1;;; mail-extr.el --- extract full name and address from RFC 822 mail header -*- coding: utf-8 -*- 2 3;; Copyright (C) 1991, 1992, 1993, 1994, 1997, 2001, 2002, 2003, 2004, 4;; 2005, 2006, 2007 Free Software Foundation, Inc. 5 6;; Author: Joe Wells <jbw@cs.bu.edu> 7;; Maintainer: FSF 8;; Keywords: mail 9 10;; This file is part of GNU Emacs. 11 12;; GNU Emacs is free software; you can redistribute it and/or modify 13;; it under the terms of the GNU General Public License as published by 14;; the Free Software Foundation; either version 2, or (at your option) 15;; any later version. 16 17;; GNU Emacs is distributed in the hope that it will be useful, 18;; but WITHOUT ANY WARRANTY; without even the implied warranty of 19;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 20;; GNU General Public License for more details. 21 22;; You should have received a copy of the GNU General Public License 23;; along with GNU Emacs; see the file COPYING. If not, write to the 24;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, 25;; Boston, MA 02110-1301, USA. 26 27;;; Commentary: 28 29;; The entry point of this code is 30;; 31;; mail-extract-address-components: (address &optional all) 32;; 33;; Given an RFC-822 ADDRESS, extract full name and canonical address. 34;; Returns a list of the form (FULL-NAME CANONICAL-ADDRESS). 35;; If no name can be extracted, FULL-NAME will be nil. 36;; ADDRESS may be a string or a buffer. If it is a buffer, the visible 37;; (narrowed) portion of the buffer will be interpreted as the address. 38;; (This feature exists so that the clever caller might be able to avoid 39;; consing a string.) 40;; If ADDRESS contains more than one RFC-822 address, only the first is 41;; returned. 42;; 43;; If ALL is non-nil, that means return info about all the addresses 44;; that are found in ADDRESS. The value is a list of elements of 45;; the form (FULL-NAME CANONICAL-ADDRESS), one per address. 46;; 47;; This code is more correct (and more heuristic) parser than the code in 48;; rfc822.el. And despite its size, it's fairly fast. 49;; 50;; There are two main benefits: 51;; 52;; 1. Higher probability of getting the correct full name for a human than 53;; any other package we know of. (On the other hand, it will cheerfully 54;; mangle non-human names/comments.) 55;; 2. Address part is put in a canonical form. 56;; 57;; The interface is not yet carved in stone; please give us suggestions. 58;; 59;; We have an extensive test-case collection of funny addresses if you want to 60;; work with the code. Developing this code requires frequent testing to 61;; make sure you're not breaking functionality. The test cases aren't included 62;; because they are over 100K. 63;; 64;; If you find an address that mail-extr fails on, please send it to the 65;; maintainer along with what you think the correct results should be. We do 66;; not consider it a bug if mail-extr mangles a comment that does not 67;; correspond to a real human full name, although we would prefer that 68;; mail-extr would return the comment as-is. 69;; 70;; Features: 71;; 72;; * Full name handling: 73;; 74;; * knows where full names can be found in an address. 75;; * avoids using empty comments and quoted text. 76;; * extracts full names from mailbox names. 77;; * recognizes common formats for comments after a full name. 78;; * puts a period and a space after each initial. 79;; * understands & referring to the mailbox name, capitalized. 80;; * strips name prefixes like "Prof.", etc. 81;; * understands what characters can occur in names (not just letters). 82;; * figures out middle initial from mailbox name. 83;; * removes funny nicknames. 84;; * keeps suffixes such as Jr., Sr., III, etc. 85;; * reorders "Last, First" type names. 86;; 87;; * Address handling: 88;; 89;; * parses rfc822 quoted text, comments, and domain literals. 90;; * parses rfc822 multi-line headers. 91;; * does something reasonable with rfc822 GROUP addresses. 92;; * handles many rfc822 noncompliant and garbage addresses. 93;; * canonicalizes addresses (after stripping comments/phrases outside <>). 94;; * converts ! addresses into .UUCP and %-style addresses. 95;; * converts rfc822 ROUTE addresses to %-style addresses. 96;; * truncates %-style addresses at leftmost fully qualified domain name. 97;; * handles local relative precedence of ! vs. % and @ (untested). 98;; 99;; It does almost no string creation. It primarily uses the built-in 100;; parsing routines with the appropriate syntax tables. This should 101;; result in greater speed. 102;; 103;; TODO: 104;; 105;; * handle all test cases. (This will take forever.) 106;; * software to pick the correct header to use (eg., "Senders-Name:"). 107;; * multiple addresses in the "From:" header (almost all of the necessary 108;; code is there). 109;; * flag to not treat `,' as an address separator. (This is useful when 110;; there is a "From:" header but no "Sender:" header, because then there 111;; is only allowed to be one address.) 112;; * mailbox name does not necessarily contain full name. 113;; * fixing capitalization when it's all upper or lowercase. (Hard!) 114;; * some of the domain literal handling is missing. (But I've never even 115;; seen one of these in a mail address, so maybe no big deal.) 116;; * arrange to have syntax tables byte-compiled. 117;; * speed hacks. 118;; * delete unused variables. 119;; * arrange for testing with different relative precedences of ! vs. @ 120;; and %. 121;; * insert documentation strings! 122;; * handle X.400-gatewayed addresses according to RFC 1148. 123 124;;; Change Log: 125;; 126;; Thu Feb 17 17:57:33 1994 Jamie Zawinski (jwz@lucid.com) 127;; 128;; * merged with jbw's latest version 129;; 130;; Wed Feb 9 21:56:27 1994 Jamie Zawinski (jwz@lucid.com) 131;; 132;; * high-bit chars in comments weren't treated as word syntax 133;; 134;; Sat Feb 5 03:13:40 1994 Jamie Zawinski (jwz@lucid.com) 135;; 136;; * call replace-match with fixed-case arg 137;; 138;; Thu Dec 16 21:56:45 1993 Jamie Zawinski (jwz@lucid.com) 139;; 140;; * some more cleanup, doc, added provide 141;; 142;; Tue Mar 23 21:23:18 1993 Joe Wells (jbw at csd.bu.edu) 143;; 144;; * Made mail-full-name-prefixes a user-customizable variable. 145;; Allow passing the address as a buffer as well as a string. 146;; Allow [ and ] as name characters (Finnish character set). 147;; 148;; Mon Mar 22 21:20:56 1993 Joe Wells (jbw at bigbird.bu.edu) 149;; 150;; * Handle "null" addresses. Handle = used for spacing in mailbox 151;; name. Fix bug in handling of ROUTE-ADDR-type addresses that are 152;; missing their brackets. Handle uppercase "JR". Extract full 153;; names from X.400 addresses encoded in RFC-822. Fix bug in 154;; handling of multiple addresses where first has trailing comment. 155;; Handle more kinds of telephone extension lead-ins. 156;; 157;; Mon Mar 22 20:16:57 1993 Joe Wells (jbw at bigbird.bu.edu) 158;; 159;; * Handle HZ encoding for embedding GB encoded chinese characters. 160;; 161;; Mon Mar 22 00:46:12 1993 Joe Wells (jbw at bigbird.bu.edu) 162;; 163;; * Fixed too broad matching of ham radio call signs. Fixed bug in 164;; handling an unmatched ' in a name string. Enhanced recognition 165;; of when . in the mailbox name terminates the name portion. 166;; Narrowed conversion of . to space to only the necessary 167;; situation. Deal with VMS's stupid date stamps. Handle a unique 168;; way of introducing an alternate address. Fixed spacing bug I 169;; introduced in switching last name order. Fixed bug in handling 170;; address with ! and % but no @. Narrowed the cases in which 171;; certain trailing words are discarded. 172;; 173;; Sun Mar 21 21:41:06 1993 Joe Wells (jbw at bigbird.bu.edu) 174;; 175;; * Fixed bugs in handling GROUP addresses. Certain words in the 176;; middle of a name no longer terminate it. Handle LISTSERV list 177;; names. Ignore comment field containing mailbox name. 178;; 179;; Sun Mar 21 14:39:38 1993 Joe Wells (jbw at bigbird.bu.edu) 180;; 181;; * Moved variant-method code back into main function. Handle 182;; underscores as spaces in comments. Handle leading nickname. Add 183;; flag to ignore single-word names. Other changes. 184;; 185;; Mon Feb 1 22:23:31 1993 Joe Wells (jbw at bigbird.bu.edu) 186;; 187;; * Added in changes by Rod Whitby and Jamie Zawinski. This 188;; includes the flag mail-extr-guess-middle-initial and the fix for 189;; handling multiple addresses correctly. (Whitby just changed 190;; a > to a <.) 191;; 192;; Mon Apr 6 23:59:09 1992 Joe Wells (jbw at bigbird.bu.edu) 193;; 194;; * Cleaned up some more. Release version 1.0 to world. 195;; 196;; Sun Apr 5 19:39:08 1992 Joe Wells (jbw at bigbird.bu.edu) 197;; 198;; * Cleaned up full name extraction extensively. 199;; 200;; Sun Feb 2 14:45:24 1992 Joe Wells (jbw at bigbird.bu.edu) 201;; 202;; * Total rewrite. Integrated mail-canonicalize-address into 203;; mail-extract-address-components. Now handles GROUP addresses more 204;; or less correctly. Better handling of lots of different cases. 205;; 206;; Fri Jun 14 19:39:50 1991 207;; * Created. 208 209;;; Code: 210 211 212(defgroup mail-extr nil 213 "Extract full name and address from RFC 822 mail header." 214 :prefix "mail-extr-" 215 :group 'mail) 216 217;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; 218;; 219;; User configuration variable definitions. 220;; 221 222(defcustom mail-extr-guess-middle-initial nil 223 "*Whether to try to guess middle initial from mail address. 224If true, then when we see an address like \"John Smith <jqs@host.com>\" 225we will assume that \"John Q. Smith\" is the fellow's name." 226 :type 'boolean 227 :group 'mail-extr) 228 229(defcustom mail-extr-ignore-single-names nil 230 "*Whether to ignore a name that is just a single word. 231If true, then when we see an address like \"Idiot <dumb@stupid.com>\" 232we will act as though we couldn't find a full name in the address." 233 :type 'boolean 234 :version "22.1" 235 :group 'mail-extr) 236 237(defcustom mail-extr-ignore-realname-equals-mailbox-name t 238"*Whether to ignore a name that is equal to the mailbox name. 239If true, then when the address is like \"Single <single@address.com>\" 240we will act as though we couldn't find a full name in the address." 241 :type 'boolean 242 :group 'mail-extr) 243 244;; Matches a leading title that is not part of the name (does not 245;; contribute to uniquely identifying the person). 246(defcustom mail-extr-full-name-prefixes 247 (purecopy 248 "\\(Prof\\|D[Rr]\\|Mrs?\\|Rev\\|Rabbi\\|SysOp\\|LCDR\\)\\.?[ \t\n]") 249 "*Matches prefixes to the full name that identify a person's position. 250These are stripped from the full name because they do not contribute to 251uniquely identifying the person." 252 :type 'regexp 253 :group 'mail-extr) 254 255(defcustom mail-extr-@-binds-tighter-than-! nil 256 "*Whether the local mail transport agent looks at ! before @." 257 :type 'boolean 258 :group 'mail-extr) 259 260(defcustom mail-extr-mangle-uucp nil 261 "*Whether to throw away information in UUCP addresses 262by translating things like \"foo!bar!baz@host\" into \"baz@bar.UUCP\"." 263 :type 'boolean 264 :group 'mail-extr) 265 266;;---------------------------------------------------------------------- 267;; what orderings are meaningful????? 268;;(defvar mail-operator-precedence-list '(?! ?% ?@)) 269;; Right operand of a % or a @ must be a domain name, period. No other 270;; operators allowed. Left operand of a @ is an address relative to that 271;; site. 272 273;; Left operand of a ! must be a domain name. Right operand is an 274;; arbitrary address. 275;;---------------------------------------------------------------------- 276 277 278 279;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; 280;; 281;; Constant definitions. 282;; 283 284;; Any character that can occur in a name, not counting characters that 285;; separate parts of a multipart name (hyphen and period). 286;; Yes, there are weird people with digits in their names. 287;; You will also notice the consideration for the 288;; Swedish/Finnish/Norwegian character set. 289(defconst mail-extr-all-letters-but-separators 290 (purecopy "][[:alnum:]{|}'~`")) 291 292;; Any character that can occur in a name in an RFC822 address including 293;; the separator (hyphen and possibly period) for multipart names. 294;; #### should . be in here? 295(defconst mail-extr-all-letters 296 (purecopy (concat mail-extr-all-letters-but-separators "---"))) 297 298;; Any character that can start a name. 299;; Keep this set as minimal as possible. 300(defconst mail-extr-first-letters (purecopy "[:alpha:]")) 301 302;; Any character that can end a name. 303;; Keep this set as minimal as possible. 304(defconst mail-extr-last-letters (purecopy "[:alpha:]`'.")) 305 306(defconst mail-extr-leading-garbage "\\W+") 307 308;; (defconst mail-extr-non-name-chars 309;; (purecopy (concat "^" mail-extr-all-letters "."))) 310;; (defconst mail-extr-non-begin-name-chars 311;; (purecopy (concat "^" mail-extr-first-letters))) 312;; (defconst mail-extr-non-end-name-chars 313;; (purecopy (concat "^" mail-extr-last-letters))) 314 315;; Matches an initial not followed by both a period and a space. 316;; (defconst mail-extr-bad-initials-pattern 317;; (purecopy 318;; (format "\\(\\([^%s]\\|\\`\\)[%s]\\)\\(\\.\\([^ ]\\)\\| \\|\\([^%s .]\\)\\|\\'\\)" 319;; mail-extr-all-letters mail-extr-first-letters mail-extr-all-letters))) 320 321;; Matches periods used instead of spaces. Must not match the period 322;; following an initial. 323(defconst mail-extr-bad-dot-pattern 324 (purecopy 325 (format "\\([%s][%s]\\)\\.+\\([%s]\\)" 326 mail-extr-all-letters 327 mail-extr-last-letters 328 mail-extr-first-letters))) 329 330;; Matches an embedded or leading nickname that should be removed. 331;; (defconst mail-extr-nickname-pattern 332;; (purecopy 333;; (format "\\([ .]\\|\\`\\)[\"'`\[\(]\\([ .%s]+\\)[\]\"'\)] " 334;; mail-extr-all-letters))) 335 336;; Matches the occurrence of a generational name suffix, and the last 337;; character of the preceding name. This is important because we want to 338;; keep such suffixes: they help to uniquely identify the person. 339;; *** Perhaps this should be a user-customizable variable. However, the 340;; *** regular expression is fairly tricky to alter, so maybe not. 341(defconst mail-extr-full-name-suffix-pattern 342 (purecopy 343 (format 344 "\\(,? ?\\([JjSs][Rr]\\.?\\|V?I+V?\\)\\)\\([^%s]\\([^%s]\\|\\'\\)\\|\\'\\)" 345 mail-extr-all-letters mail-extr-all-letters))) 346 347(defconst mail-extr-roman-numeral-pattern (purecopy "V?I+V?\\b")) 348 349;; Matches a trailing uppercase (with other characters possible) acronym. 350;; Must not match a trailing uppercase last name or trailing initial 351(defconst mail-extr-weird-acronym-pattern 352 (purecopy "\\([A-Z]+[-_/]\\|[A-Z][A-Z][A-Z]?\\b\\)")) 353 354;; Matches a mixed-case or lowercase name (not an initial). 355;; #### Match Latin1 lower case letters here too? 356;; (defconst mail-extr-mixed-case-name-pattern 357;; (purecopy 358;; (format 359;; "\\b\\([a-z][%s]*[%s]\\|[%s][%s]*[a-z][%s]*[%s]\\|[%s][%s]*[a-z]\\)" 360;; mail-extr-all-letters mail-extr-last-letters 361;; mail-extr-first-letters mail-extr-all-letters mail-extr-all-letters 362;; mail-extr-last-letters mail-extr-first-letters mail-extr-all-letters))) 363 364;; Matches a trailing alternative address. 365;; #### Match Latin1 letters here too? 366;; #### Match _ before @ here too? 367(defconst mail-extr-alternative-address-pattern 368 (purecopy "\\(aka *\\)?[a-zA-Z.]+[!@][a-zA-Z.]")) 369 370;; Matches a variety of trailing comments not including comma-delimited 371;; comments. 372(defconst mail-extr-trailing-comment-start-pattern 373 (purecopy " [-{]\\|--\\|[+@#></\;]")) 374 375;; Matches a name (not an initial). 376;; This doesn't force a word boundary at the end because sometimes a 377;; comment is separated by a `-' with no preceding space. 378(defconst mail-extr-name-pattern 379 (purecopy (format "\\b[%s][%s]*[%s]" 380 mail-extr-first-letters 381 mail-extr-all-letters 382 mail-extr-last-letters))) 383 384(defconst mail-extr-initial-pattern 385 (purecopy (format "\\b[%s]\\([. ]\\|\\b\\)" mail-extr-first-letters))) 386 387;; Matches a single name before a comma. 388;; (defconst mail-extr-last-name-first-pattern 389;; (purecopy (concat "\\`" mail-extr-name-pattern ","))) 390 391;; Matches telephone extensions. 392(defconst mail-extr-telephone-extension-pattern 393 (purecopy 394 "\\(\\([Ee]xt\\|\\|[Tt]ph\\|[Tt]el\\|[Xx]\\).?\\)? *\\+?[0-9][- 0-9]+")) 395 396;; Matches ham radio call signs. 397;; Help from: Mat Maessen N2NJZ <maessm@rpi.edu>, Mark Feit 398;; <mark@era.com>, Michael Covington <mcovingt@ai.uga.edu>. 399;; Examples: DX504 DX515 K5MRU K8DHK KA9WGN KA9WGN KD3FU KD6EUI KD6HBW 400;; KE9TV KF0NV N1API N3FU N3GZE N3IGS N4KCC N7IKQ N9HHU W4YHF W6ANK WA2SUH 401;; WB7VZI N2NJZ NR3G KJ4KK AB4UM AL7NI KH6OH WN3KBT N4TMI W1A N0NZO 402(defconst mail-extr-ham-call-sign-pattern 403 (purecopy "\\b\\(DX[0-9]+\\|[AKNW][A-Z]?[0-9][A-Z][A-Z]?[A-Z]?\\)")) 404 405;; Possible trailing suffixes: "\\(/\\(KT\\|A[AEG]\\|[R0-9]\\)\\)?" 406;; /KT == Temporary Technician (has CSC but not "real" license) 407;; /AA == Temporary Advanced 408;; /AE == Temporary Extra 409;; /AG == Temporary General 410;; /R == repeater 411;; /# == stations operating out of home district 412;; I don't include these in the regexp above because I can't imagine 413;; anyone putting them with their name in an e-mail address. 414 415;; Matches normal single-part name 416(defconst mail-extr-normal-name-pattern 417 (purecopy (format "\\b[%s][%s]+[%s]" 418 mail-extr-first-letters 419 mail-extr-all-letters-but-separators 420 mail-extr-last-letters))) 421 422;; Matches a single word name. 423;; (defconst mail-extr-one-name-pattern 424;; (purecopy (concat "\\`" mail-extr-normal-name-pattern "\\'"))) 425 426;; Matches normal two names with missing middle initial 427;; The first name is not allowed to have a hyphen because this can cause 428;; false matches where the "middle initial" is actually the first letter 429;; of the second part of the first name. 430(defconst mail-extr-two-name-pattern 431 (purecopy 432 (concat "\\`\\(" mail-extr-normal-name-pattern 433 "\\|" mail-extr-initial-pattern 434 "\\) +\\(" mail-extr-name-pattern "\\)\\(,\\|\\'\\)"))) 435 436(defconst mail-extr-listserv-list-name-pattern 437 (purecopy "Multiple recipients of list \\([-A-Z]+\\)")) 438 439(defconst mail-extr-stupid-vms-date-stamp-pattern 440 (purecopy 441 "[0-9][0-9]-[JFMASOND][aepuco][nbrylgptvc]-[0-9][0-9][0-9][0-9] [0-9]+ *")) 442 443;;; HZ -- GB (PRC Chinese character encoding) in ASCII embedding protocol 444;; 445;; In ASCII mode, a byte is interpreted as an ASCII character, unless a '~' is 446;; encountered. The character '~' is an escape character. By convention, it 447;; must be immediately followed ONLY by '~', '{' or '\n' (<LF>), with the 448;; following special meaning. 449;; 450;; o The escape sequence '~~' is interpreted as a '~'. 451;; o The escape-to-GB sequence '~{' switches the mode from ASCII to GB. 452;; o The escape sequence '~\n' is a line-continuation marker to be consumed 453;; with no output produced. 454;; 455;; In GB mode, characters are interpreted two bytes at a time as (pure) GB 456;; codes until the escape-from-GB code '~}' is read. This code switches the 457;; mode from GB back to ASCII. (Note that the escape-from-GB code '~}' 458;; ($7E7D) is outside the defined GB range.) 459(defconst mail-extr-hz-embedded-gb-encoded-chinese-pattern 460 (purecopy "~{\\([^~].\\|~[^\}]\\)+~}")) 461 462;; The leading optional lowercase letters are for a bastardized version of 463;; the encoding, as is the optional nature of the final slash. 464(defconst mail-extr-x400-encoded-address-pattern 465 (purecopy "[a-z]?[a-z]?\\(/[A-Za-z]+\\(\\.[A-Za-z]+\\)?=[^/]+\\)+/?\\'")) 466 467(defconst mail-extr-x400-encoded-address-field-pattern-format 468 (purecopy "/%s=\\([^/]+\\)\\(/\\|\\'\\)")) 469 470(defconst mail-extr-x400-encoded-address-surname-pattern 471 ;; S stands for Surname (family name). 472 (purecopy 473 (format mail-extr-x400-encoded-address-field-pattern-format "[Ss]"))) 474 475(defconst mail-extr-x400-encoded-address-given-name-pattern 476 ;; G stands for Given name. 477 (purecopy 478 (format mail-extr-x400-encoded-address-field-pattern-format "[Gg]"))) 479 480(defconst mail-extr-x400-encoded-address-full-name-pattern 481 ;; PN stands for Personal Name. When used it represents the combination 482 ;; of the G and S fields. 483 ;; "The one system I used having this field asked it with the prompt 484 ;; `Personal Name'. But they mapped it into G and S on outgoing real 485 ;; X.400 addresses. As they mapped G and S into PN on incoming..." 486 (purecopy 487 (format mail-extr-x400-encoded-address-field-pattern-format "[Pp][Nn]"))) 488 489 490 491;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; 492;; 493;; Syntax tables used for quick parsing. 494;; 495 496(defconst mail-extr-address-syntax-table (make-syntax-table)) 497(defconst mail-extr-address-comment-syntax-table (make-syntax-table)) 498(defconst mail-extr-address-domain-literal-syntax-table (make-syntax-table)) 499(defconst mail-extr-address-text-comment-syntax-table (make-syntax-table)) 500(defconst mail-extr-address-text-syntax-table (make-syntax-table)) 501(mapc 502 (lambda (pair) 503 (let ((syntax-table (symbol-value (car pair)))) 504 (dolist (item (cdr pair)) 505 (if (eq 2 (length item)) 506 ;; modifying syntax of a single character 507 (modify-syntax-entry (car item) (car (cdr item)) syntax-table) 508 ;; modifying syntax of a range of characters 509 (let ((char (nth 0 item)) 510 (bound (nth 1 item)) 511 (syntax (nth 2 item))) 512 (while (<= char bound) 513 (modify-syntax-entry char syntax syntax-table) 514 (setq char (1+ char)))))))) 515 '((mail-extr-address-syntax-table 516 (?\000 ?\037 "w") ;control characters 517 (?\040 " ") ;SPC 518 (?! ?~ "w") ;printable characters 519 (?\177 "w") ;DEL 520 (?\t " ") 521 (?\r " ") 522 (?\n " ") 523 (?\( ".") 524 (?\) ".") 525 (?< ".") 526 (?> ".") 527 (?@ ".") 528 (?, ".") 529 (?\; ".") 530 (?: ".") 531 (?\\ "\\") 532 (?\" "\"") 533 (?. ".") 534 (?\[ ".") 535 (?\] ".") 536 ;; % and ! aren't RFC822 characters, but it is convenient to pretend 537 (?% ".") 538 (?! ".") ;; this needs to be word-constituent when not in .UUCP mode 539 ) 540 (mail-extr-address-comment-syntax-table 541 (?\000 ?\377 "w") 542 (?\040 " ") 543 (?\240 " ") 544 (?\t " ") 545 (?\r " ") 546 (?\n " ") 547 (?\( "\(\)") 548 (?\) "\)\(") 549 (?\\ "\\")) 550 (mail-extr-address-domain-literal-syntax-table 551 (?\000 ?\377 "w") 552 (?\040 " ") 553 (?\240 " ") 554 (?\t " ") 555 (?\r " ") 556 (?\n " ") 557 (?\[ "\(\]") ;?????? 558 (?\] "\)\[") ;?????? 559 (?\\ "\\")) 560 (mail-extr-address-text-comment-syntax-table 561 (?\000 ?\377 "w") 562 (?\040 " ") 563 (?\240 " ") 564 (?\t " ") 565 (?\r " ") 566 (?\n " ") 567 (?\( "\(\)") 568 (?\) "\)\(") 569 (?\[ "\(\]") 570 (?\] "\)\[") 571 (?\{ "\(\}") 572 (?\} "\)\{") 573 (?\\ "\\") 574 (?\" "\"") 575 ;; (?\' "\)\`") 576 ;; (?\` "\(\'") 577 ) 578 (mail-extr-address-text-syntax-table 579 (?\000 ?\177 ".") 580 (?\200 ?\377 "w") 581 (?\040 " ") 582 (?\t " ") 583 (?\r " ") 584 (?\n " ") 585 (?A ?Z "w") 586 (?a ?z "w") 587 (?- "w") 588 (?\} "w") 589 (?\{ "w") 590 (?| "w") 591 (?\' "w") 592 (?~ "w") 593 (?0 ?9 "w")) 594 )) 595 596 597;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; 598;; 599;; Utility functions and macros. 600;; 601 602;; Fixme: There are Latin-1 nbsp below. If such characters should be 603;; included, this is the wrong thing to do -- it should use syntax (or 604;; regexp char classes). 605 606(defsubst mail-extr-skip-whitespace-forward () 607 ;; v19 fn skip-syntax-forward is more tasteful, but not byte-coded. 608 (skip-chars-forward " \t\n\r ")) 609 610(defsubst mail-extr-skip-whitespace-backward () 611 ;; v19 fn skip-syntax-backward is more tasteful, but not byte-coded. 612 (skip-chars-backward " \t\n\r ")) 613 614 615(defsubst mail-extr-undo-backslash-quoting (beg end) 616 (save-excursion 617 (save-restriction 618 (narrow-to-region beg end) 619 (goto-char (point-min)) 620 ;; undo \ quoting 621 (while (search-forward "\\" nil t) 622 (delete-char -1) 623 (or (eobp) 624 (forward-char 1)))))) 625 626(defsubst mail-extr-nuke-char-at (pos) 627 (save-excursion 628 (goto-char pos) 629 (delete-char 1) 630 (insert ?\ ))) 631 632(put 'mail-extr-nuke-outside-range 633 'edebug-form-spec '(symbolp &optional form form atom)) 634 635(defmacro mail-extr-nuke-outside-range (list-symbol 636 beg-symbol end-symbol 637 &optional no-replace) 638 "Delete all elements outside BEG..END in LIST. 639LIST-SYMBOL names a variable holding a list of buffer positions 640BEG-SYMBOL and END-SYMBOL name variables delimiting a range 641Each element of LIST-SYMBOL which lies outside of the range is 642 deleted from the list. 643Unless NO-REPLACE is true, at each of the positions in LIST-SYMBOL 644 which lie outside of the range, one character at that position is 645 replaced with a SPC." 646 (or (memq no-replace '(t nil)) 647 (error "no-replace must be t or nil, evaluable at macroexpand-time")) 648 `(let ((temp ,list-symbol) 649 ch) 650 (while temp 651 (setq ch (car temp)) 652 (when (or (> ch ,end-symbol) 653 (< ch ,beg-symbol)) 654 ,@(if no-replace 655 nil 656 `((mail-extr-nuke-char-at ch))) 657 (setcar temp nil)) 658 (setq temp (cdr temp))) 659 (setq ,list-symbol (delq nil ,list-symbol)))) 660 661(defun mail-extr-demarkerize (marker) 662 ;; if arg is a marker, destroys the marker, then returns the old value. 663 ;; otherwise returns the arg. 664 (if (markerp marker) 665 (let ((temp (marker-position marker))) 666 (set-marker marker nil) 667 temp) 668 marker)) 669 670(defun mail-extr-markerize (pos) 671 ;; coerces pos to a marker if non-nil. 672 (if (or (markerp pos) (null pos)) 673 pos 674 (copy-marker pos))) 675 676(defsubst mail-extr-safe-move-sexp (arg) 677 ;; Safely skip over one balanced sexp, if there is one. Return t if success. 678 (condition-case error 679 (progn 680 (goto-char (or (scan-sexps (point) arg) (point))) 681 t) 682 (error 683 ;; #### kludge kludge kludge kludge kludge kludge kludge !!! 684 (if (string-equal (nth 1 error) "Unbalanced parentheses") 685 nil 686 (while t 687 (signal (car error) (cdr error))))))) 688 689;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; 690;; 691;; The main function to grind addresses 692;; 693 694(defvar disable-initial-guessing-flag) ; dynamic assignment 695(defvar cbeg) ; dynamic assignment 696(defvar cend) ; dynamic assignment 697(defvar mail-extr-all-top-level-domains) ; Defined below. 698 699;;;###autoload 700(defun mail-extract-address-components (address &optional all) 701 "Given an RFC-822 address ADDRESS, extract full name and canonical address. 702Returns a list of the form (FULL-NAME CANONICAL-ADDRESS). If no 703name can be extracted, FULL-NAME will be nil. Also see 704`mail-extr-ignore-single-names' and 705`mail-extr-ignore-realname-equals-mailbox-name'. 706 707If the optional argument ALL is non-nil, then ADDRESS can contain zero 708or more recipients, separated by commas, and we return a list of 709the form ((FULL-NAME CANONICAL-ADDRESS) ...) with one element for 710each recipient. If ALL is nil, then if ADDRESS contains more than 711one recipients, all but the first is ignored. 712 713ADDRESS may be a string or a buffer. If it is a buffer, the visible 714\(narrowed) portion of the buffer will be interpreted as the address. 715\(This feature exists so that the clever caller might be able to avoid 716consing a string.)" 717 (let ((canonicalization-buffer (get-buffer-create " *canonical address*")) 718 (extraction-buffer (get-buffer-create " *extract address components*")) 719 value-list) 720 721 (with-current-buffer (get-buffer-create extraction-buffer) 722 (fundamental-mode) 723 (buffer-disable-undo extraction-buffer) 724 (set-syntax-table mail-extr-address-syntax-table) 725 (widen) 726 (erase-buffer) 727 (setq case-fold-search nil) 728 729 ;; Insert extra space at beginning to allow later replacement with < 730 ;; without having to move markers. 731 (insert ?\ ) 732 733 ;; Insert the address itself. 734 (cond ((stringp address) 735 (insert address)) 736 ((bufferp address) 737 (insert-buffer-substring address)) 738 (t 739 (error "Invalid address: %s" address))) 740 741 (set-text-properties (point-min) (point-max) nil) 742 743 (with-current-buffer (get-buffer-create canonicalization-buffer) 744 (fundamental-mode) 745 (buffer-disable-undo canonicalization-buffer) 746 (setq case-fold-search nil)) 747 748 749 ;; Unfold multiple lines. 750 (goto-char (point-min)) 751 (while (re-search-forward "\\([^\\]\\(\\\\\\\\\\)*\\)\n[ \t]" nil t) 752 (replace-match "\\1 " t)) 753 754 ;; Loop over addresses until we have as many as we want. 755 (while (and (or all (null value-list)) 756 (progn (goto-char (point-min)) 757 (skip-chars-forward " \t") 758 (not (eobp)))) 759 (let (char 760 end-of-address 761 <-pos >-pos @-pos colon-pos comma-pos !-pos %-pos \;-pos 762 group-:-pos group-\;-pos route-addr-:-pos 763 record-pos-symbol 764 first-real-pos last-real-pos 765 phrase-beg phrase-end 766 cbeg cend ; dynamically set from -voodoo 767 quote-beg quote-end 768 atom-beg atom-end 769 mbox-beg mbox-end 770 \.-ends-name 771 temp 772 ;; name-suffix 773 fi mi li ; first, middle, last initial 774 saved-%-pos saved-!-pos saved-@-pos 775 domain-pos \.-pos insert-point 776 ;; mailbox-name-processed-flag 777 disable-initial-guessing-flag) ; dynamically set from -voodoo 778 779 (set-syntax-table mail-extr-address-syntax-table) 780 (goto-char (point-min)) 781 782 ;; Insert extra space at beginning to allow later replacement with < 783 ;; without having to move markers. 784 (or (eq (following-char) ?\ ) 785 (insert ?\ )) 786 787 ;; First pass grabs useful information about address. 788 (while (progn 789 (mail-extr-skip-whitespace-forward) 790 (not (eobp))) 791 (setq char (char-after (point))) 792 (or first-real-pos 793 (if (not (eq char ?\()) 794 (setq first-real-pos (point)))) 795 (cond 796 ;; comment 797 ((eq char ?\() 798 (set-syntax-table mail-extr-address-comment-syntax-table) 799 ;; only record the first non-empty comment's position 800 (if (and (not cbeg) 801 (save-excursion 802 (forward-char 1) 803 (mail-extr-skip-whitespace-forward) 804 (not (eq ?\) (char-after (point)))))) 805 (setq cbeg (point))) 806 ;; TODO: don't record if unbalanced 807 (or (mail-extr-safe-move-sexp 1) 808 (forward-char 1)) 809 (set-syntax-table mail-extr-address-syntax-table) 810 (if (and cbeg 811 (not cend)) 812 (setq cend (point)))) 813 ;; quoted text 814 ((eq char ?\") 815 ;; only record the first non-empty quote's position 816 (if (and (not quote-beg) 817 (save-excursion 818 (forward-char 1) 819 (mail-extr-skip-whitespace-forward) 820 (not (eq ?\" (char-after (point)))))) 821 (setq quote-beg (point))) 822 ;; TODO: don't record if unbalanced 823 (or (mail-extr-safe-move-sexp 1) 824 (forward-char 1)) 825 (if (and quote-beg 826 (not quote-end)) 827 (setq quote-end (point)))) 828 ;; domain literals 829 ((eq char ?\[) 830 (set-syntax-table mail-extr-address-domain-literal-syntax-table) 831 (or (mail-extr-safe-move-sexp 1) 832 (forward-char 1)) 833 (set-syntax-table mail-extr-address-syntax-table)) 834 ;; commas delimit addresses when outside < > pairs. 835 ((and (eq char ?,) 836 (or (and (null <-pos) 837 ;; Handle ROUTE-ADDR address that is missing its <. 838 (not (eq ?@ (char-after (1+ (point)))))) 839 (and >-pos 840 ;; handle weird munged addresses 841 ;; BUG FIX: This test was reversed. Thanks to the 842 ;; brilliant Rod Whitby <rwhitby@research.canon.oz.au> 843 ;; for discovering this! 844 (< (car (last <-pos)) (car >-pos))))) 845 ;; The argument contains more than one address. 846 ;; Temporarily hide everything after this one. 847 (setq end-of-address (copy-marker (1+ (point)) t)) 848 (narrow-to-region (point-min) (1+ (point))) 849 (delete-char 1) 850 (setq char ?\() ; HAVE I NO SHAME?? 851 ) 852 ;; record the position of various interesting chars, determine 853 ;; legality later. 854 ((setq record-pos-symbol 855 (cdr (assq char 856 '((?< . <-pos) (?> . >-pos) (?@ . @-pos) 857 (?: . colon-pos) (?, . comma-pos) (?! . !-pos) 858 (?% . %-pos) (?\; . \;-pos))))) 859 (set record-pos-symbol 860 (cons (point) (symbol-value record-pos-symbol))) 861 (forward-char 1)) 862 ((eq char ?.) 863 (forward-char 1)) 864 ((memq char '( 865 ;; comment terminator illegal 866 ?\) 867 ;; domain literal terminator illegal 868 ?\] 869 ;; \ allowed only within quoted strings, 870 ;; domain literals, and comments 871 ?\\ 872 )) 873 (mail-extr-nuke-char-at (point)) 874 (forward-char 1)) 875 (t 876 (forward-word 1))) 877 (or (eq char ?\() 878 ;; At the end of first address of a multiple address header. 879 (and (eq char ?,) 880 (eobp)) 881 (setq last-real-pos (point)))) 882 883 ;; Use only the leftmost <, if any. Replace all others with spaces. 884 (while (cdr <-pos) 885 (mail-extr-nuke-char-at (car <-pos)) 886 (setq <-pos (cdr <-pos))) 887 888 ;; Use only the rightmost >, if any. Replace all others with spaces. 889 (while (cdr >-pos) 890 (mail-extr-nuke-char-at (nth 1 >-pos)) 891 (setcdr >-pos (nthcdr 2 >-pos))) 892 893 ;; If multiple @s and a :, but no < and >, insert around buffer. 894 ;; Example: @foo.bar.dom,@xxx.yyy.zzz:mailbox@aaa.bbb.ccc 895 ;; This commonly happens on the UUCP "From " line. Ugh. 896 (when (and (> (length @-pos) 1) 897 (eq 1 (length colon-pos)) ;TODO: check if between last two @s 898 (not \;-pos) 899 (not <-pos)) 900 (goto-char (point-min)) 901 (delete-char 1) 902 (setq <-pos (list (point))) 903 (insert ?<)) 904 905 ;; If < but no >, insert > in rightmost possible position 906 (when (and <-pos (null >-pos)) 907 (goto-char (point-max)) 908 (setq >-pos (list (point))) 909 (insert ?>)) 910 911 ;; If > but no <, replace > with space. 912 (when (and >-pos (null <-pos)) 913 (mail-extr-nuke-char-at (car >-pos)) 914 (setq >-pos nil)) 915 916 ;; Turn >-pos and <-pos into non-lists 917 (setq >-pos (car >-pos) 918 <-pos (car <-pos)) 919 920 ;; Trim other punctuation lists of items outside < > pair to handle 921 ;; stupid MTAs. 922 (when <-pos ; don't need to check >-pos also 923 ;; handle bozo software that violates RFC 822 by sticking 924 ;; punctuation marks outside of a < > pair 925 (mail-extr-nuke-outside-range @-pos <-pos >-pos t) 926 ;; RFC 822 says nothing about these two outside < >, but 927 ;; remove those positions from the lists to make things 928 ;; easier. 929 (mail-extr-nuke-outside-range !-pos <-pos >-pos t) 930 (mail-extr-nuke-outside-range %-pos <-pos >-pos t)) 931 932 ;; Check for : that indicates GROUP list and for : part of 933 ;; ROUTE-ADDR spec. 934 ;; Can't possibly be more than two :. Nuke any extra. 935 (while colon-pos 936 (setq temp (car colon-pos) 937 colon-pos (cdr colon-pos)) 938 (cond ((and <-pos >-pos 939 (> temp <-pos) 940 (< temp >-pos)) 941 (if (or route-addr-:-pos 942 (< (length @-pos) 2) 943 (> temp (car @-pos)) 944 (< temp (nth 1 @-pos))) 945 (mail-extr-nuke-char-at temp) 946 (setq route-addr-:-pos temp))) 947 ((or (not <-pos) 948 (and <-pos 949 (< temp <-pos))) 950 (setq group-:-pos temp)))) 951 952 ;; Nuke any ; that is in or to the left of a < > pair or to the left 953 ;; of a GROUP starting :. Also, there may only be one ;. 954 (while \;-pos 955 (setq temp (car \;-pos) 956 \;-pos (cdr \;-pos)) 957 (cond ((and <-pos >-pos 958 (> temp <-pos) 959 (< temp >-pos)) 960 (mail-extr-nuke-char-at temp)) 961 ((and (or (not group-:-pos) 962 (> temp group-:-pos)) 963 (not group-\;-pos)) 964 (setq group-\;-pos temp)))) 965 966 ;; Nuke unmatched GROUP syntax characters. 967 (when (and group-:-pos (not group-\;-pos)) 968 ;; *** Do I really need to erase it? 969 (mail-extr-nuke-char-at group-:-pos) 970 (setq group-:-pos nil)) 971 (when (and group-\;-pos (not group-:-pos)) 972 ;; *** Do I really need to erase it? 973 (mail-extr-nuke-char-at group-\;-pos) 974 (setq group-\;-pos nil)) 975 976 ;; Handle junk like ";@host.company.dom" that sendmail adds. 977 ;; **** should I remember comment positions? 978 (when group-\;-pos 979 ;; this is fine for now 980 (mail-extr-nuke-outside-range !-pos group-:-pos group-\;-pos t) 981 (mail-extr-nuke-outside-range @-pos group-:-pos group-\;-pos t) 982 (mail-extr-nuke-outside-range %-pos group-:-pos group-\;-pos t) 983 (mail-extr-nuke-outside-range comma-pos group-:-pos group-\;-pos t) 984 (and last-real-pos 985 (> last-real-pos (1+ group-\;-pos)) 986 (setq last-real-pos (1+ group-\;-pos))) 987 ;; *** This may be wrong: 988 (and cend 989 (> cend group-\;-pos) 990 (setq cend nil 991 cbeg nil)) 992 (and quote-end 993 (> quote-end group-\;-pos) 994 (setq quote-end nil 995 quote-beg nil)) 996 ;; This was both wrong and unnecessary: 997 ;;(narrow-to-region (point-min) group-\;-pos) 998 999 ;; *** The entire handling of GROUP addresses seems rather lame. 1000 ;; *** It deserves a complete rethink, except that these addresses 1001 ;; *** are hardly ever seen. 1002 ) 1003 1004 ;; Any commas must be between < and : of ROUTE-ADDR. Nuke any 1005 ;; others. 1006 ;; Hell, go ahead and nuke all of the commas. 1007 ;; **** This will cause problems when we start handling commas in 1008 ;; the PHRASE part .... no it won't ... yes it will ... ????? 1009 (mail-extr-nuke-outside-range comma-pos 1 1) 1010 1011 ;; can only have multiple @s inside < >. The fact that some MTAs 1012 ;; put de-bracketed ROUTE-ADDRs in the UUCP-style "From " line is 1013 ;; handled above. 1014 1015 ;; Locate PHRASE part of ROUTE-ADDR. 1016 (when <-pos 1017 (goto-char <-pos) 1018 (mail-extr-skip-whitespace-backward) 1019 (setq phrase-end (point)) 1020 (goto-char (or ;;group-:-pos 1021 (point-min))) 1022 (mail-extr-skip-whitespace-forward) 1023 (if (< (point) phrase-end) 1024 (setq phrase-beg (point)) 1025 (setq phrase-end nil))) 1026 1027 ;; handle ROUTE-ADDRS with real ROUTEs. 1028 ;; If there are multiple @s, then we assume ROUTE-ADDR syntax, and 1029 ;; any % or ! must be semantically meaningless. 1030 ;; TODO: do this processing into canonicalization buffer 1031 (when route-addr-:-pos 1032 (setq !-pos nil 1033 %-pos nil 1034 >-pos (copy-marker >-pos) 1035 route-addr-:-pos (copy-marker route-addr-:-pos)) 1036 (goto-char >-pos) 1037 (insert-before-markers ?X) 1038 (goto-char (car @-pos)) 1039 (while (setq @-pos (cdr @-pos)) 1040 (delete-char 1) 1041 (setq %-pos (cons (point-marker) %-pos)) 1042 (insert "%") 1043 (goto-char (1- >-pos)) 1044 (save-excursion 1045 (insert-buffer-substring extraction-buffer 1046 (car @-pos) route-addr-:-pos) 1047 (delete-region (car @-pos) route-addr-:-pos)) 1048 (or (cdr @-pos) 1049 (setq saved-@-pos (list (point))))) 1050 (setq @-pos saved-@-pos) 1051 (goto-char >-pos) 1052 (delete-char -1) 1053 (mail-extr-nuke-char-at route-addr-:-pos) 1054 (mail-extr-demarkerize route-addr-:-pos) 1055 (setq route-addr-:-pos nil 1056 >-pos (mail-extr-demarkerize >-pos) 1057 %-pos (mapcar 'mail-extr-demarkerize %-pos))) 1058 1059 ;; de-listify @-pos 1060 (setq @-pos (car @-pos)) 1061 1062 ;; TODO: remove comments in the middle of an address 1063 1064 (with-current-buffer canonicalization-buffer 1065 (widen) 1066 (erase-buffer) 1067 (insert-buffer-substring extraction-buffer) 1068 1069 (if <-pos 1070 (narrow-to-region (progn 1071 (goto-char (1+ <-pos)) 1072 (mail-extr-skip-whitespace-forward) 1073 (point)) 1074 >-pos) 1075 (if (and first-real-pos last-real-pos) 1076 (narrow-to-region first-real-pos last-real-pos) 1077 ;; ****** Oh no! What if the address is completely empty! 1078 ;; *** Is this correct? 1079 (narrow-to-region (point-max) (point-max)))) 1080 1081 (and @-pos %-pos 1082 (mail-extr-nuke-outside-range %-pos (point-min) @-pos)) 1083 (and %-pos !-pos 1084 (mail-extr-nuke-outside-range !-pos (point-min) (car %-pos))) 1085 (and @-pos !-pos (not %-pos) 1086 (mail-extr-nuke-outside-range !-pos (point-min) @-pos)) 1087 1088 ;; Error condition:?? (and %-pos (not @-pos)) 1089 1090 ;; WARNING: THIS CODE IS DUPLICATED BELOW. 1091 (when (and %-pos (not @-pos)) 1092 (goto-char (car %-pos)) 1093 (delete-char 1) 1094 (setq @-pos (point)) 1095 (insert "@") 1096 (setq %-pos (cdr %-pos))) 1097 1098 (when (and mail-extr-mangle-uucp !-pos) 1099 ;; **** I don't understand this save-restriction and the 1100 ;; narrow-to-region inside it. Why did I do that? 1101 (save-restriction 1102 (cond ((and @-pos 1103 mail-extr-@-binds-tighter-than-!) 1104 (goto-char @-pos) 1105 (setq %-pos (cons (point) %-pos) 1106 @-pos nil) 1107 (delete-char 1) 1108 (insert "%") 1109 (setq insert-point (point-max))) 1110 (mail-extr-@-binds-tighter-than-! 1111 (setq insert-point (point-max))) 1112 (%-pos 1113 (setq insert-point (car (last %-pos)) 1114 saved-%-pos (mapcar 'mail-extr-markerize %-pos) 1115 %-pos nil 1116 @-pos (mail-extr-markerize @-pos))) 1117 (@-pos 1118 (setq insert-point @-pos) 1119 (setq @-pos (mail-extr-markerize @-pos))) 1120 (t 1121 (setq insert-point (point-max)))) 1122 (narrow-to-region (point-min) insert-point) 1123 (setq saved-!-pos (car !-pos)) 1124 (while !-pos 1125 (goto-char (point-max)) 1126 (cond ((and (not @-pos) 1127 (not (cdr !-pos))) 1128 (setq @-pos (point)) 1129 (insert-before-markers "@ ")) 1130 (t 1131 (setq %-pos (cons (point) %-pos)) 1132 (insert-before-markers "% "))) 1133 (backward-char 1) 1134 (insert-buffer-substring 1135 (current-buffer) 1136 (if (nth 1 !-pos) 1137 (1+ (nth 1 !-pos)) 1138 (point-min)) 1139 (car !-pos)) 1140 (delete-char 1) 1141 (or (save-excursion 1142 (mail-extr-safe-move-sexp -1) 1143 (mail-extr-skip-whitespace-backward) 1144 (eq ?. (preceding-char))) 1145 (insert-before-markers 1146 (if (save-excursion 1147 (mail-extr-skip-whitespace-backward) 1148 (eq ?. (preceding-char))) 1149 "" 1150 ".") 1151 "uucp")) 1152 (setq !-pos (cdr !-pos)))) 1153 (and saved-%-pos 1154 (setq %-pos (append (mapcar 'mail-extr-demarkerize 1155 saved-%-pos) 1156 %-pos))) 1157 (setq @-pos (mail-extr-demarkerize @-pos)) 1158 (narrow-to-region (1+ saved-!-pos) (point-max))) 1159 1160 ;; WARNING: THIS CODE IS DUPLICATED ABOVE. 1161 (when (and %-pos (not @-pos)) 1162 (goto-char (car %-pos)) 1163 (delete-char 1) 1164 (setq @-pos (point)) 1165 (insert "@") 1166 (setq %-pos (cdr %-pos))) 1167 1168 (when (setq %-pos (nreverse %-pos)) ; implies @-pos valid 1169 (setq temp %-pos) 1170 (catch 'truncated 1171 (while temp 1172 (goto-char (or (nth 1 temp) 1173 @-pos)) 1174 (mail-extr-skip-whitespace-backward) 1175 (save-excursion 1176 (mail-extr-safe-move-sexp -1) 1177 (setq domain-pos (point)) 1178 (mail-extr-skip-whitespace-backward) 1179 (setq \.-pos (eq ?. (preceding-char)))) 1180 (when (and \.-pos 1181 ;; #### string consing 1182 (let ((s (intern-soft 1183 (buffer-substring domain-pos (point)) 1184 mail-extr-all-top-level-domains))) 1185 (and s (get s 'domain-name)))) 1186 (narrow-to-region (point-min) (point)) 1187 (goto-char (car temp)) 1188 (delete-char 1) 1189 (setq @-pos (point)) 1190 (setcdr temp nil) 1191 (setq %-pos (delq @-pos %-pos)) 1192 (insert "@") 1193 (throw 'truncated t)) 1194 (setq temp (cdr temp))))) 1195 (setq mbox-beg (point-min) 1196 mbox-end (if %-pos (car %-pos) 1197 (or @-pos 1198 (point-max)))) 1199 1200 (when @-pos 1201 ;; Make the domain-name part lowercase since it's case 1202 ;; insensitive anyway. 1203 (downcase-region (1+ @-pos) (point-max)))) 1204 1205 ;; Done canonicalizing address. 1206 ;; We are now back in extraction-buffer. 1207 1208 ;; Decide what part of the address to search to find the full name. 1209 (cond ( 1210 ;; Example: "First M. Last" <fml@foo.bar.dom> 1211 (and phrase-beg 1212 (eq quote-beg phrase-beg) 1213 (<= quote-end phrase-end)) 1214 (narrow-to-region (1+ quote-beg) (1- quote-end)) 1215 (mail-extr-undo-backslash-quoting (point-min) (point-max))) 1216 1217 ;; Example: First Last <fml@foo.bar.dom> 1218 (phrase-beg 1219 (narrow-to-region phrase-beg phrase-end)) 1220 1221 ;; Example: fml@foo.bar.dom (First M. Last) 1222 (cbeg 1223 (narrow-to-region (1+ cbeg) (1- cend)) 1224 (mail-extr-undo-backslash-quoting (point-min) (point-max)) 1225 1226 ;; Deal with spacing problems 1227 (goto-char (point-min)) 1228;;; (cond ((not (search-forward " " nil t)) 1229;;; (goto-char (point-min)) 1230;;; (cond ((search-forward "_" nil t) 1231;;; ;; Handle the *idiotic* use of underlines as spaces. 1232;;; ;; Example: fml@foo.bar.dom (First_M._Last) 1233;;; (goto-char (point-min)) 1234;;; (while (search-forward "_" nil t) 1235;;; (replace-match " " t))) 1236;;; ((search-forward "." nil t) 1237;;; ;; Fix . used as space 1238;;; ;; Example: danj1@cb.att.com (daniel.jacobson) 1239;;; (goto-char (point-min)) 1240;;; (while (re-search-forward mail-extr-bad-dot-pattern nil t) 1241;;; (replace-match "\\1 \\2" t)))))) 1242 ) 1243 1244 ;; Otherwise we try to get the name from the mailbox portion 1245 ;; of the address. 1246 ;; Example: First_M_Last@foo.bar.dom 1247 (t 1248 ;; *** Work in canon buffer instead? No, can't. Hmm. 1249 (goto-char (point-max)) 1250 (narrow-to-region (point) (point)) 1251 (insert-buffer-substring canonicalization-buffer 1252 mbox-beg mbox-end) 1253 (goto-char (point-min)) 1254 1255 ;; Example: First_Last.XXX@foo.bar.dom 1256 (setq \.-ends-name (re-search-forward "[_0-9]" nil t)) 1257 1258 (goto-char (point-min)) 1259 1260 (if (not mail-extr-mangle-uucp) 1261 (modify-syntax-entry ?! "w" (syntax-table))) 1262 1263 (while (progn 1264 (mail-extr-skip-whitespace-forward) 1265 (not (eobp))) 1266 (setq char (char-after (point))) 1267 (cond 1268 ((eq char ?\") 1269 (setq quote-beg (point)) 1270 (or (mail-extr-safe-move-sexp 1) 1271 ;; TODO: handle this error condition!!!!! 1272 (forward-char 1)) 1273 ;; take into account deletions 1274 (setq quote-end (- (point) 2)) 1275 (save-excursion 1276 (backward-char 1) 1277 (delete-char 1) 1278 (goto-char quote-beg) 1279 (or (eobp) 1280 (delete-char 1))) 1281 (mail-extr-undo-backslash-quoting quote-beg quote-end) 1282 (or (eq ?\ (char-after (point))) 1283 (insert " ")) 1284 ;; (setq mailbox-name-processed-flag t) 1285 (setq \.-ends-name t)) 1286 ((eq char ?.) 1287 (if (memq (char-after (1+ (point))) '(?_ ?=)) 1288 (progn 1289 (forward-char 1) 1290 (delete-char 1) 1291 (insert ?\ )) 1292 (if \.-ends-name 1293 (narrow-to-region (point-min) (point)) 1294 (delete-char 1) 1295 (insert " "))) 1296 ;; (setq mailbox-name-processed-flag t) 1297 ) 1298 ((memq (char-syntax char) '(?. ?\\)) 1299 (delete-char 1) 1300 (insert " ") 1301 ;; (setq mailbox-name-processed-flag t) 1302 ) 1303 (t 1304 (setq atom-beg (point)) 1305 (forward-word 1) 1306 (setq atom-end (point)) 1307 (goto-char atom-beg) 1308 (save-restriction 1309 (narrow-to-region atom-beg atom-end) 1310 (cond 1311 1312 ;; Handle X.400 addresses encoded in RFC-822. 1313 ;; *** Shit! This has to handle the case where it is 1314 ;; *** embedded in a quote too! 1315 ;; *** Shit! The input is being broken up into atoms 1316 ;; *** by periods! 1317 ((looking-at mail-extr-x400-encoded-address-pattern) 1318 1319 ;; Copy the contents of the individual fields that 1320 ;; might hold name data to the beginning. 1321 (mapc 1322 (lambda (field-pattern) 1323 (when 1324 (save-excursion 1325 (re-search-forward field-pattern nil t)) 1326 (insert-buffer-substring (current-buffer) 1327 (match-beginning 1) 1328 (match-end 1)) 1329 (insert " "))) 1330 (list mail-extr-x400-encoded-address-given-name-pattern 1331 mail-extr-x400-encoded-address-surname-pattern 1332 mail-extr-x400-encoded-address-full-name-pattern)) 1333 1334 ;; Discard the rest, since it contains stuff like 1335 ;; routing information, not part of a name. 1336 (mail-extr-skip-whitespace-backward) 1337 (delete-region (point) (point-max)) 1338 1339 ;; Handle periods used for spacing. 1340 (while (re-search-forward mail-extr-bad-dot-pattern nil t) 1341 (replace-match "\\1 \\2" t)) 1342 1343 ;; (setq mailbox-name-processed-flag t) 1344 ) 1345 1346 ;; Handle normal addresses. 1347 (t 1348 (goto-char (point-min)) 1349 ;; Handle _ and = used for spacing. 1350 (while (re-search-forward "\\([^_=]+\\)[_=]" nil t) 1351 (replace-match "\\1 " t) 1352 ;; (setq mailbox-name-processed-flag t) 1353 ) 1354 (goto-char (point-max)))))))) 1355 1356 ;; undo the dirty deed 1357 (if (not mail-extr-mangle-uucp) 1358 (modify-syntax-entry ?! "." (syntax-table))) 1359 ;; 1360 ;; If we derived the name from the mailbox part of the address, 1361 ;; and we only got one word out of it, don't treat that as a 1362 ;; name. "foo@bar" --> (nil "foo@bar"), not ("foo" "foo@bar") 1363 ;; (if (not mailbox-name-processed-flag) 1364 ;; (delete-region (point-min) (point-max))) 1365 )) 1366 1367 (set-syntax-table mail-extr-address-text-syntax-table) 1368 1369 (mail-extr-voodoo mbox-beg mbox-end canonicalization-buffer) 1370 (goto-char (point-min)) 1371 1372 ;; If name is "First Last" and userid is "F?L", then assume 1373 ;; the middle initial is the second letter in the userid. 1374 ;; Initial code by Jamie Zawinski <jwz@lucid.com> 1375 ;; *** Make it work when there's a suffix as well. 1376 (goto-char (point-min)) 1377 (when (and mail-extr-guess-middle-initial 1378 (not disable-initial-guessing-flag) 1379 (eq 3 (- mbox-end mbox-beg)) 1380 (progn 1381 (goto-char (point-min)) 1382 (looking-at mail-extr-two-name-pattern))) 1383 (setq fi (char-after (match-beginning 0)) 1384 li (char-after (match-beginning 3))) 1385 (with-current-buffer canonicalization-buffer 1386 ;; char-equal is ignoring case here, so no need to upcase 1387 ;; or downcase. 1388 (let ((case-fold-search t)) 1389 (and (char-equal fi (char-after mbox-beg)) 1390 (char-equal li (char-after (1- mbox-end))) 1391 (setq mi (char-after (1+ mbox-beg)))))) 1392 (when (and mi 1393 ;; TODO: use better table than syntax table 1394 (eq ?w (char-syntax mi))) 1395 (goto-char (match-beginning 3)) 1396 (insert (upcase mi) ". "))) 1397 1398 ;; Nuke name if it is the same as mailbox name. 1399 (let ((buffer-length (- (point-max) (point-min))) 1400 (i 0) 1401 (names-match-flag t)) 1402 (when (and (> buffer-length 0) 1403 (eq buffer-length (- mbox-end mbox-beg))) 1404 (goto-char (point-max)) 1405 (insert-buffer-substring canonicalization-buffer 1406 mbox-beg mbox-end) 1407 (while (and names-match-flag 1408 (< i buffer-length)) 1409 (or (eq (downcase (char-after (+ i (point-min)))) 1410 (downcase 1411 (char-after (+ i buffer-length (point-min))))) 1412 (setq names-match-flag nil)) 1413 (setq i (1+ i))) 1414 (delete-region (+ (point-min) buffer-length) (point-max)) 1415 (and names-match-flag 1416 mail-extr-ignore-realname-equals-mailbox-name 1417 (narrow-to-region (point) (point))))) 1418 1419 ;; Nuke name if it's just one word. 1420 (goto-char (point-min)) 1421 (and mail-extr-ignore-single-names 1422 (not (re-search-forward "[- ]" nil t)) 1423 (narrow-to-region (point) (point))) 1424 1425 ;; Record the result 1426 (setq value-list 1427 (cons (list (if (not (= (point-min) (point-max))) 1428 (buffer-string)) 1429 (with-current-buffer canonicalization-buffer 1430 (if (not (= (point-min) (point-max))) 1431 (buffer-string)))) 1432 value-list)) 1433 1434 ;; Unless one address is all we wanted, 1435 ;; delete this one from extraction-buffer 1436 ;; and get ready to extract the next address. 1437 (when all 1438 (if end-of-address 1439 (narrow-to-region 1 end-of-address) 1440 (widen)) 1441 (delete-region (point-min) (point-max)) 1442 (widen)) 1443 ))) 1444 (if all (nreverse value-list) (car value-list)) 1445 )) 1446 1447(defcustom mail-extr-disable-voodoo "\\cj" 1448 "*If it is a regexp, names matching it will never be modified. 1449If it is neither nil nor a string, modifying of names will never take 1450place. It affects how `mail-extract-address-components' works." 1451 :type '(choice (regexp :size 0) 1452 (const :tag "Always enabled" nil) 1453 (const :tag "Always disabled" t)) 1454 :group 'mail-extr) 1455 1456(defun mail-extr-voodoo (mbox-beg mbox-end canonicalization-buffer) 1457 (unless (and mail-extr-disable-voodoo 1458 (or (not (stringp mail-extr-disable-voodoo)) 1459 (progn 1460 (goto-char (point-min)) 1461 (re-search-forward mail-extr-disable-voodoo nil t)))) 1462 (let ((word-count 0) 1463 (case-fold-search nil) 1464 mixed-case-flag lower-case-flag ;;upper-case-flag 1465 suffix-flag last-name-comma-flag 1466 ;;cbeg cend 1467 initial 1468 begin-again-flag 1469 drop-this-word-if-trailing-flag 1470 drop-last-word-if-trailing-flag 1471 word-found-flag 1472 this-word-beg last-word-beg 1473 name-beg name-end 1474 name-done-flag 1475 ) 1476 (save-excursion 1477 (set-syntax-table mail-extr-address-text-syntax-table) 1478 1479 ;; Get rid of comments. 1480 (goto-char (point-min)) 1481 (while (not (eobp)) 1482 ;; Initialize for this iteration of the loop. 1483 (skip-chars-forward "^({[\"'`") 1484 (let ((cbeg (point))) 1485 (set-syntax-table mail-extr-address-text-comment-syntax-table) 1486 (if (memq (following-char) '(?\' ?\`)) 1487 (search-forward "'" nil 'move 1488 (if (eq ?\' (following-char)) 2 1)) 1489 (or (mail-extr-safe-move-sexp 1) 1490 (goto-char (point-max)))) 1491 (set-syntax-table mail-extr-address-text-syntax-table) 1492 (when (eq (char-after cbeg) ?\() 1493 ;; Delete the comment itself. 1494 (delete-region cbeg (point)) 1495 ;; Canonicalize whitespace where the comment was. 1496 (skip-chars-backward " \t") 1497 (if (looking-at "\\([ \t]+$\\|[ \t]+,\\)") 1498 (replace-match "") 1499 (setq cbeg (point)) 1500 (skip-chars-forward " \t") 1501 (if (bobp) 1502 (delete-region (point) cbeg) 1503 (just-one-space)))))) 1504 1505 ;; This was moved above. 1506 ;; Fix . used as space 1507 ;; But it belongs here because it occurs not only as 1508 ;; rypens@reks.uia.ac.be (Piet.Rypens) 1509 ;; but also as 1510 ;; "Piet.Rypens" <rypens@reks.uia.ac.be> 1511 ;;(goto-char (point-min)) 1512 ;;(while (re-search-forward mail-extr-bad-dot-pattern nil t) 1513 ;; (replace-match "\\1 \\2" t)) 1514 1515 (unless (search-forward " " nil t) 1516 (goto-char (point-min)) 1517 (cond ((search-forward "_" nil t) 1518 ;; Handle the *idiotic* use of underlines as spaces. 1519 ;; Example: fml@foo.bar.dom (First_M._Last) 1520 (goto-char (point-min)) 1521 (while (search-forward "_" nil t) 1522 (replace-match " " t))) 1523 ((search-forward "." nil t) 1524 ;; Fix . used as space 1525 ;; Example: danj1@cb.att.com (daniel.jacobson) 1526 (goto-char (point-min)) 1527 (while (re-search-forward mail-extr-bad-dot-pattern nil t) 1528 (replace-match "\\1 \\2" t))))) 1529 1530 ;; Loop over the words (and other junk) in the name. 1531 (goto-char (point-min)) 1532 (while (not name-done-flag) 1533 1534 (when word-found-flag 1535 ;; Last time through this loop we skipped over a word. 1536 (setq last-word-beg this-word-beg) 1537 (setq drop-last-word-if-trailing-flag 1538 drop-this-word-if-trailing-flag) 1539 (setq word-found-flag nil)) 1540 1541 (when begin-again-flag 1542 ;; Last time through the loop we found something that 1543 ;; indicates we should pretend we are beginning again from 1544 ;; the start. 1545 (setq word-count 0) 1546 (setq last-word-beg nil) 1547 (setq drop-last-word-if-trailing-flag nil) 1548 (setq mixed-case-flag nil) 1549 (setq lower-case-flag nil) 1550 ;; (setq upper-case-flag nil) 1551 (setq begin-again-flag nil)) 1552 1553 ;; Initialize for this iteration of the loop. 1554 (mail-extr-skip-whitespace-forward) 1555 (if (eq word-count 0) (narrow-to-region (point) (point-max))) 1556 (setq this-word-beg (point)) 1557 (setq drop-this-word-if-trailing-flag nil) 1558 1559 ;; Decide what to do based on what we are looking at. 1560 (cond 1561 1562 ;; Delete title 1563 ((and (eq word-count 0) 1564 (looking-at mail-extr-full-name-prefixes)) 1565 (goto-char (match-end 0)) 1566 (narrow-to-region (point) (point-max))) 1567 1568 ;; Stop after name suffix 1569 ((and (>= word-count 2) 1570 (looking-at mail-extr-full-name-suffix-pattern)) 1571 (mail-extr-skip-whitespace-backward) 1572 (setq suffix-flag (point)) 1573 (if (eq ?, (following-char)) 1574 (forward-char 1) 1575 (insert ?,)) 1576 ;; Enforce at least one space after comma 1577 (or (eq ?\ (following-char)) 1578 (insert ?\ )) 1579 (mail-extr-skip-whitespace-forward) 1580 (cond ((memq (following-char) '(?j ?J ?s ?S)) 1581 (capitalize-word 1) 1582 (if (eq (following-char) ?.) 1583 (forward-char 1) 1584 (insert ?.))) 1585 (t 1586 (upcase-word 1))) 1587 (setq word-found-flag t) 1588 (setq name-done-flag t)) 1589 1590 ;; Handle SCA names 1591 ((looking-at "MKA \\(.+\\)") ; "Mundanely Known As" 1592 (goto-char (match-beginning 1)) 1593 (narrow-to-region (point) (point-max)) 1594 (setq begin-again-flag t)) 1595 1596 ;; Check for initial last name followed by comma 1597 ((and (eq ?, (following-char)) 1598 (eq word-count 1)) 1599 (forward-char 1) 1600 (setq last-name-comma-flag t) 1601 (or (eq ?\ (following-char)) 1602 (insert ?\ ))) 1603 1604 ;; Stop before trailing comma-separated comment 1605 ;; THIS CASE MUST BE AFTER THE PRECEDING CASES. 1606 ;; *** This case is redundant??? 1607 ;;((eq ?, (following-char)) 1608 ;; (setq name-done-flag t)) 1609 1610 ;; Delete parenthesized/quoted comment/nickname 1611 ((memq (following-char) '(?\( ?\{ ?\[ ?\" ?\' ?\`)) 1612 (setq cbeg (point)) 1613 (set-syntax-table mail-extr-address-text-comment-syntax-table) 1614 (cond ((memq (following-char) '(?\' ?\`)) 1615 (or (search-forward "'" nil t 1616 (if (eq ?\' (following-char)) 2 1)) 1617 (delete-char 1))) 1618 (t 1619 (or (mail-extr-safe-move-sexp 1) 1620 (goto-char (point-max))))) 1621 (set-syntax-table mail-extr-address-text-syntax-table) 1622 (setq cend (point)) 1623 (cond 1624 ;; Handle case of entire name being quoted 1625 ((and (eq word-count 0) 1626 (looking-at " *\\'") 1627 (>= (- cend cbeg) 2)) 1628 (narrow-to-region (1+ cbeg) (1- cend)) 1629 (goto-char (point-min))) 1630 (t 1631 ;; Handle case of quoted initial 1632 (if (and (or (= 3 (- cend cbeg)) 1633 (and (= 4 (- cend cbeg)) 1634 (eq ?. (char-after (+ 2 cbeg))))) 1635 (not (looking-at " *\\'"))) 1636 (setq initial (char-after (1+ cbeg))) 1637 (setq initial nil)) 1638 (delete-region cbeg cend) 1639 (if initial 1640 (insert initial ". "))))) 1641 1642 ;; Handle *Stupid* VMS date stamps 1643 ((looking-at mail-extr-stupid-vms-date-stamp-pattern) 1644 (replace-match "" t)) 1645 1646 ;; Handle Chinese characters. 1647 ((looking-at mail-extr-hz-embedded-gb-encoded-chinese-pattern) 1648 (goto-char (match-end 0)) 1649 (setq word-found-flag t)) 1650 1651 ;; Skip initial garbage characters. 1652 ;; THIS CASE MUST BE AFTER THE PRECEDING CASES. 1653 ((and (eq word-count 0) 1654 (looking-at mail-extr-leading-garbage)) 1655 (goto-char (match-end 0)) 1656 ;; *** Skip backward over these??? 1657 ;; (skip-chars-backward "& \"") 1658 (narrow-to-region (point) (point-max))) 1659 1660 ;; Various stopping points 1661 ((or 1662 1663 ;; Stop before ALL CAPS acronyms, if preceded by mixed-case 1664 ;; words. Example: XT-DEM. 1665 (and (>= word-count 2) 1666 mixed-case-flag 1667 (looking-at mail-extr-weird-acronym-pattern) 1668 (not (looking-at mail-extr-roman-numeral-pattern))) 1669 1670 ;; Stop before trailing alternative address 1671 (looking-at mail-extr-alternative-address-pattern) 1672 1673 ;; Stop before trailing comment not introduced by comma 1674 ;; THIS CASE MUST BE AFTER AN EARLIER CASE. 1675 (looking-at mail-extr-trailing-comment-start-pattern) 1676 1677 ;; Stop before telephone numbers 1678 (and (>= word-count 1) 1679 (looking-at mail-extr-telephone-extension-pattern))) 1680 (setq name-done-flag t)) 1681 1682 ;; Delete ham radio call signs 1683 ((looking-at mail-extr-ham-call-sign-pattern) 1684 (delete-region (match-beginning 0) (match-end 0))) 1685 1686 ;; Fixup initials 1687 ((looking-at mail-extr-initial-pattern) 1688 (or (eq (following-char) (upcase (following-char))) 1689 (setq lower-case-flag t)) 1690 (forward-char 1) 1691 (if (eq ?. (following-char)) 1692 (forward-char 1) 1693 (insert ?.)) 1694 (or (eq ?\ (following-char)) 1695 (insert ?\ )) 1696 (setq word-found-flag t)) 1697 1698 ;; Handle BITNET LISTSERV list names. 1699 ((and (eq word-count 0) 1700 (looking-at mail-extr-listserv-list-name-pattern)) 1701 (narrow-to-region (match-beginning 1) (match-end 1)) 1702 (setq word-found-flag t) 1703 (setq name-done-flag t)) 1704 1705 ;; Handle & substitution, when & is last and is not first. 1706 ((and (> word-count 0) 1707 (eq ?\ (preceding-char)) 1708 (eq (following-char) ?&) 1709 (eq (1+ (point)) (point-max))) 1710 (delete-char 1) 1711 (capitalize-region 1712 (point) 1713 (progn 1714 (insert-buffer-substring canonicalization-buffer 1715 mbox-beg mbox-end) 1716 (point))) 1717 (setq disable-initial-guessing-flag t) 1718 (setq word-found-flag t)) 1719 1720 ;; Handle & between names, as in "Bob & Susie". 1721 ((and (> word-count 0) (eq (following-char) ?\&)) 1722 (setq name-beg (point)) 1723 (setq name-end (1+ name-beg)) 1724 (setq word-found-flag t) 1725 (goto-char name-end)) 1726 1727 ;; Regular name words 1728 ((looking-at mail-extr-name-pattern) 1729 (setq name-beg (point)) 1730 (setq name-end (match-end 0)) 1731 1732 ;; Certain words will be dropped if they are at the end. 1733 (and (>= word-count 2) 1734 (not lower-case-flag) 1735 (or 1736 ;; Trailing 4-or-more letter lowercase words preceded by 1737 ;; mixed case or uppercase words will be dropped. 1738 (looking-at "[[:lower:]]\\{4,\\}[ \t]*\\'") 1739 ;; Drop a trailing word which is terminated with a period. 1740 (eq ?. (char-after (1- name-end)))) 1741 (setq drop-this-word-if-trailing-flag t)) 1742 1743 ;; Set the flags that indicate whether we have seen a lowercase 1744 ;; word, a mixed case word, and an uppercase word. 1745 (if (re-search-forward "[[:lower:]]" name-end t) 1746 (if (progn 1747 (goto-char name-beg) 1748 (re-search-forward "[[:upper:]]" name-end t)) 1749 (setq mixed-case-flag t) 1750 (setq lower-case-flag t)) 1751 ;; (setq upper-case-flag t) 1752 ) 1753 1754 (goto-char name-end) 1755 (setq word-found-flag t)) 1756 1757 ;; Allow a number as a word, if it doesn't mean anything else. 1758 ((looking-at "[0-9]+\\>") 1759 (setq name-beg (point)) 1760 (setq name-end (match-end 0)) 1761 (goto-char name-end) 1762 (setq word-found-flag t)) 1763 1764 (t 1765 (setq name-done-flag t) 1766 )) 1767 1768 ;; Count any word that we skipped over. 1769 (if word-found-flag 1770 (setq word-count (1+ word-count)))) 1771 1772 ;; If the last thing in the name is 2 or more periods, or one or more 1773 ;; other sentence terminators (but not a single period) then keep them 1774 ;; and the preceding word. This is for the benefit of whole sentences 1775 ;; in the name field: it's better behavior than dropping the last word 1776 ;; of the sentence... 1777 (if (and (not suffix-flag) 1778 (looking-at "\\(\\.+\\|[?!;:.][?!;:.]+\\|[?!;:][?!;:.]*\\)\\'")) 1779 (goto-char (setq suffix-flag (point-max)))) 1780 1781 ;; Drop everything after point and certain trailing words. 1782 (narrow-to-region (point-min) 1783 (or (and drop-last-word-if-trailing-flag 1784 last-word-beg) 1785 (point))) 1786 1787 ;; Xerox's mailers SUCK!!!!!! 1788 ;; We simply refuse to believe that any last name is PARC or ADOC. 1789 ;; If it looks like that is the last name, that there is no meaningful 1790 ;; here at all. Actually I guess it would be best to map patterns 1791 ;; like foo.hoser@xerox.com into foo@hoser.xerox.com, but I don't 1792 ;; actually know that that is what's going on. 1793 (unless suffix-flag 1794 (goto-char (point-min)) 1795 (let ((case-fold-search t)) 1796 (if (looking-at "[-A-Za-z_]+[. ]\\(PARC\\|ADOC\\)\\'") 1797 (erase-buffer)))) 1798 1799 ;; If last name first put it at end (but before suffix) 1800 (when last-name-comma-flag 1801 (goto-char (point-min)) 1802 (search-forward ",") 1803 (setq name-end (1- (point))) 1804 (goto-char (or suffix-flag (point-max))) 1805 (or (eq ?\ (preceding-char)) 1806 (insert ?\ )) 1807 (insert-buffer-substring (current-buffer) (point-min) name-end) 1808 (goto-char name-end) 1809 (skip-chars-forward "\t ,") 1810 (narrow-to-region (point) (point-max))) 1811 1812 ;; Delete leading and trailing junk characters. 1813 ;; *** This is probably completely unneeded now. 1814 ;;(goto-char (point-max)) 1815 ;;(skip-chars-backward mail-extr-non-end-name-chars) 1816 ;;(if (eq ?. (following-char)) 1817 ;; (forward-char 1)) 1818 ;;(narrow-to-region (point) 1819 ;; (progn 1820 ;; (goto-char (point-min)) 1821 ;; (skip-chars-forward mail-extr-non-begin-name-chars) 1822 ;; (point))) 1823 1824 ;; Compress whitespace 1825 (goto-char (point-min)) 1826 (while (re-search-forward "[ \t\n]+" nil t) 1827 (replace-match (if (eobp) "" " ") t)) 1828 )))) 1829 1830 1831 1832;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; 1833;; 1834;; Table of top-level domain names. 1835;; 1836;; This is used during address canonicalization; be careful of format changes. 1837;; Keep in mind that the country abbreviations follow ISO-3166. There is 1838;; a U.S. FIPS that specifies a different set of two-letter country 1839;; abbreviations. 1840;; 1841;; Updated by the RIPE Network Coordination Centre. 1842;; 1843;; Source: ISO 3166 Maintenance Agency 1844;; http://www.iso.org/iso/en/prods-services/iso3166ma/02iso-3166-code-lists/list-en1-semic.txt 1845;; http://www.iana.org/domain-names.htm 1846;; http://www.iana.org/cctld/cctld-whois.htm 1847;; Latest change: Mon Jul 8 14:21:59 CEST 2002 1848 1849(defconst mail-extr-all-top-level-domains 1850 (let ((ob (make-vector 739 0))) 1851 (mapc 1852 (lambda (x) 1853 (put (intern (downcase (car x)) ob) 1854 'domain-name 1855 (if (nth 2 x) 1856 (format (nth 2 x) (nth 1 x)) 1857 (nth 1 x)))) 1858 '( 1859 ;; ISO 3166 codes: 1860 ("ad" "Andorra") 1861 ("ae" "United Arab Emirates") 1862 ("af" "Afghanistan") 1863 ("ag" "Antigua and Barbuda") 1864 ("ai" "Anguilla") 1865 ("al" "Albania") 1866 ("am" "Armenia") 1867 ("an" "Netherlands Antilles") 1868 ("ao" "Angola") 1869 ("aq" "Antarctica") ; continent 1870 ("ar" "Argentina" "Argentine Republic") 1871 ("as" "American Samoa") 1872 ("at" "Austria" "The Republic of %s") 1873 ("au" "Australia") 1874 ("aw" "Aruba") 1875 ("az" "Azerbaijan") 1876 ("ba" "Bosnia-Herzegovina") 1877 ("bb" "Barbados") 1878 ("bd" "Bangladesh") 1879 ("be" "Belgium" "The Kingdom of %s") 1880 ("bf" "Burkina Faso") 1881 ("bg" "Bulgaria") 1882 ("bh" "Bahrain") 1883 ("bi" "Burundi") 1884 ("bj" "Benin") 1885 ("bm" "Bermuda") 1886 ("bn" "Brunei Darussalam") 1887 ("bo" "Bolivia" "Republic of %s") 1888 ("br" "Brazil" "The Federative Republic of %s") 1889 ("bs" "Bahamas") 1890 ("bt" "Bhutan") 1891 ("bv" "Bouvet Island") 1892 ("bw" "Botswana") 1893 ("by" "Belarus") 1894 ("bz" "Belize") 1895 ("ca" "Canada") 1896 ("cc" "Cocos (Keeling) Islands") 1897 ("cd" "Congo" "The Democratic Republic of the %s") 1898 ("cf" "Central African Republic") 1899 ("cg" "Congo") 1900 ("ch" "Switzerland" "The Swiss Confederation") 1901 ("ci" "Ivory Coast") ; Cote D'ivoire 1902 ("ck" "Cook Islands") 1903 ("cl" "Chile" "The Republic of %s") 1904 ("cm" "Cameroon") ; In .fr domain 1905 ("cn" "China" "The People's Republic of %s") 1906 ("co" "Colombia") 1907 ("cr" "Costa Rica" "The Republic of %s") 1908 ("cu" "Cuba") 1909 ("cv" "Cape Verde") 1910 ("cx" "Christmas Island") 1911 ("cy" "Cyprus") 1912 ("cz" "Czech Republic") 1913 ("de" "Germany") 1914 ("dj" "Djibouti") 1915 ("dk" "Denmark") 1916 ("dm" "Dominica") 1917 ("do" "Dominican Republic" "The %s") 1918 ("dz" "Algeria") 1919 ("ec" "Ecuador" "The Republic of %s") 1920 ("ee" "Estonia") 1921 ("eg" "Egypt" "The Arab Republic of %s") 1922 ("eh" "Western Sahara") 1923 ("er" "Eritrea") 1924 ("es" "Spain" "The Kingdom of %s") 1925 ("et" "Ethiopia") 1926 ("fi" "Finland" "The Republic of %s") 1927 ("fj" "Fiji") 1928 ("fk" "Falkland Islands (Malvinas)") 1929 ("fm" "Micronesia" "Federated States of %s") 1930 ("fo" "Faroe Islands") 1931 ("fr" "France") 1932 ("ga" "Gabon") 1933 ("gb" "United Kingdom") 1934 ("gd" "Grenada") 1935 ("ge" "Georgia") 1936 ("gf" "French Guiana") 1937 ("gh" "Ghana") 1938 ("gi" "Gibraltar") 1939 ("gl" "Greenland") 1940 ("gm" "Gambia") 1941 ("gn" "Guinea") 1942 ("gp" "Guadeloupe (Fr.)") 1943 ("gq" "Equatorial Guinea") 1944 ("gr" "Greece" "The Hellenic Republic (%s)") 1945 ("gs" "South Georgia and The South Sandwich Islands") 1946 ("gt" "Guatemala") 1947 ("gu" "Guam (U.S.)") 1948 ("gw" "Guinea-Bissau") 1949 ("gy" "Guyana") 1950 ("hk" "Hong Kong") 1951 ("hm" "Heard Island and Mcdonald Islands") 1952 ("hn" "Honduras") 1953 ("hr" "Croatia" "Croatia (Hrvatska)") 1954 ("ht" "Haiti") 1955 ("hu" "Hungary" "The Hungarian Republic") 1956 ("id" "Indonesia") 1957 ("ie" "Ireland") 1958 ("il" "Israel" "The State of %s") 1959 ("im" "Isle of Man" "The %s") ; NOT in ISO 3166-1 of 2001-02-26 1960 ("in" "India" "The Republic of %s") 1961 ("io" "British Indian Ocean Territory") 1962 ("iq" "Iraq") 1963 ("ir" "Iran" "Islamic Republic of %s") 1964 ("is" "Iceland" "The Republic of %s") 1965 ("it" "Italy" "The Italian Republic") 1966 ("jm" "Jamaica") 1967 ("jo" "Jordan") 1968 ("jp" "Japan") 1969 ("ke" "Kenya") 1970 ("kg" "Kyrgyzstan") 1971 ("kh" "Cambodia") 1972 ("ki" "Kiribati") 1973 ("km" "Comoros") 1974 ("kn" "Saint Kitts and Nevis") 1975 ("kp" "Korea (North)" "Democratic People's Republic of Korea") 1976 ("kr" "Korea (South)" "Republic of Korea") 1977 ("kw" "Kuwait") 1978 ("ky" "Cayman Islands") 1979 ("kz" "Kazakhstan") 1980 ("la" "Lao People's Democratic Republic") 1981 ("lb" "Lebanon") 1982 ("lc" "Saint Lucia") 1983 ("li" "Liechtenstein") 1984 ("lk" "Sri Lanka" "The Democratic Socialist Republic of %s") 1985 ("lr" "Liberia") 1986 ("ls" "Lesotho") 1987 ("lt" "Lithuania") 1988 ("lu" "Luxembourg") 1989 ("lv" "Latvia") 1990 ("ly" "Libyan Arab Jamahiriya") 1991 ("ma" "Morocco") 1992 ("mc" "Monaco") 1993 ("md" "Moldova" "The Republic of %s") 1994 ("mg" "Madagascar") 1995 ("mh" "Marshall Islands") 1996 ("mk" "Macedonia" "The Former Yugoslav Republic of %s") 1997 ("ml" "Mali") 1998 ("mm" "Myanmar") 1999 ("mn" "Mongolia") 2000 ("mo" "Macao") 2001 ("mp" "Northern Mariana Islands") 2002 ("mq" "Martinique") 2003 ("mr" "Mauritania") 2004 ("ms" "Montserrat") 2005 ("mt" "Malta") 2006 ("mu" "Mauritius") 2007 ("mv" "Maldives") 2008 ("mw" "Malawi") 2009 ("mx" "Mexico" "The United Mexican States") 2010 ("my" "Malaysia") 2011 ("mz" "Mozambique") 2012 ("na" "Namibia") 2013 ("nc" "New Caledonia (Fr.)") 2014 ("ne" "Niger") ; In .fr domain 2015 ("nf" "Norfolk Island") 2016 ("ng" "Nigeria") 2017 ("ni" "Nicaragua" "The Republic of %s") 2018 ("nl" "Netherlands" "The Kingdom of the %s") 2019 ("no" "Norway" "The Kingdom of %s") 2020 ("np" "Nepal") ; Via .in domain 2021 ("nr" "Nauru") 2022 ("nu" "Niue") 2023 ("nz" "New Zealand") 2024 ("om" "Oman") 2025 ("pa" "Panama") 2026 ("pe" "Peru") 2027 ("pf" "French Polynesia") 2028 ("pg" "Papua New Guinea") 2029 ("ph" "Philippines" "The Republic of the %s") 2030 ("pk" "Pakistan") 2031 ("pl" "Poland") 2032 ("pm" "Saint Pierre and Miquelon") 2033 ("pn" "Pitcairn") 2034 ("pr" "Puerto Rico (U.S.)") 2035 ("ps" "Palestinian Territory, Occupied") 2036 ("pt" "Portugal" "The Portuguese Republic") 2037 ("pw" "Palau") 2038 ("py" "Paraguay") 2039 ("qa" "Qatar") 2040 ("re" "Reunion (Fr.)") ; In .fr domain 2041 ("ro" "Romania") 2042 ("ru" "Russia" "Russian Federation") 2043 ("rw" "Rwanda") 2044 ("sa" "Saudi Arabia") 2045 ("sb" "Solomon Islands") 2046 ("sc" "Seychelles") 2047 ("sd" "Sudan") 2048 ("se" "Sweden" "The Kingdom of %s") 2049 ("sg" "Singapore" "The Republic of %s") 2050 ("sh" "Saint Helena") 2051 ("si" "Slovenia") 2052 ("sj" "Svalbard and Jan Mayen") ; In .no domain 2053 ("sk" "Slovakia" "The Slovak Republic") 2054 ("sl" "Sierra Leone") 2055 ("sm" "San Marino") 2056 ("sn" "Senegal") 2057 ("so" "Somalia") 2058 ("sr" "Suriname") 2059 ("st" "Sao Tome and Principe") 2060 ("su" "U.S.S.R." "The Union of Soviet Socialist Republics") 2061 ("sv" "El Salvador") 2062 ("sy" "Syrian Arab Republic") 2063 ("sz" "Swaziland") 2064 ("tc" "Turks and Caicos Islands") 2065 ("td" "Chad") 2066 ("tf" "French Southern Territories") 2067 ("tg" "Togo") 2068 ("th" "Thailand" "The Kingdom of %s") 2069 ("tj" "Tajikistan") 2070 ("tk" "Tokelau") 2071 ("tl" "East Timor") 2072 ("tm" "Turkmenistan") 2073 ("tn" "Tunisia") 2074 ("to" "Tonga") 2075 ("tp" "East Timor") 2076 ("tr" "Turkey" "The Republic of %s") 2077 ("tt" "Trinidad and Tobago") 2078 ("tv" "Tuvalu") 2079 ("tw" "Taiwan" "%s, Province of China") 2080 ("tz" "Tanzania" "United Republic of %s") 2081 ("ua" "Ukraine") 2082 ("ug" "Uganda") 2083 ("uk" "United Kingdom" "The %s of Great Britain and Northern Ireland") 2084 ("um" "United States Minor Outlying Islands") 2085 ("us" "United States" "The %s of America") 2086 ("uy" "Uruguay" "The Eastern Republic of %s") 2087 ("uz" "Uzbekistan") 2088 ("va" "Holy See (Vatican City State)") 2089 ("vc" "Saint Vincent and the Grenadines") 2090 ("ve" "Venezuela" "The Republic of %s") 2091 ("vg" "Virgin Islands, British") 2092 ("vi" "Virgin Islands, U.S.") 2093 ("vn" "Vietnam") 2094 ("vu" "Vanuatu") 2095 ("wf" "Wallis and Futuna") 2096 ("ws" "Samoa") 2097 ("ye" "Yemen") 2098 ("yt" "Mayotte") 2099 ("yu" "Yugoslavia" "Yugoslavia, AKA Serbia-Montenegro") 2100 ("za" "South Africa" "The Republic of %s") 2101 ("zm" "Zambia") 2102 ("zw" "Zimbabwe" "Republic of %s") 2103 ;; Generic Domains: 2104 ("aero" t "Air Transport Industry") 2105 ("biz" t "Businesses") 2106 ("com" t "Commercial") 2107 ("coop" t "Cooperative Associations") 2108 ("info" t "Info") 2109 ("museum" t "Museums") 2110 ("name" t "Individuals") 2111 ("net" t "Network") 2112 ("org" t "Non-profit Organization") 2113 ;;("pro" t "Credentialed professionals") 2114 ;;("bitnet" t "Because It's Time NET") 2115 ("gov" t "United States Government") 2116 ("edu" t "Educational") 2117 ("mil" t "United States Military") 2118 ("int" t "International Treaties") 2119 ;;("nato" t "North Atlantic Treaty Organization") 2120 ("uucp" t "Unix to Unix CoPy") 2121 ;; Infrastructure Domains: 2122 ("arpa" t "Advanced Research Projects Agency (U.S. DoD)") 2123 )) 2124 ob)) 2125 2126;;;###autoload 2127(defun what-domain (domain) 2128 "Convert mail domain DOMAIN to the country it corresponds to." 2129 (interactive 2130 (let ((completion-ignore-case t)) 2131 (list (completing-read "Domain: " 2132 mail-extr-all-top-level-domains nil t)))) 2133 (or (setq domain (intern-soft (downcase domain) 2134 mail-extr-all-top-level-domains)) 2135 (error "No such domain")) 2136 (message "%s: %s" (upcase (symbol-name domain)) (get domain 'domain-name))) 2137 2138 2139;(let ((all nil)) 2140; (mapatoms #'(lambda (x) 2141; (if (and (boundp x) 2142; (string-match "^mail-extr-" (symbol-name x))) 2143; (setq all (cons x all))))) 2144; (setq all (sort all #'string-lessp)) 2145; (cons 'setq 2146; (apply 'nconc (mapcar #'(lambda (x) 2147; (list x (symbol-value x))) 2148; all)))) 2149 2150 2151(provide 'mail-extr) 2152 2153;;; arch-tag: 7785fade-1073-4ed6-b4f6-28db34a7982d 2154;;; mail-extr.el ends here 2155