When multiple servers are recycling at the same time, (e.g. due to saved
change in SCFG) they'd each call sbbs_read_ini() with a shared global_startup
struct, which in turn calls sbbs_free_ini(), which would free all the
allocated network interface lists (including the global_startup one) using
iniFreeStringList (just a wrapper for strListFree), but iniFreeStringList()
does NOT modify (NULLify) the freed-pointer, so your second or third server
that called sbbs_read_ini(), with the shared MainForm->global structure, would
*again* free the same global interface list. This bug actually has always
existed because get_ini_globals() freed the global interface list in the same
way, except it *immediately* re-allocated a new one by calling
iniGetStringList(), so the time window (opportunity) for this race condition
to occur was much smaller. Truly, SBBSCTRL should use a mutex or other
mechanism to protect the shared global_startup struct, but this is a first
step to a full fix: sbbs_free_ini() should (and now does) nullify the freed
network interface pointers by using strListFree() directly. I haven't been
able to reproduce the crash upon recycle in SBBSCTRL after making this change.

MRC client updates for Synchronet

See merge request !276

warning: format not a string literal and no format arguments

Weird this warning is happening for me with GCC 12.2 (debug or release build)

Apple Mail apparently uses tabs to indent the charset of nested MIME parts:

e.g.
--Apple-Mail-143B9F0C-6BB4-4C8E-869B-6DE05D6B58CE
Content-Type: text/plain;
<tab>charset=utf-8
Content-Transfer-Encoding: quoted-printable

This fixes issue #483 reported by Nelgin

Error: can't covert options.backup_level to an integer

Default to 10, if not set.

Looks like a typo, in Scan_For_Banned_Clients()

Fix issue #536

Ugh. This is ugly.

smtp_netmail_addr() - not currently used anywhere else, but may be someday.
load/mailutil.js's fidoaddr_to_emailaddr() has this same logic (for FTN addrs
at least and is now used by nntpservice.js).

Also, always pass a buffer to smb_faddrtoa() from the mail server since it's
multi-threaded and its unsafe to pass NULL (using a static local buffer).

smb_faddrtoa() uses a static local buffer if passed NULL and we shouldn't be
doing that from multi-threaded callers.

Use the terminal server sem file check interval instead of node_sem_check
(it was only used in the event thread) and node_stat_check was only used in
the control panel, so just support a registry over-ride, but default to 5 sec.

These settings were artifacts from SBBS v2 (WFC mode), and not really relevant
or in the proper configuration place.

Both the version and the output mode SyncTERM must be confirmed.

A similar change should be made for supports_palettes() I think.

This would cause errors reading node.dab when running xsdk doors built for 64-bit *nix targets
(e.g. SBJ, TBD, dpoker, etc.). I can't believe how long this has been broken. Wow.

Include some constification of xsdk functions

Added some comments (observed concerns) in TBD about the object[] definition
not matching (struct/type and contents) between objects.h and tbd2.c.

Returns an array of attributes (digits), if valid response

Now Nightfox, you can query SyncTERM (only) to see if it supports Pixel ops
(e.g. Sixel) is supported at the moment, using:

  cterm_lib.query_ctda(cterm_lib.cterm_device_attributes.pixelops_supported)

- will return true if pixel operations are supported

Other dynamic terminal capabilities (e.g. depdendant on the output mode that
SyncTERM is run in) can be detected with this method. But note: the values
returned in SyncTERM's CTDA query-response do not exaclty match the values
documented in cterm.txt. So use the constants from cterm_lib.js instead.

CID 451182

This is an easily overlooked setting, let's make it easier

Modem result codes. Oh, the memories...

Anacron sends mail From: Anacron <root>, so deal (convert to root@youraddr).

Also, gracefully handle a forgotten recipient list.

warning: ‘%s’ directive writing up to 16384 bytes into a region of size
between 1 and 16385 [-Wformat-overflow=]

warning: argument 1 value ‘18446744073709551608’ exceeds maximum
object size 9223372036854775807 [-Walloc-size-larger-than=]

... rather than just return false.

This will make debugging this type of issue much easier in the future.
sendmail.js was allowing empty recipient_list array arguments and the MsgBase
.status was 0 and .error just an emtpy string - unhelpful.

warning: ‘checkStamp’ accessing 20 bytes in a region of size 11
etc.

Direct users to Program Settings->Startup Screen Mode for initial/default
screen mode.

Wording update for Startup Screen Mode help.

Indent the File Locations less as these paths are usually much longer on
Windows and is ugly when wrapped here.

warning: ‘%s’ directive writing up to 128 bytes into a region of size 74
note: ‘sprintf’ output between 21 and 179 bytes into a destination of size 80

warning: format ‘%s’ expects argument of type ‘char *’, but argument 4 has type ‘void *’
warning: ‘sprintf’ writing a terminating nul past the end of the destination

(like we do for Message Groups and File Libraries already)