriding insubclasses:

  • get_environ()
  • Returns a dictionary containing the WSGI environment for a request. The defaultimplementation copies the contents of the WSGIServer object'sbase_environ dictionary attribute and then adds various headers derivedfrom the HTTP request. Each call to this method should return a new dictionarycontaining all of the relevant CGI environment variables as specified inPEP 3333.

  • get_stderr()

  • Return the object that should be used as the wsgi.errors stream. The defaultimplementation just returns sys.stderr.

  • handle()

  • Process the HTTP request. The default implementation creates a handler instanceusing a wsgiref.handlers class to implement the actual WSGI applicationinterface.

wsgiref.validate —- WSGI conformance checker

When creating new WSGI application objects, frameworks, servers, or middleware,it can be useful to validate the new code's conformance usingwsgiref.validate. This module provides a function that creates WSGIapplication objects that validate communications between a WSGI server orgateway and a WSGI application object, to check both sides for protocolconformance.

Note that this utility does not guarantee complete PEP 3333 compliance; anabsence of errors from this module does not necessarily mean that errors do notexist. However, if this module does produce an error, then it is virtuallycertain that either the server or application is not 100% compliant.

This module is based on the paste.lint module from Ian Bicking's "PythonPaste" library.

  • wsgiref.validate.validator(application)
  • Wrap application and return a new WSGI application object. The returnedapplication will forward all requests to the original application, and willcheck that both the application and the server invoking it are conforming tothe WSGI specification and to RFC 2616.

Any detected nonconformance results in an AssertionError being raised;note, however, that how these errors are handled is server-dependent. Forexample, wsgiref.simple_server and other servers based onwsgiref.handlers (that don't override the error handling methods to dosomething else) will simply output a message that an error has occurred, anddump the traceback to sys.stderr or some other error stream.

This wrapper may also generate output using the warnings module toindicate behaviors that are questionable but which may not actually beprohibited by PEP 3333. Unless they are suppressed using Python command-lineoptions or the warnings API, any such warnings will be written tosys.stderr (not wsgi.errors, unless they happen to be the sameobject).

用法示例:

  1. from wsgiref.validate import validator
  2. from wsgiref.simple_server import make_server
  3.  
  4. # Our callable object which is intentionally not compliant to the
  5. # standard, so the validator is going to break
  6. def simple_app(environ, start_response):
  7. status = '200 OK' # HTTP Status
  8. headers = [('Content-type', 'text/plain')] # HTTP Headers
  9. start_response(status, headers)
  10.  
  11. # This is going to break because we need to return a list, and
  12. # the validator is going to inform us
  13. return b"Hello World"
  14.  
  15. # This is the application wrapped in a validator
  16. validator_app = validator(simple_app)
  17.  
  18. with make_server('', 8000, validator_app) as httpd:
  19. print("Listening on port 8000....")
  20. httpd.serve_forever()

wsgiref.handlers — server/gateway base classes

This module provides base handler classes for implementing WSGI servers andgateways. These base classes handle most of the work of communicating with aWSGI application, as long as they are given a CGI-like environment, along withinput, output, and error streams.

  • class wsgiref.handlers.CGIHandler
  • CGI-based invocation via sys.stdin, sys.stdout, sys.stderr andos.environ. This is useful when you have a WSGI application and want to runit as a CGI script. Simply invoke CGIHandler().run(app), where app isthe WSGI application object you wish to invoke.

This class is a subclass of BaseCGIHandler that sets wsgi.run_onceto true, wsgi.multithread to false, and wsgi.multiprocess to true, andalways uses sys and os to obtain the necessary CGI streams andenvironment.

  • class wsgiref.handlers.IISCGIHandler
  • A specialized alternative to CGIHandler, for use when deploying onMicrosoft's IIS web server, without having set the config allowPathInfooption (IIS>=7) or metabase allowPathInfoForScriptMappings (IIS<7).

By default, IIS gives a PATH_INFO that duplicates the SCRIPT_NAME atthe front, causing problems for WSGI applications that wish to implementrouting. This handler strips any such duplicated path.

IIS can be configured to pass the correct PATH_INFO, but this causesanother bug where PATH_TRANSLATED is wrong. Luckily this variable israrely used and is not guaranteed by WSGI. On IIS<7, though, thesetting can only be made on a vhost level, affecting all other scriptmappings, many of which break when exposed to the PATH_TRANSLATED bug.For this reason IIS<7 is almost never deployed with the fix. (Even IIS7rarely uses it because there is still no UI for it.)

There is no way for CGI code to tell whether the option was set, so aseparate handler class is provided. It is used in the same way asCGIHandler, i.e., by calling IISCGIHandler().run(app), whereapp is the WSGI application object you wish to invoke.

3.2 新版功能.

  • class wsgiref.handlers.BaseCGIHandler(stdin, stdout, stderr, environ, multithread=True, multiprocess=False)
  • Similar to CGIHandler, but instead of using the sys andos modules, the CGI environment and I/O streams are specified explicitly.The multithread and multiprocess values are used to set thewsgi.multithread and wsgi.multiprocess flags for any applications run bythe handler instance.

This class is a subclass of SimpleHandler intended for use withsoftware other than HTTP "origin servers". If you are writing a gatewayprotocol implementation (such as CGI, FastCGI, SCGI, etc.) that uses aStatus: header to send an HTTP status, you probably want to subclass thisinstead of SimpleHandler.

  • class wsgiref.handlers.SimpleHandler(stdin, stdout, stderr, environ, multithread=True, multiprocess=False)
  • Similar to BaseCGIHandler, but designed for use with HTTP originservers. If you are writing an HTTP server implementation, you will probablywant to subclass this instead of BaseCGIHandler.

This class is a subclass of BaseHandler. It overrides theinit(), get_stdin(), get_stderr(), add_cgi_vars(),_write(), and _flush() methods to support explicitly setting theenvironment and streams via the constructor. The supplied environment andstreams are stored in the stdin, stdout, stderr, andenviron attributes.

The write() method of stdout should writeeach chunk in full, like io.BufferedIOBase.

  • class wsgiref.handlers.BaseHandler
  • This is an abstract base class for running WSGI applications. Each instancewill handle a single HTTP request, although in principle you could create asubclass that was reusable for multiple requests.

BaseHandler instances have only one method intended for external use:

  • run(app)
  • Run the specified WSGI application, app.

All of the other BaseHandler methods are invoked by this method in theprocess of running the application, and thus exist primarily to allowcustomizing the process.

The following methods MUST be overridden in a subclass:

  • write(_data)
  • Buffer the bytes data for transmission to the client. It's okay if thismethod actually transmits the data; BaseHandler just separates writeand flush operations for greater efficiency when the underlying system actuallyhas such a distinction.

  • _flush()

  • Force buffered data to be transmitted to the client. It's okay if this methodis a no-op (i.e., if _write() actually sends the data).

  • get_stdin()

  • Return an input stream object suitable for use as the wsgi.input of therequest currently being processed.

  • get_stderr()

  • Return an output stream object suitable for use as the wsgi.errors of therequest currently being processed.

  • add_cgi_vars()

  • Insert CGI variables for the current request into the environ attribute.

Here are some other methods and attributes you may wish to override. This listis only a summary, however, and does not include every method that can beoverridden. You should consult the docstrings and source code for additionalinformation before attempting to create a customized BaseHandlersubclass.

Attributes and methods for customizing the WSGI environment:

  • wsgi_multithread
  • The value to be used for the wsgi.multithread environment variable. Itdefaults to true in BaseHandler, but may have a different default (orbe set by the constructor) in the other subclasses.

  • wsgi_multiprocess

  • The value to be used for the wsgi.multiprocess environment variable. Itdefaults to true in BaseHandler, but may have a different default (orbe set by the constructor) in the other subclasses.

  • wsgi_run_once

  • The value to be used for the wsgi.run_once environment variable. Itdefaults to false in BaseHandler, but CGIHandler sets it totrue by default.

  • os_environ

  • The default environment variables to be included in every request's WSGIenvironment. By default, this is a copy of os.environ at the time thatwsgiref.handlers was imported, but subclasses can either create their ownat the class or instance level. Note that the dictionary should be consideredread-only, since the default value is shared between multiple classes andinstances.

  • server_software

  • If the origin_server attribute is set, this attribute's value is used toset the default SERVER_SOFTWARE WSGI environment variable, and also to set adefault Server: header in HTTP responses. It is ignored for handlers (suchas BaseCGIHandler and CGIHandler) that are not HTTP originservers.

在 3.3 版更改: The term "Python" is replaced with implementation specific term like"CPython", "Jython" etc.

  • get_scheme()
  • Return the URL scheme being used for the current request. The defaultimplementation uses the guess_scheme() function from wsgiref.utilto guess whether the scheme should be "http" or "https", based on the currentrequest's environ variables.

  • setup_environ()

  • Set the environ attribute to a fully-populated WSGI environment. Thedefault implementation uses all of the above methods and attributes, plus theget_stdin(), get_stderr(), and add_cgi_vars() methods and thewsgi_file_wrapper attribute. It also inserts a SERVER_SOFTWARE keyif not present, as long as the origin_server attribute is a true valueand the server_software attribute is set.

Methods and attributes for customizing exception handling:

  • logexception(_exc_info)
  • Log the exc_info tuple in the server log. exc_info is a (type, value,traceback) tuple. The default implementation simply writes the traceback tothe request's wsgi.errors stream and flushes it. Subclasses can overridethis method to change the format or retarget the output, mail the traceback toan administrator, or whatever other action may be deemed suitable.

  • traceback_limit

  • The maximum number of frames to include in tracebacks output by the defaultlog_exception() method. If None, all frames are included.

  • erroroutput(_environ, start_response)

  • This method is a WSGI application to generate an error page for the user. It isonly invoked if an error occurs before headers are sent to the client.

This method can access the current error information using sys.excinfo(),and should pass that information to _start_response when calling it (asdescribed in the "Error Handling" section of PEP 3333).

The default implementation just uses the error_status,error_headers, and error_body attributes to generate an outputpage. Subclasses can override this to produce more dynamic error output.

Note, however, that it's not recommended from a security perspective to spit outdiagnostics to any old user; ideally, you should have to do something special toenable diagnostic output, which is why the default implementation doesn'tinclude any.

  • error_status
  • The HTTP status used for error responses. This should be a status string asdefined in PEP 3333; it defaults to a 500 code and message.

  • error_headers

  • The HTTP headers used for error responses. This should be a list of WSGIresponse headers ((name, value) tuples), as described in PEP 3333. Thedefault list just sets the content type to text/plain.

  • error_body

  • The error response body. This should be an HTTP response body bytestring. Itdefaults to the plain text, "A server error occurred. Please contact theadministrator."

Methods and attributes for PEP 3333's "Optional Platform-Specific FileHandling" feature:

  • wsgi_file_wrapper
  • A wsgi.file_wrapper factory, or None. The default value of thisattribute is the wsgiref.util.FileWrapper class.

  • sendfile()

  • Override to implement platform-specific file transmission. This method iscalled only if the application's return value is an instance of the classspecified by the wsgi_file_wrapper attribute. It should return a truevalue if it was able to successfully transmit the file, so that the defaulttransmission code will not be executed. The default implementation of thismethod just returns a false value.

Miscellaneous methods and attributes:

  • origin_server
  • This attribute should be set to a true value if the handler's _write() and_flush() are being used to communicate directly to the client, rather thanvia a CGI-like gateway protocol that wants the HTTP status in a specialStatus: header.

This attribute's default value is true in BaseHandler, but false inBaseCGIHandler and CGIHandler.

  • http_version
  • If origin_server is true, this string attribute is used to set the HTTPversion of the response set to the client. It defaults to "1.0".
  • wsgiref.handlers.read_environ()
  • Transcode CGI variables from os.environ to PEP 3333 "bytes in unicode"strings, returning a new dictionary. This function is used byCGIHandler and IISCGIHandler in place of directly usingos.environ, which is not necessarily WSGI-compliant on all platformsand web servers using Python 3 — specifically, ones where the OS'sactual environment is Unicode (i.e. Windows), or ones where the environmentis bytes, but the system encoding used by Python to decode it is anythingother than ISO-8859-1 (e.g. Unix systems using UTF-8).

If you are implementing a CGI-based handler of your own, you probably wantto use this routine instead of just copying values out of os.environdirectly.

3.2 新版功能.

例子

This is a working "Hello World" WSGI application:

  1. from wsgiref.simple_server import make_server
  2.  
  3. # Every WSGI application must have an application object - a callable
  4. # object that accepts two arguments. For that purpose, we're going to
  5. # use a function (note that you're not limited to a function, you can
  6. # use a class for example). The first argument passed to the function
  7. # is a dictionary containing CGI-style environment variables and the
  8. # second variable is the callable object (see PEP 333).
  9. def hello_world_app(environ, start_response):
  10. status = '200 OK' # HTTP Status
  11. headers = [('Content-type', 'text/plain; charset=utf-8')] # HTTP Headers
  12. start_response(status, headers)
  13.  
  14. # The returned object is going to be printed
  15. return [b"Hello World"]
  16.  
  17. with make_server('', 8000, hello_world_app) as httpd:
  18. print("Serving on port 8000...")
  19.  
  20. # Serve until process is killed
  21. httpd.serve_forever()