lines, which were potentially garbage.

Also, with -a, dump a series of screens making up the ANSI, rather than the
last screen.

Don't make a blocking-call to send/sendsocket() without first checking
writability with select().

will now expand to "raw" rather than "telnet/rlogin/ssh". This disables IAC
handling in sexpots, for example, which isn't needed or desired when using
the passthru socket.

Also, put some conditional checks for passthru_thread_running in external()
and if it's not (e.g. for testing purposes), then lock the input_thread_mutex
instead of changing passthru_socket_active. This is currently just a way to
easily test the differences between direct and passthru socket I/O
(dependant on whether or not the passthru_thread is running which is controlled
in main.cpp.

information, made again for X/YMODEM file transfer progress (xmodem_progress).
I guess we haven't been testing X/YMODEM file transfers in win32 builds...

from a Win32 build: a ulong is not 64-bits in these builds.

use socket I/O (e.g. sexyz, door32.sys doors, sync-xsdk doors) - not just when
the client is connected via SSH. This solves a number of problems:
- programs that change socket options
- programs that don't support non-CP437 charsets (e.g. PETSCII, UTF-8)
- programs that don't support Telnet-IAC escaping

This greatly simplifies (reduces the need for) input_thread locking.

The passthru_thread now handles chunks of up to 4000 bytes per transfer
between socket and ringbuf, instead of a char at a time, greatly improving the
performance.

This allows optimizing tall ANSIs.

That much closer to eliminating direct ANSI stuff.

if fdopen() failed, we'd close the file descriptor but leave it dangling with
a >= 0 value and then call close() again on it later in the function.

depend on a socket descriptor still won't work.

"At least two arguments required (interfaces, port, and protocol)"

'=': conversion from 'time_t' to 'unsigned long', possible loss of data

Force console.line_counter to zero every time we do anything with the console
Set dk.user.seconds_remaining_from

This is time() in sbbs mode, or the date of the dropfile in other modes.

error message if failure to find the file to load less confusing.

to accommodate longer group / sub-board names without an ugly line wrap.

(MsgPostedToYouVia) to send telegrams to local users when messages posted
"to them" are imported via QWK.

from sbbsecho which uses a linked-list for the cached user list. Users are
looked up by alias or real-name (alias is has higher precedence). The found
user number is returned or 0 if not-found (or other error).

setting normal/lightgray text attributes when saving message text to file:
WHITE is 0x0f (high-intensity lightgray). LIGHTGRAY is 0x07.

when "extra raw" output is needed.
Use putcom() and comprintf() instead of rputs() and rprintf() for ANSI escape
sequences. This doesn't actually fix any problem, but since rputs() is
"less raw" than it used to be, let's use direct-output routines instead, for
ANSI output.

On Linux, using DOSEMU, he would end up with both door.sys and DOOR.SYS in the
node directory and some doors would open the wrong file and fail to initialize.
Solution: delete all case-versions of drop filenames before creating them.
Don't you just love case-sensitive file systems? :-(

(became a blank/solid blue line).
Remove one space between "CTRL-K for help" and "Insert/Overwrite Mode" to
accommodate the new longer revision. This should work for 898 more revisions.
< 80 col terminals likely have a different issue wiht the status line
we should look at as well.

which uses Ctrl-A codes, this resets the attributes back to "normal" so the
reply text doesn't "inherit" the original message's attributes.

In *nix builds of addfiles, when adding a file list (e.g. files.bbs), the
case (capitalization) of the filenames is taken literally from the file list,
and that may be incorrect in same cases. Detect the correct case of the
filename and use it when importing into the filebases.

This should fix the error BrokenMind reported.
Thanks to nelgin for more excellent bug reporting and debugging.

line counter was never cleared (upon rputs() of "\n") and the current attribute
was not saved when a new line was being buffered.
This applies to JS console.write() as well since it uses rputs().