rfc822.el 10.6 KB
Newer Older
1
;;; rfc822.el --- hairy RFC 822 (or later) parser for mail, news, etc.
Eric S. Raymond's avatar
Eric S. Raymond committed
2

Paul Eggert's avatar
Paul Eggert committed
3
;; Copyright (C) 1986-1987, 1990, 2001-2019 Free Software Foundation,
4
;; Inc.
Eric S. Raymond's avatar
Eric S. Raymond committed
5

Eric S. Raymond's avatar
Eric S. Raymond committed
6
;; Author: Richard Mlynarik <mly@eddie.mit.edu>
7
;; Maintainer: emacs-devel@gnu.org
Eric S. Raymond's avatar
Eric S. Raymond committed
8
;; Keywords: mail
Eric S. Raymond's avatar
Eric S. Raymond committed
9

Richard M. Stallman's avatar
Richard M. Stallman committed
10 11
;; This file is part of GNU Emacs.

12
;; GNU Emacs is free software: you can redistribute it and/or modify
Richard M. Stallman's avatar
Richard M. Stallman committed
13
;; it under the terms of the GNU General Public License as published by
14 15
;; the Free Software Foundation, either version 3 of the License, or
;; (at your option) any later version.
Richard M. Stallman's avatar
Richard M. Stallman committed
16 17 18 19 20 21 22

;; GNU Emacs is distributed in the hope that it will be useful,
;; but WITHOUT ANY WARRANTY; without even the implied warranty of
;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
;; GNU General Public License for more details.

;; You should have received a copy of the GNU General Public License
23
;; along with GNU Emacs.  If not, see <https://www.gnu.org/licenses/>.
Richard M. Stallman's avatar
Richard M. Stallman committed
24

25 26
;;; Commentary:

27 28
;; Support functions for parsing headers used by mail and news modes.
;; The header format is specified by Internet RFC 822 and its successors.
29

Eric S. Raymond's avatar
Eric S. Raymond committed
30 31
;;; Code:

32 33 34
(defvar rfc822-address-start)

;; uses rfc822-address-start free, throws to address
Richard M. Stallman's avatar
Richard M. Stallman committed
35 36 37
(defun rfc822-bad-address (reason)
  (save-restriction
    (insert "_^_")
38
    (narrow-to-region rfc822-address-start
Richard M. Stallman's avatar
Richard M. Stallman committed
39 40 41 42 43 44 45 46 47 48 49 50 51 52
		      (if (re-search-forward "[,;]" nil t)
			  (max (point-min) (1- (point)))
			(point-max)))
    ;; make the error string be suitable for inclusion in (...)
    (let ((losers '("\\" "(" ")" "\n")))
      (while losers
	(goto-char (point-min))
	(while (search-forward (car losers) nil t)
	  (backward-char 1)
	  (insert ?\\)
	  (forward-char 1))
	(setq losers (cdr losers))))
    (goto-char (point-min)) (insert "(Unparsable address -- "
				    reason
53
				    ": \"")
Richard M. Stallman's avatar
Richard M. Stallman committed
54 55
    (goto-char (point-max)) (insert "\")"))
  (rfc822-nuke-whitespace)
56
  (throw 'address (buffer-substring rfc822-address-start (point))))
Richard M. Stallman's avatar
Richard M. Stallman committed
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78

(defun rfc822-nuke-whitespace (&optional leave-space)
  (let (ch)
    (while (cond ((eobp)
		  nil)
		 ((= (setq ch (following-char)) ?\()
		  (forward-char 1)
		  (while (if (eobp)
			     (rfc822-bad-address "Unbalanced comment (...)")
			   (/= (setq ch (following-char)) ?\)))
		    (cond ((looking-at "[^()\\]+")
			   (replace-match ""))
			  ((= ch ?\()
			   (rfc822-nuke-whitespace))
			  ((< (point) (1- (point-max)))
			   (delete-char 2))
			  (t
			   (rfc822-bad-address "orphaned backslash"))))
		  ;; delete remaining "()"
		  (forward-char -1)
		  (delete-char 2)
		  t)
79
		 ((memq ch '(?\  ?\t ?\n))
Richard M. Stallman's avatar
Richard M. Stallman committed
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102
		  (delete-region (point)
				 (progn (skip-chars-forward " \t\n") (point)))
		  t)
		 (t
		  nil)))
    (or (not leave-space)
	(eobp)
	(bobp)
	(= (preceding-char) ?\ )
	(insert ?\ ))))

(defun rfc822-looking-at (regex &optional leave-space)
  (if (cond ((stringp regex)
	     (if (looking-at regex)
		 (progn (goto-char (match-end 0))
			t)))
	    (t
	     (if (and (not (eobp))
		      (= (following-char) regex))
		 (progn (forward-char 1)
			t))))
      (let ((tem (match-data)))
	(rfc822-nuke-whitespace leave-space)
103
	(set-match-data tem)
Richard M. Stallman's avatar
Richard M. Stallman committed
104 105 106 107 108 109 110 111
	t)))

(defun rfc822-snarf-word ()
  ;; word is atom | quoted-string
  (cond ((= (following-char) ?\")
	 ;; quoted-string
	 (or (rfc822-looking-at "\"\\([^\"\\\n]\\|\\\\.\\|\\\\\n\\)*\"")
	     (rfc822-bad-address "Unterminated quoted string")))
112
	((rfc822-looking-at "[^][\000-\037 ()<>@,;:\\\".]+")
Richard M. Stallman's avatar
Richard M. Stallman committed
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
	 ;; atom
	 )
	(t
	 (rfc822-bad-address "Rubbish in address"))))

(defun rfc822-snarf-words ()
  (rfc822-snarf-word)
  (while (rfc822-looking-at ?.)
    (rfc822-snarf-word)))

(defun rfc822-snarf-subdomain ()
  ;; sub-domain is domain-ref | domain-literal
  (cond ((= (following-char) ?\[)
	 ;; domain-ref
	 (or (rfc822-looking-at "\\[\\([^][\\\n]\\|\\\\.\\|\\\\\n\\)*\\]")
	     (rfc822-bad-address "Unterminated domain literal [...]")))
129
	((rfc822-looking-at "[^][\000-\037 ()<>@,;:\\\".]+")
Richard M. Stallman's avatar
Richard M. Stallman committed
130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182
	 ;; domain-literal = atom
	 )
	(t
	 (rfc822-bad-address "Rubbish in host/domain specification"))))

(defun rfc822-snarf-domain ()
  (rfc822-snarf-subdomain)
  (while (rfc822-looking-at ?.)
    (rfc822-snarf-subdomain)))

(defun rfc822-snarf-frob-list (name separator terminator snarfer
				    &optional return)
  (let ((first t)
	(list ())
	tem)
    (while (cond ((eobp)
		  (rfc822-bad-address
		    (format "End of addresses in middle of %s" name)))
		 ((rfc822-looking-at terminator)
		  nil)
		 ((rfc822-looking-at separator)
		  ;; multiple separators are allowed and do nothing.
		  (while (rfc822-looking-at separator))
		  t)
		 (first
		  t)
		 (t
		  (rfc822-bad-address
		    (format "Gubbish in middle of %s" name))))
      (setq tem (funcall snarfer)
	    first nil)
      (and return tem
	   (setq list (if (listp tem)
			  (nconc (reverse tem) list)
			  (cons tem list)))))
    (nreverse list)))

;; return either an address (a string) or a list of addresses
(defun rfc822-addresses-1 (&optional allow-groups)
  ;; Looking for an rfc822 `address'
  ;; Either a group (1*word ":" [#mailbox] ";")
  ;; or a mailbox (addr-spec | 1*word route-addr)
  ;;  addr-spec is (local-part "@" domain)
  ;;  route-addr is ("<" [1#("@" domain) ":"] addr-spec ">")
  ;;  local-part is (word *("." word))
  ;;  word is (atom | quoted-string)
  ;;  quoted-string is ("\([^\"\\n]\|\\.\|\\\n\)")
  ;;  atom is [^\000-\037\177 ()<>@,;:\".[]]+
  ;;  domain is sub-domain *("." sub-domain)
  ;;  sub-domain is domain-ref | domain-literal
  ;;  domain-literal is  "[" *(dtext | quoted-pair) "]"
  ;;  dtext is "[^][\\n"
  ;;  domain-ref is atom
183
  (let ((rfc822-address-start (point))
Richard M. Stallman's avatar
Richard M. Stallman committed
184 185 186 187 188 189 190 191 192 193
	(n 0))
    (catch 'address
      ;; optimize common cases:
      ;;  foo
      ;;  foo.bar@bar.zap
      ;; followed by "\\'\\|,\\|([^()\\]*)\\'"
      ;; other common cases are:
      ;;  foo bar <foo.bar@baz.zap>
      ;;  "foo bar" <foo.bar@baz.zap>
      ;;  those aren't hacked yet.
194
      (if (and (rfc822-looking-at "[^][\000-\037 ()<>@,;:\\\"]+\\(\\|@[^][\000-\037 ()<>@,;:\\\"]+\\)" t)
Richard M. Stallman's avatar
Richard M. Stallman committed
195 196 197 198 199 200 201
	       (progn (or (eobp)
			  (rfc822-looking-at ?,))))
	  (progn
	    ;; rfc822-looking-at may have inserted a space
	    (or (bobp) (/= (preceding-char) ?\ ) (delete-char -1))
	    ;; relying on the fact that rfc822-looking-at <char>
	    ;;  doesn't mung match-data
202 203
	    (throw 'address (buffer-substring rfc822-address-start (match-end 0)))))
      (goto-char rfc822-address-start)
Richard M. Stallman's avatar
Richard M. Stallman committed
204 205 206 207 208
      (while t
	(cond ((and (= n 1) (rfc822-looking-at ?@))
	       ;; local-part@domain
	       (rfc822-snarf-domain)
	       (throw 'address
209
		 (buffer-substring rfc822-address-start (point))))
Richard M. Stallman's avatar
Richard M. Stallman committed
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247
	      ((rfc822-looking-at ?:)
	       (cond ((not allow-groups)
		      (rfc822-bad-address "A group name may not appear here"))
		     ((= n 0)
		      (rfc822-bad-address "No name for :...; group")))
	       ;; group
	       (throw 'address
		 ;; return a list of addresses
		 (rfc822-snarf-frob-list ":...; group" ?\, ?\;
					 'rfc822-addresses-1 t)))
	      ((rfc822-looking-at ?<)
	       (let ((start (point))
		     (strip t))
		 (cond ((rfc822-looking-at ?>)
			;; empty path
			())
		       ((and (not (eobp)) (= (following-char) ?\@))
			;; <@foo.bar,@baz:quux@abcd.efg>
			(rfc822-snarf-frob-list "<...> address" ?\, ?\:
			  (function (lambda ()
				      (if (rfc822-looking-at ?\@)
					  (rfc822-snarf-domain)
					(rfc822-bad-address
					  "Gubbish in route-addr")))))
			(rfc822-snarf-words)
			(or (rfc822-looking-at ?@)
			    (rfc822-bad-address "Malformed <..@..> address"))
			(rfc822-snarf-domain)
			(setq strip nil))
		       ((progn (rfc822-snarf-words) (rfc822-looking-at ?@))
			; allow <foo> (losing unix seems to do this)
			(rfc822-snarf-domain)))
		 (let ((end (point)))
		   (if (rfc822-looking-at ?\>)
		       (throw 'address
			 (buffer-substring (if strip start (1- start))
					   (if strip end (1+ end))))
		     (rfc822-bad-address "Unterminated <...> address")))))
248
	      ((looking-at "[^][\000-\037 ()<>@,;:\\.]")
Richard M. Stallman's avatar
Richard M. Stallman committed
249 250 251 252
	       ;; this allows "." to be part of the words preceding
	       ;; an addr-spec, since many broken mailers output
	       ;; "Hern K. Herklemeyer III
	       ;;   <yank@megadeath.dod.gods-own-country>"
root's avatar
root committed
253 254 255 256
               (let ((again t))
                 (while again
                   (or (= n 0) (bobp) (= (preceding-char) ?\ )
                       (insert ?\ ))
Jim Blandy's avatar
Jim Blandy committed
257
                   (rfc822-snarf-words)
root's avatar
root committed
258 259
                   (setq n (1+ n))
                   (setq again (or (rfc822-looking-at ?.)
260
                                   (looking-at "[^][\000-\037 ()<>@,;:\\.]"))))))
Richard M. Stallman's avatar
Richard M. Stallman committed
261 262 263 264
	      ((= n 0)
	       (throw 'address nil))
	      ((= n 1) ; allow "foo" (losing unix seems to do this)
	       (throw 'address
265
		 (buffer-substring rfc822-address-start (point))))
root's avatar
root committed
266 267 268
              ((> n 1)
               (rfc822-bad-address "Missing comma between addresses or badly-formatted address"))
	      ((or (eobp) (= (following-char) ?,))
Richard M. Stallman's avatar
Richard M. Stallman committed
269 270 271 272
	       (rfc822-bad-address "Missing comma or route-spec"))
	      (t
	       (rfc822-bad-address "Strange character or missing comma")))))))

273

Richard M. Stallman's avatar
Richard M. Stallman committed
274
(defun rfc822-addresses (header-text)
275
  (if (string-match "\\`[ \t]*\\([^][\000-\037 ()<>@,;:\\\".]+\\)[ \t]*\\'"
Richard M. Stallman's avatar
Richard M. Stallman committed
276 277 278 279 280
                    header-text)
      ;; Make very simple case moderately fast.
      (list (substring header-text (match-beginning 1) (match-end 1)))
    (let ((buf (generate-new-buffer " rfc822")))
      (unwind-protect
281 282 283 284 285 286
          (with-current-buffer buf
            (make-local-variable 'case-fold-search)
            (setq case-fold-search nil)	;For speed(?)
            (insert header-text)
            ;; unfold continuation lines
            (goto-char (point-min))
Richard M. Stallman's avatar
Richard M. Stallman committed
287

288 289 290
            (while (re-search-forward "\\([^\\]\\(\\\\\\\\\\)*\\)\n[ \t]"
                                      nil t)
              (replace-match "\\1 " t))
Richard M. Stallman's avatar
Richard M. Stallman committed
291

292
            (goto-char (point-min))
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
	    ;; Give `rfc822-address-start' a non-nil initial value to
	    ;; prevent `rfc822-bad-address' from raising a
	    ;; `wrong-type-argument' error.
            (let* ((rfc822-address-start (point))
		   list tem
		   (err
		    (catch 'address
		      ;; Note that `rfc822-nuke-whitespace' and
		      ;; `rfc822-looking-at' can throw.
		      (rfc822-nuke-whitespace)
		      (while (not (eobp))
			(setq rfc822-address-start (point))
			(setq tem
			      (cond ((rfc822-looking-at ?\,)
				     nil)
				    ((looking-at "[][\000-\037@;:\\.>)]")
				     (forward-char)
				     (catch 'address ; For rfc822-bad-address
				       (rfc822-bad-address
					(format "Strange character \\%c found"
						(preceding-char)))))
				    (t
				     (rfc822-addresses-1 t))))
			(cond ((null tem))
			      ((stringp tem)
			       (setq list (cons tem list)))
			      (t
			       (setq list (nconc (nreverse tem) list)))))
		      nil)))
	      (nreverse (append (if err (list err)) list))))
323
	(and buf (kill-buffer buf))))))
Richard M. Stallman's avatar
Richard M. Stallman committed
324

Jim Blandy's avatar
Jim Blandy committed
325 326
(provide 'rfc822)

Eric S. Raymond's avatar
Eric S. Raymond committed
327
;;; rfc822.el ends here