xref: /dports/news/husky-areafix/husky-areafix-1.9.20191207/hpt/doc/hpt.texi

\input texinfo   @c -*-texinfo-*-

@c %**start of header
@c $Id$
@c adopted from msged.texi, written by Tobias Ernst
@setfilename hpt.info
@set HPT HPT
@set MSGED MsgEd-TE
@set VERSION 1.9
@settitle @value{HPT} @value{VERSION} - Highly Portable Tosser
@setchapternewpage odd
@c %**end of header

@titlepage
@title @value{HPT} @value{VERSION} manual
@subtitle Highly Portable Tosser
@author Matthias Tichy @@ 2:2432/645 (1997 - 1999)
@author Max Levenkov @@ 2:5000/117 (1999 - 2002)
@page
@c{empty page to get the page numbering right}
@end titlepage

@contents

@ifinfo
@dircategory Fidonet Software
@direntry
* HPT: (hpt).               HPT is the fidonet tosser
@end direntry
@end ifinfo

@iftex
@parindent=0pt
@end iftex
@iftex
@afourpaper
@end iftex

@node Top, Overview, (dir), (dir)
@top @value{HPT}
This document describes @value{HPT} @value{VERSION}, a Fidonet Message
Tosser for OS/2, Windows, BeOS and Unix.

@value{HPT} is based on FIDOCONFIG library, so read documentation of
FIDOCONFIG about location of config file, keywords ideology and about
majority of the keywords,
@xref{Top, Fidoconfig Manual, , fidoconfig, Fidoconfig manual}.

@menu
* Overview::                    What is @value{HPT}?
* Installation::                Installation. @emph{Read this!}
* Configuration Reference::     Explanation of the configuration file.
* Advanced Concepts::           Advanced features and concepts.
* Keyword Index::               Configuration File Keywords Index
@end menu

@node Overview, Installation, Top, Top
@chapter An Overview on @value{HPT}

@value{HPT} is a Fidonet message tosser and packer with areafix, written
by Matthias Tichy, 2:2432/645@@fidonet. Now project is being supported by
Husky Development Team.

Features of @value{HPT}:
@enumerate
@item tossing packets of 2, 2.0 & 2+ types
@item supporting Msg, Squish and Jam message bases
@item posting to net & echo areas
@item areafix (on the fly, from command line, limit for areas...)
@item autocreate on the fly
@item forward requests
@item pause and autopause for links
@item linking net & echo areas
@item carbon copy
@item groups & levels for personal and public access to echo areas
@item powerful dupe checker
@item link defaults
@end enumerate

The advantages of @value{HPT} are:

@enumerate
@item Open Source (GPL)
@item many supported platforms & operating systems
@item quick bug fixing
@end enumerate

The limitations of @value{HPT} is:

@enumerate
@item OPUS messagebase (type MSG) is limited to 65536 messages, this is SMAPI
implementation limit;
@item PKT password is limited to 8 characters, this is PKT 2+ limit.
@end enumerate


@node Installation, Configuration Reference, Overview, Top
@chapter Installation Procedures and Release Notes
This chapter provides you with information that is necessary to successfully
install and use @value{HPT}.

I suppose, that you already has compiled binaries. If not - read "Download"
or "Compile the Source Code" chapters.

@enumerate
@item Read FIDOCONFIG documentation about location of config-file
and majority of the configuration statements.
@item Copy sample config files to config directory
@item Edit config files for your purposes
@item Run @code{tparser} from FIDOCONFIG package to test your config
(read about @code{PublicGroup} or @code{AccessGrp} if you want to use
groups for @code{EchoAreas})
@item It is simply, isn't it? Enjoy! :-)
@end enumerate

@menu
* Download::            Download the Source Code & Binary Files
* Compiling::           How to compile the source code
* Support::             Support, Contacting the Author, Reporting Bugs, Contributing Code
* Credits::             Credits
@end menu

@node Download, Compiling, Installation, Installation
@section Download the Source Code & Binary Files

@table @asis
@item Main page - @url{http://husky.sourceforge.net/}
@item Latest win32 binaries - @url{http://elfy.nv.googlepages.com/}
@end table

@node Compiling, Support, Download, Installation
@section Compiling the Source Code

1. The smapi and fidoconf packages are required for hpt.

2. Put the fidoconfig and smapi packages in the directory where the
   other packages of fido linux reside:

@table @asis
@item /usr/src/packages/
@item -> smapi/
@item -> hpt/
@item -> fidoconfig/
@end table

3. Compile and install smapi and fidoconf packages. Use "Makefile" for
   dynamic executables and makefile.lnx (or what you need) for static
   ones.

4. Compile and install hpt:
@table @asis
@item $ make
@item $ make install
@end table
You should use the _same_ makefiles in smapi, fidoconf and hpt.

@node Support, Credits, Compiling, Installation
@section Support, Contacting the Author, Reporting Bugs, Contributing Code

There are numerous reasons why you might wish to establish contact with me,
the author of @value{HPT}.

@enumerate
@item
You have decided to use @value{HPT} on a regular basis. In this case,
please do send me an e-mail at the address listed below. How much time I will
spend on developing @value{HPT} in the future will heavily depend on the
number of mails I receive from users that tell me that they do use
@value{HPT}.

@item
You have a general questions on how to configure or on how to use a certain
feature of @value{HPT}. In other words, you need support. In this case,
you'd best post your question to one of the following echos:

@table @code
@item FIDOSOFT.HUSKY
The international Husky conference. English is the preferred language here.
@item RU.HUSKY
This Russian echo covers Husky Project. I monitor it regularly.
@item RU.ECHOPROCESSORS
Russian talks about echoprocessors. I monitor it regularly.
@end table

If you do not have access to any of these echos, you may of course also
contact me via netmail or e-mail at the addresses listed below.

@item
You want to report a bug. There are two sorts of bugs:

@enumerate a
@item Normal bugs.
You think that a certain function of @value{HPT} does not work as expected,
e.g. it is producing garbage, or doing strange things, or similar. In this
case, either post to the echos listed above, or contact me via
netmail. Please do supply all information that is necessary to understand
your problem.

@item Fatal bugs.
A fatal bug occurs if @value{HPT} crashes. Depending on your operating
system, the symptom might be a core dump, or a SYS 3175, or a general
protection fault, or a system lockup, or a spontaneous reboot. I do consider
a crash untolerable. No matter how stupid things you do, you should not be
able to crash @value{HPT}.

If you are an experienced user and get core dump, you can send me gdb
report. If you have a crash, locate @file{core} file that has been
generated. Then run $ gdb hpt core, type where. @value{HPT} must be
compiled with debug information (DEBUG=1 in @file{huskymak.cfg} file).
Then send report to address below.

If you are running any other binary version (like Windows), you will not
get a core file on a crash. Write down as much information as you can,
try to find a way to reproduce the crash and contact me at the addresses
below.
@end enumerate

@item You want to contribute to @value{HPT}.
If you are a programmer and have fixed a problem in @value{MSGED} on your
own, please submit your changes to me. The preferred way for doing this is to
send me a difference file in GNU diff format (with -c parameter). Your
work will be highly appreciated and honored in an appropriate place. If
you want to regularly work on @value{HPT}, we also have a CVS server
online that you can have access to if you like.

If you want to write a new feature for @value{HPT}, please contact me
beforehand to avoid that we do duplicate work. Again, I will appreciate and
honor eny efforts done by you. Please note that for writing a @value{HPT}
enhancement, you should be familiar with C. Also, @value{HPT} uses a
special indentation style throughout the source code, that I would like
you to adhere to.
@end enumerate

So here are my addresses if you want to get in contact with me:

@itemize @bullet
@item Fidonet:
Max Levenkov @ 2:5000/117

@item e-mail:
sackett@@mail.ru

@end itemize

@node Credits,  , Support, Installation
@section Credits

Thanks to:

@enumerate
@item Matthias Tichy (basic code)
@item Max Levenkov (areafix, group code, carbon copy, code clean up & speed up and more)
@item Tobias Ernst (some features, stable releases, cvs & www support)
@item Sacha M. Silbe (features, makefiles)
@item Kolya Nesterov (unpack, post, link)
@item Fedor Lizunkov (areafix, autocreate reports, group code, JAM code in SMAPI)
@item Serguei Revtov (win32 version support, hptlink, features & good patches)
@item Alexander Vernigora (dupe detection)
@item Pavel Gulchouk (Perl support, JAM & link hacking)
@item Alex Semenyaka (netmailExtern, processPkt)
@item Alex Bronin (win32 patches)
@item Nikolay Nestyurkin (makefiles)
@item Lev Serebryakov (new %list)
@item ... and all other people (sorry, if I've forget somebody)
@end enumerate
P.S. all this people make bug fixes, so I don't mention it.

@node Configuration Reference, Advanced Concepts, Installation, Top
@chapter @value{HPT} @value{VERSION} Configuration Reference

@value{HPT} is based on FIDOCONFIG library, so read documentation of
FIDOCONFIG about location of config file, keywords ideology and about
majority of the keywords,
@xref{Top, Fidoconfig Manual, , fidoconfig, Fidoconfig manual}.

@menu
* Using config::                Notes about fidoconfig usage in hpt
* Keywords::                    This chapter describes keywords
* Files and Paths::             File & Path keywords
* Link Keywords::               Links configuration
* Carbon Copy::                 Carbon Copy keywords
@end menu

@node Using config, Keywords, Configuration Reference, Configuration Reference
@section Using config

Although HPT uses fidocongig, not all config statements refer to hpt. Hpt
ignores all unused config statements.The Husky Fidonet Software project
concept is to use one config file for all fidonet programs.


@subsection Predefined config variables

In some cases need use different values for one statement in hpt, htick
and/or other husky programs (called @dfn{modules}). It solve using config
variable @code{module}: hpt define the @code{module} variable with value
@code{hpt} and you may use if-elseif-else-endif conditions statements
sequence like as:

@example
if "[module]"=="hpt"
  Origin   High Portable Tosser at my node
elseif "[module]"=="htick"
  Origin   High Portable Ticker at my node
else
  Origin   My node
endif
@end example

HPT define config variable @code{version} with value set to hpt version
like @code{MAJOR.MINOR.PATHLEVEL}, i.e. @code{@value{VERSION}.0} or
@code{1.4.1}.
You may use this variable too for make common config for several hpt versions.

To test config for hpt only you may call tparser tool with parameters
@code{-Dmodule=hpt} and (if need) @code{-Dversion=@value{VERSION}.0}
(tparser included into fidoconfig package) like this:

@example
tparser -Dmodule=hpt
tparser -Dmodule=hpt -Dversion=@value{VERSION}.0
@end example


@subsection Statements order

Common rule for config statements order is:
@strong{Define object before using it!}.

Examples:
@itemize
@item
define addresses before any link, area and carbon definitions
@item
define packers before any link definitions
@item
define links before any area and route definitions
@item
define links and areas before any carbon definitions
@end itemize


@node Keywords, Files and Paths, Using config, Configuration Reference
@section General Keywords

@value{HPT} is based on FIDOCONFIG library, so read documentation of
FIDOCONFIG about location of config file, keywords ideology and about
majority of the keywords,
@xref{Top, Fidoconfig Manual, , fidoconfig, Fidoconfig manual}.

@menu
* AddToSeen::                   add addr to seen-by's
* AfterUnpack::                 execute string after unpack arcmail bundles
* AreafixFromPkt::              process areafix requests on the fly
* AreasFileNameCase::           case of filenames of autocreated areas
* AutoPassive::                 set pause for links who don't poll our system
* BeforePack::                  execute string before pack pkt files
* BundleNameStyle::             style of filenames of arcmail bundles
* DefArcmailSize::              default size for arcmail bundle
* DisableTID::                  disable adding TID to messages
* DisablePID::                  disable adding PID to messages
* DisableKludgeRescanned::      disable adding kludge @@RESCANNED to messages
* IgnoreCapWord::               ignore Capability Word in pkt files
* IgnoreSeen::                  ignore SEEN-BYs and pack mail for link
* KeepTrsFiles::                keep transit files
* KeepTrsMail::                 keep transit netmail messages
* KludgeAreaNetmail::           how to process netmail with kludge "AREA:NETMAIL"
* LinkWithImportLog::           list of areas only which should be linked
* NetmailFlag::                 create flag after unpacking netmail msg
* NoProcessBundles::            no unpack arcmail bundles
* Origin::                      add specified origin to hpt messages
* PackNetMailOnScan::           pack netmail found in EchoTossLog file when scan
* ProcessPkt::                  execute string before tossing pkts
* RecodeMsgBase::               allow recoding of message base
* Remap::                       remap mail to other address
* Route::                       routing of netmails
* RouteFile::                   routing of netmails with file attached
* RouteMail::                   routing of netmails without file attached
* SeparateBundles::             move echomail for each link to his own directory
* SetConsoleTitle::             schange conesole title under win32
* SortEchoList::                echolist sort mode
* TearLine::                    add specified tearline to hpt messages
* TossingExt::                  extension of mail files while tossing
@end menu

@node AddToSeen, AfterUnpack, Keywords, Keywords
@subsection AddToSeen
@findex AddToSeen
@table @asis
@item Syntax:
@code{addToSeen <addr> [<addr> ...]}
@item Example:
@code{addToSeen 99/100 99/101}
@end table

Add <addr> to SEEN-BYs. If <addr> is present or not - it is always adds.

This statement can be repeated.

@node AfterUnpack, AreafixFromPkt, AddToSeen, Keywords
@subsection AfterUnpack
@findex AfterUnpack
@table @asis
@item Syntax:
@code{afterUnpack <string>}
@item Example:
@code{afterUnpack pktpack /home/fido/in.tmp/*.pkt}
@end table

This <string> is executed after unpacking arcmail bundle. You may process
your pkt files in @code{tempInbound} directory with external utility.

This statement cannot be repeated.

@node AreafixFromPkt, AreasFileNameCase, AfterUnpack, Keywords
@subsection AreafixFromPkt
@findex AreafixFromPkt
@table @asis
@item Syntax:
@code{areafixFromPkt <bool>}
@item Example:
@code{areafixFromPkt}
@end table

Process areafix requests on the fly. Check "areafix", "areamgr" &
"hpt" keywords in toUserName field. Request messages don't saves.

This statement cannot be repeated.

@xref{RobotNames,,,fidoconfig, fidoconfig}.

@node AreasFileNameCase, AutoPassive, AreafixFromPkt, Keywords
@subsection AreasFileNameCase
@findex AreasFileNameCase
@table @asis
@item Syntax:
@code{areasFileNameCase (Lower|Upper)}
@item Example:
@code{areasFileNameCase Upper}
@end table

This statement defines case of filemanes of autocreated areas.
Default is lower case.

This statement cannot be repeated.

@node AutoPassive, BeforePack, AreasFileNameCase, Keywords
@subsection AutoPassive
@findex AutoPassive
@table @asis
@item Syntax:
@code{autopassive <bool>}
@item Example:
@code{autopassive}
@end table

If this statement is defined HPT will check for old bundles in every run.
Fileboxes are not checked. Links, for which there are bundles that are
at least "@ref{AutoPause}" days old, will be paused. It stops exporting
echomail for them. If @ref{AutoPause} is not defined or is equal to 0 for a link
then AutoPassive is ignored for this link. If AutoPassive is not defined
then @ref{AutoPause} is ignored for all links.

The paused link is unsubscribed from the areas which are passthrough with no
downlinks besides him (see -def in @ref{EchoArea,,,fidoconfig,fidoconfig} to make it work)
AND which are allowed to be paused
(see -pause, -paused, -noautoareapause, -autoareapause in @ref{EchoArea,,,fidoconfig,fidoconfig}
and @ref{AutoAreaPause,,,fidoconfig,fidoconfig}).
An unsubscribe request is sent to the uplink for those areas. Note that
this feature is unavailable when using %PAUSE Areafix command.

By using "hpt pause" command-line parameter instead of defining AutoPassive in
the configuration file, you can do the check for old bundles and all the rest
of the functionality of AutoPassive only once.

This statement cannot be repeated.

See also @ref{AutoPause} and @ref{Pause,,,fidoconfig,fidoconfig} sections.

@node BeforePack, BundleNameStyle, AutoPassive, Keywords
@subsection BeforePack
@findex BeforePack
@table @asis
@item Syntax:
@code{beforePack <string>}
@item Example:
@code{beforePack pktpack /home/fido/out.tmp/*.pkt}
@end table

This <string> is executed before packing pkt files to arcmail bundles. You
may process your pkt files in @code{tempOutbound} directory with
external utility.

This statement cannot be repeated.

@node BundleNameStyle, DefArcmailSize, BeforePack, Keywords
@subsection BundleNameStyle
@findex BundleNameStyle
@table @asis
@item Syntax:
@code{bundleNameStyle <timeStamp | addrDiff | addrsCRC32 | addrDiffAlways | addrsCRC32Always | Amiga>}
@item Example:
@code{bundleNameStyle timeStamp}
@end table

This statement sets a rule for creating names of arcmail bundles:

@table @asis
@item @code{timeStamp}
some kind of random names (current time + some counter).
@item @code{addrDiff}
create name from difference of source and destination addreses. Always
the same filename for every pair of links. Automatically  switches to
@code{timeStamp} when all alowed extensions have been used.
@item @code{addrsCRC32}
create name from CRC32 of string composed from "hpt", source and destination
addreses. Always the same filename for every pair of links, very unique.
Automatically switches to @code{timeStamp} when all alowed extensions have
been used.
@item @code{addrDiffAlways}
the same as addrDiff but tries to use lower free numbers of extension in
case when all higher numbers are already busy.
@item @code{addrsCRC32Always}
the same as addrsCRC32 but tryes to use lower free numbers of extension in
case when all higher numbers are already busy.
@item @code{Amiga)}
Amiga Style Outbound (ASO). Bundles are created in @ref{outbound,,,fidoconfig,FidoConfig Manual} directory
like this:
@example
2.5000.117.1.hut
2.5000.117.1.mo0
@end example
If @ref{SeparateBundles} is @code{on} and @ref{packer,,,fidoconfig,FidoConfig Manual} not defined, pkt is moved to
@code{2.5000.117.1.sep} directory.

Extensions & prefixes in flo files are created as in @code{addrDiffAlways} algorithm.
@end table

Default value is @code{timeStamp}

This statement cannot be repeated.

See also @ref{LinkBundleNameStyle}.


@node DefArcmailSize, DisableTID, BundleNameStyle, Keywords
@subsection DefArcmailSize
@findex DefArcmailSize
@table @asis
@item Syntax:
@code{defarcmailsize <integer>}
@item Example:
@code{defarcmailsize 1024}
@end table

default arcmail size in kb for all links.
500kb if not defined.

This statement cannot be repeated.

@node DisableTID, DisablePID, DefArcmailSize, Keywords
@subsection DisableTID
@findex DisableTID
@table @asis
@item Syntax:
@code{DisableTID <bool>}
@item Example:
@code{DisableTID}
@end table

Don't add TID-line to scanned messages.

This statement cannot be repeated.

@node DisablePID, DisableKludgeRescanned, DisableTID, Keywords
@subsection DisablePID
@findex DisablePID
@table @asis
@item Syntax:
@code{DisablePID <bool>}
@item Example:
@code{DisablePID}
@end table

Don't add PID-line to generated messages.

This statement cannot be repeated.


@node DisableKludgeRescanned, IgnoreCapWord, DisablePID, Keywords
@subsection DisableKludgeRescanned
@findex DisableKludgeRescanned
@table @asis
@item Syntax:
@code{DisableKludgeRescanned <bool>}
@item Example:
@code{DisableKludgeRescanned}
@end table

Don't add kludge RESCANNED to generated messages.

This statement cannot be repeated.


@node IgnoreCapWord, IgnoreSeen, DisableKludgeRescanned, Keywords
@subsection IgnoreCapWord
@findex IgnoreCapWord
@table @asis
@item Syntax:
@code{IgnoreCapWord <bool>}
@item Example:
@code{IgnoreCapWord}
@end table

Ignoring Capability Word in pkt files. If some pkt moved to bad. This
may help, but not recommended. It is better to change old software.

This statement cannot be repeated.

@node IgnoreSeen, KeepTrsFiles, IgnoreCapWord, Keywords
@subsection IgnoreSeen
@findex IgnoreSeen
@table @asis
@item Syntax:
@code{IgnoreSeen <addr> [<addr> ...]}
@item Example:
@code{IgnoreSeen 99/150}
@end table

Ignore this SEEN-BY & pack mail for link.
But no pack it back if the mail was from him.

This statement can be repeated.

@node KeepTrsFiles, KeepTrsMail, IgnoreSeen, Keywords
@subsection KeepTrsFiles
@findex KeepTrsFiles
@table @asis
@item Syntax:
@code{keepTrsFiles <bool>}
@item Example:
@code{keepTrsFiles}
@end table

Leave transit (routed) files in @code{Inbound} directory. If this
statement is ommitted transit files wouldn't be kept.
Default is "off".

File route possible with "Att" attribute in message header and file name
in subject line.

This statement cannot be repeated.

@xref{RouteFile}.

@node KeepTrsMail, KludgeAreaNetmail, KeepTrsFiles, Keywords
@subsection KeepTrsMail
@findex KeepTrsMail
@table @asis
@item Syntax:
@code{keepTrsMail <bool>}
@item Example:
@code{keepTrsMail}
@end table

Save transit messages in NetmailArea. If this statement is ommitted
transit messages wouldn't be kept in NetmailArea. Default is "off".

This statement cannot be repeated.


@node KludgeAreaNetmail, LinkWithImportLog, KeepTrsMail, Keywords
@subsection KludgeAreaNetmail
@findex KludgeAreaNetmail
@table @asis
@item Syntax:
@code{kludgeAreaNetmail <kill | ignore | echomail>}
@item Example:
@code{kludgeAreaNetmail kill}
@end table

Default is "kill"

If message started with "AREA:NETMAIL" we have three ways:

1. kill this kludge. process message as netmail.
2. ignore this kludge. process message as netmail.
3. process message as echomail.

This statement cannot be repeated.

@node LinkWithImportLog, NetmailFlag, KludgeAreaNetmail, Keywords
@subsection LinkWithImportLog
@findex LinkWithImportLog
@table @asis
@item Syntax:
@code{linkWithImportLog <yes | no | kill>}
@item Example:
@code{linkWithImportLog yes}
@end table

This statement specifies if the importlog-file should be used to determine
which echomail areas need to be linked.

@table @asis
@item yes
importlog-file will be read. areas which are in importlog, will be linked.
the importlog-file will not be erased.
@item kill
like yes, but the importlog-file will be killed after using it.
@item no
DEFAULT. all areas will be linked.
@end table

This statement cannot be repeated.

@node NetmailFlag, NoProcessBundles, LinkWithImportLog, Keywords
@subsection NetmailFlag
@findex NetmailFlag
@table @asis
@item Syntax:
@code{netmailFlag <file>}
@item Example:
@code{netmailFlag /etc/ftn/flags/netmail}
@end table

Create file-flag after unpacking netmail msg. This feature can be used
for execute netmail trackers after tossing.

This file also created after scannig @code{NetmailArea} without route
definition. Scanning is stopped but file-flag created for netmail
tracker.

This statement cannot be repeated.

@node NoProcessBundles, Origin, NetmailFlag, Keywords
@subsection NoProcessBundles
@findex NoProcessBundles
@table @asis
@item Syntax:
@code{noProcessBundles <bool>}
@item Example:
@code{noProcessBundles}
@end table

Don't unpack arcmail bundles.

This statement cannot be repeated.

@node Origin, PackNetMailOnScan, NoProcessBundles, Keywords
@subsection Origin
@findex Origin
@table @asis
@item Syntax:
@code{origin <string>}
@item Example:
@code{origin mega cool station}
@end table

Add this Origin to hpt messages: post, reports about new areas created.

This statement cannot be repeated.

@node PackNetMailOnScan, ProcessPkt, Origin, Keywords
@subsection PackNetMailOnScan
@findex PackNetMailOnScan
@table @asis
@item Syntax:
@code{PackNetMailOnScan <bool>}
@item Example:
@code{PackNetMailOnScan off}
@end table

When PackNetMailOnScan is "on" (default) hpt packs netmail when doing
"hpt scan" and netmail area is found in EchoTossLog file. When it is
"off" hpt leaves netmail area(s) in EchoTossLog file until "hpt pack" is
invoked.

This statement can be repeated (overrides old setting).

@node ProcessPkt, RecodeMsgBase, PackNetMailOnScan, Keywords
@subsection ProcessPkt
@findex ProcessPkt
@table @asis
@item Syntax:
@code{processPkt <string>}
@item Example:
@code{processPkt pktdate}
@end table

The <string> is executed before tossing each pkt file. You can
fix your pkts using pktdate or any other tool before tossing them. Note that
pkt file may be renamed depending of @code{tossingExt} token value.

This statement cannot be repeated.

@node RecodeMsgBase, Remap, ProcessPkt, Keywords
@subsection RecodeMsgBase
@findex RecodeMsgBase
@table @asis
@item Syntax:
@code{recodeMsgBase <bool>}
@item Example:
@code{recodeMsgBase off}
@item Defaults:
@code{recodeMsgBase on}
@end table

This statement specifies if we should recode messages when put/get them
to/from message base. So you can keep message base in transport charset and
configs/other local files in local one.

This statement cannot be repeated.

@node Remap, Route, RecodeMsgBase, Keywords
@subsection Remap
@findex Remap
@table @asis
@item Syntax:
@code{Remap <ToName>,<ToAddress>,<NewAddress>}
@item Examples:
@itemize
@item @code{Remap Max Levenkov,2:5000/117,2:5000/117.1\n}
@item @code{Remap *,2:5000/117,2:5000/117.1}
@item @code{Remap Max Levenkov,*,2:5000/117.1}
@end itemize
@end table

Remap mail to other address.
ToName @strong{OR} ToAddress can be replaced with asterisk to match any
value of ToName or ToAddress. Two asterisks can't be defined together.
@strong{This statement does not touch TOPT & FMPT kludges.}

This statement can be repeated.

@node Route, RouteFile, Remap, Keywords
@subsection Route
@findex Route
@table @asis
@item Syntax:
@code{route <flavour> <routeVia> <target> [<target> ...]}@*
@code{route nopack <target> [<target> ...]}
@item Example:
@code{route crash 2:2433/1245 2:2433/* 2:2432/*}@*
@code{route nopack 2:5020/98*}
@end table

The @strong{route} statement defines a route to destination address namely
the node address via which we send the netmail containing the specific
destination address.

There are two variants of the statement. The first one defines a route to
<target>. The second one with 'nopack' (or 'no-pack') just after the
'route' keyword tells not to process netmail when its destination coinsides
with an address in <target>.

@strong{<target>} is one specific destination address or a DOS-style
pattern containing characters @strong{?} and/or @strong{*} and defining a set of destination
addresses.

@strong{<flavour>} is an instruction to mailer how it should handle the netmail
and it can be one of the following keywords:
@itemize @bullet
@item hold
@item normal
@item crash
@item direct
@item immediate
@end itemize

If the flavour is 'hold', then the mailer should transfer the mail during
an incoming session only. So the mail will be placed into outbound directory
and it will wait for an incoming session. If the flavour differs from 'hold',
then the mail may be transferred both during an incoming and an outgoing
session. Interpretation of all flavours except 'hold' depends on the mailer.

The file extension of the netmail file or the flow file which the tosser puts
into the outbound depends on the flavour:

@multitable @columnfractions .46 .30 .24
@headitem flavour @tab netmail @tab flow
@item hold
@tab .hut
@tab .hlo
@item normal
@tab .out
@tab .flo
@item crash
@tab .cut
@tab .clo
@item direct
@tab .dut
@tab .dlo
@item immediate
@tab .iut
@tab .ilo
@end multitable

@strong{<routeVia>} defines the node via which the mail will be sent to
destination. It may take the following values:
@itemize @bullet
@item <address> -- mail is routed to destination via this specific address.
@item host -- when this keyword is used, mail will be transmitted via the
node whose address is formed from the destination address with node and a
point numbers equated to zero.
@item hub -- this keyword denotes hub-routing, which does not use the hubs
defined in the nodelist but uses the node address defined by formula:

@code{via_address = (destination_address / 100) * 100}

Here '/ 100' denotes integer division discarding the remainder. E.g. a
mail to node 2:2433/1245 is sent via 2:2433/1200, but a mail to node
2:2433/355 is sent via 2:2433/300 which currently does not exist. BE CAREFUL!
@item boss -- this keyword is used when destination system is a point.
If the destination address is zone:net/node.point then mail is routed
via zone:net/node.0
@item no-route (or noroute) -- this keyword is a misnomer in a way and
in fact it means route to destination via itself.
@end itemize

One may use the second variant of the route statement in the form

@code{route <flavour> nopack <target> [<target> ...]}

Here the <flavour> is ignored.  Such use is deprecated and it is left for
compatibility with old versions.

There can be several route statements. They are parsed in the order they
are written until the statement with the target containing the destination
address from the netmail is found.

@strong{NOTE!} This statement must occur after "link" sections.

@node RouteFile, RouteMail, Route, Keywords
@subsection RouteFile
@findex RouteFile
@table @asis
@item Syntax:
@code{routeFile <flavour> <routeVia> <target> [<target> ...]}@*
@code{routeFile nopack <target> [<target> ...]}
@item Example:
@code{routeFile crash 2:2433/1245 2:2433/* 2:2432/*}@*
@code{routeFile nopack 2:5020/98*}
@end table

This statement is the same as the @ref{Route} statement, but applies
only to messages with file attaches. Files are routed together with netmail
messages. If no @code{RouteFile} defined, then files will be not routed at all!

This statement can be repeated.

@node RouteMail, SeparateBundles, RouteFile, Keywords
@subsection RouteMail
@findex RouteMail
@table @asis
@item Syntax:
@code{routeMail <flavour> <routeVia> <target> [<target> ...]}@*
@code{routeMail nopack <target> [<target> ...]}
@item Example:
@code{routeMail crash 2:2433/1245 2:2433/* 2:2432/*}@*
@code{routeMail nopack 2:5020/98*}
@end table

This statement is the same as the @ref{Route} statement.

This statement can be repeated.

@node SeparateBundles, SetConsoleTitle, RouteMail, Keywords
@subsection SeparateBundles
@findex SeparateBundles
@table @asis
@item Syntax:
@code{SeparateBundles <bool>}
@item Example:
@code{SeparateBundles}
@end table

Move echomail for all links to their own directories.

This statement cannot be repeated.

@node SetConsoleTitle, SortEchoList, SeparateBundles, Keywords
@subsection SetConsoleTitle
@findex SetConsoleTitle
@table @asis
@item Syntax:
@code{SetConsoleTitle <bool>}
@item Example:
@code{SetConsoleTitle}
@end table

Set hpt console title while tossing. (@strong{WIN32 Only!})

This statement cannot be repeated.

@node SortEchoList, TearLine, SetConsoleTitle, Keywords
@subsection SortEchoList
@findex SortEchoList
@table @asis
@item Syntax:
@code{SortEchoList none|name|group|group,name}
@item Example:
@code{SortEchoList name}
@end table

This keyword determines sorting mode for echolist (%LIST, %QUERY, etc):
@table @asis
@item @code{none}
unsorted (order as presents in config file)
@item @code{name}
sort by name
@item @code{group}
sort by group (-g in echoArea) only
@item @code{group,name}
sort by group and name.
@end table

When sorting is carried out by group, group descriptions
(see grpDesc keyword) are printed before groups' areas.

By default echolist is sorted by name.

This statement cannot be repeated.


@node TearLine, TossingExt, SortEchoList, Keywords
@subsection TearLine
@findex TearLine
@table @asis
@item Syntax:
@code{tearline <string>}
@item Example:
@code{tearline We Love HPT! :)}
@end table

Add this tearline to hpt messages (post, reports about new areas created)

This statement cannot be repeated.

@node TossingExt,  , TearLine, Keywords
@subsection TossingExt
@findex TossingExt
@table @asis
@item Syntax:
@code{TossingExt [<string>]}
@item Example:
@code{TossingExt tos}
@end table

Extension of bundle & packet files will be changed to <string> before
tossing (and before processPkt string executed if set). That may be used
for preventing of permanent tossing fault because of bad file in inbound.
Default extension: tos.
@code{TossingExt} without parameter disables files renaming in tossing.

This statement cannot be repeated.

@node Files and Paths, Link Keywords, Keywords, Configuration Reference
@section Files and Paths

@value{HPT} is based on FIDOCONFIG library, so read documentation of
FIDOCONFIG about location of config file, keywords ideology and about
majority of the keywords,
@xref{Top, Fidoconfig Manual, , fidoconfig, Fidoconfig manual}.

@menu
* AdvStatisticsFile::           binary statistic file
* AreasMaxDupeAge::             max age for dupes in CommonDupeBase
* DupeBaseType::                type of dupe base
* DupeHistoryDir::              path for dupe files
* EchoTossLog::                 file which contains the names of the areas to scan
* FileBoxesDir::                directory where link fileBoxes are created
* HptPerlFile::                 file which contains perl filter functions
* ImportLog::                   file which contains the names of the areas tossed in
* Intab::                       file for recoding incoming msgs
* MinDiskFreeSpace::            minimum free space on the drive
* MsgBaseDir::                  base directory of areas
* NotValidFileNameChars::       chars that will be converted to hex values
* Outtab::                      file for recoding outgoing msgs
* RulesDir::                    directory with area rules
* StatLog::                     file whith counter of personal net- and echo mails
* TempInbound::                 temporary inbound path
* TempOutbound::                temporary outbound path
@end menu

@node AdvStatisticsFile, AreasMaxDupeAge, Files and Paths, Files and Paths
@subsection AdvStatisticsFile
@findex AdvStatisticsFile
@table @asis
@item Syntax:
@code{AdvStatisticsFile <filename>}
@item Example:
@code{AdvStatisticsFile /home/val/fido/log/hpt.sta}
@end table

Define binary statistic file for calculating links and echo traffic.
Parsed by hpt/misc/adv-stat-hpt.pl

@node AreasMaxDupeAge, DupeBaseType, AdvStatisticsFile, Files and Paths
@subsection AreasMaxDupeAge
@findex AreasMaxDupeAge
@table @asis
@item Syntax:
@code{areasMaxDupeAge <integer>}
@item Example:
@code{areasMaxDupeAge 10}
@end table

Set maximum days for storing you hashes in CommonDupeBase.
Default value is 5.

For any other dupe base type please use @code{-DupeHistory}
option in @ref{EchoArea,,,fidoconfig, fidoconfig} or
@ref{EchoAreaDefaults,,,fidoconfig, fidoconfig}.

This statement cannot be repeated.

@node DupeBaseType, DupeHistoryDir, AreasMaxDupeAge, Files and Paths
@subsection DupeBaseType
@findex DupeBaseType
@table @asis
@item Syntax:
@code{dupeBaseType <TextDupes | HashDupes | HashDupesWMsgId | CommonDupeBase>}
@item Example:
@code{dupeBaseType HashDupesWMsgId}
@end table

@table @asis
@item TextDupes
stores from, to, subj & msgid as text lines.
@item HashDupes
stores src32 of from + to + subj + msgid.
@item HashDupesWMsgId
same as HashDupes, but stores also msgid as text.
@item CommonDupeBase
stores hashes of from + to + subj + areatag + msgid in one file (hpt_base.dpa)
@end table

Default is @code{HashDupesWMsgId}.

This statement cannot be repeated.

@node DupeHistoryDir, EchoTossLog, DupeBaseType, Files and Paths
@subsection DupeHistoryDir
@findex DupeHistoryDir
@table @asis
@item Syntax:
@code{dupeHistoryDir <path>}
@item Example:
@code{dupeHistoryDir /var/spool/fido/dupes}
@end table

This command specifies the path where the dupe history files are stored.
The format and the names of the dupe-files are not standardized.

This statement cannot be repeated.

@node EchoTossLog, FileBoxesDir, DupeHistoryDir, Files and Paths
@subsection EchoTossLog
@findex EchoTossLog
@table @asis
@item Syntax:
@code{echotosslog <file>}
@item Example:
@code{echotosslog /var/spool/fido/echotoss.log}
@end table

This statement specifies the file which is filled by a message editor
or "hpt post" with the names of the areas where new netmails or echomails
have been entered. Each line contains one areaname.
When "hpt scan" is invoked, only echomail areas listed in this file will be
scanned, when we start "hpt pack" - the same is for netmail areas.
If area was processed, its name is removed from echotosslog file.
When hpt has finished processing all listed areas and this file became
empty, it is removed, otherwise kept for futher processing (maybe by other programs).
If this file is not found - all areas will be scanned
(depending on value of PackNetMailOnScan).
@xref{PackNetMailOnScan}, but if '-f' command line flag specified, scanning
will stop. Filename after -f is optional. If it is not set, value of
EchoTossLogFile will be taken from config, and from command line otherwise.
If scanning from EchoTossLogFile did not make any area scanning, all areas
will be processed (depending on 'scan' or 'pack' mode)

This statement cannot be repeated.

@node FileBoxesDir, HptPerlFile, EchoTossLog, Files and Paths
@subsection FileBoxesDir
@findex FileBoxesDir
@table @asis
@item Syntax:
@code{FileBoxesDir <directory>}
@item Example:
@code{FileBoxesDir ../boxes}
@end table

@xref{FileBoxesDir, , ,fidoconfig, FidoConfig Manual}, for full description.

@node HptPerlFile, ImportLog, FileBoxesDir, Files and Paths
@subsection HptPerlFile
@findex HptPerlFile
@table @asis
@item Syntax:
@code{hptperlfile <file>}
@item Example:
@code{hptperlfile /etc/ftn/filter.pl}
@end table

This statement specifies the file which contains perl filter functions.
If not specified, perl support will be switched off.

This statement cannot be repeated.

@node ImportLog, Intab, HptPerlFile, Files and Paths
@subsection ImportLog
@findex ImportLog
@table @asis
@item Syntax:
@code{importlog <file>}
@item Example:
@code{importlog /var/spool/fido/import.log}
@end table

This statement specifies the file which a tosser fills with the names
of the areas where echomails has been tossed in.

This statement cannot be repeated.

@node Intab, MinDiskFreeSpace, ImportLog, Files and Paths
@subsection Intab
@findex Intab
@table @asis
@item Syntax:
@code{intab <file>}
@item Example:
@code{intab /var/spool/fido/recode/outaltkoi8}
@end table

This statement specifies the file which should be used to recode the
characters of the incoming messages from transport to internal
charset. It is useful in russia.
If you do not use this statement no recoding will be done.

This statement cannot be repeated.

@node MinDiskFreeSpace, MsgBaseDir, Intab, Files and Paths
@subsection MinDiskFreeSpace
@findex MinDiskFreeSpace
@table @asis
@item Syntax:
@code{MinDiskFreeSpace <integer>}
@item Example:
@code{MinDiskFreeSpace 10}
@end table

This is the minimum disk free space in MB to run @value{HPT}.
The following directories are checked: @code{TempInbound},
@code{MsgBaseDir}, @code{LinkMsgBaseDir}s.

The default value is 10.

This statement cannot be repeated.

@node MsgBaseDir, NotValidFileNameChars, MinDiskFreeSpace, Files and Paths
@subsection MsgBaseDir
@findex MsgBaseDir
@table @asis
@item Syntax:
@code{msgBaseDir <path>}
@item Example:
@code{msgBaseDir /var/spool/fido/msgb}
@end table

This command specifies the path where msgBases of autocreated areas are
stored. For example: if an area called @code{LINUX.GER} was autocreated
and the msgBaseDir is @code{/var/spool/fido/msgb} the resulting
msgBaseName is

@example
/var/spool/fido/msgb/linux.ger.sqd
@end example

If you specify the msgbasedir as PASSTHROUGH, the areas will be created
as passthrough areas.

This statement cannot be repeated.

@xref{LinkMsgBaseDir}.

@node NotValidFileNameChars, Outtab, MsgBaseDir, Files and Paths
@subsection NotValidFileNameChars
@findex NotValidFileNameChars
@table @asis
@item Syntax:
@code{NotValidFileNameChars <string>}
@item Example:
@code{NotValidFileNameChars :;<>=*?\/}
@end table

This characters in message and dupebase filenames will be replaced
with %hex analogs. If not defined, default string used: "*/:;<=>?\|%`'&+

This statement cannot be repeated.

@node Outtab, RulesDir, NotValidFileNameChars, Files and Paths
@subsection Outtab
@findex Outtab
@table @asis
@item Syntax:
@code{outtab <file>}
@item Example:
@code{outtab /var/spool/fido/recode/outkoi8alt}
@end table

This statement specifies the file which should be used to recode the
characters of the outgoing messages from internal to transport
charset. It is useful in russia.
If you do not use this statement no recoding will be done.

This statement cannot be repeated.

@node RulesDir, StatLog, Outtab, Files and Paths
@subsection RulesDir
@findex RulesDir
@table @asis
@item Syntax:
@code{rulesdir <path>}
@item Example:
@code{rulesdir /home/ftn/rules}
@end table

This statement specifies where areafix searches for files with echoarea
rules which are sent to link via netmail when link is subscribed to
areas. File name is the same as message base file with trailing
".rul" (and consistently .ru1-.ru9 files with additional information).
Areaname is used when no message base file is set.
See also @code{NoRules} statement to avoid areafix from sending rules to
link.

This statement cannot be repeated.

@node StatLog, TempInbound, RulesDir, Files and Paths
@subsection StatLog
@findex StatLog
@table @asis
@item Syntax:
@code{statlog <file>}
@item Example:
@code{statlog /var/spool/fido/stat.log}
@end table

After tossing (hpt toss), it is checked if there is new personal mail
(netmail or personal echo mail). If that's true, it is checked if
statlog is defined in config. If yes, then it is checked if the log file
exists. If not, then it is created and the number of received netmails
or/and personal echo mails is written to the log file. Otherwise, if the
file exists, the old counter will be read, added to the new counter and
written the actual counter.
The log file looks like this:
 netmail: x
 CC: x
Whereas x is the number of mails.
If the log file exists, one of the two or both lines exist.

This is for utils, which can show you how many personal mail you got.
The log file is not removed by hpt, only by the util that uses it.
Or you could write in your toss script something like this:
 if exist stat.log type stat.log
 if exist stat.log del stat.log>nul

This statement cannot be repeated.

@node TempInbound, TempOutbound, StatLog, Files and Paths
@subsection TempInbound
@findex TempInbound
@table @asis
@item Syntax:
@code{tempinbound <path>}
@item Example:
@code{tempinbound /var/spool/fido/in.tmp}
@end table

This command specifies a path which is used while tossing. The incoming packets
are unpacked there.

This statement cannot be repeated.

@node TempOutbound,  , TempInbound, Files and Paths
@subsection TempOutbound
@findex TempOutbound
@table @asis
@item Syntax:
@code{tempoutbound <path>}
@item Example:
@code{tempoutbound /var/spool/fido/out.tmp}
@end table

This command specifies your temporary outbound path. It is used for
storing outgoing pkt-files before packing.

This statement cannot be repeated.


@node Link Keywords, Carbon Copy, Files and Paths, Configuration Reference
@section Link Keywords

@menu
* AdvancedAreafix::             advanced areafix
* AllowEmptyPktPwd::            allow empty packet passwords
* AllowPktAddrDiffer::          allow addresses in PKT and MSG differ for areafix req.
* AllowRemoteControl::          allow remote control using areafix
* ArcmailSize::                 size of arcmail bundle
* ArcNetmail::                  pack netmail for this link into arcmail bundles
* Areafix::                     areafix access for this link
* AutoAreaCreateSubdirs::       create structured or flat message base?
* AutoPause::                   stop export mail for this link after xx days
* AvailList::                   specify variant for reply to %avail
* DailyBundles::                start a new arcmail bundle for the link each day
* DenyRescan::                  deny rescan of areas
* EchoMailFlavour::             flavour for outgoing echomail
* Flavour::                     flavour for netmail, echomail and fileechoes
* ForwardPkts::                 shall we forward inbound packets for this node?
* FileBox::                     filebox directory
* FileBoxAlways::               pack to filebox even if link is not busy
* LinkBundleNameStyle::         style of filenames of arcmail bundles for link
* LinkGrp::                     main group for the link
* LinkMsgBaseDir::              base directory of areas for link
* NetMailFlavour::              flavour for outgoing netmail
* NoRules::                     don't send area rules to link on subscription
* PackAka::
* PktSize::                     max size of pkt in arcmail bundle
* ReducedSeenBY::               reduced seen-by mode (FSC-0093)
* RescanGrp::                   groups that @code{denyRescan} token applies to
* RescanLimit::                 limit max number of messages sent by rescan
* SendNotifyMessages::          send messages about query changes, bad posts and so on
* UnsubscribeOnAreaDelete::     send unsubscribe request when subscribed area is deleted
@end menu
  @ref{AreafixReportsAttr,,,fidoconfig,fidoconfig}::          set flags to areafix reports for the link

@value{HPT} is based on FIDOCONFIG library, so read documentation of
FIDOCONFIG about location of config file, keywords ideology and about
majority of the keywords,
@xref{Top, Fidoconfig Manual, , fidoconfig, Fidoconfig manual}.


@node AdvancedAreafix, AllowEmptyPktPwd, Link Keywords, Link Keywords
@subsection AdvancedAreafix
@findex AdvancedAreafix
@table @asis
@item Syntax:
@code{advancedareafix <bool>}
@item Example:
@code{advancedareafix on}
@end table

If this statement is "on" and our system wants to delete an area from
remote config then unsubscribe messages to areafix robot of this link
look like "~areaname". The area on the remote system will be deleted
from config file, if our system is allowed to delete this area ("allow
area delete" in FastEcho, and @ref{LinkGrp} must be the same as
EchoArea's @code{-g <group>} in @value{HPT}).

Our system sends deleting command only for systems with
@code{AdvancedAreafix} and with @ref{UnsubscribeOnAreaDelete} set to "on"
and when default uplink (@code{-def}) unsubscribes from an EchoArea or
some link deletes this area by "~areaname" command. Of course if this
link has the rights to delete this area.

You may also delete an area yourself by running the following command:@*@*
@code{hpt afix <your main AKA> ~area}@*@*
Here @code{<your main AKA>} is your main address. @xref{address,,,fidoconfig,FidoConfig Manual}.

If @code{AdvancedAreafix} for the link is "off" (default), then unsubscribe
command "-areaname" is sent instead of delete command "~areaname".

This statement can only be repeated for different links.

@node AllowEmptyPktPwd, AllowPktAddrDiffer, AdvancedAreafix, Link Keywords
@subsection AllowEmptyPktPwd
@findex AllowEmptyPktPwd
@table @asis
@item Syntax:
@code{allowEmptyPktPwd <off|secure|on>}
@item Example:
@code{allowEmptyPktPwd on}
@end table

This flag is useful if you want to generate packet passwords for this link,
but do not want to check the packet passwords that your link sends to you.
This is sometimes necessary as a workaraound if your link sends you netmail
packets without packet passwords, for example.

The default state is off. In this case the incoming packet password must
match the packet password that you defined. This is the most secure option.

If you set this switch to @code{secure}, packets that do not contain a
packet password and are received in the protected inbound will be processed.
You can use this if your uplink sometimes sends you packets without packet
passwords, as a workaround until the uplink has fixed his system. Still, if
you receive packets with @emph{wrong} packet passwords, they will be
rejected.

The setting @code{on} works like the @code{secure} setting, with the
difference that packets without packet passwords are allowed even in the
unprotected inbound directory. It is not recommended to use this setting.

This statement can only be repeated for different links.

@node AllowPktAddrDiffer, AllowRemoteControl, AllowEmptyPktPwd, Link Keywords
@subsection AllowPktAddrDiffer
@findex AllowPktAddrDiffer
@table @asis
@item Syntax:
@code{allowPktAddrDiffer <bool>}
@item Example:
@code{allowPktAddrDiffer on}
@end table

This keyword is useful if your link has more than one addresses, and therefore
the address in PKT header of his areafix request may be different from the
address in MSG header.

The default state if @code{off}: in this case such letters won't be processed
by your areafix. The state @code{on} makes areafix ignore this error.

This statement can only be repeated for different links.

@node AllowRemoteControl, ArcmailSize, AllowPktAddrDiffer, Link Keywords
@subsection AllowRemoteControl
@findex AllowRemoteControl
@table @asis
@item Syntax:
@code{allowRemoteControl <bool>}
@item Example:
@code{allowRemoteControl on}
@end table

If this keyword is @code{on} then link can do remote control using areafix.
Sending command "%from <addr>" (where <addr> is the address of any link,
defined in config) he makes areafix behave as if the message came from that link.

You should enable this keyword only for yourself or cosysop, as it allows
changing other links' subscription and settings without knowing their passwords.

The default state if @code{off}: remote control is disabled.

This statement can only be repeated for different links.

@node ArcmailSize, ArcNetmail, AllowRemoteControl, Link Keywords
@subsection ArcmailSize
@findex ArcmailSize
@table @asis
@item Syntax:
@code{arcmailsize <integer>}
@item Example:
@code{arcmailsize 300}
@end table

Maximum arcmail size in kb for this link. Default is 500kb.

This statement can only be repeated for different links.

@node ArcNetmail, Areafix, ArcmailSize, Link Keywords
@subsection ArcNetmail
@findex ArcNetmail
@table @asis
@item Syntax:
@code{arcnetmail <bool>}
@item Example:
@code{arcnetmail on}
@end table

This keyword is useful if you want to compress netmail for this link and
pack it into arcmail bundles or to filebox like echomail.
A netmail message will be packed and compressed into a bundle only if
 its flavour is equal to @code{EchoMailFlavour} for this link.

The default state if @code{off}: in this case netmail will be written
to ut-file (i.e. *.?ut) and will not compressed.

This statement can only be repeated for different links.

@node Areafix, AutoAreaCreateSubdirs, ArcNetmail, Link Keywords
@subsection Areafix
@findex Areafix
@table @asis
@item Syntax:
@code{areafix <bool>}
@item Example:
@code{areafix off}
@end table

By default areafix is "on". You can turn off using of areafix by this link.

This statement can only be repeated for different links.

@node AutoAreaCreateSubdirs, AutoPause, Areafix, Link Keywords
@subsection AutoAreaCreateSubdirs
@findex AutoAreaCreateSubdirs
@table @asis
@item Syntax:
@code{autoareacreatesubdirs <bool>}
@item Example:
@code{autoareacreatesubdirs on}
@end table

This switch is turned off by default. When turned off, hpt will create a flat
message base. This means that a file echo @samp{FIDOSOFT.HUSKY} would be
created in the message base directory with a base file name of
@file{fidosoft.husky}, like in @file{/var/spool/msgbase/fidosoft.husky.sqd}
and so on.

When turned on, a structured message base will be created. Each dot in the
echo name is treated as directory separator. Thus, @samp{FIDOSOFT.HUSKY}
would be created with a base name of @file{husky} in the @file{fidosoft}
directory, e.g. @file{/var/spool/msgbase/fidosoft/husky.sqd}.

Please note that this option will @emph{not} currently work if the
@code{-dosfile} option is used to create filenames in 8.3 convention.
@xref{EchoArea,,EchoArea,fidoconfig,fidoconfig}, for more information on the @code{-dosfile} option.

@node AutoPause, AvailList, AutoAreaCreateSubdirs, Link Keywords
@subsection AutoPause
@findex AutoPause
@table @asis
@item Syntax:
@code{autopause <integer>}
@item Example:
@code{autopause 10}
@end table

AutoPause sets the number of days after which HPT will stop exporting echomail
for the link if he has not fetched his echomail for this period of time.
@ref{AutoPassive} should be defined for AutoPause to take effect.

Technically, if you define "autopause 10" for a link then HPT looks for
10 (or more) days old files that are listed to be sent for this link in
the .?lo files with `#' or `^' prefixes and reside in the outbound directory.
If at least one such file exists, exporting echomail for the link is paused.

This statement can only be repeated for different links.

See also @ref{AutoPassive} and @ref{Pause,,,fidoconfig,fidoconfig} sections.

@node AvailList, DailyBundles, AutoPause, Link Keywords
@subsection AvailList
@findex AvailList
@table @asis
@item Syntax:
@code{AvailList (Full|Unique|UniqueOne)}
@item Example:
@code{AvailList Full}
@end table

This statement specify variant for reply to %avail command for current link:

@multitable {* UniqueOne}{aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa}
@item * Full
@tab produce full areas list for each uplink with avaiable forward subscribe for link
@item * Unique
@tab like @code{Full} but exclude dulicated areas for each next uplink
@item * UniqueOne
@tab produce one list: all areas from all uplinks without duplicates
@end multitable

Default value is @code{AvailList Full}.

This statement can only be repeated for different links.

@node DailyBundles, DenyRescan, AvailList, Link Keywords
@subsection DailyBundles
@findex DailyBundles
@table @asis
@item Syntax:
@code{dailyBundles <bool>}
@item Example:
@code{dailyBundles}
@end table

If DailyBundles is off, hpt will use existing arcmail bundles to add packets
for this link, even if they were created in the previous days. If
DailyBundles is set, a new arcmail bundle for the link will be started when
the new day begins.

This statement can only be repeated for different links.

@node DenyRescan, EchoMailFlavour, DailyBundles, Link Keywords
@subsection DenyRescan
@findex DenyRescan
@table @asis
@item Syntax:
@code{denyRescan <bool>}
@item Example:
@code{denyRescan}
@end table

Don't allow to rescan areas (via areafix) to your links.

This token can be applied to one or several area groups (see @code{RescanGrp}
token for details).

This statement can only be repeated for different links.

@node EchoMailFlavour, Flavour, DenyRescan, Link Keywords
@subsection EchoMailFlavour
@findex EchoMailFlavour
@table @asis
@item Syntax:
@code{echoMailFlavour <hold | normal | crash | direct | immediate>}
@item Example:
@code{echoMailFlavour hold}
@end table

This statement sets the flavour which outgoing echomails for this link get.
For example set echomailFlavour for points to hold and for uplink crash.

This statement can only be repeated for different links.

@node Flavour, ForwardPkts, EchoMailFlavour, Link Keywords
@subsection Flavour
@findex Flavour
@table @asis
@item Syntax:
@code{flavour <hold | normal | crash | direct | immediate>}
@item Example:
@code{flavour hold}
@end table

This statement sets the flavours for netmail, echomail and fileechoes
simultaneously. Note that you can combine this with @code{netMailFlavour},
@code{echoMailFlavour} or @code{fileEchoFlavour}; in this case the latest
flavour is applied.

This statement can only be repeated for different links.

@node ForwardPkts, FileBox, Flavour, Link Keywords
@subsection ForwardPkts
@findex ForwardPkts
@table @asis
@item Syntax:
@code{forwardPkts <off|secure|on>}
@item Example:
@code{forwardPkts yes}
@end table

If we receive a PKT file that is not addressed to our system, but to this
link of us, this flag controls if the PKT file should be binary
forwarded to this link. The default behaviour is not to forward the pkt
file, but to remain it to @code{<filename>.ntu} and leave it in the
inbound. If you specify @code{<yes>}, the file will instead be forwarded
to the destination link (i.E. put into his arcmail bundle). If you
specify @code{<secure>}, the file will only be forwarded if we have
received it in the secure inbound. You should specify @code{secure} if
the destination link does not check packet passwords.

PKT forwarding can be useful for tunneling purposes, for
instance. Another example is if you are running two nodes, one IP node
at your company and one PSTN node at your home. If you want to show both
node numbers at both mailers, the tossers at each node must forward PKT
files that are addressed to the other node, because they themselves
cannot process it (each tosser has a different node number, because the
systems operate on distinct outbound structures and distinct message
base areas).

This statement can only be repeated for different links.

WARNING! This statement also enables file forwarding in htick!

@node FileBox, FileBoxAlways, ForwardPkts, Link Keywords
@subsection FileBox
@findex FileBox
@table @asis
@item Syntax:
@code{FileBox <directory>}
@item Example:
@code{FileBox boxes/2.5021.19.1}
@end table

This statement defines directory where outgoing files for link would be
putten instead of putting them in @code{Outbound} tree. See also
@code{FileBoxAlways} for details.

Currently hpt can put only echomail into fileBoxes.

This statement can only be repeated for different links.

@node FileBoxAlways, LinkBundleNameStyle, FileBox, Link Keywords
@subsection FileBoxAlways
@findex FileBoxAlways
@table @asis
@item Syntax:
@code{FileBoxAlways <bool>}
@item Example:
@code{FileBoxAlways}
@end table

Pack to link @code{FileBox} even if link is not busy. By default
outbound is used when link is not busy.

This statement can only be repeated for different links. Works only for
links which have @code{FileBox} or when @code{FileBoxesDir} is set.

@xref{FileBox}, @xref{FileBoxesDir}.

@node LinkBundleNameStyle, LinkGrp, FileBoxAlways, Link Keywords
@subsection LinkBundleNameStyle
@findex LinkBundleNameStyle
@table @asis
@item Syntax:
@code{linkBundleNameStyle <timeStamp | addrDiff | addrDiffAlways | addrsCRC32 | addrsCRC32Always | Amiga>}
@item Example:
@code{linkbundleNameStyle addrDiff}
@end table

This statement sets rule for creating names of arcmail bundles for link.
It is similar @code{BundleNameStyle} keyword.

@xref{BundleNameStyle}.

This statement can only be repeated for different links.

@node LinkGrp, LinkMsgBaseDir, LinkBundleNameStyle, Link Keywords
@subsection LinkGrp
@findex LinkGrp
@table @asis
@item Syntax:
@code{LinkGrp <Group name>}
@item Example:
@code{LinkGrp INTERNATIONAL}
@end table

   Specifies a group to use for areas auto-created from this link. Also
link can delete echo areas in his @code{LinkGrp} group.

   In this exmaple:

   1) string @code{INTERNATIONAL} will be added to a new area record unless
@code{AutoCreateDefaults} defines implicit group;

   2) links that have no access to group @code{INTERNATIONAL} can't forward requests to uplink with @code{LinkGrp INTERNATIONAL};

   3) this link can delete echo areas with @code{-g INTERNATIONAL} group if @code{AdvancedAreafix} is enabled for this link.

@xref{AdvancedAreafix}.

This statement can only be repeated for different links.

@node LinkMsgBaseDir, NetMailFlavour, LinkGrp, Link Keywords
@subsection LinkMsgBaseDir
@findex LinkMsgBaseDir
@table @asis
@item Syntax:
@code{linkMsgBaseDir <path>}
@item Example:
@code{linkMsgBaseDir /var/spool/fido/msgb}
@end table

Same as @code{MsgBaseDir}, but you can set it for different links.
@strong{Do not use it with} @code{LinkDefaults}!

This statement can only be repeated for different links.

@xref{MsgBaseDir}.

@node NetMailFlavour, NoRules, LinkMsgBaseDir, Link Keywords
@subsection NetMailFlavour
@findex NetMailFlavour
@table @asis
@item Syntax:
@code{netMailFlavour <hold | normal | crash | direct | immediate>}
@item Example:
@code{netMailFlavour crash}
@end table

This statement sets the flavour for outgoing netmail routed via this link.

This statement can only be repeated for different links.

@node NoRules, PackAka, NetMailFlavour, Link Keywords
@subsection NoRules
@findex NoRules
@table @asis
@item Syntax:
@code{norules <bool>}
@item Example:
@code{norules on}
@end table

By default "norules off" i.e. send rules.

This statement points to areafix to not send area rules to link when
link is subscribed to area and corresponding file is found in
@code{RulesDir}.

This statement can only be repeated for different links.

@node PackAka, PktSize, NoRules, Link Keywords
@subsection PackAka
@findex PackAka
@table @asis
@item Syntax:
@code{PackAka <addr>}
@item Example:
@code{PackAka 2:4600/220}
@end table

This statement sets the aka to which the echomail and fileechos will be sent
(Pack echomail and fileechos via third link). The PackAka link used only in
BSO, fileboxes and arcmail filename creation, therefore the PackAka link may
not present in config.

This feature may be used for simple echomail routing.

This statement can only be repeated for different links.


@node PktSize, ReducedSeenBY, PackAka, Link Keywords
@subsection PktSize
@findex PktSize
@table @asis
@item Syntax:
@code{pktsize <integer>}
@item Example:
@code{pktsize 300}
@end table

Maximum pkt size in kb for this link. Default - unlimited.

This statement can only be repeated for different links.

@node ReducedSeenBY, RescanGrp, PktSize, Link Keywords
@subsection ReducedSeenBY
@findex ReducedSeenBY
@table @asis
@item Syntax:
@code{ReducedSeenBY <bool>}
@item Example:
@code{ReducedSeenBY on}
@end table

This statement turns on/off reduced seen-by algorithm (FSC-0093)

@example
RSB algorithm
-------------
1) add own system to the PATH.
2) all area links not contained in the RSB qualify as recipients.
3) strip RSB addresses not matching an address in the PATH, then
   add own address(es) to the RSB set if not already contained.
4) add recipients to RSB, sort RSB and forward mail to recipients.
@end example

This statement can only be repeated for different links.

@node RescanGrp, RescanLimit, ReducedSeenBY, Link Keywords
@subsection RescanGrp
@findex RescanGrp
@table @asis
@item Syntax:
@code{RescanGrp <string>[,<string>...]}
@item Example:
@code{RescanGrp A,B,C,Local}
@end table

If defined, @code{denyRescan} token applies to specified area groups only,
while other area groups are applied with the opposite value.

Example:

There are areas with the groups A, B, C and D in config. The following table
demonstrates how combination of options allows you to specify different
rights for link.

@table @asis
@item denyRescan off
All area groups are allowed to rescan.
@item denyRescan on
All area groups are denied to rescan.
@item denyRescan off, rescanGrp A,B
Area groups A and B are allowed to rescan, while C and D are denied.
@item denyRescan on, rescanGrp A,B
Area groups A and B are denied to rescan, while C and D are allowed.
@end table

This statement can only be repeated for different links.

@node RescanLimit, SendNotifyMessages, RescanGrp, Link Keywords
@subsection RescanLimit
@findex RescanLimit
@table @asis
@item Syntax:
@code{RescanLimit <number>}
@item Example:
@code{RescanGrp 100}
@end table

This keyword is used to limit max number of messages sent by %rescan command.
If it's set to 0 (default), the link is not limited.

This statement can only be repeated for different links.

@node SendNotifyMessages, UnsubscribeOnAreaDelete, RescanLimit, Link Keywords
@subsection SendNotifyMessages
@findex SendNotifyMessages
@table @asis
@item Syntax:
@code{sendNotifyMessages <bool>}
@item Example:
@code{sendNotifyMessages on}
@end table

If sendNotifyMessages is true, link will receive notification messages from
areafix about link's requests status, bad posts or deletion of subscribed area.

By default is off, i.e. do not send any notification messages.

This statement can only be repeated for different links.

@node UnsubscribeOnAreaDelete,  , SendNotifyMessages, Link Keywords
@subsection UnsubscribeOnAreaDelete
@findex UnsubscribeOnAreaDelete
@table @asis
@item Syntax:
@code{unsubscribeOnAreaDelete <bool>}
@item Example:
@code{unsubscribeOnAreaDelete yes}
@end table

If unsubscribeOnAreaDelete is true, link will receive unsubscribe request
("-area" if @ref{AdvancedAreafix} is off for the link and "~area" if it is on)
to his areafix when subscribed area is deleted. An area may be deleted when
default uplink (@code{-def}) unsubscribes from the area or some link deletes
this area by "~area" command (@ref{LinkGrp} for the link that deletes an area
must be the same as EchoArea's @code{-g <group>}). You may also delete an area
yourself by running the following command:@*@*
@code{hpt afix <your main AKA> ~area}@*@*
Here @code{<your main AKA>} is your main address. @xref{address,,,fidoconfig,FidoConfig Manual}.

By default is off, i.e. do not send unsubscribe request.

This statement can only be repeated for different links.


@node Carbon Copy,  , Link Keywords, Configuration Reference
@section Carbon Copy

@menu
* CarbonAddr::                  make carbonCopy using from address of msg
* CarbonAndQuit::               carbon message only one time
* CarbonCopy::                  specify the area where the carbon copies go into
* CarbonDelete::                specify that the carbon msg should be deleted
* CarbonExcludeFwdFrom::        exclude " * Forward from area " string
* CarbonExtern::                specify the external program to pass msg text to
* CarbonFrom::                  make carbonCopy using from-field of msg
* CarbonFromArea::              carbon messages from area
* CarbonGroups::                make carbonbonCopy from this groups
* CarbonKeepSb::                keep seen-by's and path
* CarbonKludge::                make carbonCopy using kludge lines of msg
* CarbonMove::                  specify the area where the carbon msg get moved into
* CarbonOut::                   carbon outgoing messages
* CarbonReason::                string of carbon copy reason
* CarbonRule::                  how to combine Carbon expressions
* CarbonSubj::                  make carbonCopy using subject-field of msg
* CarbonText::                  make carbonCopy using text-field of msg
* CarbonTo::                    make carbonCopy using to-field of msg
* ExcludePassthroughCarbon::    don't carbon in passthrough areas
* NetmailExtern::               specify the external Program to pass netmail msgs to

@end menu

@value{HPT} is based on FIDOCONFIG library, so read documentation of
FIDOCONFIG about location of config file, keywords ideology and about
majority of the keywords,
@xref{Top, Fidoconfig Manual, , fidoconfig, Fidoconfig manual}.


@node CarbonAddr, CarbonAndQuit, Carbon Copy, Carbon Copy
@subsection  CarbonAddr
@findex CarbonAddr
@table @asis
@item Syntax:
@code{carbonAddr <addr>}
@item Example:
@code{carbonAddr 2:5000/100}@*
@code{carbonCopy mail.from.100}
@end table

If an echomail is tossed whose from address field is the same as <addr>,
the echomail is copied to the area specified by the @code{carbonCopy}
keyword.

This statement can be repeated.

@node CarbonAndQuit, CarbonCopy, CarbonAddr, Carbon Copy
@subsection CarbonAndQuit
@findex CarbonAndQuit
@table @asis
@item Syntax:
@code{carbonAndQuit <bool>}
@item Example:
@code{carbonAndQuit}
@end table

By default one message can be processed by Carbon Copy several times.
For example: you set CarbonTo "max" to echoarea MY.MAIL and CarbonSubj
"beer" to echoarea MY.PLEASURE :-) And message to "max" with this
subject line carbons to each @code{EchoArea}.

If you turn on @code{CarbonAndQuit} message will be copyed to MY.MAIL
only.

It is however possible to override this setting, by putting an '*'
before the action:
@example
     CarbonCopy *my.mail
     CarbonMove *my.mail
     CarbonExtern *<command line>
@end example

This '*' does not work for CarbonDelete of course.

This statement cannot be repeated.

@node CarbonCopy, CarbonDelete, CarbonAndQuit, Carbon Copy
@subsection CarbonCopy
@findex CarbonCopy
@table @asis
@item Syntax:
@code{carbonCopy <area-tag>}
@item Example:
@code{carbonCopy written.from.points}
@end table

This statement sets the area for the previous
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.

If no @code{EchoArea} defined to copy, the carbon msgs goes to the
@code{BadArea}.

Note: You cannot carbonCopy a message from EchoArea to NetmailArea
and vice versa.

This statement can be placed after different
carbon@{Addr|To|From|Kludge|Subj|Text@} statements.

@xref{CarbonMove}.

@node CarbonDelete, CarbonExcludeFwdFrom, CarbonCopy, Carbon Copy
@subsection CarbonDelete
@findex CarbonDelete
@table @asis
@item Syntax:
@code{carbonDelete}
@item Example:
@code{carbonDelete}
@end table

This statement specifies that selected by previous
carbon@{Addr|To|From|Kludge|Subj|Text@} statement msgs should not be
stored into @code{EchoArea}.

This statement can be repeated for each different
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.

@node CarbonExcludeFwdFrom, CarbonExtern, CarbonDelete, Carbon Copy
@subsection CarbonExcludeFwdFrom
@findex CarbonExcludeFwdFrom
@table @asis
@item Syntax:
@code{CarbonExcludeFwdFrom <bool>}
@item Example:
@code{CarbonExcludeFwdFrom}
@end table

Don't add to begin of carbon msg text " * Forward from area <areatag>"

This statement cannot be repeated.

@node CarbonExtern, CarbonFrom, CarbonExcludeFwdFrom, Carbon Copy
@subsection CarbonExtern
@findex CarbonExtern
@table @asis
@item Syntax:
@code{carbonExtern <program name>}
@item Example:
@code{carbonExtern douuedecode}
@end table

This statement specifies external program to run for the previous
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.

If this statement is ommitted and no @code{EchoArea} defined for move or
copy, the carbon msgs gets copyed to the @code{BadArea}.

If program name is prepended with '|' sign, hpt tries to pass data thru
the pipe, not temporary file.

This statement can be repeated for each different
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.

@node CarbonFrom, CarbonFromArea, CarbonExtern, Carbon Copy
@subsection CarbonFrom
@findex CarbonFrom
@table @asis
@item Syntax:
@code{carbonFrom <pattern>}
@item Example:
@example
carbonFrom Matthias Tichy
carbonCopy my.echomail
@end example
@end table

If an echomail is tossed whose from-field matched the pattern <pattern>,
the echomail is copied to the area specified by the @code{CarbonCopy}
keyword or moved to the area specified by the @code{CarbonMove} keyword.
The names must be an exact match.

You can enclose <string> into quotes ("<string>").

This statement can be repeated.

@node CarbonFromArea, CarbonGroups, CarbonFrom, Carbon Copy
@subsection CarbonFromArea
@findex CarbonFromArea
@table @asis
@item Syntax:
@code{CarbonFromArea <pattern>}
@item Example:
@code{CarbonFromArea Interuser}
@end table

Only messages that are written in this areas will be copied. Use it in
combination with CarbonRule to prevent that all messages from an area
are copied.

@xref{CarbonRule}.

Another example:
@example
CarbonFromArea Interuser
CarbonRule NOT
CarbonText " Israel "
CarbonCopy my.news
@end example

<pattern> is the mask for area name, defined after EchoArea or NetMailArea.

This statement can be repeated.

@node CarbonGroups, CarbonKeepSb, CarbonFromArea, Carbon Copy
@subsection CarbonGroups
@findex CarbonGroups
@table @asis
@item Syntax:
@code{carbonGroups <string> [,<string>...]}
@item Example:
@code{carbonGroups Fido, LifeNet}@*
@code{carbonCopy my.echomail}
@end table

With this keyword you can define from which groups messages should be copied.
It is wise when you use this keyword together with other selection criteria
or else the result will give many messages.

@xref{CarbonRule}.

This statement can be repeated.

@node CarbonKeepSb, CarbonKludge, CarbonGroups, Carbon Copy
@subsection CarbonKeepSb
@findex CarbonKeepSb
@table @asis
@item Syntax:
@code{carbonKeepSb <bool>}
@item Example:
@code{carbonKeepSb}
@end table

For each CC message SEEN-BY's and PATH fields are killed. If you set up
@code{CarbonKeepSb} then all kludges will be saved. But you may also
set up @code{-keepsb} in options of @code{EchoArea} where you carbon
messages.

This statement cannot be repeated.

@node CarbonKludge, CarbonMove, CarbonKeepSb, Carbon Copy
@subsection CarbonKludge
@findex CarbonKludge
@table @asis
@item Syntax:
@code{carbonKludge <pattern>}
@item Example:
@code{carbonKludge MSGID: 2:5000/117.}@*
@code{carbonCopy written.by.points}
@end table

If an echomail is tossed which has a kludge line which includes the
substring matches <pattern>, the echomail is copied to the area
specified by the @code{CarbonCopy} keyword.

You can enclose <pattern> into quotes ("<pattern>").
@example
carbonKludge "REPLY: 2:5000/117 "
@end example

This statement can be repeated.

@node CarbonMove, CarbonOut, CarbonKludge, Carbon Copy
@subsection CarbonMove
@findex CarbonMove
@table @asis
@item Syntax:
@code{carbonMove <area-tag>}
@item Example:
@code{carbonMove my.echomail}
@end table

This statement sets the area for the previous
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.

If no @code{EchoArea} defined to copy (or move), the carbon msgs goes to
the @code{BadArea}.

Note: You cannot carbonMove a message from EchoArea to NetmailArea
and vice versa.

Unlike @code{CarbonCopy} msg gets moved, not copied into this area.

This statement can be placed after different
carbon@{Addr|To|From|Kludge|Subj|Text@} statements.

@xref{CarbonCopy}.

@node CarbonOut, CarbonReason, CarbonMove, Carbon Copy
@subsection CarbonOut
@findex CarbonOut
@table @asis
@item Syntax:
@code{carbonOut <bool>}
@item Example:
@code{carbonOut}
@end table

This statement makes possible carbon your outgoing mail (scanned &
packed). Don't forget to set carbon rules.

This statement cannot be repeated.

@node CarbonReason, CarbonRule, CarbonOut, Carbon Copy
@subsection CarbonReason
@findex CarbonReason
@table @asis
@item Syntax:
@code{carbonReason <string>}
@item Example:
@code{carbonText msged}@*
@code{carbonReason Found 'msged' in message}@*
@code{carbonCopy my.searches.echo}
@end table

This statement sets the 'reason string' for the previous
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.

This string would be printed after ' * Forwarded ...' line in message
begining. That's useful when many carbons are made to single area.
No reason string is printed if this statement is ommitted.

This statement can be repeated for each different
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.

@node CarbonRule, CarbonSubj, CarbonReason, Carbon Copy
@subsection CarbonRule
@findex CarbonRule
@table @asis
@item Syntax:
@code{carbonRule AND|NOT|OR}
@item Example:
@example
carbonGroups fido
carbonRule AND
carbonText programming
carbonCopy my.echomail
@end example
@end table

With this keyword you can make it possible to have more than 1 selection
criteria for a message. The default rule is AND. So without using CarbonRule,
all expressions will be AND-ed.
With Carbonrule AND, the two expressions above and below the AND must be
true or else a message will not be copied.
With Carbonrule you can define how the expressions are joined together.

CarbonRule NOT, is in fact AND NOT. So

@example
        CarbonText beatles
        CarbonRule NOT
        CarbonText bugs
        CarbonCopy my.mail
@end example

...will only copy messages when 'beatles' is in the text AND NOT 'bugs'
in the text.

In general, it is possible to define more than one set of rules for
1 single carbonArea.
For example:
@example
        CarbonText female
        CarbonRule NOT
        CarbonText connector
        CarbonRule OR
        CarbonSubj beer
        CarbonRule AND
        CarbonText drink
        CarbonText holliday
        CarbonCopy my.mail
@end example
Here are two sets of criteria and for all of them there is only 1 carbonarea
defined. After an OR, a new set of expressions start.
The above example can also be written as:
@example
        CarbonRule NOT
        CarbonText connector
        CarbonRule AND
        CarbonText female
        CarbonRule OR
        CarbonRule AND
        CarbonSubj beer
        CarbonText drink
        CarbonText holliday
        CarbonCopy my.mail
@end example

Messages that have 'female' in the text, but not the word 'connector',
will be copied to the my.mail area.

Also messages with 'beer' in the text AND 'drink' in the text AND 'holliday'
in the text, will be copied to the my.mail area.

A set of expressions is true when they are OR-ed and at least one
of them is true.

A set of expressions is true when they are AND-ed and ALL of them are true.
Expressions are evaluated from top to bottom.

This statement can be repeated.

A CarbonRule is valid until a next rule is defined of until an action.
After an action (CarbonCopy, CarbonMove, etc), the CarbonRule is AND again
and new set of expressions starts.

@node CarbonSubj, CarbonText, CarbonRule, Carbon Copy
@subsection CarbonSubj
@findex CarbonSubj
@table @asis
@item Syntax:
@code{carbonSubj <pattern>}
@item Example:
@code{carbonSubj beer}@*
@code{carbonCopy cc.beer}
@end table

If an echomail is tossed which has a subject line matched <pattern>,
this echomail is copied to the area specified by the @code{CarbonCopy}
keyword or moved to the area specified by the @code{CarbonMove} keyword.
You can enclose <pattern> into quotes ("<pattern>").

This statement can be repeated.

@node CarbonText, CarbonTo, CarbonSubj, Carbon Copy
@subsection CarbonText
@findex CarbonText
@table @asis
@item Syntax:
@code{carbonText <pattern>}
@item Example:
@code{carbonText cool beer}@*
@code{carbonCopy cc.beer}
@end table

If an echomail is tossed which has a kludge line which includes the
substring matches <pattern>, the echomail is copied to the area
specified by the @code{CarbonCopy} keyword or moved to the area
specified by the @code{CarbonMove} keyword.

You can enclose <pattern> into quotes ("<pattern>").

This statement can be repeated.

@node CarbonTo, ExcludePassthroughCarbon, CarbonText, Carbon Copy
@subsection CarbonTo
@findex CarbonTo
@table @asis
@item Syntax:
@code{carbonTo <pattern>}
@item Example:
@code{carbonTo Max Levenkov}@*
@code{carbonCopy my.echomail}
@end table

If an echomail is tossed whose to-field matched <pattern>, the
echomail is copied to the area specified by the @code{CarbonCopy}
keyword or moved to the area specified by the @code{CarbonMove}
keyword. The names must be an exact match.

You can enclose <pattern> into quotes ("<pattern>").

This statement can be repeated.

@node ExcludePassthroughCarbon, NetmailExtern, CarbonTo, Carbon Copy
@subsection ExcludePassthroughCarbon
@findex ExcludePassthroughCarbon
@table @asis
@item Syntax:
@code{excludePassthroughCarbon <bool>}
@item Example:
@code{excludePassthroughCarbon}
@end table

Don't carbon from passthrough areas.

This statement cannot be repeated.

@node NetmailExtern,  , ExcludePassthroughCarbon, Carbon Copy
@subsection NetmailExtern
@findex NetmailExtern
@table @asis
@item Syntax:
@code{netmailExtern <program name>}
@item Example:
@code{netmailExtern douuedecode}
@end table

This statement specifies external program to run for the previous
carbon@{Addr|To|From|Kludge|Subj|Text@} statement for netmail
messages...

If this statement is ommitted and no @code{NetMailArea} defined for move or
copy, the carbon msgs gets copyed to the @code{BadArea}.

If program name is prepended with '|' sign, hpt tries to pass data thru the
pipe, not temporary file.

This statement can be repeated for each different
carbon@{Addr|To|From|Kludge|Subj|Text@} statement.


@node Advanced Concepts, Keyword Index, Configuration Reference, Top
@chapter Advanced Concepts in @value{HPT}

This chapter describes how to use some features...
After you have managed to perform basic functions with @value{HPT}, like
defining paths & areas, this chapter will introduce you into some
advanced concepts in @value{HPT} that deserve special attention, like
how to use several netmail areas and similar.

@menu
* Exit::            hpt exit codes
* Suffixes::        renaming suffixes for pkt files
* Netmail::         defining several netmail areas for several users
* ZoneGating::      how zonegating works for echomail
* Perl support::    Perl support in HPT
* Remote control::  how remote control using Areafix works
@end menu

@node Exit, Suffixes, Advanced Concepts, Advanced Concepts
@section HPT Exit Codes

@table @asis
@item 0 - successful termination
@item 64 - command line usage error
@item 66 - cannot open input (file)
@item 69 - service unavailable [config file not found]
@item 70 - internal software error [not enough memory, etc.]
@item 73 - can't create (user) output file [lockfile]
@item 74 - input/output error
@item 75 - temp failure; user is invited to retry [lockfile used by second copy of hpt]
@item 78 - configuration error
@end table

@node Suffixes, Netmail, Exit, Advanced Concepts
@section Renaming Suffixes For Pkt Files

@table @asis
@item .tos - hpt crashes on this pkt, report to developers!
@item .sec - a) pktpwd problem or link not found; b) bundles in unsecure inbound
@item .acs - could not open pkt
@item .bad - a) not/wrong pkt; b) bundle unpacking problems
@item .ntu - not to us
@item .err - msg tossing problem (couldn't open badArea/couldn't write msg)
@item .flt - perl filter (process_pkt()) returns bad retcode
@end table

@node Netmail, ZoneGating, Suffixes, Advanced Concepts
@section Defining Several Netmail Areas For Several Users

If you want receive netmail for one address in first netmail area, and
for second address in second netmail area respectively, you must add
@code{-a <ourAka>} to netmail area, like this:

@table @asis
@item Address 2:5000/117
@item Address 2:90/190
@item NetmailArea NetMail /home/ml/fido/msgb/netmail -b squish -a 2:5000/117
@item NetmailArea OtherNetMail /home/ml/fido/msgb/othernetmail -b squish -a 2:90/190
@end table

@node ZoneGating, Perl support, Netmail, Advanced Concepts
@section How Zone Gating works for echomail and what seen-bys is stripped

We have EchoArea with links from zone1 & zone2. Echo->useAka with
zone2. Defined by "-a <addr>" or the by the first Echo->address from
config if not defined in EchoArea line.

Program algorythm:

@table @asis
@item 1.
check pkt zone address
@item 2.
split links in to groups: with (pktaddr.zone == linkaddr.zone) and
(pktaddr.zone != linkaddr.zone)
@item 3.
send msg to first group of links (their aka adding to seen-bys and
echo->useAka too if not in seen-bys yet)
@item 4.
if (pktaddr.zone != echo->useAka.zone) seen-bys cleaned.
@item 5.
send msg to second group of links (their aka adding to seen-bys and
echo->useAka too if not in seen-bys yet)
@item 6.
store msg in msgbase
@end table

Examples:

a) msg comes from z1. for z1 links we sending seen-bys with z2
   nodes. but z1 links must clean seen-bys by themselves (this is z2
   echo) or download this echo from z1 link. after that seen-bys cleaned
   and msg forwarded to z2 links. in the msgbase seen-bys from z2 links
   only.

b) msg comes from z2. forward it to z2 links. then forward to z1
   links. seen-bys not cleaned: this is z2 echo and z1 links must clean
   it by themselves or receive echo from z1 link. store in msgbase with
   all seen-bys from z1 & z2 nodes.

c) scan works the same: first sending to z2 links, adding to seen-by's
   z1 links and export to them.

@node Perl support, Remote control, ZoneGating, Advanced Concepts
@section Perl support in HPT
HPT may be compiled with support for calling Perl subs in some parts of its
work. In this case it also provides API for such subs (Perl hooks).
Perl subs are located in file filter.pl, which usually located
in same directory as HPT's executable file but it's name and location
may changed via @code{HptPerlFile} configuration option.
The following Perl hooks are supported:
@table @asis
@item 1.
 after_unpack - called after unpacking an echomail bundle to @code{TempInbound}
@item 2.
 before_pack - called before packing echomail bundle to one of links
@item 3.
 process_pkt - called before processing pkt, the following variables available:
@table @asis
@item $pktname - name of pkt,
@item $secure - defined if this pkt from secure link
@end table
 hook must return "" for normal pkt processing or other string to rename pkt to .flt
@item 4.
 pkt_done - called after pkt processing, the following variables available:
@table @asis
@item $pktname - name of pkt,
@item $rc - exit code(0-ok),
@item $res - reason(exit code in text form):
@table @asis
@item 0 - OK ($res undefined),
@item 1 - Security violation,
@item 2 - Can't open pkt,
@item 3 - Bad pkt format,
@item 4 - Not to us,
@item 5 - Msg tossing problem
@end table
@end table
@item 5.
  hpt_exit - called before hpt completely exit if any other Perl hook was called
  during this session.
@item 6.
  route - called just before routing netmail message, the following variables
  availble:
@table @asis
@item $addr - message destination address,
@item $from - message originating address,
@item $toname - destination user name,
@item $fromname - originating user name,
@item $subject - message subject line,
@item $date - message date and time,
@item $text - message text,
@item $attr - message attributes,
@item $route - default route for this message (derermined
  via @code{Route} statements in config file (may be empty, this means
  that either no route at all for this message or it will be routed via
  one-to-multi routing(Route normal noroute 2:5004/73.*)).
@end table

  Before return you can set $flavour - to hold|normal|crash|direct|immediate
  for required flavour of message.
  return "" for default routing or address via which this message should be sent.
  example:

@example
  sub route @{
    if ($from eq "2:5004/75.73") return "2:5004/75.0";
    else return "";
  @}
@end example

  if source address of message is 2:5004/75.73 then route it via 2:5004/75,
  otherwise route it using default routing
@item 7.
 scan - called while scanning messages (hpt scan or hpt pack). The following
 variables available:
@table @asis
@item $fromname - originating user name,
@item $fromaddr - message originating address,
@item $toname - destination user name,
@item $toaddr - message destination address (for netmail),
@item $area - message area (for echomail),
@item $subject - message subject line,
@item $date - message date and time,
@item $text - message text,
@item $attr - message attributes.
@end table
  Set $change to update $text, $subject, $fromaddr, $toaddr, $fromname,
$toname, $attr.
  If returns non-empty string (reason), the message will not pack to downlinks.
@item 8.
 filter - called for processing every message while tossing.
The following variables available:
@table @asis
@item $fromname - originating user name,
@item $fromaddr - message originating address,
@item $toname - destination user name,
@item $toaddr - message destination address (for netmail),
@item $area - message area (for echomail),
@item $subject - message subject line,
@item $date - message date and time,
@item $text - message text,
@item $attr - message attributes,
@item $pktfrom - address of originating pkt,
@item $secure - defined if the message received from secure link.
@end table
  Set $change to update $text, $subject, $fromaddr, $toaddr, $fromname,
$toname, $attr.
  Set $kill for kill message.
  If returns non-empty string (reason), the message will be moved to badarea.
@item 9.
 tossbad - called when message will be put in badArea.
The following variables available:
@table @asis
@item $fromname - originating user name,
@item $fromaddr - message originating address,
@item $toname - destination user name,
@item $toaddr - message destination address (for netmail),
@item $area - message area (for echomail),
@item $subject - message subject line,
@item $date - message date and time,
@item $text - message text,
@item $attr - message attributes,
@item $pktfrom - address of originating pkt,
@item $reason - reason, why badarea (text string).
@end table
  Set $change to update $text, $subject, $fromaddr, $toaddr, $fromname,
$toname, $attr.
  If returns non-empty string (reason) for kill the message.

@end table

@node Remote control,  , Perl support, Advanced Concepts
@section How remote control using Areafix works

You can remotely control and change other links' subscription and options
using Areafix. Remote control will not give you full access on your system,
but will help you (or your cosysop) with some most common tasks.

Some terms first.

Controlling link is a link that exercise remote control. He should be defined
in config and should have allowRemoteControl token set to "on".

Controlled link is a link whose subscription and options are controlled. He should
be defined in config too.

Controlling link is allowed to send "%from <addr>" command to Areafix, where
<addr> is an address of controlled link. All the next commands till the end of
message or till the next "%from" command will be executed as if the message came
from controlled link. All the reports will be sent to controlling link. There can
be as much "%from" commands in message as you like. If there was an error while
processing "%from" command (for example, address was not specified in "%from"
command), message processing will be stopped and controlling link will receive
notification message.

Example of Areafix message:

@example
%from 2:450/210.1       <-- all next actions will be made with 2:450/210.1
+ru.husky               <-- subscribe to ru.husky
%areafixpwd testpwd     <-- change areafix password to 'testpwd'
%from 2:450/210.5       <-- all next actions will be made with 2:450/210.5
-bel.sysop              <-- unsubscribe from bel.sysop
%linked                 <-- send list of linked areas
@end example

As this function allows remote control of other links' subscription and options
without knowing their areafix passwords, you should enable allowRemoteControl
token only for yourself or your cosysop.

@node Keyword Index,  , Advanced Concepts, Top

@value{HPT} is based on FIDOCONFIG library, so read documentation of
FIDOCONFIG about location of config file, keywords ideology and about
majority of the keywords,
@xref{Top, Fidoconfig Manual, , fidoconfig, Fidoconfig manual}.

@appendix Configuration File Keyword Index
@printindex fn

@bye