markus schnalke [Fri, 3 Feb 2012 09:42:49 +0000 (10:42 +0100)]
Fixed abort of e.g. comp when draft folder is empty.
In this case mp->lowoff is 1 but mp->lowmsg equal mp->hghmsg equal 0.
Hence we have an access at mp->msgstats[-1], which corrupted our mem.
Gdb hadn't helped me for this, but valgrind showed me the problem.
markus schnalke [Mon, 30 Jan 2012 15:16:26 +0000 (16:16 +0100)]
Scan listings shall not contain body content. Hence, removed this feature.
Scan listings shall operator on message headers and non-message information
only. Displaying the beginning of the body complicates everything too much.
That's no surprise, because it's something completely different. If you
want to examine the body, then use show(1)/mhshow(1).
Changed the default scan formats accordingly.
markus schnalke [Mon, 30 Jan 2012 11:42:06 +0000 (12:42 +0100)]
Removed -pause from mhshow(1).
My goal is to have show/mhshow generate a single stream of bytes to be
piped through a pager. Content not convertable to text shall not be displayed
directly. This is my idea currently.
markus schnalke [Mon, 30 Jan 2012 11:00:49 +0000 (12:00 +0100)]
Bugfix: Removed -reverse switch from scan(1).
No more to say but to quote MTR:
-[no]reverse under #ifdef BERK (I really HATE this)
And JLR:
Lists messages in reverse order with the `-reverse' switch.
This should be considered a bug.
markus schnalke [Mon, 30 Jan 2012 10:33:59 +0000 (11:33 +0100)]
Alias: Removed the address group `*', which meant everyone on the system.
Therefore could remove `everyone' and `noshell' from mts.conf.
If one really needs to write mail to everyone on the system (not just
all regular users, which probably are in a group `user'), then he
shurely can invest a bit more effort to do so. Likely, the simple condition
to uids > 200 and not the shell `noshell' would anyways be to limited.
Better use awk or the like and do it by hand.
markus schnalke [Mon, 30 Jan 2012 10:03:29 +0000 (11:03 +0100)]
Rest of the MMDF-style mail drop support removal.
Only Unix-style (mbox) mail drops are supported now. Their location
is in `mailspool', which is reported in the configure summary now.
If users like to have different locations, they can use the environment
variable MAILDROP or the profile entry MailDrop to overrule the default.
From mts.conf, mmdfldir and mmdflfil were removed.
msgchk(1) received several cleanups.
I touched sbr/m_getfld.c ... Wooohh ...
markus schnalke [Sat, 28 Jan 2012 15:05:54 +0000 (16:05 +0100)]
forw/repl/dist annotations won't add address lists anymore.
spost does not support it and I wonder if we need it.
Remove the undocumented -idanno switch, actually a no-op, from spost.
markus schnalke [Sat, 28 Jan 2012 12:35:42 +0000 (13:35 +0100)]
Removed the ATTVIBUG compile option.
If you do have the bug, then workaround it like described in the FAQ:
Subject: 06.01 What to do with "Problems with edit - draft removed".
From: John Romine <jromine at ics.uci.edu>
Date: Mon, 1 May 1995 00:00:00 -0800
If your users are using an AT&T version of "vi", it's exiting with
non-zero status (supposedly a count of the "errors" during the
edit). Move "vi" to "broken_vi" and put it its place :
#! /bin/sh
/usr/ucb/broken_vi "$@"
exit 0
Alternatively, compile MH with the ATTVIBUG option.
Then complain to your vendor that "vi" is broken, and they shouldfix it.
markus schnalke [Sat, 28 Jan 2012 11:26:36 +0000 (12:26 +0100)]
Removed options to deactivate dashstuffing. Always dashstuff.
Maybe there are few RFC 934 messages around, but dashstuffing is IMO
more useful than bad.
markus schnalke [Wed, 25 Jan 2012 21:56:43 +0000 (22:56 +0100)]
Removed the UCI code to read the signatur (= name) from ~/.signature.
If we want to read something from ~/.signature, then those lines that
form the footer of a mail message. What is called ``signature'' here
is the name of the sender; the name that is put into the From: line.
There are already several options to put it:
- environment variable: SIGNATURE (to change temporary)
- profile entry: Signature: (to set only for the mail system)
- GECOS field in passwd (fallback and global)
These should be enough.
markus schnalke [Wed, 25 Jan 2012 21:36:49 +0000 (22:36 +0100)]
Removed the UCI alternative name of anno(1): `fanno'.
fanno didn't add the date. If you want this, run `anno -nodate'. Hey, you
can create a shell alias for it. What about that? ;-)
markus schnalke [Wed, 25 Jan 2012 21:29:34 +0000 (22:29 +0100)]
Removed the -header option from scan(1).
The heading had been very simple since the advent of format strings anyway.
If someone wants a heading, he can run folder(1) before.
markus schnalke [Wed, 25 Jan 2012 21:21:56 +0000 (22:21 +0100)]
Removed (allmost all) JLR-specific code.
There still is one fragment in fmt_scan(), of which I don't understand
the logic. I don't want to remove stuff I don't understand.
markus schnalke [Wed, 25 Jan 2012 12:31:36 +0000 (13:31 +0100)]
Finally fixed the annoying `Invalid cross-device link' problem.
Due to a missing slash (/), temporary files were put into the parent
directory instead of the same one. See:
mhbuild: unable to rename output /home/meillo/Mail/send0zxfDu to
/home/meillo/MailmhbuildzQjMUm: Invalid cross-device link
markus schnalke [Tue, 24 Jan 2012 23:01:51 +0000 (00:01 +0100)]
Fixed -mime for forw and repl.
If the value of an attachment header starts with `+', then it is treated
as the specification of a message in MH style: ``+folder msg''
markus schnalke [Tue, 24 Jan 2012 19:08:35 +0000 (20:08 +0100)]
Activated Jon's attachment system by default and steamlined it.
I had done similar work already some time ago. See:
http://www.mail-archive.com/nmh-workers@nongnu.org/msg02112.html
This time it's the second go.
I changed Jon's attachment system to have a default attachment header
field name, or get it from the profile. I removed the command line options.
The new profile name is: `Attachment-Header'. Stored in attach_fmt.
I removed the the choice between several attachment formats.
A message gets mimeified if it has attachment headers or (and this is new)
contains any non-ASCII character. The body won't be scanned for lines
matching /^#/. The MIME structure is created automatically.
The content-typing of unknown attachments is still poor and needs to
be improved.
You shouldn't use the `mime' command at the whatnow prompt anymore.
markus schnalke [Thu, 5 Jan 2012 18:18:12 +0000 (19:18 +0100)]
Explicit checks for return values of strcmp().
The return value of strcmp() is unintuitive for string equality tests.
I prefer explicit comparisions against 0. I changed this everywhere.
markus schnalke [Thu, 5 Jan 2012 11:26:09 +0000 (12:26 +0100)]
Removed isprefix() and replaced it with strncmp().
isprefix() had been ssequal(), which had not been what it advertised to be.
Now, all this is gone.
We could get problems if one of the arguments is NULL, as isprefix()
handled this situation, but strncmp() does not. It appears as if this case
does not show up in the code.
markus schnalke [Thu, 5 Jan 2012 10:52:03 +0000 (11:52 +0100)]
Removed the non-LOCALE character code and the #ifdefs and simplified.
POSIX says: tolower() won't change its argument if it is not isupper().
And isupper() doesn't need an additional check for isalpha().
markus schnalke [Thu, 5 Jan 2012 10:11:13 +0000 (11:11 +0100)]
Much more straight-forward handling of the beyond sequence.
Could remove the message flag SELECT_EMPTY in the same go.
This frees one more space for a user-defined sequence.
markus schnalke [Wed, 4 Jan 2012 17:29:23 +0000 (18:29 +0100)]
Renamed all standard sequences (e.g. cur->c) and made them globally changeable
The full list:
cur -> c
first -> f
last -> l
next -> n
prev -> p
all -> a
new -> b (mnemonic: one beyond the last)
They are globally changeable in config/config.c
Could be that I'll add the old names again, as convenience aliases.
markus schnalke [Wed, 4 Jan 2012 10:16:12 +0000 (11:16 +0100)]
Fixed bug in brkstring(), newly introduced by the switch to strchr().
strchr() treats the trailing null byte surprisingly:
``The terminating null byte is considered to be part of the string.''
markus schnalke [Wed, 4 Jan 2012 09:06:51 +0000 (10:06 +0100)]
Globally defined the names for the standard sequences.
I'll rework the whole code to use them and not have hard-coded names anymore.
The `new' sequence will be called `beyond' sequence (`b') from now on.
The `unseen' sequence will be set by default to `u' from now on.
The sequence negation will be set by default to `!' from now on.
markus schnalke [Mon, 2 Jan 2012 09:57:24 +0000 (10:57 +0100)]
Renames: s/+outbox/+sent/ and s/cc:/Cc:/
IMO ``sent'' specifies more exactly what it contains. ``outbox'' would
mean to me that these messages are prepared for delivery, but are not
already sent.
markus schnalke [Fri, 23 Dec 2011 11:09:55 +0000 (12:09 +0100)]
Reworked the man page system and some man page contents (mmh-intro).
Now it's possible to have two equally named pages in two different sections.
This appeared to be interesting for mmh(X), but I don't use it now. Let's
see if we keep the new system or even rework it to become simpler.
markus schnalke [Fri, 9 Dec 2011 15:34:57 +0000 (16:34 +0100)]
Completely reworked the path convertion functions
Moved everything (from sbr/getfolder.c and sbr/m_maildir.c) into
sbr/path.c, but actually replaced the code almost completely.
See h/prototypes.h for the function changes.
sbr/path.c provides explaining comments on the functions.
None of them allocates memory automatically.
Additionally:
- Like for other ``files'', `inc -audit file' places file relative
to the cwd, not relative to the mh-dir. This is for consistency.
- Replaced add(foo, NULL) with getcpy(foo), which ist clearer.
markus schnalke [Sun, 20 Nov 2011 11:38:17 +0000 (12:38 +0100)]
The existence of the profile entry Path is already enforced by context_read().
Hence, removed the obscure ``free(path("./", TFOLDER));'' calls.
Still I don't see the sense in these lines, but they won't get executed,
as the condition is checked before (and the program aborts then).
Probably that's obsolete stuff that remained through several reworks.
markus schnalke [Sat, 5 Nov 2011 16:56:07 +0000 (17:56 +0100)]
Introduced FDEF and FCUR for speaking arguments to getfolder().
Maybe it would be better to two separate functions, but I cannot
imagine good names for them ... :-(
markus schnalke [Sat, 5 Nov 2011 13:04:02 +0000 (14:04 +0100)]
Renamed r1bindex() to mhbasename(), to make its function becomes clear.
All calls to r1bindex() had '/' as second argument. For everything else
one can use strrchr(3) instead of r1bindex().