Based on qwk.ssjs

I tested this using:
wget --auth-no-challenge --post-file=file.ext --http-user=username --http-password=passwd vert.synchro.net/upload.ssjs?filename=file.ext\?desc=description-text

This currently requires an "uploads" directory to be configured by the sysop.

I'm not sure if there's a better/more-standard way for the posted filename or
content to be included in the request. This is just a sort of proof-of-concept
as a solution for issue #554. Perhaps an ecbwebv4 page displays a prompt to the
user for their filename and it redirects to this page. Not sure how that'll
work.

Tested with this:
wget --content-disposition --post-file=nodelist.010 http://synchro.net/fido-nodelist-syncterm.ssjs

Reported by AL@TRMB.

Resolves this errors appearing on Vertrauen where the sub=<code> value appears
to be from some *other* Synchronet BBS:
Unrecognized msgbase code: dove-firearms, Request: /msgs/msgs.ssjs?msg_sub=dove-firearms

msgs.ssjs line 6: TypeError: msgbase is undefined

Instead, just display an error to the user/browser (assuming there is one).

"msg.ssjs line 42: Error: Error -1 reading user number"

Fixed an errant semicolon in the process. (15 year old minor bug)

This excludes votes and polls automatically.

I'm not sure how long this has been like this, but the last_msg
of a sub-board can be a super high number (e.g. in the billions), so using
that value as the highest message offset would cause tons of message index
read failures and just cause this loop to interate unneccessarily through
tons of non-existent messages. Also, this code is utter crap.

The days of browsers rendering HTML served-up via FTP are over, so remove this feature. This also removes all JavaScript support from the FTP server and that is a bit odd as it was one of the first Synchronet components for which I added JS support.

Removing this feature was pretty painless; much easier than adding it was. The main motivation was less cruft to port to the file base in the works. There should be no more references to 00index.html anywhere at this point. Bye bye cool feature, we'll miss you.

So don't bother linking to 00index*, just the ftp root dir should work
(for a while yet, anyway)

Double-check that the user has access to both the directory and the containing
library before allowing a download a file.

directory or library does not exist. (Thx DM)

Fixes error: ReferenceError: MGS_HFIELDS_UTF8 is not defined
when attempting to post a poll.

(Jump back to #fTelnet anchor when doing the click thing on
a clickable click thing to do the thing.)
Complaint courtesy of Android8675309

Don't display the name of messages posted with the anonymous attribute
and force messages posted to anonymous-only subs as anonymous.

total credits.
Set an 'error' property on reply if file not found or not
enough credits.

The User.downloaded_file() method has been enhanced to accept a directory
code and a filename to do "all the things" expected after a file has been
successfully downloaded. Use the method.

in the mail and forum pages.

If modopts -> [web] -> forum_no_spam, then filter spam messages
in the forum.  (Maybe make this more advanced in the future so
users can see these messages if they want to.)

When posting a reply-message, if the original "from" field is the
current user's alias, then reply to the original message's "to"
value instead of the "from" value. This is also how sbbs handles
reply 'to' fields.

Using Linux top, long-running (hundreds of hours) httpSess threads were seen
consuming ~15% of a CPU core. This loop calls time() and yield() in a tight
loop, so instead mswait(50) (milliseconds), which reduces the CPU utilization
considerable (down to < 1%), but will slightly impact the frequency with
which the events callbacks are invoked. A better scheme would be to block
while waiting for an event or a timeout (e.g. 1 second), so this is not a
great solution. More of a work-around.

Some messages come from non-English locales with non-UTF-8 charsets
(e.g. CP437), so their to/from fields may contain non-ASCII/non-UTF-8
characters (e.g. "Björn Felte") - so convert these fields (to/from)
to UTF-8, as needed. Uses the new(ish) utf8_encode() function, so you
must have a recent v3.17c build.
I did not transcode message subjects as I'm not clear how they are stored
and copied here. That's probably something better handled by echicken.

We don't (yet) support UTF-8 sequences in telegrams (short messages to users).

Respect the local "allow dupe realnames" setting.
Slightly more careful checking of some things.
More to come.

newuser_level
newuser_flags1
newuser_flags2
newuser_flags3
newuser_flags4
newuser_exemptions
newuser_restrictions

When message header.is_utf8 is true, don't do the CP437->HTML entity
encoding dance.

There's a setting in SCFG->System->Toggle Options:
"Include Sysop Activity in System Statistics"

This option prevents sysop logons from being "counted" for stats
purposes as well as from being added to the logon list. Make this
so for web logons too.

When run with a build that includes utf8_encode() (new global JS
function), use it to convert the user's .sig file from CP437 to UTF-8.

When creating a new message (post or email), set the auxattr flag
MSG_HFIELDS_UTF8 since we support UTF-8 chars (not CP437) in message
header fields (e.g. to/from/subject). Requires updated load/smbdefs.js

using the FTN charset method (since we're not creating MIME-encoded messages
here).
Convert .sig files to straight-ASCII, for now. We really need a CP437->UTF-8
conversion routine, don't have one yet (in JS).
Also check the MsgBase.open() return value.

validation.