the file and sets the descriptor to -1 (if not already < 0).

deprecation warning from MSVC.

https://patchwork.kernel.org/patch/9289177/

debug-builds only.

- PWAGE - age (in days) of current user's password
- PWDAYS - maximum age (in days) before forced-password-change (or "unlimited")
- PWDATE (alias for MEMO) - date of last change of the current user's password
- AUTODEL - days of user inactivity before auto-deletion (or "unlimited")

(user) Inactivity" system setting (in days, 0 = disabled).

Eliminate duplicate case statements added in previous commit (some of these were already mapped using numeric constants).

Added more "black square" -> CP437 0xFE mappings.
Added Unicode FULL/HALF block -> CP437 mappings.

definitions.

translating from Unicode to CP437.
Use the UNICODE_HOUSE macro instead of the codepoint hex-value.

was not a recoverable message header (e.g. the toUserName field was *not*
at offset 0x22):

xx 02 00 01 00 01 00 6A - 02 FE 01 00 00 00 00 30   : .......j.......0
34 20 46 65 62 20 31 31 - 39 20 20 32 30 3A 32 36   : 4 Feb 119  20:26
3A 33 32 04 00 4B 75 72 - 74 20 57 65 69 73 6B 65   : :32..Kurt Weiske

So restore the DateTime field validation: the 20th byte must be null.

through debug-level log messages exactly why a QWK vote message failed to
import.

created in the global scope when when a scope is passed to load(), require()
or js.exec().

While having 'var x = 0;' and 'x = 0;' put x in different scopes is bad, we
actually want them all in the passed scope, not the global scope... and that's
not an option.

It looks like js.exec() will actually need to set scope as the global object
and add a second global object to the original scope.

contains one or more UTF-16 encoded characters.

utf8_encode() can now accept a UTF-16 encoded string (e.g. "\u0f00") and will
return the UTF-8 encoded version.

New global functions:
- str_is_utf8()
- str_is_ascii()
- str_has_ctrl()

Return a more sensible value when 'null' is passed to the utf8_* functions.
Updated some JSDOCS strings in regards to "extended-ASCII" and what that
means (in some cases, specifically CP437, in others, any non-ASCII chars).

any control (ASCII 0x01 - 0x1F) characters.

the "js" object.

These almost certainly improve performance a bunch, but may result in
new behaviours.

child scope.

object.

not the global scope.  This appears to be a systemic issue, and could even
be the root cause of the background log problem...

The procedure entry point inet_ntop could not be located in dynamic link
library ws2_32.dll

run with the '-p' or '-!' options. This resolve the issue of (on Win32)
having to enter *twice* to close the program in this mode.

This resolves the issue where if an external program closes the
client_socket_dup, the passthru_thread can't write to the passthru_socket,
so it terminates itself - by definition, deactivating the passthru_socket, but
it didn't clear passthru_socket_active, so the input_thread would keep trying
to write to it rather than sending the input from the user to the BBS.

Also, terminate the passthru_thread (without any special log message) when
sbbs->online is non-zero.

Also overhauled a lot of the passthru-related log messages. Passthru ain't just
for SSH anymore.