gmane.mail.nail.devel

S-nail has been coverity scanned (project 444)...

Steffen "Daode" Nurpmeso — 2013-05-21T10:16:08

Heya,
filling the void i want to mention that i registered S-nail (it's
project 444) at coverity.com [1].  Out of curiosity...

And that was a good thing, since it responded with 104 issues for
the first build and five for the second, many of which were really
bugs, some even catastrophic!  Damn!!

So i've spent the last one and a half day to iterate through those
and can finally report that S-nail's *next* branch doesn't produce
coverity errors anymore (and as of today)!

The severity of some of the bugs made me think about releasing a
14.3, however, since it'll take quite a long time until the
original 14.3 issues will be addressed; i.e., spending this week
on IMAP / POP3 testing, doing reviews of the few things that have
been done development-wise on *next*, and release a 14.3 once this
is done.  Then, release a 14.4 with all the things that were
planned for 14.3, like fixed thread selection etc. at the end of
this summer.

Yep, so i'll come back to you in a few days with a 14.3!
Ciao,

[1] <http://scan2.

First step to control pipe-MIMETYPE command execution

Steffen "Daode" Nurpmeso — 2013-05-03T14:42:18

Hello,
i was kind of „impressed“ that the pipe-MIMETYPE mechanism as it
is today really gets used for anything else than preparation of
HTML MIME parts, because i personally felt much too insecure to
do so because of some **** mails which come through and contain
attachments of various sort, and i usually use 'p:n' to look into
my boxes…

So, whereas it is not really possible to offer as much control as,
e.g., mutt(1), or any other mailer that has a GUI, especially not
today with our recursive shared parser/sender (display
preparator), here is a patch that extends the pipe-MIMETYPE syntax
to offer at least a little bit of control.
The attached diff is for S-nail 14.2.2 and thus also the *master*
branch from the repo (patch -p1 < pipe-prefix.diff).

  (Of course you could also upgrade to the *next* branch instead,
  though i wouldn't recommend it; why not use a temporary branch as in
    git co -b experimental-pipe-control
    patch -p1 < pipe-prefix.diff
    git add --all
    git ci -m 'seems to work f

Announcing S-nail 14.2.2

Steffen "Daode" Nurpmeso — 2013-05-01T14:38:35

Sigh.
So here is S-nail 14.2.2.
S-nail(1) is a fork of Heirloom mailx(1), the former nail(1).

The release tarball can be downloaded (e.g., via 'curl -vv -L')
from, and its checksums are:

  <https://sourceforge.net/projects/s-nail/files/latest/download?source=files>

  MD5    = 8331beb5307c329e5b2c80a28e362b5e
  SHA1   = ec4aba4b26fcf2a60ef5b50c8aa27febd814d00b
  SHA256 = 602a31085573426d22f042b2a8dd6243e79450130de7bf5a4d8613c5c8d1314a

  Online manual: <http://sdaoden.users.sourceforge.net/code-nail.html>
  [Web site    : <http://sdaoden.users.sourceforge.net/code.html#s-nail>]
  git(1) repo  : <git.code.sf.net/p/s-nail/code>
  git(1) browse: <http://sourceforge.net/p/s-nail/code/>

  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
  ! NOTE:                                                           !
  ! the release tarball for 14.2.1 will be removed from the server! !
  ! While there, i'll also drop 13.3.                               !
  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Attachments not working

Gavin Troy — 2013-05-01T13:00:10

Hi list,


Really briefly, attachments aren't working for me (since 14.1 I think). eg,
I can not write & then view a jpeg, nor use pipe->cat->file to save a PDF
attachment.

Is this only an error on my end?

Regards,
Gavin Troy
------------------------------------------------------------------------------
Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
Get 100% visibility into your production application - at no cost.
Code-level diagnostics for performance bottlenecks with <2% overhead
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap1_______________________________________________
nail-devel mailing list
nail-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/nail-devel

Announcing S-nail 14.2.1

Steffen "Daode" Nurpmeso — 2013-04-30T16:28:29

welcome to S-nail 14.2.1.
S-nail(1) is a fork of Heirloom mailx(1), the former nail(1).

The release tarball can be downloaded (e.g., via 'curl -vv -L')
from, and its checksums are:

  <https://sourceforge.net/projects/s-nail/files/latest/download?source=files>

  MD5    = 76cab492c9b15233171493dd125b71dd
  SHA1   = 9d0530a2b8a4276e19986ab0f57de85770bd55a2
  SHA256 = 9c87736973165c3b61a434ac18224bec05e6fc5fa2b0364273e0ed28e49bcaa5

  Online manual: <http://sdaoden.users.sourceforge.net/code-nail.html>
  [Web site    : <http://sdaoden.users.sourceforge.net/code.html#s-nail>]
  git(1) repo  : <git.code.sf.net/p/s-nail/code>
  git(1) browse: <http://sourceforge.net/p/s-nail/code/>

It follows the NEWS entry for 14.2.1; the full changelog of the
commits in between s-nail-14.2 and s-nail-14.2.1 can be inspected by
issuing the following git(1) command:

  $ git log --reverse --topo-order s-nail-14.2..s-nail-14.2.1

v14.2.1, 2013-04-30
-------------------

An unplanned (minor) bugfix release after i've found two bu

..on cherry-picked bugfixes

Steffen "Daode" Nurpmeso — 2013-04-30T14:13:29

Hello all,
i've detected two bugs of mine and want to report they're fixed
and cherry-picked onto the *master* branch.
The plain diffs are also attached below and may be patched into
S-nail 14.2 (small offset, but will succeed).

. The first bug is a possible buffer overflow in a snprintf(3)
  statement that may happen when displaying a mail.
  In short -- we yet did not reserve any space for printing two
  integers but assumed that the numbers would well fit into the 16
  format characters that disappear during the snprintf(3).
  This is not true when the second bug hitted though, since then
  UINT64_MAX (may) happen(s), and that is 18446744073709551615.

. The second one is a partial reverse of (Tweak MIME boundary
  detection.., 2012-12-20), or, to be exact, we'll use the
  original if() condition again (but adjusted to new codeflow).
  A Microsoft Word-created mail on the ICU list revealed
  a weakness in the boundary code that caused some boundary to
  be missed.  (We _want_ the single-pass MIME part pa

Fwd: Re: Testing s-nail 14.2

Steffen "Daode" Nurpmeso — 2013-03-27T17:22:55

Sorry, missed the list..

-------- Original Message --------
Date: Wed, 27 Mar 2013 18:14:18 +0100
From: Steffen "Daode" Nurpmeso <sdaoden-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
To: peters-0bfbc7pHNvkQoVVkSDTaOQ< at >public.gmane.org
Subject: Re: Testing s-nail 14.2

Hello,

peters-0bfbc7pHNvkQoVVkSDTaOQ< at >public.gmane.org wrote:
 |Hi,
 |
 |I am currently testing version 14.2 of s-nail. First thing I noticed:
 |If "bsdcompat" is set, the header summary looks garbled: It prints
 |a newline before the closing quote of the subject.

I just pushed the appended fix to the *master* branch.
Many thanks for reporting this issue!

 |Regards
 |Peters

Ciao, and i hope the second look is more appealing..

--steffen

Date:   2013-03-27 18:04:11 +0100

    cmd1.c:hprf(): fix subject length calculation..

    (printhead(): support "-" for all fields etc., 2013-03-09)
    added support for field alignment for (almost) all fields.
    It was missed that subtracting a value from a negative value
    actually increases the res

Announcing S-nail 14.2

Steffen "Daode" Nurpmeso — 2013-03-15T11:24:28

Well, hello all again ;,

welcome to S-nail 14.2.
S-nail(1) is a fork of Heirloom mailx(1), the former nail(1).

The release tarball can be downloaded (e.g., via 'curl -vv -L')
from, and its checksums are:

  <https://sourceforge.net/projects/s-nail/files/latest/download?source=files>

  MD5    = 268a34f65aa6e34003cd3173df88b4f0
  SHA1   = 81c2b1a5dcbc9e4d2971f00790ef03563bf5d0b2
  SHA256 = 4f4d98e6d0a1a2d0aaaec14223d0499ae3db34fb46c556bb27bd2a354734a735

  Online manual: <http://sdaoden.users.sourceforge.net/code-nail.html>
  [Web site    : <http://sdaoden.users.sourceforge.net/code.html#s-nail>]
  git(1) repo  : <git.code.sf.net/p/s-nail/code>
  git(1) browse: <http://sourceforge.net/p/s-nail/code/>

Note:
  The big big MIME and send layer rewrite will come and cumulate in
  v15.0, most likely mid to late 2014.
  Until then there is one more minor version to be expected, one that
  fixes thread sorting, thread selection, adds a %n format for
  *headline* and other minor stuff that springs into existence in

Bugfix for S-nail 14.1

Steffen "Daode" Nurpmeso — 2013-03-14T18:53:55

I need to create automatized tests which catch such cases.
So there has to be v14.2 pretty soon, and the mentioned *headline*
%n / thread sorting and selection release will be v14.3.
Aaach -- too stupid to add this pure code slacking commit shortly
before the release, *damn*.

Date:   2013-03-14 19:47:05 +0100

    mime_fromhdr(): never return NULL output..
    
    S-nail 14.1 can be crashed with an empty header field; i thought
    this ends up as NULL/non-existent, but no, it doesn't.
    Therefore we try to terminate a non-existent buffer, which ends
    up as a SEGV.
    
    Also, mime_fromhdr() *must* return with a *terminated*, *newly
    allocated* buffer to satisfy its callers, so just do so.
    This is XXX, of course.
    
    Excuse:
    With the head at the Unicode stuff that i'm about to do
    i wanted to get that thing into the release, and obviously did
    a premature commit.  And here it is what happens.
---
 cmd1.c |    7 +++----
 mime.c |   11 +++++++++--
 2 files changed, 12 inserti

Announcing S-nail 14.1

Steffen "Daode" Nurpmeso — 2013-03-12T18:49:26

Hello all,

welcome to S-nail 14.1.
S-nail(1) is a fork of Heirloom mailx(1), the former nail(1).

The release tarball can be downloaded (e.g., via 'curl -vv -L')
from, and its checksums are:

  <https://sourceforge.net/projects/s-nail/files/latest/download?source=files>

  MD5    = 83d243c677ec2d0bae6524d6e6a4d28d
  SHA1   = 7d1c67e16fb4bfaefdf38b28ca55f288375c504d
  SHA256 = 1e267137dc4698b2366a97a6f864940f260d877583faf916cd3905db4585197f

  Online manual: <http://sdaoden.users.sourceforge.net/code-nail.html>
  [Web site    : <http://sdaoden.users.sourceforge.net/code.html#s-nail>]
  git(1) repo  : <git.code.sf.net/p/s-nail/code>
  git(1) browse: <http://sourceforge.net/p/s-nail/code/>

Note:
  The big big MIME and send layer rewrite will come and cumulate in
  v15.0, most likely mid to late 2014.
  Until then there is one more minor version to be expected, one that
  fixes thread sorting, thread selection, adds a %n format for
  *headline* and other minor stuff that springs into existence in the
  meanwhi

smtp-auth-* with user< at >host fixed - please DO reportbugs!

Steffen "Daode" Nurpmeso — 2013-03-04T10:26:14

Hi list,
i've just stumbled over the fact that smtp-auth-XY-user< at >host
syntax was broken.
I googled and - voilá - that has been reported in 2011 and even
been patched in january 2012 by Gavin Troy!

  So *please*, if you do know about misbehaviours or bugs in
  S-nail(1), would you be so kind and report them?

I am subscribed to nail-devel< at > since 2012-05-04.
I am maintaining the codebase since 2012-09-18.
I did not *think* the codebase from the bottom up, but am more
like "El Condor pasa" on top of 37205 lines and 134375 words of
C code as of now but that grew over more than three decades.
Worse, i am a somewhat insular user.
Worst, there are no tests.

I want to move this shit forward.
Please do report any bugs or misbehaviours you know of.
Thank you!

--steffen

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb
_________

heirloom mailx or s-nail?

2013-03-01T20:48:59

What features does s-nail have that Heirloom mailx
does not ?




------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb

patch that makes set work for me

Stephen Isard — 2013-03-01T20:40:02

See Subject :-)

Stephen Isarddiff --git a/Makefile b/Makefile
index c316b39..df22ca3 100644
--- a/Makefile
+++ b/Makefile
< at >< at > -24,7 +24,7 < at >< at > INSTALL= /usr/bin/install
 # aliasing etc., did not yet exist.  Until the codebase has been overhauled
 # (and the overhauling has itself been debugged :) you may read INSTALL for
 # hints but are otherwise on your own if you turn on fancy compiler features!
-CFLAGS= -O1
+CFLAGS= -g
 #CFLAGS= -std=c89 -O2 -g -fstrict-aliasing
 #WARN= -Wall -Wextra -pedantic -Wbad-function-cast -Wcast-align \
 #-Winit-self -Wwrite-strings -Wcast-qual -Wunused
diff --git a/cmd3.c b/cmd3.c
index cafbe1b..f289a66 100644
--- a/cmd3.c
+++ b/cmd3.c
< at >< at > -124,7 +124,7 < at >< at > _set_show_all(void)
 for (i = u.j = 0; i < HSHSIZE; ++i)
 for (vp = variables[i]; vp != NULL; vp = vp->v_link)
 ++u.j;
-vacp = (char**)salloc(u.j * sizeof(*vacp));
+vacp = (char**)salloc((u.j + 1) * sizeof(*vacp));
 for (i = 0, p = vacp; i < HSHSIZE; ++i)
 for (vp = variables[i]; vp

backtrace

Stephen Isard — 2013-03-01T16:47:45

Giving the set command while running s-nail in gdb:

Program received signal SIGSEGV, Segmentation fault.
0x0000000000455f8b in canonify (
     vn=0x682d7265646c6f66 <Address 0x682d7265646c6f66 out of bounds>)
     at vars.c:91
91if (upperchar(*vn))
(gdb) bt
#0  0x0000000000455f8b in canonify (
     vn=0x682d7265646c6f66 <Address 0x682d7265646c6f66 out of bounds>)
     at vars.c:91
#1  0x000000000045629c in value (
     name=0x682d7265646c6f66 <Address 0x682d7265646c6f66 out of bounds>)
     at vars.c:193
#2  0x000000000040b0d5 in _set_show_all () at cmd3.c:138
#3  0x000000000040c4dd in set (v=0x7fff8567b130) at cmd3.c:700
#4  0x000000000042ea62 in execute (linebuf=0x1295030 "set", contxt=0,
     linesize=3) at lex.c:659
#5  0x000000000042e41c in commands () at lex.c:452
#6  0x00000000004379ff in main (argc=3, argv=0x7fff8567d458) at 
main.c:616

Hope that helps.

Regards,

Stephen Isard

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.

segfault on set with no arguments

Stephen Isard — 2013-03-01T01:06:45

I've just pulled the "next" branch from git and 
datefield-markout-older works for me, but I still get a segfault if I 
call set without an argument.  I'm calling s-nail with -f and a local 
mbox folder.  I'm running Scientific Linux 6.3 (clone of RHEL 6.3, like 
Centos 6.3).

Regards,

Stephen Isard

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb

Fwd: Re: patch to make date in s-nail behave like ls -l

Steffen "Daode" Nurpmeso — 2013-02-28T13:23:54

Doesn't seem to make it through to the list ?

-------- Original Message --------
Date: Thu, 28 Feb 2013 12:53:01 +0100
From: Steffen "Daode" Nurpmeso <sdaoden-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
To: "Stephen Isard" <3s9xh9m02-O/bDAPVd7B0N+BqQ9rBEUg< at >public.gmane.org>
Cc: nail-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
Subject: Re: [nail-devel] patch to make date in s-nail behave like ls -l

Hello,

"Stephen Isard" <3s9xh9m02-O/bDAPVd7B0N+BqQ9rBEUg< at >public.gmane.org> wrote:
 |Hello,
 |
 |I sometimes use mailx to view archives of old mail messages and in that 
 |situation I would prefer the date to be displayed the way unix "ls -l"
 |does it, giving the year instead of the time of day for messages more 
 |than six months old.  I have attached a patch that has this effect if 
 |the variable date_like_ls is set.  I have only been able to test it on 
 |my own message archives, so there may be cases I haven't considered.

I really like the idea, but i think we should then follow the
POSIX standard

14.0 suffers from CRAM-MD5 regressions, fixed on next

Steffen "Daode" Nurpmeso — 2013-02-28T12:16:05

Hi list,
and yet a third bug made it into 14.0, which should have prevented
any CRAM-MD5 based IMAP/SMTP login/usage successfully.

The bug(s) is/are fixed on the *next* branch and will make it into
14.1, which i hope to finish this weekend.

Background:
-----------

It is actually a misunderstanding and one newly introduced bug.

The misunderstanding was that the newly implemented Base64 layer
used whitechar() to trim away whitespace surrounding a Base64
encoded string, matching \n (LF, U+000A) and ' ' (SP, U+0020)
instead of spacechar() that also matches \r (CR, U+000D); the
current codebase however directly passes data from the wire,
including the line terminator CRLF ...
(This is a really strange codebase that even *appends* newlines
when there are none, whereas normal people try to remove them as
fast as possible, chomping them off, maximally tracking wether
there was one, and which one.)

The bug was that the original codebase used strtob64() in
cram_md5_string(), working on a \0 (NUL, U+0000) terminat

patch to make date in s-nail behave like ls -l

Stephen Isard — 2013-02-28T01:45:26

Hello,

I sometimes use mailx to view archives of old mail messages and in that 
situation I would prefer the date to be displayed the way unix "ls -l"
does it, giving the year instead of the time of day for messages more 
than six months old.  I have attached a patch that has this effect if 
the variable date_like_ls is set.  I have only been able to test it on 
my own message archives, so there may be cases I haven't considered.

Btw, the s-nail I pulled from git about ten days ago segfaults when I 
give the "set" command with no arguments.

Best regards,

Stephen Isard
diff --git a/cmd1.c b/cmd1.c
index 7d1d7d6..d6e6abd 100644
--- a/cmd1.c
+++ b/cmd1.c
< at >< at > -412,6 +412,7 < at >< at > hprf(const char *fmt, int mesg, FILE *f, int threaded, const char *attrlist)
 char const *name, *date;
 int B, c, i, n, s, fromlen,
 subjlen = scrnwidth, isto = 0, isaddr = 0;
+time_t now;
 
 date = NULL;
 if (value("datefield")) {
< at >< at > -551,7 +552,17 < at >< at > hprf(const char *fmt, int mesg, FILE *f, int threaded, const char *

S-nail 14.0 fixes; and expect 14.1 in February

Steffen "Daode" Nurpmeso — 2013-02-16T19:22:07

Hello list,

two bugs made it into S-nail 14.0, one of which may cause an
infinite loop.  I've hoped to make it without such ugly fixes, but
unfortunately ... not.

I will release a 14.1, maybe next week, but surely in February.
For those of you who have a git(1) clone (i'm happy -- S-nail(1)
really *has* users!  Thank you!!  And, while at it: thanks to the
funny guy who downloaded the release tarball 13 times -- i hope it
was a shell loop :-)) an update should possibly be considered; you
may howevery also patch in the diffs below by calling

  $ diff -Napru -p1 < THIS-MAIL-MESSAGE

For 14.1 there will also be two new features at least, a new option
*mimetypes-load-control* and also alias expansion for Reply-To:
list entries, as has been suggested and basically patched in by
Martin Neitzel.

The diffs, and note that it is the first which forces 14.1; the
second is of course also a bug, but i didn't see it doing any harm
in thousands of mails, including weird test mails.  Of course it
can trigger, though, and

Gunnar Ritter

Pierre-Jean — 2012-11-12T10:22:57

Dear list,

Does anyone has some news of Gunnar Ritter?
Does someone have some other contact with him than its public email?

It seems that there's a long long time that he didn't reply anymore,
nor on this list, nor on others, nor on its public email.

I hope that he found some better occupation than heirloom
software, but I fear that this silence has some other
reasons.

Should'nt we try to know what's going on?

Cheers,

Pierre-Jean.

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_nov

Announcing S-nail v13.3

Steffen "Daode" Nurpmeso — 2012-11-11T00:35:00

Hmhm, hi list,

deepest apologies -- this release will finally work even for users
which use a MTA gateway and have *record* set!

That fix is the main reason why i spent some more time and did yet
another release, S-nail(1) v13.3; but please read on.
The release tarball for S-nail(1) v13.3 can be downloaded from

  http://sourceforge.net/projects/s-nail/files/s-nailv13.3.tar.gz/download
  http://sourceforge.net/projects/s-nail/files/s-nail.tar.gz/download

The latter is a symlink to the latest release.  The checksum of
the release tarball is:

  MD5(s-nailv13.3.tar.gz)= 5f0a82b487a62c76be46bcb63cfaaa45
  SHA1(s-nailv13.3.tar.gz)= 866375b87659d06507c3bddb6176d02735afc767
  SHA256(s-nailv13.3.tar.gz)= \
    f13f981a55ff61c00cd8987a112fee39d69072c2dfd38dbff1c6941456eb3b80

The full changelog of the commits in between s-nailv13.2 and
s-nailv13.3 can be inspected by issuing a

  $ git log --reverse --topo-order s-nailv13.2..s-nailv13.3

So thanks for using S-nail(1), the most feature rich but only
second best Be