diff options
author | Stefan Monnier <monnier@iro.umontreal.ca> | 2022-09-25 16:15:16 -0400 |
---|---|---|
committer | Stefan Monnier <monnier@iro.umontreal.ca> | 2022-09-25 16:15:16 -0400 |
commit | 650c20f1ca4e07591a727e1cfcc74b3363d15985 (patch) | |
tree | 85d11f6437cde22f410c25e0e5f71a3131ebd07d /lisp/url/url-http.el | |
parent | 8869332684c2302b5ba1ead4568bbc7ba1c0183e (diff) | |
parent | 4b85ae6a24380fb67a3315eaec9233f17a872473 (diff) | |
download | emacs-650c20f1ca4e07591a727e1cfcc74b3363d15985.tar.gz emacs-650c20f1ca4e07591a727e1cfcc74b3363d15985.tar.bz2 emacs-650c20f1ca4e07591a727e1cfcc74b3363d15985.zip |
Merge 'master' into noverlay
Diffstat (limited to 'lisp/url/url-http.el')
-rw-r--r-- | lisp/url/url-http.el | 591 |
1 files changed, 336 insertions, 255 deletions
diff --git a/lisp/url/url-http.el b/lisp/url/url-http.el index 51f158e5c21..94ef156108c 100644 --- a/lisp/url/url-http.el +++ b/lisp/url/url-http.el @@ -1,6 +1,6 @@ ;;; url-http.el --- HTTP retrieval routines -*- lexical-binding:t -*- -;; Copyright (C) 1999, 2001, 2004-2017 Free Software Foundation, Inc. +;; Copyright (C) 1999, 2001, 2004-2022 Free Software Foundation, Inc. ;; Author: Bill Perry <wmperry@gnu.org> ;; Maintainer: emacs-devel@gnu.org @@ -36,6 +36,7 @@ (defvar url-current-object) (defvar url-http-after-change-function) (defvar url-http-chunked-counter) +(defvar url-http-chunked-last-crlf-missing) (defvar url-http-chunked-length) (defvar url-http-chunked-start) (defvar url-http-connection-opened) @@ -54,6 +55,7 @@ (defvar url-http-target-url) (defvar url-http-transfer-encoding) (defvar url-show-status) +(defvar url-http-referer) (require 'url-gw) (require 'url-parse) @@ -65,7 +67,7 @@ (defconst url-http-default-port 80 "Default HTTP port.") (defconst url-http-asynchronous-p t "HTTP retrievals are asynchronous.") -(defalias 'url-http-expand-file-name 'url-default-expander) +(defalias 'url-http-expand-file-name #'url-default-expander) (defvar url-http-real-basic-auth-storage nil) (defvar url-http-proxy-basic-auth-storage nil) @@ -149,16 +151,7 @@ request.") ;; These routines will allow us to implement persistent HTTP ;; connections. (defsubst url-http-debug (&rest args) - (if quit-flag - (let ((proc (get-buffer-process (current-buffer)))) - ;; The user hit C-g, honor it! Some things can get in an - ;; incredibly tight loop (chunked encoding) - (if proc - (progn - (set-process-sentinel proc nil) - (set-process-filter proc nil))) - (error "Transfer interrupted!"))) - (apply 'url-debug 'http args)) + (apply #'url-debug 'http args)) (defun url-http-mark-connection-as-busy (host port proc) (url-http-debug "Marking connection as busy: %s:%d %S" host port proc) @@ -233,11 +226,40 @@ request.") (os-info (unless (and (listp url-privacy-level) (memq 'os url-privacy-level)) (format "(%s; %s)" url-system-type url-os-type))) - (url-info (format "URL/%s" url-version))) + (url-info "URL/Emacs")) (string-join (delq nil (list package-info url-info emacs-info os-info)) " "))) +(defun url-http--get-referer (url) + (url-http-debug "getting referer from buffer: buffer:%S target-url:%S lastloc:%S" (current-buffer) url url-current-lastloc) + (when url-current-lastloc + (if (not (url-p url-current-lastloc)) + (setq url-current-lastloc (url-generic-parse-url url-current-lastloc))) + (let ((referer (copy-sequence url-current-lastloc))) + (setf (url-host referer) (puny-encode-domain (url-host referer))) + (let ((referer-string (url-recreate-url referer))) + (when (and (not (memq url-privacy-level '(low high paranoid))) + (not (and (listp url-privacy-level) + (memq 'lastloc url-privacy-level)))) + ;; url-privacy-level allows referer. But url-lastloc-privacy-level + ;; may restrict who we send it to. + (cl-case url-lastloc-privacy-level + (host-match + (let ((referer-host (url-host referer)) + (url-host (url-host url))) + (when (string= referer-host url-host) + referer-string))) + (domain-match + (let ((referer-domain (url-domain referer)) + (url-domain (url-domain url))) + (when (and referer-domain + url-domain + (string= referer-domain url-domain)) + referer-string))) + (otherwise + referer-string))))))) + ;; Building an HTTP request (defun url-http-user-agent-string () "Compute a User-Agent string. @@ -254,8 +276,9 @@ The string is based on `url-privacy-level' and `url-user-agent'." ((eq url-user-agent 'default) (url-http--user-agent-default-string)))))) (if ua-string (format "User-Agent: %s\r\n" (string-trim ua-string)) ""))) -(defun url-http-create-request (&optional ref-url) - "Create an HTTP request for `url-http-target-url', referred to by REF-URL." +(defun url-http-create-request () + "Create an HTTP request for `url-http-target-url'. +Use `url-http-referer' as the Referer-header (subject to `url-privacy-level')." (let* ((extra-headers) (request nil) (no-cache (cdr-safe (assoc "Pragma" url-http-extra-headers))) @@ -268,13 +291,14 @@ The string is based on `url-privacy-level' and `url-user-agent'." 'url-http-proxy-basic-auth-storage)) (url-get-authentication url-http-proxy nil 'any nil)))) (real-fname (url-filename url-http-target-url)) - (host (url-http--encode-string (url-host url-http-target-url))) + (host (url-host url-http-target-url)) (auth (if (cdr-safe (assoc "Authorization" url-http-extra-headers)) nil (url-get-authentication (or (and (boundp 'proxy-info) proxy-info) - url-http-target-url) nil 'any nil)))) + url-http-target-url) nil 'any nil))) + (ref-url (url-http--encode-string url-http-referer))) (if (equal "" real-fname) (setq real-fname "/")) (setq no-cache (and no-cache (string-match "no-cache" no-cache))) @@ -288,12 +312,6 @@ The string is based on `url-privacy-level' and `url-user-agent'." (string= ref-url ""))) (setq ref-url nil)) - ;; We do not want to expose the referrer if the user is paranoid. - (if (or (memq url-privacy-level '(low high paranoid)) - (and (listp url-privacy-level) - (memq 'lastloc url-privacy-level))) - (setq ref-url nil)) - ;; url-http-extra-headers contains an assoc-list of ;; header/value pairs that we need to put into the request. (setq extra-headers (mapconcat @@ -312,7 +330,13 @@ The string is based on `url-privacy-level' and `url-user-agent'." ;; The request (or url-http-method "GET") " " (url-http--encode-string - (if using-proxy (url-recreate-url url-http-target-url) real-fname)) + (if (and using-proxy + ;; Bug#35969. + (not (equal "https" (url-type url-http-target-url)))) + (let ((url (copy-sequence url-http-target-url))) + (setf (url-host url) (puny-encode-domain (url-host url))) + (url-recreate-url url)) + real-fname)) " HTTP/" url-http-version "\r\n" ;; Version of MIME we speak "MIME-Version: 1.0\r\n" @@ -329,9 +353,11 @@ The string is based on `url-privacy-level' and `url-user-agent'." (url-scheme-get-property (url-type url-http-target-url) 'default-port)) (format - "Host: %s:%d\r\n" (puny-encode-domain host) + "Host: %s:%d\r\n" (url-http--encode-string + (puny-encode-domain host)) (url-port url-http-target-url)) - (format "Host: %s\r\n" (puny-encode-domain host))) + (format "Host: %s\r\n" + (url-http--encode-string (puny-encode-domain host)))) ;; Who its from (if url-personal-mail-address (concat @@ -434,49 +460,57 @@ Return the number of characters removed." auth (strength 0)) - ;; find strongest supported auth - (dolist (this-auth auths) - (setq this-auth (url-eat-trailing-space - (url-strip-leading-spaces - this-auth))) - (let* ((this-type - (downcase (if (string-match "[ \t]" this-auth) - (substring this-auth 0 (match-beginning 0)) - this-auth))) - (registered (url-auth-registered this-type)) - (this-strength (cddr registered))) - (when (and registered (> this-strength strength)) - (setq auth this-auth - type this-type - strength this-strength)))) - - (if (not (url-auth-registered type)) - (progn - (widen) - (goto-char (point-max)) - (insert "<hr>Sorry, but I do not know how to handle " (or type auth url "") - " authentication. If you'd like to write it," - " please use M-x report-emacs-bug RET.<hr>") - ;; We used to set a `status' var (declared "special") but I can't - ;; find the corresponding let-binding, so it's probably an error. - ;; FIXME: Maybe it was supposed to set `success', i.e. to return t? - ;; (setq status t) - nil) ;; Not success yet. - - (let* ((args (url-parse-args (subst-char-in-string ?, ?\; auth))) - (auth (url-get-authentication auth-url - (cdr-safe (assoc "realm" args)) - type t args))) - (if (not auth) - t ;Success. - (push (cons (if proxy "Proxy-Authorization" "Authorization") auth) - url-http-extra-headers) - (let ((url-request-method url-http-method) - (url-request-data url-http-data) - (url-request-extra-headers url-http-extra-headers)) - (url-retrieve-internal url url-callback-function - url-callback-arguments)) - nil))))) ;; Not success yet. + ;; If we're here, then we got a 40x Unauthorized response from the + ;; server. If we already have "Authorization" in the extra + ;; headers, then this means that we've already tried sending + ;; credentials to the server, and they were wrong, so just give + ;; up. + (let ((authorization (assoc "Authorization" url-http-extra-headers))) + (if (and authorization + (not (string-match "^NTLM " (cdr authorization)))) ;Bug#43566 + t ;; Instruct caller to signal an error. Bug#50511 + ;; Find strongest supported auth. + (dolist (this-auth auths) + (setq this-auth (string-trim this-auth)) + (let* ((this-type + (downcase (if (string-match "[ \t]" this-auth) + (substring this-auth 0 (match-beginning 0)) + this-auth))) + (registered (url-auth-registered this-type)) + (this-strength (cddr registered))) + (when (and registered (> this-strength strength)) + (setq auth this-auth + type this-type + strength this-strength)))) + + (if (not (url-auth-registered type)) + (progn + (widen) + (goto-char (point-max)) + (insert "<hr>Sorry, but I do not know how to handle " + (or type auth url "") + " authentication. If you'd like to write it," + " please use M-x report-emacs-bug RET.<hr>") + ;; We used to set a `status' var (declared "special") but I can't + ;; find the corresponding let-binding, so it's probably an error. + ;; FIXME: Maybe it was supposed to set `success', i.e. to return t? + ;; (setq status t) + nil) ;; Not success yet. + + (let* ((args (url-parse-args (subst-char-in-string ?, ?\; auth))) + (auth (url-get-authentication auth-url + (cdr-safe (assoc "realm" args)) + type t args))) + (if (not auth) + t ;Success. + (push (cons (if proxy "Proxy-Authorization" "Authorization") auth) + url-http-extra-headers) + (let ((url-request-method url-http-method) + (url-request-data url-http-data) + (url-request-extra-headers url-http-extra-headers)) + (url-retrieve-internal url url-callback-function + url-callback-arguments)) + nil))))))) ;; Not success yet. (defun url-http-parse-response () "Parse just the response code." @@ -485,11 +519,11 @@ Return the number of characters removed." (url-http-debug "url-http-parse-response called in (%s)" (buffer-name)) (goto-char (point-min)) (skip-chars-forward " \t\n") ; Skip any blank crap - (skip-chars-forward "HTTP/") ; Skip HTTP Version + (skip-chars-forward "/HPT") ; Skip HTTP Version "HTTP/". (setq url-http-response-version (buffer-substring (point) (progn - (skip-chars-forward "[0-9].") + (skip-chars-forward "0-9.") (point)))) (setq url-http-response-status (read (current-buffer)))) @@ -511,6 +545,24 @@ work correctly." (declare-function gnutls-peer-status "gnutls.c" (proc)) (declare-function gnutls-negotiate "gnutls.el" t t) +(defun url-http--insert-file-helper (buffer url &optional visit) + (with-current-buffer buffer + (when (bound-and-true-p url-http-response-status) + ;; Don't signal an error if VISIT is non-nil, because + ;; 'insert-file-contents' doesn't. This is required to + ;; support, e.g., 'browse-url-emacs', which is a fancy way of + ;; visiting the HTML source of a URL: in that case, we want to + ;; display a file buffer even if the URL does not exist and + ;; 'url-retrieve-synchronously' returns 404 or whatever. + (unless (or visit + (or (and (>= url-http-response-status 200) + (< url-http-response-status 300)) + (= url-http-response-status 304))) ; "Not modified" + (let ((desc (nth 2 (assq url-http-response-status url-http-codes)))) + (kill-buffer buffer) + ;; Signal file-error per bug#16733. + (signal 'file-error (list url desc))))))) + (defun url-http-parse-headers () "Parse and handle HTTP specific headers. Return t if and only if the current buffer is still active and @@ -535,6 +587,13 @@ should be shown to the user." (url-http-debug "url-http-parse-headers called in (%s)" (buffer-name)) (url-http-parse-response) (mail-narrow-to-head) + (when url-debug + (save-excursion + (goto-char (point-min)) + (while (not (eobp)) + (url-http-debug "Response: %s" + (buffer-substring (point) (line-end-position))) + (forward-line 1)))) ;;(narrow-to-region (point-min) url-http-end-of-headers) (let ((connection (mail-fetch-field "Connection"))) ;; In HTTP 1.0, keep the connection only if there is a @@ -585,7 +644,7 @@ should be shown to the user." ;; 206 Partial content ;; 207 Multi-status (Added by DAV) (pcase status-symbol - ((or `no-content `reset-content) + ((or 'no-content 'reset-content) ;; No new data, just stay at the same document (url-mark-buffer-as-dead buffer)) (_ @@ -606,7 +665,7 @@ should be shown to the user." (let ((redirect-uri (or (mail-fetch-field "Location") (mail-fetch-field "URI")))) (pcase status-symbol - (`multiple-choices ; 300 + ('multiple-choices ; 300 ;; Quoth the spec (section 10.3.1) ;; ------------------------------- ;; The requested resource corresponds to any one of a set of @@ -623,40 +682,38 @@ should be shown to the user." ;; We do not support agent-driven negotiation, so we just ;; redirect to the preferred URI if one is provided. nil) - (`see-other ; 303 + ('found ; 302 + ;; 302 Found was ambiguously defined in the standards, but + ;; it's now recommended that it's treated like 303 instead + ;; of 307, since that's what most servers expect. + (setq url-http-method "GET" + url-http-data nil)) + ('see-other ; 303 ;; The response to the request can be found under a different ;; URI and SHOULD be retrieved using a GET method on that ;; resource. (setq url-http-method "GET" url-http-data nil)) - (`not-modified ; 304 + ('not-modified ; 304 ;; The 304 response MUST NOT contain a message-body. (url-http-debug "Extracting document from cache... (%s)" (url-cache-create-filename (url-view-url t))) (url-cache-extract (url-cache-create-filename (url-view-url t))) (setq redirect-uri nil success t)) - (`use-proxy ; 305 + ('use-proxy ; 305 ;; The requested resource MUST be accessed through the ;; proxy given by the Location field. The Location field ;; gives the URI of the proxy. The recipient is expected ;; to repeat this single request via the proxy. 305 ;; responses MUST only be generated by origin servers. - (error "Redirection thru a proxy server not supported: %s" + (error "Redirection through a proxy server not supported: %s" redirect-uri)) (_ ;; Treat everything like '300' nil)) (when redirect-uri - ;; Clean off any whitespace and/or <...> cruft. - (if (string-match "\\([^ \t]+\\)[ \t]" redirect-uri) - (setq redirect-uri (match-string 1 redirect-uri))) - (if (string-match "^<\\(.*\\)>$" redirect-uri) - (setq redirect-uri (match-string 1 redirect-uri))) - - ;; Some stupid sites (like sourceforge) send a - ;; non-fully-qualified URL (ie: /), which royally confuses - ;; the URL library. + ;; Handle relative redirect URIs. (if (not (string-match url-nonrelative-link redirect-uri)) ;; Be careful to use the real target URL, otherwise we may ;; compute the redirection relative to the URL of the proxy. @@ -695,12 +752,12 @@ should be shown to the user." ;; without changing the API. Instead url-retrieve should ;; either simply not return the "destination" buffer, or it ;; should take an optional `dest-buf' argument. - (set (make-local-variable 'url-redirect-buffer) - (url-retrieve-internal - redirect-uri url-callback-function - url-callback-arguments - (url-silent url-current-object) - (not (url-use-cookies url-current-object)))) + (setq-local url-redirect-buffer + (url-retrieve-internal + redirect-uri url-callback-function + url-callback-arguments + (url-silent url-current-object) + (not (url-use-cookies url-current-object)))) (url-mark-buffer-as-dead buffer)) ;; We hit url-max-redirections, so issue an error and ;; stop redirecting. @@ -734,50 +791,50 @@ should be shown to the user." ;; 424 Failed Dependency (setq success (pcase status-symbol - (`unauthorized ; 401 + ('unauthorized ; 401 ;; The request requires user authentication. The response ;; MUST include a WWW-Authenticate header field containing a ;; challenge applicable to the requested resource. The ;; client MAY repeat the request with a suitable ;; Authorization header field. (url-http-handle-authentication nil)) - (`payment-required ; 402 + ('payment-required ; 402 ;; This code is reserved for future use (url-mark-buffer-as-dead buffer) (error "Somebody wants you to give them money")) - (`forbidden ; 403 + ('forbidden ; 403 ;; The server understood the request, but is refusing to ;; fulfill it. Authorization will not help and the request ;; SHOULD NOT be repeated. t) - (`not-found ; 404 + ('not-found ; 404 ;; Not found t) - (`method-not-allowed ; 405 + ('method-not-allowed ; 405 ;; The method specified in the Request-Line is not allowed ;; for the resource identified by the Request-URI. The ;; response MUST include an Allow header containing a list of ;; valid methods for the requested resource. t) - (`not-acceptable ; 406 + ('not-acceptable ; 406 ;; The resource identified by the request is only capable of ;; generating response entities which have content ;; characteristics not acceptable according to the accept ;; headers sent in the request. t) - (`proxy-authentication-required ; 407 + ('proxy-authentication-required ; 407 ;; This code is similar to 401 (Unauthorized), but indicates ;; that the client must first authenticate itself with the ;; proxy. The proxy MUST return a Proxy-Authenticate header ;; field containing a challenge applicable to the proxy for ;; the requested resource. (url-http-handle-authentication t)) - (`request-timeout ; 408 + ('request-timeout ; 408 ;; The client did not produce a request within the time that ;; the server was prepared to wait. The client MAY repeat ;; the request without modifications at any later time. t) - (`conflict ; 409 + ('conflict ; 409 ;; The request could not be completed due to a conflict with ;; the current state of the resource. This code is only ;; allowed in situations where it is expected that the user @@ -786,11 +843,11 @@ should be shown to the user." ;; information for the user to recognize the source of the ;; conflict. t) - (`gone ; 410 + ('gone ; 410 ;; The requested resource is no longer available at the ;; server and no forwarding address is known. t) - (`length-required ; 411 + ('length-required ; 411 ;; The server refuses to accept the request without a defined ;; Content-Length. The client MAY repeat the request if it ;; adds a valid Content-Length header field containing the @@ -800,29 +857,29 @@ should be shown to the user." ;; `url-http-create-request' automatically calculates the ;; content-length. t) - (`precondition-failed ; 412 + ('precondition-failed ; 412 ;; The precondition given in one or more of the ;; request-header fields evaluated to false when it was ;; tested on the server. t) - ((or `request-entity-too-large `request-uri-too-large) ; 413 414 + ((or 'request-entity-too-large 'request-uri-too-large) ; 413 414 ;; The server is refusing to process a request because the ;; request entity|URI is larger than the server is willing or ;; able to process. t) - (`unsupported-media-type ; 415 + ('unsupported-media-type ; 415 ;; The server is refusing to service the request because the ;; entity of the request is in a format not supported by the ;; requested resource for the requested method. t) - (`requested-range-not-satisfiable ; 416 + ('requested-range-not-satisfiable ; 416 ;; A server SHOULD return a response with this status code if ;; a request included a Range request-header field, and none ;; of the range-specifier values in this field overlap the ;; current extent of the selected resource, and the request ;; did not include an If-Range request-header field. t) - (`expectation-failed ; 417 + ('expectation-failed ; 417 ;; The expectation given in an Expect request-header field ;; could not be met by this server, or, if the server is a ;; proxy, the server has unambiguous evidence that the @@ -849,16 +906,16 @@ should be shown to the user." ;; 507 Insufficient storage (setq success t) (pcase url-http-response-status - (`not-implemented ; 501 + ('not-implemented ; 501 ;; The server does not support the functionality required to ;; fulfill the request. nil) - (`bad-gateway ; 502 + ('bad-gateway ; 502 ;; The server, while acting as a gateway or proxy, received ;; an invalid response from the upstream server it accessed ;; in attempting to fulfill the request. nil) - (`service-unavailable ; 503 + ('service-unavailable ; 503 ;; The server is currently unable to handle the request due ;; to a temporary overloading or maintenance of the server. ;; The implication is that this is a temporary condition @@ -867,19 +924,19 @@ should be shown to the user." ;; header. If no Retry-After is given, the client SHOULD ;; handle the response as it would for a 500 response. nil) - (`gateway-timeout ; 504 + ('gateway-timeout ; 504 ;; The server, while acting as a gateway or proxy, did not ;; receive a timely response from the upstream server ;; specified by the URI (e.g. HTTP, FTP, LDAP) or some other ;; auxiliary server (e.g. DNS) it needed to access in ;; attempting to complete the request. nil) - (`http-version-not-supported ; 505 + ('http-version-not-supported ; 505 ;; The server does not support, or refuses to support, the ;; HTTP protocol version that was used in the request ;; message. nil) - (`insufficient-storage ; 507 (DAV) + ('insufficient-storage ; 507 (DAV) ;; The method could not be performed on the resource ;; because the server is unable to store the representation ;; needed to successfully complete the request. This @@ -899,16 +956,21 @@ should be shown to the user." class url-http-response-status))) (if (not success) (url-mark-buffer-as-dead buffer) + ;; Narrow the buffer for url-handle-content-transfer-encoding to + ;; find only the headers relevant to this transaction. + (and (not (buffer-narrowed-p)) + (mail-narrow-to-head)) (url-handle-content-transfer-encoding)) (url-http-debug "Finished parsing HTTP headers: %S" success) (widen) (goto-char (point-min)) success)) -(declare-function zlib-decompress-region "decompress.c" (start end)) +(declare-function zlib-decompress-region "decompress.c" + (start end &optional allow-partial)) (defun url-handle-content-transfer-encoding () - (let ((encoding (mail-fetch-field "content-encoding"))) + (let ((encoding (mail-fetch-field "content-encoding" nil nil nil t))) (when (and encoding (fboundp 'zlib-available-p) (zlib-available-p) @@ -917,7 +979,7 @@ should be shown to the user." (widen) (goto-char (point-min)) (when (search-forward "\n\n") - (zlib-decompress-region (point) (point-max))))))) + (zlib-decompress-region (point) (point-max) t)))))) ;; Miscellaneous (defun url-http-activate-callback () @@ -966,14 +1028,17 @@ should be shown to the user." (setq url-using-proxy (url-generic-parse-url url-using-proxy))) (url-http url-current-object url-callback-function - url-callback-arguments (current-buffer))))) + url-callback-arguments (current-buffer) + (and (string= "https" (url-type url-current-object)) + 'tls))))) ((url-http-parse-headers) (url-http-activate-callback)))))) (defun url-http-simple-after-change-function (_st _nd _length) ;; Function used when we do NOT know how long the document is going to be ;; Just _very_ simple 'downloaded %d' type of info. - (url-lazy-message "Reading %s..." (file-size-human-readable (buffer-size)))) + (url-lazy-message "Reading %s..." + (funcall byte-count-to-string-function (buffer-size)))) (defun url-http-content-length-after-change-function (_st nd _length) "Function used when we DO know how long the document is going to be. @@ -981,28 +1046,23 @@ More sophisticated percentage downloaded, etc. Also does minimal parsing of HTTP headers and will actually cause the callback to be triggered." (if url-http-content-type - (url-display-percentage + (url-display-message "Reading [%s]... %s of %s (%d%%)" - (url-percentage (- nd url-http-end-of-headers) - url-http-content-length) url-http-content-type - (file-size-human-readable (- nd url-http-end-of-headers)) - (file-size-human-readable url-http-content-length) + (funcall byte-count-to-string-function (- nd url-http-end-of-headers)) + (funcall byte-count-to-string-function url-http-content-length) (url-percentage (- nd url-http-end-of-headers) url-http-content-length)) - (url-display-percentage + (url-display-message "Reading... %s of %s (%d%%)" - (url-percentage (- nd url-http-end-of-headers) - url-http-content-length) - (file-size-human-readable (- nd url-http-end-of-headers)) - (file-size-human-readable url-http-content-length) + (funcall byte-count-to-string-function (- nd url-http-end-of-headers)) + (funcall byte-count-to-string-function url-http-content-length) (url-percentage (- nd url-http-end-of-headers) url-http-content-length))) (if (> (- nd url-http-end-of-headers) url-http-content-length) (progn ;; Found the end of the document! Wheee! - (url-display-percentage nil nil) (url-lazy-message "Reading... done.") (if (url-http-parse-headers) (url-http-activate-callback))))) @@ -1012,87 +1072,97 @@ the callback to be triggered." Cannot give a sophisticated percentage, but we need a different function to look for the special 0-length chunk that signifies the end of the document." - (save-excursion - (goto-char st) - (let ((read-next-chunk t) - (case-fold-search t) - (regexp nil) - (no-initial-crlf nil)) - ;; We need to loop thru looking for more chunks even within - ;; one after-change-function call. - (while read-next-chunk - (setq no-initial-crlf (= 0 url-http-chunked-counter)) - (if url-http-content-type - (url-display-percentage nil - "Reading [%s]... chunk #%d" - url-http-content-type url-http-chunked-counter) - (url-display-percentage nil - "Reading... chunk #%d" - url-http-chunked-counter)) - (url-http-debug "Reading chunk %d (%d %d %d)" - url-http-chunked-counter st nd length) - (setq regexp (if no-initial-crlf - "\\([0-9a-z]+\\).*\r?\n" - "\r?\n\\([0-9a-z]+\\).*\r?\n")) - - (if url-http-chunked-start - ;; We know how long the chunk is supposed to be, skip over - ;; leading crap if possible. - (if (> nd (+ url-http-chunked-start url-http-chunked-length)) - (progn - (url-http-debug "Got to the end of chunk #%d!" - url-http-chunked-counter) - (goto-char (+ url-http-chunked-start - url-http-chunked-length))) - (url-http-debug "Still need %d bytes to hit end of chunk" - (- (+ url-http-chunked-start - url-http-chunked-length) - nd)) - (setq read-next-chunk nil))) - (if (not read-next-chunk) - (url-http-debug "Still spinning for next chunk...") - (if no-initial-crlf (skip-chars-forward "\r\n")) - (if (not (looking-at regexp)) - (progn - ;; Must not have received the entirety of the chunk header, - ;; need to spin some more. - (url-http-debug "Did not see start of chunk @ %d!" (point)) - (setq read-next-chunk nil)) - (add-text-properties (match-beginning 0) (match-end 0) - (list 'start-open t - 'end-open t - 'chunked-encoding t - 'face 'cursor - 'invisible t)) - (setq url-http-chunked-length (string-to-number (buffer-substring - (match-beginning 1) - (match-end 1)) - 16) - url-http-chunked-counter (1+ url-http-chunked-counter) - url-http-chunked-start (set-marker - (or url-http-chunked-start - (make-marker)) - (match-end 0))) -; (if (not url-http-debug) - (delete-region (match-beginning 0) (match-end 0));) - (url-http-debug "Saw start of chunk %d (length=%d, start=%d" - url-http-chunked-counter url-http-chunked-length - (marker-position url-http-chunked-start)) - (if (= 0 url-http-chunked-length) - (progn - ;; Found the end of the document! Wheee! - (url-http-debug "Saw end of stream chunk!") - (setq read-next-chunk nil) - (url-display-percentage nil nil) - ;; Every chunk, even the last 0-length one, is - ;; terminated by CRLF. Skip it. - (when (looking-at "\r?\n") - (url-http-debug "Removing terminator of last chunk") - (delete-region (match-beginning 0) (match-end 0))) - (if (re-search-forward "^\r?\n" nil t) - (url-http-debug "Saw end of trailers...")) - (if (url-http-parse-headers) - (url-http-activate-callback)))))))))) + (if url-http-chunked-last-crlf-missing + (progn + (goto-char url-http-chunked-last-crlf-missing) + (if (not (looking-at "\r\n")) + (url-http-debug + "Still spinning for the terminator of last chunk...") + (url-http-debug "Saw the last CRLF.") + (delete-region (match-beginning 0) (match-end 0)) + (when (url-http-parse-headers) + (url-http-activate-callback)))) + (save-excursion + (goto-char st) + (let ((read-next-chunk t) + (case-fold-search t) + (regexp nil) + (no-initial-crlf nil)) + ;; We need to loop thru looking for more chunks even within + ;; one after-change-function call. + (while read-next-chunk + (setq no-initial-crlf (= 0 url-http-chunked-counter)) + (url-http-debug "Reading chunk %d (%d %d %d)" + url-http-chunked-counter st nd length) + (setq regexp (if no-initial-crlf + "\\([0-9a-z]+\\).*\r?\n" + "\r?\n\\([0-9a-z]+\\).*\r?\n")) + + (if url-http-chunked-start + ;; We know how long the chunk is supposed to be, skip over + ;; leading crap if possible. + (if (> nd (+ url-http-chunked-start url-http-chunked-length)) + (progn + (url-http-debug "Got to the end of chunk #%d!" + url-http-chunked-counter) + (goto-char (+ url-http-chunked-start + url-http-chunked-length))) + (url-http-debug "Still need %d bytes to hit end of chunk" + (- (+ url-http-chunked-start + url-http-chunked-length) + nd)) + (setq read-next-chunk nil))) + (if (not read-next-chunk) + (url-http-debug "Still spinning for next chunk...") + (if no-initial-crlf (skip-chars-forward "\r\n")) + (if (not (looking-at regexp)) + (progn + ;; Must not have received the entirety of the chunk header, + ;; need to spin some more. + (url-http-debug "Did not see start of chunk @ %d!" (point)) + (setq read-next-chunk nil)) + ;; The data we got may have started in the middle of the + ;; initial chunk header, so move back to the start of the + ;; line and re-compute. + (when (= url-http-chunked-counter 0) + (beginning-of-line) + (looking-at regexp)) + (add-text-properties (match-beginning 0) (match-end 0) + (list 'chunked-encoding t + 'face 'cursor + 'invisible t)) + (setq url-http-chunked-length + (string-to-number (buffer-substring (match-beginning 1) + (match-end 1)) + 16) + url-http-chunked-counter (1+ url-http-chunked-counter) + url-http-chunked-start (set-marker + (or url-http-chunked-start + (make-marker)) + (match-end 0))) + (delete-region (match-beginning 0) (match-end 0)) + (url-http-debug "Saw start of chunk %d (length=%d, start=%d" + url-http-chunked-counter url-http-chunked-length + (marker-position url-http-chunked-start)) + (if (= 0 url-http-chunked-length) + (progn + ;; Found the end of the document! Wheee! + (url-http-debug "Saw end of stream chunk!") + (setq read-next-chunk nil) + ;; Every chunk, even the last 0-length one, is + ;; terminated by CRLF. Skip it. + (if (not (looking-at "\r?\n")) + (progn + (url-http-debug + "Spinning for the terminator of last chunk...") + (setq url-http-chunked-last-crlf-missing + (point))) + (url-http-debug "Removing terminator of last chunk") + (delete-region (match-beginning 0) (match-end 0)) + (when (re-search-forward "^\r?\n" nil t) + (url-http-debug "Saw end of trailers...")) + (when (url-http-parse-headers) + (url-http-activate-callback)))))))))))) (defun url-http-wait-for-headers-change-function (_st nd _length) ;; This will wait for the headers to arrive and then splice in the @@ -1146,8 +1216,7 @@ the end of the document." ;; We got back a headerless malformed response from the ;; server. (url-http-activate-callback)) - ((or (= url-http-response-status 204) - (= url-http-response-status 205)) + ((memq url-http-response-status '(204 205)) (url-http-debug "%d response must have headers only (%s)." url-http-response-status (buffer-name)) (when (url-http-parse-headers) @@ -1182,11 +1251,11 @@ the end of the document." (url-http-debug "Saw HTTP/0.9 response, connection closed means end of document.") (setq url-http-after-change-function - 'url-http-simple-after-change-function)) + #'url-http-simple-after-change-function)) ((equal url-http-transfer-encoding "chunked") (url-http-debug "Saw chunked encoding.") (setq url-http-after-change-function - 'url-http-chunked-encoding-after-change-function) + #'url-http-chunked-encoding-after-change-function) (when (> nd url-http-end-of-headers) (url-http-debug "Calling initial chunked-encoding for extra data at end of headers") @@ -1197,7 +1266,7 @@ the end of the document." (url-http-debug "Got a content-length, being smart about document end.") (setq url-http-after-change-function - 'url-http-content-length-after-change-function) + #'url-http-content-length-after-change-function) (cond ((= 0 url-http-content-length) ;; We got a NULL body! Activate the callback @@ -1218,7 +1287,7 @@ the end of the document." (t (url-http-debug "No content-length, being dumb.") (setq url-http-after-change-function - 'url-http-simple-after-change-function))))) + #'url-http-simple-after-change-function))))) ;; We are still at the beginning of the buffer... must just be ;; waiting for a response. (url-http-debug "Spinning waiting for headers...") @@ -1234,7 +1303,7 @@ passing it an updated value of CBARGS as arguments. The first element in CBARGS should be a plist describing what has happened so far during the request, as described in the docstring of `url-retrieve' (if in doubt, specify nil). The current buffer -then CALLBACK is executed is the retrieval buffer. +when CALLBACK is executed is the retrieval buffer. Optional arg RETRY-BUFFER, if non-nil, specifies the buffer of a previous `url-http' call, which is being re-attempted. @@ -1246,9 +1315,7 @@ The return value of this function is the retrieval buffer." (cl-check-type url url "Need a pre-parsed URL.") (let* (;; (host (url-host (or url-using-proxy url))) ;; (port (url-port (or url-using-proxy url))) - (nsm-noninteractive (or url-request-noninteractive - (and (boundp 'url-http-noninteractive) - url-http-noninteractive))) + (nsm-noninteractive (not (url-interactive-p))) ;; The following binding is needed in url-open-stream, which ;; is called from url-http-find-free-connection. (url-current-object url) @@ -1258,7 +1325,8 @@ The return value of this function is the retrieval buffer." (mime-accept-string url-mime-accept-string) (buffer (or retry-buffer (generate-new-buffer - (format " *http %s:%d*" (url-host url) (url-port url)))))) + (format " *http %s:%d*" (url-host url) (url-port url))))) + (referer (url-http--encode-string (url-http--get-referer url)))) (if (not connection) ;; Failed to open the connection for some reason (progn @@ -1278,6 +1346,7 @@ The return value of this function is the retrieval buffer." url-http-after-change-function url-http-response-version url-http-response-status + url-http-chunked-last-crlf-missing url-http-chunked-length url-http-chunked-counter url-http-chunked-start @@ -1293,7 +1362,8 @@ The return value of this function is the retrieval buffer." url-http-no-retry url-http-connection-opened url-mime-accept-string - url-http-proxy)) + url-http-proxy + url-http-referer)) (set (make-local-variable var) nil)) (setq url-http-method (or url-request-method "GET") @@ -1301,6 +1371,7 @@ The return value of this function is the retrieval buffer." url-http-noninteractive url-request-noninteractive url-http-data url-request-data url-http-process connection + url-http-chunked-last-crlf-missing nil url-http-chunked-length nil url-http-chunked-start nil url-http-chunked-counter 0 @@ -1311,15 +1382,16 @@ The return value of this function is the retrieval buffer." url-http-no-retry retry-buffer url-http-connection-opened nil url-mime-accept-string mime-accept-string - url-http-proxy url-using-proxy) + url-http-proxy url-using-proxy + url-http-referer referer) (set-process-buffer connection buffer) - (set-process-filter connection 'url-http-generic-filter) + (set-process-filter connection #'url-http-generic-filter) (pcase (process-status connection) - (`connect + ('connect ;; Asynchronous connection (set-process-sentinel connection 'url-http-async-sentinel)) - (`failed + ('failed ;; Asynchronous connection failed (error "Could not create connection to %s:%d" (url-host url) (url-port url))) @@ -1328,19 +1400,28 @@ The return value of this function is the retrieval buffer." (url-type url-current-object))) (url-https-proxy-connect connection) (set-process-sentinel connection - 'url-http-end-of-document-sentinel) + #'url-http-end-of-document-sentinel) (process-send-string connection (url-http-create-request))))))) buffer)) (defun url-https-proxy-connect (connection) - (setq url-http-after-change-function 'url-https-proxy-after-change-function) - (process-send-string connection (format (concat "CONNECT %s:%d HTTP/1.1\r\n" - "Host: %s\r\n" - "\r\n") - (url-host url-current-object) - (or (url-port url-current-object) - url-https-default-port) - (url-host url-current-object)))) + (setq url-http-after-change-function #'url-https-proxy-after-change-function) + (process-send-string + connection + (format + (concat "CONNECT %s:%d HTTP/1.1\r\n" + "Host: %s\r\n" + (let ((proxy-auth (let ((url-basic-auth-storage + 'url-http-proxy-basic-auth-storage)) + (url-get-authentication url-http-proxy nil + 'any nil)))) + (and proxy-auth + (concat "Proxy-Authorization: " proxy-auth "\r\n"))) + "\r\n") + (puny-encode-domain (url-host url-current-object)) + (or (url-port url-current-object) + url-https-default-port) + (puny-encode-domain (url-host url-current-object))))) (defun url-https-proxy-after-change-function (_st _nd _length) (let* ((process-buffer (current-buffer)) @@ -1362,17 +1443,17 @@ The return value of this function is the retrieval buffer." (condition-case e (let ((tls-connection (gnutls-negotiate :process proc - :hostname (url-host url-current-object) + :hostname (puny-encode-domain (url-host url-current-object)) :verify-error nil))) ;; check certificate validity (setq tls-connection (nsm-verify-connection tls-connection - (url-host url-current-object) + (puny-encode-domain (url-host url-current-object)) (url-port url-current-object))) (with-current-buffer process-buffer (erase-buffer)) (set-process-buffer tls-connection process-buffer) (setq url-http-after-change-function - 'url-http-wait-for-headers-change-function) + #'url-http-wait-for-headers-change-function) (set-process-filter tls-connection 'url-http-generic-filter) (process-send-string tls-connection (url-http-create-request))) @@ -1381,10 +1462,10 @@ The return value of this function is the retrieval buffer." (error "gnutls-error: %s" e)) (error (url-http-activate-callback) - (error "error: %s" e))) - (error "error: gnutls support needed!"))) + (error "Error: %s" e))) + (error "Error: gnutls support needed!"))) (t - (message "error response: %d" url-http-response-status) + (url-http-debug "error response: %d" url-http-response-status) (url-http-activate-callback)))))) (defun url-http-async-sentinel (proc why) @@ -1424,24 +1505,25 @@ The return value of this function is the retrieval buffer." ;; Sometimes we get a zero-length data chunk after the process has ;; been changed to 'free', which means it has no buffer associated ;; with it. Do nothing if there is no buffer, or 0 length data. - (and (process-buffer proc) - (/= (length data) 0) - (with-current-buffer (process-buffer proc) - (url-http-debug "Calling after change function `%s' for `%S'" url-http-after-change-function proc) - (funcall url-http-after-change-function - (point-max) - (progn - (goto-char (point-max)) - (insert data) - (point-max)) - (length data))))) + (let ((b (process-buffer proc))) + (when (and (buffer-live-p b) (not (zerop (length data)))) + (with-current-buffer b + (url-http-debug "Calling after change function `%s' for `%S'" + url-http-after-change-function proc) + (funcall url-http-after-change-function + (point-max) + (progn + (goto-char (point-max)) + (insert data) + (point-max)) + (length data)))))) ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;;; file-name-handler stuff from here on out ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; (defalias 'url-http-symbol-value-in-buffer (if (fboundp 'symbol-value-in-buffer) - 'symbol-value-in-buffer + #'symbol-value-in-buffer (lambda (symbol buffer &optional unbound-value) "Return the value of SYMBOL in BUFFER, or UNBOUND-VALUE if it is unbound." (with-current-buffer buffer @@ -1563,7 +1645,6 @@ p3p ;; HTTPS. This used to be in url-https.el, but that file collides ;; with url-http.el on systems with 8-character file names. -(require 'tls) (defconst url-https-asynchronous-p t "HTTPS retrievals are asynchronous.") |