Commit | Line | Data |
---|---|---|
72c0ae01 ER |
1 | ;;; mail-extr.el --- extract full name and address from RFC 822 mail header. |
2 | ||
d1782bd8 | 3 | ;; Copyright (C) 1991, 1992, 1993, 1994, 1997 Free Software Foundation, Inc. |
3a801d0c | 4 | |
72c0ae01 | 5 | ;; Author: Joe Wells <jbw@cs.bu.edu> |
d1782bd8 | 6 | ;; Maintainer: FSF |
72c0ae01 ER |
7 | ;; Keywords: mail |
8 | ||
72c0ae01 ER |
9 | ;; This file is part of GNU Emacs. |
10 | ||
11 | ;; GNU Emacs is free software; you can redistribute it and/or modify | |
12 | ;; it under the terms of the GNU General Public License as published by | |
154b3e39 | 13 | ;; the Free Software Foundation; either version 2, or (at your option) |
72c0ae01 ER |
14 | ;; any later version. |
15 | ||
16 | ;; GNU Emacs is distributed in the hope that it will be useful, | |
17 | ;; but WITHOUT ANY WARRANTY; without even the implied warranty of | |
18 | ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
19 | ;; GNU General Public License for more details. | |
20 | ||
21 | ;; You should have received a copy of the GNU General Public License | |
b578f267 EN |
22 | ;; along with GNU Emacs; see the file COPYING. If not, write to the |
23 | ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
24 | ;; Boston, MA 02111-1307, USA. | |
72c0ae01 ER |
25 | |
26 | ;;; Commentary: | |
27 | ||
154b3e39 RS |
28 | ;; The entry point of this code is |
29 | ;; | |
d1782bd8 | 30 | ;; mail-extract-address-components: (address &optional all) |
154b3e39 RS |
31 | ;; |
32 | ;; Given an RFC-822 ADDRESS, extract full name and canonical address. | |
33 | ;; Returns a list of the form (FULL-NAME CANONICAL-ADDRESS). | |
34 | ;; If no name can be extracted, FULL-NAME will be nil. | |
35 | ;; ADDRESS may be a string or a buffer. If it is a buffer, the visible | |
36 | ;; (narrowed) portion of the buffer will be interpreted as the address. | |
37 | ;; (This feature exists so that the clever caller might be able to avoid | |
38 | ;; consing a string.) | |
39 | ;; If ADDRESS contains more than one RFC-822 address, only the first is | |
40 | ;; returned. | |
41 | ;; | |
d1782bd8 KH |
42 | ;; If ALL is non-nil, that means return info about all the addresses |
43 | ;; that are found in ADDRESS. The value is a list of elements of | |
44 | ;; the form (FULL-NAME CANONICAL-ADDRESS), one per address. | |
45 | ;; | |
154b3e39 RS |
46 | ;; This code is more correct (and more heuristic) parser than the code in |
47 | ;; rfc822.el. And despite its size, it's fairly fast. | |
48 | ;; | |
72c0ae01 | 49 | ;; There are two main benefits: |
154b3e39 | 50 | ;; |
72c0ae01 | 51 | ;; 1. Higher probability of getting the correct full name for a human than |
154b3e39 | 52 | ;; any other package we know of. (On the other hand, it will cheerfully |
72c0ae01 ER |
53 | ;; mangle non-human names/comments.) |
54 | ;; 2. Address part is put in a canonical form. | |
154b3e39 RS |
55 | ;; |
56 | ;; The interface is not yet carved in stone; please give us suggestions. | |
57 | ;; | |
58 | ;; We have an extensive test-case collection of funny addresses if you want to | |
72c0ae01 | 59 | ;; work with the code. Developing this code requires frequent testing to |
154b3e39 RS |
60 | ;; make sure you're not breaking functionality. The test cases aren't included |
61 | ;; because they are over 100K. | |
62 | ;; | |
63 | ;; If you find an address that mail-extr fails on, please send it to the | |
64 | ;; maintainer along with what you think the correct results should be. We do | |
65 | ;; not consider it a bug if mail-extr mangles a comment that does not | |
66 | ;; correspond to a real human full name, although we would prefer that | |
67 | ;; mail-extr would return the comment as-is. | |
68 | ;; | |
72c0ae01 | 69 | ;; Features: |
154b3e39 | 70 | ;; |
72c0ae01 | 71 | ;; * Full name handling: |
154b3e39 | 72 | ;; |
72c0ae01 ER |
73 | ;; * knows where full names can be found in an address. |
74 | ;; * avoids using empty comments and quoted text. | |
75 | ;; * extracts full names from mailbox names. | |
76 | ;; * recognizes common formats for comments after a full name. | |
77 | ;; * puts a period and a space after each initial. | |
154b3e39 RS |
78 | ;; * understands & referring to the mailbox name, capitalized. |
79 | ;; * strips name prefixes like "Prof.", etc. | |
72c0ae01 ER |
80 | ;; * understands what characters can occur in names (not just letters). |
81 | ;; * figures out middle initial from mailbox name. | |
82 | ;; * removes funny nicknames. | |
83 | ;; * keeps suffixes such as Jr., Sr., III, etc. | |
84 | ;; * reorders "Last, First" type names. | |
154b3e39 | 85 | ;; |
72c0ae01 | 86 | ;; * Address handling: |
154b3e39 | 87 | ;; |
72c0ae01 ER |
88 | ;; * parses rfc822 quoted text, comments, and domain literals. |
89 | ;; * parses rfc822 multi-line headers. | |
90 | ;; * does something reasonable with rfc822 GROUP addresses. | |
91 | ;; * handles many rfc822 noncompliant and garbage addresses. | |
92 | ;; * canonicalizes addresses (after stripping comments/phrases outside <>). | |
93 | ;; * converts ! addresses into .UUCP and %-style addresses. | |
94 | ;; * converts rfc822 ROUTE addresses to %-style addresses. | |
95 | ;; * truncates %-style addresses at leftmost fully qualified domain name. | |
96 | ;; * handles local relative precedence of ! vs. % and @ (untested). | |
154b3e39 | 97 | ;; |
72c0ae01 ER |
98 | ;; It does almost no string creation. It primarily uses the built-in |
99 | ;; parsing routines with the appropriate syntax tables. This should | |
100 | ;; result in greater speed. | |
154b3e39 | 101 | ;; |
72c0ae01 | 102 | ;; TODO: |
154b3e39 | 103 | ;; |
72c0ae01 ER |
104 | ;; * handle all test cases. (This will take forever.) |
105 | ;; * software to pick the correct header to use (eg., "Senders-Name:"). | |
106 | ;; * multiple addresses in the "From:" header (almost all of the necessary | |
107 | ;; code is there). | |
108 | ;; * flag to not treat `,' as an address separator. (This is useful when | |
109 | ;; there is a "From:" header but no "Sender:" header, because then there | |
110 | ;; is only allowed to be one address.) | |
111 | ;; * mailbox name does not necessarily contain full name. | |
112 | ;; * fixing capitalization when it's all upper or lowercase. (Hard!) | |
113 | ;; * some of the domain literal handling is missing. (But I've never even | |
114 | ;; seen one of these in a mail address, so maybe no big deal.) | |
115 | ;; * arrange to have syntax tables byte-compiled. | |
116 | ;; * speed hacks. | |
117 | ;; * delete unused variables. | |
118 | ;; * arrange for testing with different relative precedences of ! vs. @ | |
119 | ;; and %. | |
72c0ae01 ER |
120 | ;; * insert documentation strings! |
121 | ;; * handle X.400-gatewayed addresses according to RFC 1148. | |
122 | ||
123 | ;;; Change Log: | |
124 | ;; | |
154b3e39 RS |
125 | ;; Thu Feb 17 17:57:33 1994 Jamie Zawinski (jwz@lucid.com) |
126 | ;; | |
127 | ;; * merged with jbw's latest version | |
128 | ;; | |
129 | ;; Wed Feb 9 21:56:27 1994 Jamie Zawinski (jwz@lucid.com) | |
130 | ;; | |
131 | ;; * high-bit chars in comments weren't treated as word syntax | |
132 | ;; | |
133 | ;; Sat Feb 5 03:13:40 1994 Jamie Zawinski (jwz@lucid.com) | |
134 | ;; | |
135 | ;; * call replace-match with fixed-case arg | |
136 | ;; | |
137 | ;; Thu Dec 16 21:56:45 1993 Jamie Zawinski (jwz@lucid.com) | |
138 | ;; | |
139 | ;; * some more cleanup, doc, added provide | |
140 | ;; | |
141 | ;; Tue Mar 23 21:23:18 1993 Joe Wells (jbw at csd.bu.edu) | |
142 | ;; | |
143 | ;; * Made mail-full-name-prefixes a user-customizable variable. | |
144 | ;; Allow passing the address as a buffer as well as as a string. | |
145 | ;; Allow [ and ] as name characters (Finnish character set). | |
146 | ;; | |
147 | ;; Mon Mar 22 21:20:56 1993 Joe Wells (jbw at bigbird.bu.edu) | |
148 | ;; | |
149 | ;; * Handle "null" addresses. Handle = used for spacing in mailbox | |
150 | ;; name. Fix bug in handling of ROUTE-ADDR-type addresses that are | |
151 | ;; missing their brackets. Handle uppercase "JR". Extract full | |
152 | ;; names from X.400 addresses encoded in RFC-822. Fix bug in | |
153 | ;; handling of multiple addresses where first has trailing comment. | |
154 | ;; Handle more kinds of telephone extension lead-ins. | |
155 | ;; | |
156 | ;; Mon Mar 22 20:16:57 1993 Joe Wells (jbw at bigbird.bu.edu) | |
157 | ;; | |
158 | ;; * Handle HZ encoding for embedding GB encoded chinese characters. | |
159 | ;; | |
160 | ;; Mon Mar 22 00:46:12 1993 Joe Wells (jbw at bigbird.bu.edu) | |
161 | ;; | |
162 | ;; * Fixed too broad matching of ham radio call signs. Fixed bug in | |
163 | ;; handling an unmatched ' in a name string. Enhanced recognition | |
164 | ;; of when . in the mailbox name terminates the name portion. | |
165 | ;; Narrowed conversion of . to space to only the necessary | |
166 | ;; situation. Deal with VMS's stupid date stamps. Handle a unique | |
167 | ;; way of introducing an alternate address. Fixed spacing bug I | |
168 | ;; introduced in switching last name order. Fixed bug in handling | |
169 | ;; address with ! and % but no @. Narrowed the cases in which | |
170 | ;; certain trailing words are discarded. | |
171 | ;; | |
172 | ;; Sun Mar 21 21:41:06 1993 Joe Wells (jbw at bigbird.bu.edu) | |
173 | ;; | |
174 | ;; * Fixed bugs in handling GROUP addresses. Certain words in the | |
175 | ;; middle of a name no longer terminate it. Handle LISTSERV list | |
176 | ;; names. Ignore comment field containing mailbox name. | |
177 | ;; | |
178 | ;; Sun Mar 21 14:39:38 1993 Joe Wells (jbw at bigbird.bu.edu) | |
179 | ;; | |
180 | ;; * Moved variant-method code back into main function. Handle | |
181 | ;; underscores as spaces in comments. Handle leading nickname. Add | |
182 | ;; flag to ignore single-word names. Other changes. | |
183 | ;; | |
184 | ;; Mon Feb 1 22:23:31 1993 Joe Wells (jbw at bigbird.bu.edu) | |
185 | ;; | |
186 | ;; * Added in changes by Rod Whitby and Jamie Zawinski. This | |
187 | ;; includes the flag mail-extr-guess-middle-initial and the fix for | |
6b07c06e RS |
188 | ;; handling multiple addresses correctly. (Whitby just changed |
189 | ;; a > to a <.) | |
154b3e39 | 190 | ;; |
72c0ae01 ER |
191 | ;; Mon Apr 6 23:59:09 1992 Joe Wells (jbw at bigbird.bu.edu) |
192 | ;; | |
193 | ;; * Cleaned up some more. Release version 1.0 to world. | |
194 | ;; | |
195 | ;; Sun Apr 5 19:39:08 1992 Joe Wells (jbw at bigbird.bu.edu) | |
196 | ;; | |
197 | ;; * Cleaned up full name extraction extensively. | |
198 | ;; | |
199 | ;; Sun Feb 2 14:45:24 1992 Joe Wells (jbw at bigbird.bu.edu) | |
200 | ;; | |
201 | ;; * Total rewrite. Integrated mail-canonicalize-address into | |
202 | ;; mail-extract-address-components. Now handles GROUP addresses more | |
203 | ;; or less correctly. Better handling of lots of different cases. | |
204 | ;; | |
205 | ;; Fri Jun 14 19:39:50 1991 | |
206 | ;; * Created. | |
207 | ||
208 | ;;; Code: | |
209 | \f | |
72c0ae01 | 210 | |
154b3e39 RS |
211 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; |
212 | ;; | |
213 | ;; User configuration variable definitions. | |
214 | ;; | |
215 | ||
216 | (defvar mail-extr-guess-middle-initial nil | |
217 | "*Whether to try to guess middle initial from mail address. | |
218 | If true, then when we see an address like \"John Smith <jqs@host.com>\" | |
219 | we will assume that \"John Q. Smith\" is the fellow's name.") | |
220 | ||
221 | (defvar mail-extr-ignore-single-names t | |
222 | "*Whether to ignore a name that is just a single word. | |
223 | If true, then when we see an address like \"Idiot <dumb@stupid.com>\" | |
224 | we will act as though we couldn't find a full name in the address.") | |
225 | ||
226 | ;; Matches a leading title that is not part of the name (does not | |
227 | ;; contribute to uniquely identifying the person). | |
228 | (defvar mail-extr-full-name-prefixes | |
229 | (purecopy | |
230 | "\\(Prof\\|D[Rr]\\|Mrs?\\|Rev\\|Rabbi\\|SysOp\\|LCDR\\)\\.?[ \t\n]") | |
231 | "*Matches prefixes to the full name that identify a person's position. | |
232 | These are stripped from the full name because they do not contribute to | |
233 | uniquely identifying the person.") | |
234 | ||
235 | (defvar mail-extr-@-binds-tighter-than-! nil | |
236 | "*Whether the local mail transport agent looks at ! before @.") | |
237 | ||
238 | (defvar mail-extr-mangle-uucp nil | |
239 | "*Whether to throw away information in UUCP addresses | |
240 | by translating things like \"foo!bar!baz@host\" into \"baz@bar.UUCP\".") | |
72c0ae01 ER |
241 | |
242 | ;;---------------------------------------------------------------------- | |
243 | ;; what orderings are meaningful????? | |
244 | ;;(defvar mail-operator-precedence-list '(?! ?% ?@)) | |
245 | ;; Right operand of a % or a @ must be a domain name, period. No other | |
246 | ;; operators allowed. Left operand of a @ is an address relative to that | |
247 | ;; site. | |
248 | ||
249 | ;; Left operand of a ! must be a domain name. Right operand is an | |
250 | ;; arbitrary address. | |
251 | ;;---------------------------------------------------------------------- | |
252 | ||
154b3e39 | 253 | \f |
72c0ae01 | 254 | |
154b3e39 RS |
255 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; |
256 | ;; | |
257 | ;; Constant definitions. | |
258 | ;; | |
259 | ||
260 | ;; Codes in | |
261 | ;; Names in ISO 8859-1 Name | |
262 | ;; ISO 10XXX ISO 8859-2 in | |
263 | ;; ISO 6937 ISO 10646 RFC Swedish | |
264 | ;; etc. Hex Oct 1345 TeX Split ASCII Description | |
265 | ;; --------- ---------- ---- --- ----- ----- ------------------------------- | |
266 | ;; %a E4 344 a: \"a ae { latin small a + diaeresis d | |
267 | ;; %o F6 366 o: \"o oe | latin small o + diaeresis v | |
268 | ;; @a E5 345 aa \oa aa } latin small a + ring above e | |
269 | ;; %u FC 374 u: \"u ue ~ latin small u + diaeresis | | |
270 | ;; /e E9 351 e' \'e ` latin small e + acute i | |
271 | ;; %A C4 304 A: \"A AE [ latin capital a + diaeresis D | |
272 | ;; %O D6 326 O: \"O OE \ latin capital o + diaeresis V | |
273 | ;; @A C5 305 AA \oA AA ] latin capital a + ring above E | |
274 | ;; %U DC 334 U: \"U UE ^ latin capital u + diaeresis \ | |
275 | ;; /E C9 311 E' \'E @ latin capital e + acute I | |
276 | ||
277 | ;; NOTE: @a and @A are not in ISO 8859-2 (the codes mentioned above invoke | |
278 | ;; /l and /L). Some of this data was retrieved from | |
279 | ;; listserv@jhuvm.hcf.jhu.edu. | |
72c0ae01 ER |
280 | |
281 | ;; Any character that can occur in a name, not counting characters that | |
154b3e39 RS |
282 | ;; separate parts of a multipart name (hyphen and period). |
283 | ;; Yes, there are weird people with digits in their names. | |
284 | ;; You will also notice the consideration for the | |
285 | ;; Swedish/Finnish/Norwegian character set. | |
154b3e39 | 286 | (defconst mail-extr-all-letters-but-separators |
fa25dbbc | 287 | (purecopy "][A-Za-z{|}'~0-9`\200-\377")) |
154b3e39 RS |
288 | |
289 | ;; Any character that can occur in a name in an RFC822 address including | |
290 | ;; the separator (hyphen and possibly period) for multipart names. | |
291 | ;; #### should . be in here? | |
292 | (defconst mail-extr-all-letters | |
293 | (purecopy (concat mail-extr-all-letters-but-separators "---"))) | |
294 | ||
295 | ;; Any character that can start a name. | |
296 | ;; Keep this set as minimal as possible. | |
fa25dbbc | 297 | (defconst mail-extr-first-letters (purecopy "A-Za-z\200-\377")) |
72c0ae01 ER |
298 | |
299 | ;; Any character that can end a name. | |
154b3e39 | 300 | ;; Keep this set as minimal as possible. |
fa25dbbc | 301 | (defconst mail-extr-last-letters (purecopy "A-Za-z\200-\377`'.")) |
72c0ae01 | 302 | |
154b3e39 RS |
303 | (defconst mail-extr-leading-garbage |
304 | (purecopy (format "[^%s]+" mail-extr-first-letters))) | |
72c0ae01 | 305 | |
154b3e39 RS |
306 | ;; (defconst mail-extr-non-name-chars |
307 | ;; (purecopy (concat "^" mail-extr-all-letters "."))) | |
308 | ;; (defconst mail-extr-non-begin-name-chars | |
309 | ;; (purecopy (concat "^" mail-extr-first-letters))) | |
310 | ;; (defconst mail-extr-non-end-name-chars | |
311 | ;; (purecopy (concat "^" mail-extr-last-letters))) | |
72c0ae01 | 312 | |
154b3e39 RS |
313 | ;; Matches an initial not followed by both a period and a space. |
314 | ;; (defconst mail-extr-bad-initials-pattern | |
315 | ;; (purecopy | |
316 | ;; (format "\\(\\([^%s]\\|\\`\\)[%s]\\)\\(\\.\\([^ ]\\)\\| \\|\\([^%s .]\\)\\|\\'\\)" | |
317 | ;; mail-extr-all-letters mail-extr-first-letters mail-extr-all-letters))) | |
72c0ae01 ER |
318 | |
319 | ;; Matches periods used instead of spaces. Must not match the period | |
320 | ;; following an initial. | |
154b3e39 RS |
321 | (defconst mail-extr-bad-dot-pattern |
322 | (purecopy | |
323 | (format "\\([%s][%s]\\)\\.+\\([%s]\\)" | |
324 | mail-extr-all-letters | |
325 | mail-extr-last-letters | |
326 | mail-extr-first-letters))) | |
72c0ae01 ER |
327 | |
328 | ;; Matches an embedded or leading nickname that should be removed. | |
154b3e39 RS |
329 | ;; (defconst mail-extr-nickname-pattern |
330 | ;; (purecopy | |
331 | ;; (format "\\([ .]\\|\\`\\)[\"'`\[\(]\\([ .%s]+\\)[\]\"'\)] " | |
332 | ;; mail-extr-all-letters))) | |
72c0ae01 ER |
333 | |
334 | ;; Matches the occurrence of a generational name suffix, and the last | |
154b3e39 RS |
335 | ;; character of the preceding name. This is important because we want to |
336 | ;; keep such suffixes: they help to uniquely identify the person. | |
337 | ;; *** Perhaps this should be a user-customizable variable. However, the | |
338 | ;; *** regular expression is fairly tricky to alter, so maybe not. | |
339 | (defconst mail-extr-full-name-suffix-pattern | |
340 | (purecopy | |
341 | (format | |
342 | "\\(,? ?\\([JjSs][Rr]\\.?\\|V?I+V?\\)\\)\\([^%s]\\([^%s]\\|\\'\\)\\|\\'\\)" | |
343 | mail-extr-all-letters mail-extr-all-letters))) | |
344 | ||
345 | (defconst mail-extr-roman-numeral-pattern (purecopy "V?I+V?\\b")) | |
72c0ae01 ER |
346 | |
347 | ;; Matches a trailing uppercase (with other characters possible) acronym. | |
348 | ;; Must not match a trailing uppercase last name or trailing initial | |
154b3e39 RS |
349 | (defconst mail-extr-weird-acronym-pattern |
350 | (purecopy "\\([A-Z]+[-_/]\\|[A-Z][A-Z][A-Z]?\\b\\)")) | |
72c0ae01 ER |
351 | |
352 | ;; Matches a mixed-case or lowercase name (not an initial). | |
154b3e39 RS |
353 | ;; #### Match Latin1 lower case letters here too? |
354 | ;; (defconst mail-extr-mixed-case-name-pattern | |
355 | ;; (purecopy | |
356 | ;; (format | |
357 | ;; "\\b\\([a-z][%s]*[%s]\\|[%s][%s]*[a-z][%s]*[%s]\\|[%s][%s]*[a-z]\\)" | |
358 | ;; mail-extr-all-letters mail-extr-last-letters | |
359 | ;; mail-extr-first-letters mail-extr-all-letters mail-extr-all-letters | |
360 | ;; mail-extr-last-letters mail-extr-first-letters mail-extr-all-letters))) | |
72c0ae01 ER |
361 | |
362 | ;; Matches a trailing alternative address. | |
154b3e39 RS |
363 | ;; #### Match Latin1 letters here too? |
364 | ;; #### Match _ before @ here too? | |
365 | (defconst mail-extr-alternative-address-pattern | |
366 | (purecopy "\\(aka *\\)?[a-zA-Z.]+[!@][a-zA-Z.]")) | |
72c0ae01 ER |
367 | |
368 | ;; Matches a variety of trailing comments not including comma-delimited | |
369 | ;; comments. | |
154b3e39 RS |
370 | (defconst mail-extr-trailing-comment-start-pattern |
371 | (purecopy " [-{]\\|--\\|[+@#></\;]")) | |
72c0ae01 ER |
372 | |
373 | ;; Matches a name (not an initial). | |
374 | ;; This doesn't force a word boundary at the end because sometimes a | |
375 | ;; comment is separated by a `-' with no preceding space. | |
154b3e39 RS |
376 | (defconst mail-extr-name-pattern |
377 | (purecopy (format "\\b[%s][%s]*[%s]" | |
378 | mail-extr-first-letters | |
379 | mail-extr-all-letters | |
380 | mail-extr-last-letters))) | |
72c0ae01 | 381 | |
154b3e39 RS |
382 | (defconst mail-extr-initial-pattern |
383 | (purecopy (format "\\b[%s]\\([. ]\\|\\b\\)" mail-extr-first-letters))) | |
72c0ae01 ER |
384 | |
385 | ;; Matches a single name before a comma. | |
154b3e39 RS |
386 | ;; (defconst mail-extr-last-name-first-pattern |
387 | ;; (purecopy (concat "\\`" mail-extr-name-pattern ","))) | |
72c0ae01 ER |
388 | |
389 | ;; Matches telephone extensions. | |
154b3e39 RS |
390 | (defconst mail-extr-telephone-extension-pattern |
391 | (purecopy | |
392 | "\\(\\([Ee]xt\\|\\|[Tt]ph\\|[Tt]el\\|[Xx]\\).?\\)? *\\+?[0-9][- 0-9]+")) | |
72c0ae01 ER |
393 | |
394 | ;; Matches ham radio call signs. | |
154b3e39 RS |
395 | ;; Help from: Mat Maessen N2NJZ <maessm@rpi.edu>, Mark Feit |
396 | ;; <mark@era.com>, Michael Covington <mcovingt@ai.uga.edu>. | |
397 | ;; Examples: DX504 DX515 K5MRU K8DHK KA9WGN KA9WGN KD3FU KD6EUI KD6HBW | |
398 | ;; KE9TV KF0NV N1API N3FU N3GZE N3IGS N4KCC N7IKQ N9HHU W4YHF W6ANK WA2SUH | |
399 | ;; WB7VZI N2NJZ NR3G KJ4KK AB4UM AL7NI KH6OH WN3KBT N4TMI W1A N0NZO | |
400 | (defconst mail-extr-ham-call-sign-pattern | |
401 | (purecopy "\\b\\(DX[0-9]+\\|[AKNW][A-Z]?[0-9][A-Z][A-Z]?[A-Z]?\\)")) | |
402 | ||
403 | ;; Possible trailing suffixes: "\\(/\\(KT\\|A[AEG]\\|[R0-9]\\)\\)?" | |
404 | ;; /KT == Temporary Technician (has CSC but not "real" license) | |
405 | ;; /AA == Temporary Advanced | |
406 | ;; /AE == Temporary Extra | |
407 | ;; /AG == Temporary General | |
408 | ;; /R == repeater | |
409 | ;; /# == stations operating out of home district | |
410 | ;; I don't include these in the regexp above because I can't imagine | |
411 | ;; anyone putting them with their name in an e-mail address. | |
72c0ae01 ER |
412 | |
413 | ;; Matches normal single-part name | |
154b3e39 RS |
414 | (defconst mail-extr-normal-name-pattern |
415 | (purecopy (format "\\b[%s][%s]+[%s]" | |
416 | mail-extr-first-letters | |
417 | mail-extr-all-letters-but-separators | |
418 | mail-extr-last-letters))) | |
419 | ||
420 | ;; Matches a single word name. | |
421 | ;; (defconst mail-extr-one-name-pattern | |
422 | ;; (purecopy (concat "\\`" mail-extr-normal-name-pattern "\\'"))) | |
423 | ||
72c0ae01 | 424 | ;; Matches normal two names with missing middle initial |
154b3e39 RS |
425 | ;; The first name is not allowed to have a hyphen because this can cause |
426 | ;; false matches where the "middle initial" is actually the first letter | |
427 | ;; of the second part of the first name. | |
428 | (defconst mail-extr-two-name-pattern | |
429 | (purecopy | |
430 | (concat "\\`\\(" mail-extr-normal-name-pattern | |
431 | "\\|" mail-extr-initial-pattern | |
432 | "\\) +\\(" mail-extr-name-pattern "\\)\\(,\\|\\'\\)"))) | |
433 | ||
434 | (defconst mail-extr-listserv-list-name-pattern | |
435 | (purecopy "Multiple recipients of list \\([-A-Z]+\\)")) | |
436 | ||
437 | (defconst mail-extr-stupid-vms-date-stamp-pattern | |
438 | (purecopy | |
439 | "[0-9][0-9]-[JFMASOND][aepuco][nbrylgptvc]-[0-9][0-9][0-9][0-9] [0-9]+ *")) | |
440 | ||
441 | ;;; HZ -- GB (PRC Chinese character encoding) in ASCII embedding protocol | |
442 | ;; | |
443 | ;; In ASCII mode, a byte is interpreted as an ASCII character, unless a '~' is | |
444 | ;; encountered. The character '~' is an escape character. By convention, it | |
445 | ;; must be immediately followed ONLY by '~', '{' or '\n' (<LF>), with the | |
446 | ;; following special meaning. | |
447 | ;; | |
448 | ;; o The escape sequence '~~' is interpreted as a '~'. | |
449 | ;; o The escape-to-GB sequence '~{' switches the mode from ASCII to GB. | |
450 | ;; o The escape sequence '~\n' is a line-continuation marker to be consumed | |
451 | ;; with no output produced. | |
452 | ;; | |
453 | ;; In GB mode, characters are interpreted two bytes at a time as (pure) GB | |
454 | ;; codes until the escape-from-GB code '~}' is read. This code switches the | |
455 | ;; mode from GB back to ASCII. (Note that the escape-from-GB code '~}' | |
456 | ;; ($7E7D) is outside the defined GB range.) | |
457 | (defconst mail-extr-hz-embedded-gb-encoded-chinese-pattern | |
458 | (purecopy "~{\\([^~].\\|~[^\}]\\)+~}")) | |
459 | ||
460 | ;; The leading optional lowercase letters are for a bastardized version of | |
461 | ;; the encoding, as is the optional nature of the final slash. | |
462 | (defconst mail-extr-x400-encoded-address-pattern | |
463 | (purecopy "[a-z]?[a-z]?\\(/[A-Za-z]+\\(\\.[A-Za-z]+\\)?=[^/]+\\)+/?\\'")) | |
464 | ||
465 | (defconst mail-extr-x400-encoded-address-field-pattern-format | |
466 | (purecopy "/%s=\\([^/]+\\)\\(/\\|\\'\\)")) | |
467 | ||
468 | (defconst mail-extr-x400-encoded-address-surname-pattern | |
469 | ;; S stands for Surname (family name). | |
470 | (purecopy | |
471 | (format mail-extr-x400-encoded-address-field-pattern-format "[Ss]"))) | |
472 | ||
473 | (defconst mail-extr-x400-encoded-address-given-name-pattern | |
474 | ;; G stands for Given name. | |
475 | (purecopy | |
476 | (format mail-extr-x400-encoded-address-field-pattern-format "[Gg]"))) | |
477 | ||
478 | (defconst mail-extr-x400-encoded-address-full-name-pattern | |
479 | ;; PN stands for Personal Name. When used it represents the combination | |
480 | ;; of the G and S fields. | |
481 | ;; "The one system I used having this field asked it with the prompt | |
482 | ;; `Personal Name'. But they mapped it into G and S on outgoing real | |
483 | ;; X.400 addresses. As they mapped G and S into PN on incoming..." | |
484 | (purecopy | |
485 | (format mail-extr-x400-encoded-address-field-pattern-format "[Pp][Nn]"))) | |
486 | ||
487 | \f | |
488 | ||
489 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; | |
490 | ;; | |
491 | ;; Syntax tables used for quick parsing. | |
492 | ;; | |
493 | ||
494 | (defconst mail-extr-address-syntax-table (make-syntax-table)) | |
495 | (defconst mail-extr-address-comment-syntax-table (make-syntax-table)) | |
496 | (defconst mail-extr-address-domain-literal-syntax-table (make-syntax-table)) | |
497 | (defconst mail-extr-address-text-comment-syntax-table (make-syntax-table)) | |
498 | (defconst mail-extr-address-text-syntax-table (make-syntax-table)) | |
72c0ae01 ER |
499 | (mapcar |
500 | (function | |
501 | (lambda (pair) | |
502 | (let ((syntax-table (symbol-value (car pair)))) | |
503 | (mapcar | |
504 | (function | |
505 | (lambda (item) | |
506 | (if (eq 2 (length item)) | |
154b3e39 | 507 | ;; modifying syntax of a single character |
72c0ae01 | 508 | (modify-syntax-entry (car item) (car (cdr item)) syntax-table) |
154b3e39 RS |
509 | ;; modifying syntax of a range of characters |
510 | (let ((char (nth 0 item)) | |
511 | (bound (nth 1 item)) | |
512 | (syntax (nth 2 item))) | |
72c0ae01 ER |
513 | (while (<= char bound) |
514 | (modify-syntax-entry char syntax syntax-table) | |
515 | (setq char (1+ char))))))) | |
516 | (cdr pair))))) | |
154b3e39 RS |
517 | '((mail-extr-address-syntax-table |
518 | (?\000 ?\037 "w") ;control characters | |
519 | (?\040 " ") ;SPC | |
520 | (?! ?~ "w") ;printable characters | |
521 | (?\177 "w") ;DEL | |
522 | (?\200 ?\377 "w") ;high-bit-on characters | |
523 | (?\240 " ") ;nobreakspace | |
72c0ae01 ER |
524 | (?\t " ") |
525 | (?\r " ") | |
526 | (?\n " ") | |
527 | (?\( ".") | |
528 | (?\) ".") | |
529 | (?< ".") | |
530 | (?> ".") | |
531 | (?@ ".") | |
532 | (?, ".") | |
533 | (?\; ".") | |
534 | (?: ".") | |
535 | (?\\ "\\") | |
536 | (?\" "\"") | |
537 | (?. ".") | |
538 | (?\[ ".") | |
539 | (?\] ".") | |
540 | ;; % and ! aren't RFC822 characters, but it is convenient to pretend | |
541 | (?% ".") | |
154b3e39 | 542 | (?! ".") ;; this needs to be word-constituent when not in .UUCP mode |
72c0ae01 | 543 | ) |
154b3e39 RS |
544 | (mail-extr-address-comment-syntax-table |
545 | (?\000 ?\377 "w") | |
546 | (?\040 " ") | |
547 | (?\240 " ") | |
548 | (?\t " ") | |
549 | (?\r " ") | |
550 | (?\n " ") | |
72c0ae01 ER |
551 | (?\( "\(\)") |
552 | (?\) "\)\(") | |
553 | (?\\ "\\")) | |
154b3e39 RS |
554 | (mail-extr-address-domain-literal-syntax-table |
555 | (?\000 ?\377 "w") | |
556 | (?\040 " ") | |
557 | (?\240 " ") | |
558 | (?\t " ") | |
559 | (?\r " ") | |
560 | (?\n " ") | |
72c0ae01 ER |
561 | (?\[ "\(\]") ;?????? |
562 | (?\] "\)\[") ;?????? | |
563 | (?\\ "\\")) | |
154b3e39 RS |
564 | (mail-extr-address-text-comment-syntax-table |
565 | (?\000 ?\377 "w") | |
566 | (?\040 " ") | |
567 | (?\240 " ") | |
568 | (?\t " ") | |
569 | (?\r " ") | |
570 | (?\n " ") | |
72c0ae01 ER |
571 | (?\( "\(\)") |
572 | (?\) "\)\(") | |
573 | (?\[ "\(\]") | |
574 | (?\] "\)\[") | |
575 | (?\{ "\(\}") | |
576 | (?\} "\)\{") | |
577 | (?\\ "\\") | |
578 | (?\" "\"") | |
579 | ;; (?\' "\)\`") | |
580 | ;; (?\` "\(\'") | |
581 | ) | |
154b3e39 RS |
582 | (mail-extr-address-text-syntax-table |
583 | (?\000 ?\177 ".") | |
584 | (?\200 ?\377 "w") | |
585 | (?\040 " ") | |
586 | (?\t " ") | |
587 | (?\r " ") | |
588 | (?\n " ") | |
72c0ae01 ER |
589 | (?A ?Z "w") |
590 | (?a ?z "w") | |
591 | (?- "w") | |
592 | (?\} "w") | |
593 | (?\{ "w") | |
594 | (?| "w") | |
595 | (?\' "w") | |
596 | (?~ "w") | |
597 | (?0 ?9 "w")) | |
598 | )) | |
599 | ||
600 | \f | |
154b3e39 RS |
601 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; |
602 | ;; | |
72c0ae01 | 603 | ;; Utility functions and macros. |
154b3e39 RS |
604 | ;; |
605 | ||
606 | (defmacro mail-extr-delete-char (n) | |
607 | ;; in v19, delete-char is compiled as a function call, but delete-region | |
608 | ;; is byte-coded, so it's much much faster. | |
609 | (list 'delete-region '(point) (list '+ '(point) n))) | |
610 | ||
611 | (defmacro mail-extr-skip-whitespace-forward () | |
612 | ;; v19 fn skip-syntax-forward is more tasteful, but not byte-coded. | |
613 | '(skip-chars-forward " \t\n\r\240")) | |
614 | ||
615 | (defmacro mail-extr-skip-whitespace-backward () | |
616 | ;; v19 fn skip-syntax-backward is more tasteful, but not byte-coded. | |
617 | '(skip-chars-backward " \t\n\r\240")) | |
618 | ||
72c0ae01 | 619 | |
154b3e39 | 620 | (defmacro mail-extr-undo-backslash-quoting (beg end) |
72c0ae01 ER |
621 | (`(save-excursion |
622 | (save-restriction | |
623 | (narrow-to-region (, beg) (, end)) | |
624 | (goto-char (point-min)) | |
625 | ;; undo \ quoting | |
154b3e39 RS |
626 | (while (search-forward "\\" nil t) |
627 | (mail-extr-delete-char -1) | |
628 | (or (eobp) | |
629 | (forward-char 1)) | |
72c0ae01 ER |
630 | ))))) |
631 | ||
154b3e39 | 632 | (defmacro mail-extr-nuke-char-at (pos) |
72c0ae01 ER |
633 | (` (save-excursion |
634 | (goto-char (, pos)) | |
154b3e39 RS |
635 | (mail-extr-delete-char 1) |
636 | (insert ?\ )))) | |
637 | ||
638 | (put 'mail-extr-nuke-outside-range | |
639 | 'edebug-form-spec '(symbolp &optional form form atom)) | |
640 | ||
641 | (defmacro mail-extr-nuke-outside-range (list-symbol | |
642 | beg-symbol end-symbol | |
643 | &optional no-replace) | |
644 | ;; LIST-SYMBOL names a variable holding a list of buffer positions | |
645 | ;; BEG-SYMBOL and END-SYMBOL name variables delimiting a range | |
646 | ;; Each element of LIST-SYMBOL which lies outside of the range is | |
647 | ;; deleted from the list. | |
648 | ;; Unless NO-REPLACE is true, at each of the positions in LIST-SYMBOL | |
649 | ;; which lie outside of the range, one character at that position is | |
650 | ;; replaced with a SPC. | |
651 | (or (memq no-replace '(t nil)) | |
29565a87 | 652 | (error "no-replace must be t or nil, evaluable at macroexpand-time")) |
154b3e39 RS |
653 | (` (let ((temp (, list-symbol)) |
654 | ch) | |
72c0ae01 | 655 | (while temp |
154b3e39 RS |
656 | (setq ch (car temp)) |
657 | (cond ((or (> ch (, end-symbol)) | |
658 | (< ch (, beg-symbol))) | |
659 | (,@ (if no-replace | |
660 | nil | |
661 | (` ((mail-extr-nuke-char-at ch))))) | |
72c0ae01 ER |
662 | (setcar temp nil))) |
663 | (setq temp (cdr temp))) | |
664 | (setq (, list-symbol) (delq nil (, list-symbol)))))) | |
665 | ||
154b3e39 RS |
666 | (defun mail-extr-demarkerize (marker) |
667 | ;; if arg is a marker, destroys the marker, then returns the old value. | |
668 | ;; otherwise returns the arg. | |
669 | (if (markerp marker) | |
670 | (let ((temp (marker-position marker))) | |
671 | (set-marker marker nil) | |
672 | temp) | |
673 | marker)) | |
674 | ||
675 | (defun mail-extr-markerize (pos) | |
676 | ;; coerces pos to a marker if non-nil. | |
677 | (if (or (markerp pos) (null pos)) | |
678 | pos | |
679 | (copy-marker pos))) | |
680 | ||
681 | (defmacro mail-extr-last (list) | |
682 | ;; Returns last element of LIST. | |
683 | ;; Could be a subst. | |
72c0ae01 ER |
684 | (` (let ((list (, list))) |
685 | (while (not (null (cdr list))) | |
686 | (setq list (cdr list))) | |
687 | (car list)))) | |
688 | ||
154b3e39 RS |
689 | (defmacro mail-extr-safe-move-sexp (arg) |
690 | ;; Safely skip over one balanced sexp, if there is one. Return t if success. | |
72c0ae01 ER |
691 | (` (condition-case error |
692 | (progn | |
16f45d1b | 693 | (goto-char (or (scan-sexps (point) (, arg)) (point))) |
72c0ae01 ER |
694 | t) |
695 | (error | |
154b3e39 | 696 | ;; #### kludge kludge kludge kludge kludge kludge kludge !!! |
72c0ae01 ER |
697 | (if (string-equal (nth 1 error) "Unbalanced parentheses") |
698 | nil | |
699 | (while t | |
700 | (signal (car error) (cdr error)))))))) | |
72c0ae01 | 701 | \f |
154b3e39 RS |
702 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; |
703 | ;; | |
72c0ae01 | 704 | ;; The main function to grind addresses |
154b3e39 RS |
705 | ;; |
706 | ||
707 | (defvar disable-initial-guessing-flag) ; dynamic assignment | |
708 | (defvar cbeg) ; dynamic assignment | |
709 | (defvar cend) ; dynamic assignment | |
72c0ae01 | 710 | |
154b3e39 | 711 | ;;;###autoload |
d1782bd8 KH |
712 | (defun mail-extract-address-components (address &optional all) |
713 | "Given an RFC-822 address ADDRESS, extract full name and canonical address. | |
154b3e39 RS |
714 | Returns a list of the form (FULL-NAME CANONICAL-ADDRESS). |
715 | If no name can be extracted, FULL-NAME will be nil. | |
d1782bd8 KH |
716 | |
717 | If the optional argument ALL is non-nil, then ADDRESS can contain zero | |
718 | or more recipients, separated by commas, and we return a list of | |
719 | the form ((FULL-NAME CANONICAL-ADDRESS) ...) with one element for | |
720 | each recipient. If ALL is nil, then if ADDRESS contains more than | |
721 | one recipients, all but the first is ignored. | |
722 | ||
154b3e39 RS |
723 | ADDRESS may be a string or a buffer. If it is a buffer, the visible |
724 | (narrowed) portion of the buffer will be interpreted as the address. | |
725 | (This feature exists so that the clever caller might be able to avoid | |
d1782bd8 | 726 | consing a string.)" |
154b3e39 RS |
727 | (let ((canonicalization-buffer (get-buffer-create " *canonical address*")) |
728 | (extraction-buffer (get-buffer-create " *extract address components*")) | |
d1782bd8 KH |
729 | value-list) |
730 | ||
72c0ae01 ER |
731 | (save-excursion |
732 | (set-buffer extraction-buffer) | |
154b3e39 | 733 | (fundamental-mode) |
e8a57935 | 734 | (buffer-disable-undo extraction-buffer) |
154b3e39 | 735 | (set-syntax-table mail-extr-address-syntax-table) |
72c0ae01 ER |
736 | (widen) |
737 | (erase-buffer) | |
738 | (setq case-fold-search nil) | |
739 | ||
740 | ;; Insert extra space at beginning to allow later replacement with < | |
741 | ;; without having to move markers. | |
154b3e39 RS |
742 | (insert ?\ ) |
743 | ||
744 | ;; Insert the address itself. | |
745 | (cond ((stringp address) | |
746 | (insert address)) | |
747 | ((bufferp address) | |
748 | (insert-buffer-substring address)) | |
749 | (t | |
29565a87 | 750 | (error "Invalid address: %s" address))) |
03007ccc RS |
751 | |
752 | (set-text-properties (point-min) (point-max) nil) | |
d1782bd8 KH |
753 | |
754 | (save-excursion | |
755 | (set-buffer canonicalization-buffer) | |
756 | (fundamental-mode) | |
757 | (buffer-disable-undo canonicalization-buffer) | |
758 | (set-syntax-table mail-extr-address-syntax-table) | |
759 | (setq case-fold-search nil)) | |
760 | ||
72c0ae01 | 761 | |
72c0ae01 ER |
762 | ;; Unfold multiple lines. |
763 | (goto-char (point-min)) | |
764 | (while (re-search-forward "\\([^\\]\\(\\\\\\\\\\)*\\)\n[ \t]" nil t) | |
765 | (replace-match "\\1 " t)) | |
766 | ||
d1782bd8 KH |
767 | ;; Loop over addresses until we have as many as we want. |
768 | (while (and (or all (null value-list)) | |
769 | (progn (goto-char (point-min)) | |
770 | (skip-chars-forward " \t") | |
771 | (not (eobp)))) | |
772 | (let (char | |
773 | end-of-address | |
774 | <-pos >-pos @-pos :-pos comma-pos !-pos %-pos \;-pos | |
775 | group-:-pos group-\;-pos route-addr-:-pos | |
776 | record-pos-symbol | |
777 | first-real-pos last-real-pos | |
778 | phrase-beg phrase-end | |
779 | cbeg cend ; dynamically set from -voodoo | |
780 | quote-beg quote-end | |
781 | atom-beg atom-end | |
782 | mbox-beg mbox-end | |
783 | \.-ends-name | |
784 | temp | |
785 | ;; name-suffix | |
786 | fi mi li ; first, middle, last initial | |
787 | saved-%-pos saved-!-pos saved-@-pos | |
788 | domain-pos \.-pos insert-point | |
789 | ;; mailbox-name-processed-flag | |
790 | disable-initial-guessing-flag) ; dynamically set from -voodoo | |
791 | ||
792 | (goto-char (point-min)) | |
793 | ||
794 | ;; Insert extra space at beginning to allow later replacement with < | |
795 | ;; without having to move markers. | |
796 | (or (eq (following-char) ?\ ) | |
797 | (insert ?\ )) | |
798 | ||
799 | ;; First pass grabs useful information about address. | |
800 | (while (progn | |
801 | (mail-extr-skip-whitespace-forward) | |
802 | (not (eobp))) | |
803 | (setq char (char-after (point))) | |
804 | (or first-real-pos | |
805 | (if (not (eq char ?\()) | |
806 | (setq first-real-pos (point)))) | |
807 | (cond | |
808 | ;; comment | |
809 | ((eq char ?\() | |
810 | (set-syntax-table mail-extr-address-comment-syntax-table) | |
811 | ;; only record the first non-empty comment's position | |
812 | (if (and (not cbeg) | |
813 | (save-excursion | |
814 | (forward-char 1) | |
815 | (mail-extr-skip-whitespace-forward) | |
816 | (not (eq ?\) (char-after (point)))))) | |
817 | (setq cbeg (point))) | |
818 | ;; TODO: don't record if unbalanced | |
819 | (or (mail-extr-safe-move-sexp 1) | |
820 | (forward-char 1)) | |
821 | (set-syntax-table mail-extr-address-syntax-table) | |
822 | (if (and cbeg | |
823 | (not cend)) | |
824 | (setq cend (point)))) | |
825 | ;; quoted text | |
826 | ((eq char ?\") | |
827 | ;; only record the first non-empty quote's position | |
828 | (if (and (not quote-beg) | |
829 | (save-excursion | |
830 | (forward-char 1) | |
831 | (mail-extr-skip-whitespace-forward) | |
832 | (not (eq ?\" (char-after (point)))))) | |
833 | (setq quote-beg (point))) | |
834 | ;; TODO: don't record if unbalanced | |
835 | (or (mail-extr-safe-move-sexp 1) | |
836 | (forward-char 1)) | |
837 | (if (and quote-beg | |
838 | (not quote-end)) | |
839 | (setq quote-end (point)))) | |
840 | ;; domain literals | |
841 | ((eq char ?\[) | |
842 | (set-syntax-table mail-extr-address-domain-literal-syntax-table) | |
843 | (or (mail-extr-safe-move-sexp 1) | |
844 | (forward-char 1)) | |
845 | (set-syntax-table mail-extr-address-syntax-table)) | |
846 | ;; commas delimit addresses when outside < > pairs. | |
847 | ((and (eq char ?,) | |
848 | (or (and (null <-pos) | |
849 | ;; Handle ROUTE-ADDR address that is missing its <. | |
850 | (not (eq ?@ (char-after (1+ (point)))))) | |
851 | (and >-pos | |
852 | ;; handle weird munged addresses | |
853 | ;; BUG FIX: This test was reversed. Thanks to the | |
854 | ;; brilliant Rod Whitby <rwhitby@research.canon.oz.au> | |
855 | ;; for discovering this! | |
856 | (< (mail-extr-last <-pos) (car >-pos))))) | |
857 | ;; The argument contains more than one address. | |
858 | ;; Temporarily hide everything after this one. | |
859 | (setq end-of-address (copy-marker (1+ (point)))) | |
860 | (narrow-to-region (point-min) (1+ (point))) | |
861 | (mail-extr-delete-char 1) | |
862 | (setq char ?\() ; HAVE I NO SHAME?? | |
863 | ) | |
864 | ;; record the position of various interesting chars, determine | |
865 | ;; legality later. | |
866 | ((setq record-pos-symbol | |
867 | (cdr (assq char | |
868 | '((?< . <-pos) (?> . >-pos) (?@ . @-pos) | |
869 | (?: . :-pos) (?, . comma-pos) (?! . !-pos) | |
870 | (?% . %-pos) (?\; . \;-pos))))) | |
871 | (set record-pos-symbol | |
872 | (cons (point) (symbol-value record-pos-symbol))) | |
72c0ae01 | 873 | (forward-char 1)) |
d1782bd8 | 874 | ((eq char ?.) |
72c0ae01 | 875 | (forward-char 1)) |
d1782bd8 KH |
876 | ((memq char '( |
877 | ;; comment terminator illegal | |
878 | ?\) | |
879 | ;; domain literal terminator illegal | |
880 | ?\] | |
881 | ;; \ allowed only within quoted strings, | |
882 | ;; domain literals, and comments | |
883 | ?\\ | |
884 | )) | |
885 | (mail-extr-nuke-char-at (point)) | |
72c0ae01 | 886 | (forward-char 1)) |
d1782bd8 KH |
887 | (t |
888 | (forward-word 1))) | |
889 | (or (eq char ?\() | |
890 | ;; At the end of first address of a multiple address header. | |
891 | (and (eq char ?,) | |
892 | (eobp)) | |
893 | (setq last-real-pos (point)))) | |
894 | ||
895 | ;; Use only the leftmost <, if any. Replace all others with spaces. | |
896 | (while (cdr <-pos) | |
897 | (mail-extr-nuke-char-at (car <-pos)) | |
898 | (setq <-pos (cdr <-pos))) | |
899 | ||
900 | ;; Use only the rightmost >, if any. Replace all others with spaces. | |
901 | (while (cdr >-pos) | |
902 | (mail-extr-nuke-char-at (nth 1 >-pos)) | |
903 | (setcdr >-pos (nthcdr 2 >-pos))) | |
904 | ||
905 | ;; If multiple @s and a :, but no < and >, insert around buffer. | |
906 | ;; Example: @foo.bar.dom,@xxx.yyy.zzz:mailbox@aaa.bbb.ccc | |
907 | ;; This commonly happens on the UUCP "From " line. Ugh. | |
908 | (cond ((and (> (length @-pos) 1) | |
909 | (eq 1 (length :-pos)) ;TODO: check if between last two @s | |
910 | (not \;-pos) | |
911 | (not <-pos)) | |
912 | (goto-char (point-min)) | |
154b3e39 | 913 | (mail-extr-delete-char 1) |
d1782bd8 KH |
914 | (setq <-pos (list (point))) |
915 | (insert ?<))) | |
916 | ||
917 | ;; If < but no >, insert > in rightmost possible position | |
918 | (cond ((and <-pos | |
919 | (null >-pos)) | |
920 | (goto-char (point-max)) | |
921 | (setq >-pos (list (point))) | |
922 | (insert ?>))) | |
923 | ||
924 | ;; If > but no <, replace > with space. | |
925 | (cond ((and >-pos | |
926 | (null <-pos)) | |
927 | (mail-extr-nuke-char-at (car >-pos)) | |
928 | (setq >-pos nil))) | |
929 | ||
930 | ;; Turn >-pos and <-pos into non-lists | |
931 | (setq >-pos (car >-pos) | |
932 | <-pos (car <-pos)) | |
933 | ||
934 | ;; Trim other punctuation lists of items outside < > pair to handle | |
935 | ;; stupid MTAs. | |
936 | (cond (<-pos ; don't need to check >-pos also | |
937 | ;; handle bozo software that violates RFC 822 by sticking | |
938 | ;; punctuation marks outside of a < > pair | |
939 | (mail-extr-nuke-outside-range @-pos <-pos >-pos t) | |
940 | ;; RFC 822 says nothing about these two outside < >, but | |
941 | ;; remove those positions from the lists to make things | |
942 | ;; easier. | |
943 | (mail-extr-nuke-outside-range !-pos <-pos >-pos t) | |
944 | (mail-extr-nuke-outside-range %-pos <-pos >-pos t))) | |
945 | ||
946 | ;; Check for : that indicates GROUP list and for : part of | |
947 | ;; ROUTE-ADDR spec. | |
948 | ;; Can't possibly be more than two :. Nuke any extra. | |
949 | (while :-pos | |
950 | (setq temp (car :-pos) | |
951 | :-pos (cdr :-pos)) | |
952 | (cond ((and <-pos >-pos | |
953 | (> temp <-pos) | |
954 | (< temp >-pos)) | |
955 | (if (or route-addr-:-pos | |
956 | (< (length @-pos) 2) | |
957 | (> temp (car @-pos)) | |
958 | (< temp (nth 1 @-pos))) | |
959 | (mail-extr-nuke-char-at temp) | |
960 | (setq route-addr-:-pos temp))) | |
961 | ((or (not <-pos) | |
962 | (and <-pos | |
963 | (< temp <-pos))) | |
964 | (setq group-:-pos temp)))) | |
965 | ||
966 | ;; Nuke any ; that is in or to the left of a < > pair or to the left | |
967 | ;; of a GROUP starting :. Also, there may only be one ;. | |
968 | (while \;-pos | |
969 | (setq temp (car \;-pos) | |
970 | \;-pos (cdr \;-pos)) | |
971 | (cond ((and <-pos >-pos | |
972 | (> temp <-pos) | |
973 | (< temp >-pos)) | |
974 | (mail-extr-nuke-char-at temp)) | |
975 | ((and (or (not group-:-pos) | |
976 | (> temp group-:-pos)) | |
977 | (not group-\;-pos)) | |
978 | (setq group-\;-pos temp)))) | |
979 | ||
980 | ;; Nuke unmatched GROUP syntax characters. | |
981 | (cond ((and group-:-pos (not group-\;-pos)) | |
982 | ;; *** Do I really need to erase it? | |
983 | (mail-extr-nuke-char-at group-:-pos) | |
984 | (setq group-:-pos nil))) | |
985 | (cond ((and group-\;-pos (not group-:-pos)) | |
986 | ;; *** Do I really need to erase it? | |
987 | (mail-extr-nuke-char-at group-\;-pos) | |
988 | (setq group-\;-pos nil))) | |
989 | ||
990 | ;; Handle junk like ";@host.company.dom" that sendmail adds. | |
991 | ;; **** should I remember comment positions? | |
992 | (cond | |
993 | (group-\;-pos | |
994 | ;; this is fine for now | |
995 | (mail-extr-nuke-outside-range !-pos group-:-pos group-\;-pos t) | |
996 | (mail-extr-nuke-outside-range @-pos group-:-pos group-\;-pos t) | |
997 | (mail-extr-nuke-outside-range %-pos group-:-pos group-\;-pos t) | |
998 | (mail-extr-nuke-outside-range comma-pos group-:-pos group-\;-pos t) | |
999 | (and last-real-pos | |
1000 | (> last-real-pos (1+ group-\;-pos)) | |
1001 | (setq last-real-pos (1+ group-\;-pos))) | |
1002 | ;; *** This may be wrong: | |
1003 | (and cend | |
1004 | (> cend group-\;-pos) | |
1005 | (setq cend nil | |
1006 | cbeg nil)) | |
1007 | (and quote-end | |
1008 | (> quote-end group-\;-pos) | |
1009 | (setq quote-end nil | |
1010 | quote-beg nil)) | |
1011 | ;; This was both wrong and unnecessary: | |
1012 | ;;(narrow-to-region (point-min) group-\;-pos) | |
1013 | ||
1014 | ;; *** The entire handling of GROUP addresses seems rather lame. | |
1015 | ;; *** It deserves a complete rethink, except that these addresses | |
1016 | ;; *** are hardly ever seen. | |
1017 | )) | |
1018 | ||
1019 | ;; Any commas must be between < and : of ROUTE-ADDR. Nuke any | |
1020 | ;; others. | |
1021 | ;; Hell, go ahead an nuke all of the commas. | |
1022 | ;; **** This will cause problems when we start handling commas in | |
1023 | ;; the PHRASE part .... no it won't ... yes it will ... ????? | |
1024 | (mail-extr-nuke-outside-range comma-pos 1 1) | |
1025 | ||
1026 | ;; can only have multiple @s inside < >. The fact that some MTAs | |
1027 | ;; put de-bracketed ROUTE-ADDRs in the UUCP-style "From " line is | |
1028 | ;; handled above. | |
1029 | ||
1030 | ;; Locate PHRASE part of ROUTE-ADDR. | |
1031 | (cond (<-pos | |
1032 | (goto-char <-pos) | |
154b3e39 | 1033 | (mail-extr-skip-whitespace-backward) |
d1782bd8 KH |
1034 | (setq phrase-end (point)) |
1035 | (goto-char (or ;;group-:-pos | |
1036 | (point-min))) | |
1037 | (mail-extr-skip-whitespace-forward) | |
1038 | (if (< (point) phrase-end) | |
1039 | (setq phrase-beg (point)) | |
1040 | (setq phrase-end nil)))) | |
1041 | ||
1042 | ;; handle ROUTE-ADDRS with real ROUTEs. | |
1043 | ;; If there are multiple @s, then we assume ROUTE-ADDR syntax, and | |
1044 | ;; any % or ! must be semantically meaningless. | |
1045 | ;; TODO: do this processing into canonicalization buffer | |
1046 | (cond (route-addr-:-pos | |
1047 | (setq !-pos nil | |
1048 | %-pos nil | |
1049 | >-pos (copy-marker >-pos) | |
1050 | route-addr-:-pos (copy-marker route-addr-:-pos)) | |
1051 | (goto-char >-pos) | |
1052 | (insert-before-markers ?X) | |
1053 | (goto-char (car @-pos)) | |
1054 | (while (setq @-pos (cdr @-pos)) | |
154b3e39 | 1055 | (mail-extr-delete-char 1) |
d1782bd8 KH |
1056 | (setq %-pos (cons (point-marker) %-pos)) |
1057 | (insert "%") | |
1058 | (goto-char (1- >-pos)) | |
1059 | (save-excursion | |
1060 | (insert-buffer-substring extraction-buffer | |
1061 | (car @-pos) route-addr-:-pos) | |
1062 | (delete-region (car @-pos) route-addr-:-pos)) | |
1063 | (or (cdr @-pos) | |
1064 | (setq saved-@-pos (list (point))))) | |
1065 | (setq @-pos saved-@-pos) | |
1066 | (goto-char >-pos) | |
1067 | (mail-extr-delete-char -1) | |
1068 | (mail-extr-nuke-char-at route-addr-:-pos) | |
1069 | (mail-extr-demarkerize route-addr-:-pos) | |
1070 | (setq route-addr-:-pos nil | |
1071 | >-pos (mail-extr-demarkerize >-pos) | |
1072 | %-pos (mapcar 'mail-extr-demarkerize %-pos)))) | |
1073 | ||
1074 | ;; de-listify @-pos | |
1075 | (setq @-pos (car @-pos)) | |
1076 | ||
1077 | ;; TODO: remove comments in the middle of an address | |
1078 | ||
1079 | (save-excursion | |
1080 | (set-buffer canonicalization-buffer) | |
1081 | ||
1082 | (widen) | |
1083 | (erase-buffer) | |
1084 | (insert-buffer-substring extraction-buffer) | |
1085 | ||
1086 | (if <-pos | |
1087 | (narrow-to-region (progn | |
1088 | (goto-char (1+ <-pos)) | |
1089 | (mail-extr-skip-whitespace-forward) | |
1090 | (point)) | |
1091 | >-pos) | |
1092 | (if (and first-real-pos last-real-pos) | |
1093 | (narrow-to-region first-real-pos last-real-pos) | |
1094 | ;; ****** Oh no! What if the address is completely empty! | |
1095 | ;; *** Is this correct? | |
1096 | (narrow-to-region (point-max) (point-max)) | |
1097 | )) | |
1098 | ||
1099 | (and @-pos %-pos | |
1100 | (mail-extr-nuke-outside-range %-pos (point-min) @-pos)) | |
1101 | (and %-pos !-pos | |
1102 | (mail-extr-nuke-outside-range !-pos (point-min) (car %-pos))) | |
1103 | (and @-pos !-pos (not %-pos) | |
1104 | (mail-extr-nuke-outside-range !-pos (point-min) @-pos)) | |
1105 | ||
1106 | ;; Error condition:?? (and %-pos (not @-pos)) | |
1107 | ||
1108 | ;; WARNING: THIS CODE IS DUPLICATED BELOW. | |
1109 | (cond ((and %-pos | |
1110 | (not @-pos)) | |
1111 | (goto-char (car %-pos)) | |
1112 | (mail-extr-delete-char 1) | |
1113 | (setq @-pos (point)) | |
1114 | (insert "@") | |
1115 | (setq %-pos (cdr %-pos)))) | |
1116 | ||
1117 | (if mail-extr-mangle-uucp | |
1118 | (cond (!-pos | |
1119 | ;; **** I don't understand this save-restriction and the | |
1120 | ;; narrow-to-region inside it. Why did I do that? | |
1121 | (save-restriction | |
1122 | (cond ((and @-pos | |
1123 | mail-extr-@-binds-tighter-than-!) | |
1124 | (goto-char @-pos) | |
1125 | (setq %-pos (cons (point) %-pos) | |
1126 | @-pos nil) | |
1127 | (mail-extr-delete-char 1) | |
1128 | (insert "%") | |
1129 | (setq insert-point (point-max))) | |
1130 | (mail-extr-@-binds-tighter-than-! | |
1131 | (setq insert-point (point-max))) | |
1132 | (%-pos | |
1133 | (setq insert-point (mail-extr-last %-pos) | |
1134 | saved-%-pos (mapcar 'mail-extr-markerize %-pos) | |
1135 | %-pos nil | |
1136 | @-pos (mail-extr-markerize @-pos))) | |
1137 | (@-pos | |
1138 | (setq insert-point @-pos) | |
1139 | (setq @-pos (mail-extr-markerize @-pos))) | |
1140 | (t | |
1141 | (setq insert-point (point-max)))) | |
1142 | (narrow-to-region (point-min) insert-point) | |
1143 | (setq saved-!-pos (car !-pos)) | |
1144 | (while !-pos | |
1145 | (goto-char (point-max)) | |
1146 | (cond ((and (not @-pos) | |
1147 | (not (cdr !-pos))) | |
1148 | (setq @-pos (point)) | |
1149 | (insert-before-markers "@ ")) | |
1150 | (t | |
1151 | (setq %-pos (cons (point) %-pos)) | |
1152 | (insert-before-markers "% "))) | |
1153 | (backward-char 1) | |
1154 | (insert-buffer-substring | |
1155 | (current-buffer) | |
1156 | (if (nth 1 !-pos) | |
1157 | (1+ (nth 1 !-pos)) | |
1158 | (point-min)) | |
1159 | (car !-pos)) | |
1160 | (mail-extr-delete-char 1) | |
1161 | (or (save-excursion | |
1162 | (mail-extr-safe-move-sexp -1) | |
1163 | (mail-extr-skip-whitespace-backward) | |
1164 | (eq ?. (preceding-char))) | |
1165 | (insert-before-markers | |
1166 | (if (save-excursion | |
1167 | (mail-extr-skip-whitespace-backward) | |
1168 | (eq ?. (preceding-char))) | |
1169 | "" | |
1170 | ".") | |
1171 | "uucp")) | |
1172 | (setq !-pos (cdr !-pos)))) | |
1173 | (and saved-%-pos | |
1174 | (setq %-pos (append (mapcar 'mail-extr-demarkerize | |
1175 | saved-%-pos) | |
1176 | %-pos))) | |
1177 | (setq @-pos (mail-extr-demarkerize @-pos)) | |
1178 | (narrow-to-region (1+ saved-!-pos) (point-max))))) | |
1179 | ||
1180 | ;; WARNING: THIS CODE IS DUPLICATED ABOVE. | |
1181 | (cond ((and %-pos | |
1182 | (not @-pos)) | |
1183 | (goto-char (car %-pos)) | |
1184 | (mail-extr-delete-char 1) | |
1185 | (setq @-pos (point)) | |
1186 | (insert "@") | |
1187 | (setq %-pos (cdr %-pos)))) | |
1188 | ||
1189 | (setq %-pos (nreverse %-pos)) | |
1190 | (cond (%-pos ; implies @-pos valid | |
1191 | (setq temp %-pos) | |
1192 | (catch 'truncated | |
1193 | (while temp | |
1194 | (goto-char (or (nth 1 temp) | |
1195 | @-pos)) | |
1196 | (mail-extr-skip-whitespace-backward) | |
1197 | (save-excursion | |
1198 | (mail-extr-safe-move-sexp -1) | |
1199 | (setq domain-pos (point)) | |
1200 | (mail-extr-skip-whitespace-backward) | |
1201 | (setq \.-pos (eq ?. (preceding-char)))) | |
1202 | (cond ((and \.-pos | |
1203 | ;; #### string consing | |
1204 | (let ((s (intern-soft | |
1205 | (buffer-substring domain-pos (point)) | |
1206 | mail-extr-all-top-level-domains))) | |
1207 | (and s (get s 'domain-name)))) | |
1208 | (narrow-to-region (point-min) (point)) | |
1209 | (goto-char (car temp)) | |
1210 | (mail-extr-delete-char 1) | |
1211 | (setq @-pos (point)) | |
1212 | (setcdr temp nil) | |
1213 | (setq %-pos (delq @-pos %-pos)) | |
1214 | (insert "@") | |
1215 | (throw 'truncated t))) | |
1216 | (setq temp (cdr temp)))))) | |
1217 | (setq mbox-beg (point-min) | |
1218 | mbox-end (if %-pos (car %-pos) | |
1219 | (or @-pos | |
1220 | (point-max))))) | |
1221 | ||
1222 | ;; Done canonicalizing address. | |
1223 | ;; We are now back in extraction-buffer. | |
1224 | ||
1225 | ;; Decide what part of the address to search to find the full name. | |
1226 | (cond ( | |
1227 | ;; Example: "First M. Last" <fml@foo.bar.dom> | |
1228 | (and phrase-beg | |
1229 | (eq quote-beg phrase-beg) | |
1230 | (<= quote-end phrase-end)) | |
1231 | (narrow-to-region (1+ quote-beg) (1- quote-end)) | |
1232 | (mail-extr-undo-backslash-quoting (point-min) (point-max))) | |
1233 | ||
1234 | ;; Example: First Last <fml@foo.bar.dom> | |
1235 | (phrase-beg | |
1236 | (narrow-to-region phrase-beg phrase-end)) | |
1237 | ||
1238 | ;; Example: fml@foo.bar.dom (First M. Last) | |
1239 | (cbeg | |
1240 | (narrow-to-region (1+ cbeg) (1- cend)) | |
1241 | (mail-extr-undo-backslash-quoting (point-min) (point-max)) | |
1242 | ||
1243 | ;; Deal with spacing problems | |
1244 | (goto-char (point-min)) | |
1245 | ;;; (cond ((not (search-forward " " nil t)) | |
1246 | ;;; (goto-char (point-min)) | |
1247 | ;;; (cond ((search-forward "_" nil t) | |
1248 | ;;; ;; Handle the *idiotic* use of underlines as spaces. | |
1249 | ;;; ;; Example: fml@foo.bar.dom (First_M._Last) | |
1250 | ;;; (goto-char (point-min)) | |
1251 | ;;; (while (search-forward "_" nil t) | |
1252 | ;;; (replace-match " " t))) | |
1253 | ;;; ((search-forward "." nil t) | |
1254 | ;;; ;; Fix . used as space | |
1255 | ;;; ;; Example: danj1@cb.att.com (daniel.jacobson) | |
1256 | ;;; (goto-char (point-min)) | |
1257 | ;;; (while (re-search-forward mail-extr-bad-dot-pattern nil t) | |
1258 | ;;; (replace-match "\\1 \\2" t)))))) | |
154b3e39 | 1259 | ) |
d1782bd8 KH |
1260 | |
1261 | ;; Otherwise we try to get the name from the mailbox portion | |
1262 | ;; of the address. | |
1263 | ;; Example: First_M_Last@foo.bar.dom | |
72c0ae01 | 1264 | (t |
d1782bd8 KH |
1265 | ;; *** Work in canon buffer instead? No, can't. Hmm. |
1266 | (goto-char (point-max)) | |
1267 | (narrow-to-region (point) (point)) | |
1268 | (insert-buffer-substring canonicalization-buffer | |
1269 | mbox-beg mbox-end) | |
1270 | (goto-char (point-min)) | |
1271 | ||
1272 | ;; Example: First_Last.XXX@foo.bar.dom | |
1273 | (setq \.-ends-name (re-search-forward "[_0-9]" nil t)) | |
1274 | ||
1275 | (goto-char (point-min)) | |
1276 | ||
1277 | (if (not mail-extr-mangle-uucp) | |
1278 | (modify-syntax-entry ?! "w" (syntax-table))) | |
1279 | ||
1280 | (while (progn | |
1281 | (mail-extr-skip-whitespace-forward) | |
1282 | (not (eobp))) | |
1283 | (setq char (char-after (point))) | |
154b3e39 | 1284 | (cond |
d1782bd8 KH |
1285 | ((eq char ?\") |
1286 | (setq quote-beg (point)) | |
1287 | (or (mail-extr-safe-move-sexp 1) | |
1288 | ;; TODO: handle this error condition!!!!! | |
1289 | (forward-char 1)) | |
1290 | ;; take into account deletions | |
1291 | (setq quote-end (- (point) 2)) | |
1292 | (save-excursion | |
1293 | (backward-char 1) | |
1294 | (mail-extr-delete-char 1) | |
1295 | (goto-char quote-beg) | |
1296 | (or (eobp) | |
1297 | (mail-extr-delete-char 1))) | |
1298 | (mail-extr-undo-backslash-quoting quote-beg quote-end) | |
1299 | (or (eq ?\ (char-after (point))) | |
1300 | (insert " ")) | |
1301 | ;; (setq mailbox-name-processed-flag t) | |
1302 | (setq \.-ends-name t)) | |
1303 | ((eq char ?.) | |
1304 | (if (memq (char-after (1+ (point))) '(?_ ?=)) | |
1305 | (progn | |
1306 | (forward-char 1) | |
1307 | (mail-extr-delete-char 1) | |
1308 | (insert ?\ )) | |
1309 | (if \.-ends-name | |
1310 | (narrow-to-region (point-min) (point)) | |
1311 | (mail-extr-delete-char 1) | |
1312 | (insert " "))) | |
1313 | ;; (setq mailbox-name-processed-flag t) | |
1314 | ) | |
1315 | ((memq (char-syntax char) '(?. ?\\)) | |
1316 | (mail-extr-delete-char 1) | |
1317 | (insert " ") | |
1318 | ;; (setq mailbox-name-processed-flag t) | |
154b3e39 | 1319 | ) |
154b3e39 | 1320 | (t |
d1782bd8 KH |
1321 | (setq atom-beg (point)) |
1322 | (forward-word 1) | |
1323 | (setq atom-end (point)) | |
1324 | (goto-char atom-beg) | |
1325 | (save-restriction | |
1326 | (narrow-to-region atom-beg atom-end) | |
1327 | (cond | |
1328 | ||
1329 | ;; Handle X.400 addresses encoded in RFC-822. | |
1330 | ;; *** Shit! This has to handle the case where it is | |
1331 | ;; *** embedded in a quote too! | |
1332 | ;; *** Shit! The input is being broken up into atoms | |
1333 | ;; *** by periods! | |
1334 | ((looking-at mail-extr-x400-encoded-address-pattern) | |
1335 | ||
1336 | ;; Copy the contents of the individual fields that | |
1337 | ;; might hold name data to the beginning. | |
1338 | (mapcar | |
1339 | (function | |
1340 | (lambda (field-pattern) | |
1341 | (cond | |
1342 | ((save-excursion | |
1343 | (re-search-forward field-pattern nil t)) | |
1344 | (insert-buffer-substring (current-buffer) | |
1345 | (match-beginning 1) | |
1346 | (match-end 1)) | |
1347 | (insert " "))))) | |
1348 | (list mail-extr-x400-encoded-address-given-name-pattern | |
1349 | mail-extr-x400-encoded-address-surname-pattern | |
1350 | mail-extr-x400-encoded-address-full-name-pattern)) | |
1351 | ||
1352 | ;; Discard the rest, since it contains stuff like | |
1353 | ;; routing information, not part of a name. | |
1354 | (mail-extr-skip-whitespace-backward) | |
1355 | (delete-region (point) (point-max)) | |
1356 | ||
1357 | ;; Handle periods used for spacing. | |
1358 | (while (re-search-forward mail-extr-bad-dot-pattern nil t) | |
1359 | (replace-match "\\1 \\2" t)) | |
1360 | ||
1361 | ;; (setq mailbox-name-processed-flag t) | |
1362 | ) | |
1363 | ||
1364 | ;; Handle normal addresses. | |
1365 | (t | |
1366 | (goto-char (point-min)) | |
1367 | ;; Handle _ and = used for spacing. | |
1368 | (while (re-search-forward "\\([^_=]+\\)[_=]" nil t) | |
1369 | (replace-match "\\1 " t) | |
1370 | ;; (setq mailbox-name-processed-flag t) | |
1371 | ) | |
1372 | (goto-char (point-max)))))))) | |
1373 | ||
1374 | ;; undo the dirty deed | |
1375 | (if (not mail-extr-mangle-uucp) | |
1376 | (modify-syntax-entry ?! "." (syntax-table))) | |
1377 | ;; | |
1378 | ;; If we derived the name from the mailbox part of the address, | |
1379 | ;; and we only got one word out of it, don't treat that as a | |
1380 | ;; name. "foo@bar" --> (nil "foo@bar"), not ("foo" "foo@bar") | |
1381 | ;; (if (not mailbox-name-processed-flag) | |
1382 | ;; (delete-region (point-min) (point-max))) | |
1383 | )) | |
1384 | ||
1385 | (set-syntax-table mail-extr-address-text-syntax-table) | |
1386 | ||
1387 | (mail-extr-voodoo mbox-beg mbox-end canonicalization-buffer) | |
1388 | (goto-char (point-min)) | |
1389 | ||
1390 | ;; If name is "First Last" and userid is "F?L", then assume | |
1391 | ;; the middle initial is the second letter in the userid. | |
1392 | ;; Initial code by Jamie Zawinski <jwz@lucid.com> | |
1393 | ;; *** Make it work when there's a suffix as well. | |
1394 | (goto-char (point-min)) | |
1395 | (cond ((and mail-extr-guess-middle-initial | |
1396 | (not disable-initial-guessing-flag) | |
1397 | (eq 3 (- mbox-end mbox-beg)) | |
1398 | (progn | |
1399 | (goto-char (point-min)) | |
1400 | (looking-at mail-extr-two-name-pattern))) | |
1401 | (setq fi (char-after (match-beginning 0)) | |
1402 | li (char-after (match-beginning 3))) | |
1403 | (save-excursion | |
1404 | (set-buffer canonicalization-buffer) | |
1405 | ;; char-equal is ignoring case here, so no need to upcase | |
1406 | ;; or downcase. | |
1407 | (let ((case-fold-search t)) | |
1408 | (and (char-equal fi (char-after mbox-beg)) | |
1409 | (char-equal li (char-after (1- mbox-end))) | |
1410 | (setq mi (char-after (1+ mbox-beg)))))) | |
1411 | (cond ((and mi | |
1412 | ;; TODO: use better table than syntax table | |
1413 | (eq ?w (char-syntax mi))) | |
1414 | (goto-char (match-beginning 3)) | |
1415 | (insert (upcase mi) ". "))))) | |
1416 | ||
1417 | ;; Nuke name if it is the same as mailbox name. | |
1418 | (let ((buffer-length (- (point-max) (point-min))) | |
1419 | (i 0) | |
1420 | (names-match-flag t)) | |
1421 | (cond ((and (> buffer-length 0) | |
1422 | (eq buffer-length (- mbox-end mbox-beg))) | |
1423 | (goto-char (point-max)) | |
1424 | (insert-buffer-substring canonicalization-buffer | |
1425 | mbox-beg mbox-end) | |
1426 | (while (and names-match-flag | |
1427 | (< i buffer-length)) | |
1428 | (or (eq (downcase (char-after (+ i (point-min)))) | |
1429 | (downcase | |
1430 | (char-after (+ i buffer-length (point-min))))) | |
1431 | (setq names-match-flag nil)) | |
1432 | (setq i (1+ i))) | |
1433 | (delete-region (+ (point-min) buffer-length) (point-max)) | |
1434 | (if names-match-flag | |
1435 | (narrow-to-region (point) (point)))))) | |
1436 | ||
1437 | ;; Nuke name if it's just one word. | |
1438 | (goto-char (point-min)) | |
1439 | (and mail-extr-ignore-single-names | |
1440 | (not (re-search-forward "[- ]" nil t)) | |
1441 | (narrow-to-region (point) (point))) | |
1442 | ||
1443 | ;; Record the result | |
1444 | (setq value-list | |
1445 | (cons (list (if (not (= (point-min) (point-max))) | |
1446 | (buffer-string)) | |
1447 | (save-excursion | |
1448 | (set-buffer canonicalization-buffer) | |
1449 | (if (not (= (point-min) (point-max))) | |
1450 | (buffer-string)))) | |
1451 | value-list)) | |
1452 | ||
1453 | ;; Unless one address is all we wanted, | |
1454 | ;; delete this one from extraction-buffer | |
1455 | ;; and get ready to extract the next address. | |
1456 | (when all | |
1457 | (if end-of-address | |
1458 | (narrow-to-region 1 end-of-address) | |
1459 | (widen)) | |
1460 | (delete-region (point-min) (point-max)) | |
1461 | (widen)) | |
1462 | ))) | |
1463 | (if all (nreverse value-list) (car value-list)) | |
1464 | )) | |
72c0ae01 | 1465 | |
154b3e39 RS |
1466 | (defun mail-extr-voodoo (mbox-beg mbox-end canonicalization-buffer) |
1467 | (let ((word-count 0) | |
1468 | (case-fold-search nil) | |
1469 | mixed-case-flag lower-case-flag ;;upper-case-flag | |
72c0ae01 | 1470 | suffix-flag last-name-comma-flag |
154b3e39 RS |
1471 | ;;cbeg cend |
1472 | initial | |
1473 | begin-again-flag | |
1474 | drop-this-word-if-trailing-flag | |
1475 | drop-last-word-if-trailing-flag | |
1476 | word-found-flag | |
1477 | this-word-beg last-word-beg | |
1478 | name-beg name-end | |
1479 | name-done-flag | |
72c0ae01 ER |
1480 | ) |
1481 | (save-excursion | |
154b3e39 | 1482 | (set-syntax-table mail-extr-address-text-syntax-table) |
e432fc74 RS |
1483 | |
1484 | ;; Get rid of comments. | |
1485 | (goto-char (point-min)) | |
1486 | (while (not (eobp)) | |
1487 | ;; Initialize for this iteration of the loop. | |
1488 | (skip-chars-forward "^({[\"'`") | |
1489 | (let ((cbeg (point))) | |
1490 | (set-syntax-table mail-extr-address-text-comment-syntax-table) | |
1491 | (cond ((memq (following-char) '(?\' ?\`)) | |
b9c22adf | 1492 | (search-forward "'" nil 'move |
e432fc74 RS |
1493 | (if (eq ?\' (following-char)) 2 1))) |
1494 | (t | |
1495 | (or (mail-extr-safe-move-sexp 1) | |
1496 | (goto-char (point-max))))) | |
1497 | (set-syntax-table mail-extr-address-text-syntax-table) | |
1498 | (when (eq (char-after cbeg) ?\() | |
1499 | ;; Delete the comment itself. | |
1500 | (delete-region cbeg (point)) | |
1501 | ;; Canonicalize whitespace where the comment was. | |
1502 | (skip-chars-backward " \t") | |
1503 | (if (looking-at "\\([ \t]+$\\|[ \t]+,\\)") | |
1504 | (replace-match "") | |
1505 | (setq cbeg (point)) | |
1506 | (skip-chars-forward " \t") | |
1507 | (if (bobp) | |
1508 | (delete-region (point) cbeg) | |
1509 | (just-one-space)))))) | |
72c0ae01 | 1510 | |
154b3e39 | 1511 | ;; This was moved above. |
72c0ae01 | 1512 | ;; Fix . used as space |
154b3e39 RS |
1513 | ;; But it belongs here because it occurs not only as |
1514 | ;; rypens@reks.uia.ac.be (Piet.Rypens) | |
1515 | ;; but also as | |
1516 | ;; "Piet.Rypens" <rypens@reks.uia.ac.be> | |
1517 | ;;(goto-char (point-min)) | |
1518 | ;;(while (re-search-forward mail-extr-bad-dot-pattern nil t) | |
1519 | ;; (replace-match "\\1 \\2" t)) | |
1520 | ||
1521 | (cond ((not (search-forward " " nil t)) | |
1522 | (goto-char (point-min)) | |
1523 | (cond ((search-forward "_" nil t) | |
1524 | ;; Handle the *idiotic* use of underlines as spaces. | |
1525 | ;; Example: fml@foo.bar.dom (First_M._Last) | |
1526 | (goto-char (point-min)) | |
1527 | (while (search-forward "_" nil t) | |
1528 | (replace-match " " t))) | |
1529 | ((search-forward "." nil t) | |
1530 | ;; Fix . used as space | |
1531 | ;; Example: danj1@cb.att.com (daniel.jacobson) | |
1532 | (goto-char (point-min)) | |
1533 | (while (re-search-forward mail-extr-bad-dot-pattern nil t) | |
1534 | (replace-match "\\1 \\2" t)))))) | |
1535 | ||
154b3e39 | 1536 | ;; Loop over the words (and other junk) in the name. |
72c0ae01 | 1537 | (goto-char (point-min)) |
154b3e39 RS |
1538 | (while (not name-done-flag) |
1539 | ||
1540 | (cond (word-found-flag | |
1541 | ;; Last time through this loop we skipped over a word. | |
1542 | (setq last-word-beg this-word-beg) | |
1543 | (setq drop-last-word-if-trailing-flag | |
1544 | drop-this-word-if-trailing-flag) | |
1545 | (setq word-found-flag nil))) | |
1546 | ||
1547 | (cond (begin-again-flag | |
1548 | ;; Last time through the loop we found something that | |
1549 | ;; indicates we should pretend we are beginning again from | |
1550 | ;; the start. | |
1551 | (setq word-count 0) | |
1552 | (setq last-word-beg nil) | |
1553 | (setq drop-last-word-if-trailing-flag nil) | |
1554 | (setq mixed-case-flag nil) | |
1555 | (setq lower-case-flag nil) | |
1556 | ;; (setq upper-case-flag nil) | |
1557 | (setq begin-again-flag nil) | |
1558 | )) | |
1559 | ||
1560 | ;; Initialize for this iteration of the loop. | |
1561 | (mail-extr-skip-whitespace-forward) | |
1562 | (if (eq word-count 0) (narrow-to-region (point) (point-max))) | |
1563 | (setq this-word-beg (point)) | |
1564 | (setq drop-this-word-if-trailing-flag nil) | |
1565 | ||
1566 | ;; Decide what to do based on what we are looking at. | |
1567 | (cond | |
1568 | ||
1569 | ;; Delete title | |
1570 | ((and (eq word-count 0) | |
1571 | (looking-at mail-extr-full-name-prefixes)) | |
1572 | (goto-char (match-end 0)) | |
1573 | (narrow-to-region (point) (point-max))) | |
1574 | ||
1575 | ;; Stop after name suffix | |
1576 | ((and (>= word-count 2) | |
1577 | (looking-at mail-extr-full-name-suffix-pattern)) | |
1578 | (mail-extr-skip-whitespace-backward) | |
1579 | (setq suffix-flag (point)) | |
1580 | (if (eq ?, (following-char)) | |
1581 | (forward-char 1) | |
1582 | (insert ?,)) | |
1583 | ;; Enforce at least one space after comma | |
1584 | (or (eq ?\ (following-char)) | |
1585 | (insert ?\ )) | |
1586 | (mail-extr-skip-whitespace-forward) | |
1587 | (cond ((memq (following-char) '(?j ?J ?s ?S)) | |
1588 | (capitalize-word 1) | |
1589 | (if (eq (following-char) ?.) | |
1590 | (forward-char 1) | |
1591 | (insert ?.))) | |
1592 | (t | |
1593 | (upcase-word 1))) | |
1594 | (setq word-found-flag t) | |
1595 | (setq name-done-flag t)) | |
1596 | ||
1597 | ;; Handle SCA names | |
1598 | ((looking-at "MKA \\(.+\\)") ; "Mundanely Known As" | |
1599 | (goto-char (match-beginning 1)) | |
1600 | (narrow-to-region (point) (point-max)) | |
1601 | (setq begin-again-flag t)) | |
1602 | ||
1603 | ;; Check for initial last name followed by comma | |
1604 | ((and (eq ?, (following-char)) | |
1605 | (eq word-count 1)) | |
1606 | (forward-char 1) | |
1607 | (setq last-name-comma-flag t) | |
1608 | (or (eq ?\ (following-char)) | |
1609 | (insert ?\ ))) | |
1610 | ||
1611 | ;; Stop before trailing comma-separated comment | |
1612 | ;; THIS CASE MUST BE AFTER THE PRECEDING CASES. | |
1613 | ;; *** This case is redundant??? | |
1614 | ;;((eq ?, (following-char)) | |
1615 | ;; (setq name-done-flag t)) | |
1616 | ||
1617 | ;; Delete parenthesized/quoted comment/nickname | |
1618 | ((memq (following-char) '(?\( ?\{ ?\[ ?\" ?\' ?\`)) | |
1619 | (setq cbeg (point)) | |
1620 | (set-syntax-table mail-extr-address-text-comment-syntax-table) | |
1621 | (cond ((memq (following-char) '(?\' ?\`)) | |
1622 | (or (search-forward "'" nil t | |
1623 | (if (eq ?\' (following-char)) 2 1)) | |
1624 | (mail-extr-delete-char 1))) | |
1625 | (t | |
1626 | (or (mail-extr-safe-move-sexp 1) | |
1627 | (goto-char (point-max))))) | |
1628 | (set-syntax-table mail-extr-address-text-syntax-table) | |
1629 | (setq cend (point)) | |
72c0ae01 | 1630 | (cond |
154b3e39 | 1631 | ;; Handle case of entire name being quoted |
72c0ae01 | 1632 | ((and (eq word-count 0) |
154b3e39 RS |
1633 | (looking-at " *\\'") |
1634 | (>= (- cend cbeg) 2)) | |
1635 | (narrow-to-region (1+ cbeg) (1- cend)) | |
1636 | (goto-char (point-min))) | |
1637 | (t | |
1638 | ;; Handle case of quoted initial | |
1639 | (if (and (or (= 3 (- cend cbeg)) | |
1640 | (and (= 4 (- cend cbeg)) | |
1641 | (eq ?. (char-after (+ 2 cbeg))))) | |
1642 | (not (looking-at " *\\'"))) | |
1643 | (setq initial (char-after (1+ cbeg))) | |
1644 | (setq initial nil)) | |
1645 | (delete-region cbeg cend) | |
1646 | (if initial | |
1647 | (insert initial ". "))))) | |
1648 | ||
154b3e39 RS |
1649 | ;; Handle *Stupid* VMS date stamps |
1650 | ((looking-at mail-extr-stupid-vms-date-stamp-pattern) | |
1651 | (replace-match "" t)) | |
1652 | ||
1653 | ;; Handle Chinese characters. | |
1654 | ((looking-at mail-extr-hz-embedded-gb-encoded-chinese-pattern) | |
1655 | (goto-char (match-end 0)) | |
1656 | (setq word-found-flag t)) | |
1657 | ||
1658 | ;; Skip initial garbage characters. | |
1659 | ;; THIS CASE MUST BE AFTER THE PRECEDING CASES. | |
1660 | ((and (eq word-count 0) | |
1661 | (looking-at mail-extr-leading-garbage)) | |
1662 | (goto-char (match-end 0)) | |
1663 | ;; *** Skip backward over these??? | |
1664 | ;; (skip-chars-backward "& \"") | |
1665 | (narrow-to-region (point) (point-max))) | |
1666 | ||
1667 | ;; Various stopping points | |
1668 | ((or | |
72c0ae01 | 1669 | |
154b3e39 RS |
1670 | ;; Stop before ALL CAPS acronyms, if preceded by mixed-case |
1671 | ;; words. Example: XT-DEM. | |
1672 | (and (>= word-count 2) | |
1673 | mixed-case-flag | |
1674 | (looking-at mail-extr-weird-acronym-pattern) | |
1675 | (not (looking-at mail-extr-roman-numeral-pattern))) | |
72c0ae01 | 1676 | |
154b3e39 RS |
1677 | ;; Stop before trailing alternative address |
1678 | (looking-at mail-extr-alternative-address-pattern) | |
72c0ae01 | 1679 | |
154b3e39 RS |
1680 | ;; Stop before trailing comment not introduced by comma |
1681 | ;; THIS CASE MUST BE AFTER AN EARLIER CASE. | |
1682 | (looking-at mail-extr-trailing-comment-start-pattern) | |
72c0ae01 | 1683 | |
154b3e39 RS |
1684 | ;; Stop before telephone numbers |
1685 | (looking-at mail-extr-telephone-extension-pattern)) | |
1686 | (setq name-done-flag t)) | |
1687 | ||
1688 | ;; Delete ham radio call signs | |
1689 | ((looking-at mail-extr-ham-call-sign-pattern) | |
1690 | (delete-region (match-beginning 0) (match-end 0))) | |
1691 | ||
1692 | ;; Fixup initials | |
1693 | ((looking-at mail-extr-initial-pattern) | |
1694 | (or (eq (following-char) (upcase (following-char))) | |
1695 | (setq lower-case-flag t)) | |
1696 | (forward-char 1) | |
1697 | (if (eq ?. (following-char)) | |
1698 | (forward-char 1) | |
1699 | (insert ?.)) | |
1700 | (or (eq ?\ (following-char)) | |
1701 | (insert ?\ )) | |
1702 | (setq word-found-flag t)) | |
1703 | ||
1704 | ;; Handle BITNET LISTSERV list names. | |
1705 | ((and (eq word-count 0) | |
1706 | (looking-at mail-extr-listserv-list-name-pattern)) | |
1707 | (narrow-to-region (match-beginning 1) (match-end 1)) | |
1708 | (setq word-found-flag t) | |
1709 | (setq name-done-flag t)) | |
1710 | ||
524d06f5 RS |
1711 | ;; Handle & substitution, when & is last and is not first. |
1712 | ((and (> word-count 0) | |
1713 | (eq ?\ (preceding-char)) | |
1714 | (eq (following-char) ?&) | |
1715 | (eq (1+ (point)) (point-max))) | |
1716 | (mail-extr-delete-char 1) | |
1717 | (capitalize-region | |
1718 | (point) | |
1719 | (progn | |
1720 | (insert-buffer-substring canonicalization-buffer | |
1721 | mbox-beg mbox-end) | |
1722 | (point))) | |
1723 | (setq disable-initial-guessing-flag t) | |
1724 | (setq word-found-flag t)) | |
1725 | ||
1726 | ;; Handle & between names, as in "Bob & Susie". | |
1727 | ((and (> word-count 0) (eq (following-char) ?\&)) | |
1728 | (setq name-beg (point)) | |
1729 | (setq name-end (1+ name-beg)) | |
1730 | (setq word-found-flag t) | |
1731 | (goto-char name-end)) | |
1732 | ||
154b3e39 RS |
1733 | ;; Regular name words |
1734 | ((looking-at mail-extr-name-pattern) | |
1735 | (setq name-beg (point)) | |
1736 | (setq name-end (match-end 0)) | |
1737 | ||
1738 | ;; Certain words will be dropped if they are at the end. | |
1739 | (and (>= word-count 2) | |
1740 | (not lower-case-flag) | |
1741 | (or | |
1742 | ;; A trailing 4-or-more letter lowercase words preceded by | |
1743 | ;; mixed case or uppercase words will be dropped. | |
1744 | (looking-at "[a-z][a-z][a-z][a-z]+[ \t]*\\'") | |
1745 | ;; Drop a trailing word which is terminated with a period. | |
1746 | (eq ?. (char-after (1- name-end)))) | |
1747 | (setq drop-this-word-if-trailing-flag t)) | |
1748 | ||
1749 | ;; Set the flags that indicate whether we have seen a lowercase | |
1750 | ;; word, a mixed case word, and an uppercase word. | |
1751 | (if (re-search-forward "[a-z]" name-end t) | |
1752 | (if (progn | |
1753 | (goto-char name-beg) | |
1754 | (re-search-forward "[A-Z]" name-end t)) | |
1755 | (setq mixed-case-flag t) | |
72c0ae01 | 1756 | (setq lower-case-flag t)) |
154b3e39 RS |
1757 | ;; (setq upper-case-flag t) |
1758 | ) | |
1759 | ||
1760 | (goto-char name-end) | |
1761 | (setq word-found-flag t)) | |
72c0ae01 | 1762 | |
154b3e39 RS |
1763 | (t |
1764 | (setq name-done-flag t) | |
1765 | )) | |
1766 | ||
1767 | ;; Count any word that we skipped over. | |
1768 | (if word-found-flag | |
1769 | (setq word-count (1+ word-count)))) | |
72c0ae01 | 1770 | |
154b3e39 RS |
1771 | ;; If the last thing in the name is 2 or more periods, or one or more |
1772 | ;; other sentence terminators (but not a single period) then keep them | |
4802af76 | 1773 | ;; and the preceding word. This is for the benefit of whole sentences |
154b3e39 RS |
1774 | ;; in the name field: it's better behavior than dropping the last word |
1775 | ;; of the sentence... | |
1776 | (if (and (not suffix-flag) | |
1777 | (looking-at "\\(\\.+\\|[?!;:.][?!;:.]+\\|[?!;:][?!;:.]*\\)\\'")) | |
1778 | (goto-char (setq suffix-flag (point-max)))) | |
1779 | ||
1780 | ;; Drop everything after point and certain trailing words. | |
1781 | (narrow-to-region (point-min) | |
1782 | (or (and drop-last-word-if-trailing-flag | |
1783 | last-word-beg) | |
1784 | (point))) | |
1785 | ||
1786 | ;; Xerox's mailers SUCK!!!!!! | |
1787 | ;; We simply refuse to believe that any last name is PARC or ADOC. | |
1788 | ;; If it looks like that is the last name, that there is no meaningful | |
1789 | ;; here at all. Actually I guess it would be best to map patterns | |
1790 | ;; like foo.hoser@xerox.com into foo@hoser.xerox.com, but I don't | |
1791 | ;; actually know that that is what's going on. | |
72c0ae01 ER |
1792 | (cond ((not suffix-flag) |
1793 | (goto-char (point-min)) | |
154b3e39 RS |
1794 | (let ((case-fold-search t)) |
1795 | (if (looking-at "[-A-Za-z_]+[. ]\\(PARC\\|ADOC\\)\\'") | |
1796 | (erase-buffer))))) | |
1797 | ||
72c0ae01 ER |
1798 | ;; If last name first put it at end (but before suffix) |
1799 | (cond (last-name-comma-flag | |
1800 | (goto-char (point-min)) | |
1801 | (search-forward ",") | |
154b3e39 | 1802 | (setq name-end (1- (point))) |
72c0ae01 | 1803 | (goto-char (or suffix-flag (point-max))) |
154b3e39 RS |
1804 | (or (eq ?\ (preceding-char)) |
1805 | (insert ?\ )) | |
1806 | (insert-buffer-substring (current-buffer) (point-min) name-end) | |
1807 | (goto-char name-end) | |
1808 | (skip-chars-forward "\t ,") | |
1809 | (narrow-to-region (point) (point-max)))) | |
72c0ae01 | 1810 | |
154b3e39 | 1811 | ;; Delete leading and trailing junk characters. |
4802af76 | 1812 | ;; *** This is probably completely unneeded now. |
154b3e39 RS |
1813 | ;;(goto-char (point-max)) |
1814 | ;;(skip-chars-backward mail-extr-non-end-name-chars) | |
1815 | ;;(if (eq ?. (following-char)) | |
1816 | ;; (forward-char 1)) | |
1817 | ;;(narrow-to-region (point) | |
1818 | ;; (progn | |
1819 | ;; (goto-char (point-min)) | |
1820 | ;; (skip-chars-forward mail-extr-non-begin-name-chars) | |
1821 | ;; (point))) | |
72c0ae01 ER |
1822 | |
1823 | ;; Compress whitespace | |
1824 | (goto-char (point-min)) | |
1825 | (while (re-search-forward "[ \t\n]+" nil t) | |
154b3e39 | 1826 | (replace-match (if (eobp) "" " ") t)) |
72c0ae01 ER |
1827 | ))) |
1828 | ||
154b3e39 | 1829 | \f |
72c0ae01 | 1830 | |
154b3e39 RS |
1831 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; |
1832 | ;; | |
1833 | ;; Table of top-level domain names. | |
1834 | ;; | |
1835 | ;; This is used during address canonicalization; be careful of format changes. | |
72c0ae01 ER |
1836 | ;; Keep in mind that the country abbreviations follow ISO-3166. There is |
1837 | ;; a U.S. FIPS that specifies a different set of two-letter country | |
1838 | ;; abbreviations. | |
1839 | ||
154b3e39 RS |
1840 | (defconst mail-extr-all-top-level-domains |
1841 | (let ((ob (make-vector 509 0))) | |
1842 | (mapcar | |
1843 | (function | |
1844 | (lambda (x) | |
1845 | (put (intern (downcase (car x)) ob) | |
1846 | 'domain-name | |
1847 | (if (nth 2 x) | |
1848 | (format (nth 2 x) (nth 1 x)) | |
1849 | (nth 1 x))))) | |
e1a34b58 RS |
1850 | '( |
1851 | ;; ISO 3166 codes: | |
1852 | ("ae" "United Arab Emirates") | |
1853 | ("ag" "Antigua and Barbuda") | |
1854 | ("al" "Albania") | |
1855 | ("ao" "Angola") | |
1856 | ("aq" "Antarctica") ; continent | |
154b3e39 | 1857 | ("ar" "Argentina" "Argentine Republic") |
154b3e39 RS |
1858 | ("at" "Austria" "The Republic of %s") |
1859 | ("au" "Australia") | |
e1a34b58 | 1860 | ("az" "Azerbaijan") |
154b3e39 | 1861 | ("bb" "Barbados") |
e1a34b58 | 1862 | ("bd" "Bangladesh") |
154b3e39 | 1863 | ("be" "Belgium" "The Kingdom of %s") |
e1a34b58 | 1864 | ("bf" "Burkina Faso") |
154b3e39 | 1865 | ("bg" "Bulgaria") |
e1a34b58 RS |
1866 | ("bh" "Bahrain") |
1867 | ("bm" "Bermuda") | |
154b3e39 RS |
1868 | ("bo" "Bolivia" "Republic of %s") |
1869 | ("br" "Brazil" "The Federative Republic of %s") | |
1870 | ("bs" "Bahamas") | |
e1a34b58 RS |
1871 | ("bw" "Botswana") |
1872 | ("by" "Belarus") | |
154b3e39 RS |
1873 | ("bz" "Belize") |
1874 | ("ca" "Canada") | |
e1a34b58 | 1875 | ("cg" "Congo") |
154b3e39 | 1876 | ("ch" "Switzerland" "The Swiss Confederation") |
e1a34b58 | 1877 | ("ci" "Ivory Coast") |
154b3e39 | 1878 | ("cl" "Chile" "The Republic of %s") |
e1a34b58 | 1879 | ("cm" "Cameroon") ; In .fr domain |
154b3e39 | 1880 | ("cn" "China" "The People's Republic of %s") |
e1a34b58 | 1881 | ("co" "Colombia") |
154b3e39 RS |
1882 | ("cr" "Costa Rica" "The Republic of %s") |
1883 | ("cs" "Czechoslovakia") | |
e1a34b58 RS |
1884 | ("cu" "Cuba") |
1885 | ("cy" "Cyprus") | |
1886 | ("cz" "Czech Republic") | |
154b3e39 RS |
1887 | ("de" "Germany") |
1888 | ("dk" "Denmark") | |
1889 | ("dm" "Dominica") | |
1890 | ("do" "Dominican Republic" "The %s") | |
e1a34b58 | 1891 | ("dz" "Algeria") |
154b3e39 | 1892 | ("ec" "Ecuador" "The Republic of %s") |
e1a34b58 | 1893 | ("ee" "Estonia") |
154b3e39 | 1894 | ("eg" "Egypt" "The Arab Republic of %s") |
e1a34b58 | 1895 | ("er" "Eritrea") |
154b3e39 RS |
1896 | ("es" "Spain" "The Kingdom of %s") |
1897 | ("fi" "Finland" "The Republic of %s") | |
1898 | ("fj" "Fiji") | |
e1a34b58 | 1899 | ("fo" "Faroe Islands") |
154b3e39 | 1900 | ("fr" "France") |
e1a34b58 RS |
1901 | ("gb" "Great Britain") |
1902 | ("gd" "Grenada") | |
1903 | ("ge" "Georgia") | |
1904 | ("gf" "Guyana (Fr.)") | |
1905 | ("gp" "Guadeloupe (Fr.)") | |
154b3e39 | 1906 | ("gr" "Greece" "The Hellenic Republic (%s)") |
e1a34b58 RS |
1907 | ("gt" "Guatemala") |
1908 | ("gu" "Guam (U.S.)") | |
154b3e39 | 1909 | ("hk" "Hong Kong") |
e1a34b58 RS |
1910 | ("hn" "Honduras") |
1911 | ("hr" "Croatia") | |
1912 | ("ht" "Haiti") | |
449c2cd8 | 1913 | ("hu" "Hungary" "The Hungarian Republic") ;??? |
e1a34b58 | 1914 | ("id" "Indonesia") |
154b3e39 RS |
1915 | ("ie" "Ireland") |
1916 | ("il" "Israel" "The State of %s") | |
1917 | ("in" "India" "The Republic of %s") | |
e1a34b58 | 1918 | ("ir" "Iran") |
154b3e39 RS |
1919 | ("is" "Iceland" "The Republic of %s") |
1920 | ("it" "Italy" "The Italian Republic") | |
1921 | ("jm" "Jamaica") | |
1922 | ("jp" "Japan") | |
e1a34b58 RS |
1923 | ("ke" "Kenya") |
1924 | ("kn" "St. Kitts, Nevis, and Anguilla") | |
1925 | ("kp" "Korea (North)") | |
1926 | ("kr" "Korea (South)") | |
1927 | ("kw" "Kuwait") | |
4802af76 | 1928 | ("kz" "Kazakhstan") |
e1a34b58 | 1929 | ("lb" "Lebanon") |
154b3e39 | 1930 | ("lc" "St. Lucia") |
e1a34b58 | 1931 | ("li" "Liechtenstein") |
154b3e39 | 1932 | ("lk" "Sri Lanka" "The Democratic Socialist Republic of %s") |
e1a34b58 RS |
1933 | ("ls" "Lesotho") |
1934 | ("lt" "Lithuania") | |
1935 | ("lu" "Luxembourg") | |
1936 | ("lv" "Latvia") | |
1937 | ("ma" "Morocco") | |
1938 | ("md" "Moldova") | |
1939 | ("mg" "Madagascar") | |
1940 | ("mk" "Macedonia") | |
1941 | ("ml" "Mali") | |
1942 | ("mo" "Macau") | |
1943 | ("mt" "Malta") | |
1944 | ("mu" "Mauritius") | |
1945 | ("mw" "Malawi") | |
154b3e39 RS |
1946 | ("mx" "Mexico" "The United Mexican States") |
1947 | ("my" "Malaysia" "%s (changed to Myanmar?)") ;??? | |
e1a34b58 | 1948 | ("mz" "Mozambique") |
154b3e39 | 1949 | ("na" "Namibia") |
e1a34b58 RS |
1950 | ("nc" "New Caledonia (Fr.)") |
1951 | ("ne" "Niger") ; In .fr domain | |
154b3e39 RS |
1952 | ("ni" "Nicaragua" "The Republic of %s") |
1953 | ("nl" "Netherlands" "The Kingdom of the %s") | |
1954 | ("no" "Norway" "The Kingdom of %s") | |
e1a34b58 | 1955 | ("np" "Nepal") ; Via .in domain |
154b3e39 | 1956 | ("nz" "New Zealand") |
e1a34b58 | 1957 | ("pa" "Panama") |
154b3e39 | 1958 | ("pe" "Peru") |
e1a34b58 | 1959 | ("pf" "Polynesia (Fr.)") |
154b3e39 RS |
1960 | ("pg" "Papua New Guinea") |
1961 | ("ph" "Philippines" "The Republic of the %s") | |
e1a34b58 | 1962 | ("pk" "Pakistan") |
154b3e39 | 1963 | ("pl" "Poland") |
e1a34b58 | 1964 | ("pr" "Puerto Rico (U.S.)") |
4802af76 | 1965 | ("pt" "Portugal" "The Portuguese Republic") |
154b3e39 | 1966 | ("py" "Paraguay") |
e1a34b58 RS |
1967 | ("re" "Reunion (Fr.)") ; In .fr domain |
1968 | ("ro" "Romania") | |
1969 | ("ru" "Russian Federation") | |
1970 | ("sa" "Saudi Arabia") | |
1971 | ("sc" "Seychelles") | |
1972 | ("sd" "Sudan") | |
154b3e39 RS |
1973 | ("se" "Sweden" "The Kingdom of %s") |
1974 | ("sg" "Singapore" "The Republic of %s") | |
e1a34b58 RS |
1975 | ("si" "Slovenia") |
1976 | ("sj" "Svalbard and Jan Mayen Is.") ; In .no domain | |
1977 | ("sk" "Slovakia" "The Slovak Republic") | |
1978 | ("sn" "Senegal") | |
154b3e39 RS |
1979 | ("sr" "Suriname") |
1980 | ("su" "Soviet Union") | |
e1a34b58 RS |
1981 | ("sz" "Swaziland") |
1982 | ("tg" "Togo") | |
154b3e39 | 1983 | ("th" "Thailand" "The Kingdom of %s") |
e1a34b58 | 1984 | ("tm" "Turkmenistan") ; In .su domain |
154b3e39 RS |
1985 | ("tn" "Tunisia") |
1986 | ("tr" "Turkey" "The Republic of %s") | |
1987 | ("tt" "Trinidad and Tobago") | |
1988 | ("tw" "Taiwan") | |
e1a34b58 | 1989 | ("ua" "Ukraine") |
44a1338a | 1990 | ("uk" "United Kingdom" "The %s of Great Britain and Northern Ireland") |
e1a34b58 | 1991 | ("us" "United States" "The %s of America") |
154b3e39 RS |
1992 | ("uy" "Uruguay" "The Eastern Republic of %s") |
1993 | ("vc" "St. Vincent and the Grenadines") | |
1994 | ("ve" "Venezuela" "The Republic of %s") | |
e1a34b58 RS |
1995 | ("vi" "Virgin Islands (U.S.)") |
1996 | ("vn" "Vietnam") | |
1997 | ("vu" "Vanuatu") | |
154b3e39 | 1998 | ("yu" "Yugoslavia" "The Socialist Federal Republic of %s") |
154b3e39 RS |
1999 | ("za" "South Africa" "The Republic of %s (or Zambia? Zaire?)") |
2000 | ("zw" "Zimbabwe" "Republic of %s") | |
e1a34b58 RS |
2001 | ;; Special top-level domains: |
2002 | ("arpa" t "Advanced Research Projects Agency (U.S. DoD)") | |
2003 | ("bitnet" t "Because It's Time NET") | |
2004 | ("com" t "Commercial") | |
2005 | ("edu" t "Educational") | |
2006 | ("gov" t "Government (U.S.)") | |
2007 | ("int" t "International (NATO)") | |
2008 | ("mil" t "Military (U.S.)") | |
2009 | ("nato" t "North Atlantic Treaty Organization") | |
2010 | ("net" t "Network") | |
2011 | ("org" t "Non-profit Organization") | |
2012 | ;;("unter-dom" t "? (Ger.)") | |
2013 | ("uucp" t "Unix to Unix CoPy") | |
2014 | ;;("fipnet" nil "?") | |
154b3e39 RS |
2015 | )) |
2016 | ob)) | |
2017 | ||
2018 | ;;;###autoload | |
2019 | (defun what-domain (domain) | |
e1cd65fd | 2020 | "Convert mail domain DOMAIN to the country it corresponds to." |
154b3e39 RS |
2021 | (interactive |
2022 | (let ((completion-ignore-case t)) | |
2023 | (list (completing-read "Domain: " | |
2024 | mail-extr-all-top-level-domains nil t)))) | |
2025 | (or (setq domain (intern-soft (downcase domain) | |
2026 | mail-extr-all-top-level-domains)) | |
29565a87 | 2027 | (error "No such domain")) |
154b3e39 | 2028 | (message "%s: %s" (upcase (symbol-name domain)) (get domain 'domain-name))) |
72c0ae01 ER |
2029 | |
2030 | \f | |
154b3e39 RS |
2031 | ;(let ((all nil)) |
2032 | ; (mapatoms #'(lambda (x) | |
2033 | ; (if (and (boundp x) | |
2034 | ; (string-match "^mail-extr-" (symbol-name x))) | |
2035 | ; (setq all (cons x all))))) | |
2036 | ; (setq all (sort all #'string-lessp)) | |
2037 | ; (cons 'setq | |
2038 | ; (apply 'nconc (mapcar #'(lambda (x) | |
2039 | ; (list x (symbol-value x))) | |
2040 | ; all)))) | |
72c0ae01 ER |
2041 | |
2042 | \f | |
154b3e39 | 2043 | (provide 'mail-extr) |
72c0ae01 ER |
2044 | |
2045 | ;;; mail-extr.el ends here |