"SfR Fresh" - the SfR Freeware/Shareware Archive 
Member "inn-2.4.5/doc/hook-perl" of archive inn-2.4.5.tar.gz:
As a special service "SfR Fresh" has tried to format the requested source page into HTML format using source code syntax highlighting with prefixed line numbers.
Alternatively you can here view or download the uninterpreted source code file.
That can be also achieved for any archive member file by clicking within an archive contents listing on the first character of the file(path) respectively on the according byte size field.
1 INN Perl Filtering and Authentication Support
2
3 This is $Revision: 7880 $ dated $Date: 2008-06-07 14:46:49 +0200 (Sat,
4 07 Jun 2008) $.
5
6 This file documents INN's built-in support for Perl filtering and reader
7 authentication. The code is based very heavily on work by Christophe
8 Wolfhugel <wolf@pasteur.fr>, and his work was in turn inspired by the
9 existing TCL support. Please send any bug reports to inn-bugs@isc.org,
10 not to Christophe, as the code has been modified heavily since he
11 originally wrote it.
12
13 The Perl filtering support is described in more detail below.
14 Basically, it allows you to supply a Perl function that is invoked on
15 every article received by innd from a peer (the innd filter) or by nnrpd
16 from a reader (the nnrpd filter). This function can decide whether to
17 accept or reject the article, and can optionally do other, more
18 complicated processing (such as add history entries, cancel articles,
19 spool local posts into a holding area, or even modify the headers of
20 locally submitted posts). The Perl authentication hooks allow you to
21 replace or supplement the readers.conf mechanism used by nnrpd.
22
23 For Perl filtering support, you need to have Perl version 5.004 or
24 newer. Earlier versions of Perl will fail with a link error at
25 compilation time. http://language.perl.com/info/software.html should
26 have the latest Perl version.
27
28 To enable Perl support, you have to specify --with-perl when you run
29 configure. See INSTALL for more information.
30
31 The innd Perl Filter
32
33 When innd starts, it first loads the file _PATH_PERL_STARTUP_INND
34 (defined in include/paths.h, by default startup_innd.pl) and then loads
35 the file _PATH_PERL_FILTER_INND (also defined in include/paths.h, by
36 default filter_innd.pl). Both of these files must be located in the
37 directory specified by pathfilter in inn.conf
38 (/usr/local/news/bin/filter by default). The default directory for
39 filter code can be specified at configure time by giving the flag
40 --with-filter-dir to configure.
41
42 INN doesn't care what Perl functions you define in which files. The
43 only thing that's different about the two files is when they're loaded.
44 startup_innd.pl is loaded only once, when innd first starts, and is
45 never reloaded as long as innd is running. Any modifications to that
46 file won't be noticed by innd; only stopping and restarting innd can
47 cause it to be reloaded.
48
49 filter_innd.pl, on the other hand, can be reloaded on command (with
50 "ctlinnd reload filter.perl 'reason'"). Whenever filter_innd.pl is
51 loaded, including the first time at innd startup, the Perl function
52 filter_before_reload() is called before it's reloaded and the function
53 filter_after_reload() is called after it's reloaded (if the functions
54 exist). Additionally, any code in either startup_innd.pl or
55 filter_innd.pl at the top level (in other words, not inside a sub { })
56 is automatically executed by Perl when the files are loaded.
57
58 This allows one to do things like write out filter statistics whenever
59 the filter is reloaded, load a cache into memory, flush cached data to
60 disk, or other similar operations that should only happen at particular
61 times or with manual intervention. Remember, any code not inside
62 functions in startup_innd.pl is executed when that file is loaded, and
63 it's loaded only once when innd first starts. That makes it the ideal
64 place to put initialization code that should only run once, or code to
65 load data that was preserved on disk across a stop and restart of innd
66 (perhaps using filter_mode() -- see below).
67
68 As mentioned above, "ctlinnd reload filter.perl 'reason'" (or "ctlinnd
69 reload all 'reason'") will cause filter_innd.pl to be reloaded. If the
70 function filter_art() is defined after the file has been reloaded,
71 filtering is turned on. Otherwise, filtering is turned off. (Note that
72 due to the way Perl stores functions, once you've defined filter_art(),
73 you can't undefine it just by deleting it from the file and reloading
74 the filter. You'll need to replace it with an empty sub.)
75
76 The Perl function filter_art() is the heart of a Perl filter. Whenever
77 an article is received from a peer, via either IHAVE or TAKETHIS,
78 filter_art() is called if Perl filtering is turned on. It receives no
79 arguments, and should return a single scalar value. That value should
80 be the empty string to indicate that INN should accept the article, or
81 some rejection message to indicate that the article should be rejected.
82
83 filter_art() has access to a global hash named %hdr, which contains all
84 of the standard headers present in the article and their values. The
85 standard headers are:
86
87 Also-Control, Approved, Bytes, Cancel-Key, Cancel-Lock,
88 Content-Base, Content-Disposition, Content-Transfer-Encoding,
89 Content-Type, Control, Date, Date-Received, Distribution, Expires,
90 Face, Followup-To, From, In-Reply-To, Injection-Date, Injection-Info,
91 Keywords, Lines, List-ID, Message-ID, MIME-Version, Newsgroups,
92 NNTP-Posting-Date, NNTP-Posting-Host, Organization, Originator,
93 Path, Posted, Posting-Version, Received, References, Relay-Version,
94 Reply-To, Sender, Subject, Supersedes, User-Agent,
95 X-Auth, X-Canceled-By, X-Cancelled-By, X-Complaints-To, X-Face,
96 X-HTTP-UserAgent, X-HTTP-Via, X-Mailer, X-Modbot, X-Modtrace,
97 X-Newsposter, X-Newsreader, X-No-Archive, X-Original-Message-ID,
98 X-Original-Trace, X-Originating-IP, X-PGP-Key, X-PGP-Sig,
99 X-Poster-Trace, X-Postfilter, X-Proxy-User, X-Submissions-To,
100 X-Trace, X-Usenet-Provider, Xref.
101
102 Note that all the above headers are as they arrived, not modified by
103 your INN (especially, the Xref: header, if present, is the one of the
104 remote site which sent you the article, and not yours).
105
106 For example, the Newsgroups: header of the article is accessible inside
107 the Perl filter as $hdr{'Newsgroups'}. In addition, $hdr{'__BODY__'}
108 will contain the full body of the article and $hdr{'__LINES__'} will
109 contain the number of lines in the body of the article.
110
111 The contents of the %hdr hash for a typical article may therefore look
112 something like this:
113
114 %hdr = (Subject => 'MAKE MONEY FAST!!',
115 From => 'Joe Spamer <him@example.com>',
116 Date => '10 Sep 1996 15:32:28 UTC',
117 Newsgroups => 'alt.test',
118 Path => 'news.example.com!not-for-mail',
119 Organization => 'Spammers Anonymous',
120 Lines => '5',
121 Distribution => 'usa',
122 'Message-ID' => '<6.20232.842369548@example.com>',
123 __BODY__ => 'Send five dollars to the ISC, c/o ...',
124 __LINES__ => 5
125 );
126
127 Note that the value of $hdr{Lines} is the contents of the Lines: header
128 of the article and may bear no resemblence to the actual length of the
129 article. $hdr{__LINES__} is the line count calculated by INN, and is
130 guaranteed to be accurate.
131
132 The %hdr hash should not be modified inside filter_art(). Instead, if
133 any of the contents need to be modified temporarily during filtering
134 (smashing case, for example), copy them into a seperate variable first
135 and perform the modifications on the copy. Currently, $hdr{__BODY__} is
136 the only data that will cause your filter to die if you modify it, but
137 in the future other keys may also contain live data. Modifying live INN
138 data in Perl will hopefully only cause a fatal exception in your Perl
139 code that disables Perl filtering until you fix it, but it's possible
140 for it to cause article munging or even core dumps in INN. So always,
141 always make a copy first.
142
143 As mentioned above, if filter_art() returns the empty string (''), the
144 article is accepted. Note that this must be the empty string, not 0 or
145 undef. Otherwise, the article is rejected, and whatever scalar
146 filter_art() returns (typically a string) will be taken as the reason
147 why the article was rejected. This reason will be returned to the
148 remote peer as well as logged to the news logs. (innreport, in its
149 nightly report, will summarize the number of articles rejected by the
150 Perl filter and include a count of how many articles were rejected with
151 each reason string.)
152
153 One other type of filtering is also supported. If Perl filtering is
154 turned on and the Perl function filter_messageid() is defined, that
155 function will be called for each message ID received from a peer (via
156 either CHECK or IHAVE). The function receives a single argument, the
157 message ID, and like filter_art() should return an empty string to
158 accept the article or an error string to refuse the article. This
159 function is called before any history lookups and for every article
160 offered to innd with CHECK or IHAVE (before the actual article is sent).
161 Accordingly, the message ID is the only information it has about the
162 article (the %hdr hash will be empty). This code would sit in a
163 performance-critical hot path in a typical server, and therefore should
164 be as fast as possible, but it can do things like refuse articles from
165 certain hosts or cancels for already rejected articles (if they follow
166 the $alz convention) without having to take the network bandwidth hit of
167 accepting the entire article first.
168
169 Note that you cannot rely on filter_messageid() being called for every
170 incoming article; articles sent via TAKETHIS without an earlier CHECK
171 will never pass through filter_messageid() and will only go through
172 filter_art().
173
174 Finally, whenever ctlinnd throttle, ctlinnd pause, or ctlinnd go is run,
175 the Perl function filter_mode() is called if it exists. It receives no
176 arguments and returns no value, but it has access to a global hash %mode
177 that contains three values:
178
179 Mode The current server mode (throttled, paused, or running)
180 NewMode The new mode the server is going to
181 reason The reason that was given to ctlinnd
182
183 One possible use for this function is to save filter state across a
184 restart of innd. There isn't any Perl function which is called when INN
185 shuts down, but using filter_mode() the Perl filter can dump it's state
186 to disk whenever INN is throttled. Then, if the news administrator
187 follows the strongly recommended shutdown procedure of throttling the
188 server before shutting it down, the filter state will be safely saved to
189 disk and can be reloaded when innd restarts (possibly by
190 startup_innd.pl).
191
192 The state of the Perl interpretor in which all of these Perl functions
193 run is preserved over the lifetime of innd. In other words, it's
194 permissible for the Perl code to create its own global Perl variables,
195 data structures, saved state, and the like, and all of that will be
196 available to filter_art() and filter_messageid() each time they're
197 called. The only variable INN fiddles with (or pays any attention to at
198 all) is %hdr, which is cleared after each call to filter_art().
199
200 Perl filtering can be turned off with "ctlinnd perl n" and back on again
201 with "ctlinnd perl y". Perl filtering is turned off automatically if
202 loading of the filter fails or if the filter code returns any sort of a
203 fatal error (either due to Perl itself or due to a "die" in the Perl
204 code).
205
206 Supported innd Callbacks
207
208 innd makes seven functions available to any of its embedded Perl code.
209 Those are:
210
211 INN::addhist(*messageid*, *arrival*, *articledate*, *expire*, *paths*)
212 Adds *messageid* to the history database. All of the arguments
213 except the first one are optional; the times default to the current
214 time and the paths field defaults to the empty string. (For those
215 unfamiliar with the fields of a history(5) database entry, the
216 *arrival* is normally the time at which the server accepts the
217 article, the *articledate* is from the Date header of the article,
218 the *expire* is from the Expires header of the article, and the
219 *paths* field is the storage API token. All three times as measured
220 as a time_t since the epoch.) Returns true on success, false
221 otherwise.
222
223 INN::article(*messageid*)
224 Returns the full article (as a simple string) identified by
225 *messageid*, or undef if it isn't found. Each line will end with a
226 simple \n, but leading periods may still be doubled if the article
227 is stored in wire format.
228
229 INN::cancel(*messageid*)
230 Cancels *messageid*. (This is equivalent to "ctlinnd cancel"; it
231 cancels the message on the local server, but doesn't post a cancel
232 message or do anything else that affects anything other than the
233 local server.) Returns true on success, false otherwise.
234
235 INN::filesfor(*messageid*)
236 Returns the *paths* field of the history entry for the given
237 *messageid*. This will be the storage API token for the message.
238 If *messageid* isn't found in the history database, returns undef.
239
240 INN::havehist(*messageid*)
241 Looks up *messageid* in the history database and returns true if
242 it's found, false otherwise.
243
244 INN::head(*messageid*)
245 Returns the header (as a simple string) of the article identified by
246 *messageid*, or undef if it isn't found. Each line will end with a
247 simple \n (in other words, regardless of the format of article
248 storage, the returned string won't be in wire format).
249
250 INN::newsgroup(*newsgroup*)
251 Returns the status of *newsgroup* (the last field of the active file
252 entry for that newsgroup). See active(5) for a description of the
253 possible values and their meanings (the most common are "y" for an
254 unmoderated group and "m" for a moderated group). If *newsgroup*
255 isn't in the active file, returns undef.
256
257 These functions can only be used from inside the innd Perl filter;
258 they're not available in the nnrpd filter.
259
260 Common Callbacks
261
262 The following additional function is available from inside filters
263 embedded in innd, and is also available from filters embedded in nnrpd
264 (see below):
265
266 INN::syslog(level, message)
267 Logs a message via syslog(2). This is quite a bit more reliable and
268 portable than trying to use Sys::Syslog from inside the Perl filter.
269 Only the first character of the level argument matters; the valid
270 letters are the first letters of ALERT, CRIT, ERR, WARNING, NOTICE,
271 INFO, and DEBUG (case-insensitive) and specify the priority at which
272 the message is logged. If a level that doesn't match any of those
273 levels is given, the default priority level is LOG_NOTICE. The
274 second argument is the message to log; it will be prefixed by
275 "filter: " and logged to syslog with facility LOG_NEWS.
276
277 The nnrpd Posting Filter
278
279 Whenever Perl support is needed in nnrpd, it first loads the file
280 _PATH_PERL_FILTER_NNRPD (defined in include/paths.h, by default
281 filter_nnrpd.pl). This file must be located in the directory specified
282 by pathfilter in inn.conf (/usr/local/news/bin/filter by default). The
283 default directory for filter code can be specified at configure time by
284 giving the flag --with-filter-dir to configure.
285
286 If filter_nnrpd.pl loads successfully and defines the Perl function
287 filter_post(), Perl filtering is turned on. Otherwise, it's turned off.
288 If filter_post() ever returns a fatal error (either from Perl or from a
289 "die" in the Perl code), Perl filtering is turned off for the life of
290 that nnrpd process and any further posts made during that session won't
291 go through the filter.
292
293 While Perl filtering is on, every article received by nnrpd via the POST
294 command is passed to the filter_post() Perl function before it is passed
295 to INN (or mailed to the moderator of a moderated newsgroup). If
296 filter_post() returns an empty string (''), the article is accepted and
297 normal processing of it continues. Otherwise, the article is rejected
298 and the string returned by filter_post() is returned to the client as
299 the error message (with some exceptions; see below).
300
301 filter_post() has access to a global hash %hdr, which contains all of
302 the headers of the article. (Unlike the innd Perl filter, %hdr for the
303 nnrpd Perl filter contains *all* of the headers, not just the standard
304 ones. If any of the headers are duplicated, though, %hdr will contain
305 only the value of the last occurance of the header. nnrpd will reject
306 the article before the filter runs if any of the standard headers are
307 duplicated.) It also has access to the full body of the article in the
308 variable $body, and if the poster authenticated via AUTHINFO (or if
309 either Perl authentication or a readers.conf authentication method is
310 used and produces user information), it has access to the authenticated
311 username of the poster in the variable $user.
312
313 Unlike the innd Perl filter, the nnrpd Perl filter can modify the %hdr
314 hash. In fact, if the Perl variable $modify_headers is set to true
315 after filter_post() returns, the contents of the %hdr hash will be
316 written back to the article replacing the original headers.
317 filter_post() can therefore make any modifications it wishes to the
318 headers and those modifications will be reflected in the article as it's
319 finally posted. The article body cannot be modified in this way; any
320 changes to $body will just be ignored.
321
322 Be careful when using the ability to modify headers. filter_post() runs
323 after all the normal consistency checks on the headers and after server
324 supplied headers (like Message-ID: and Date:) are filled in. Deleting
325 required headers or modifying headers that need to follow a strict
326 format can result in nnrpd trying to post nonsense articles (which will
327 probably then be rejected by innd). If $modify_headers is set,
328 *everything* in the %hdr hash is taken to be article headers and added
329 to the article.
330
331 If filter_post() returns something other than the empty string, this
332 message is normally returned to the client as an error. There are two
333 exceptions: If the string returned begins with "DROP", the post will be
334 silently discarded and success returned to the client. If the string
335 begins with "SPOOL", success is returned to the client, but the post is
336 saved in a directory named "spam" under the directory specified by
337 pathincoming in inn.conf (in a directory named "spam/mod" if the post is
338 to a moderated group). This is intended to allow manual inspection of
339 the suspect messages; if they should be posted, they can be manually
340 moved out of the subdirectory to the directory specified by pathincoming
341 in inn.conf, where they can be posted by running "rnews -U". If you use
342 this functionality, make sure those directories exist.
343
344 Changes to Perl Authentication Support for nnrpd
345
346 The old authentication functionality has been combined with the new
347 readers.conf mechanism by Erik Klavon <erik@eriq.org>; bug reports
348 should however go to inn-bugs@isc.org, not Erik.
349
350 The remainder of this section is an introduction to the new mechanism
351 (which uses the perl_auth: and perl_access: readers.conf parameters)
352 with porting/migration suggestions for people familiar with the old
353 mechanism (identifiable by the nnrpperlauth: parameter in inn.conf).
354
355 Other people should skip this section.
356
357 The perl_auth parameter allows the use of Perl to authenticate a user.
358 Scripts (like those from the old mechanism) are listed in readers.conf
359 using perl_auth in the same manner other authenticators are using auth:
360
361 perl_auth: "/path/to/script/auth1.pl"
362
363 The file given as argument to perl_auth should contain the same
364 procedures as before. The global hash %attributes remains the same,
365 except for the removal of the "type" entry which is no longer needed in
366 this modification and the addition of several new entries (port,
367 intipaddr, intport) described below. The return array now only contains
368 either two or three elements, the first of which is the NNTP return
369 code. The second is an error string which is passed to the client if the
370 error code indicates that the authentication attempt has failed. This
371 allows a specific error message to be generated by the perl script in
372 place of "Authentication failed". An optional third return element if
373 present will be used to match the connection with the users: parameter
374 in access groups and will also be the username logged. If this element
375 is absent, the username supplied by the client during authentication
376 will be used as was the previous behavior.
377
378 The perl_access parameter (described below) is also new; it allows the
379 dynamic generation of an access group for an incoming connection using a
380 Perl script. If a connection matches an auth group which has a
381 perl_access parameter, all access groups in readers.conf are ignored;
382 instead the procedure described below is used to generate an access
383 group. This concept is due to Jeffrey M. Vinocur.
384
385 The new functionality should provide all of the existing capabilities of
386 the Perl hook, in combination with the flexibility of readers.conf and
387 the use of other authentication and resolving programs. To use Perl
388 authentication code that predates the readers.conf mechanism, you would
389 need to modify the code slightly (see below for the new specification)
390 and supply a simple readers.conf file. If you don't want to modify your
391 code, the samples directory has nnrpd_auth_wrapper.pl and
392 nnrpd_access_wrapper.pl which should allow you to use your old code
393 without needing to change it.
394
395 However, before trying to use your old Perl code, you may want to
396 consider replacing it entirely with non-Perl authentication. (With
397 readers.conf and the regular authenticator and resolver programs, much
398 of what once required Perl can be done directly.) Even if the
399 functionality is not available directly, you may wish to write a new
400 authenticator or resolver (which can be done in whatever language you
401 prefer to work in).
402
403 Perl Authentication Support for nnrpd
404
405 Support for authentication via Perl is provided in nnrpd by the
406 inclusion of a perl_auth: parameter in a readers.conf auth group.
407 perl_auth: works exactly like the auth: parameter in readers.conf,
408 except that it calls the script given as argument using the Perl hook
409 rather then treating it as an external program.
410
411 If the processing of readers.conf requires that a perl_auth: statement
412 be used for authentication, Perl is loaded (if it has yet to be) and the
413 file given as argument to the perl_auth: parameter is loaded as well. If
414 a Perl function auth_init() is defined by that file, it is called
415 immediately after the file is loaded. It takes no arguments and returns
416 nothing.
417
418 Provided the file loads without errors, auth_init() (if present) runs
419 without fatal errors, and a Perl function authenticate() is defined,
420 authenticate() will then be called. authenticate() takes no arguments,
421 but it has access to a global hash %attributes which contains
422 information about the connection as follows: $attributes{hostname} will
423 contain the hostname (or the IP address if it doesn't resolve) of the
424 client machine, $attributes{ipaddress} will contain its IP address (as a
425 string), $attributes{port} will contain the client port (as an integer),
426 $attributes{interface} contains the hostname of the interface the client
427 connected on, $attributes{intipaddr} contains the IP address (as a
428 string) of the interface the client connected on, $attributes{intport}
429 contains the port (as an integer) on the interface the client connected
430 on, $attributes{username} will contain the provided username and
431 $attributes{password} the password.
432
433 authenticate() should return a two or three element array. The first
434 element is the NNTP response code to return to the client, the second
435 element is an error string which is passed to the client if the response
436 code indicates that the authentication attempt has failed. An optional
437 third return element if present will be used to match the connection
438 with the users: parameter in access groups and will also be the username
439 logged. If this element is absent, the username supplied by the client
440 during authentication will be used for matching and logging.
441
442 The NNTP response code should probably be either 281 (authentication
443 successful) or 502 (authentication unsuccessful). If the code returned
444 is anything other than 281, nnrpd will print an authentication error
445 message and drop the connection and exit.
446
447 If authenticate() dies (either due to a Perl error or due to calling
448 die), or if it returns anything other than the two or three element
449 array described above, an internal error will be reported to the client,
450 the exact error will be logged to syslog, and nnrpd will drop the
451 connection and exit.
452
453 Dynamic Generation of Access Groups
454
455 A Perl script may be used to dynamically generate an access group which
456 is then used to determine the access rights of the client. This occurs
457 whenever the perl_access: is specified in an auth group which has
458 successfully matched the client. Only one perl_access: statement is
459 allowed in an auth group. This parameter should not be mixed with a
460 python_access: statement in the same auth group.
461
462 When a perl_access: parameter is encountered, Perl is loaded (if it has
463 yet to be) and the file given as argument is loaded as well. Provided
464 the file loads without errors, and a Perl function access() is defined,
465 access() will then be called. access() takes no arguments, but it has
466 access to a global hash %attributes which contains information about the
467 connection as follows: $attributes{hostname} will contain the hostname
468 (or the IP address if it doesn't resolve) of the client machine,
469 $attributes{ipaddress} will contain its IP address (as a string),
470 $attributes{port} will contain the client port (as an integer),
471 $attributes{interface} contains the hostname of the interface the client
472 connected on, $attributes{intipaddr} contains the IP address (as a
473 string) of the interface the client connected on, $attributes{intport}
474 contains the port (as an integer) on the interface the client connected
475 on, $attributes{username} will contain the provided username and domain
476 (in username@domain form).
477
478 access() returns a hash, containing the desired access parameters and
479 values. Here is an untested example showing how to dynamically generate
480 a list of newsgroups based on the client's username and domain.
481
482 my %hosts = ( "example.com" => "example.*", "isc.org" => "isc.*" );
483
484 sub access {
485 %return_hash = (
486 "max_rate" => "10000",
487 "addnntppostinghost" => "true",
488 # ...
489 );
490 if( defined $attributes{username} &&
491 $attributes{username} =~ /.*@(.*)/ )
492 {
493 $return_hash{"virtualhost"} = "true";
494 $return_hash{"path"} = $1;
495 $return_hash{"newsgroups"} = $hosts{$1};
496 } else {
497 $return_hash{"read"} = "*";
498 $return_hash{"post"} = "local.*"
499 }
500 return %return_hash;
501 }
502
503 Note that both the keys and values are quoted strings. These values are
504 to be returned to a C program and must be quoted strings. For values
505 containing one or more spaces, it is not necessary to include extra
506 quotes inside the string.
507
508 While you may include the users: parameter in a dynamically generated
509 access group, some care should be taken (unless your pattern is just *
510 which is equivalent to leaving the parameter out). The group created
511 with the values returned from the Perl script is the only one considered
512 when nnrpd attempts to find an access group matching the connection. If
513 a users: parameter is included and it doesn't match the connection, then
514 the client will be denied access since there are no other access groups
515 which could match the connection.
516
517 If access() dies (either due to a Perl error or due to calling die), or
518 if it returns anything other than a hash as described above, an internal
519 error will be reported to the client, the exact error will be logged to
520 syslog, and nnrpd will drop the connection and exit.
521
522 Notes on Writing Embedded Perl
523
524 All Perl evaluation is done inside an implicit eval block, so calling
525 die in Perl code will not kill the innd or nnrpd process. Neither will
526 Perl errors (such as syntax errors). However, such errors will have
527 negative effects (fatal errors in the innd or nnrpd filter will cause
528 filtering to be disabled, and fatal errors in the nnrpd authentication
529 code will cause the client connection to be terminated).
530
531 Calling exit directly, however, *will* kill the innd or nnrpd process,
532 so don't do that. Similarly, you probably don't want to call fork (or
533 any other function that results in a fork such as system,
534 IPC::Open3::open3(), or any use of backticks) since there are possibly
535 unflushed buffers that could get flushed twice, lots of open state that
536 may not get closed properly, and innumerable other potential problems.
537 In general, be aware that all Perl code is running inside a large and
538 complicated C program, and Perl code that impacts the process as a whole
539 is best avoided.
540
541 You can use print and warn inside Perl code to send output to STDOUT or
542 STDERR, but you probably shouldn't. Instead, open a log file and print
543 to it instead (or, in the innd filter, use INN::syslog() to write
544 messages via syslog like the rest of INN). If you write to STDOUT or
545 STDERR, where that data will go depends on where the filter is running;
546 inside innd, it will go to the news log or the errlog, and inside nnrpd
547 it will probably go nowhere but could go to the client. The nnrpd
548 filter takes some steps to try to keep output from going across the
549 network connection to the client (which would probably result in a very
550 confused client), but best not to take the chance.
551
552 For similar reasons, try to make your Perl code -w clean, since Perl
553 warnings are written to STDERR. (INN won't run your code under -w, but
554 better safe than sorry, and some versions of Perl have some mandatory
555 warnings you can't turn off.)
556
557 You *can* use modules in your Perl code, just like you would in an
558 ordinary Perl script. You can even use modules that dynamically load C
559 code. Just make sure that none of the modules you use go off behind
560 your back to do any of the things above that are best avoided.
561
562 Whenever you make any modifications to the Perl code, and particularly
563 before starting INN or reloading filter.perl with new code, you should
564 run perl -wc on the file. This will at least make sure you don't have
565 any glaring syntax errors. Remember, if there are errors in your code,
566 filtering will be disabled, which could mean that posts you really
567 wanted to reject will leak through and authentication of readers may be
568 totally broken.
569
570 The samples directory has example startup_innd.pl, filter_innd.pl,
571 filter_nnrpd.pl, and nnrpd_auth.pl files that contain some simplistic
572 examples. Look them over as a starting point when writing your own.
573
574 Available Packages
575
576 This is an unofficial list of known filtering packages at the time of
577 publication. This is not an endorsement of these filters by the ISC or
578 the INN developers, but is included as assistance in locating packages
579 which make use of this filter mechanism.
580
581 CleanFeed Jeremy Nixon <jeremy@exit109.com>
582 <URL:http://www.exit109.com/~jeremy/news/cleanfeed.html>
583 A spam filter catching excessive multi-posting and a host of
584 other things. Uses filter_innd.pl exclusively, requires the MD5
585 Perl module. Probably the most popular and widely-used Perl
586 filter around.
587
588 Usenet II Filter Edward S. Marshall <emarshal@xnet.com>
589 <URL:http://www.xnet.com/~emarshal/inn/filter_nnrpd.pl>
590 Checks for "soundness" according to Usenet II guidelines in the
591 net.* hierarchy. Designed to use filter_nnrpd.pl.
592
593 News Gizmo Aidan Cully <aidan@panix.com>
594 <URL:http://www.panix.com/gizmo/>
595 A posting filter for helping a site enforce Usenet-II soundness,
596 and for quotaing the number of messages any user can post to
597 Usenet daily.