Before commit 5da26eca, you could pass Msgbase() the path to an SMB
on the disk and open() it, no configuration needed. As of 2 years
ago, I broke that, and passing a path to an SMB would open the "mail" base
instead - most unexpected. This is a feature of smb_open_sub() which we
switched to using (from smb_open()), so go back to using smb_open() when
an unrecognized code is pass to the constructor.

This has the negative consequence that the "mail" base can't be created
via JS. Probably should fix that.

The DoorFrame door library will about with "Overflow Error" if it reads a door.sys file with lines 16 (calls), 42 (minute credits), or 52 (posted messages) with a value > 32767, the maximum positive value of a signed 16-bit integer. So cap these values in the door.sys file at that maximum. This does potentially break/limit doors that give minutes using the door.sys drop file, so use that "modify user data" option in SCFG with care.

Electrosys initially reported this problem with the Lemonade! door game which was barfing on a line 42 value of 449632.

The extra strength bug fix needs a new version number.

You get str_gained, not str. :(

This is what was causing the 'is_utf8' message header property to be created as a Number instead of a Boolean.

This prevents re-entry to the door configurations from returning to the previously highlighted menu option. Not sure why this was, but clearly re-zeroing static variables makes no sense in the first place.

If a sysop wants to prevent new users from using an email address of an existing user, they can set SCFG->System->New User Values->Question Toggles->Force Unique E-mail/NetMail to "Yes".

This partially addresses issue #127, at least for the terminal server.

On *nix, log() output can go to syslog() and on Windows, it can go to the
event log.

This addresses issue #360.

Fixes issue #395

CID 351997 and 351998

ftello() returns 0 after fnopen(..., O_APPEND). Use filelength() instead.

For the following log files:
  hungup.log
  error.log
  crash.log
  hack.log
  spam.log
  guru.log

... they can be limited in size, in bytes, by the sysop, along with a maximum retention limit (number of *.#.log files). Configured in SCFG->System->Advanced.

By default, the limit and keep numbers are 0, so "unlimited" (same behavior as before). Once sufficiently tested, I'll change the default in a stock main.cnf file to use a sensible limit (e.g. 10 log files of 10MB each, something like that).

DDFileLister: Dates shown in file desriptions when extended descriptions are enabled.

See merge request !169

When extended file descriptions are enabled, the file date is now shown
with the file description on the last line.

e.g. "Grunged message (type 2) from 4:80/1 at offset 58 in packet: (null)"

off_t is 64-bit, but long in 32-bit (on Win32), so needs a typecast here in
lprintf() call (otherwise, the upper 4 bytes of the offset are interpreted
as the string address for the corresponding '%s' argument, the filename).

JS doors with the "Use Shell or New Context" option enabled in SCFG and JS
modules installed a global hot key handlers would not automatically terminate
when the user disconnected (and js.auto_terminate was true, the default).
I'm not sure why the operation callback was only installed when scope==NULL
but always installing it fixes the issue with some global hot key modules
and JS doors becoming "zombies" when a user disconnects while running them.

A user disconnecting while running tw2 would just result in an infinite loop
checking node status and waiting for the user's allowed time to run out (which
may never happen) or for the server to be locally terminated.

This has been happening for years on Vertrauen. Finally now I see why.

Fixes issue #390

Especially for Guest (G-Exempt) accounts which support multiple
concurrent logons - the 'curxtrn' value stored in the user's record
may not match the external program that was actually last executed
on the node in question, so use the .aux property value (from
node.dab) as is done in the C++ code (e.g. printnodedat()). This insures
that the node_status() output here matches the C/C++ code output
(e.g. when logging on the terminal server).

Still support the 'code-based' external program name look-up since
that usage is relied upon by other modules, but when a number is
passed, use the number to find the right external progarm name
(and we need to subtract one, since aux is a 1-based number).

I've looked at this problem a couple of times before and not sure why
I didn't see the solution. <shrug>

Issue #389 reported by Keyop:
srvc 0324 IMAPS <Keyop> !JavaScript  /sbbs/exec/imapservice.js line 653:
TypeError: args[1].toUpperCase is not a function

I'm not sure how you reproduce this error or what type args[1] is
(apparently, not a string), but this fix it regardless. Please let me
know.

Broken in commit ce01d8a7.
F*ck the English language.

Fixes isue #388 - no change in behavior

Since this script can be reloaded in the same scope.

Fixes issue #387.

This converts the body text as well as the to, from, and subject header
fields (first to CP437, then to ASCII).

This should address the issue raised in issue #386 with exporting UTF-8
sequences to some message networks.

Also changed output to reflect slog.c more closely. I suppose only one of
of these 2 "utilities" to read csts.tab should exist at some point. :-)

Also added suport for "-r" and "-t" to be more modern and *nixy.

To match behavior of changed xpdev->byte_estimated_to_str().

If the caller wants a 'B' suffix, they can add it.

Configured in sbbs.ini [web] by setting:

 FileVpathPrefix = /files/
 FileIndexScript = webfileindex.ssjs

And optionally (if you want your vhosts to get filebase access):

 FileVPathForVHosts = true

You can use an .xjs file if you prefer.

When loadfiles() calls sortfiles(), only the file's index records have been
read in, so trying to sort on any header field won't work.

This bug wasn't observable when sorting by date ascending, since that's the
natural index order of the files already (order imported/added), only
observed when sorting by date descending (newest at the top).

to accommodate larger numbers/strings.

... from ERROR to WARNING, but log the human-readable logon time too.

This condition can easily happen if the system clock is corrected (e.g. a few seconds via NTP adjustment) during a quick connect/disconnect.

Don't call logoffstats() unless a user actually successfully logged-in. I don't think sysops actually want all the time bots sit at the logon prompt counted as "time online" in their stats anyway.