But also, this needs some revision since I'm supposed to send
the header all at once and probably shouldn't be doing a bunch
of individual sendBin calls.

Added incomplete send of haproxy protocol header. Need to sort
out lookup of target (BBS) IP address.

other unnecessary work - if the BBS is using HAProxy but we can't
do the thing with the stuff because we don't have the things.

then throw an error if X-Forwarded-For header was not received,
or otherwise do the HAProxy thing upon connecting.

like to mangle these unless you tell them not to.

Capture the value of the X-Forwarded-For header, if present, and
stash it in FWebSocketHeader['X-Forwarded-For'].

As a work around for TradeWars 2 (v11.14 and v11.20) FOSSIL support:
This door game would never call the FOSSIL "get received character"
function (02h) . But I did notice that it was calling the "Keyboard
read" functions. I never implemented the "Keyboard read" functions
(figured the door game could read the keyboard itself if it wanted to),
but appparently TW2 won't ever read the COM port (using function 02h)
unless the Keyboard read function returns AX=0xFFFF (no keyboard data
available). So now, I just do that and TW2 works (with FOSSIL I/O).

Incremented the revision to 26.
Logging the Git branch/hash and other helpful FOSSIL-debug stuff now.

Needed the actually mapping of XTRN_UART -> EX_UART mode flag here.

When configuring an external program, a Win32 sysop can now choose to *only* enable the virtual UART (and not the FOSSIL driver) feature of the Synchronet virtual UART/FOSSIL driver (dosxtrn.exe/sbbsexec.dll). Some programs (e.g. TradeWars 2) will always use the FOSSIL driver if one is detected and in order to force the use of COM/UART driver for I/O (if that is desired), then one must disable the FOSSIL driver. It's a rare use case, but I can see the potential need.

Renamed the "Intercept I/O" option in SCFG to "I/O Method" and make it clear that "Socket" (for native programs) and "FOSSIL or UART" (for 16-bit DOS programs) is the default method (what was previously identified as "No" I/O Interception).

If a sysop want to disable the virtual UART support (on Win32, e.g. so *only* FOSSIL is available to one or all DOS programs), they do that via their sbbsexec.ini file.

This change needed to build dosxtrn.exe (again).

warning: format ‘%lx’ expects argument of type ‘long unsigned int’

The internal line editor's quoting feature add some hard-coded strings ("Done" and "All") and the (L)ist key was hard-coded. Use the text.dat string (new and pre-existing) for these now.

Also, use the new sbbs_t *_key() methods for referencing the configured key bindings (via text.dat) for these common key-stroke commands.
Convert the text.dat strings for keys (e.g. YNQP) to uppercase always as well.

For localization purposes, no more hard-coded text strings or characters in the new-file and file scan configuration menus.

This change fixes issue 232.

A Hungarian sysop also requested this change via Facebook not long ago.

sbbsctrl.exe is built with Borland C++ still which only has a 32-bit time_t, so to keep the scfg_t definition compatible between MSVC and Borland builds, use time32_t (32-bit time_t) values only.

This fixes the reported sbbsctrl.exe error:
Error loading configuration
cfg->size (23944) != sizeof(scfg_t) (23952)

Introduced in commit e535aaac.
Reported via DOVE-Net by  Daryl Stout (TBOLT), confirmed by Nightfox (DIGDIST).

Since mail server log messages have their white-space condensed, it was not obvious why some sender's name would match a twitlist.cfg line that filters names beginning with a space: "\ *" because the initial space of the sender's name was condensed/combined with space before it in the log message.

As found and reported by Talisam author, apam (HAPPYLND).

sscanf() won't initialize the buffer of the argument passed if the format string is not matched, so when there was no hexadecimal/SMB timezone value, the timezone specified in the ISO date/time string (if any) would be overwritten with 0, converted from the uninitialized 'zone' variable (unlikely to be valid hex-ASCII chars). Thanks apam!

where "address" is a QWKnet ID or path or a FidoNet originating address. Wildcards can be used (e.g. "*@VERT").

For Trikester.

%+ will now expand to the current user's real name, automatically enclosed in quotes if it contains any spaces.

The "serialno" field is a bad/naive one (a time_t value), but this is really just an experiment for Ragnarok to see if this fixes issue #306 (with hotdoged receiving AreaFix responses/notices).

Found during review of previous commit.
Introduced as part of the new file base branch merge (commit fcf58640).

Environment variables are shared among all threads of a process (e.g. all nodes of a instance of sbbs), so this scheme of passing the uploaded filename and description to an external file tester (upload processor) via environment variables has always been fraught with the possibility of failure or flakiness in Synchronet v3. Since I very much doubt that any upload processor is actually using this scheme, just remove it. Upload processors can and should)use the sbbsfile.nam and sbbsfile.des text files created in the node directory instead.

When a linked-node (in echocfg->Linked Nodes) contains one of the local system's FTN addresses, display/log an error and exit immediately, e.g.:
Configuration ERROR: Linked node #20 is your own address: 1:103/705

I'd perform this check in EchoCfg, but:
1. echocfg doesn't read the BBS's configuration files (where the local FTN addresses are configured), and
2. some sysops just directly edit their sbbsecho.ini file

With the old method, it was possible for a certificate to remain used
eternally, and letsyncrypt.js can change it relatively often.

Now correctly rejects UTF-8 encoded Unicode surrogates and does not support
5 and 6 byte UTF-8 encodings.

For reference:
https://github.com/openssl/openssl/commit/ba64e5a92a6f009e311ad1c3565817820a1632a4

Optional, for easier use with game servers that take one of the 3 rlogin
negotation parameters as the name/code of the door to execute. A telgate
mode flag value argument must be provided (use 0 for default behavior)
if you want to provide any of the other arguments to override the defaults
(the user's alias, real name, and current detected terminal type).

When there's no "uploads" directory configured by the sysop, cfg.upload_dir will be set to INVALID_DIR, which cannot be used as an index into cfg.dir[] to determine if the time used for uploading the files should be "given back" to the user.

In v3.18, we used the first file in the upload queue, if there was one, else fell back to the "uploads" dir (which had to have been defined if there were no files in the queue). So this illegal array indexing was a regression in v3.19.

Reported by Zoltán Gábor on Facebook

Per https://man7.org/linux/man-pages/man3/pthread_yield.3.html
	This call is nonstandard, but present on several other systems.
        Use the standardized sched_yield(2) instead.

Should fix #299.

We don't support platforms where '/' doesn't work as a path element separator.

Per https://man7.org/linux/man-pages/man3/pthread_yield.3.html
       This call is nonstandard, but present on several other systems.
       Use the standardized sched_yield(2) instead.

Should fix #299.

As reported by Ragnarok (DOCKSUD) via DOVE-Net, when packing (packetizing) netmail messages, the source and destination zones from the netmail header were discarded (since they are defined as "optional" in FTS-1) and the zone values were replaced with the system's primary/default zone number. If the netmail message included an INTL kludge, the source and destination zone values would be parsed from that kludge line and this issue would not be noticed. 

So the fix is: Only if the netmail message header contains a source or destination zone number of 0, replace that zone with the system's primary/default zone number. As before, the INTL kludge zone information can override the header fields.

Also, there's no reason to force the point numbers from the header to zero, since if they were not valid (e.g. "fill" in the original FTS-1 specification), they would be already zero. So use the source and destination point numbers, as is, from the netmail header too (with the FMPT and TOPT kludge override support left in-tact).