- This option instruts JSexec to *not* change the current working directory to
  the control (CTRL) directory
- This is useful when using jsexec to deal with files and passing relative
  paths or wildcards on the command-line
- Warning: scripts executed with this option may behave differently than the
  same script executed from SBBS where the current working directory is always
  the CTRL directory.

- FILE_AREA
- FILE_AREA_DESC
- FILE_LIB
- FILE_LIB_DESC
- FILE_LIB_NUM
- FILE_DIR
- FILE_DIR_DESC
- FILE_DIR_CODE
- FILE_DIR_NUM
- FILE_NAME
- FILE_DESC
- FILE_UPLOADER
- FILE_SIZE
- FILE_CREDITS
- FILE_TIME
- FILE_TIME_ULED
- FILE_TIME_DLED
- FILE_DATE
- FILE_DATE_ULED
- FILE_DAT_DLED
- FILE_TIMES_DLED
Fixed MSG_ATTR @-code (missing SPAM attribute)
New text.dat line: FileHdrDescSeparator, displayed between file detail display
(name, date, uploader, etc.) and extended description, if there is one. This
is used to execute showfileavatar.js.
New JS 'bbs' object properties:
- file_name
- file_description
- file_dir_number
- file_attr
- file_date
- file_credits
- file_uploader
- file_update_date
- file_download_date
- file_download_count
Inspiried by Lon Seidman's now infamous (but likely common) Synchronet sysop
blunder (removing a file rather than downloading it while demonstrating his BBS
on his youtube channel:
- New text.dat line: RemoveFileQ <- using this rather than the vague:
  AreYouSureQ, when prompting to remove a file
- The sysop command to remove files from the batch file flagging prompt has
  been changed from 'D' (delete) to 'R' (remove).
- The 'D' command from the batch file flagging prompt may now be used for
  immediate Download of any flagged file(s).
- When flagging files with '*', treat the same as '*.*'

  MsgHdrBodySeparator
This is display between the msg header text.dat lines and the message body
(previously, just hard-coded to a blank line).
This enables us to now execute showmsgavatar.js much faster than loading
msghdr.asc file, and using it to execute showmsghdr.js.
So... if you're upgrading to this build, you can get rid of msghdr.asc and
get a message viewing performance boost. The results should look the same
(but the message header will display immediately, with no discernable delay).

- When the Ctrl-A code is encountered, the terminal server will output a
  new-line sequence (CRLF), but only if the current column is not the first
  column. This is helpful for text.dat lines where you cannot make assumptions
  about the previously displayed text and you want to insure the cursor is at
  the beginning of a new-line without creating unnecessary blank lines in the
  output.

check the protocol.log (e.g. DSZ.LOG) for successful transfer even when the
protocol was so-configured in SCFG. Fixed.
Additionally, while sendfile() would return a proper boolean result based on
the errorlevel returned by the protocol driver process, the log and console
output would report success and statistics adjusted as though the xfer
was successful, regardless.

attempted to be sent (e.g. there are file open failures for all files), then we
won't call zmodem_get_zfin() - we never sent a ZINIT! Just return -1.

Rastus:
file_to_netmail() assumed the file pointer was at the end of the file.
This was not the case with the areamgr.hlp file (just opened, not read or
seeked).
Also, added some more error logging in case of file_to_netmail() failure.

1) Only call crpytFlushData() if there *is* data.
2) Only call cryptFlushData() once per call to do_CryptFlush()
3) If cryptPushData() does a short write, flush immediately, regardless of
   the preferences
4) Pass the whole private data pointer to do_CryptFlush() so it can only
   call cryptFlushData() if there's unflushed data
5) Extend the private data structure to track unflushed data

line (CRLF) after the header and before the message body. If you want a blank
line, it must be in the msghdr.asc file. This change is to be consistent with
the behavior in v3.16.

- If the first lines sent after a key-press are blank lines, don't count them
  in the line-counter (lncntr) - maximize the viewable data before pause.
- Assume tabs chars are expanded to 8-space tab-stops when detecting auto
  line-wrapped output (e.g. when ;LISTing text files with tabs)

Don't ask the user if their terminal supports RIP unless they already answered
yes to ANSI-support (can't have RIP without ANSI).

- if the sysop had text/menu/<filename>.bak (for example), the optoinal menu
  would attempt to load, resulting in an error ("<filename>.asc not found")
- if the sysop specified menu_file or menu_dir (e.g. via Baja), the existing
  checks for optional menu file existance would not work

<ESC>[5~ and <ESC>[6~ (these appear to be VT-220 sequences).

Not defined in MSVC2013.
Appears to work again/still.

input on Windows (e.g. the prompt() method no longer worked) broke JSDoor,
and caused Windows and Linux console input to behave differently.  Create
functions to cook/thaw the tty, and call them around stdin things.
This change was related to his postpoll.js work, so that might be broken now.

than non-SPAM-tagged email.

xpbeep.c assumes you have sys/ioctl.h on *nix now. Hope you do, mr. fbsd.

SPAM-tagged messages will not be listed/downloaded. This is useful if for
example you use POP3 to download mail to your phone and have limited
storage or sorting options. The SPAM-tagged messages will remain in your
inbox on the BBS so you should use another POP3 or local mail client to
download and delete those messages (which could potentially include
false positives).

and Ctrl-N (next) instead Ctrl-U and Ctrl-D as that wasn't going to work with
fseditor.js's (Ctrl-U is quote).

SBBSCTRL is set. Also changed the library versioning from 3.16 to 3.17.

Created 'install' and 'symlinks' targets to ease the copying (install) or creating of symlinks after build, requires SBBSEXEC env var

(helpful in determining proper BSO directory for an "othernet" zone).

-> Advanced tab

lines - don't check the (invalid) sbbs_t:useron.qwk settings for QWKE
support in this case (!).
If you've been seeing the occasional message with "Subject:" as the first
line followed by some @-kludge lines (e.g. @MSGID, @TZ, etc.) this would
be the cause: you have QWKE support enabled on your hub BBS (e.g. VERT),
which adds the "Subject:" kludge lines for subjects > 25 chars and when
importing QWK messages from a *hub*, these lines weren't (normally)
parsed/stripped.

sequence is currently just the <esc>[V and <esc>[U sequences (supported by
SyncTERM).

exists in the message body.

When displaying message body text, skip all initial ocurrences of CR or LF
(rather than all initial white-space). Some people use white-space for
centering or justifying blocks of text and we don't want to mess up the
presentation of their messages.