[M-Dev] Re: GPG (and S/MIME)

Like I said -- if you just put the basic wxExec functionality in
there, GPG support will be easy, and you can leave it to others like
myself.  The two solutions (repeat, repeat) are not mutually
exclusive.

I'm not trying to say that S/MIME == PGP and don't follow how you
extracted that from my comments.  What I am saying is that, however
interesting a development project it may be, it is not anywhere near
the market penetration (hence importance) of plain old PGP.  You have
not persuaded me that S/MIME is anywhere near as important in an email
client.

Moreover if I want to send a message to someone on MailVault or
HushMail, I need to create (generate) OpenPGP messages, not just read
them.

M.

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Gadgets, caffeine, t-shirts, fun stuff.
http://thinkgeek.com/sf

Re: [M-Dev] install_strip target question

On Thu, 11 Jul 2002 20:50:47 +0200 (CEST) Thomas Finneid <tfinneid <at> ifi.uio.no> wrote:

TF> I found out that the solution of modifying
TF> configure.in and makeopts.in is not needed. The reason is that the binary
TF> is compiled with or without debug info anyway. So there is only a need to
TF> modify the existing install_bin target, which has the -s switch hardcoded, 
TF> and add a new target called, as we know by now, install_strip.

 Ok, fine.

TF> My solution is removing the -s switch in the install command and replace it
TF> with a variable which is set if the target install_strip is run (because
TF> in the other cases, the debug info inclusion is already decided by the
TF> --eanble-debug option to configure).

 I.e. something like:

install:
        $(INSTALL) $(INSTALL_STRIP) ...

install_strip:
        $(MAKE) INSTALL_STRIP="-s" install

?

TF> Any suggestions? If not I'll do it by sub make arguments.

 I don't see any other way but I'm far from being a make expert...

 Regards,

Re[2]: [M-Dev] Frankenstein messages

On Fri, 12 Jul 2002 08:52:32 +0200 Xavier Nodet <xavier.nodet <at> free.fr> wrote:

XN> Of course. Unfortunately, I did not even succeed in launching M under
XN> Purify yet. Purify reports a few harmless (for this case) errors, but
XN> then seems to fail to 'activate' M.

 This is bad news -- it used to work in the past. I must have used Purify
on M a few months ago for the last time but it always worked flawlessly.
Just a hint: have you used /FIXED:NO option to the linker? It is required
by Purify.

 Thanks,
VZ

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Gadgets, caffeine, t-shirts, fun stuff.
http://thinkgeek.com/sf

[M-Dev] diff command with M?

Hi

What is the desired diff command(with arguments) I should use with M?

--

-- 
Thomas Finneid

email: tfinneid <at> ifi.uio.no

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: diff command with M?

On Sat, 13 Jul 2002 12:58:07 +0200 (CEST) Thomas Finneid <tfinneid <at> ifi.uio.no> wrote:

TF> What is the desired diff command(with arguments) I should use with M?

 Use diff -u2 please.

 Thanks!
VZ

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re[3]: Forwarded message: [rwelty <at> averillpark.net: email problems]

[replying to the list to let the others know about this]

On Fri, 12 Jul 2002 21:38:39 -0400 (EDT) Richard Welty <rwelty <at> averillpark.net> wrote:

RW> ok, here's what i'm seeing:
RW> 
RW> context -- redhat linux 7.3, exim Mail Transport Agent installed locally
RW>    on my laptop
RW> 
RW> if i use SMTP server of 127.0.0.1, then there is no ^M problem
RW> 
RW> if i use a local command of /usr/local/sbin/exim -t, there is a ^M problem.
RW> the MTA expects properly formed ^M^J pairs in an SMTP transaction, but in
RW> a local un*x delivery, it wants to just get ^J characters in the stream
RW> like any other un*x program.

 Ok, this indeed explains it -- thanks a lot for taking time to investigate
it. We always use "\r\n" for the outgoing message, regardless of where does
it go (SMTP server or local MTA).

RW> i think sendmail has a fixup mode, and i can get exim to do ^M^J fixup with
RW> a config flag, but i'd still argue that this is really an MUA bug, not an
RW> MTA bug.

 Yes, sure. I'll fix soon as it should be quite easy.

 However I'm afraid this doesn't explain the problems reported by Alan G
Isaac as he is under Windows and so probably doesn't have any local MTA
(and if he did, it would be using "\r\n" line termination anyhow). So there
is some other problem as well -- but it's already good to know about this

Re[4]: Forwarded message: [rwelty <at> averillpark.net: email problems]

On Sat, 13 Jul 2002 14:03:42 +0200 (Romance Daylight Time) Vadim Zeitlin
<zeitlin <at> dptmaths.ens-cachan.fr> wrote:

> [replying to the list to let the others know about this]

> On Fri, 12 Jul 2002 21:38:39 -0400 (EDT) Richard Welty
> <rwelty <at> averillpark.net> wrote:
> RW> i think sendmail has a fixup mode, and i can get exim to do ^M^J
> fixup with
> RW> a config flag, but i'd still argue that this is really an MUA bug,
> not an
> RW> MTA bug.

FYI the exim command line option is -dropcr, so a local MTA command of

/usr/local/sbin/exim -t -dropcr

(or whatever path to exim you're using) will clean up the extra cr
characters. i have no idea what you'd do with sendmail, but it likely has
something.

richard
--
Richard Welty                                         rwelty <at> averillpark.net
Averill Park Networking                                         518-573-7592
              Unix, Linux, IP Network Engineering, Security

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.

install_strip patch

Hi

I managed to find a way to do it by environment variable passing to the sub
makes instead of by argument passing, the problem with the argument passing
was that make proceeded with failures all the time or behaved strangely.
I will write a small tech note on make and configure etc tomorrow.

--

-- 
Thomas Finneid

email: tfinneid <at> ifi.uio.no

M going into a loop

Hi

I got M going into a loop while downloading (or filtering) messages.
I had to use mozilla to empty the mailbox and afterwards it was ok.
This may be related to the frankenstein messages problem.

Here is the gdb trace, I hope it helps:

#0  0x40a149ae in pthread_setcanceltype (type=0, oldtype=0x0) at internals.h:384
#1  0x40a1b995 in read (fd=12, buf=0x40bf1000, count=4096) at wrapsyscall.c:153
#2  0x407a31a6 in _IO_file_read () from /lib/libc.so.6
#3  0x407a232c in _IO_file_underflow () from /lib/libc.so.6
#4  0x407a4bf9 in _IO_default_uflow () from /lib/libc.so.6
#5  0x407a3c62 in __uflow () from /lib/libc.so.6
#6  0x4079f1f0 in getc () from /lib/libc.so.6
#7  0x082de8b7 in file_string_next (s=0xbfffe690) at flstring.c:68
#8  0x082aba2a in rfc822_parse_content (body=0x8573108, bs=0xbfffe690, h=0x839c749
".MISSING-HOST-NAME.", depth=0, 
    flags=1572) at rfc822.c:534
#9  0x082ab89d in rfc822_parse_msg_full (en=0x857294c, bdy=0x8572950, s=0x8575b22 "", i=0,
bs=0xbfffe690, 
    host=0x839c749 ".MISSING-HOST-NAME.", depth=0, flags=1572) at rfc822.c:496
#10 0x0829196e in mail_fetch_structure (stream=0x8557c58, msgno=2, body=0x857f918, flags=1) at mail.c:1212
#11 0x081f0c3a in MessageCC::GetBody ()
#12 0x081f0ec7 in MessageCC::WriteToString ()
#13 0x081d5466 in MailFolderCC::AppendMessage ()
#14 0x081e3762 in MailFolderCmn::SaveMessages ()
#15 0x081d5e70 in MailFolderCC::SaveMessages ()
#16 0x081e3bc5 in MailFolderCmn::SaveMessages ()
#17 0x081e682e in MailFolderCmn::CollectNewMail ()

Re: Frankenstein messages

On Sat, 13 Jul 2002 01:16:29 +0200 (Romance Daylight Time) Vadim Zeitlin
<zeitlin <at> dptmaths.ens-cachan.fr> wrote:

VZ> Just a hint: have you used /FIXED:NO option to the linker? It is required
VZ> by Purify.

I will check... Thanks.

--

-- 
Xavier Nodet
PGP fingerprint: EFE0 0E75 C4DC 2654 5C36  4E2D 107C 19DE 0646 1FE2

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf