| blob | 7047e6b5f13ee33eb57ef7b08e5b6e774e3cc1b5 |
1 ;;; url-http.el --- HTTP retrieval routines
3 ;; Copyright (C) 1999, 2001, 2004-2013 Free Software Foundation, Inc.
5 ;; Author: Bill Perry <wmperry@gnu.org>
6 ;; Keywords: comm, data, processes
8 ;; This file is part of GNU Emacs.
9 ;;
10 ;; GNU Emacs is free software: you can redistribute it and/or modify
11 ;; it under the terms of the GNU General Public License as published by
12 ;; the Free Software Foundation, either version 3 of the License, or
13 ;; (at your option) any later version.
15 ;; GNU Emacs is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 ;; GNU General Public License for more details.
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
23 ;;; Commentary:
25 ;;; Code:
74 "What version of HTTP we advertise, as a string.
75 Valid values are 1.1 and 1.0.
76 This is only useful when debugging the HTTP subsystem.
78 Setting this to 1.0 will tell servers not to send chunked encoding,
82 "Whether to use a single TCP connection multiple times in HTTP.
83 This is only useful when debugging the HTTP subsystem. Setting to
84 nil will explicitly close the connection to the server after every
136 ;(eval-when-compile
137 ;; These are all macros so that they are hidden from external sight
138 ;; when the file is byte-compiled.
139 ;;
140 ;; This allows us to expose just the entry points we want.
142 ;; These routines will allow us to implement persistent HTTP
143 ;; connections.
147 ;; The user hit C-g, honor it! Some things can get in an
148 ;; incredibly tight loop (chunked encoding)
161 url-http-open-connections)
162 proc)
172 url-http-open-connections))
173 nil)
194 ;; `url-open-stream' needs a buffer in which to do things
195 ;; like authentication. But we use another buffer afterwards.
198 ;; url-open-stream might return nil.
200 ;; Drop the temp buffer link before killing the buffer.
203 ;; If there was an error on connect, make sure we don't
204 ;; get queried.
212 ;; Building an HTTP request
217 ""
222 url-version)))
225 "Create an HTTP request for `url-http-target-url', referred to by REF-URL."
231 url-http-extra-headers))
233 nil
240 nil
243 proxy-info)
253 ;; Protection against stupid values in the referrer
258 ;; We do not want to expose the referrer if the user is paranoid.
264 ;; url-http-extra-headers contains an assoc-list of
265 ;; header/value pairs that we need to put into the request.
273 ;; This was done with a call to `format'. Concatenating parts has
274 ;; the advantage of keeping the parts of each header together and
275 ;; allows us to elide null lines directly, at the cost of making
276 ;; the layout less clear.
278 ;; We used to concat directly, but if one of the strings happens
279 ;; to being multibyte (even if it only contains pure ASCII) then
280 ;; every string gets converted with `string-MAKE-multibyte' which
281 ;; turns the 127-255 codes into things like latin-1 accented chars
282 ;; (it would work right if it used `string-TO-multibyte' instead).
283 ;; So to avoid the problem we force every string to be unibyte.
285 ;; FIXME: Instead of `string-AS-unibyte' we'd want
286 ;; `string-to-unibyte', so as to properly signal an error if one
287 ;; of the strings contains a multibyte char.
288 'string-as-unibyte
291 ;; The request
295 ;; Version of MIME we speak
297 ;; (maybe) Try to keep the connection open
301 ;; HTTP extensions we support
305 ;; Who we want to talk to
312 ;; Who its from
316 ;; Encodings we understand
323 ;; Languages we understand
327 ;; Types we understand
329 ;; User agent
331 ;; Proxy Authorization
332 proxy-auth
333 ;; Authorization
334 auth
335 ;; Cookies
338 host real-fname
340 ;; If-modified-since
347 ;; Whence we came
350 extra-headers
351 ;; Length of data
357 ;; End request
359 ;; Any data
360 url-http-data
361 ;; If `url-http-data' is nil, avoid two CRLFs (Bug#8931).
365 request))
367 ;; Parsing routines
370 This allows us to use `mail-fetch-field', etc.
371 Return the number of characters removed."
383 nil nil t))
389 url-http-proxy
390 url-current-object)))
392 ;; Cheating, but who cares? :)
393 'url-http-proxy-basic-auth-storage
395 auth
398 ;; find strongest supported auth
402 this-auth)))
406 this-auth)))
411 type this-type
412 strength this-strength))))
419 " authentication. If you'd like to write it,"
421 ;; We used to set a `status' var (declared "special") but I can't
422 ;; find the corresponding let-binding, so it's probably an error.
423 ;; FIXME: Maybe it was supposed to set `success', i.e. to return t?
424 ;; (setq status t)
430 type t args)))
432 t ;Success.
434 url-http-extra-headers)
439 url-callback-arguments))
443 "Parse just the response code."
458 "Handle all set-cookie / set-cookie2 headers in an HTTP response.
459 The buffer must already be narrowed to the headers, so `mail-fetch-field' will
460 work correctly."
467 ;;; (while cookies2
468 ;;; (url-cookie-handle-set-cookie2 (pop cookies)))
469 )
470 )
473 "Parse and handle HTTP specific headers.
474 Return t if and only if the current buffer is still active and
475 should be shown to the user."
476 ;; The comments after each status code handled are taken from RFC
477 ;; 2616 (HTTP/1.1)
480 url-http-process)
489 ;;(narrow-to-region (point-min) url-http-end-of-headers)
491 ;; In HTTP 1.0, keep the connection only if there is a
492 ;; "Connection: keep-alive" header.
493 ;; In HTTP 1.1 (and greater), keep the connection unless there is a
494 ;; "Connection: close" header
507 ;; other status symbols: jewelry and luxury cars
510 class url-http-response-status)
515 ;; Classes of response codes
516 ;;
517 ;; 5xx = Server Error
518 ;; 4xx = Client Error
519 ;; 3xx = Redirection
520 ;; 2xx = Successful
521 ;; 1xx = Informational
523 ;; 100 = Continue with request
524 ;; 101 = Switching protocols
525 ;; 102 = Processing (Added by DAV)
528 url-http-response-status))
530 ;; 200 Ok
531 ;; 201 Created
532 ;; 202 Accepted
533 ;; 203 Non-authoritative information
534 ;; 204 No content
535 ;; 205 Reset content
536 ;; 206 Partial content
537 ;; 207 Multi-status (Added by DAV)
540 ;; No new data, just stay at the same document
543 ;; Generic success for all others. Store in the cache, and
544 ;; mark it as successful.
550 ;; 300 Multiple choices
551 ;; 301 Moved permanently
552 ;; 302 Found
553 ;; 303 See other
554 ;; 304 Not modified
555 ;; 305 Use proxy
556 ;; 307 Temporary redirect
561 ;; Quoth the spec (section 10.3.1)
562 ;; -------------------------------
563 ;; The requested resource corresponds to any one of a set of
564 ;; representations, each with its own specific location and
565 ;; agent-driven negotiation information is being provided so
566 ;; that the user can select a preferred representation and
567 ;; redirect its request to that location.
568 ;; [...]
569 ;; If the server has a preferred choice of representation, it
570 ;; SHOULD include the specific URI for that representation in
571 ;; the Location field; user agents MAY use the Location field
572 ;; value for automatic redirection.
573 ;; -------------------------------
574 ;; We do not support agent-driven negotiation, so we just
575 ;; redirect to the preferred URI if one is provided.
576 nil)
578 ;; If the 301|302 status code is received in response to a
579 ;; request other than GET or HEAD, the user agent MUST NOT
580 ;; automatically redirect the request unless it can be
581 ;; confirmed by the user, since this might change the
582 ;; conditions under which the request was issued.
586 ;; The response to the request can be found under a different
587 ;; URI and SHOULD be retrieved using a GET method on that
588 ;; resource.
590 url-http-data nil))
592 ;; The 304 response MUST NOT contain a message-body.
597 success t))
599 ;; The requested resource MUST be accessed through the
600 ;; proxy given by the Location field. The Location field
601 ;; gives the URI of the proxy. The recipient is expected
602 ;; to repeat this single request via the proxy. 305
603 ;; responses MUST only be generated by origin servers.
605 redirect-uri))
607 ;; Treat everything like '300'
608 nil))
610 ;; Clean off any whitespace and/or <...> cruft.
616 ;; Some stupid sites (like sourceforge) send a
617 ;; non-fully-qualified URL (ie: /), which royally confuses
618 ;; the URL library.
620 ;; Be careful to use the real target URL, otherwise we may
621 ;; compute the redirection relative to the URL of the proxy.
627 ;; Check existing number of redirects
638 ;; url-max-redirections hasn't been reached, so go
639 ;; ahead and redirect.
641 ;; Remember that the request was redirected.
645 ;; Put in the current buffer a forwarding pointer to the new
646 ;; destination buffer.
647 ;; FIXME: This is a hack to fix url-retrieve-synchronously
648 ;; without changing the API. Instead url-retrieve should
649 ;; either simply not return the "destination" buffer, or it
650 ;; should take an optional `dest-buf' argument.
653 redirect-uri url-callback-function
654 url-callback-arguments
658 ;; We hit url-max-redirections, so issue an error and
659 ;; stop redirecting.
663 redirect-uri))
667 ;; 400 Bad Request
668 ;; 401 Unauthorized
669 ;; 402 Payment required
670 ;; 403 Forbidden
671 ;; 404 Not found
672 ;; 405 Method not allowed
673 ;; 406 Not acceptable
674 ;; 407 Proxy authentication required
675 ;; 408 Request time-out
676 ;; 409 Conflict
677 ;; 410 Gone
678 ;; 411 Length required
679 ;; 412 Precondition failed
680 ;; 413 Request entity too large
681 ;; 414 Request-URI too large
682 ;; 415 Unsupported media type
683 ;; 416 Requested range not satisfiable
684 ;; 417 Expectation failed
685 ;; 422 Unprocessable Entity (Added by DAV)
686 ;; 423 Locked
687 ;; 424 Failed Dependency
691 ;; The request requires user authentication. The response
692 ;; MUST include a WWW-Authenticate header field containing a
693 ;; challenge applicable to the requested resource. The
694 ;; client MAY repeat the request with a suitable
695 ;; Authorization header field.
698 ;; This code is reserved for future use
702 ;; The server understood the request, but is refusing to
703 ;; fulfill it. Authorization will not help and the request
704 ;; SHOULD NOT be repeated.
705 t)
707 ;; Not found
708 t)
710 ;; The method specified in the Request-Line is not allowed
711 ;; for the resource identified by the Request-URI. The
712 ;; response MUST include an Allow header containing a list of
713 ;; valid methods for the requested resource.
714 t)
716 ;; The resource identified by the request is only capable of
717 ;; generating response entities which have content
718 ;; characteristics not acceptable according to the accept
719 ;; headers sent in the request.
720 t)
722 ;; This code is similar to 401 (Unauthorized), but indicates
723 ;; that the client must first authenticate itself with the
724 ;; proxy. The proxy MUST return a Proxy-Authenticate header
725 ;; field containing a challenge applicable to the proxy for
726 ;; the requested resource.
729 ;; The client did not produce a request within the time that
730 ;; the server was prepared to wait. The client MAY repeat
731 ;; the request without modifications at any later time.
732 t)
734 ;; The request could not be completed due to a conflict with
735 ;; the current state of the resource. This code is only
736 ;; allowed in situations where it is expected that the user
737 ;; might be able to resolve the conflict and resubmit the
738 ;; request. The response body SHOULD include enough
739 ;; information for the user to recognize the source of the
740 ;; conflict.
741 t)
743 ;; The requested resource is no longer available at the
744 ;; server and no forwarding address is known.
745 t)
747 ;; The server refuses to accept the request without a defined
748 ;; Content-Length. The client MAY repeat the request if it
749 ;; adds a valid Content-Length header field containing the
750 ;; length of the message-body in the request message.
751 ;;
752 ;; NOTE - this will never happen because
753 ;; `url-http-create-request' automatically calculates the
754 ;; content-length.
755 t)
757 ;; The precondition given in one or more of the
758 ;; request-header fields evaluated to false when it was
759 ;; tested on the server.
760 t)
762 ;; The server is refusing to process a request because the
763 ;; request entity|URI is larger than the server is willing or
764 ;; able to process.
765 t)
767 ;; The server is refusing to service the request because the
768 ;; entity of the request is in a format not supported by the
769 ;; requested resource for the requested method.
770 t)
772 ;; A server SHOULD return a response with this status code if
773 ;; a request included a Range request-header field, and none
774 ;; of the range-specifier values in this field overlap the
775 ;; current extent of the selected resource, and the request
776 ;; did not include an If-Range request-header field.
777 t)
779 ;; The expectation given in an Expect request-header field
780 ;; could not be met by this server, or, if the server is a
781 ;; proxy, the server has unambiguous evidence that the
782 ;; request could not be met by the next-hop server.
783 t)
785 ;; The request could not be understood by the server due to
786 ;; malformed syntax. The client SHOULD NOT repeat the
787 ;; request without modifications.
788 t)))
789 ;; Tell the callback that an error occurred, and what the
790 ;; status code was.
796 ;; 500 Internal server error
797 ;; 501 Not implemented
798 ;; 502 Bad gateway
799 ;; 503 Service unavailable
800 ;; 504 Gateway time-out
801 ;; 505 HTTP version not supported
802 ;; 507 Insufficient storage
806 ;; The server does not support the functionality required to
807 ;; fulfill the request.
808 nil)
810 ;; The server, while acting as a gateway or proxy, received
811 ;; an invalid response from the upstream server it accessed
812 ;; in attempting to fulfill the request.
813 nil)
815 ;; The server is currently unable to handle the request due
816 ;; to a temporary overloading or maintenance of the server.
817 ;; The implication is that this is a temporary condition
818 ;; which will be alleviated after some delay. If known, the
819 ;; length of the delay MAY be indicated in a Retry-After
820 ;; header. If no Retry-After is given, the client SHOULD
821 ;; handle the response as it would for a 500 response.
822 nil)
824 ;; The server, while acting as a gateway or proxy, did not
825 ;; receive a timely response from the upstream server
826 ;; specified by the URI (e.g. HTTP, FTP, LDAP) or some other
827 ;; auxiliary server (e.g. DNS) it needed to access in
828 ;; attempting to complete the request.
829 nil)
831 ;; The server does not support, or refuses to support, the
832 ;; HTTP protocol version that was used in the request
833 ;; message.
834 nil)
836 ;; The method could not be performed on the resource
837 ;; because the server is unable to store the representation
838 ;; needed to successfully complete the request. This
839 ;; condition is considered to be temporary. If the request
840 ;; which received this status code was the result of a user
841 ;; action, the request MUST NOT be repeated until it is
842 ;; requested by a separate user action.
843 nil))
844 ;; Tell the callback that an error occurred, and what the
845 ;; status code was.
852 class url-http-response-status)))
859 success))
873 ;; Miscellaneous
875 "Activate callback specified when this buffer was created."
878 url-http-process)
882 ;; )
884 ;; These unfortunately cannot be macros... please ignore them!
886 "Remove (now defunct) process PROC from the list of open connections."
890 url-http-open-connections))
893 ;; Sentinel used to handle (i) terminated old HTTP/0.9 connections,
894 ;; and (ii) closed connection due to reusing a HTTP connection which
895 ;; we believed was still alive, but which the server closed on us.
896 ;; We handle case (ii) by calling `url-http' again.
905 ;; HTTP/0.9 just gets passed back no matter what
907 ;; Call `url-http' again if our connection expired.
918 ;; Function used when we do NOT know how long the document is going to be
919 ;; Just _very_ simple 'downloaded %d' type of info.
923 "Function used when we DO know how long the document is going to be.
924 More sophisticated percentage downloaded, etc.
925 Also does minimal parsing of HTTP headers and will actually cause
926 the callback to be triggered."
929 "Reading [%s]... %s of %s (%d%%)"
931 url-http-content-length)
932 url-http-content-type
936 url-http-content-length))
938 "Reading... %s of %s (%d%%)"
940 url-http-content-length)
944 url-http-content-length)))
948 ;; Found the end of the document! Wheee!
955 "Function used when dealing with 'chunked' encoding.
956 Cannot give a sophisticated percentage, but we need a different
957 function to look for the special 0-length chunk that signifies
958 the end of the document."
965 ;; We need to loop thru looking for more chunks even within
966 ;; one after-change-function call.
971 "Reading [%s]... chunk #%d"
972 url-http-content-type url-http-chunked-counter)
974 "Reading... chunk #%d"
975 url-http-chunked-counter))
977 url-http-chunked-counter st nd length)
983 ;; We know how long the chunk is supposed to be, skip over
984 ;; leading crap if possible.
988 url-http-chunked-counter)
990 url-http-chunked-length)))
993 url-http-chunked-length)
994 nd))
1001 ;; Must not have received the entirety of the chunk header,
1002 ;; need to spin some more.
1007 'end-open t
1008 'chunked-encoding t
1020 ; (if (not url-http-debug)
1023 url-http-chunked-counter url-http-chunked-length
1027 ;; Found the end of the document! Wheee!
1031 ;; Every chunk, even the last 0-length one, is
1032 ;; terminated by CRLF. Skip it.
1042 ;; This will wait for the headers to arrive and then splice in the
1043 ;; next appropriate after-change-function, etc.
1054 ;; Not HTTP/x.y data, must be 0.9
1055 ;; God, I wish this could die.
1057 url-http-end-of-headers 0
1058 old-http t)
1059 ;; Blank line at end of headers.
1062 ;; Saw the end of the headers
1066 end-of-headers t)
1070 ;; Haven't seen the end of the headers yet, need to wait
1071 ;; for more data to arrive.
1072 nil
1089 ;; We got back a headerless malformed response from the
1090 ;; server.
1099 ;; A HEAD request is _ALWAYS_ terminated by the header
1100 ;; information, regardless of any entity headers,
1101 ;; according to section 4.4 of the HTTP/1.1 draft.
1107 ;; A CONNECT request is finished, but we cannot stick this
1108 ;; back on the free connection list
1113 ;; Only allowed to have a header section. We have to handle
1114 ;; this here instead of in url-http-parse-headers because if
1115 ;; you have a cached copy of something without a known
1116 ;; content-length, and try to retrieve it from the cache, we'd
1117 ;; fall into the 'being dumb' section and wait for the
1118 ;; connection to terminate, which means we'd wait for 10
1119 ;; seconds for the keep-alives to time out on some servers.
1123 ;; HTTP/0.9 always signaled end-of-connection by closing the
1124 ;; connection.
1146 ;; We got a NULL body! Activate the callback
1147 ;; immediately!
1153 ;; Have some leftover data
1157 nd
1160 nil)))
1165 ;; We are still at the beginning of the buffer... must just be
1166 ;; waiting for a response.
1172 "Retrieve URL via HTTP asynchronously.
1173 URL must be a parsed URL. See `url-generic-parse-url' for details.
1175 When retrieval is completed, execute the function CALLBACK, using
1176 the arguments listed in CBARGS. The first element in CBARGS
1177 should be a plist describing what has happened so far during the
1178 request, as described in the docstring of `url-retrieve' (if in
1179 doubt, specify nil).
1181 Optional arg RETRY-BUFFER, if non-nil, specifies the buffer of a
1182 previous `url-http' call, which is being re-attempted."
1191 ;; Failed to open the connection for some reason
1202 url-http-content-type
1203 url-http-content-length
1204 url-http-transfer-encoding
1205 url-http-after-change-function
1206 url-http-response-version
1207 url-http-response-status
1208 url-http-chunked-length
1209 url-http-chunked-counter
1210 url-http-chunked-start
1211 url-callback-function
1212 url-callback-arguments
1213 url-show-status
1214 url-http-process
1215 url-http-method
1216 url-http-extra-headers
1217 url-http-data
1218 url-http-target-url
1219 url-http-no-retry
1220 url-http-connection-opened
1221 url-http-proxy))
1225 url-http-extra-headers url-request-extra-headers
1226 url-http-data url-request-data
1227 url-http-process connection
1228 url-http-chunked-length nil
1229 url-http-chunked-start nil
1230 url-http-chunked-counter 0
1231 url-callback-function callback
1232 url-callback-arguments cbargs
1233 url-http-after-change-function 'url-http-wait-for-headers-change-function
1234 url-http-target-url url-current-object
1235 url-http-no-retry retry-buffer
1236 url-http-connection-opened nil
1237 url-http-proxy url-using-proxy)
1242 (`connect
1243 ;; Asynchronous connection
1245 (`failed
1246 ;; Asynchronous connection failed
1252 buffer))
1255 ;; We are performing an asynchronous connection, and a status change
1256 ;; has occurred.
1278 ;; Since Emacs 19/20 does not allow you to change the
1279 ;; `after-change-functions' hook in the midst of running them, we fake
1280 ;; an after change by hooking into the process filter and inserting
1281 ;; the data ourselves. This is slightly less efficient, but there
1282 ;; were tons of weird ways the after-change code was biting us in the
1283 ;; shorts.
1284 ;; FIXME this can probably be simplified since the above is no longer true.
1286 ;; Sometimes we get a zero-length data chunk after the process has
1287 ;; been changed to 'free', which means it has no buffer associated
1288 ;; with it. Do nothing if there is no buffer, or 0 length data.
1292 (url-http-debug "Calling after change function `%s' for `%S'" url-http-after-change-function proc)
1301 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
1302 ;;; file-name-handler stuff from here on out
1303 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
1306 'symbol-value-in-buffer
1308 "Return the value of SYMBOL in BUFFER, or UNBOUND-VALUE if it is unbound."
1311 unbound-value
1336 nil ;dir / link / normal file
1339 nil nil nil ;atime ; mtime ; ctime
1354 "Return a property list describing options available for URL.
1355 This list is retrieved using the `OPTIONS' HTTP method.
1357 Property list members:
1359 methods
1360 A list of symbols specifying what HTTP methods the resource
1361 supports.
1363 dav
1364 A list of numbers specifying what DAV protocol/schema versions are
1365 supported.
1367 dasl
1368 A list of supported DASL search types supported (string form)
1370 ranges
1371 A list of the units available for use in partial document fetches.
1373 p3p
1374 The `Platform For Privacy Protection' description for the resource.
1375 Currently this is just the raw header contents. This is likely to
1376 change once P3P is formally supported by the URL package or
1377 Emacs/W3."
1385 ;; Only parse the options if we got a 2xx response code!
1391 ;; Figure out what methods are supported.
1394 options 'methods
1397 ;; Check for DAV
1400 options 'dav
1405 ;; Now for DASL
1408 options 'dasl
1411 ;; P3P - should get more detailed here. FIXME
1415 ;; Check for whether they accept byte-range requests.
1418 options 'ranges
1422 ))))
1424 options))
1426 ;; HTTPS. This used to be in url-https.el, but that file collides
1427 ;; with url-http.el on systems with 8-character file names.
1433 ;; FIXME what is the point of this alias being an autoload?
1434 ;; Trying to use it will not cause url-http to be loaded,
1435 ;; since the full alias just gets dumped into loaddefs.el.
1437 ;;;###autoload (autoload 'url-default-expander "url-expand")
1438 ;;;###autoload
1448 ;;;###autoload (autoload 'url-https "url-http")
1450 ;;;###autoload (autoload 'url-https-file-exists-p "url-http")
1452 ;;;###autoload (autoload 'url-https-file-readable-p "url-http")
1454 ;;;###autoload (autoload 'url-https-file-attributes "url-http")
1459 ;;; url-http.el ends here