aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Properly log return value of sendmail invocation if it fails.Steffen Kaiser2008-03-061-1/+1
|
* add note about using mailtrain for other purposes (suggested by Steffen Kaiser)Johannes Berg2008-03-051-0/+13
|
* Fix file descriptor leak in mailtrain backendSteffen Kaiser2008-03-051-2/+3
|
* Leaked a t_pop() call in I/O handler 0x805b350Steffen Kaiser2008-03-051-2/+5
| | | | | | | | | | | | | | | Raw backtrace: imap [0x80ad741] -> imap [0x80ad65c] -> imap(io_loop_handler _run+0x1ce) [0x80b35ce] -> imap(io_loop_run+0x28) [0x80b29a8] -> imap(main+0x5a0) [0x8063590] -> /lib/tls/i686/cmov/libc.so.6(__lib c_start_main+0xc8) [0xb7e37ea8] -> imap [0x8055e31] child 18478 (imap) killed with signal 6 Cause is, if sendmail fails with an exit code != 0, the function returns with -1 immediately skipping the t_pop() before the end of the function. This fixes it.
* allow appends to spam if configured (offlineimap support)Johannes Berg2008-03-054-5/+76
|
* disallow APPEND to unsure/spam foldersJohannes Berg2008-02-191-5/+20
|
* init/reset movetype/save_hack variables appropriatelyJohannes Berg2008-02-181-0/+15
|
* use want_mail variableJohannes Berg2008-02-181-4/+3
|
* remove --stdout from dspam-exec's syslog messageJohannes Berg2008-02-021-1/+1
|
* dspam backend: don't force --stdout and recommend --deliver=Johannes Berg2008-02-022-6/+5
|
* allow arbitrary command line arguments for mailtrain backendJohannes Berg2008-02-022-9/+39
|
* clarify backendJohannes Berg2007-10-291-1/+1
|
* fix readme bugsJohannes Berg2007-10-221-2/+2
|
* indent everything consistentlyJohannes Berg2007-10-151-2/+2
|
* make manual page out of READMEJohannes Berg2007-10-152-161/+166
|
* add note about mailtrain/full message to NOTESJohannes Berg2007-10-151-1/+1
|
* use -WextraJohannes Berg2007-10-155-8/+12
|
* fix compiler warning (missing const)Johannes Berg2007-10-151-1/+1
|
* fix init messageJohannes Berg2007-10-151-1/+1
|
* fix assertion failure bugJohannes Berg2007-10-151-0/+18
| | | | | We need to override 'box->v.save_init' to be able to set 'want_mail' to TRUE since we do need the mail to extract the signature.
* make settings per-pluginJohannes Berg2007-10-149-42/+70
|
* really print out dspam binaryJohannes Berg2007-10-141-1/+1
|
* always print dspam binary in debug outputJohannes Berg2007-10-141-3/+2
|
* use $(CONFIG) everywhereJohannes Berg2007-10-141-3/+3
|
* clean all .so filesJohannes Berg2007-10-141-1/+1
|
* allow setting plugin name from config, using CONFIG=conffile instead of .configJohannes Berg2007-10-143-9/+22
|
* fix typo in startup messageJohannes Berg2007-10-081-1/+1
|
* don't break dspam-exec backend when DEBUG=stderrJohannes Berg2007-10-031-0/+6
|
* change build system a bitJohannes Berg2007-10-021-5/+8
| | | | | | Before this change it used to rm $objs after building, also this change introduces a new PLUGIN_NAME variable so you can override the plugin name from the make command line.
* document crm114 backendJohannes Berg2007-10-021-0/+27
|
* fix crm114, it worksJohannes Berg2007-10-021-6/+7
|
* use configured signature in crm114Johannes Berg2007-10-013-3/+4
|
* fix crm114 build bugJohannes Berg2007-10-011-0/+3
|
* add experimental crm114 backendJohannes Berg2007-10-012-0/+202
|
* fix error in config handlingJohannes Berg2007-10-011-1/+1
|
* introduce 'unsure' folder feature, fix bug from previous commitJohannes Berg2007-10-014-7/+55
|
* use 'enum classification' instead of 'bool from_spam'Johannes Berg2007-10-016-28/+58
|
* don't configure a backend by default and make a nice error for that caseJohannes Berg2007-10-012-1/+8
|
* add mailtrain backend and docsJohannes Berg2007-10-013-1/+333
|
* flush stderr when logging thereJohannes Berg2007-10-011-0/+1
|
* remove dict-backend from NOTESJohannes Berg2007-10-011-19/+0
|
* revert default backend to dspam-execJohannes Berg2007-10-011-2/+2
|
* add non-working SQL logger backendJohannes Berg2007-10-013-2/+151
| | | | | | | | | | It doesn't work with sqlite for the following reasons: (1) sqlite is buggy and needs sql_connect() before starting a transaction although the API states that isn't the case (2) sqlite doesn't understand ON DUPLICATE (1) leads to a segfault when used with dict-proxy. Also see the comment in the source code for the broken transaction handling in this backend.
* improve signature extraction APIJohannes Berg2007-10-013-7/+27
|
* remove some #if 0'ed codeJohannes Berg2007-10-011-17/+0
|
* use last signature header lineJohannes Berg2007-10-011-4/+7
| | | | | | | Sometimes you'll get incoming mail that already had a signature header line from somebody else's misconfigured antispam setup. Use the last found signature header line to avoid retraining failures because the wrong signature was used.
* adjust notes for new internal APIJohannes Berg2007-10-011-6/+2
|
* rename, add install targetJohannes Berg2007-10-014-12/+20
|
* fix defconfigJohannes Berg2007-10-011-1/+1
|
* fix debug outputJohannes Berg2007-10-011-1/+1
|