lisp/url/url-parse.el

   1 ;;; url-parse.el --- Uniform Resource Locator parser
   2 ;; Author: $Author: miles $
   3 ;; Created: $Date: 2004/04/04 04:44:10 $
   4 ;; Version: $Revision: 1.3 $
   5 ;; Keywords: comm, data, processes
   6
   7 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
   8 ;;; Copyright (c) 1993 - 1996, 2004 by William M. Perry <wmperry@cs.indiana.edu>
   9 ;;; Copyright (c) 1996 - 1999 Free Software Foundation, Inc.
  10 ;;;
  11 ;;; This file is part of GNU Emacs.
  12 ;;;
  13 ;;; GNU Emacs is free software; you can redistribute it and/or modify
  14 ;;; it under the terms of the GNU General Public License as published by
  15 ;;; the Free Software Foundation; either version 2, or (at your option)
  16 ;;; any later version.
  17 ;;;
  18 ;;; GNU Emacs is distributed in the hope that it will be useful,
  19 ;;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  20 ;;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  21 ;;; GNU General Public License for more details.
  22 ;;;
  23 ;;; You should have received a copy of the GNU General Public License
  24 ;;; along with GNU Emacs; see the file COPYING.  If not, write to the
  25 ;;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  26 ;;; Boston, MA 02111-1307, USA.
  27 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
  28 (require 'url-vars)
  29
  30 (autoload 'url-scheme-get-property "url-methods")
  31
  32 (defmacro url-type (urlobj)
  33   `(aref ,urlobj 0))
  34
  35 (defmacro url-user (urlobj)
  36   `(aref ,urlobj 1))
  37
  38 (defmacro url-password (urlobj)
  39   `(aref ,urlobj 2))
  40
  41 (defmacro url-host (urlobj)
  42   `(aref ,urlobj 3))
  43
  44 (defmacro url-port (urlobj)
  45   `(or (aref ,urlobj 4)
  46       (if (url-fullness ,urlobj)
  47           (url-scheme-get-property (url-type ,urlobj) 'default-port))))
  48
  49 (defmacro url-filename (urlobj)
  50   `(aref ,urlobj 5))
  51
  52 (defmacro url-target (urlobj)
  53   `(aref ,urlobj 6))
  54
  55 (defmacro url-attributes (urlobj)
  56   `(aref ,urlobj 7))
  57
  58 (defmacro url-fullness (urlobj)
  59   `(aref ,urlobj 8))
  60
  61 (defmacro url-set-type (urlobj type)
  62   `(aset ,urlobj 0 ,type))
  63
  64 (defmacro url-set-user (urlobj user)
  65   `(aset ,urlobj 1 ,user))
  66
  67 (defmacro url-set-password (urlobj pass)
  68   `(aset ,urlobj 2 ,pass))
  69
  70 (defmacro url-set-host (urlobj host)
  71   `(aset ,urlobj 3 ,host))
  72
  73 (defmacro url-set-port (urlobj port)
  74   `(aset ,urlobj 4 ,port))
  75
  76 (defmacro url-set-filename (urlobj file)
  77   `(aset ,urlobj 5 ,file))
  78
  79 (defmacro url-set-target (urlobj targ)
  80   `(aset ,urlobj 6 ,targ))
  81
  82 (defmacro url-set-attributes (urlobj targ)
  83   `(aset ,urlobj 7 ,targ))
  84
  85 (defmacro url-set-full (urlobj val)
  86   `(aset ,urlobj 8 ,val))
  87
  88 ;;;###autoload
  89 (defun url-recreate-url (urlobj)
  90   (concat (url-type urlobj) ":" (if (url-host urlobj) "//" "")
  91           (if (url-user urlobj)
  92               (concat (url-user urlobj)
  93                       (if (url-password urlobj)
  94                           (concat ":" (url-password urlobj)))
  95                       "@"))
  96           (url-host urlobj)
  97           (if (and (url-port urlobj)
  98                    (not (equal (url-port urlobj)
  99                                (url-scheme-get-property (url-type urlobj) 'default-port))))
 100               (format ":%d" (url-port urlobj)))
 101           (or (url-filename urlobj) "/")
 102           (if (url-target urlobj)
 103               (concat "#" (url-target urlobj)))
 104           (if (url-attributes urlobj)
 105               (concat ";"
 106                       (mapconcat
 107                        (function
 108                         (lambda (x)
 109                           (if (cdr x)
 110                               (concat (car x) "=" (cdr x))
 111                             (car x)))) (url-attributes urlobj) ";")))))
 112
 113 ;;;###autoload
 114 (defun url-generic-parse-url (url)
 115   "Return a vector of the parts of URL.
 116 Format is:
 117 \[proto username password hostname portnumber file reference attributes fullp\]"
 118   (cond
 119    ((null url)
 120     (make-vector 9 nil))
 121    ((or (not (string-match url-nonrelative-link url))
 122         (= ?/ (string-to-char url)))
 123     (let ((retval (make-vector 9 nil)))
 124       (url-set-filename retval url)
 125       (url-set-full retval nil)
 126       retval))
 127    (t
 128     (save-excursion
 129       (set-buffer (get-buffer-create " *urlparse*"))
 130       (set-syntax-table url-parse-syntax-table)
 131       (let ((save-pos nil)
 132             (prot nil)
 133             (user nil)
 134             (pass nil)
 135             (host nil)
 136             (port nil)
 137             (file nil)
 138             (refs nil)
 139             (attr nil)
 140             (full nil)
 141             (inhibit-read-only t))
 142         (erase-buffer)
 143         (insert url)
 144         (goto-char (point-min))
 145         (setq save-pos (point))
 146         (if (not (looking-at "//"))
 147             (progn
 148               (skip-chars-forward "a-zA-Z+.\\-")
 149               (downcase-region save-pos (point))
 150               (setq prot (buffer-substring save-pos (point)))
 151               (skip-chars-forward ":")
 152               (setq save-pos (point))))
 153
 154         ;; We are doing a fully specified URL, with hostname and all
 155         (if (looking-at "//")
 156             (progn
 157               (setq full t)
 158               (forward-char 2)
 159               (setq save-pos (point))
 160               (skip-chars-forward "^/")
 161               (setq host (buffer-substring save-pos (point)))
 162               (if (string-match "^\\([^@]+\\)@" host)
 163                   (setq user (match-string 1 host)
 164                         host (substring host (match-end 0) nil)))
 165               (if (and user (string-match "\\([^:]+\\):\\(.*\\)" user))
 166                   (setq pass (match-string 2 user)
 167                         user (match-string 1 user)))
 168               (if (string-match ":\\([0-9+]+\\)" host)
 169                   (setq port (string-to-int (match-string 1 host))
 170                         host (substring host 0 (match-beginning 0))))
 171               (if (string-match ":$" host)
 172                   (setq host (substring host 0 (match-beginning 0))))
 173               (setq host (downcase host)
 174                     save-pos (point))))
 175
 176         (if (not port)
 177             (setq port (url-scheme-get-property prot 'default-port)))
 178
 179         ;; Gross hack to preserve ';' in data URLs
 180
 181         (setq save-pos (point))
 182
 183         (if (string= "data" prot)
 184             (goto-char (point-max))
 185           ;; Now check for references
 186           (skip-chars-forward "^#")
 187           (if (eobp)
 188               nil
 189             (delete-region
 190              (point)
 191              (progn
 192                (skip-chars-forward "#")
 193                (setq refs (buffer-substring (point) (point-max)))
 194                (point-max))))
 195           (goto-char save-pos)
 196           (skip-chars-forward "^;")
 197           (if (not (eobp))
 198               (setq attr (url-parse-args (buffer-substring (point) (point-max)) t)
 199                     attr (nreverse attr))))
 200
 201         (setq file (buffer-substring save-pos (point)))
 202         (if (and host (string-match "%[0-9][0-9]" host))
 203             (setq host (url-unhex-string host)))
 204         (vector prot user pass host port file refs attr full))))))
 205
 206 (provide 'url-parse)
 207
 208 ;;; arch-tag: f338325f-71ab-4bee-93cc-78fb9a03d403