rogomatic.6.in

.TH ROGOMATIC 6 "2026 Aug 10"
.UC 4
.SH NAME
rogomatic, run_rogo, rerun_rogo, mrerun_rogo, mstop_rogo, end_player, unstuck_player \- Automatically Exploring The Dungeons of Doom
.SH SYNOPSIS
.I rogomatic
.in +0.5i
[
.B \-h
]
[
.B \-V
]
.br
[
.BI \-a\  secs
]
[
.B \-c
]
[
.B \-d
]
[
.BI \-D\  rgmdir
]
[
.B \-e
]
[
.B \-E
]
[
.BI \-f\  rogue
]
[
.BI \-G\  goodlvl
]
.br
[
.B \-H
]
[
.B \-p
]
[
.BI \-P\  player
]
[
.B \-q
]
[
.B \-r
]
[
.BI \-S\  ROGOSEED
]
[
.B \-t
]
[
.B \-u
]
[
.BI \-U\  usec
]
[
.B \-w
]
.br
[
.B \-s
[
.I rogue_ver
]
|
.I r_file
]
.in -0.5i
.PP
.I run_rogo
.in +0.5i
[
.B \-h
]
[
.B \-V
]
[
.BI \-v\  level
]
[
.B \-n
]
[
.B \-N
]
.br
[
.BI \-a\  secs
]
[
.BI \-C\  coredir
]
[
.B \-d
]
[
.BI \-D\  rgmdir
]
[
.B \-e
]
[
.BI \-f\  rogue
]
[
.BI \-G\  goodlvl
]
[
.B \-H
]
.br
[
.BI \-P\  player
]
[
.BI \-r\  rogomatic
]
[
.BI \-S\  ROGOSEED
]
[
.BI \-U\  usec
]
[
.B \-Z
]
.in -0.5i
.PP
.I rerun_rogo
.in +0.5i
[
.B \-h
]
[
.B \-V
]
[
.BI \-v\  level
]
[
.B \-n
]
[
.B \-N
]
.br
[
.BI \-i\  idlesec
]
[
.BI \-R\  run_rogo
]
[
.BI \-s\  stopfile
]
.br
[
.BI \-a\  secs
]
[
.BI \-C\  coredir
]
[
.B \-d
]
[
.BI \-D\  rgmdir
]
[
.B \-e
]
[
.BI \-f\  rogue
]
[
.BI \-G\  goodlvl
]
[
.B \-H
]
.br
[
.BI \-P\  player
]
[
.BI \-r\  rogomatic
]
[
.BI \-S\  ROGOSEED
]
[
.BI \-U\  usec
]
[
.B \-Z
]
.in -0.5i
.PP
.I end_player
.in +0.5i
[
.B \-h
]
[
.B \-V
]
[
.BI \-v\  level
]
[
.B \-n
]
[
.B \-N
]
.br
[
.BI \-u\  user
]
.PP
.I mrerun_rogo
.in +0.5i
[
.B \-h
]
[
.B \-V
]
[
.BI \-v\  level
]
[
.B \-n
]
[
.B \-N
]
.br
[
.BI \-i\  idlesec
]
[
.BI \-o\  unstuck_player
]
[
.BI \-O\  rerun_rogo
]
[
.BI \-R\  run_rogo
]
[
.BI \-s\  stopfile
]
.br
[
.BI \-a\  secs
]
[
.BI \-C\  coredir
]
[
.B \-d
]
[
.BI \-D\  base_rgmdir
]
[
.B \-e
]
[
.BI \-f\  rogue
]
[
.BI \-G\  goodlvl
]
[
.B \-H
]
.br
[
.BI \-P\  player
]
[
.BI \-r\  rogomatic
]
[
.BI \-S\  ROGOSEED
]
[
.BI \-U\  usec
]
[
.B \-Z
]
.I ID
.PP
.I mstop_rogo
.in +0.5i
[
.B \-h
]
[
.B \-V
]
[
.BI \-v\  level
]
[
.B \-n
]
[
.B \-N
]
.br
[
.BI \-D\  base_rgmdir
]
[ \fIID ..\fP ]
.in -0.5i
.PP
.I end_player
.in +0.5i
[
.B \-h
]
[
.B \-V
]
[
.BI \-v\  level
]
[
.B \-n
]
[
.B \-N
]
.br
[
.BI \-u\  user
]
.PP
.I unstuck_player
.in +0.5i
[
.B \-h
]
[
.B \-V
]
[
.BI \-v\  level
]
[
.B \-n
]
[
.B \-N
]
.br
[
.BI \-D\  rgmdir
]
[
.BI \-m\  missing_sec
]
[
.BI \-r\  recheck_sec
]
.br
[
.BI \-R\  restart_sec
]
[
.BI \-u\  user
]
[
.BI \-z\  loop_sec
]
[
.I unstuck.log
]
.in -0.5i
.SH DESCRIPTION
.PP
.I Rog-O-Matic
(hereafter referred to as
.BR rogomatic )
is a computer fantasy game with a newer twist.
It is CRT (Cathode Rat Tube) oriented and the
object of the game is to automatically survive the attacks of various
monsters and get a lot of gold, rather than having to actually type the
required commands yourself.
.PP
The
.B rogomatic
plays the game of
.BR rogue (6)
on your behalf while you watch.
.PP
Once you have entered the world of
.B rogomatic
you will stay there, passively watching the game unfold before you much
like a catatonic teenager watching television.
Should you wish to intervene, type the
.B t
command and
.B rogomatic
will enter transparent mode.
See the
.B COMMANDS
section below for details on commands, and on transparent mode.
.PP
When
.B rogomatic
launches
.BR rogue (6)
via the
.B player
program, it sets the
.B ROGUEOPTS
environment variable to a value such that the
rogue name used will start with
.BR rogo\- .
The rogue score path to
.BR @RGMDIR@/rogue.scr ,
which is
.B NOT
the default rogue score file.
Thus by default,
.B rogomatic
game stats will not compete with the humans who are playing
.BR rogue (6)
in the default way.
.SH OPTIONS
.TP
.B \-h
Print the usage message,
and exit with a status of
.BR 2 .
.TP
.B \-V
Print the
.B rogomatic
version string
and exit with a status of
.BR 2 .
.TP
.BI \-a \ secs
Set the timeout timer to
.B secs
seconds,
and set the
.I ROGOTIMER
environment variable to
.BR secs .
.sp
The
.B secs
must be a value >= 0,
and it may an integer, or
a real value (i.e., 4.5).
By default,
.B secs
is set to
.BR 0.0 ,
which has the effect of not setting a timeout timer.
.sp
For example, using
.BI \-a \ 4.5
on the command line will cause the program attempt to resume normal
operation after a hang while reading a message from
.BR rogue (6).
.sp
Before
.BR rogomatic (6)
attempts to read from
.BR rogue (6),
a timeout alarm will be set.
If the read hangs for more than \fIsecs\fP seconds, a
.B SIGARLM
signal will be delivered to the player program.
The
.B SIGARLM
is then caught and a
.BR siglongjmp (3)
performed a non-local jump to a control point where player program
is much more likely to resume normal operation.
Nevertheless, there are circumstances where non-local jump still hangs.
.sp
See the section on \fBBUGS\fP below for more information on hanging.
.TP
.B \-c
Turn on "\fIcreative mode\fP".
.sp
In "\fIcreative mode\fP", a number of strategies are employed,
many of which are used by wide variety of human players,
such as using trap arrows.
.sp
Unfortunately a number of these "\fIcreative mode\fP strategies mostly
work on very old version of
.BR rogue (6)
that are no longer supported.
.sp
Historically, this mode was called "cheat mode" even though
the same strategies were used by wide variety of human players.
.sp
When using with
.BR rogue (6)
with \fBrogue version 5.4.5\fP or later,
the "\fIcreative mode\fP" is enabled by default.
.sp
See also the
.B c
command in the in transparent mode in the
.B COMMANDS
section below.
.TP
.B \-d
Use a UTC date and time sub-directory under rogomatic directory path
to store files.
.sp
By default,
.B rogomatic
creates the rogomatic directory:
.sp
.in +0.5i
.IR @RGMDIR@
.in -0.5i
.sp
With the
.B \-d
option,
.B rogomatic
creates a new rogomatic sub-directory, using a UTC date and time, of the form:
.sp
.in +0.5i
.BI @RGMDIR@ /YYYYMMDD_HHMMSS
.in -0.5i
.sp
where
.I YYYYMMDD_HHMMSS
is the year, month, day, hour, minute and second, in the UTC timezone,
of when player program started.
.sp
Because
.B \-d
creates a new rogomatic sub-directory, the environment starts out in an initial state.
This will make is easier to use a debugger such as
.BR lldb (1),
especially when combined with
.BI \-S\  ROGOSEED
making the action much more deterministic from a debugging standpoint.
See the
.B DEBUGGING
section near the bottom of this man page for more details.
.TP
.BI \-D \ rgmdir
Change the rogomatic directory path to
.IR rgmdir .
.sp
The use of
.B \-d
command line option combined with
.BI \-D \ rgmdir
will case
.B rogomatic
to create and use files under:
.sp
.in +0.5i
.IR rgmdir /YYYYMMDD_HHMMSS
.in -0.5i
.TP
.B \-e
Turn
.B OFF
rogomatic game logging.
.sp
By default, rogomatic game logging is enabled, and the
.I gamelog
file is written to:
.sp
.in +0.5i
.BI @RGMDIR@ /gamelog
.in -0.5i
.sp
.BR NOTE :
Prior to version 14.2.6, this command line option turned on rogomatic game logging.
Now that rogomatic game logging is enabled by default, and the
.B \-e
disables rogomatic game logging.
.sp
See also the
.B e
command in transparent mode.
.TP
.B \-E
Start up in emacs mode.
.TP
.BI \-f \ rogue
Set the path of
.BR rogue ,
overriding the default.
.sp
By default,
.B rogomatic
first looks for:
.sp
.in +0.5i
.B ./rogue
.in -0.5i
.sp
in the current directory, then it looks for:
.sp
.in +0.5i
.B ../rogue5.4/rogue
.in -0.5i
.sp
before looking for
.B rogue (6)
in the default installation location of:
.BR @BINDIR@/rogue .
.sp
.BR IMPORTANT\ NOTE :
The \fB$PATH\fP
environment variable is
.B NOT
used to find
.B rogue (6).
Therefore, the
.I rogue
argument needs to be a path.
.sp
The
.BR rogomatic (6)
command might be able to  play with versions 3.6, 5.2, 5.3, 5.4, 5.4.4, and 5.4.5.
.sp
You can run
.B rogomatic
on modified rogues using this option,
but be warned that the interface must be completely compatible with
one of the standard versions of
.B rogue
listed, and the version number
printed must be either "3.6," "5.2," "5.3", "5.4", or "5.4.5".
.sp
.BR IMPORTANT\ NOTE :
Only \fBrogue version 5.4.5\fP
as found in the GitHub repo:
.sp
.in +0.5i
.I https://github.com/lcn2/rogue5.4
.in -0.5i
.sp
has been tested with this version of
.BR rogomatic (6).
While other versions of rogue may work, only \fBrogue version 5.4.5\fP
with a minimum date of
.B 2026-08-03
is supported.
.TP
.BI \-G\  goodlvl
Set the "\fBgood game\fP" level to
.IR goodlvl .
.sp
A "\fBgood game\fP" is when the rogue reaches a dungeon level of at least
.IR goodlvl .
When a rogue has the \fIAmulet of Yendor\fP
at the end of the game, then the game is also considered a "\fBgood game\fP".
.sp
By default,
.I goodlvl
level is
.BR @GOODGAME@ .
.sp
When a "\fBgood game\fP" exits, the rogomatic game log file:
.sp
.in +0.5i
.BI @RGMDIR@ /gamelog
.in -0.5i
.sp
is preserved by renaming it to a file of the form:
.sp
.in +0.5i
.BI @RGMDIR@ /goodgame.YYYYMMDD.HHMMSS.max_level.score.mnst
.in -0.5i
.sp
See the
.B FILES
section below for details on the goodname file form.
.TP
.B \-H
Disable the so-called rogomatic "\fIhalftime show\fP".
.sp
When
.B rogomatic
detects that
.B rogue
has reached level 2, it plays a short character based "movie" on the screen.
.sp
The use of
.B \-t
or
.B \-E
implies
.BR \-H .
.sp
Historically, this option was
.BR \-h ,
however recently that option became the "help option",
and so it was named as
.BR \-H .
.TP
.B \-p
Replay the rogomatic game log file.
.sp
The default rogomatic game log file is:
.sp
.in +0.5i
.BI @RGMDIR@ /gamelog
.in -0.5i
.sp
When combined with a final
.I r_file
command line argument,
the replay file will be
.IR r_file .
.TP
.BI \-P\  player
Set the path to the player program.
.TP
.B \-q
Set \fBquiet mode\fP, and disable
.BR ncurses (3X)
output.
.sp
By default, output from
.RB rogue (6),
along with additional messages from the
.B player
process are written to standard output so
that a human may watch the game as it is played.
The use of
.B \-q
will silence nearly all output, allowing in theory,
the game to be played faster.
.sp
When quite mode, all file logging continues as normal.
One may wish to tail a log file such as
.BR total.lvllog ,
or even the more verbose
.BR gamelog
to track progress.
Such files are under the rogomatic directory.
When using the
.BR tail (1)
utility, one would use both the \fB\-f\fP, and the \fB\-F\fP command line options:
.sp
.in +0.5i
\fBtail \-f \-F \fP\fIRGMDIR\fP\fB/total.lvllog\fP
.in -0.5i
.sp
where \fIRGMDIR\fP is the rogomatic directory
(such as \fB@RGMDIR@\fP or \fB/var/tmp/rogo\fP or \fB/var/tmp/mrogo/rogo.\fP\fInumber\fP).
.sp
The use of
.B \-q
overrides and disables any use of user mode (disables \fB\-u\fP),
and use of watched mode (disables \fB\-w\fP).
.sp
See the
.B FILES
section for more details on log files.
.TP
.B \-r
Restore a saved rogue game.
.sp
Without a final
.I r_file
command line option, run
.B rogue
with the
.BR \-r \ @RGMDIR@/rogue.sav
command line option and restore
the rogue game stored in that file.
.sp
If a final
.I r_file
command line argument is given, run
.BR rogue (6)
will be run with
.BI \-r \ r_file
so that rogue game will be restored from
.IR r_file .
.TP
.BI \-S\  ROGOSEED
Seed the pseudo-random number generator to \fIROGOSEED\fP
for both
.BR rogue (6)
and rogomatic.
.sp
When combined with
.B \-d
this creates much more deterministic run from a debugging standpoint.
.TP
.B \-t
Terse mode.
In terse mode,
.B rogomatic
will only print a summary after each level explored.
.TP
.B \-u
Start up in user mode.
This tells
.B rogomatic
to start up in transparent mode, so the user can type commands before
.B rogomatic
starts exploring level 1.
.sp
The use of \fB\-q\fP disables user mode, and overrides this flag.
.TP
.BI \-U\  usec
Set the sleep time between actions to
.I usec
microseconds.
.sp
Between writes to
.BR rogue (6)
the player program will sleep
.I usec
microseconds.
This has the effect of slowing down the actions on the screen to be somewhat viewable by humans.
.sp
To have no such delays, use \fB\-H \-U 0\fP on the command line.
.sp
By default,
.I usec
is set to
.B @USLEEP@
microseconds.
.TP
.B \-w
Start up in watched mode.
.sp
The use of \fB\-q\fP disables watch mode, and overrides this flag.
.TP
.B \-s\  [\ \fIrogue_ver\fP\ ]
Print the \fBrogomatic scoreboard\fP, reverse sorted by the ending gold amount,
of the rogomatic game results.
.sp
The \fBrogomatic scoreboard\fP is printed with the following fields:
.sp
.RS
.RS
.nr step 0 1
.nr PI 3n
.IP \n+[step].\ Year 0.5i
End of game 4-digit year (YYYY), in the local timezone
.IP \n+[step].\ Month 0.5i
End of game 3-letter month (Mon), in the local timezone
.IP \n+[step].\ Day 0.5i
End of game 2-digit day (01-31), in the local timezone
.IP \n+[step].\ Time 0.5i
End of game time of day (24-hour HH:MM:SS format, in the local timezone
.IP \n+[step].\ User 0.5i
rogue username that will start with \fRrogo-\fP.
.IP \n+[step].\ Gold 0.5i
End of game gold that will include the 10% penalty when killed,
or payment for inventory items if a total winner
.IP \n+[step].\ Killed\ by 0.5i
Cause of death, or "\fRtotal winner\fP
.IP \n+[step].\ Level 0.5i
The deepest dungeon level reached
.IP \n+[step].\ Hp 0.5i
End of game points
.IP \n+[step].\ Str 0.5i
End of game strength
.IP \n+[step].\ Ac 0.5i
End of game armor class that was being worn
.IP \n+[step].\ Level/Exp 0.5i
End of game experience level followed by / experience points
.IP \n+[step].\ Game\ number 0.5i
The logged game number, starting with 0
.IP \n+[step].\ Dungeon 0.5i
The rogue dungeon number
.RE
.RE
.sp
.RS
The default
.B rogue_ver
is
.BR 5.4.5 .
.sp
The default rogomatic scoreboard is:
.sp
.in +0.5i
.BI @RGMDIR@/rgmscore 5.4.5
.in -0.5i
.sp
If
.BI \-s\  rogue_ver
is given, then the rogomatic scoreboard printed will be from:
.sp
.in +0.5i
.BI @RGMDIR@/rgmscore rogue_ver
.in -0.5i
.sp
where
.B rogue_ver
is the
.B rogue
version number, such as
.IR 5.4.5 .
.sp
If a rogomatic delta file is found:
.sp
.in +0.5i
.BI @RGMDIR@/rgmdelta rogue_ver
.in -0.5i
.sp
The rogomatic delta is merged into the rogomatic scoreboard,
and then the rogomatic delta file is removed.
.sp
If just the
.BR \-s\  command
line option is used, then the
.BR \-s\  command
line option
.BR MUST\ BE\ THE\ LAST\ OPTION\ ON\ THE\ COMMAND\ LINE ,
and cannot be used in conjunction with a final
.IR r_file .
.RE
.TP
[ \fIr_file\fP ]
When used with
.BR \-p ,
.I r_file
specifies the rogomatic game to replay from.
.sp
When used with
.BR \-r ,
.I r_file
specifies the rogue game to restore from.
.sp
Use of
.I r_file
is \fBNOT\fP compatible with using
.B \-s
by itself.
.SH USEFUL UTILITIES
Several useful utilities are installed under the
.B @BINDIR@
directory,
or may be run from top of the source code directory.
.TP
.B run_rogo
The
.B run_rogo
command is a very useful way to run
.BR rogomatic (6)
and set the rogomatic directory to:
.sp
.in +0.5i
.BI /var/tmp/rogo
.in -0.5i
.sp
instead of the usual default rogomatic directory.
By changing the rogomatic directory to a non-default value,
manual runs of
.BR rogomatic (6)
will not interfere (such a lock file contention).
.sp
In addition, the
.B run_rogo
command will first look in the local directory for commands, and use them if found,
otherwise it search and use commands along the \fB$PATH\fP environment variable path.
.sp
See \fBrun_rogo -h\fP for usage details.
.TP
.B rerun_rogo
The
.B rerun_rogo
command is a very useful to run
.BR rogomatic (6)
multiple times in a row.
Currently, when the rogue dies in the dungeon,
.BR rogomatic (6)
exits.
When run with
.B rerun_rogo
a new
.BR rogomatic (6)
will be started after the previous
.BR rogomatic (6)
exits.
.sp
The
.B rerun_rogo
uses the
.B run_rogo
command to run
.BR rogomatic (6).
As such, the rogomatic directory will be set to:
.sp
.in +0.5i
.B /var/tmp/rogo
.in -0.5i
.sp
If the file
.B .stopfile
is created in the rogomatic directory while
.B rerun_rogo
is running, then the next time
.BR rogue (6)
exits, the
.B .stopfile
file is removed, and rerun cycle will be stopped.
This is a graceful way to end the rerun cycle.
.sp
Because
.B rerun_rogo
changes the default rogomatic directory to \fB/var/tmp/rogo\fP, the default way to
top the rerun cycle is to use the command:
.sp
.in +0.5i
.B touch /var/tmp/rogo/.stopfile
.in -0.5i
.sp
Changing the rogomatic directory by supplying
.BI \-D\  rgmdir
implies
.BI \-s\  rgmdir/.stopfile
as well.
.sp
To deal with the possibility of
.BR rogomatic (6)
hanging, and to speed up the automated game play,
we recommend
.B NO
relay between writes to
.BR rogue (6),
as well as a timeout timer:
.sp
.in +0.5i
.B rerun_rogo\ \-H\ \-U\ 0\ \-a\ 4.5
.in -0.5i
.sp
To deal with the possibility of
.BR rogomatic (6)
hanging, we recommend running
.B unstuck_player
in a separate window while running this tool:
.sp
.in +0.5i
.B unstuck_player\ \-v\ 1
.in -0.5i
.sp
See the section on \fBBUGS\fP below for more information on hanging.
.sp
See \fBrerun_rogo -h\fP for usage details.
.TP
.B mrerun_rogo
The
.B mrerun_rogo
command is a very useful to run multiple instances of the
.B rerun_rogo
command.
.sp
The
.B mrerun_rogo
command will establish a rogomatic directory of the form:
.sp
.in +0.5i
.B /var/tmp/mrogo/rogo.ID
.in -0.5i
.sp
where \fIID\fP is the argument supplied to the
.B rerun_rogo
command.
.sp
An instance of the
.B unstuck_player
utility with
.BI \-D\  /var/tmp/mrogo/rogo.ID
being set.
.sp
Then the
.B mrerun_rogo
command will be run with the same
.BI \-D\  /var/tmp/mrogo/rogo.ID
being set.
.sp
Any other options, except
.BI \-O\  rerun_rogo
will be passed down to the subsequent utilities as needed.
.TP
.B end_player
The
.B end_player
command will attempt to end a
.BR rogomatic (6)
and
.BR rogue (6)
run.
.sp
It will first attempt to send a
.B SIGHUP
signal to
.BR rogue (6)
by way of the
.BR killall (1)
command.
This allows
.BR rogue (6)
to gracefully exit, and potentially record a score
in the rogue game file.
The cause of death will be reported as "\fIby signal\fP".
When the happens,
.BR rogomatic (6)
will consider the rogue as having undergone a normal death
and will update the rogomatic state accordingly.
.sp
See \fBend_player -h\fP for usage details.
.TP
.B unstuck_player
The
.B unstuck_player
command attempt to detect if a
.BR rogue (6)
program is stalled by observing that the process isn't accumulating CPU time
by checking the output of a
.BR ps (1)
command.
The common reason for a stalled
.BR rogue (6)
program is that
.BR rogomatic (6)
has hung.
.sp
See the section on \fBBUGS\fP below for more information on hanging.
.sp
We recommended that a minor level of verbosity be used with this tool, as in:
.sp
.in +0.5i
.B unstuck_player\ \-v\ 1
.in -0.5i
.sp
When a
.BR rogue (6)
program has not accumulated CPU time after a period of time
(default: after 16 seconds), a
.B SIGHUP
signal is set to the
.BR rogue (6)
program.
.sp
See
.B end_player
above for what happens when a
.B SIGHUP
signal is set to the
.BR rogue (6)
program.
.sp
If a
.B SIGHUP
signal cannot be delivered to a
.BR rogue (6)
program, say because no such program is found, then a
.B SIGHUP
signal is set to the
.B player
program.
.sp
See \fBunstuck_player -h\fP for usage details.
.PP
The following tools are less well documented and less well supported:
.TP
.B rgmgene
The
.B rgmgene
command summarizes the current population in the gene pool.
.TP
.B rgmscatter
The
.B rgmscatter
command prints a scatter plot of all scores in the scorefile.
.TP
.B rgmhist
The
.B rgmhist
command prints a histogram of score (or levels if the
.B -l
option is given).
.SH COMMANDS
The following commands may be entered from the keyboard while
.BR rogomatic (6)
is running.
.PP
Some useful keyboard commands are:
.TP
.B t
.I Toggle,
toggles the setting of the
.I transparent
mode flag.
In transparent mode,
.B rogomatic
will not automatically attack monsters, pick up treasure, or explore
around.
Unfortunately, there is no way to enter complicated commands
like
.I quaff
or
.I read
in transparent mode.
Typing any
.B rogue
command, or single-step command also enables transparent mode.
.sp
The
.B t
command
will cause
.B rogomatic
will enter transparent mode.
.sp
In transparent mode, you can type several commands, the
most useful of which is the
.B ?
command, which prints out the next help line.
Press the
.B ?
key repeatedly to see all of the help.
Some useful commands are:
.B h
.B j
.B k
.B l
.B H
.B J
.B K
.B L
.B b
.B n
.B y
.B u
.B f
.B s
.B >
.B <
.B S
.B Q.
These have the same meaning as in normal
.BR rogue .
To return to
.B rogomatic
mode, type the
.B t
command again.
.TP
.B Q
.I Quit,
tells
.B rogomatic
to quit the game.
The score is logged in the score file, and possibly the
.B rogue
top ten, and the game is terminated.
.TP
.B S
.I Save,
the game is saved to the default save file, set by the
.I <file>
argument to
.I rogomatic
or the
.I ROGUEOPTS
environment variable.
The game may be restarted later.
.TP
.B c
.IR creative\ mode ,
tells
.B rogomatic
to take advantage of idiosyncrasies in the
.B rogue
game to get a higher score.
.sp
Note that this option should only be used on very old 3.6 version
.BR rogue ,
since the VMS version does not exhibit the same bugs.
.sp
The strategies enabled by
.I creative\ mode
are known and used by a fair number of human players.
.TP
.B d
.I Debugging,
toggles the debugging flag, which is useless to non-wizard
users (although it is entertaining if your terminal has reverse video).
.TP
.B e
.I Toggle\ game\ logging.
.sp
By default rogomatic game logging is enabled (unless the
.B \-e
command line option was used).
.sp
When enable, the rogomatic game log file is:
.sp
.in +0.5i
.BI @RGMDIR@ /gamelog
.in -0.5i
.sp
If the rogomatic game logging was enabled the
.B e
command will disable it; otherwise
rogomatic game logging will be re-enabled.
.sp
The
.B -p
option will replay a saved game.
Additionally, a stop action command
(carriage return gives a single step) has been added.
.TP
.B i
.I Inventory,
dumps the current inventory on the screen.
.TP
.B m
.I Monster list,
shows the current long term memory evaluation of the monsters.
.B rogomatic
nows keeps a long term memory file (for each version of
.BR rogue )
in the
score directory, and records the meanness of each monster it finds.
This command prints out the letter, name, and strength of each monster.
Strength is listed as three numbers: expected damage (scaled by 10) and
maximum damage (unscaled) are shown in parentheses, and the mean number
of arrows required to kill a monster is shown in square brackets.
.TP
.B CR
.I Single step,
a carriage return enables
.I transparent
mode, and causes
.B rogomatic
to single-step.
The command sent is printed at the
bottom right hand of the screen.
.TP
.B /
.I Snapshot,
a timestamp, the current screen, and the current inventory are appended
to file
.I snapshot.rgm
in the current directory.
.TP
.B $
.I Weaponrank,
dumps
.BR rogomatic 's
current estimate of his weapons rankings.
.TP
.B %
.I Armorrank,
dumps
.BR rogomatic 's
current estimate of his armor rankings.
.TP
.B =
.I Ringrank,
dumps
.BR rogomatic 's
current estimate of his ring rankings.
.TP
.B -
.I Status,
prints out a status line indicating various things like whether
.BR rogomatic 's
is confused, has red hands, is wearing cursed armor or wielding a cursed
weapon, and which monsters have been genocided.
It also prints the number of turns elapsed in the game.
.TP
.B `
.I Summary,
a backquote prints a short summary of the game.
.TP
.B ~
.I Version,
prints the version of
.B rogue
and
.B rogomatic
being used.
It also prints the current goal score.
.TP
.B G
.I Genotype,
displays the settings of the genetically learned parameters of the
genotype playing now.
See section
.I Genes
for more information about these parameters.
.TP
.B R
.I Replay position,
allows you to skip around in a log file when replaying an old game.
This key must be followed by either
.B f, p, c, n,
or
.B l.
These indicate you wish to see the
.I first, previous, current, next,
or
.B last
levels of the game.
The first time you enter the
.B R
command, there will be a pause of several seconds while the whole log
file is digested.
After this initialization, there should be no delay
when skipping around in the log file.
After you have skipped once, the
program won't terminate at the end of the file until you enter the
.B Q
.I (quit)
ommand.
This way you can replay the last level over (and over...).
.IR
.PP
When the game ends, either by your death, when you quit, or when
.B rogomatic
dies horribly of some bug, you will not get a list of the top-ten scorers.
The scoring is based entirely upon useless criteria anyway.
.B rogomatic
does keep a list of its games.
To see the best
.B rogomatic
games, use "rogomatic -s" which dumps the current scores in reverse order.
.PP
For more detailed directions, as the expression goes:
.B RTFS
(\fIhttps://en.wikipedia.org/wiki/RTFS\fP) :-).
See:
.sp
.in +0.5i
.I https://www.github.com/lcn2/rogomatic
.in -0.5i
.SH GENES
The
.B G
command lists the current settings for each
.BR rogomatic 's
``DNA'' settings.
These numbers (from 0 to 99) specify some of
.BR rogomatic 's
behaviors.
A gene pool file is kept in the score directory that
records the current population of explorers.
A variant of Holland's
``Genetic Algorithm'' is used to manipulate this gene pool.
The result
is that
.B rogomatic
learns from his past games how to play better.
The following genes are
currently used:
.TP
.B Sr
How many times to search each square for traps (0 to 4, depending on
amount of food and how far below level 7 he is).
.TP
.B Dr
How many times to search each square for secret doors (0 to 19, more for
dead end passages not in mazes).
.TP
.B Re
How much to rest up when injured (until 40% to 80% healed, based on
experience level).
.TP
.B Ar
When to shoot arrows at enemies, the higher the number, the more likely
.B rogomatic
is to shoot arrows at monsters (both smaller monsters, and more arrows).
.TP
.B Ex
How soon to experiment with unknown potions and scrolls (between levels
1 and 10).
.TP
.B Rn
How likely he is to run from an opponent (the higher the number, the sooner
he will run).
.TP
.B Wk
How likely he is to wake a sleeping opponent (the higher the number, the
bigger the monster he will attack, even though it is sleeping).
.TP
.B Fd
How likely he is to hoard extra food.
This affects how many rings he
wears, and how much he searches for traps.
.PP
.SH FILES
There are two executable files:
.B rogomatic
and
.BR player .
The
.B rogomatic
program is a small program which forks and execs the
.B rogue
game connected to the
.B player
program.
You must have access to all of these files.
.PP
These programs are searched for in the current directory.
If they are not found in the current directory, then, the
.B rogomatic
program
will look for then under the
.B @BINDIR@
directory.
.PP
By default,
.B rogomatic
creates
.B @RGMDIR@
known as the rogomatic directory.
.PP
The rogomatic directory may be changed by the command line option
.BR \-D \ path
command line option.
.PP
If the
.B \-d
command line option is used, a new rogomatic directory
is created and used in sub-directory of the form:
.sp
.in +0.5i
.BI @RGMDIR@ /YYYYMMDD_HHMMSS
.in -0.5i
.sp
where
.I YYYYMMDD_HHMMSS
is the year, month, day, hour, minute and second, in the UTC timezone,
of when player program started.
.TP
.B debuglog.frogue
This file:
.sp
.in +0.5i
.BI @RGMDIR@/debuglog.frogue
.in -0.5i
.sp
records the standard output (stdout) from the current
.BR rogue (6)
process.
When
.BR rogomatic (6)
starts, this file is truncated to a zero size.
.sp
The player process when starting, will create this file if it does not exist,
or truncate it to zero size otherwise.
Thus, this file only holds data about the most recent run.
.TP
.B debuglog.player
When
.BR rogomatic (6)
launches player, this file:
.sp
.in +0.5i
.BI @RGMDIR@/debuglog.player
.in -0.5i
.sp
records tokens (output) written to the current
.BR rogue (6)
process.
.sp
The player program when starting, will create this file if it does not exist,
or truncate it to zero size otherwise.
Thus, this file only holds data about the most recent run.
.TP
.B errlog
.br
Output written to standard error (stderr) is appended to:
.sp
.in +0.5i
.BI @RGMDIR@/errlog
.in -0.5i
.sp
If this file does not exist, it is created, otherwise it is appended to by the player program.
Thus, this file may contain data about previous runs.
.TP
.B error5.4.5
A log of internal errors and bugs is kept in:
.sp
.in +0.5i
.BI @RGMDIR@/error version
.in -0.5i
.sp
where
.I version
is the
.B rogue
version number, such as
.IR 5.4.5 .
.sp
If this file does not exist, it is created, otherwise it is appended to by the player program.
Thus, this file may contain data about previous runs.
.TP
.B gamelog
This file is the rogomatic game log file:
.sp
.in +0.5i
.BI @RGMDIR@/gamelog
.in -0.5i
.sp
Lines that begin with
.I C:
record commands sent to
.BR rogue (6)
by
.BR rogomatic (6).
Lines that begin with
.I R:
record the responses sent from
.BR rogue (6)
to
.BR rogomatic (6).
.sp
By using the
.B \-p
command line option, one may replay such a rogomatic game log.
When combined with a final
.I r_file
command line argument, the replay file will be
.IR r_file .
.sp
When a "\fBgood game\fP" exits, the rogomatic game log is renamed to a file of the form:
.sp
.in +0.5i
.BI @RGMDIR@ /goodgame.YYYYMMDD.HHMMSS.max_level.score.mnst
.in -0.5i
.sp
See
.BI \-G\  goodlvl
in the
.B OPTIONS
section for information on what constitutes a "\fBgood game\fP".
.sp
The player program when starting, will create this file if it does not exist,
or truncate it to zero size otherwise.
Thus, if this file exists, only holds data about the most recent run
if the run wasn't a "\fBgood game\fP".
.TP
.B GeneLock545
Before
.BR rogomatic (6)
and
player writes a gene pool related file,
this file is locked:
.sp
.in +0.5i
.BI @RGMDIR@/GeneLoc version
.in -0.5i
.sp
where
.I version
is the
.B rogue
version number, without dots, such as
.IR 545 .
This file is locked via the
.BR flock (2)
system call.
.sp
The player program when starting, will create this file if it does not exist.
This is a zero length file.
.TP
.B GeneLog545
A log of all mutations and crossovers is kept in:
.sp
.in +0.5i
.BI @RGMDIR@/GeneLog version
.in -0.5i
.sp
where
.I version
is the
.B rogue
version number, without dots, such as
.IR 545 .
.sp
If this file does not exist, it is created, otherwise it is appended to by the player program.
Thus, this file may contain data about previous runs.
.TP
.B GenePool545
The gene pool is kept in:
.sp
.in +0.5i
.BI @RGMDIR@/GenePool version
.in -0.5i
.sp
where
.I version
is the
.B rogue
version number, without dots, such as
.IR 545 .
.sp
If this file does not exist, it is created, otherwise it is appended to by the player program.
Thus, this file may contain data about previous runs.
.TP
.B goodgame.YYYYMMDD.HHMMSS.max_level.score.mnst
When
.BR rogomatic (6)
co
then the
.B gamelog
will be renamed under the rogomatic directory as
.sp
.in +0.5i
.BI @RGMDIR@ /goodgame.YYYYMMDD.HHMMSS.max_level.score.mnst
.in -0.5i
.sp
where
.I YYYYMMDD_HHMMSS
is the year, month, day, hour, minute and second, in the UTC timezone,
and
.I max_level
is the maximum dungeon level that was reached,
and
.I score
is the final rogue score,
and
.I mnst
is the first 4 characters of how the rogue died, or "\fItota\fP" if the rouge was a total winner.
.sp
This file is only created, by renaming
.BR gamelog ,
when a "\fBgood game\fP" ends.
See
.BI \-G\  goodlvl
in the
.B OPTIONS
section for information on what constitutes a "\fBgood game\fP".
.sp
.TP
.B ltm545
.br
The long term memory of monster attributes is kept in:
.sp
.in +0.5i
.BI @RGMDIR@/ltm version
.in -0.5i
.sp
where
.I version
is the
.B rogue
version number, without dots, such as
.IR 545 .
.sp
If this file does not exist, it is created, otherwise it is update by the player process.
Thus, this file may contain data about previous runs.
.TP
.B lvllog
.br
Each time player detects that a new level has been reached,
and when player exits (due to rogue death, player quit, or
even total winner), this file:
.sp
.in +0.5i
.BI @RGMDIR@/lvllog
.in -0.5i
.sp
will be appended with a line containing the following space separated fields:
.RS
.RS
.nr step 0 1
.nr PI 3n
.IP \n+[step].\ timestamp 0.5i
microseconds since the epoch
.IP \n+[step].\ UTC\ date 0.5i
YYYY-MM-DD format
.IP \n+[step].\ UTC\ time 0.5i
time of day in HH:MM:DD format
.IP \n+[step].\ HEADER:\ Level: 0.5i
a string ending in a colon \:
.IP \n+[step].\ level\ number 0.5i
dungeon level number
.IP \n+[step].\ HEADER:\ Gold: 0.5i
a string ending in a colon \:
.IP \n+[step].\ gold\ collected 0.5i
amount of Gold collected so far
.IP \n+[step].\ HEADER:\ Hp: 0.5i
a string ending in a colon \:
.IP \n+[step].\ hit\ points 0.5i
current hit points followed by ( maximum hit points )
.IP \n+[step].\ HEADER:\ Str: 0.5i
a string ending in a colon \:
.IP \n+[step].\ strength 0.5i
current strength followed by ( maximum strength )
.IP \n+[step].\ HEADER:\ Ac: 0.5i
a string ending in a colon \:
.IP \n+[step].\ armor\ class 0.5i
current armor class
.IP \n+[step].\ HEADER:\ Exp: 0.5i
a string ending in a colon \:
.IP \n+[step].\ experience 0.5i
experience level followed by / experience points
.IP \n+[step].\ HEADER:\ Am: 0.5i
a string ending in a colon \:
.IP \n+[step].\ amulet\ status 0.5i
- ==> amulet not found , ==> \fIAmulet of Yendor\fP found
.IP \n+[step].\ end\ status 0.5i
... ==> still in dungeon, else ==> end game status
.RE
.RE
.RE
.sp
.RS
The player program, when starting, will create this file if it does not exist,
or truncate it to zero size otherwise.
Thus, this file only holds data about the most recent run.
.RE
.TP
.B pidlog
.br
Early on in player, this file:
.sp
.in +0.5i
.BI @RGMDIR@/pidlog
.in -0.5i
.sp
is appended with a line containing the following space separated fields:
.RS
.RS
.nr step 0 1
.nr PI 3n
.IP \n+[step].\ timestamp 0.5i
microseconds since the epoch
.IP \n+[step].\ UTC\ date 0.5i
YYYY-MM-DD format
.IP \n+[step].\ UTC\ time 0.5i
time of day in HH:MM:DD format
.IP \n+[step].\ rogue-pid: 0.5i
a string ending in a colon \:
.IP \n+[step].\ rogue\ PID
rogue game process ID
.IP \n+[step].\ HEADER:\ player-pid: 0.5i
a string ending in a colon \:
.IP \n+[step].\ player\ PID
player executable process ID
.IP \n+[step].\ HEADER:\ dungeon: 0.5i
a string ending in a colon \:
.IP \n+[step].\ dnum
dungeon number
.RE
.RE
.sp
.RS
If this file does not exist, it is created, otherwise it is appended to by the player program.
Thus, this file may contain data about previous runs.
.RE
.TP
.B player.lck
When
.BR rogomatic (6)
launches, this file:
.sp
.in +0.5i
.BI @RGMDIR@/player.lck
.in -0.5i
.sp
is locked via
.BR flock (2)
and then it released.
If the file cannot be locked, then
.BR rogomatic (6)
will exit.
The purpose of this initial temporary lock is an attempt to detect if a player
program has completed its initial setup and is running.
.sp
When
.BR rogomatic (6)
executes the player program, early on this file is locked via
.BR flock (2)
and is held until the program exits.
This lock helps prevent more than one
.BR rogomatic (6)
from using the same rogomatic directory.
.sp
The player program, when starting, will create this file if it does not exist.
This is a zero length file.
.TP
.B Rgm.Lock
Before
.BR rogomatic (6)
writes to a critical file (e.g.,
.I ltm545
or
.I rgmdelta5.4.5
)
this file:
.sp
.in +0.5i
.BI @RGMDIR@/rgmdelta5.4.5
.in -0.5i
.sp
is locked via the
.BR flock (2)
and then is released after the write is complete.
.sp
The player program, when starting, will create this file if it does not exist.
This is a zero length file.
.TP
.B rgmdelta5.4.5
The rogomatic delta file is:
.sp
.in +0.5i
.BI @RGMDIR@/rgmdelta version
.in -0.5i
.sp
where
.I version
is the
.B rogue
version number, such as
.IR 5.4.5 .
.sp
When the
.B \-s
command line option is used to print the rogomatic scoreboard,
the rogomatic delta file is merged into the rogomatic score file
(the rogomatic score file is created if needed), and then the
rogomatic delta file is removed before printing the result.
.sp
If this file does not exist, it is created, otherwise it is appended to by the player process.
Thus, this file may contain data about previous runs.
.TP
.B rgmscore5.4.5
The rogomatic score file is kept in:
.sp
.in +0.5i
.BI @RGMDIR@/rgmscore version
.in -0.5i
.sp
where
.I version
is the
.B rogue
version number, such as
.IR 5.4.5 .
.sp
If this file does not exist, it is created when the
.B \-s
command line option is used.
Thus, this file may contain data about previous runs.
.TP
.B rogue.lck
A lock file used by
.BR rogue (6)
to manage writing to its local files.
.sp
When
.BR rogomatic (6)
launches
.BR rogue (6),
it does so in such a way that the program's default path is changed to:
.sp
.in +0.5i
.BI @RGMDIR@/rogue.lck
.in -0.5i
.sp
If this file does not exist, it is created by
.BR rogue (6).
This is a zero length file.
.TP
.B rogue.scr
A score file containing the top scores of
.BR rogue (6).
.sp
When
.BR rogomatic (6)
launches
.BR rogue (6),
it does so in such a way that the program's default path is changed to:
.sp
.in +0.5i
.BI @RGMDIR@/rogue.sav
.in -0.5i
.sp
To view the rogue score file:
.sp
.in +0.5i
\fBrogue -s @RGMDIR@/rogue.lck\fP
.in -0.5i
.sp
If this file does not exist, it is created by
.BR rogue (6).
.TP
.B saved_termattr
Before
.BR rogomatic (6)
prepares the screen for
.BR ncurses (3X)
managed I/O (such as raw terminal mode), it saves the terminal state into the file:
.sp
.in +0.5i
.BI @RGMDIR@/saved_termattr
.in -0.5i
.sp
On exit,
.BR rogomatic (6)
will use the contents of this file to the previous (hopefully) sane terminal state.
.sp
While running
.BR rogue (6)
will create this file, and as it is exiting, the file will be removed.
.TP
.B total.lvllog
Each time player detects that a new level has been reached,
and when player exits (due to rogue death, player quit, or
even total winner), this file:
.sp
.in +0.5i
.BI @RGMDIR@/total.lvllog
.in -0.5i
.sp
See
.B lvllog
above for a description of the contents of this file.
.sp
Unlike
.BR lvllog ,
this file isn't truncated at the start of each run.
.sp
If this file does not exist, it is created, otherwise it is appended to by the player process.
Thus, this file may contain data about previous runs.
.SH SEE ALSO
.BR flock (2),
.BR ncurses (3X),
.BR rogue (6)
.SH SOURCE CODE
The source code for
.B rogomatic
is available is available at:
.PP
.in +0.5i
.I https://github.com/lcn2/rogomatic
.in -0.5i
.PP
in the form of a GitHub repo.
.PP
The above GitHub repo is maintained by:
.B Landon\ Curt\ Noll
( GitHub user
.BR @lcn2 :
.IR https://github.com/lcn2 \ )
with help from the
GitHub user @jenoola ( \fIhttps://github.com/jenoola\fP ).
.PP
The port to ISO C 2017 with GNU extensions, the port to various platforms
such as NetBSD, Linux, and Cygwin, and many rogue bug fixes applied
by Landon Curt Noll user \fB@lcn2\fP, and the GitHub user \fB@jenoola\fP.
.SH BUGS
It’s easy to imagine that the original authors were more focused on crafting
a scholarly paper rather than on creating well written code.
.sp
A number of people have attempted to improve this code.
Many, many bug fixes have been applied to this code.
.B Landon\ Curt\ Noll
alone has a change log of mostly bug fixes, that in mid 2026,
exceeded 1500 lines of change description text!
Even so, a \fBsignificant number of bugs\fP remain in this code.
.sp
Bug fixes in the form of GitHub pull requests are \fBvery welcome\fP at:
.sp
.in +0.5i
.I https://github.com/lcn2/rogomatic/pulls
.in -0.5i
.sp
You are likely to notice that
.BR rogomatic (6)
will occasionally hang.
Often the code will hang when the state machine, expect a response from
.BR rogue (6)
encounters an unexpected message (e.g., announcing a new level,
reporting hunger, coming under attack, removal of a temporary impairment, etc.)
.sp
Occasionally
.B rogomatic
will dead-lock waiting for input from
.BR rogue .
When this happens, try to type
.I t
in an effort to toggle
.I transparent
mode.
If that is successful, type
.I t
again in the hopes of resuming the run.
.sp
Sometimes
you will have to use transparent mode to get it down to the next level
before turning it loose again.
.sp
In some hanging cases, try interrupting (usually by typing control C),
and then entering the
.I S
command to save the game.
Once saved, re-start with "rogomatic -r" and then \fBhope\fP that the restart will be OK.
.sp
One work-a-round to hanging is to use the
.BI \-a\  secs
command line option, such as using:
.BI \-a\  4.5
on the command line.
.sp
When the
.BI \-a\  secs
command line option is used, then before
.BR rogomatic (6)
attempts to read from
.BR rogue (6),
a timeout alarm will be set.
If the read hangs for more than \fIsecs\fP seconds, a
.B SIGARLM
signal will be delivered to the player process.
The
.B SIGARLM
is then caught and a
.BR siglongjmp (3)
performed a non-local jump to a control point where player process
is much more likely to resume normal operation.
Nevertheless, there are circumstances where non-local jump still hangs.
.sp
While less often,
.BR rogomatic (6)
will enter a loop, such as cycling between two positions, or
cycling between pick up and dropping an item.
The above mentioned use of keyboard intervention may help, if you catch the loop in time.
Without some form of intervention, the rogue will eventually die of starvation.
As with real life, death tends to interrupt the flow of things.  :-)
.SH DEBUGGING
See the
.B BUGS
section above for why you are welcome to debug
.BR rogomatic (6).
.sp
See also any open GitHub issues by visiting:
.sp
.in +0.5i
.I https://github.com/lcn2/rogomatic/issues
.in -0.5i
.sp
If you wish to debug
.BR rogomatic (6)
we recommend compiling with the
.BR clang (1)
compiler, and debugging with the
.BR lldb (1)
debugger as these tools do well to handle to debugging across
.B fork (2)
system calls, and cross process executions such as making
.B execl (2)
system calls.
.sp
You should first recompile
.BR rogomatic (6)
with full debugging and C tags:
.sp
.in +0.5i
.B make clobber clang tags
.in -0.5i
.sp
When debugging, you should use the
.B \-d
command line option to always create a new rogomatic sub-directory
so that the rogomatic the environment starts out in an initial state.
Attempting to debug an observed problem in a previous run set
(such as what will happen when using the
.B rerun_rogo
command)
will be problematic because
.BR rogomatic (6)
is designed to evolve its game play.
By using the
.B \-d
command line option an observed behaviour is more likely to be reproducible,
and a more likely to be successfully debugged.
.sp
Using the
.B \-d
command line option will cause an accumulation of rogomatic sub-directories,
so you might wish to clean out old rogomatic sub-directories
from time to time by doing:
.sp
.in +0.5i
.B rm -rf /var/tmp/rogomatic/2*
.in -0.5i
.sp
Using the
.BR \-S\  seed
command line option will cause the
.BR random (3)
pseudo-random number generators to be seeded with a fixed value via a
.BR srandom (3)
call.
By default, both
.BR rogomatic (6)
and
.BR rogue (6)
with a seed that changes on each new invocation.
By using the
.BR \-S\  seed
command line option you force both programs to seed with the same value each time.
This will make behaviour is more likely to be reproducible,
and a more likely to be successfully debugged.
.sp
By using the
.B \-H
and the
.BI \-U\  0
command line option, you won't slow down the rogomatic operations,
allowing one to debug deeper dungeon conditions sooner.
.sp
The
.I seed
value can be any integer between 0 and 2147483647.
For example:
.sp
.in +0.5i
.B ./rogomatic \-H -U 0 \-d \-S 1989014291
.in -0.5i
.sp
Because
.BR rogomatic (6)
makes extensive use of ncurses, and because puts the terminal into raw mode,
simply running the command from within the debugger will very problematic.
Both
.BR rogomatic (6)
and your debugger will fight for control of the terminal, and you will
.B NOT
be the winner.
Moreover, the use of ncurses by dueling programs will produce a mess on the screen.
Fortunately
.BR lldb (1)
offers a clean way around this problem.
.sp
Start the debugger on
.BR rogomatic (6)
with both the
.B \-d
and the
.BR \-S\  seed
command line option.
For example:
.sp
.in +0.5i
.B lldb -- ./rogomatic \-H -U 0 \-d \-S 3
.in -0.5i
.sp
Once the debugging has started, set a breakpoint somewhere with the player process:
.sp
.in +0.5i
.B b getrogue
.in -0.5i
.sp
Don't worry about the debugger complaining about not being able resolve breakpoint to any actual locations.
That will resolve itself later on.
.sp
Next, launch the process in a separate terminal window.
This will cause another terminal window to pop-up:
.sp
.in +0.5i
.B process launch --tty
.in -0.5i
.sp
.BR NOTE :
Depending on your terminal manager, keyboard focus may shift to the new terminal window pop-up.
You may have the click back on original terminal window so that your keyboard actions don't
get consumed by the new terminal window.
.sp
Have the debugger continue the process:
.sp
.in +0.5i
.B c
.in -0.5i
.sp
Best wishes in debugging.
If you find a bug fix, please consider submitting a GitHub pull requests are \fBvery welcome\fP at:
.sp
.in +0.5i
.I https://github.com/lcn2/rogomatic/pulls
.in -0.5i
.sp
.SH HISTORY
For an outdated, and historic history of some changes, see the
.I ORIG_DOC/old.history.txt
file in the source directory.