looks better, more consistent with the log messages before/after it.

We removed the large-scope 'i', but didn't update the Unix-only blocks of code
(dealing with UNIX domain spy sockets).

This is just a code quality (readability) improvement. No functional change.

The other server stats (e.g. error_count, served) are zeroed upon startup, so
the highwater mark should be zeroed too.

Like the other servers.

A sysop likely would want to know which services/protocols had what
utilization (as a portion of the total highwater mark), so that'll have to
be added later.

We don't log highwater valeus of 1, though those are published to MQTT.

Not dealing with 'services' server yet, cause that's kind of tricky with the
sysop likely wanting to know per-service stats (served counter, highwater
mark, etc.)

Only log the hightwater mark when it's > 1. :-)

Fix CID 497098

If a file descriptor is passed to getnodedat() and the lock retry counter was
reached, the file would be closed, but the passed file descriptor reference
would not be set to -1. This could result in exceptions (from subsequent read
attempts on the referenced file descriptor) in cases where the node.dab could
not be locked or read by getnodedat() and was thus closed.

The set/get_node_* helper functions (used by MQTT) were not initializing the
node.dab file descriptor (i.e. to -1), so it's possible getnodedat() could
try to read from and close an invalid/wrong open file descriptor. If the local
variable happened to be initialized to a value <= 0, then, no problem, but
this is undefined behavior (UB).

This is the proper way to manually-restore the most recently saved
window (via the WIN_SAV mode flag to the list() method).

I wanted something that was backward comaptible with earlier builds
of SBBS, so just doing a magic thing when decrementing the save_num
rather than adding a new method seemed the seamless way to do it.

This allows for menu options on an "active" list (with both WIN_SAVE and
WIN_ACT mode flags) to close the window and restore the most recently saved
window.

Also, for the range check on the save buffer count in ulist(), if we've
hit the bounds, don't *just* beep (though, that was useful), but turn off the
WIN_SAV mode flag so we don't overflow the sav[] array and corrupt the heap!

... for UIFC apps/scripts that have screen-width dependant behavior.

caught by valgrind

We need to read the file header (not just index) entry to get the actual
full filename.

Also, removed some extraneous calls to getfilepath() that served no
purposed.

Don't actually truncate the string, just the display length

This was only a problem when these methods were passed a file-meta-object
instead of the file name. In hindsight, that's kind of a dumb API "feature"
since we're just parsing the name property out of the object.

The file_t.file_idx.name is limited to 64 chars, so don't use that struct
element to store the parsed "name" property of the passed file-meta-object.

uifc option lists are always NULL-terminated string lists (not static
arrays of char arrays, as is often used in other UIFC utils), so grow
the ability to have blank options (list items).

This was needed for fileman.js to accurately edit extended file
descriptions that contain blank lines.

This requires that the option list be NULL-terminated, when used.

For use with js_uifc.c (JS uifc.list method).

For some reason, this exposed an issue where "(null)" would be printed for
the last line when combined with WIN_XTR mode flag, so worked around that
bug too.

Working on fileman.js, when moving files to another base, the output of unzip
(trying to extra a FILE_ID.DIZ) was corrupting the display. We don't really
want a DIZ file to get imported when moving files between bases anyway.

So the second parameter to FileBase.add() can be either:
- undefined: will try to extract/use DIZ only if there's no ext desc already
- true: will try to extract/use a DIZ even if there's an ext desc already
- false: will never try to extract/use a DIZ

If the directory is not configured for DIZ support, then the DIZ extraction/use
will never occur under any condition.

For use with fileman.js:

The space bar can be used by the user to "tag" items, or Ctrl-A to toggle
tag status for *all* items. Returns MSK_TAG|with the option tagged/untagged
or MSG_TAGALL when all tags are being toggled (via ^A).

Also, fix bug where ^Find command would not restore the status bar after
ulist() would overwrite it with the string-input/edit hot-keys available.

Cosmetic change only, output of 'smbutil s' command when used on a filebase.

Increment the SMBUTIL version from 3.19 to 3.20

This addresses GCC warnings about not using the return value of send().

This rule should really be in build/Common.gmake or similar

Not sent in connection banner.

I noticed these redundant properties were being logged for every logon in
logon.jsonl. We don't need these properties separately enumerable, and really
they shouldn't be used anyway.

computer is an alias of host_name
modem is an alias of connection

Fixes cosmetic/readability issue when imported DIZ files that have line/word
wrapped text and the space between words was being removed.

Just for testing/experimenting with proper recording locking support. e.g.
when one process or network node (on a LAN) has the SMB header locked, another
process/node should not be able to lock it or reading the status header.

Now can include command-line specifiers.

The exception is: DOS programs on Linux (e.g. Linux DOSEMU), due to the
funky drive letter mappng, this can't be supported for those external
programs.

As requested by Fluid via IRC, for Frotz support, to store user-unique
data files in data/user/####.something (e.g. "%juser/%4.zork").

And warning about signed/unsigned compare

Meaning, it'll negotiate all Telnet options with the remote server (assuming
not an RLogin server), so long as the TG_PASSTHRU option is *not* set.

This allows more Telnet servers to work as expected, especially when the
client (BBS user) is not using a Telnet client!

Fix issue #432