1) Comments do not start with a ;
   Any line that doesn't start with an @ is either a parameter to
   the last command (such as in @show), or is ignored.  Further,
   "extra" arguments are also ignored.  This means you can literally
   type comments almost anywhere.
2) There is a maximum "call stack" of two "frames".  If you nest
   subroutines three deep, the first one will never be returned to.
3) Implement @routineabort.
4) If @run is used in a subroutine, it acts as an immediate return
   The arguments are not consulted, and no error is displayed.
5) Of course, if you call subroutines three deep and return from
   two of them, the first subroutine is now the topmost frame, so
   run behaves normally there.

Since the filename is in the subject, we can't allow the user to change the subject (duh). Also, don't bother with the "Fwd:" prefix thing as that could potentially cause problems in other places where the filename (only) is assumed to be in the subject.

Also, make sure to create the destination directory when moving file attachments. These changes fix issue #230 reported by Keyop.

I noticed "DAILY ... Semaphore signaled" in the event thread log output. Now fixed.

show splash screen and await response before restoring system status. explicitly disconnect from json client when exiting.

Fixes issue in healers huts.

Documented here: https://pubs.opengroup.org/onlinepubs/009695399/functions/xsh_chap02_05.html#tag_02_05_01
This provides the best chance at read_raw() reading from the right
location and updating the file pointer as expected while avoiding
behaviour which is explicitly undefined by POSIX.

This allows more @whoison to be more accurate.
Also, fix a misplaced ) while I'm here.

(More issues with rape)

REF files need to set it manually if they want it.

1) ` vars which are always expanded.
2) & vars which are expanded "sometimes" (generally for display stuff).
3) "pure" vars which are only expanded when they are the entire string.

The main change here is in the & var handling... lw() no longer
expands these, so they can be displayed to the user... they are now
expanded in the following cases:

1) Anything that calculates the displayed length.  The assumption
   here is that the string will be passed to one of the following
   things.
2) Bar updates, either @quebar or @saybar
3) @do addlog
4) @do write
5) @moremap
6) @progname
7) @show
8) @writefile

I think I was intending to strip & and prevent those variables,
but it looks like the original doesn't.

However, it also looks like the original doesn't *always* replace
string variables... will need to look into that harder in the future.

reverse the logic in my previous commit :| also dont show the next turn alert if hidden_names == true

separated list of interfaces to bind to.

Prevents invsible other players. :D

While we're here, remove the janky redraw avoidance and fix various
other item menu issues.

- Fix update_space() to move to correct location every tim
  Resulted in "ghost" player character after using smackrod
- Load the map and set lastx/lasty when loading an existing
  player record.  Undefined lastx/lasty would crash the game,
  and using the smackrod on joining allowed that.

While it would be nice to serialize it and do a single write, there's
no native functions to create binary values from numbers in the
Synchronet JS object model at this time.

Should eliminate ghost "busy" status.

Should fix the issue with rape.

Arr.

1) Call update_update() more often when player data changes.
   Should hopefully solve issue with players staying red after a
   battle on other people's screens.
2) When erasing other players, also erase ones that have moved off
   your map.

As of this commit, LORD2 keeps track of the last time a key was
pressed, and a timeout occurs five minutes after that if it's
waiting for a key.  This has issues as the getstr() timeout does
not behave in this manner.

However, if the timeout callback style works, getstr() and the other
dk functions can grow callback support pretty easily, and this seems
like it may be the best way forward.

Further, disconnect handling should be tied up in this, and dorkit
does not have very robust features for this.  In most cases, a
disconnect will be treated the same as a timeout, so the same solution
should likely be usable by both.

The biggest challenge is preventing the need for all the code to
have exceptional condition handling, which can be quite tricky as
battles can end by running a REF file (For example, the training
centre).  This would mean all code paths reachable via REF calls
would need to fail nicely in the face of a connection loss/timeout.

There were 2 bug identified by issue #230:
- the "Fwd:" prefix being added to the message subject was being treated as a filename. I first thought to just remove this subject tag, but then thought it best to just ignore obviously invalid filenames in the subject in the first place.

- when forwarding files to a netmail address, the 'to' extension (user number) is 0, so the file will be in the data/file/####.out directory of the sender instead.

We have 4 places (at least) where the message subjects are parsed and only one of those places currently supports quoted filenames (e.g. with spaces in them) and some of the others (e.g. QWK) don't support multiple filenames at all. That should be fixed.

Renamed from *.asc to *.msg
Replaced mouse hot-spot @-codes with the color-coded hotspots
Removed trailling newline in head.* to avoid line-wrap in PabloDraw
on @ include line
The Sysop Menu line still wraps (in PD) - no easy solution for that
just yet.

For nickshanks1 in IRC.