lisp/url/url.el

   1 ;;; url.el --- Uniform Resource Locator retrieval tool
   2
   3 ;; Copyright (C) 1996, 1997, 1998, 1999, 2001, 2004,
   4 ;;   2005, 2006, 2007, 2008, 2009, 2010, 2011  Free Software Foundation, Inc.
   5
   6 ;; Author: Bill Perry <wmperry@gnu.org>
   7 ;; Keywords: comm, data, processes, hypermedia
   8
   9 ;; This file is part of GNU Emacs.
  10 ;;
  11 ;; GNU Emacs is free software: you can redistribute it and/or modify
  12 ;; it under the terms of the GNU General Public License as published by
  13 ;; the Free Software Foundation, either version 3 of the License, or
  14 ;; (at your option) any later version.
  15
  16 ;; GNU Emacs is distributed in the hope that it will be useful,
  17 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  19 ;; GNU General Public License for more details.
  20
  21 ;; You should have received a copy of the GNU General Public License
  22 ;; along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.
  23
  24 ;;; Commentary:
  25
  26 ;; Registered URI schemes: http://www.iana.org/assignments/uri-schemes
  27
  28 ;;; Code:
  29
  30 (eval-when-compile (require 'cl))
  31
  32 (eval-when-compile
  33   (require 'mm-decode)
  34   (require 'mm-view))
  35
  36 (require 'mailcap)
  37 (require 'url-vars)
  38 (require 'url-cookie)
  39 (require 'url-history)
  40 (require 'url-expand)
  41 (require 'url-privacy)
  42 (require 'url-methods)
  43 (require 'url-proxy)
  44 (require 'url-parse)
  45 (require 'url-util)
  46
  47
  48 (defcustom url-configuration-directory
  49   (locate-user-emacs-file "url/" ".url/")
  50   "Directory used by the URL package for cookies, history, etc."
  51   :type 'directory
  52   :group 'url)
  53
  54 (defun url-do-setup ()
  55   "Setup the URL package.
  56 This is to avoid conflict with user settings if URL is dumped with
  57 Emacs."
  58   (unless url-setup-done
  59
  60     ;; Make OS/2 happy
  61     ;;(push '("http" "80") tcp-binary-process-input-services)
  62
  63     (mailcap-parse-mailcaps)
  64     (mailcap-parse-mimetypes)
  65
  66     ;; Register all the authentication schemes we can handle
  67     (url-register-auth-scheme "basic" nil 4)
  68     (url-register-auth-scheme "digest" nil 7)
  69
  70     (setq url-cookie-file
  71           (or url-cookie-file
  72               (expand-file-name "cookies" url-configuration-directory)))
  73
  74     (setq url-history-file
  75           (or url-history-file
  76               (expand-file-name "history" url-configuration-directory)))
  77
  78     ;; Parse the global history file if it exists, so that it can be used
  79     ;; for URL completion, etc.
  80     (url-history-parse-history)
  81     (url-history-setup-save-timer)
  82
  83     ;; Ditto for cookies
  84     (url-cookie-setup-save-timer)
  85     (url-cookie-parse-file url-cookie-file)
  86
  87     ;; Read in proxy gateways
  88     (let ((noproxy (and (not (assoc "no_proxy" url-proxy-services))
  89                         (or (getenv "NO_PROXY")
  90                             (getenv "no_PROXY")
  91                             (getenv "no_proxy")))))
  92       (if noproxy
  93           (setq url-proxy-services
  94                 (cons (cons "no_proxy"
  95                             (concat "\\("
  96                                     (mapconcat
  97                                      (lambda (x)
  98                                        (cond
  99                                         ((= x ?,) "\\|")
 100                                         ((= x ? ) "")
 101                                         ((= x ?.) (regexp-quote "."))
 102                                         ((= x ?*) ".*")
 103                                         ((= x ??) ".")
 104                                         (t (char-to-string x))))
 105                                      noproxy "") "\\)"))
 106                       url-proxy-services))))
 107
 108     (url-setup-privacy-info)
 109     (run-hooks 'url-load-hook)
 110     (setq url-setup-done t)))
 111
 112 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 113 ;;; Retrieval functions
 114 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 115
 116 (defvar url-redirect-buffer nil
 117   "New buffer into which the retrieval will take place.
 118 Sometimes while retrieving a URL, the URL library needs to use another buffer
 119 than the one returned initially by `url-retrieve'.  In this case, it sets this
 120 variable in the original buffer as a forwarding pointer.")
 121
 122 ;;;###autoload
 123 (defun url-retrieve (url callback &optional cbargs)
 124   "Retrieve URL asynchronously and call CALLBACK with CBARGS when finished.
 125 URL is either a string or a parsed URL.
 126
 127 CALLBACK is called when the object has been completely retrieved, with
 128 the current buffer containing the object, and any MIME headers associated
 129 with it.  It is called as (apply CALLBACK STATUS CBARGS).
 130 STATUS is a list with an even number of elements representing
 131 what happened during the request, with most recent events first,
 132 or an empty list if no events have occurred.  Each pair is one of:
 133
 134 \(:redirect REDIRECTED-TO) - the request was redirected to this URL
 135 \(:error (ERROR-SYMBOL . DATA)) - an error occurred.  The error can be
 136 signaled with (signal ERROR-SYMBOL DATA).
 137
 138 Return the buffer URL will load into, or nil if the process has
 139 already completed (i.e. URL was a mailto URL or similar; in this case
 140 the callback is not called).
 141
 142 The variables `url-request-data', `url-request-method' and
 143 `url-request-extra-headers' can be dynamically bound around the
 144 request; dynamic binding of other variables doesn't necessarily
 145 take effect."
 146 ;;; XXX: There is code in Emacs that does dynamic binding
 147 ;;; of the following variables around url-retrieve:
 148 ;;; url-standalone-mode, url-gateway-unplugged, w3-honor-stylesheets,
 149 ;;; url-confirmation-func, url-cookie-multiple-line,
 150 ;;; url-cookie-{{,secure-}storage,confirmation}
 151 ;;; url-standalone-mode and url-gateway-unplugged should work as
 152 ;;; usual.  url-confirmation-func is only used in nnwarchive.el and
 153 ;;; webmail.el; the latter should be updated.  Is
 154 ;;; url-cookie-multiple-line needed anymore?  The other url-cookie-*
 155 ;;; are (for now) only used in synchronous retrievals.
 156   (url-retrieve-internal url callback (cons nil cbargs)))
 157
 158 (defun url-retrieve-internal (url callback cbargs)
 159   "Internal function; external interface is `url-retrieve'.
 160 CBARGS is what the callback will actually receive - the first item is
 161 the list of events, as described in the docstring of `url-retrieve'."
 162   (url-do-setup)
 163   (url-gc-dead-buffers)
 164   (if (stringp url)
 165        (set-text-properties 0 (length url) nil url))
 166   (if (not (vectorp url))
 167       (setq url (url-generic-parse-url url)))
 168   (if (not (functionp callback))
 169       (error "Must provide a callback function to url-retrieve"))
 170   (unless (url-type url)
 171     (error "Bad url: %s" (url-recreate-url url)))
 172   (let ((loader (url-scheme-get-property (url-type url) 'loader))
 173         (url-using-proxy (if (url-host url)
 174                              (url-find-proxy-for-url url (url-host url))))
 175         (buffer nil)
 176         (asynch (url-scheme-get-property (url-type url) 'asynchronous-p)))
 177     (if url-using-proxy
 178         (setq asynch t
 179               loader 'url-proxy))
 180     (if asynch
 181         (setq buffer (funcall loader url callback cbargs))
 182       (setq buffer (funcall loader url))
 183       (if buffer
 184           (with-current-buffer buffer
 185             (apply callback cbargs))))
 186     (if url-history-track
 187         (url-history-update-url url (current-time)))
 188     buffer))
 189
 190 ;;;###autoload
 191 (defun url-retrieve-synchronously (url)
 192   "Retrieve URL synchronously.
 193 Return the buffer containing the data, or nil if there are no data
 194 associated with it (the case for dired, info, or mailto URLs that need
 195 no further processing).  URL is either a string or a parsed URL."
 196   (url-do-setup)
 197
 198   (lexical-let ((retrieval-done nil)
 199                 (asynch-buffer nil))
 200     (setq asynch-buffer
 201           (url-retrieve url (lambda (&rest ignored)
 202                               (url-debug 'retrieval "Synchronous fetching done (%S)" (current-buffer))
 203                               (setq retrieval-done t
 204                                     asynch-buffer (current-buffer)))))
 205     (if (null asynch-buffer)
 206         ;; We do not need to do anything, it was a mailto or something
 207         ;; similar that takes processing completely outside of the URL
 208         ;; package.
 209         nil
 210       (let ((proc (get-buffer-process asynch-buffer)))
 211         ;; If the access method was synchronous, `retrieval-done' should
 212         ;; hopefully already be set to t.  If it is nil, and `proc' is also
 213         ;; nil, it implies that the async process is not running in
 214         ;; asynch-buffer.  This happens e.g. for FTP files.  In such a case
 215         ;; url-file.el should probably set something like a `url-process'
 216         ;; buffer-local variable so we can find the exact process that we
 217         ;; should be waiting for.  In the mean time, we'll just wait for any
 218         ;; process output.
 219         (while (not retrieval-done)
 220           (url-debug 'retrieval
 221                      "Spinning in url-retrieve-synchronously: %S (%S)"
 222                      retrieval-done asynch-buffer)
 223           (if (buffer-local-value 'url-redirect-buffer asynch-buffer)
 224               (setq proc (get-buffer-process
 225                           (setq asynch-buffer
 226                                 (buffer-local-value 'url-redirect-buffer
 227                                                     asynch-buffer))))
 228             (if (and proc (memq (process-status proc)
 229                                 '(closed exit signal failed))
 230                      ;; Make sure another process hasn't been started.
 231                      (eq proc (or (get-buffer-process asynch-buffer) proc)))
 232                 ;; FIXME: It's not clear whether url-retrieve's callback is
 233                 ;; guaranteed to be called or not.  It seems that url-http
 234                 ;; decides sometimes consciously not to call it, so it's not
 235                 ;; clear that it's a bug, but even then we need to decide how
 236                 ;; url-http can then warn us that the download has completed.
 237                 ;; In the mean time, we use this here workaround.
 238                 ;; XXX: The callback must always be called.  Any
 239                 ;; exception is a bug that should be fixed, not worked
 240                 ;; around.
 241                 (progn ;; Call delete-process so we run any sentinel now.
 242                   (delete-process proc)
 243                   (setq retrieval-done t)))
 244             ;; We used to use `sit-for' here, but in some cases it wouldn't
 245             ;; work because apparently pending keyboard input would always
 246             ;; interrupt it before it got a chance to handle process input.
 247             ;; `sleep-for' was tried but it lead to other forms of
 248             ;; hanging.  --Stef
 249             (unless (or (with-local-quit
 250                           (accept-process-output proc))
 251                         (null proc))
 252               ;; accept-process-output returned nil, maybe because the process
 253               ;; exited (and may have been replaced with another).  If we got
 254               ;; a quit, just stop.
 255               (when quit-flag
 256                 (delete-process proc))
 257               (setq proc (and (not quit-flag)
 258                               (get-buffer-process asynch-buffer)))))))
 259       asynch-buffer)))
 260
 261 (defun url-mm-callback (&rest ignored)
 262   (let ((handle (mm-dissect-buffer t)))
 263     (url-mark-buffer-as-dead (current-buffer))
 264     (with-current-buffer
 265         (generate-new-buffer (url-recreate-url url-current-object))
 266       (if (eq (mm-display-part handle) 'external)
 267           (progn
 268             (set-process-sentinel
 269              ;; Fixme: this shouldn't have to know the form of the
 270              ;; undisplayer produced by `mm-display-part'.
 271              (get-buffer-process (cdr (mm-handle-undisplayer handle)))
 272              `(lambda (proc event)
 273                 (mm-destroy-parts (quote ,handle))))
 274             (message "Viewing externally")
 275             (kill-buffer (current-buffer)))
 276         (display-buffer (current-buffer))
 277         (add-hook 'kill-buffer-hook
 278                   `(lambda () (mm-destroy-parts ',handle))
 279                   nil
 280                   t)))))
 281
 282 (defun url-mm-url (url)
 283   "Retrieve URL and pass to the appropriate viewing application."
 284   ;; These requires could advantageously be moved to url-mm-callback or
 285   ;; turned into autoloads, but I suspect that it would introduce some bugs
 286   ;; because loading those files from a process sentinel or filter may
 287   ;; result in some undesirable carner cases.
 288   (require 'mm-decode)
 289   (require 'mm-view)
 290   (url-retrieve url 'url-mm-callback nil))
 291
 292 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 293 ;;; Miscellaneous
 294 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 295 (defvar url-dead-buffer-list nil)
 296
 297 (defun url-mark-buffer-as-dead (buff)
 298   (push buff url-dead-buffer-list))
 299
 300 (defun url-gc-dead-buffers ()
 301   (let ((buff))
 302     (while (setq buff (pop url-dead-buffer-list))
 303       (if (buffer-live-p buff)
 304           (kill-buffer buff)))))
 305
 306 (cond
 307  ((fboundp 'display-warning)
 308   (defalias 'url-warn 'display-warning))
 309  ((fboundp 'warn)
 310   (defun url-warn (class message &optional level)
 311     (warn "(%s/%s) %s" class (or level 'warning) message)))
 312  (t
 313   (defun url-warn (class message &optional level)
 314     (with-current-buffer (get-buffer-create "*URL-WARNINGS*")
 315       (goto-char (point-max))
 316       (save-excursion
 317         (insert (format "(%s/%s) %s\n" class (or level 'warning) message)))
 318       (display-buffer (current-buffer))))))
 319
 320 (provide 'url)
 321
 322 ;; arch-tag: bc182f1f-d187-4f10-b961-47af2066579a
 323 ;;; url.el ends here