| 1 | Exim's interfaces to mail filtering |
| 2 | |
| 3 | Philip Hazel |
| 4 | |
| 5 | Copyright (c) 2014 University of Cambridge |
| 6 | |
| 7 | Revision 4.89 07 Mar 2017 PH |
| 8 | |
| 9 | ------------------------------------------------------------------------------- |
| 10 | |
| 11 | TABLE OF CONTENTS |
| 12 | |
| 13 | 1. Forwarding and filtering in Exim |
| 14 | |
| 15 | 1.1. Introduction |
| 16 | 1.2. Filter operation |
| 17 | 1.3. Testing a new filter file |
| 18 | 1.4. Installing a filter file |
| 19 | 1.5. Testing an installed filter file |
| 20 | 1.6. Details of filtering commands |
| 21 | |
| 22 | 2. Sieve filter files |
| 23 | |
| 24 | 2.1. Recognition of Sieve filters |
| 25 | 2.2. Saving to specified folders |
| 26 | 2.3. Strings containing header names |
| 27 | 2.4. Exists test with empty list of headers |
| 28 | 2.5. Header test with invalid MIME encoding in header |
| 29 | 2.6. Address test for multiple addresses per header |
| 30 | 2.7. Semantics of keep |
| 31 | 2.8. Semantics of fileinto |
| 32 | 2.9. Semantics of redirect |
| 33 | 2.10. String arguments |
| 34 | 2.11. Number units |
| 35 | 2.12. RFC compliance |
| 36 | |
| 37 | 3. Exim filter files |
| 38 | |
| 39 | 3.1. Format of Exim filter files |
| 40 | 3.2. Data values in filter commands |
| 41 | 3.3. String expansion |
| 42 | 3.4. Some useful general variables |
| 43 | 3.5. Header variables |
| 44 | 3.6. User variables |
| 45 | 3.7. Current directory |
| 46 | 3.8. Significant deliveries |
| 47 | 3.9. Filter commands |
| 48 | 3.10. The add command |
| 49 | 3.11. The deliver command |
| 50 | 3.12. The save command |
| 51 | 3.13. The pipe command |
| 52 | 3.14. Mail commands |
| 53 | 3.15. Logging commands |
| 54 | 3.16. The finish command |
| 55 | 3.17. The testprint command |
| 56 | 3.18. The fail command |
| 57 | 3.19. The freeze command |
| 58 | 3.20. The headers command |
| 59 | 3.21. Obeying commands conditionally |
| 60 | 3.22. String testing conditions |
| 61 | 3.23. Numeric testing conditions |
| 62 | 3.24. Testing for significant deliveries |
| 63 | 3.25. Testing for error messages |
| 64 | 3.26. Testing a list of addresses |
| 65 | 3.27. Testing for personal mail |
| 66 | 3.28. Alias addresses for the personal condition |
| 67 | 3.29. Details of the personal condition |
| 68 | 3.30. Testing delivery status |
| 69 | 3.31. Multiple personal mailboxes |
| 70 | 3.32. Ignoring delivery errors |
| 71 | 3.33. Examples of Exim filter commands |
| 72 | |
| 73 | |
| 74 | |
| 75 | =============================================================================== |
| 76 | 1. FORWARDING AND FILTERING IN EXIM |
| 77 | |
| 78 | This document describes the user interfaces to Exim's in-built mail filtering |
| 79 | facilities, and is copyright (c) University of Cambridge 2014. It corresponds |
| 80 | to Exim version 4.89. |
| 81 | |
| 82 | |
| 83 | 1.1 Introduction |
| 84 | ---------------- |
| 85 | |
| 86 | Most Unix mail transfer agents (programs that deliver mail) permit individual |
| 87 | users to specify automatic forwarding of their mail, usually by placing a list |
| 88 | of forwarding addresses in a file called .forward in their home directories. |
| 89 | Exim extends this facility by allowing the forwarding instructions to be a set |
| 90 | of rules rather than just a list of addresses, in effect providing ".forward |
| 91 | with conditions". Operating the set of rules is called filtering, and the file |
| 92 | that contains them is called a filter file. |
| 93 | |
| 94 | Exim supports two different kinds of filter file. An Exim filter contains |
| 95 | instructions in a format that is unique to Exim. A Sieve filter contains |
| 96 | instructions in the Sieve format that is defined by RFC 3028. As this is a |
| 97 | standard format, Sieve filter files may already be familiar to some users. |
| 98 | Sieve files should also be portable between different environments. However, |
| 99 | the Exim filtering facility contains more features (such as variable |
| 100 | expansion), and better integration with the host environment (such as the use |
| 101 | of external processes and pipes). |
| 102 | |
| 103 | The choice of which kind of filter to use can be left to the end-user, provided |
| 104 | that the system administrator has configured Exim appropriately for both kinds |
| 105 | of filter. However, if interoperability is important, Sieve is the only choice. |
| 106 | |
| 107 | The ability to use filtering or traditional forwarding has to be enabled by the |
| 108 | system administrator, and some of the individual facilities can be separately |
| 109 | enabled or disabled. A local document should be provided to describe exactly |
| 110 | what has been enabled. In the absence of this, consult your system |
| 111 | administrator. |
| 112 | |
| 113 | This document describes how to use a filter file and the format of its |
| 114 | contents. It is intended for use by end-users. Both Sieve filters and Exim |
| 115 | filters are covered. However, for Sieve filters, only issues that relate to the |
| 116 | Exim implementation are discussed, since Sieve itself is described elsewhere. |
| 117 | |
| 118 | The contents of traditional .forward files are not described here. They |
| 119 | normally contain just a list of addresses, file names, or pipe commands, |
| 120 | separated by commas or newlines, but other types of item are also available. |
| 121 | The full details can be found in the chapter on the redirect router in the Exim |
| 122 | specification, which also describes how the system administrator can set up and |
| 123 | control the use of filtering. |
| 124 | |
| 125 | |
| 126 | 1.2 Filter operation |
| 127 | -------------------- |
| 128 | |
| 129 | It is important to realize that, in Exim, no deliveries are actually made while |
| 130 | a filter or traditional .forward file is being processed. Running a filter or |
| 131 | processing a traditional .forward file sets up future delivery operations, but |
| 132 | does not carry them out. |
| 133 | |
| 134 | The result of filter or .forward file processing is a list of destinations to |
| 135 | which a message should be delivered. The deliveries themselves take place |
| 136 | later, along with all other deliveries for the message. This means that it is |
| 137 | not possible to test for successful deliveries while filtering. It also means |
| 138 | that any duplicate addresses that are generated are dropped, because Exim never |
| 139 | delivers the same message to the same address more than once. |
| 140 | |
| 141 | |
| 142 | 1.3 Testing a new filter file |
| 143 | ----------------------------- |
| 144 | |
| 145 | Filter files, especially the more complicated ones, should always be tested, as |
| 146 | it is easy to make mistakes. Exim provides a facility for preliminary testing |
| 147 | of a filter file before installing it. This tests the syntax of the file and |
| 148 | its basic operation, and can also be used with traditional .forward files. |
| 149 | |
| 150 | Because a filter can do tests on the content of messages, a test message is |
| 151 | required. Suppose you have a new filter file called myfilter and a test message |
| 152 | in a file called test-message. Assuming that Exim is installed with the |
| 153 | conventional path name /usr/sbin/sendmail (some operating systems use /usr/lib/ |
| 154 | sendmail), the following command can be used: |
| 155 | |
| 156 | /usr/sbin/sendmail -bf myfilter <test-message |
| 157 | |
| 158 | The -bf option tells Exim that the following item on the command line is the |
| 159 | name of a filter file that is to be tested. There is also a -bF option, which |
| 160 | is similar, but which is used for testing system filter files, as opposed to |
| 161 | user filter files, and which is therefore of use only to the system |
| 162 | administrator. |
| 163 | |
| 164 | The test message is supplied on the standard input. If there are no |
| 165 | message-dependent tests in the filter, an empty file (/dev/null) can be used. A |
| 166 | supplied message must start with header lines or the "From " message separator |
| 167 | line that is found in many multi-message folder files. Note that blank lines at |
| 168 | the start terminate the header lines. A warning is given if no header lines are |
| 169 | read. |
| 170 | |
| 171 | The result of running this command, provided no errors are detected in the |
| 172 | filter file, is a list of the actions that Exim would try to take if presented |
| 173 | with the message for real. For example, for an Exim filter, the output |
| 174 | |
| 175 | Deliver message to: gulliver@lilliput.fict.example |
| 176 | Save message to: /home/lemuel/mail/archive |
| 177 | |
| 178 | means that one copy of the message would be sent to |
| 179 | gulliver@lilliput.fict.example, and another would be added to the file /home/ |
| 180 | lemuel/mail/archive, if all went well. |
| 181 | |
| 182 | The actions themselves are not attempted while testing a filter file in this |
| 183 | way; there is no check, for example, that any forwarding addresses are valid. |
| 184 | For an Exim filter, if you want to know why a particular action is being taken, |
| 185 | add the -v option to the command. This causes Exim to output the results of any |
| 186 | conditional tests and to indent its output according to the depth of nesting of |
| 187 | if commands. Further additional output from a filter test can be generated by |
| 188 | the testprint command, which is described below. |
| 189 | |
| 190 | When Exim is outputting a list of the actions it would take, if any text |
| 191 | strings are included in the output, non-printing characters therein are |
| 192 | converted to escape sequences. In particular, if any text string contains a |
| 193 | newline character, this is shown as "\n" in the testing output. |
| 194 | |
| 195 | When testing a filter in this way, Exim makes up an "envelope" for the message. |
| 196 | The recipient is by default the user running the command, and so is the sender, |
| 197 | but the command can be run with the -f option to supply a different sender. For |
| 198 | example, |
| 199 | |
| 200 | /usr/sbin/sendmail -bf myfilter \ |
| 201 | -f islington@never.where <test-message |
| 202 | |
| 203 | Alternatively, if the -f option is not used, but the first line of the supplied |
| 204 | message is a "From " separator from a message folder file (not the same thing |
| 205 | as a From: header line), the sender is taken from there. If -f is present, the |
| 206 | contents of any "From " line are ignored. |
| 207 | |
| 208 | The "return path" is the same as the envelope sender, unless the message |
| 209 | contains a Return-path: header, in which case it is taken from there. You need |
| 210 | not worry about any of this unless you want to test out features of a filter |
| 211 | file that rely on the sender address or the return path. |
| 212 | |
| 213 | It is possible to change the envelope recipient by specifying further options. |
| 214 | The -bfd option changes the domain of the recipient address, while the -bfl |
| 215 | option changes the "local part", that is, the part before the @ sign. An |
| 216 | adviser could make use of these to test someone else's filter file. |
| 217 | |
| 218 | The -bfp and -bfs options specify the prefix or suffix for the local part. |
| 219 | These are relevant only when support for multiple personal mailboxes is |
| 220 | implemented; see the description in section 3.31 below. |
| 221 | |
| 222 | |
| 223 | 1.4 Installing a filter file |
| 224 | ---------------------------- |
| 225 | |
| 226 | A filter file is normally installed under the name .forward in your home |
| 227 | directory - it is distinguished from a conventional .forward file by its first |
| 228 | line (described below). However, the file name is configurable, and some system |
| 229 | administrators may choose to use some different name or location for filter |
| 230 | files. |
| 231 | |
| 232 | |
| 233 | 1.5 Testing an installed filter file |
| 234 | ------------------------------------ |
| 235 | |
| 236 | Testing a filter file before installation cannot find every potential problem; |
| 237 | for example, it does not actually run commands to which messages are piped. |
| 238 | Some "live" tests should therefore also be done once a filter is installed. |
| 239 | |
| 240 | If at all possible, test your filter file by sending messages from some other |
| 241 | account. If you send a message to yourself from the filtered account, and |
| 242 | delivery fails, the error message will be sent back to the same account, which |
| 243 | may cause another delivery failure. It won't cause an infinite sequence of such |
| 244 | messages, because delivery failure messages do not themselves generate further |
| 245 | messages. However, it does mean that the failure won't be returned to you, and |
| 246 | also that the postmaster will have to investigate the stuck message. |
| 247 | |
| 248 | If you have to test an Exim filter from the same account, a sensible precaution |
| 249 | is to include the line |
| 250 | |
| 251 | if error_message then finish endif |
| 252 | |
| 253 | as the first filter command, at least while testing. This causes filtering to |
| 254 | be abandoned for a delivery failure message, and since no destinations are |
| 255 | generated, the message goes on to be delivered to the original address. Unless |
| 256 | there is a good reason for not doing so, it is recommended that the above test |
| 257 | be left in all Exim filter files. (This does not apply to Sieve files.) |
| 258 | |
| 259 | |
| 260 | 1.6 Details of filtering commands |
| 261 | --------------------------------- |
| 262 | |
| 263 | The filtering commands for Sieve and Exim filters are completely different in |
| 264 | syntax and semantics. The Sieve mechanism is defined in RFC 3028; in the next |
| 265 | chapter we describe how it is integrated into Exim. The subsequent chapter |
| 266 | covers Exim filtering commands in detail. |
| 267 | |
| 268 | |
| 269 | |
| 270 | =============================================================================== |
| 271 | 2. SIEVE FILTER FILES |
| 272 | |
| 273 | The code for Sieve filtering in Exim was contributed by Michael Haardt, and |
| 274 | most of the content of this chapter is taken from the notes he provided. Since |
| 275 | Sieve is an extensible language, it is important to understand "Sieve" in this |
| 276 | context as "the specific implementation of Sieve for Exim". |
| 277 | |
| 278 | This chapter does not contain a description of Sieve, since that can be found |
| 279 | in RFC 3028, which should be read in conjunction with these notes. |
| 280 | |
| 281 | The Exim Sieve implementation offers the core as defined by RFC 3028, |
| 282 | comparison tests, the subaddress parameter, the copy, envelope, fileinto, |
| 283 | notify, and vacation extensions, but not the reject extension. Exim does not |
| 284 | support message delivery notifications (MDNs), so adding it just to the Sieve |
| 285 | filter (as required for reject) makes little sense. |
| 286 | |
| 287 | In order for Sieve to work properly in Exim, the system administrator needs to |
| 288 | make some adjustments to the Exim configuration. These are described in the |
| 289 | chapter on the redirect router in the full Exim specification. |
| 290 | |
| 291 | |
| 292 | 2.1 Recognition of Sieve filters |
| 293 | -------------------------------- |
| 294 | |
| 295 | A filter file is interpreted as a Sieve filter if its first line is |
| 296 | |
| 297 | # Sieve filter |
| 298 | |
| 299 | This is what distinguishes it from a conventional .forward file or an Exim |
| 300 | filter file. |
| 301 | |
| 302 | |
| 303 | 2.2 Saving to specified folders |
| 304 | ------------------------------- |
| 305 | |
| 306 | If the system administrator has set things up as suggested in the Exim |
| 307 | specification, and you use keep or fileinto to save a mail into a folder, |
| 308 | absolute files are stored where specified, relative files are stored relative |
| 309 | to $home, and inbox goes to the standard mailbox location. |
| 310 | |
| 311 | |
| 312 | 2.3 Strings containing header names |
| 313 | ----------------------------------- |
| 314 | |
| 315 | RFC 3028 does not specify what happens if a string denoting a header field does |
| 316 | not contain a valid header name, for example, it contains a colon. This |
| 317 | implementation generates an error instead of ignoring the header field in order |
| 318 | to ease script debugging, which fits in with the common picture of Sieve. |
| 319 | |
| 320 | |
| 321 | 2.4 Exists test with empty list of headers |
| 322 | ------------------------------------------ |
| 323 | |
| 324 | The exists test succeeds only if all the specified headers exist. RFC 3028 does |
| 325 | not explicitly specify what happens on an empty list of headers. This |
| 326 | implementation evaluates that condition as true, interpreting the RFC in a |
| 327 | strict sense. |
| 328 | |
| 329 | |
| 330 | 2.5 Header test with invalid MIME encoding in header |
| 331 | ---------------------------------------------------- |
| 332 | |
| 333 | Some MUAs process invalid base64 encoded data, generating junk. Others ignore |
| 334 | junk after seeing an equal sign in base64 encoded data. RFC 2047 does not |
| 335 | specify how to react in this case, other than stating that a client must not |
| 336 | forbid to process a message for that reason. RFC 2045 specifies that invalid |
| 337 | data should be ignored (apparently looking at end of line characters). It also |
| 338 | specifies that invalid data may lead to rejecting messages containing them (and |
| 339 | there it appears to talk about true encoding violations), which is a clear |
| 340 | contradiction to ignoring them. |
| 341 | |
| 342 | RFC 3028 does not specify how to process incorrect MIME words. This |
| 343 | implementation treats them literally, as it does if the word is correct but its |
| 344 | character set cannot be converted to UTF-8. |
| 345 | |
| 346 | |
| 347 | 2.6 Address test for multiple addresses per header |
| 348 | -------------------------------------------------- |
| 349 | |
| 350 | A header may contain multiple addresses. RFC 3028 does not explicitly specify |
| 351 | how to deal with them, but since the address test checks if anything matches |
| 352 | anything else, matching one address suffices to satisfy the condition. That |
| 353 | makes it impossible to test if a header contains a certain set of addresses and |
| 354 | no more, but it is more logical than letting the test fail if the header |
| 355 | contains an additional address besides the one the test checks for. |
| 356 | |
| 357 | |
| 358 | 2.7 Semantics of keep |
| 359 | --------------------- |
| 360 | |
| 361 | The keep command is equivalent to |
| 362 | |
| 363 | fileinto "inbox"; |
| 364 | |
| 365 | It saves the message and resets the implicit keep flag. It does not set the |
| 366 | implicit keep flag; there is no command to set it once it has been reset. |
| 367 | |
| 368 | |
| 369 | 2.8 Semantics of fileinto |
| 370 | ------------------------- |
| 371 | |
| 372 | RFC 3028 does not specify whether fileinto should try to create a mail folder |
| 373 | if it does not exist. This implementation allows the sysadmin to configure that |
| 374 | aspect using the appendfile transport options create_directory, create_file, |
| 375 | and file_must_exist. See the appendfile transport in the Exim specification for |
| 376 | details. |
| 377 | |
| 378 | |
| 379 | 2.9 Semantics of redirect |
| 380 | ------------------------- |
| 381 | |
| 382 | Sieve scripts are supposed to be interoperable between servers, so this |
| 383 | implementation does not allow mail to be redirected to unqualified addresses, |
| 384 | because the domain would depend on the system being used. On systems with |
| 385 | virtual mail domains, the default domain is probably not what the user expects |
| 386 | it to be. |
| 387 | |
| 388 | |
| 389 | 2.10 String arguments |
| 390 | --------------------- |
| 391 | |
| 392 | There has been confusion if the string arguments to require are to be matched |
| 393 | case-sensitively or not. This implementation matches them with the match type |
| 394 | :is (default, see section 2.7.1 of the RFC) and the comparator i;ascii-casemap |
| 395 | (default, see section 2.7.3 of the RFC). The RFC defines the command defaults |
| 396 | clearly, so any different implementations violate RFC 3028. The same is valid |
| 397 | for comparator names, also specified as strings. |
| 398 | |
| 399 | |
| 400 | 2.11 Number units |
| 401 | ----------------- |
| 402 | |
| 403 | There is a mistake in RFC 3028: the suffix G denotes gibi-, not tebibyte. The |
| 404 | mistake is obvious, because RFC 3028 specifies G to denote 2^30 (which is gibi, |
| 405 | not tebi), and that is what this implementation uses as the scaling factor for |
| 406 | the suffix G. |
| 407 | |
| 408 | |
| 409 | 2.12 RFC compliance |
| 410 | ------------------- |
| 411 | |
| 412 | Exim requires the first line of a Sieve filter to be |
| 413 | |
| 414 | # Sieve filter |
| 415 | |
| 416 | Of course the RFC does not specify that line. Do not expect examples to work |
| 417 | without adding it, though. |
| 418 | |
| 419 | RFC 3028 requires the use of CRLF to terminate a line. The rationale was that |
| 420 | CRLF is universally used in network protocols to mark the end of the line. This |
| 421 | implementation does not embed Sieve in a network protocol, but uses Sieve |
| 422 | scripts as part of the Exim MTA. Since all parts of Exim use LF as the newline |
| 423 | character, this implementation does, too, by default, though the system |
| 424 | administrator may choose (at Exim compile time) to use CRLF instead. |
| 425 | |
| 426 | Exim violates RFC 2822, section 3.6.8, by accepting 8-bit header names, so this |
| 427 | implementation repeats this violation to stay consistent with Exim. This is in |
| 428 | preparation for UTF-8 data. |
| 429 | |
| 430 | Sieve scripts cannot contain NUL characters in strings, but mail headers could |
| 431 | contain MIME encoded NUL characters, which could never be matched by Sieve |
| 432 | scripts using exact comparisons. For that reason, this implementation extends |
| 433 | the Sieve quoted string syntax with \0 to describe a NUL character, violating \ |
| 434 | 0 being the same as 0 in RFC 3028. Even without using \0, the following tests |
| 435 | are all true in this implementation. Implementations that use C-style strings |
| 436 | will only evaluate the first test as true. |
| 437 | |
| 438 | Subject: =?iso-8859-1?q?abc=00def |
| 439 | |
| 440 | header :contains "Subject" ["abc"] |
| 441 | header :contains "Subject" ["def"] |
| 442 | header :matches "Subject" ["abc?def"] |
| 443 | |
| 444 | Note that by considering Sieve to be an MUA, RFC 2047 can be interpreted in a |
| 445 | way that NUL characters truncating strings is allowed for Sieve |
| 446 | implementations, although not recommended. It is further allowed to use encoded |
| 447 | NUL characters in headers, but that's not recommended either. The above example |
| 448 | shows why. |
| 449 | |
| 450 | RFC 3028 states that if an implementation fails to convert a character set to |
| 451 | UTF-8, two strings cannot be equal if one contains octets greater than 127. |
| 452 | Assuming that all unknown character sets are one-byte character sets with the |
| 453 | lower 128 octets being US-ASCII is not sound, so this implementation violates |
| 454 | RFC 3028 and treats such MIME words literally. That way at least something |
| 455 | could be matched. |
| 456 | |
| 457 | The folder specified by fileinto must not contain the character sequence ".." |
| 458 | to avoid security problems. RFC 3028 does not specify the syntax of folders |
| 459 | apart from keep being equivalent to |
| 460 | |
| 461 | fileinto "INBOX"; |
| 462 | |
| 463 | This implementation uses inbox instead. |
| 464 | |
| 465 | Sieve script errors currently cause messages to be silently filed into inbox. |
| 466 | RFC 3028 requires that the user is notified of that condition. This may be |
| 467 | implemented in the future by adding a header line to mails that are filed into |
| 468 | inbox due to an error in the filter. |
| 469 | |
| 470 | |
| 471 | |
| 472 | =============================================================================== |
| 473 | 3. EXIM FILTER FILES |
| 474 | |
| 475 | This chapter contains a full description of the contents of Exim filter files. |
| 476 | |
| 477 | |
| 478 | 3.1 Format of Exim filter files |
| 479 | ------------------------------- |
| 480 | |
| 481 | Apart from leading white space, the first text in an Exim filter file must be |
| 482 | |
| 483 | # Exim filter |
| 484 | |
| 485 | This is what distinguishes it from a conventional .forward file or a Sieve |
| 486 | filter file. If the file does not have this initial line (or the equivalent for |
| 487 | a Sieve filter), it is treated as a conventional .forward file, both when |
| 488 | delivering mail and when using the -bf testing mechanism. The white space in |
| 489 | the line is optional, and any capitalization may be used. Further text on the |
| 490 | same line is treated as a comment. For example, you could have |
| 491 | |
| 492 | # Exim filter <<== do not edit or remove this line! |
| 493 | |
| 494 | The remainder of the file is a sequence of filtering commands, which consist of |
| 495 | keywords and data values. For example, in the command |
| 496 | |
| 497 | deliver gulliver@lilliput.fict.example |
| 498 | |
| 499 | the keyword is "deliver" and the data value is |
| 500 | "gulliver@lilliput.fict.example". White space or line breaks separate the |
| 501 | components of a command, except in the case of conditions for the if command, |
| 502 | where round brackets (parentheses) also act as separators. Complete commands |
| 503 | are separated from each other by white space or line breaks; there are no |
| 504 | special terminators. Thus, several commands may appear on one line, or one |
| 505 | command may be spread over a number of lines. |
| 506 | |
| 507 | If the character # follows a separator anywhere in a command, everything from # |
| 508 | up to the next newline is ignored. This provides a way of including comments in |
| 509 | a filter file. |
| 510 | |
| 511 | |
| 512 | 3.2 Data values in filter commands |
| 513 | ---------------------------------- |
| 514 | |
| 515 | There are two ways in which a data value can be input: |
| 516 | |
| 517 | * If the text contains no white space, it can be typed verbatim. However, if |
| 518 | it is part of a condition, it must also be free of round brackets |
| 519 | (parentheses), as these are used for grouping in conditions. |
| 520 | |
| 521 | * Otherwise, text must be enclosed in double quotation marks. In this case, |
| 522 | the character \ (backslash) is treated as an "escape character" within the |
| 523 | string, causing the following character or characters to be treated |
| 524 | specially: |
| 525 | |
| 526 | \n is replaced by a newline |
| 527 | \r is replaced by a carriage return |
| 528 | \t is replaced by a tab |
| 529 | |
| 530 | Backslash followed by up to three octal digits is replaced by the character |
| 531 | specified by those digits, and "\x" followed by up to two hexadecimal digits is |
| 532 | treated similarly. Backslash followed by any other character is replaced by the |
| 533 | second character, so that in particular, "\"" becomes """ and "\\" becomes "\". |
| 534 | A data item enclosed in double quotes can be continued onto the next line by |
| 535 | ending the first line with a backslash. Any leading white space at the start of |
| 536 | the continuation line is ignored. |
| 537 | |
| 538 | In addition to the escape character processing that occurs when strings are |
| 539 | enclosed in quotes, most data values are also subject to string expansion (as |
| 540 | described in the next section), in which case the characters "$" and "\" are |
| 541 | also significant. This means that if a single backslash is actually required in |
| 542 | such a string, and the string is also quoted, "\\\\" has to be entered. |
| 543 | |
| 544 | The maximum permitted length of a data string, before expansion, is 1024 |
| 545 | characters. |
| 546 | |
| 547 | |
| 548 | 3.3 String expansion |
| 549 | -------------------- |
| 550 | |
| 551 | Most data values are expanded before use. Expansion consists of replacing |
| 552 | substrings beginning with "$" with other text. The full expansion facilities |
| 553 | available in Exim are extensive. If you want to know everything that Exim can |
| 554 | do with strings, you should consult the chapter on string expansion in the Exim |
| 555 | documentation. |
| 556 | |
| 557 | In filter files, by far the most common use of string expansion is the |
| 558 | substitution of the contents of a variable. For example, the substring |
| 559 | |
| 560 | $reply_address |
| 561 | |
| 562 | is replaced by the address to which replies to the message should be sent. If |
| 563 | such a variable name is followed by a letter or digit or underscore, it must be |
| 564 | enclosed in curly brackets (braces), for example, |
| 565 | |
| 566 | ${reply_address} |
| 567 | |
| 568 | If a "$" character is actually required in an expanded string, it must be |
| 569 | escaped with a backslash, and because backslash is also an escape character in |
| 570 | quoted input strings, it must be doubled in that case. The following two |
| 571 | examples illustrate two different ways of testing for a "$" character in a |
| 572 | message: |
| 573 | |
| 574 | if $message_body contains \$ then ... |
| 575 | if $message_body contains "\\$" then ... |
| 576 | |
| 577 | You can prevent part of a string from being expanded by enclosing it between |
| 578 | two occurrences of "\N". For example, |
| 579 | |
| 580 | if $message_body contains \N$$$$\N then ... |
| 581 | |
| 582 | tests for a run of four dollar characters. |
| 583 | |
| 584 | |
| 585 | 3.4 Some useful general variables |
| 586 | --------------------------------- |
| 587 | |
| 588 | A complete list of the available variables is given in the Exim documentation. |
| 589 | This shortened list contains the ones that are most likely to be useful in |
| 590 | personal filter files: |
| 591 | |
| 592 | $body_linecount: The number of lines in the body of the message. |
| 593 | |
| 594 | $body_zerocount: The number of binary zero characters in the body of the |
| 595 | message. |
| 596 | |
| 597 | $home: In conventional configurations, this variable normally contains the |
| 598 | user's home directory. The system administrator can, however, change this. |
| 599 | |
| 600 | $local_part: The part of the email address that precedes the @ sign - normally |
| 601 | the user's login name. If support for multiple personal mailboxes is enabled |
| 602 | (see section 3.31 below) and a prefix or suffix for the local part was |
| 603 | recognized, it is removed from the string in this variable. |
| 604 | |
| 605 | $local_part_prefix: If support for multiple personal mailboxes is enabled (see |
| 606 | section 3.31 below), and a local part prefix was recognized, this variable |
| 607 | contains the prefix. Otherwise it contains an empty string. |
| 608 | |
| 609 | $local_part_suffix: If support for multiple personal mailboxes is enabled (see |
| 610 | section 3.31 below), and a local part suffix was recognized, this variable |
| 611 | contains the suffix. Otherwise it contains an empty string. |
| 612 | |
| 613 | $message_body: The initial portion of the body of the message. By default, up |
| 614 | to 500 characters are read into this variable, but the system administrator can |
| 615 | configure this to some other value. Newlines in the body are converted into |
| 616 | single spaces. |
| 617 | |
| 618 | $message_body_end: The final portion of the body of the message, formatted and |
| 619 | limited in the same way as $message_body. |
| 620 | |
| 621 | $message_body_size: The size of the body of the message, in bytes. |
| 622 | |
| 623 | $message_exim_id: The message's local identification string, which is unique |
| 624 | for each message handled by a single host. |
| 625 | |
| 626 | $message_headers: The header lines of the message, concatenated into a single |
| 627 | string, with newline characters between them. |
| 628 | |
| 629 | $message_size: The size of the entire message, in bytes. |
| 630 | |
| 631 | $original_local_part: When an address that arrived with the message is being |
| 632 | processed, this contains the same value as the variable $local_part. However, |
| 633 | if an address generated by an alias, forward, or filter file is being |
| 634 | processed, this variable contains the local part of the original address. |
| 635 | |
| 636 | $reply_address: The contents of the Reply-to: header, if the message has one; |
| 637 | otherwise the contents of the From: header. It is the address to which normal |
| 638 | replies to the message should be sent. |
| 639 | |
| 640 | $return_path: The return path - that is, the sender field that will be |
| 641 | transmitted as part of the message's envelope if the message is sent to another |
| 642 | host. This is the address to which delivery errors are sent. In many cases, |
| 643 | this variable has the same value as $sender_address, but if, for example, an |
| 644 | incoming message to a mailing list has been expanded, $return_path may have |
| 645 | been changed to contain the address of the list maintainer. |
| 646 | |
| 647 | $sender_address: The sender address that was received in the envelope of the |
| 648 | message. This is not necessarily the same as the contents of the From: or |
| 649 | Sender: header lines. For delivery error messages ("bounce messages") there is |
| 650 | no sender address, and this variable is empty. |
| 651 | |
| 652 | $tod_full: A full version of the time and date, for example: Wed, 18 Oct 1995 |
| 653 | 09:51:40 +0100. The timezone is always given as a numerical offset from GMT. |
| 654 | |
| 655 | $tod_log: The time and date in the format used for writing Exim's log files, |
| 656 | without the timezone, for example: 1995-10-12 15:32:29. |
| 657 | |
| 658 | $tod_zone: The local timezone offset, for example: +0100. |
| 659 | |
| 660 | |
| 661 | 3.5 Header variables |
| 662 | -------------------- |
| 663 | |
| 664 | There is a special set of expansion variables containing the header lines of |
| 665 | the message being processed. These variables have names beginning with $header_ |
| 666 | followed by the name of the header line, terminated by a colon. For example, |
| 667 | |
| 668 | $header_from: |
| 669 | $header_subject: |
| 670 | |
| 671 | The whole item, including the terminating colon, is replaced by the contents of |
| 672 | the message header line. If there is more than one header line with the same |
| 673 | name, their contents are concatenated. For header lines whose data consists of |
| 674 | a list of addresses (for example, From: and To:), a comma and newline is |
| 675 | inserted between each set of data. For all other header lines, just a newline |
| 676 | is used. |
| 677 | |
| 678 | Leading and trailing white space is removed from header line data, and if there |
| 679 | are any MIME "words" that are encoded as defined by RFC 2047 (because they |
| 680 | contain non-ASCII characters), they are decoded and translated, if possible, to |
| 681 | a local character set. Translation is attempted only on operating systems that |
| 682 | have the iconv() function. This makes the header line look the same as it would |
| 683 | when displayed by an MUA. The default character set is ISO-8859-1, but this can |
| 684 | be changed by means of the headers command (see below). |
| 685 | |
| 686 | If you want to see the actual characters that make up a header line, you can |
| 687 | specify $rheader_ instead of $header_. This inserts the "raw" header line, |
| 688 | unmodified. |
| 689 | |
| 690 | There is also an intermediate form, requested by $bheader_, which removes |
| 691 | leading and trailing space and decodes MIME "words", but does not do any |
| 692 | character translation. If an attempt to decode what looks superficially like a |
| 693 | MIME "word" fails, the raw string is returned. If decoding produces a binary |
| 694 | zero character, it is replaced by a question mark. |
| 695 | |
| 696 | The capitalization of the name following $header_ is not significant. Because |
| 697 | any printing character except colon may appear in the name of a message's |
| 698 | header (this is a requirement of RFC 2822, the document that describes the |
| 699 | format of a mail message) curly brackets must not be used in this case, as they |
| 700 | will be taken as part of the header name. Two shortcuts are allowed in naming |
| 701 | header variables: |
| 702 | |
| 703 | * The initiating $header_, $rheader_, or $bheader_ can be abbreviated to $h_, |
| 704 | $rh_, or $bh_, respectively. |
| 705 | |
| 706 | * The terminating colon can be omitted if the next character is white space. |
| 707 | The white space character is retained in the expanded string. However, this |
| 708 | is not recommended, because it makes it easy to forget the colon when it |
| 709 | really is needed. |
| 710 | |
| 711 | If the message does not contain a header of the given name, an empty string is |
| 712 | substituted. Thus it is important to spell the names of headers correctly. Do |
| 713 | not use $header_Reply_to when you really mean $header_Reply-to. |
| 714 | |
| 715 | |
| 716 | 3.6 User variables |
| 717 | ------------------ |
| 718 | |
| 719 | There are ten user variables with names $n0 - $n9 that can be incremented by |
| 720 | the add command (see section 3.10). These can be used for "scoring" messages in |
| 721 | various ways. If Exim is configured to run a "system filter" on every message, |
| 722 | the values left in these variables are copied into the variables $sn0 - $sn9 at |
| 723 | the end of the system filter, thus making them available to users' filter |
| 724 | files. How these values are used is entirely up to the individual installation. |
| 725 | |
| 726 | |
| 727 | 3.7 Current directory |
| 728 | --------------------- |
| 729 | |
| 730 | The contents of your filter file should not make any assumptions about the |
| 731 | current directory. It is best to use absolute paths for file names; you can |
| 732 | normally make use of the $home variable to refer to your home directory. The |
| 733 | save command automatically inserts $home at the start of non-absolute paths. |
| 734 | |
| 735 | |
| 736 | 3.8 Significant deliveries |
| 737 | -------------------------- |
| 738 | |
| 739 | When in the course of delivery a message is processed by a filter file, what |
| 740 | happens next, that is, after the filter file has been processed, depends on |
| 741 | whether or not the filter sets up any significant deliveries. If at least one |
| 742 | significant delivery is set up, the filter is considered to have handled the |
| 743 | entire delivery arrangements for the current address, and no further processing |
| 744 | of the address takes place. If, however, no significant deliveries are set up, |
| 745 | Exim continues processing the current address as if there were no filter file, |
| 746 | and typically sets up a delivery of a copy of the message into a local mailbox. |
| 747 | In particular, this happens in the special case of a filter file containing |
| 748 | only comments. |
| 749 | |
| 750 | The delivery commands deliver, save, and pipe are by default significant. |
| 751 | However, if such a command is preceded by the word "unseen", its delivery is |
| 752 | not considered to be significant. In contrast, other commands such as mail and |
| 753 | vacation do not set up significant deliveries unless preceded by the word |
| 754 | "seen". The following example commands set up significant deliveries: |
| 755 | |
| 756 | deliver jack@beanstalk.example |
| 757 | pipe $home/bin/mymailscript |
| 758 | seen mail subject "message discarded" |
| 759 | seen finish |
| 760 | |
| 761 | The following example commands do not set up significant deliveries: |
| 762 | |
| 763 | unseen deliver jack@beanstalk.example |
| 764 | unseen pipe $home/bin/mymailscript |
| 765 | mail subject "message discarded" |
| 766 | finish |
| 767 | |
| 768 | |
| 769 | 3.9 Filter commands |
| 770 | ------------------- |
| 771 | |
| 772 | The filter commands that are described in subsequent sections are listed below, |
| 773 | with the section in which they are described in brackets: |
| 774 | |
| 775 | add increment a user variable (section 3.10) |
| 776 | deliver deliver to an email address (section 3.11) |
| 777 | fail force delivery failure (sysadmin use) (section 3.18) |
| 778 | finish end processing (section 3.16) |
| 779 | freeze freeze message (sysadmin use) (section 3.19) |
| 780 | headers set the header character set (section 3.20) |
| 781 | if test condition(s) (section 3.21) |
| 782 | logfile define log file (section 3.15) |
| 783 | logwrite write to log file (section 3.15) |
| 784 | mail send a reply message (section 3.14) |
| 785 | pipe pipe to a command (section 3.13) |
| 786 | save save to a file (section 3.12) |
| 787 | testprint print while testing (section 3.17) |
| 788 | vacation tailored form of mail (section 3.14) |
| 789 | |
| 790 | The headers command has additional parameters that can be used only in a system |
| 791 | filter. The fail and freeze commands are available only when Exim's filtering |
| 792 | facilities are being used as a system filter, and are therefore usable only by |
| 793 | the system administrator and not by ordinary users. They are mentioned only |
| 794 | briefly in this document; for more information, see the main Exim |
| 795 | specification. |
| 796 | |
| 797 | |
| 798 | 3.10 The add command |
| 799 | -------------------- |
| 800 | |
| 801 | add <number> to <user variable> |
| 802 | e.g. add 2 to n3 |
| 803 | |
| 804 | There are 10 user variables of this type, with names $n0 - $n9. Their values |
| 805 | can be obtained by the normal expansion syntax (for example $n3) in other |
| 806 | commands. At the start of filtering, these variables all contain zero. Both |
| 807 | arguments of the add command are expanded before use, making it possible to add |
| 808 | variables to each other. Subtraction can be obtained by adding negative |
| 809 | numbers. |
| 810 | |
| 811 | |
| 812 | 3.11 The deliver command |
| 813 | ------------------------ |
| 814 | |
| 815 | deliver <mail address> |
| 816 | e.g. deliver "Dr Livingstone <David@somewhere.africa.example>" |
| 817 | |
| 818 | This command provides a forwarding operation. The delivery that it sets up is |
| 819 | significant unless the command is preceded by "unseen" (see section 3.8). The |
| 820 | message is sent on to the given address, exactly as happens if the address had |
| 821 | appeared in a traditional .forward file. If you want to deliver the message to |
| 822 | a number of different addresses, you can use more than one deliver command |
| 823 | (each one may have only one address). However, duplicate addresses are |
| 824 | discarded. |
| 825 | |
| 826 | To deliver a copy of the message to your normal mailbox, your login name can be |
| 827 | given as the address. Once an address has been processed by the filtering |
| 828 | mechanism, an identical generated address will not be so processed again, so |
| 829 | doing this does not cause a loop. |
| 830 | |
| 831 | However, if you have a mail alias, you should not refer to it here. For |
| 832 | example, if the mail address L.Gulliver is aliased to lg303 then all references |
| 833 | in Gulliver's .forward file should be to lg303. A reference to the alias will |
| 834 | not work for messages that are addressed to that alias, since, like .forward |
| 835 | file processing, aliasing is performed only once on an address, in order to |
| 836 | avoid looping. |
| 837 | |
| 838 | Following the new address, an optional second address, preceded by "errors_to" |
| 839 | may appear. This changes the address to which delivery errors on the forwarded |
| 840 | message will be sent. Instead of going to the message's original sender, they |
| 841 | go to this new address. For ordinary users, the only value that is permitted |
| 842 | for this address is the user whose filter file is being processed. For example, |
| 843 | the user lg303 whose mailbox is in the domain lilliput.example could have a |
| 844 | filter file that contains |
| 845 | |
| 846 | deliver jon@elsewhere.example errors_to lg303@lilliput.example |
| 847 | |
| 848 | Clearly, using this feature makes sense only in situations where not all |
| 849 | messages are being forwarded. In particular, bounce messages must not be |
| 850 | forwarded in this way, as this is likely to create a mail loop if something |
| 851 | goes wrong. |
| 852 | |
| 853 | |
| 854 | 3.12 The save command |
| 855 | --------------------- |
| 856 | |
| 857 | save <file name> |
| 858 | e.g. save $home/mail/bookfolder |
| 859 | |
| 860 | This command specifies that a copy of the message is to be appended to the |
| 861 | given file (that is, the file is to be used as a mail folder). The delivery |
| 862 | that save sets up is significant unless the command is preceded by "unseen" |
| 863 | (see section 3.8). |
| 864 | |
| 865 | More than one save command may be obeyed; each one causes a copy of the message |
| 866 | to be written to its argument file, provided they are different (duplicate save |
| 867 | commands are ignored). |
| 868 | |
| 869 | If the file name does not start with a / character, the contents of the $home |
| 870 | variable are prepended, unless it is empty, or the system administrator has |
| 871 | disabled this feature. In conventional configurations, this variable is |
| 872 | normally set in a user filter to the user's home directory, but the system |
| 873 | administrator may set it to some other path. In some configurations, $home may |
| 874 | be unset, or prepending may be disabled, in which case a non-absolute path name |
| 875 | may be generated. Such configurations convert this to an absolute path when the |
| 876 | delivery takes place. In a system filter, $home is never set. |
| 877 | |
| 878 | The user must of course have permission to write to the file, and the writing |
| 879 | of the file takes place in a process that is running as the user, under the |
| 880 | user's primary group. Any secondary groups to which the user may belong are not |
| 881 | normally taken into account, though the system administrator can configure Exim |
| 882 | to set them up. In addition, the ability to use this command at all is |
| 883 | controlled by the system administrator - it may be forbidden on some systems. |
| 884 | |
| 885 | An optional mode value may be given after the file name. The value for the mode |
| 886 | is interpreted as an octal number, even if it does not begin with a zero. For |
| 887 | example: |
| 888 | |
| 889 | save /some/folder 640 |
| 890 | |
| 891 | This makes it possible for users to override the system-wide mode setting for |
| 892 | file deliveries, which is normally 600. If an existing file does not have the |
| 893 | correct mode, it is changed. |
| 894 | |
| 895 | An alternative form of delivery may be enabled on your system, in which each |
| 896 | message is delivered into a new file in a given directory. If this is the case, |
| 897 | this functionality can be requested by giving the directory name terminated by |
| 898 | a slash after the save command, for example |
| 899 | |
| 900 | save separated/messages/ |
| 901 | |
| 902 | There are several different formats for such deliveries; check with your system |
| 903 | administrator or local documentation to find out which (if any) are available |
| 904 | on your system. If this functionality is not enabled, the use of a path name |
| 905 | ending in a slash causes an error. |
| 906 | |
| 907 | |
| 908 | 3.13 The pipe command |
| 909 | --------------------- |
| 910 | |
| 911 | pipe <command> |
| 912 | e.g. pipe "$home/bin/countmail $sender_address" |
| 913 | |
| 914 | This command specifies that the message is to be delivered to the specified |
| 915 | command using a pipe. The delivery that it sets up is significant unless the |
| 916 | command is preceded by "unseen" (see section 3.8). Remember, however, that no |
| 917 | deliveries are done while the filter is being processed. All deliveries happen |
| 918 | later on. Therefore, the result of running the pipe is not available to the |
| 919 | filter. |
| 920 | |
| 921 | When the deliveries are done, a separate process is run, and a copy of the |
| 922 | message is passed on its standard input. The process runs as the user, under |
| 923 | the user's primary group. Any secondary groups to which the user may belong are |
| 924 | not normally taken into account, though the system administrator can configure |
| 925 | Exim to set them up. More than one pipe command may appear; each one causes a |
| 926 | copy of the message to be written to its argument pipe, provided they are |
| 927 | different (duplicate pipe commands are ignored). |
| 928 | |
| 929 | When the time comes to transport the message, the command supplied to pipe is |
| 930 | split up by Exim into a command name and a number of arguments. These are |
| 931 | delimited by white space except for arguments enclosed in double quotes, in |
| 932 | which case backslash is interpreted as an escape, or in single quotes, in which |
| 933 | case no escaping is recognized. Note that as the whole command is normally |
| 934 | supplied in double quotes, a second level of quoting is required for internal |
| 935 | double quotes. For example: |
| 936 | |
| 937 | pipe "$home/myscript \"size is $message_size\"" |
| 938 | |
| 939 | String expansion is performed on the separate components after the line has |
| 940 | been split up, and the command is then run directly by Exim; it is not run |
| 941 | under a shell. Therefore, substitution cannot change the number of arguments, |
| 942 | nor can quotes, backslashes or other shell metacharacters in variables cause |
| 943 | confusion. |
| 944 | |
| 945 | Documentation for some programs that are normally run via this kind of pipe |
| 946 | often suggest that the command should start with |
| 947 | |
| 948 | IFS=" " |
| 949 | |
| 950 | This is a shell command, and should not be present in Exim filter files, since |
| 951 | it does not normally run the command under a shell. |
| 952 | |
| 953 | However, there is an option that the administrator can set to cause a shell to |
| 954 | be used. In this case, the entire command is expanded as a single string and |
| 955 | passed to the shell for interpretation. It is recommended that this be avoided |
| 956 | if at all possible, since it can lead to problems when inserted variables |
| 957 | contain shell metacharacters. |
| 958 | |
| 959 | The default PATH set up for the command is determined by the system |
| 960 | administrator, usually containing at least /bin and /usr/bin so that common |
| 961 | commands are available without having to specify an absolute file name. |
| 962 | However, it is possible for the system administrator to restrict the pipe |
| 963 | facility so that the command name must not contain any / characters, and must |
| 964 | be found in one of the directories in the configured PATH. It is also possible |
| 965 | for the system administrator to lock out the use of the pipe command |
| 966 | altogether. |
| 967 | |
| 968 | When the command is run, a number of environment variables are set up. The |
| 969 | complete list for pipe deliveries may be found in the Exim reference manual. |
| 970 | Those that may be useful for pipe deliveries from user filter files are: |
| 971 | |
| 972 | DOMAIN the domain of the address |
| 973 | HOME your home directory |
| 974 | LOCAL_PART see below |
| 975 | LOCAL_PART_PREFIX see below |
| 976 | LOCAL_PART_SUFFIX see below |
| 977 | LOGNAME your login name |
| 978 | MESSAGE_ID the unique id of the message |
| 979 | PATH the command search path |
| 980 | RECIPIENT the complete recipient address |
| 981 | SENDER the sender of the message |
| 982 | SHELL /bin/sh |
| 983 | USER see below |
| 984 | |
| 985 | LOCAL_PART, LOGNAME, and USER are all set to the same value, namely, your login |
| 986 | id. LOCAL_PART_PREFIX and LOCAL_PART_SUFFIX may be set if Exim is configured to |
| 987 | recognize prefixes or suffixes in the local parts of addresses. For example, a |
| 988 | message addressed to pat-suf2@domain.example may cause the filter for user pat |
| 989 | to be run. If this sets up a pipe delivery, LOCAL_PART_SUFFIX is "-suf2" when |
| 990 | the pipe command runs. The system administrator has to configure Exim specially |
| 991 | for this feature to be available. |
| 992 | |
| 993 | If you run a command that is a shell script, be very careful in your use of |
| 994 | data from the incoming message in the commands in your script. RFC 2822 is very |
| 995 | generous in the characters that are permitted to appear in mail addresses, and |
| 996 | in particular, an address may begin with a vertical bar or a slash. For this |
| 997 | reason you should always use quotes round any arguments that involve data from |
| 998 | the message, like this: |
| 999 | |
| 1000 | /some/command '$SENDER' |
| 1001 | |
| 1002 | so that inserted shell meta-characters do not cause unwanted effects. |
| 1003 | |
| 1004 | Remember that, as was explained earlier, the pipe command is not run at the |
| 1005 | time the filter file is interpreted. The filter just defines what deliveries |
| 1006 | are required for one particular addressee of a message. The deliveries |
| 1007 | themselves happen later, once Exim has decided everything that needs to be done |
| 1008 | for the message. |
| 1009 | |
| 1010 | A consequence of this is that you cannot inspect the return code from the pipe |
| 1011 | command from within the filter. Nevertheless, the code returned by the command |
| 1012 | is important, because Exim uses it to decide whether the delivery has succeeded |
| 1013 | or failed. |
| 1014 | |
| 1015 | The command should return a zero completion code if all has gone well. Most |
| 1016 | non-zero codes are treated by Exim as indicating a failure of the pipe. This is |
| 1017 | treated as a delivery failure, causing the message to be returned to its |
| 1018 | sender. However, there are some completion codes that are treated as temporary |
| 1019 | errors. The message remains on Exim's spool disk, and the delivery is tried |
| 1020 | again later, though it will ultimately time out if the delivery failures go on |
| 1021 | too long. The completion codes to which this applies can be specified by the |
| 1022 | system administrator; the default values are 73 and 75. |
| 1023 | |
| 1024 | The pipe command should not normally write anything to its standard output or |
| 1025 | standard error file descriptors. If it does, whatever is written is normally |
| 1026 | returned to the sender of the message as a delivery error, though this action |
| 1027 | can be varied by the system administrator. |
| 1028 | |
| 1029 | |
| 1030 | 3.14 Mail commands |
| 1031 | ------------------ |
| 1032 | |
| 1033 | There are two commands that cause the creation of a new mail message, neither |
| 1034 | of which count as a significant delivery unless the command is preceded by the |
| 1035 | word "seen" (see section 3.8). This is a powerful facility, but it should be |
| 1036 | used with care, because of the danger of creating infinite sequences of |
| 1037 | messages. The system administrator can forbid the use of these commands |
| 1038 | altogether. |
| 1039 | |
| 1040 | To help prevent runaway message sequences, these commands have no effect when |
| 1041 | the incoming message is a bounce (delivery error) message, and messages sent by |
| 1042 | this means are treated as if they were reporting delivery errors. Thus, they |
| 1043 | should never themselves cause a bounce message to be returned. The basic |
| 1044 | mail-sending command is |
| 1045 | |
| 1046 | mail [to <address-list>] |
| 1047 | [cc <address-list>] |
| 1048 | [bcc <address-list>] |
| 1049 | [from <address>] |
| 1050 | [reply_to <address>] |
| 1051 | [subject <text>] |
| 1052 | [extra_headers <text>] |
| 1053 | [text <text>] |
| 1054 | [[expand] file <filename>] |
| 1055 | [return message] |
| 1056 | [log <log file name>] |
| 1057 | [once <note file name>] |
| 1058 | [once_repeat <time interval>] |
| 1059 | e.g. mail text "Your message about $h_subject: has been received" |
| 1060 | |
| 1061 | Each <address-list> can contain a number of addresses, separated by commas, in |
| 1062 | the format of a To: or Cc: header line. In fact, the text you supply here is |
| 1063 | copied exactly into the appropriate header line. It may contain additional |
| 1064 | information as well as email addresses. For example: |
| 1065 | |
| 1066 | mail to "Julius Caesar <jc@rome.example>, \ |
| 1067 | <ma@rome.example> (Mark A.)" |
| 1068 | |
| 1069 | Similarly, the texts supplied for from and reply_to are copied into their |
| 1070 | respective header lines. |
| 1071 | |
| 1072 | As a convenience for use in one common case, there is also a command called |
| 1073 | vacation. It behaves in the same way as mail, except that the defaults for the |
| 1074 | subject, file, log, once, and once_repeat options are |
| 1075 | |
| 1076 | subject "On vacation" |
| 1077 | expand file .vacation.msg |
| 1078 | log .vacation.log |
| 1079 | once .vacation |
| 1080 | once_repeat 7d |
| 1081 | |
| 1082 | respectively. These are the same file names and repeat period used by the |
| 1083 | traditional Unix vacation command. The defaults can be overridden by explicit |
| 1084 | settings, but if a file name is given its contents are expanded only if |
| 1085 | explicitly requested. |
| 1086 | |
| 1087 | Warning: The vacation command should always be used conditionally, subject to |
| 1088 | at least the personal condition (see section 3.27 below) so as not to send |
| 1089 | automatic replies to non-personal messages from mailing lists or elsewhere. |
| 1090 | Sending an automatic response to a mailing list or a mailing list manager is an |
| 1091 | Internet Sin. |
| 1092 | |
| 1093 | For both commands, the key/value argument pairs can appear in any order. At |
| 1094 | least one of text or file must appear (except with vacation, where there is a |
| 1095 | default for file); if both are present, the text string appears first in the |
| 1096 | message. If expand precedes file, each line of the file is subject to string |
| 1097 | expansion before it is included in the message. |
| 1098 | |
| 1099 | Several lines of text can be supplied to text by including the escape sequence |
| 1100 | "\n" in the string wherever a newline is required. If the command is output |
| 1101 | during filter file testing, newlines in the text are shown as "\n". |
| 1102 | |
| 1103 | Note that the keyword for creating a Reply-To: header is reply_to, because Exim |
| 1104 | keywords may contain underscores, but not hyphens. If the from keyword is |
| 1105 | present and the given address does not match the user who owns the forward |
| 1106 | file, Exim normally adds a Sender: header to the message, though it can be |
| 1107 | configured not to do this. |
| 1108 | |
| 1109 | The extra_headers keyword allows you to add custom header lines to the message. |
| 1110 | The text supplied must be one or more syntactically valid RFC 2822 header |
| 1111 | lines. You can use "\n" within quoted text to specify newlines between headers, |
| 1112 | and also to define continued header lines. For example: |
| 1113 | |
| 1114 | extra_headers "h1: first\nh2: second\n continued\nh3: third" |
| 1115 | |
| 1116 | No newline should appear at the end of the final header line. |
| 1117 | |
| 1118 | If no to argument appears, the message is sent to the address in the |
| 1119 | $reply_address variable (see section 3.3 above). An In-Reply-To: header is |
| 1120 | automatically included in the created message, giving a reference to the |
| 1121 | message identification of the incoming message. |
| 1122 | |
| 1123 | If return message is specified, the incoming message that caused the filter |
| 1124 | file to be run is added to the end of the message, subject to a maximum size |
| 1125 | limitation. |
| 1126 | |
| 1127 | If a log file is specified, a line is added to it for each message sent. |
| 1128 | |
| 1129 | If a once file is specified, it is used to hold a database for remembering who |
| 1130 | has received a message, and no more than one message is ever sent to any |
| 1131 | particular address, unless once_repeat is set. This specifies a time interval |
| 1132 | after which another copy of the message is sent. The interval is specified as a |
| 1133 | sequence of numbers, each followed by the initial letter of one of "seconds", |
| 1134 | "minutes", "hours", "days", or "weeks". For example, |
| 1135 | |
| 1136 | once_repeat 5d4h |
| 1137 | |
| 1138 | causes a new message to be sent if at least 5 days and 4 hours have elapsed |
| 1139 | since the last one was sent. There must be no white space in a time interval. |
| 1140 | |
| 1141 | Commonly, the file name specified for once is used as the base name for |
| 1142 | direct-access (DBM) file operations. There are a number of different DBM |
| 1143 | libraries in existence. Some operating systems provide one as a default, but |
| 1144 | even in this case a different one may have been used when building Exim. With |
| 1145 | some DBM libraries, specifying once results in two files being created, with |
| 1146 | the suffixes .dir and .pag being added to the given name. With some others a |
| 1147 | single file with the suffix .db is used, or the name is used unchanged. |
| 1148 | |
| 1149 | Using a DBM file for implementing the once feature means that the file grows as |
| 1150 | large as necessary. This is not usually a problem, but some system |
| 1151 | administrators want to put a limit on it. The facility can be configured not to |
| 1152 | use a DBM file, but instead, to use a regular file with a maximum size. The |
| 1153 | data in such a file is searched sequentially, and if the file fills up, the |
| 1154 | oldest entry is deleted to make way for a new one. This means that some |
| 1155 | correspondents may receive a second copy of the message after an unpredictable |
| 1156 | interval. Consult your local information to see if your system is configured |
| 1157 | this way. |
| 1158 | |
| 1159 | More than one mail or vacation command may be obeyed in a single filter run; |
| 1160 | they are all honoured, even when they are to the same recipient. |
| 1161 | |
| 1162 | |
| 1163 | 3.15 Logging commands |
| 1164 | --------------------- |
| 1165 | |
| 1166 | A log can be kept of actions taken by a filter file. This facility is normally |
| 1167 | available in conventional configurations, but there are some situations where |
| 1168 | it might not be. Also, the system administrator may choose to disable it. Check |
| 1169 | your local information if in doubt. |
| 1170 | |
| 1171 | Logging takes place while the filter file is being interpreted. It does not |
| 1172 | queue up for later like the delivery commands. The reason for this is so that a |
| 1173 | log file need be opened only once for several write operations. There are two |
| 1174 | commands, neither of which constitutes a significant delivery. The first |
| 1175 | defines a file to which logging output is subsequently written: |
| 1176 | |
| 1177 | logfile <file name> |
| 1178 | e.g. logfile $home/filter.log |
| 1179 | |
| 1180 | The file name must be fully qualified. You can use $home, as in this example, |
| 1181 | to refer to your home directory. The file name may optionally be followed by a |
| 1182 | mode for the file, which is used if the file has to be created. For example, |
| 1183 | |
| 1184 | logfile $home/filter.log 0644 |
| 1185 | |
| 1186 | The number is interpreted as octal, even if it does not begin with a zero. The |
| 1187 | default for the mode is 600. It is suggested that the logfile command normally |
| 1188 | appear as the first command in a filter file. Once a log file has been obeyed, |
| 1189 | the logwrite command can be used to write to it: |
| 1190 | |
| 1191 | logwrite "<some text string>" |
| 1192 | e.g. logwrite "$tod_log $message_id processed" |
| 1193 | |
| 1194 | It is possible to have more than one logfile command, to specify writing to |
| 1195 | different log files in different circumstances. Writing takes place at the end |
| 1196 | of the file, and a newline character is added to the end of each string if |
| 1197 | there isn't one already there. Newlines can be put in the middle of the string |
| 1198 | by using the "\n" escape sequence. Lines from simultaneous deliveries may get |
| 1199 | interleaved in the file, as there is no interlocking, so you should plan your |
| 1200 | logging with this in mind. However, data should not get lost. |
| 1201 | |
| 1202 | |
| 1203 | 3.16 The finish command |
| 1204 | ----------------------- |
| 1205 | |
| 1206 | The command finish, which has no arguments, causes Exim to stop interpreting |
| 1207 | the filter file. This is not a significant action unless preceded by "seen". A |
| 1208 | filter file containing only "seen finish" is a black hole. |
| 1209 | |
| 1210 | |
| 1211 | 3.17 The testprint command |
| 1212 | -------------------------- |
| 1213 | |
| 1214 | It is sometimes helpful to be able to print out the values of variables when |
| 1215 | testing filter files. The command |
| 1216 | |
| 1217 | testprint <text> |
| 1218 | e.g. testprint "home=$home reply_address=$reply_address" |
| 1219 | |
| 1220 | does nothing when mail is being delivered. However, when the filtering code is |
| 1221 | being tested by means of the -bf option (see section 1.3 above), the value of |
| 1222 | the string is written to the standard output. |
| 1223 | |
| 1224 | |
| 1225 | 3.18 The fail command |
| 1226 | --------------------- |
| 1227 | |
| 1228 | When Exim's filtering facilities are being used as a system filter, the fail |
| 1229 | command is available, to force delivery failure. Because this command is |
| 1230 | normally usable only by the system administrator, and not enabled for use by |
| 1231 | ordinary users, it is described in more detail in the main Exim specification |
| 1232 | rather than in this document. |
| 1233 | |
| 1234 | |
| 1235 | 3.19 The freeze command |
| 1236 | ----------------------- |
| 1237 | |
| 1238 | When Exim's filtering facilities are being used as a system filter, the freeze |
| 1239 | command is available, to freeze a message on the queue. Because this command is |
| 1240 | normally usable only by the system administrator, and not enabled for use by |
| 1241 | ordinary users, it is described in more detail in the main Exim specification |
| 1242 | rather than in this document. |
| 1243 | |
| 1244 | |
| 1245 | 3.20 The headers command |
| 1246 | ------------------------ |
| 1247 | |
| 1248 | The headers command can be used to change the target character set that is used |
| 1249 | when translating the contents of encoded header lines for insertion by the |
| 1250 | $header_ mechanism (see section 3.5 above). The default can be set in the Exim |
| 1251 | configuration; if not specified, ISO-8859-1 is used. The only currently |
| 1252 | supported format for the headers command in user filters is as in this example: |
| 1253 | |
| 1254 | headers charset "UTF-8" |
| 1255 | |
| 1256 | That is, headers is followed by the word "charset" and then the name of a |
| 1257 | character set. This particular example would be useful if you wanted to compare |
| 1258 | the contents of a header to a UTF-8 string. |
| 1259 | |
| 1260 | In system filter files, the headers command can be used to add or remove header |
| 1261 | lines from the message. These features are described in the main Exim |
| 1262 | specification. |
| 1263 | |
| 1264 | |
| 1265 | 3.21 Obeying commands conditionally |
| 1266 | ----------------------------------- |
| 1267 | |
| 1268 | Most of the power of filtering comes from the ability to test conditions and |
| 1269 | obey different commands depending on the outcome. The if command is used to |
| 1270 | specify conditional execution, and its general form is |
| 1271 | |
| 1272 | if <condition> |
| 1273 | then <commands> |
| 1274 | elif <condition> |
| 1275 | then <commands> |
| 1276 | else <commands> |
| 1277 | endif |
| 1278 | |
| 1279 | There may be any number of elif and then sections (including none) and the else |
| 1280 | section is also optional. Any number of commands, including nested if commands, |
| 1281 | may appear in any of the <commands> sections. |
| 1282 | |
| 1283 | Conditions can be combined by using the words and and or, and round brackets |
| 1284 | (parentheses) can be used to specify how several conditions are to combine. |
| 1285 | Without brackets, and is more binding than or. For example: |
| 1286 | |
| 1287 | if |
| 1288 | $h_subject: contains "Make money" or |
| 1289 | $h_precedence: is "junk" or |
| 1290 | ($h_sender: matches ^\\d{8}@ and not personal) or |
| 1291 | $message_body contains "this is not spam" |
| 1292 | then |
| 1293 | seen finish |
| 1294 | endif |
| 1295 | |
| 1296 | A condition can be preceded by not to negate it, and there are also some |
| 1297 | negative forms of condition that are more English-like. |
| 1298 | |
| 1299 | |
| 1300 | 3.22 String testing conditions |
| 1301 | ------------------------------ |
| 1302 | |
| 1303 | There are a number of conditions that operate on text strings, using the words |
| 1304 | "begins", "ends", "is", "contains" and "matches". If you want to apply the same |
| 1305 | test to more than one header line, you can easily concatenate them into a |
| 1306 | single string for testing, as in this example: |
| 1307 | |
| 1308 | if "$h_to:, $h_cc:" contains me@domain.example then ... |
| 1309 | |
| 1310 | If a string-testing condition name is written in lower case, the testing of |
| 1311 | letters is done without regard to case; if it is written in upper case (for |
| 1312 | example, "CONTAINS"), the case of letters is taken into account. |
| 1313 | |
| 1314 | <text1> begins <text2> |
| 1315 | <text1> does not begin <text2> |
| 1316 | e.g. $header_from: begins "Friend@" |
| 1317 | |
| 1318 | A "begins" test checks for the presence of the second string at the start of |
| 1319 | the first, both strings having been expanded. |
| 1320 | |
| 1321 | <text1> ends <text2> |
| 1322 | <text1> does not end <text2> |
| 1323 | e.g. $header_from: ends "public.com.example" |
| 1324 | |
| 1325 | An "ends" test checks for the presence of the second string at the end of the |
| 1326 | first, both strings having been expanded. |
| 1327 | |
| 1328 | <text1> is <text2> |
| 1329 | <text1> is not <text2> |
| 1330 | e.g. $local_part_suffix is "-foo" |
| 1331 | |
| 1332 | An "is" test does an exact match between the strings, having first expanded |
| 1333 | both strings. |
| 1334 | |
| 1335 | <text1> contains <text2> |
| 1336 | <text1> does not contain <text2> |
| 1337 | e.g. $header_subject: contains "evolution" |
| 1338 | |
| 1339 | A "contains" test does a partial string match, having expanded both strings. |
| 1340 | |
| 1341 | <text1> matches <text2> |
| 1342 | <text1> does not match <text2> |
| 1343 | e.g. $sender_address matches "(bill|john)@" |
| 1344 | |
| 1345 | For a "matches" test, after expansion of both strings, the second one is |
| 1346 | interpreted as a regular expression. Exim uses the PCRE regular expression |
| 1347 | library, which provides regular expressions that are compatible with Perl. |
| 1348 | |
| 1349 | The match succeeds if the regular expression matches any part of the first |
| 1350 | string. If you want a regular expression to match only at the start or end of |
| 1351 | the subject string, you must encode that requirement explicitly, using the "^" |
| 1352 | or "$" metacharacters. The above example, which is not so constrained, matches |
| 1353 | all these addresses: |
| 1354 | |
| 1355 | bill@test.example |
| 1356 | john@some.example |
| 1357 | spoonbill@example.com |
| 1358 | littlejohn@example.com |
| 1359 | |
| 1360 | To match only the first two, you could use this: |
| 1361 | |
| 1362 | if $sender_address matches "^(bill|john)@" then ... |
| 1363 | |
| 1364 | Care must be taken if you need a backslash in a regular expression, because |
| 1365 | backslashes are interpreted as escape characters both by the string expansion |
| 1366 | code and by Exim's normal processing of strings in quotes. For example, if you |
| 1367 | want to test the sender address for a domain ending in .com the regular |
| 1368 | expression is |
| 1369 | |
| 1370 | \.com$ |
| 1371 | |
| 1372 | The backslash and dollar sign in that expression have to be escaped when used |
| 1373 | in a filter command, as otherwise they would be interpreted by the expansion |
| 1374 | code. Thus, what you actually write is |
| 1375 | |
| 1376 | if $sender_address matches \\.com\$ |
| 1377 | |
| 1378 | An alternative way of handling this is to make use of the "\N" expansion flag |
| 1379 | for suppressing expansion: |
| 1380 | |
| 1381 | if $sender_address matches \N\.com$\N |
| 1382 | |
| 1383 | Everything between the two occurrences of "\N" is copied without change by the |
| 1384 | string expander (and in fact you do not need the final one, because it is at |
| 1385 | the end of the string). If the regular expression is given in quotes (mandatory |
| 1386 | only if it contains white space) you have to write either |
| 1387 | |
| 1388 | if $sender_address matches "\\\\.com\\$" |
| 1389 | |
| 1390 | or |
| 1391 | |
| 1392 | if $sender_address matches "\\N\\.com$\\N" |
| 1393 | |
| 1394 | If the regular expression contains bracketed sub-expressions, numeric variable |
| 1395 | substitutions such as $1 can be used in the subsequent actions after a |
| 1396 | successful match. If the match fails, the values of the numeric variables |
| 1397 | remain unchanged. Previous values are not restored after endif. In other words, |
| 1398 | only one set of values is ever available. If the condition contains several |
| 1399 | sub-conditions connected by and or or, it is the strings extracted from the |
| 1400 | last successful match that are available in subsequent actions. Numeric |
| 1401 | variables from any one sub-condition are also available for use in subsequent |
| 1402 | sub-conditions, because string expansion of a condition occurs just before it |
| 1403 | is tested. |
| 1404 | |
| 1405 | |
| 1406 | 3.23 Numeric testing conditions |
| 1407 | ------------------------------- |
| 1408 | |
| 1409 | The following conditions are available for performing numerical tests: |
| 1410 | |
| 1411 | <number1> is above <number2> |
| 1412 | <number1> is not above <number2> |
| 1413 | <number1> is below <number2> |
| 1414 | <number1> is not below <number2> |
| 1415 | e.g. $message_size is not above 10k |
| 1416 | |
| 1417 | The <number> arguments must expand to strings of digits, optionally followed by |
| 1418 | one of the letters K or M (upper case or lower case) which cause multiplication |
| 1419 | by 1024 and 1024x1024 respectively. |
| 1420 | |
| 1421 | |
| 1422 | 3.24 Testing for significant deliveries |
| 1423 | --------------------------------------- |
| 1424 | |
| 1425 | You can use the delivered condition to test whether or not any previously |
| 1426 | obeyed filter commands have set up a significant delivery. For example: |
| 1427 | |
| 1428 | if not delivered then save mail/anomalous endif |
| 1429 | |
| 1430 | "Delivered" is perhaps a poor choice of name for this condition, because the |
| 1431 | message has not actually been delivered; rather, a delivery has been set up for |
| 1432 | later processing. |
| 1433 | |
| 1434 | |
| 1435 | 3.25 Testing for error messages |
| 1436 | ------------------------------- |
| 1437 | |
| 1438 | The condition error_message is true if the incoming message is a bounce (mail |
| 1439 | delivery error) message. Putting the command |
| 1440 | |
| 1441 | if error_message then finish endif |
| 1442 | |
| 1443 | at the head of your filter file is a useful insurance against things going |
| 1444 | wrong in such a way that you cannot receive delivery error reports. Note: |
| 1445 | error_message is a condition, not an expansion variable, and therefore is not |
| 1446 | preceded by "$". |
| 1447 | |
| 1448 | |
| 1449 | 3.26 Testing a list of addresses |
| 1450 | -------------------------------- |
| 1451 | |
| 1452 | There is a facility for looping through a list of addresses and applying a |
| 1453 | condition to each of them. It takes the form |
| 1454 | |
| 1455 | foranyaddress <string> (<condition>) |
| 1456 | |
| 1457 | where <string> is interpreted as a list of RFC 2822 addresses, as in a typical |
| 1458 | header line, and <condition> is any valid filter condition or combination of |
| 1459 | conditions. The "group" syntax that is defined for certain header lines that |
| 1460 | contain addresses is supported. |
| 1461 | |
| 1462 | The parentheses surrounding the condition are mandatory, to delimit it from |
| 1463 | possible further sub-conditions of the enclosing if command. Within the |
| 1464 | condition, the expansion variable $thisaddress is set to the non-comment |
| 1465 | portion of each of the addresses in the string in turn. For example, if the |
| 1466 | string is |
| 1467 | |
| 1468 | B.Simpson <bart@sfld.example>, lisa@sfld.example (his sister) |
| 1469 | |
| 1470 | then $thisaddress would take on the values "bart@sfld.example" and |
| 1471 | "lisa@sfld.example" in turn. |
| 1472 | |
| 1473 | If there are no valid addresses in the list, the whole condition is false. If |
| 1474 | the internal condition is true for any one address, the overall condition is |
| 1475 | true and the loop ends. If the internal condition is false for all addresses in |
| 1476 | the list, the overall condition is false. This example tests for the presence |
| 1477 | of an eight-digit local part in any address in a To: header: |
| 1478 | |
| 1479 | if foranyaddress $h_to: ( $thisaddress matches ^\\d{8}@ ) then ... |
| 1480 | |
| 1481 | When the overall condition is true, the value of $thisaddress in the commands |
| 1482 | that follow then is the last value it took on inside the loop. At the end of |
| 1483 | the if command, the value of $thisaddress is reset to what it was before. It is |
| 1484 | best to avoid the use of multiple occurrences of foranyaddress, nested or |
| 1485 | otherwise, in a single if command, if the value of $thisaddress is to be used |
| 1486 | afterwards, because it isn't always clear what the value will be. Nested if |
| 1487 | commands should be used instead. |
| 1488 | |
| 1489 | Header lines can be joined together if a check is to be applied to more than |
| 1490 | one of them. For example: |
| 1491 | |
| 1492 | if foranyaddress $h_to:,$h_cc: .... |
| 1493 | |
| 1494 | This scans through the addresses in both the To: and the Cc: headers. |
| 1495 | |
| 1496 | |
| 1497 | 3.27 Testing for personal mail |
| 1498 | ------------------------------ |
| 1499 | |
| 1500 | A common requirement is to distinguish between incoming personal mail and mail |
| 1501 | from a mailing list, or from a robot or other automatic process (for example, a |
| 1502 | bounce message). In particular, this test is normally required for "vacation |
| 1503 | messages". |
| 1504 | |
| 1505 | The personal condition checks that the message is not a bounce message and that |
| 1506 | the current user's email address appears in the To: header. It also checks that |
| 1507 | the sender is not the current user or one of a number of common daemons, and |
| 1508 | that there are no header lines starting List- in the message. Finally, it |
| 1509 | checks the content of the Precedence: header line, if there is one. |
| 1510 | |
| 1511 | You should always use the personal condition when generating automatic |
| 1512 | responses. This example shows the use of personal in a filter file that is |
| 1513 | sending out vacation messages: |
| 1514 | |
| 1515 | if personal then |
| 1516 | mail to $reply_address |
| 1517 | subject "I am on holiday" |
| 1518 | file $home/vacation/message |
| 1519 | once $home/vacation/once |
| 1520 | once_repeat 10d |
| 1521 | endif |
| 1522 | |
| 1523 | It is tempting, when writing commands like the above, to quote the original |
| 1524 | subject in the reply. For example: |
| 1525 | |
| 1526 | subject "Re: $h_subject:" |
| 1527 | |
| 1528 | There is a danger in doing this, however. It may allow a third party to |
| 1529 | subscribe you to an opt-in mailing list, provided that the list accepts bounce |
| 1530 | messages as subscription confirmations. (Messages sent from filters are always |
| 1531 | sent as bounce messages.) Well-managed lists require a non-bounce message to |
| 1532 | confirm a subscription, so the danger is relatively small. |
| 1533 | |
| 1534 | If prefixes or suffixes are in use for local parts - something which depends on |
| 1535 | the configuration of Exim (see section 3.31 below) - the tests for the current |
| 1536 | user are done with the full address (including the prefix and suffix, if any) |
| 1537 | as well as with the prefix and suffix removed. If the system is configured to |
| 1538 | rewrite local parts of mail addresses, for example, to rewrite "dag46" as |
| 1539 | "Dirk.Gently", the rewritten form of the address is also used in the tests. |
| 1540 | |
| 1541 | |
| 1542 | 3.28 Alias addresses for the personal condition |
| 1543 | ----------------------------------------------- |
| 1544 | |
| 1545 | It is quite common for people who have mail accounts on a number of different |
| 1546 | systems to forward all their mail to one system, and in this case a check for |
| 1547 | personal mail should test all their various mail addresses. To allow for this, |
| 1548 | the personal condition keyword can be followed by |
| 1549 | |
| 1550 | alias <address> |
| 1551 | |
| 1552 | any number of times, for example: |
| 1553 | |
| 1554 | if personal alias smith@else.where.example |
| 1555 | alias jones@other.place.example |
| 1556 | then ... |
| 1557 | |
| 1558 | The alias addresses are treated as alternatives to the current user's email |
| 1559 | address when testing the contents of header lines. |
| 1560 | |
| 1561 | |
| 1562 | 3.29 Details of the personal condition |
| 1563 | -------------------------------------- |
| 1564 | |
| 1565 | The basic personal test is roughly equivalent to the following: |
| 1566 | |
| 1567 | not error_message and |
| 1568 | $message_headers does not contain "\nList-Id:" and |
| 1569 | $message_headers does not contain "\nList-Help:" and |
| 1570 | $message_headers does not contain "\nList-Subscribe:" and |
| 1571 | $message_headers does not contain "\nList-Unsubscribe:" and |
| 1572 | $message_headers does not contain "\nList-Post:" and |
| 1573 | $message_headers does not contain "\nList-Owner:" and |
| 1574 | $message_headers does not contain "\nList-Archive:" and |
| 1575 | ( |
| 1576 | "${if def:h_auto-submitted:{present}{absent}}" is "absent" or |
| 1577 | $header_auto-submitted: is "no" |
| 1578 | ) and |
| 1579 | $header_precedence: does not contain "bulk" and |
| 1580 | $header_precedence: does not contain "list" and |
| 1581 | $header_precedence: does not contain "junk" and |
| 1582 | foranyaddress $header_to: |
| 1583 | ( $thisaddress contains "$local_part$domain" ) and |
| 1584 | not foranyaddress $header_from: |
| 1585 | ( |
| 1586 | $thisaddress contains "$local_part@$domain" or |
| 1587 | $thisaddress contains "server@" or |
| 1588 | $thisaddress contains "daemon@" or |
| 1589 | $thisaddress contains "root@" or |
| 1590 | $thisaddress contains "listserv@" or |
| 1591 | $thisaddress contains "majordomo@" or |
| 1592 | $thisaddress contains "-request@" or |
| 1593 | $thisaddress matches "^owner-[^@]+@" |
| 1594 | ) |
| 1595 | |
| 1596 | The variable $local_part contains the local part of the mail address of the |
| 1597 | user whose filter file is being run - it is normally your login id. The $domain |
| 1598 | variable contains the mail domain. As explained above, if aliases or rewriting |
| 1599 | are defined, or if prefixes or suffixes are in use, the tests for the current |
| 1600 | user are also done with alternative addresses. |
| 1601 | |
| 1602 | |
| 1603 | 3.30 Testing delivery status |
| 1604 | ---------------------------- |
| 1605 | |
| 1606 | There are two conditions that are intended mainly for use in system filter |
| 1607 | files, but which are available in users' filter files as well. The condition |
| 1608 | first_delivery is true if this is the first process that is attempting to |
| 1609 | deliver the message, and false otherwise. This indicator is not reset until the |
| 1610 | first delivery process successfully terminates; if there is a crash or a power |
| 1611 | failure (for example), the next delivery attempt is also a "first delivery". |
| 1612 | |
| 1613 | In a user filter file first_delivery will be false if there was previously an |
| 1614 | error in the filter, or if a delivery for the user failed owing to, for |
| 1615 | example, a quota error, or if forwarding to a remote address was deferred for |
| 1616 | some reason. |
| 1617 | |
| 1618 | The condition manually_thawed is true if the message was "frozen" for some |
| 1619 | reason, and was subsequently released by the system administrator. It is |
| 1620 | unlikely to be of use in users' filter files. |
| 1621 | |
| 1622 | |
| 1623 | 3.31 Multiple personal mailboxes |
| 1624 | -------------------------------- |
| 1625 | |
| 1626 | The system administrator can configure Exim so that users can set up variants |
| 1627 | on their email addresses and handle them separately. Consult your system |
| 1628 | administrator or local documentation to see if this facility is enabled on your |
| 1629 | system, and if so, what the details are. |
| 1630 | |
| 1631 | The facility involves the use of a prefix or a suffix on an email address. For |
| 1632 | example, all mail addressed to lg303-<something> would be the property of user |
| 1633 | lg303, who could determine how it was to be handled, depending on the value of |
| 1634 | <something>. |
| 1635 | |
| 1636 | There are two possible ways in which this can be set up. The first possibility |
| 1637 | is the use of multiple .forward files. In this case, mail to lg303-foo, for |
| 1638 | example, is handled by looking for a file called .forward-foo in lg303's home |
| 1639 | directory. If such a file does not exist, delivery fails and the message is |
| 1640 | returned to its sender. |
| 1641 | |
| 1642 | The alternative approach is to pass all messages through a single .forward |
| 1643 | file, which must be a filter file so that it can distinguish between the |
| 1644 | different cases by referencing the variables $local_part_prefix or |
| 1645 | $local_part_suffix, as in the final example in section 3.33 below. |
| 1646 | |
| 1647 | It is possible to configure Exim to support both schemes at once. In this case, |
| 1648 | a specific .forward-foo file is first sought; if it is not found, the basic |
| 1649 | .forward file is used. |
| 1650 | |
| 1651 | The personal test (see section 3.27) includes prefixes and suffixes in its |
| 1652 | checking. |
| 1653 | |
| 1654 | |
| 1655 | 3.32 Ignoring delivery errors |
| 1656 | ----------------------------- |
| 1657 | |
| 1658 | As was explained above, filtering just sets up addresses for delivery - no |
| 1659 | deliveries are actually done while a filter file is active. If any of the |
| 1660 | generated addresses subsequently suffers a delivery failure, an error message |
| 1661 | is generated in the normal way. However, if a filter command that sets up a |
| 1662 | delivery is preceded by the word "noerror", errors for that delivery, and any |
| 1663 | deliveries consequent on it (that is, from alias, forwarding, or filter files |
| 1664 | it invokes) are ignored. |
| 1665 | |
| 1666 | |
| 1667 | 3.33 Examples of Exim filter commands |
| 1668 | ------------------------------------- |
| 1669 | |
| 1670 | Simple forwarding: |
| 1671 | |
| 1672 | # Exim filter |
| 1673 | deliver baggins@rivendell.middle-earth.example |
| 1674 | |
| 1675 | Vacation handling using traditional means, assuming that the .vacation.msg and |
| 1676 | other files have been set up in your home directory: |
| 1677 | |
| 1678 | # Exim filter |
| 1679 | unseen pipe "/usr/ucb/vacation \"$local_part\"" |
| 1680 | |
| 1681 | Vacation handling inside Exim, having first created a file called .vacation.msg |
| 1682 | in your home directory: |
| 1683 | |
| 1684 | # Exim filter |
| 1685 | if personal then vacation endif |
| 1686 | |
| 1687 | File some messages by subject: |
| 1688 | |
| 1689 | # Exim filter |
| 1690 | if $header_subject: contains "empire" or |
| 1691 | $header_subject: contains "foundation" |
| 1692 | then |
| 1693 | save $home/mail/f+e |
| 1694 | endif |
| 1695 | |
| 1696 | Save all non-urgent messages by weekday: |
| 1697 | |
| 1698 | # Exim filter |
| 1699 | if $header_subject: does not contain "urgent" and |
| 1700 | $tod_full matches "^(...)," |
| 1701 | then |
| 1702 | save $home/mail/$1 |
| 1703 | endif |
| 1704 | |
| 1705 | Throw away all mail from one site, except from postmaster: |
| 1706 | |
| 1707 | # Exim filter |
| 1708 | if $reply_address contains "@spam.site.example" and |
| 1709 | $reply_address does not contain "postmaster@" |
| 1710 | then |
| 1711 | seen finish |
| 1712 | endif |
| 1713 | |
| 1714 | Handle multiple personal mailboxes: |
| 1715 | |
| 1716 | # Exim filter |
| 1717 | if $local_part_suffix is "-foo" |
| 1718 | then |
| 1719 | save $home/mail/foo |
| 1720 | elif $local_part_suffix is "-bar" |
| 1721 | then |
| 1722 | save $home/mail/bar |
| 1723 | endif |
| 1724 | |