lisp/url/url-parse.el

   1 ;;; url-parse.el --- Uniform Resource Locator parser
   2
   3 ;; Copyright (C) 1996, 1997, 1998, 1999, 2004,
   4 ;;   2005, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
   5
   6 ;; Keywords: comm, data, processes
   7
   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.
  14
  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.
  19
  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/>.
  22
  23 ;;; Commentary:
  24
  25 ;;; Code:
  26
  27 (require 'url-vars)
  28 (require 'auth-source)
  29 (eval-when-compile (require 'cl))
  30
  31 (autoload 'url-scheme-get-property "url-methods")
  32
  33 (defstruct (url
  34             (:constructor nil)
  35             (:constructor url-parse-make-urlobj
  36                           (&optional type user password host portspec filename
  37                                      target attributes fullness))
  38             (:copier nil))
  39   type user password host portspec filename target attributes fullness)
  40
  41 (defsubst url-port (urlobj)
  42   (or (url-portspec urlobj)
  43       (if (url-fullness urlobj)
  44           (url-scheme-get-property (url-type urlobj) 'default-port))))
  45
  46 (defsetf url-port (urlobj) (port) `(setf (url-portspec ,urlobj) ,port))
  47
  48 ;;;###autoload
  49 (defun url-recreate-url (urlobj)
  50   "Recreate a URL string from the parsed URLOBJ."
  51   (concat (url-type urlobj) ":" (if (url-host urlobj) "//" "")
  52           (if (url-user urlobj)
  53               (concat (url-user urlobj)
  54                       (if (url-password urlobj)
  55                           (concat ":" (url-password urlobj)))
  56                       "@"))
  57           (url-host urlobj)
  58           (if (and (url-port urlobj)
  59                    (not (equal (url-port urlobj)
  60                                (url-scheme-get-property (url-type urlobj) 'default-port))))
  61               (format ":%d" (url-port urlobj)))
  62           (or (url-filename urlobj) "/")
  63           (url-recreate-url-attributes urlobj)
  64           (if (url-target urlobj)
  65               (concat "#" (url-target urlobj)))))
  66
  67 (defun url-recreate-url-attributes (urlobj)
  68   "Recreate the attributes of an URL string from the parsed URLOBJ."
  69   (when (url-attributes urlobj)
  70     (concat ";"
  71             (mapconcat (lambda (x)
  72                          (if (cdr x)
  73                              (concat (car x) "=" (cdr x))
  74                            (car x)))
  75                        (url-attributes urlobj) ";"))))
  76
  77 ;;;###autoload
  78 (defun url-generic-parse-url (url)
  79   "Return an URL-struct of the parts of URL.
  80 The CL-style struct contains the following fields:
  81 TYPE USER PASSWORD HOST PORTSPEC FILENAME TARGET ATTRIBUTES FULLNESS."
  82   ;; See RFC 3986.
  83   (cond
  84    ((null url)
  85     (url-parse-make-urlobj))
  86    ((or (not (string-match url-nonrelative-link url))
  87         (= ?/ (string-to-char url)))
  88     ;; This isn't correct, as a relative URL can be a fragment link
  89     ;; (e.g. "#foo") and many other things (see section 4.2).
  90     ;; However, let's not fix something that isn't broken, especially
  91     ;; when close to a release.
  92     (url-parse-make-urlobj nil nil nil nil nil url))
  93    (t
  94     (with-temp-buffer
  95       ;; Don't let those temp-buffer modifications accidentally
  96       ;; deactivate the mark of the current-buffer.
  97       (let ((deactivate-mark nil))
  98         (set-syntax-table url-parse-syntax-table)
  99         (let ((save-pos nil)
 100               (prot nil)
 101               (user nil)
 102               (pass nil)
 103               (host nil)
 104               (port nil)
 105               (file nil)
 106               (refs nil)
 107               (attr nil)
 108               (full nil)
 109               (inhibit-read-only t))
 110           (erase-buffer)
 111           (insert url)
 112           (goto-char (point-min))
 113           (setq save-pos (point))
 114
 115           ;; 3.1. Scheme
 116           (unless (looking-at "//")
 117             (skip-chars-forward "a-zA-Z+.\\-")
 118             (downcase-region save-pos (point))
 119             (setq prot (buffer-substring save-pos (point)))
 120             (skip-chars-forward ":")
 121             (setq save-pos (point)))
 122
 123           ;; 3.2. Authority
 124           (when (looking-at "//")
 125             (setq full t)
 126             (forward-char 2)
 127             (setq save-pos (point))
 128             (skip-chars-forward "^/")
 129             (setq host (buffer-substring save-pos (point)))
 130             (if (string-match "^\\([^@]+\\)@" host)
 131                 (setq user (match-string 1 host)
 132                       host (substring host (match-end 0) nil)))
 133             (if (and user (string-match "\\([^:]+\\):\\(.*\\)" user))
 134                 (setq pass (match-string 2 user)
 135                       user (match-string 1 user)))
 136             ;; This gives wrong results for IPv6 literal addresses.
 137             (if (string-match ":\\([0-9+]+\\)" host)
 138                 (setq port (string-to-number (match-string 1 host))
 139                       host (substring host 0 (match-beginning 0))))
 140             (if (string-match ":$" host)
 141                 (setq host (substring host 0 (match-beginning 0))))
 142             (setq host (downcase host)
 143                   save-pos (point)))
 144
 145           (if (not port)
 146               (setq port (url-scheme-get-property prot 'default-port)))
 147
 148           ;; 3.3. Path
 149           ;; Gross hack to preserve ';' in data URLs
 150           (setq save-pos (point))
 151
 152           ;; 3.4. Query
 153           (if (string= "data" prot)
 154               (goto-char (point-max))
 155             ;; Now check for references
 156             (skip-chars-forward "^#")
 157             (if (eobp)
 158                 nil
 159               (delete-region
 160                (point)
 161                (progn
 162                  (skip-chars-forward "#")
 163                  (setq refs (buffer-substring (point) (point-max)))
 164                  (point-max))))
 165             (goto-char save-pos)
 166             (skip-chars-forward "^;")
 167             (unless (eobp)
 168               (setq attr (url-parse-args (buffer-substring (point) (point-max))
 169                                          t)
 170                     attr (nreverse attr))))
 171
 172           (setq file (buffer-substring save-pos (point)))
 173           (if (and host (string-match "%[0-9][0-9]" host))
 174               (setq host (url-unhex-string host)))
 175           (url-parse-make-urlobj
 176            prot user pass host port file refs attr full)))))))
 177
 178 (defmacro url-bit-for-url (method lookfor url)
 179   `(let* ((urlobj (url-generic-parse-url url))
 180           (bit (funcall ,method urlobj))
 181           (methods (list 'url-recreate-url
 182                          'url-host)))
 183      (while (and (not bit) (> (length methods) 0))
 184        (setq bit
 185              (auth-source-user-or-password
 186               ,lookfor (funcall (pop methods) urlobj) (url-type urlobj))))
 187      bit))
 188
 189 (defun url-user-for-url (url)
 190   "Attempt to use .authinfo to find a user for this URL."
 191   (url-bit-for-url 'url-user "login" url))
 192
 193 (defun url-password-for-url (url)
 194   "Attempt to use .authinfo to find a password for this URL."
 195   (url-bit-for-url 'url-password "password" url))
 196
 197 (provide 'url-parse)
 198
 199 ;; arch-tag: f338325f-71ab-4bee-93cc-78fb9a03d403
 200 ;;; url-parse.el ends here