courier-maildrop Mailing List for Courier Mail Server
Brought to you by:
mrsam
You can subscribe to this list here.
2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(45) |
Nov
(29) |
Dec
(58) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2002 |
Jan
(62) |
Feb
(38) |
Mar
(90) |
Apr
(62) |
May
(53) |
Jun
(61) |
Jul
(118) |
Aug
(83) |
Sep
(139) |
Oct
(72) |
Nov
(135) |
Dec
(90) |
2003 |
Jan
(138) |
Feb
(57) |
Mar
(83) |
Apr
(192) |
May
(79) |
Jun
(88) |
Jul
(50) |
Aug
(78) |
Sep
(112) |
Oct
(129) |
Nov
(73) |
Dec
(24) |
2004 |
Jan
(64) |
Feb
(75) |
Mar
(113) |
Apr
(91) |
May
(67) |
Jun
(83) |
Jul
(86) |
Aug
(70) |
Sep
(67) |
Oct
(37) |
Nov
(105) |
Dec
(72) |
2005 |
Jan
(53) |
Feb
(74) |
Mar
(83) |
Apr
(38) |
May
(47) |
Jun
(49) |
Jul
(73) |
Aug
(38) |
Sep
(71) |
Oct
(87) |
Nov
(94) |
Dec
(77) |
2006 |
Jan
(78) |
Feb
(62) |
Mar
(75) |
Apr
(54) |
May
(9) |
Jun
(35) |
Jul
(47) |
Aug
(33) |
Sep
(40) |
Oct
(66) |
Nov
(56) |
Dec
(31) |
2007 |
Jan
(61) |
Feb
(48) |
Mar
(80) |
Apr
(35) |
May
(17) |
Jun
(37) |
Jul
(31) |
Aug
(35) |
Sep
(16) |
Oct
(20) |
Nov
(19) |
Dec
(31) |
2008 |
Jan
(18) |
Feb
(9) |
Mar
(18) |
Apr
(8) |
May
(14) |
Jun
(13) |
Jul
(15) |
Aug
(16) |
Sep
(22) |
Oct
(1) |
Nov
(2) |
Dec
(11) |
2009 |
Jan
|
Feb
|
Mar
(13) |
Apr
(11) |
May
(9) |
Jun
(27) |
Jul
(11) |
Aug
|
Sep
(4) |
Oct
|
Nov
(10) |
Dec
(3) |
2010 |
Jan
(19) |
Feb
(14) |
Mar
(35) |
Apr
(10) |
May
(14) |
Jun
(12) |
Jul
|
Aug
(10) |
Sep
(10) |
Oct
(1) |
Nov
(24) |
Dec
(8) |
2011 |
Jan
(17) |
Feb
(2) |
Mar
(2) |
Apr
(1) |
May
(27) |
Jun
(14) |
Jul
|
Aug
(10) |
Sep
(7) |
Oct
(14) |
Nov
(19) |
Dec
(20) |
2012 |
Jan
(1) |
Feb
(5) |
Mar
|
Apr
(4) |
May
|
Jun
(6) |
Jul
|
Aug
(4) |
Sep
(3) |
Oct
(2) |
Nov
(1) |
Dec
|
2013 |
Jan
(22) |
Feb
|
Mar
(9) |
Apr
(4) |
May
(2) |
Jun
(38) |
Jul
(3) |
Aug
(6) |
Sep
(1) |
Oct
(4) |
Nov
(4) |
Dec
(15) |
2014 |
Jan
(20) |
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
(17) |
Jul
|
Aug
|
Sep
(5) |
Oct
(2) |
Nov
(1) |
Dec
(2) |
2015 |
Jan
|
Feb
(6) |
Mar
|
Apr
(1) |
May
(3) |
Jun
(10) |
Jul
(6) |
Aug
(16) |
Sep
(2) |
Oct
|
Nov
|
Dec
(22) |
2016 |
Jan
|
Feb
(16) |
Mar
|
Apr
(8) |
May
(1) |
Jun
|
Jul
|
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
(5) |
2017 |
Jan
(6) |
Feb
|
Mar
(4) |
Apr
(10) |
May
(1) |
Jun
(5) |
Jul
(2) |
Aug
(1) |
Sep
|
Oct
(4) |
Nov
(9) |
Dec
(14) |
2018 |
Jan
(3) |
Feb
(3) |
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
(3) |
Sep
(4) |
Oct
|
Nov
(2) |
Dec
(6) |
2019 |
Jan
|
Feb
|
Mar
(6) |
Apr
|
May
|
Jun
(4) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
|
2020 |
Jan
|
Feb
|
Mar
|
Apr
(2) |
May
(2) |
Jun
(6) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
|
2021 |
Jan
|
Feb
(1) |
Mar
(7) |
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(11) |
Dec
|
2022 |
Jan
(1) |
Feb
|
Mar
(3) |
Apr
(1) |
May
|
Jun
(4) |
Jul
|
Aug
(5) |
Sep
(7) |
Oct
|
Nov
(1) |
Dec
(1) |
2023 |
Jan
|
Feb
(7) |
Mar
|
Apr
(3) |
May
|
Jun
(8) |
Jul
|
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
2024 |
Jan
|
Feb
|
Mar
(2) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Sam V. <mr...@co...> - 2024-03-13 11:31:51
|
Philip Rhoades via Courier-maildrop writes: > People, > > I have started seeing long maildrop texts in my IndiMail (QMail) logs - not > sure what I have done - see the attached (I have added newlines after colons > to make it easier for me to read) but I am wondering about the last line . . > and how to allow the long report . . The mails are getting delivered OK . . This looks like maildrop is getting invoked with the -V option to deliver mail. -V should be used only when manually running maildrop, for diagnostic and debugging purposes. Just remove the -V option, to get rid of all of that. |
From: Philip R. <ph...@pr...> - 2024-03-13 08:35:58
|
People, I have started seeing long maildrop texts in my IndiMail (QMail) logs - not sure what I have done - see the attached (I have added newlines after colons to make it easier for me to read) but I am wondering about the last line . . and how to allow the long report . . The mails are getting delivered OK . . Thanks, Phil. -- Philip Rhoades PO Box 896 Cowra NSW 2794 Australia E-mail: ph...@pr... |
From: Sam V. <mr...@co...> - 2023-08-14 14:38:54
|
I created a new PGP key, to replace the expiring one, in addition to a minor release of courier, maildrop, sqwebmail, and cone packages Download: https://www.courier-mta.org/download.html Changes: • maildrop: fix a compilation error when some non-default options are specified to the configure script. • sqwebmail, cone: Recognize names of the newer EC-based public key algorithms when displaying available PGP keys. |
From: Sam V. <mr...@co...> - 2023-06-21 10:47:55
|
Bengt Thuree writes: > <URL:https://wiki.musl-libc.org/functional-differences-from- > glibc.html>https://wiki.musl-libc.org/functional-differences-from-glibc.html This does explain the observed behavior. It seems that the POSIX behavior makes it impossible to determine that the encoding conversion is invalid. That's unfortunate, so only with glibc would cone and sqwebmail report a transcoding error. This test will need to be adjusted, in order to pass, however I lack the ability to verify any changes to it, you'll need to give this your best shot, and see what happens. |
From: Bengt T. <be...@th...> - 2023-06-21 06:13:47
|
Hmmm Looks like iconv implementation is a bit wrong? https://sourceware.org/bugzilla/show_bug.cgi?id=14008 https://wiki.musl-libc.org/functional-differences-from-glibc.html Bengt On Wed, 21 June 2023, 10:28 Sam Varshavchik, <mr...@co...> wrote: > Bengt Thuree writes: > > > « HTML content follows » > > > > > > Hi Sam, > > > > And thank you very much for your reply and explanation. Much appreciated. > > I will dig a bit deeper and try to figure out what is happening here. > > My C is a bit rusty though, but good time to refresh it. > > > > > > Any suggestions on an area to look more into? > > The current version of the iconv(3) manual page: > > https://man7.org/linux/man-pages/man3/iconv.3.html > > courier-unicode expects iconv to return EILSEQ, as described by the first > failure case, there. This will be encountered by convert_flush_iconv() > which > eventually gets called from unicode_convert_deinit(). > > This does require not merely knowing C, but having debugging know-how, to > debug the code and see exactly what iconv returns. > > From https://man7.org/linux/man-pages/man3/iconv_open.3.html the closest > matching description of your observed results is the //IGNORE option, > where > unrepresentable characters get discarded; but there is no option to > replace > them with a substitute character. > > Reading up on Alpine Linux, it seems to be using its own C library, so it > looks to me like its implementation of iconv differs in behavior from the > one in glibc. Perhaps there's a way to set it up to match glibc's iconv > behavior, in this instance. > > _______________________________________________ > Courier-maildrop mailing list > Cou...@li... > https://lists.sourceforge.net/lists/listinfo/courier-maildrop > |
From: Sam V. <mr...@co...> - 2023-06-21 00:28:06
|
Bengt Thuree writes: > « HTML content follows » > > > Hi Sam, > > And thank you very much for your reply and explanation. Much appreciated. > I will dig a bit deeper and try to figure out what is happening here. > My C is a bit rusty though, but good time to refresh it. > > > Any suggestions on an area to look more into? The current version of the iconv(3) manual page: https://man7.org/linux/man-pages/man3/iconv.3.html courier-unicode expects iconv to return EILSEQ, as described by the first failure case, there. This will be encountered by convert_flush_iconv() which eventually gets called from unicode_convert_deinit(). This does require not merely knowing C, but having debugging know-how, to debug the code and see exactly what iconv returns. >From https://man7.org/linux/man-pages/man3/iconv_open.3.html the closest matching description of your observed results is the //IGNORE option, where unrepresentable characters get discarded; but there is no option to replace them with a substitute character. Reading up on Alpine Linux, it seems to be using its own C library, so it looks to me like its implementation of iconv differs in behavior from the one in glibc. Perhaps there's a way to set it up to match glibc's iconv behavior, in this instance. |
From: Bengt T. <be...@th...> - 2023-06-20 17:52:37
|
Hi Sam, And thank you very much for your reply and explanation. Much appreciated. I will dig a bit deeper and try to figure out what is happening here. My C is a bit rusty though, but good time to refresh it. Any suggestions on an area to look more into? Cheers Bengt On Tue, Jun 20, 2023 at 8:55 PM Sam Varshavchik <mr...@co...> wrote: > Bengt Thuree writes: > > > > And when I try the unicodetest by itself, sure enough, that error message pop > > > ped up. > > I added some extra printouts, and I got the following output > > > > $ ./unicodetest > > 1cb.cnt: 0 > > cb.ptr: > > checkflag: 0 > > 2cb.cnt: 0 > > cb.ptr: > > checkflag: 0 > > 3cb.cnt: 0 > > cb.ptr: > > checkflag: 0 > > 4cb.cnt: 4 > > cb.ptr: A**B > > checkflag: 0 > > Unexpected result from convert() > > 5cb.cnt: 4 > > cb.ptr: A**B > > checkflag: 0 > > > I also started an ubuntu server in a virtual box, and on this server, > > the make check command works like a charm. > > So there are some differences between Alpine and Ubuntu here, > > but I'm not sure where I should start. > > locale -a gives the same (C, and C.UTF-8, but ubuntu also has POSIX) > > I tried with bash (instead of default ash), but no difference > > Would anyone have any thoughts, or suggestions on this issue? > > This test checks what happens when text cannot be converted to a > different > character set. The expected result is that the non-convertible text gets > skipped. Your output shows that, instead, non-convertible characters get > replaced by asterisk characters. > > That's all that can be determined, based on the above. courier-unicode > does > need to know when text cannot be properly converted from one encoding to > another. Pretending that everything is cool, and just replacing the non- > convertible characters with placeholder characters, is not cool. > > Some further investigation would be needed to understand the difference > in > behavior, and figure out the solution. > _______________________________________________ > Courier-maildrop mailing list > Cou...@li... > https://lists.sourceforge.net/lists/listinfo/courier-maildrop > |
From: Sam V. <mr...@co...> - 2023-06-20 10:55:00
|
Bengt Thuree writes: > And when I try the unicodetest by itself, sure enough, that error message pop > ped up. > I added some extra printouts, and I got the following output > > $ ./unicodetest > 1cb.cnt: 0 > cb.ptr: > checkflag: 0 > 2cb.cnt: 0 > cb.ptr: > checkflag: 0 > 3cb.cnt: 0 > cb.ptr: > checkflag: 0 > 4cb.cnt: 4 > cb.ptr: A**B > checkflag: 0 > Unexpected result from convert() > 5cb.cnt: 4 > cb.ptr: A**B > checkflag: 0 > I also started an ubuntu server in a virtual box, and on this server, > the make check command works like a charm. > So there are some differences between Alpine and Ubuntu here, > but I'm not sure where I should start. > locale -a gives the same (C, and C.UTF-8, but ubuntu also has POSIX) > I tried with bash (instead of default ash), but no difference > Would anyone have any thoughts, or suggestions on this issue? This test checks what happens when text cannot be converted to a different character set. The expected result is that the non-convertible text gets skipped. Your output shows that, instead, non-convertible characters get replaced by asterisk characters. That's all that can be determined, based on the above. courier-unicode does need to know when text cannot be properly converted from one encoding to another. Pretending that everything is cool, and just replacing the non- convertible characters with placeholder characters, is not cool. Some further investigation would be needed to understand the difference in behavior, and figure out the solution. |
From: Bengt T. <be...@th...> - 2023-06-20 07:27:02
|
Hello everyone I am trying to port maildrop to Alpine, mainly because I need the reformail command. But maildrop needed courier-unicode, so I need to port this one first. But I am running into one issue which I am not sure what to do with. So I am just checking here to see if there are any dependencies that I am not aware of. Configure works fine Make works fine make check fails 12345678alpinedev:~/aports/testing/courier-unicode/src/courier-unicode-2.2.6$ make check make check-am make[1]: Entering directory '/home/bengt/aports/testing/courier-unicode/src/courier-unicode-2.2.6' ./unicodetest Unexpected result from convert() make[1]: *** [Makefile:1651: check-am] Error 1 make[1]: Leaving directory '/home/bengt/aports/testing/courier-unicode/src/courier-unicode-2.2.6' make: *** [Makefile:1333: check] Error 2 And when I try the unicodetest by itself, sure enough, that error message popped up. I added some extra printouts, and I got the following output $ ./unicodetest 1cb.cnt: 0 cb.ptr: checkflag: 0 2cb.cnt: 0 cb.ptr: checkflag: 0 3cb.cnt: 0 cb.ptr: checkflag: 0 4cb.cnt: 4 cb.ptr: A**B checkflag: 0 Unexpected result from convert() 5cb.cnt: 4 cb.ptr: A**B checkflag: 0 and the code looks like this static void test1() { static const char teststr[]= { 0x00, 0x00, 0x00, 0x41, 0x00, 0x00, 0x04, 0x14, 0x00, 0x00, 0x04, 0x30, 0x00, 0x00, 0x00, 0x42}; char outputbuf[12]; struct collect_buf cb; unicode_convert_handle_t h; int checkflag; cb.ptr=outputbuf; cb.cnt=0; cb.size=sizeof(outputbuf); fprintf(stderr, "1cb.cnt: %zu\n", cb.cnt); // Print cb.cnt value fprintf(stderr, "cb.ptr: %.*s\n", (int)cb.cnt, cb.ptr); // Print cb.ptr value fprintf(stderr, "checkflag: %d\n", checkflag); // Print checkflag value if ((h=unicode_convert_init("UCS-4BE", "ISO-8859-1", save_output, &cb)) == NULL) { perror("unicode_convert_init"); exit(1); } fprintf(stderr, "2cb.cnt: %zu\n", cb.cnt); // Print cb.cnt value fprintf(stderr, "cb.ptr: %.*s\n", (int)cb.cnt, cb.ptr); // Print cb.ptr value fprintf(stderr, "checkflag: %d\n", checkflag); // Print checkflag value unicode_convert(h, teststr, sizeof(teststr)); fprintf(stderr, "3cb.cnt: %zu\n", cb.cnt); // Print cb.cnt value fprintf(stderr, "cb.ptr: %.*s\n", (int)cb.cnt, cb.ptr); // Print cb.ptr value fprintf(stderr, "checkflag: %d\n", checkflag); // Print checkflag value if (unicode_convert_deinit(h, &checkflag)) { perror("unicode_convert_deinit"); exit(1); } fprintf(stderr, "4cb.cnt: %zu\n", cb.cnt); // Print cb.cnt value fprintf(stderr, "cb.ptr: %.*s\n", (int)cb.cnt, cb.ptr); // Print cb.ptr value fprintf(stderr, "checkflag: %d\n", checkflag); // Print checkflag value if (cb.cnt != 2 || memcmp(cb.ptr, "AB", 2) || !checkflag) { fprintf(stderr, "Unexpected result from convert()\n"); fprintf(stderr, "5cb.cnt: %zu\n", cb.cnt); // Print cb.cnt value fprintf(stderr, "cb.ptr: %.*s\n", (int)cb.cnt, cb.ptr); // Print cb.ptr value fprintf(stderr, "checkflag: %d\n", checkflag); // Print checkflag value exit(1); } } I also started an ubuntu server in a virtual box, and on this server, the make check command works like a charm. So there are some differences between Alpine and Ubuntu here, but I'm not sure where I should start. locale -a gives the same (C, and C.UTF-8, but ubuntu also has POSIX) I tried with bash (instead of default ash), but no difference Would anyone have any thoughts, or suggestions on this issue? Cheers Bengt |
From: Mahlon E. S. <ma...@ma...> - 2023-06-12 23:24:13
|
I've recently been rebuilding my mail environment, and finally pushing procmail out in favor of maildrop. (I know, I know. I'm waaay more than a little late to the game. :) Running into a wall. Specifically, any use of xfilter seems to timeout. The entirety of my .mailfilter is as follows: SHELL="/bin/sh" DEFAULT=$HOME/Maildir LOGFILE=$HOME/maildrop.log logfile "$LOGFILE" xfilter "/bin/cat" My understanding of this config is that the message should pass through 'cat' unmolested, and delivered to $DEFAULT. Upon new incoming mail, the maildrop process appears to just... hang out, until it gets a timeout. They stack up. mahlon 30754 0.0 0.0 18032 5356 - IJ 15:37 0:00.01 /usr/local/bin/maildrop mahlon 31352 0.0 0.0 17844 5240 - IJ 15:38 0:00.01 /usr/local/bin/maildrop mahlon 31380 0.0 0.0 17844 5240 - IJ 15:38 0:00.01 /usr/local/bin/maildrop If I comment out the xfilter, it delivers instantly. This is using Maildrop 3.1.5, using (net)qmail on FreeBSD 13.1. My qmail delivery instructions are simply: | /usr/local/bin/maildrop Final message is: "maildrop: Timeout quota exceeded." I can't escape the feeling that I am doing something patently silly here, but I'm just not seeing it. I appreciate any cluebats to be bludgeoned with, or hints on how to provide better debug. Thanks! -- Mahlon E. Smith http://www.martini.nu/ |
From: Mahlon E. S. <ma...@ma...> - 2023-06-12 23:13:06
|
> > I can't escape the feeling that I am doing something patently silly here, but > I'm just not seeing it. I appreciate any cluebats to be bludgeoned with, or > hints on how to provide better debug. Immediately after sending this, I tried just: % /usr/local/bin/maildrop .mailfilter < message And it works great. Perhaps there's some trick with qmail delivery line that I've missed, someone can share? -- Mahlon E. Smith http://www.martini.nu/ |
From: Sam V. <mr...@co...> - 2023-04-30 13:34:27
|
Download: https://www.courier-mta.org/download.html New releases of courier, courier-imap, sqwebmail, maildrop, and cone packages. Changes: * all: fix warning message from gcc 13 (false positives). * courier, courier-imap, cone: update configure script to check for the standardized location of the CA certificate bundle, in addition to various legacy compatibility paths. * courier: update the ESMTP server to recognize alternative Courier ESMTP extension names, in addition to the current ones. A future version of Courier will switch the syntax of its custom ESMTP extension names to be strictly compliant with the ESMTP specification. The names remain unchanged for now, but this version of Courier recognizes both the existing names and the new names. * courier: update internal scripts, replace deprecated fgrep alias with grep -F. |
From: Sam V. <mr...@co...> - 2023-04-06 12:33:35
|
Download: https://www.courier-mta.org/download.html maildrop 3.1.3 fixes a build issue when using Dovecot authentication. |
From: Sam V. <mr...@co...> - 2023-04-05 01:55:33
|
Download: https://www.courier-mta.org/download.html New releases courier, courier-imap, sqwebmail, maildrop, and cone packages. Changes: • maildrop: overhaul of internal code, updating pre-C++03 code to C++11. As part of the code update, the "foreach" command now supports subpattern matches setting $MATCH<n>. Fix latent, rare bugs that could theoretically result in crashes, that were lurking around for years. The -V option initializes the VERBOSE variable, for consistency. • all: check for minimum libidn version in the configure script. • courier: fix issues with system accounts that are in a large number of groups. |
From: Sam V. <mr...@co...> - 2023-02-20 00:39:57
|
Download: https://www.courier-mta.org/download.html New releases of all Courier packages. Changes: * courier: adjust esmtp server's ulimit to match the imap server's ulimit * courier: fix error handling with some versions of OpenSSL that resulted in an error if the certificate file did not have the optional DH parameter section. * courier-imap: fix parsing of the IMAp APPEND command. * maildrop: reimplement how timeouts works, to prevent a potential crash in the rare event of a lock timeout. |
From: Francesco A. <fa...@ar...> - 2023-02-07 13:24:52
|
Hello Sam, Il 07 febbraio 2023 alle 07:35 Sam Varshavchik ha scritto: > getmail is not involved here. This is how the mail was sent. It's strange to > observe variation in formats of automated github spam. Maybe they're doing > A/B testing, and either the A, or the B, is sending broken mail. > > > if ( /^@[a-zA-Z0-9_-]+ pushed [0-9]+ commits?.$/:bD ) > > The period actually matches any character. This is a Perl RE. Try: > > > if ( /^@[a-zA-Z0-9_-]+ pushed [0-9]+ commits?\.\r?$/:bD ) > > to match a real period, and an optional CR, which is what that encodes. > maildrop should be able to decode it, and apply the regexp. That worked perfectly. I am quite happy with my mutt/maildrop/getmail setup, many thanks —F |
From: Sam V. <mr...@co...> - 2023-02-07 12:35:36
|
Francesco Ariis writes: > Hello, > I have finished setting up my ~/.mailfilter. Almost everything > functions well. > One thing that gives me troubles is possibly related to message > encoding. > > > There are some GitHub push messages that I do want to see, like > > @Xyz Pushed 3 commits > > Hence I wrote this: > > if ( /^@[a-zA-Z0-9_-]+ pushed [0-9]+ commits?.$/:bD ) > to $MAILFOLDER/spam > > (notice the ‘$’ at the end.) > > Now, *some* messages are caught by this filter, some not. I inspected > two of them with vim > > ----==_mimepart_xyz > Content-Type: text/plain; > charset=UTF-8 > Content-Transfer-Encoding: 7bit > > @someone pushed 5 commits. > > ----==_mimepart_qwz > Content-Type: text/plain; > charset=UTF-8 > Content-Transfer-Encoding: quoted-printable > > @else pushed 6 commits.=0D > > The message which does not trigger the pattern has an additional “=0D” > at the end. > > I know that I could just remove the ‘$’ from my regexp to make it work > on both messages, but I wonder whether there is a more principled solution, > if I should report this to my MRA (getmail) or what else. getmail is not involved here. This is how the mail was sent. It's strange to observe variation in formats of automated github spam. Maybe they're doing A/B testing, and either the A, or the B, is sending broken mail. > if ( /^@[a-zA-Z0-9_-]+ pushed [0-9]+ commits?.$/:bD ) The period actually matches any character. This is a Perl RE. Try: > if ( /^@[a-zA-Z0-9_-]+ pushed [0-9]+ commits?\.\r?$/:bD ) to match a real period, and an optional CR, which is what that encodes. maildrop should be able to decode it, and apply the regexp. |
From: Francesco A. <fa...@ar...> - 2023-02-07 09:18:54
|
Hello, I have finished setting up my ~/.mailfilter. Almost everything functions well. One thing that gives me troubles is possibly related to message encoding. There are some GitHub push messages that I do want to see, like @Xyz Pushed 3 commits Hence I wrote this: if ( /^@[a-zA-Z0-9_-]+ pushed [0-9]+ commits?.$/:bD ) to $MAILFOLDER/spam (notice the ‘$’ at the end.) Now, *some* messages are caught by this filter, some not. I inspected two of them with vim ----==_mimepart_xyz Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit @someone pushed 5 commits. ----==_mimepart_qwz Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable @else pushed 6 commits.=0D The message which does not trigger the pattern has an additional “=0D” at the end. I know that I could just remove the ‘$’ from my regexp to make it work on both messages, but I wonder whether there is a more principled solution, if I should report this to my MRA (getmail) or what else. Thanks —F |
From: Francesco A. <fa...@ar...> - 2023-02-06 06:46:18
|
Il 05 febbraio 2023 alle 22:31 Sam Varshavchik ha scritto: > That should be doable as > > if ( /^List-Id:.*<discourse.something.org>/:h && \ > !/^Subject:.*\[PM\]/:h) > Thanks Sam, that did it —F |
From: Sam V. <mr...@co...> - 2023-02-06 03:31:31
|
Francesco Ariis writes: > Hello, > I am a new maildrop user. > Today I was writing a pattern like this: > > if ( /^List-Id:.*<discourse.something.org>/:h || && \ > /^Subject:.*\[PM\]/:h || ) > to $MAILFOLDER/lists > > I would like to tell maildrop “deliver to `lists` if the first pattern > is matched but *not* the second”. How can I do that? That should be doable as if ( /^List-Id:.*<discourse.something.org>/:h && \ !/^Subject:.*\[PM\]/:h) |
From: Francesco A. <fa...@ar...> - 2023-02-05 20:41:47
|
Hello, I am a new maildrop user. Today I was writing a pattern like this: if ( /^List-Id:.*<discourse.something.org>/:h || && \ /^Subject:.*\[PM\]/:h || ) to $MAILFOLDER/lists I would like to tell maildrop “deliver to `lists` if the first pattern is matched but *not* the second”. How can I do that? Thanks in advance —F |
From: Sam V. <mr...@co...> - 2022-12-04 17:56:33
|
Download: https://www.courier-mta.org/download.html New releases of all Courier packages. Changes: • Use libidn2 instead of libidn. • "make install" creates relative, instead of absolute, symbolic links. • debuild script: update lintian overrides, add pkg-config to build dependencies. |
From: Sam V. <mr...@co...> - 2022-11-14 12:54:41
|
Download: https://www.courier-mta.org/download.html New releases of all Courier packages. Changes: • all: update deb packaging. Check /etc/lsb-release and include the distribution release in the deb package version, to faciliate updating to the same version of the package in an updated release. Fix build dependencies. • sysconftool: add "use strict+warnings" • courier: add a MISC section in the esmtpd-ssl configuration file, for stashing any custom config settings. • courier: update rpm packaging. Remove constructs for historical rpm-based Linux distributions. • maildrop: update deliverquota man page. |
From: Milan O. <cou...@di...> - 2022-09-10 11:35:34
|
On Sat, 10 Sep 2022 07:24:44 -0400 Sam Varshavchik <mr...@co...> wrote: [ snip ] > Your locale is not UTF-8? > > $ locale > LANG=en_US.UTF-8 > LC_CTYPE="en_US.UTF-8" > LC_NUMERIC="en_US.UTF-8" > LC_TIME="en_US.UTF-8" > LC_COLLATE="en_US.UTF-8" > LC_MONETARY="en_US.UTF-8" > LC_MESSAGES="en_US.UTF-8" > LC_PAPER="en_US.UTF-8" > LC_NAME="en_US.UTF-8" > LC_ADDRESS="en_US.UTF-8" > LC_TELEPHONE="en_US.UTF-8" > LC_MEASUREMENT="en_US.UTF-8" > LC_IDENTIFICATION="en_US.UTF-8" > LC_ALL= > > reformime reads the encoding from the system locale. You can use -c > UTF-8 to override it. It's all in the man page. > Thank you! This was the missing point. I overlooked it in the man page, adding -c UTF-8 does exactly what was expected. Regards, Milan |
From: Sam V. <mr...@co...> - 2022-09-10 11:24:52
|
Milan Obuch writes: > Creating file acc with some accented characters, so > > % cat acc > ľšťž > > and executing > > % reformime -o "`cat acc`" | hd > 00000000 0a |.| > 00000001 > > points to reformime itself being the culprit... Could this be > eventually some issue being fixed in newer version? I have Courier > suite 1.1.4 installed and Courier Unicode library 2.2.4, if that's > relevant. Your locale is not UTF-8? $ locale LANG=en_US.UTF-8 LC_CTYPE="en_US.UTF-8" LC_NUMERIC="en_US.UTF-8" LC_TIME="en_US.UTF-8" LC_COLLATE="en_US.UTF-8" LC_MONETARY="en_US.UTF-8" LC_MESSAGES="en_US.UTF-8" LC_PAPER="en_US.UTF-8" LC_NAME="en_US.UTF-8" LC_ADDRESS="en_US.UTF-8" LC_TELEPHONE="en_US.UTF-8" LC_MEASUREMENT="en_US.UTF-8" LC_IDENTIFICATION="en_US.UTF-8" LC_ALL= reformime reads the encoding from the system locale. You can use -c UTF-8 to override it. It's all in the man page. |