home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
OS/2 Shareware BBS: 2 BBS
/
02-BBS.zip
/
orion336.zip
/
README.TXT
< prev
next >
Wrap
Text File
|
1998-04-05
|
226KB
|
4,875 lines
Orion CBCS/2
*****************************************************************************
DisClaimer:
1> This archive can be circulated, but it is primarily for my own and a couple
other peoples use.
2> If you got a hold of this archive somehow and are bent on running the
orion.exe and associated files in this archive your entirely on your
own, there is *NO* support for this software. I work on it at my leisure.
(Hey what do you expect for free?)
3> That said, updates can be had at ftp to dogstar.bbs.dyndns.com
or via telnet to the bbs at the same address.
\Orion_CBCS_Updates is the directory name they will be placed in.
Orionfull.zip is for first time installs, the updates are called
Orion331.zip for version 3.31 update etc....
4> email me at mariod@ican.net
5> ftp/telnet or http://dogstar.bbs.dyndns.com
******************************************************************************
3.36c
1> Found a major nasty that would trap Orion if u toggled DTR while a user
was downloading files in a native telnet session, fixed it and did some
other cleanup while i was at it.
2> The ftp server will no longer trap orion on shutdown.
3> Would like to report that Ray Gwinns SIO Version 2 and it's Vmodem work
just DANDY with orion, I'm running it here and beleive me it's worth
every penny of my pre-registration money even in it's beta state. See
http://gwinn.com for the latest archive.
4> I feel I must make a clean break with AdeptXbbs, future builds of Orion
will Likely not include features added to adept after 1.11x and vice
versa adept will not have my code incorporated into it. If I feel a feature
is worth putting in I'll see what I can do, as I'm sure Nitin Chandra will
do for adeptx.
3.36b
1> More minor changes and cleanup.
2> Came up with a workaround to that rexx problem, (I think this is IBM's problem)
not ours, anyway instead of using Address Cmd SomeOs2Command, use
Start SomeOs2Command
Call SysSleep(5)
or however many seconds u think your command will take to execute to completion
and this will prevent the hung nodes problem caused by rexx scripts using
Address CMD statements. Works here, should work everywhere, even in ahem... AdeptXbbs.
3.36a
1> Made a few internal changes, this requires u do an orion /nodes x /setup to
use this version, fire it up in setup mode and edit the mailer general settings
screen for each line, it should be obvious what needs fixing and what does not.
Shut it down and fire it up normally and u should be in business.
3.35h
1> Fixed the ftp problem, it now transfers files correctly.
2> Recompiled all support files from scratch.
3> General cleanup of some stuff.
3.35g
1> Fixed the cserver problem, it now works properly.
3.35f
1> Fixed that rexx problem with menutype 20, it now works as advertised.
2> Fixed a problem where usenet messages would not be exported by fastecho
or squishp (I hope) BK this needs some testing.
3> Fixed a possible problem in the new chatserver if a user hung up in a certain
spot, cpu load would jump to 100%. This should now not happen no matter where
he drops carrier.
4> Including my e-mail address for feedback if your so inclined, don't get upset
if you don't get a reply...
e-mail to mariod@ican.net
5> Theres still a problem with rexx where if a node is in a dos door and another
node executes an os/2 command from a rexx script it hangs that node till the
other logs off, a workaround is to enable bbswatcher threads in action flags
this will recycle the node after the user hangs up in disgust...
6> Made a change to try to prevent nntp dupes when exporting nntp to echo areas,
BK this needs your testing...
7> Made a change to the menudrvr, should be more stable now even when called from
rexx functions.
3.35c
1> Did some more work on rexx functions, there remains an obscure problem
where menutype 20 traps in show message header if called from a rexx script,
still looking at that one, other types appear to be fine.
2> I'm having a problem with cserver, everything appears to be functioning
properly, but no chat takes place, it may or may not work for you, if it
does not work, try the cserver out of the adept distribution archive, it
may work better...
3.35
1> Added a new feature, if u address a message to **All, it will automatically
go to every user in your user base. When any user in the user base logs on
he will have a new mail waiting for him, and when he reads it it will be
addressed to him personally, users can also post messages to **All (i was
of two minds about this) and have the message seen by everybody. Xmsg correctly
supports this feature to (new xmsg included in the update archive and full
install archive). Oh it has to be in a PRIVATE message area for it to work.
2> Fixed a nuisance trap in adeptrexx library, having to do with memory allocation.
3.34
1> New OrionFtp.Dll, fixes the crashing on disconnect during file xfer problems
3.33
1> New Xmsg.exe, this one has the ability to gate echo messages into Orions
NNTP engine for posting to usenet. Note this works for SQUISH areas only.
So u define yur usenet areas as squish areas, scan em to your downlinks
with fastecho or squishp or whatever, then after u toss incoming mail,
invoke XMSG -W, this will take all newly arrived messages and post them
to NNTP, then next time Orion polls the server, the messages will be posted
to the appropriate groups.
2> Fixed a nasty bug that would hang the desktop on a newuser logon occasionally.
3> Fixed the nasty adeptrexx script hanging a node when address cmd statemnts
were used, now instead of using ADDRESS CMD "Some OS2 Command" use
rc = "Some OS2 Command" and your golden. This applies to all adeptrexx scripts.
Typical hangs were in atend.cmd scripts, look for address cmd statments, put this
exe in place and your hanging nodes should dissapear, they did here.
4> There is a known problem with orionftp, if a user logs on starts a transfer
then kills his client during the transfer, orion will shut down, no trap.log,
no popuplog.os2, it just dissapears, a normal abort is fine, just killing the
client during a transfer kills orion for some reason, I'm still working on this one
so u may or may not want to use ftp in the meantime, but at least you know
about it now.
3.31
1> We now have a working ftp server, simply place OrionFtp.Dll in yur \Orion dir
and enable ftp server in bbs action flags, your file areas are now available for ftp
anonymous login or username login, all file restrictions apply.
I have tested it with os/2 command line ftp (works perfectly) netscape (works
if u rememeber to put the / after directory names) and with gibbon ftp (works
as expected.) Other clients may or may not work.
2> Fixed a small DTR problem when hanging up the modem on an offhook shutdown.
3.30
1> Fixed a trap condition in startup related to loading addresses for each line.
2> Made the msg and file area loading routine more friendly to the cpu, startup
with 10000 areas defined takes about 30 seconds to complete now. Be Patient.
3> Fixed a problem in file uploading entering descriptions, this was self inflicted
but i never noticed as i don't allow uploads here, neway its fixed.
4> Corrected the slow userlist building problem in useredit.
5> Fixed a problem in Cserver's display
3.28
1> Fixed a trap in the new display code if there was no event file defined or
loaded, orion would trap, now it does not.
3.27
1> Added a display of time of mext event to the Current field of the
status window displays.
3.25
1> Fixed a transient logoff induced trap.
3.24
1> Added a "What area to start at" prompt to the list message areas function,
on a system with lots of areas, it was tedious having to restart at 1 all the
time. This way a user can interrupt the list and restart where he left off
or jump to a location near the end of the message areas. When he hits ? at
the prompt to list areas, he will be prompted for a name or number, simply
hitting enter lists starting at 1. Note this function respects the users
set preferences for message areas, if the start area is not in his preferences
it won't list anything.
2> Added Trashcan2 and trashcan2.asc/ans/avt to the system, trashcan2 if it exists
in the system directory will be checked before saving a message entered with
the full screen or line editors. If a match is found trashcan2.a?? is displayed,
the users time used for today is maxed out and he is then disconnected.
Search algorithm is the same as that used by trashcan logic, so check the docs
to refresh your memory if u want to use this feature.
3> Recompiled with a bigger default bbs stack.
3.22
1> FINALLY!!! this squish msg dates business is put to bed, no more funky dates
in max from orion generated messages, and vice versa. This should fix the
nntp bad dates when reading from an offline reader the areas written by
orion too. I successfully share message bases between my orion setup and a
test maximus setup and everybody is happy, I regularly toss 5000 messages
a day into my echo areas with squishp, and again everybodys happy, I import
nntp (a few areas) into my squish base and everybodys happy. I run sqpack
nightly and again, no probs. Fleetstreet is happy too.
2> Ground up compile.
3.20
1> Fixed a bug in nntp importing to squish message areas, the signature (squish)
was not being written correctly.
2> Made a small change in the date routine for the above function too. Hopefully
now offline readers wont have the date field screwed up in their display.
3> Fixed a problem where e-mail messages were not having the read bit set if
the area was a squish base. PMIIndex would find the same message over and over
even if it had been read by the addressee.
4> Fixed a problem with squish message bases, Max now has NO problems with
Orion created bases, cept max displays the date differently somehow, don't
know why since it's stored as a utc date entry, but orion has a prob with
max generated dates too so i need to figure this out yet, but it's working
cleanly in all other respects, max/orion can share squish bases if u like.
3.19
1> Worked on squish some more, it's now possible to see the hidden kludge lines
from adepts reader via the @ key just like in adepts message base format.
2> Added a PID: to all messages written in adepts editor.
3.18
1> Fresh compile
2> New Xmsg.exe, loads the areas file for message base 40 times faster, this
cuts run time down a lot, especially if u have 10000 areas to load like i do.
It was also loading it twice if u specified just the -z switch (pmi's only run)
now it avoids the double load if just pmi is selected.
3.17
1> Fixed a problem where Orion could not read a squish base freshly created
by squishp or fastecho, it now reads them fine.
2> Recompiled from scratch.
3.15
1> Fixed the offline mail traps, instead of simply letting it trap and log
an error message in the trap.log, it now handles a logoff without
generating an error or trap.
2> Worked on the callerlog function, its now possible to see *ALL* the callers,
a page at a time, instead of just the last 18 to the system, the nonstop
function will scroll the whole file if requested.
3> Added a total calls for the day display to Callerlog, it now displays totall
calls for each day at the end of the days display..
3.14
1> Changed the names of the cfg files from Adept.Ini to Orion.Ini and from
NC_Colors.Ini to MD_Colors.Ini, rename or copy those two files if you want
to use your existing colors and positions, otherwise defaults will be created
that you can customize to your liking.
2> Corrected a problem in offline mail exporting, if a user selected all
message areas with monitor processing turned off then started a download
and then logged off before downloading the mail, Orion would trap, now it
just logs an excpetion and carries on.
3.12
1> Worked on the adjust message/adjust file areas options.
2> Worked on BK's call comes in, orion calls back out after call is complete
problem. Needs testing...
3> LxLited all the dll's, they are included in this archive, BACKUP yur existing
dll's first. They work ok here but u never know...
3.11
1> Fixed a problem in the telnet client where some characters were being missed
if used from a local login, remotes logins worked fine. Locals tho were
screwed up.
2> Changed the escape character to end a telnet or socket session from ~ to
control x (note this is for the local login only) since some people may want to
send a ~ character during a session.
3.10
1> Added default port to the telnet client, user can now just hit enter
for default port of 23.
3.09
1> Squish pmi's work properly now.
2> Removed userbase flush code and uses Nitins usersrvr.dll (included in this
archive) which has that code in it.
3> Lxlited the exe, cuts the file size in half.
4> Recompiled from scratch.
3.08
1> Attempted to make orion ( and xmsg) support the msgread bit flag in squish
msg bases, I need U bk to test this out, see if the pmi make works any better
now when running xmsg and uve read a message in orion... lemme know...
2> Fixed the broken {ZZ meta variable, should work as before now instead of
coming up unknown meta variable....
3> More work on squish PMI's, new xmsg in this archive.. and orion now *should*
properly support the message read bits in squish bases... BK lemme know..
3.07
1> Did some more work on the new chat server/client arrangement.
2> Adjusted the time slicing a bit, it's friendlier than ever now.
3> Changed the /Invite function in the new chat a bit, it now prompts for a
node number to send the invitation to.
4> Recompiled from scratch.
5> New Xmsg that properly updates PMI index for squish bases as well as the
regular adept style bases. It does both in one run, one after the other
so u can mix msg base styles if u like.
6> Made it even friendlier, Orion with 8 lines waiting for call now consumes
approx 1% cpu time as measured by topcpu monitor.
3.06
1> Added the who's on line display to the new chatserver/client, who's online is
invoked by typing /o
2> Added invite another node to the chat lounge function to the new chat
server/client. This is invoked by /# this would invite node # to the chat
lounge. IE /2 would send the invite to node 2. Note there is *SOME* protection
to ensure an invite is not sent
to a user downloading a file or off-line mail, seems to work but well have to see.
3.05 update
1> Added menu type 937, what this does is use os2chat3.zip files as the chat server
basically a replacement chat system. Be sure to unzip os2chat3.zip into a directory
somewhere, edit the os2chat.pwd file to change the default passwords, then in
that dir type runchat and your new chat system will be on the air.
There is documentation included with the chat server, be sure to have a look
at it. The old cserver is still there of course, but this gives an alternative
system.
2> Made it possible to use the telnet client and socket client from a local
login, it will now function as if u were using it from a remote login, with
one important exception, to exit u must hit ~ which will terminate the
telnet or socket linkup. This ~ escape mechanism is required since there is
no carrier detection on a local login.
3> Finished the AdeptxBBS to Orion naming changes, I think I got em all and
changed gatekpr and xmsg to be consistent too (included in this archive)
lemme know if I missed anyplace.
3.04
1> Added a "You are out of time, would u like to withdraw from the bank [y/N]"
prompt to the login sequence *IF* a user is outta time for the day *AND*
they have time in the bank, otherwise they get kicked as before.
2> Complete Recompile
3> Included a new language.text to reflect the name change in all the
strings and prompts.
3.03
1> Changed the name to Orion.Exe and the wav file it plays to Orion.Wav,
a sample is included in this archive, just place it in the same dir
u start Orion from.
2> Orion now consumes just 2% of the cpu with 8 nodes waiting for call, a side
effect of this is it can take up to 2 rings before adept answers a call
even if u have answer on ring 1 set. No biggie as far as i'm concerened.
3> Recompiled the whole shebang with different options.
3.02
1> Changed the name to Orion CBCS/2
2> New Cserver.Exe Included in this archive, logging is now controlled by
/Log on the command line, default is logging off, /Log turns it on, default
logfile is cserver.log in the dir cserver is started from.
3> Added NotValid.Asc, this file will be shown when the user hits a key thats
not a valid menu item, if the file does not exist the menu will simply be
redisplayed on an invalid entry.
4> Fixed the at end watcher thread, it was sometimes killing off Atend.cmd
before it had a chance to complete
3.01
1> Added a display file called NotEnabled.Asc if Portflags for a menu item
are set to 4, this file will be shown to the user and that menu item
will not be executed.
3.00
1> Expanded the name field of the telnet client option to 50 chars.
2> Implemented a socket client, menu type 936, works same as for telnet
client but inititates a socket connection stead of a telnet connection.
NOTE: Both these clients require an active connection (modem/telnet) to link
to, so these two menu types will not work over a local login. Telnet in or
call in to test em out.. they will work with the local loopback too, so u
can telnet/socket connect to yerself...
2.99
1> Fixed a problem in internode chats private messages were not working.
2> Fixed and implemented Nitin's Telnet Client Code, menu type 935 now
activates the built in telnet client, it will prompt the user for an address
and port number, then try to establish a telnet session with that address
and port, if successfull, the user is connected and in a regular telnet session.
if it cant connect or times out, it returns the user to the menu that called
the type 935.
2.97
1> Recompiled cserver
2> ReCompiled Avi.Dll
3> Recompiled GIOmt.dll
4> ReCompiled Adeptx and AdeptRexx
5> Fixed a bug i introduced in the time per day calcs while fixing the
time used today set to greater than time allowed per day function.
This can no longer happen, even if they use the time bank.
2.96
1> Did a little work on the prompts subsystem.
2.95
1> Added a Kill Session to the Line popup menu on the adet status screen, the
way this works is it sets the users logoff time to *NOW*, adept sees that
and logs the user off. Great for line 0 where toggling dtr does nothing.
2> Enabled BBSWatch threads for line 0 now too, used in conjunction with the
above feature, u can now have a line 0 bbs session with a hung rexx script
be autokilled instead of having to shut adept down to recover. Simply
click on Kill Session for the local login and the bbswatch thread will
kick in and kill the hung bbs session, (Can take a few minutes) and let
it cycle back to waiting for caller.
I did this cause I'm always testing adeptrexx scripts on the local login
and have many many times hung the local session, requiring a shutdown
to recover from it. Hopefully those days are gone for good...
2.94a update
1> Fixed the broken sysop chat
2> New Avi.Dll cause I got a trap with the old one.
3> Removed a lot of kludges I had put in for the 107f giomt.dll.
2.94
1> Incresed the stack space for mailer and bbs threads to a accomodate a change i made
earlier, 2.93 would trap out on startup with stack overflow cause of that change,
fixed in this build.
Recompiled from scratch, both adeptx and adeptrexx.
2.93
1> Possibily fixed those locked files, has not happened here since putting this build
in place. Lets see.
2> Added GioMTErr.Log file logging to giomt.dll, lets see if this helps track
down some of those obscure traps in GioMT.Dll.
3> Added the chatlog (Cserver.Log) back into cserver.exe
2.91
1> Did some more work on the locked files thing.
2> Made a change to user timebank withdrawl routines.
3> Made a change to name entries on log on, should not be any functional
differences.. so far so good..
2.90
1> Had those locked files again, I'm beginning to suspect it's an os/2 problem
since this time popuplog.os2 was locked too, and I know adept does not
touch that file in any way shape or form, that file belongs to os/2.
Recompiled from scratch with a couple different compile time options, Avi
too, so lets see how it goes this time.
2> New cserver.exe in this archive, another attempt at curing those phantom
chat sessions in cserver.
3> New GioMT.dll too.
2.89c update
1> Fixed a problem I introduced into srif processing, where the cmd was being
ignored. It now properly processes the command.
2> Fixed a mite (long standing) in the request session logging, the logfile
had an anomaly in it, it's now correctly logged.
2.89b update
1> fixed a problem with the portflags keyword in menuitems, if a port was telnet
and u restricted a menuitem to comm ports only, it was not being honored,
now it is.
2.89a
1> More work on those locked files symptoms.
2> Got a trap in the menu driver, did a bit of work there, hopefully no more
menu driver traps.
3> External sessions revisited, in hopes of clearing up those locked files.
This is a shot in the dark, just in case it's related.
4> Fixed a prob in the user handle setting routines, it was allowing way more
chars than the field holds.
5> Worked on the who's online routines and associated meta convert routines
basically looking for overwrites, didn't find any but expanded the fields
just in case.
6> Fixed that quirk in GioMt.dll re starting adept before cserver, now it does
not matter which u start first. Works either way now.
2.89
1> Fixed a trap in smtp email entry code, if a user filled in the entire To:
field and hit enter, Adept would trap. This is corrected now.
2.88b
1> One quirk about this new GioMt.Dll, I've found that it traps on completion
of an inbound mail run *IF* u started chatserver first. If u start Adept
first, then start chatserver, this problem goes away. The trap is in the c runtime
code, so i'm not sure if it's a watcom compiler thing or what, but for now
start adept before u start cserver, and all should be well.
2.88a
1> A new giomt.dll, the other would trap whenever a user attemted to telnet
out of adept using the telnet script. This one fixes that.
2.88
1> Another attempt at solving locked files problem.
2> A freshly compiled GioMt.Dll, this is my first compile of gio here since
getting the source code, seems to be working ok here, at least if this one
traps, I may be able to do something about it.
3> A newly compiled Avi.Dll too, trying a few different compile time options.
Seems to be working ok so far here too.
4> A newly compiled cserver.exe too, again, different options used at compile
time. Looks good so far too.
5> You should use all the dll's and exe's out of this archive, after backing
up your working setup, and please send me any traps u may get. Thanks...
2.87
1> I'm still getting locked asc/ans and menu files occasionally, this build
is another attempt to fix that annoying bug. Lets see... BTW is anybody
else seeing this ??
2.86d update
1> Whole new set of I/O dlls, use *ALL* the dll's supplied in this archive.
2.86c update
1> Fixed another rexx trap.
2> New Avi.Dll to fix an optimization problem in the last release.
2.86b update
1> Fixed two more trap conditions in adeptrexx (very rare occurrences) that
my ever curious users found when playing with adeptirc.srx.
2.86a update
1> Corrected a trap in the count new files thread, 2.86 would trap at logon
if there were new files on the system. This build corrects that.
2.86
1> Added a Reload Events File option to the programs pulldown menu, clicking
on this option will reload the events file for the active line.
2> At long last the no new files flag in the files areas works as designed, adept
will now ignore areas wit this flag set when doing a new files scan, either
from the menu options, or a new files scan at logon, if that flag is checked
the area is simply skipped without being scanned for any new files.
2.85
1> Adept will now hang up on the users online when u invoke a shutdown, I got
tired of manually hanging them up when i wanted to shutdown.
2> Corrected a problem in AVI.DLL, it would trap when trying to logon to a
/secondcopy of adept. It now works correctly.
2.84
1> Had a problem where the main menu ans screen got locked somehow during
a users session, result, no main menu display, had to reboot os/2 to unlock
the file. This build is an attempt to prevent that sort of thing from
happening again, lets see... (it was not a recurring problem, just
happened once in a very long while).
2> Included is a new avi.dll to (hopefully) cure stack overflow probs with the
older one.
3> Worked on the adeptshow routine to make it more freindly to meta codes in
the text, seems to work better now.
2.83a update
1> If the users file could not be opened for some reason adept would still
try doing a login, resulting in a crash. It now just logs the fact in the
linex and adept logfiles, and recycles to wait for another call.
2.83
1> PPPSniffer will now honor users time per day and time per call settings, so
a user can't stay in a ppp session indefinatly.
2> Cleaned up a trap condition in the logoff sequence, it would sometimes
very rarely trap on a user logoff, should not happen anymore.
This showed up in PPPsniffer if a user dropped carrier while adept was
trying to determine a ppp session it would sometimes trap.
3> Found that starting 64 nodes would still sometimes result in some nodes
remaining stuck in a starting state, this is now hopefully cured. I tried
10 64 node startups in a row here and was unable to hang any nodes in
starting state, before I would get one or two every 3 or 4 startups.
Hopefully this is the end of that nonsense.
2.82c update
1> Fixed the multi nodelist lookup to find more than just the first 2 objects
being looked for, now clicking on doit will find the next match till all
matches have been found, hitting clear resets the search data, so if u
enter another search string, the first match will again be pulled up.
2.82b update
1> Adept now properly fills in the SYSTEM field in srif files with the name
of the remotes software, ie binkley, frontdoor or whatever the remote is
using.
2.82a
1> Corrected another problem in the bbswatcher threads, they were killing off
a bbs session before the bbs session had a chance to kill a hung door here.
The hung door is now properly killed off before bbs watcher kicks in and
kills the bbs session (only if necessary now).
2.82
1> Well that trap handler is only for VAC, will not work on wat builds, so
this build has all that extraneous code removed.
I'm still interested in any trap.logs or popuplog.os2 entries generated
by adept, so please e-mail these to me if u get them, yur logfiles at time
of trap would also be useful.
2> Fixed a problem in AtEnd watcher, it was terminating before doing nething,
now it properly watches the atend processing.
2.81
1> Added Nitins Enhanced trap handler to this build, please turn exception
handling flag on in bbs->config->action flags, then look for a trap.log
in \adept dir. If u get one e-mail it to me at mariod@ican.net. Note you
will have to check for the log, there is no other indication that a problem
occurred, other than the action that caused it will be aborted.
2.80b update
1> Fixed another trap condition (rare) that would show up when a user dropped
carrier at various places in the logon sequence. Hopefully there aren't
any more like this, if u get traps, especially in GIOMT.DLL please try
to figure out what was going on when it trapped, ie user dropped carrier
in the message editor or whatever, it'll help a lot to know what the circumstances
at the time of the trap were. The logfiles can help u track down what the guy
was doing at the time, especially if u start adept with the /NC_VERBOSE
switch.
2.80a update
1> Did some work on cserver, removed the truncating at 255 chars when sending
chat msgs. Made another small change in cserver trying to eliminate this
phantom chat syndrome that shows up on busy chat systems...
2.80
1> Did some more work on those pesky GIOMT.DLL traps, found a couple more
conditions that could cause it to trap in gio, these are now gone.
2.79b update
1> Revisited the startup sequence once again, on large systems, nodes were
still getting stuck in a "starting" state. This build should cure that
once and for all.
2> Found a couple more spots that could cause giomt.dll traps if a user
dropped carrier on the system, these too have now been patched to avoid
the giomt traps. Lets see how it works out now...
3> Fixed a rexx trap that showed up when a user quit the ftp script after
getting a few files, rexx would trap on exiting the script, this is now
fixed. Same symptom I suspect plagued the telnet snoop script, altho I haven't
texted that one, it too should now work as advertised. If it doesn't, please
e-mail me the trap vector (if it still traps on exit) and I'll see about fixing
that one too...
2.79a
1> This is an experiment, I have tried to patch around the various GIOMT.DLL
traps when using the older set that fixes the zmodem telnet downloads, I'm
gonna run it here for a while before putting it out (if it's a success that is)
for general distribution. Let's see how it goes.
2> Srif command, if enabled will now be executed on receipt of any file instead
of just req files, this is so those using netmail .pkt files to make a request
will still invoke the srif command, this also means u'll have to move yur
mail.cmd into srif.cmd if u use srif since yur mail will not be processed
otherwise, if srif is disabled, mail.cmd will be invoked as usual.
3> Worked on the startup sequence once again, hopefully for the last time.
4> I'm including the 107f dll's that fix the telnet zmodem download problem
in this archive, hopefully Ive patched every spot in adept that was causing
the giomt.dll traps evident in these dll's, so far they have not trapped on
my system, so i'm gonna release this build with these dll's.
5> Did a huge amount of work on the bbswatcher threads, these ones should be
a whole lot more reliable, it will now also check the users time and if
time has expired, it will kill the bbs session, this means no more rexx
script users overrunning their time limits.
2.78b Update
1> So ends the experiment of the older set of dll's, while they fix the telnet
zmodem problem, giomt.dll is too unstable to use, it traps randomly when a
user drops carrier in a door or in chat, so this archive contains the set
that works here. Telnet downloaders (adept telnet, not vmodem telnet) will
just have to use xmodem instead of zmodem if they wanna do downloading, at
least till we can (if ever) get the source to giomt.dll and fix it.
2.78a update
1> Corrected a trap condition when using bbswatcher threads and a user
dropped carrier, it would sometimes trap in giomt.dll, this has been
corrected.
2> Included a new cserver.exe, recompiled it with the same headers used in
adept to resolve some inconsistincies in the structures used by cserver.
2.78
1> Worked on the startup sequence some more, was still hanging nodes on
large systems.. lets see if this works ne better.
2> A bug has been reported in GIOMT.DLL, this bug causes zmodem downloads
not to work when a port is set up for adepts native telnet, Included in
this archive are a set of dll's to fix this problem. Backup your current
dll's first then put all these in place.
2.77b update
1> <sigh> corrected another trap which only showed up when mixing rexx
and standard menus (as i do here), if u don't mix the two you would
never see this trap. Anuhoo... it's fixed now.
2.77a update
1> Corrected an memory allocation error (possible system lockup) in menu type
301 (show text file) when doing searches.
2.77
1> Fixed another problem with the offline exporting functions and compressed
message bases, they now export the compressed messages properly.
2> Fixed the PPPSniffer code so the user does not get knocked offline after
15 minutes nemore. He will now stay connected till he drops carrier or
your autoppp script hangs up on him.
2.76
1> Did some work on the bbswatcher threads, now they will check to see if the
bbs thread has ended before issuing the killthread signal, instead of blindly
issuing it on a carrier drop. Should make it a tad more reliable...
2> Did a bit more work on the time slicing routines for idling nodes, should
be a lot more friendly now, note that if u have a *BIG* events file for
any node, startup of that node will take a few seconds more than before,
as will shutdown, but the friendliness gained is worth this small performance
hit i think. Events will still execute as per normal tho.
2.75
1> Fixed a problem where the offline packers were skipping messages if compression
was being used in the message bases. They were not packing some comprsessed
messages. They now properly pack *ALL* messages in the base even with
compression turned on. This has been there for a while but I just recently
started using compression and noticed it, sorry it took so long to see it<G>.
2.74a update
1> I had a couple traps here, this build is sposed to address thhose,
note that there are new avi and usersrvr dll's in this archive too.
2.74
1> Reworked the *.msg routines, so they no longer write the adept msgbase
files as well as the *.msg files, the only adept style file now is the
LMR file which is still needed, otherwise u get *.msg and nothing else.
2> Found there is a problem with that telnetsnoop script a lot of people
are using for telnet out, till I can get it resolved, i have included
the telnet.srx file I use here, it works and works well, you just won't
be able to monitor what the user is typing since it is a *TRUE* telnet
link to the users comm port, unlike that snoop which did block i/o (and
burned cpu at the speed of light, I might add) but if u want fast reliable
telnet out capabilities, this script is the way to go.
2.73c update
1> Hokay... found the *.msg problem, I can now read/write *.msgs created by
an external source as well as those created by adept, sorry it took so
long bk...
2.73b update
1> Found a situation where the squish msg files were not being closed under
certain circumstances, this results in 3rd party apps not being able to
access the squish base in question, they should now be properly closed
in all cases.
2> Removed the *.msg fix i had done earlier, trying it another way, lets see...
2.73a update
1> Changed the default behaviour of autoppp.cmd to non-cached, it was cached
before, and this could cause probs during setup and what not, so now it's
non-cached. Note the file *MUST* reside in \adept\menus for it to be executed
by the sniffer code, i mistakenly told dutchie it belonged in \adept, sorry
Dutch..it belongs in \adept\menus dir.
2.73
1> Made the PPPsniffer routines into a sysop option, they will only be enabled
if the bbs action flag "Enable PPP Sniffer" is turned on in bbs config action
flags, if this flag is not turned on the ppp auto detect code is bypassed
altogether.
2.72
1> Tweaked the time slice routines, instant response now while not giving up
the cpu friendliness.
2> Added PPP session detection to adept, now after a user enters name and
password, a check is made for a ppp session, if it's a ppp session the
cmd file AutoPPP.Cmd is executed if it exists, after this rexx script has
completed, the user is logged off. A log entry is made in the linex.logfile
as to what type of session is being executed, would appreciate some feedback
on this option as i have no fixed ip#'s to test it with here, all i have been
able to do is detect a ppp session and start the rexx script. You should be able
to modify the ppp door to run when a ppp session is detected, or modify the
enclosed ppp.cmd file from Greg Varga and have it spawn ppp for your users
who call in with a ppp session, the configuration requirments are all
documented in the script, look for autoppp.cmd in this archive.
2.71d update
1> seems i got carried away with giving up time slices <G>, this build restores
the bbs threads to instant response while maintaining the cpu friendliness
of the previous builds.
2.71c update
1> Did some more work on the menu driver, hopefully the last od the exceptions
is repaired now.. lets see...
2.71b update
1> Fiddled some more with that domain name stuff, lets see if it's fixed now.
2> Note.. due to the reduced cpu requirements, some operations, like reiniting
the modem or hitting immediate call in the outbound may take a few seconds to
be acted upon, no biggie, they'll be executed next clock tick.
2.71a upgrade
1> Added a newly recompiled cserver (faster, less cpu hungry) to the archive.
2> Further tweaked adepts cpu demands, it's now almost impossible to measure
the cpu usage adept uses, even with users online, I have here a total of
25 tasks running with 8 nodes of adept up and a user online reading
messages, avg cpu usage is less than 3% and most of that's due to the other
tasks, not adept... as measured by pmpatrol with the time critcial meausring
switch turned on.... adept is truly cpu friendly now... :>
3> Took the events scanning and put it back the way it was, found a better
way to make it friendly, u can disregard the modem reinit timer stuff, it's
just for the modem reinits again...
2.71
1> Undid the damagae i did to the old style menudrvr, it should work again
as it did in previous versions.
2> Took out the hard coded limit of 8 chars on the domain name comparison
in the recvd emsi data... Gordon, yur linearnet address should parse
correctly now.
3> Upped the idling sleep by a few more milliseconds, cpu usage is cut
drastically now...
2.70a update
1> Tied the event end check to the modem re-init timer, what this means is that
instead of constanlty checking to see if an event has ended ( and hitting
the cpu all the time) it will now only check it every x minutes, x being the
number of minutes u set the modem reinit time to. So set the reinit time
to something less than yur shortest event, ie if yur shortest event lasts
20 minutes, set yur modem reinit to 15 or 10, it will then chek the events
every 10 or 15 minutes..
2.70
1> Fixed another trap in the menudrvr, note i made sum changes to the old style
menudriver too, would appreciate any feedback on problems in that driver
since i don't have old style menus here to test with.
2> Made the wait for call threads *REALLY* cpu friendly, i get virtually 0% cpu
use if all nodes (8 here) are waiting for call or event.
2.69a
1> Added Hexer's *.Msg fix for reading *.msg messages not created by adept,
this should now work properly... confirmation neone ?? I have nuthing capable
of creating *.msg files here cept adept itself..
2> Corrected a problem in the menudrvr, which could result in a trap in menudriver
on occasion, or if u had exception handling turned on this would show as an
entry in trap.log, usually showed after a new user logon seq was completed,
yeah, check yur logs folks..
2.69
1> Expanded the domain name field to 50 chars, this outta take care of networks
with Looooonnnnggggg domain names, adept was honoring the old fidonet limit
of 8 chars for a domain name, the 1.11x series had this fixed quite sum time
ago.. I just plumb forgot about it, sorry... :>
2> Tweaked a bit more speed outta the code, this build should be even faster
than the last, will be very noticeable on >26400 connects...
3> *NOTE* because of the address domain name expansion, yur linex configs
must be edited or recreated, yur choice.. I simply del mine and recreate
them with adept /nodes x /setup...
2.68
1> Fixed a long standing telnet bug in adept, telnet connects would sometimes
hang the nodes, even if using vmodem, note this would only happen if internet
timing was checked for the node in question...
2.67
1> Ok, now that this stealth business is over and done with, we can bring
the local logon and local bbs screens back into the picture for those folks
who simply cannot or will not trust the kids or their users. I just wanted
to see what kind of outcry there would be, and it seems minimal, I respect
the right of each sysop to run their particular bbs as they see fit, sooo...
In my case the only time I want to even *LOOK* at the bbs is if I see the
sysop mail flag waiting for me, otherwise i could care less what the folks
are using it for.. I trust them to police themselves.... nuff said..
2> In the last build i changed the linex configs, sorry i forgot to mention
the bbs config needs to be deleted and recreated too, just delete
bbs_configuration in the \adept dir and let adept recreate it, the speed
increases is dramatic folks. Also don't forget to delete *.bin in the
\adept\menus directory...
This version should be virtually bug free folks, also included is a new faster
usersrvr.dll, backup yur old usersrvr.dll file and drop this one in place
along with the new exe, and be sure to put the action flag all files
commit on timer to on in the bbs action flags, this will force user file
to be flushed to disk on a regular basis (if a user is logged on of course)
That's it folks, thanks for this vote of confidence, there was really
very little public outcry about this stealth business <grin>.
3> Fixed the bug where short messages were not being saved if compression
was enabled in bbs action flags, they now save properly, and u already know
gatekpr and xmsg honor the compression schemes, there is no limit to how
small a message can be to be compressed, if u like u can set gatekpr to
pack size of 1, it will work properly, but i recommend a value of 128 bytes
for gatekprs packsize... good stuff folks, saves disk space..
Just enable the flag in bbs action flags, xmsg automatically compresses
uncomprsessed messages and put the pack size keyword in gatekpr.ctl
2.66
1> The rexx bug workaround does *NOT* work, it's definilty a bug in OS/2 rexx
boys and girls, IBM is on the case and it will be resolved. They promised.
Had a request to spawn an external message reader for the adept message
base, since there is no longer a local login, this seems a reasonable request
, so we have added to the pulldown (programs pulldown) the external mail reader
command, this menu choice will activate the cmd file "ReadMail.CMD" if it exists
otherwise it does nothing, u can have readmail.cmd execute whatever u wish, it
is intended tho to fire up golded or sumthin similar so u can read/write messages
without logging on, I prefer logging on but i understand there's lots of folks
who wish the exetenal mail procedure.
2> We have once again sped up adept by a factor of ten or so, this change requires
you to delete all LINE_X_CONFIGURATION files and LOCAL_CONFIGURATION (no longer used
neway but...just to be complete) and recreate em, save you old init strings before
destroying the old config files. (Fuq right off frank, I'm busy at the moment, listen
to Sunny Came Home Asshole). Once u fire adept back up in /setup mode with the
number of nodes you run, simply edit the init strings and com ports numbers as
per usual, then shutdown and fire it up normally. You're now ready for callers
and yu'll be amazed at the speed difference. Have fun...
2.65a update
1> Worked on that rexx bug some more, found a possible workaround, it's in
this build, let's see how it goes...
This one is for Jason.... :>
2.65
1> Been a while since a new hagar build was put out folks, (see, very few bugs
left in this code, and those that are left, well... they won't last long)
Anyway... this build is mainly a ground up recompile of the whole entire
thing, rexx included. Speaking of rexx, we've found a problem with adeptrexx
spawning multi-threads from a single script, this problem is buried deep in
the guts of os/2's rexx que interface, IBM is aware of our problem and has
promised to do their very best to eliminate this long standing mite (it was
reported previously) and I have no doubt at all that they will fix it. IBM
does good work folks, don't doubt them, I don't :>
Thanks to Jason in OZ for spotting this particular bug, even tho it's not
in adept, we tried a few work arounds, but IBM informed us it'll be fixed
asap. Current Fixpak for warp3 users is fixpak31, but be aware this breaks popuplog.os2
logging, the next fixpak has this fixed, and will contain the rexx que fixx too.
People using rexx scripts will have lots less probs when this fix is released.
God Speed Folks...
2.62
1> Solved a problem with Avi.Dll trapping under certain circumstances, this
was brought to light by out old friends at lakebbs, included in this archive
is a new avi.dll and cserver.exe, as well as a groundup recompile of adeptrexx
.lib and adeptx.exe
2.61e update
1> Played with the startup sequence again, hopefully no more stuck nodes in a
starting state.
2> Rebuilt Avi.Dll from scratch to try and help this same problem.. put this avi
dll in place after backing up your old avi.dll
2.61d update
1> Fixed a path problem in the generated srifx.txt files, should now be proper.
2.61c update
1> Fixed the srif entry field, I broke it last couple of builds, it now
works as before..
2.61b update
1> Fixed the broken local upload function, it now works as advertised..
2.61a update
1> Corrected a trap condition in srif prcoessing, if a file that was not in
the file base was written into srifx.txt, adept would trap after failing
to find the file, now it just ignores it..
2.61
1> Recompiled with a small change to try and sort out logoff traps on some
large systems, can't duplicate it here, but mattis gets them regularly on
lakebbs in oz. Nebody else getting traps in menutype 1000 ??
2> Included in this archive is a newly compiled Avi.Dll, may or may not
ease the avi.dll traps some people have been getting..
2.59b update
1> Added a little feature, if the system finds an asc/ans file with the same
name of the user logging into the system in the default text dir, this file
will be shown to the user then deleted.Ie if a file name Mario Dulisse.Asc
existed in yur text dir and i logged on it would show me this file and then
delete it, all metas can be used in those files. The file is shown after
atlogin.menu has completed, but before newuser.menu has run, so you can get
really creative here, and have your atlogin.menu create a custom file that
will be shown to newusers at login time, then deleted, it's a simple matter
to run an adeptrexx script in atlogin.menu that would automatically create
a custom weclome screen for each new user who logs on, then delete it. And
since u can use the name meta in that screen u can even personalize it to
the user logging in... thanks to north star bbs software for this idea..
2.59a update
1> Was still having stuck nodes in starting state problems, this is another
attempt to address that, usually shows up when starting lots of nodes.
2.59
1> Fixed the Jam Message base support so it works now <g>, there were quite
a few spots I overlooked plugging in the jam support on the first pass.
I Have tested read/write and they both work as advertised. Read the 1.11
series readme for the caveats on using Jam bases...
2.58
1> Plugged in hexers JAM code, dunno if it works or not, I have no way of
testing it here, note that adept will not create a jam base, it must already
exist. Same conditions apply as the 1.11 builds jam support (natch, it's the
same code).
2> Sorry folks i misspoke, adept will indeed create the jam base if it does not
exist, seems i read an old readme of the 1.11 series, the current code does
create if necessary..
2.57a update
1> Fixed Srif file creation so proper path and file to req list is generated,
note: If u are going to use srif files u *MUST* set the Adept envirnment
variable, example: SET ADEPT=D:\ADEPT
this path will be used for the base path of all srif files, the files will
be created in in this case d:\adept\system\linex directory, and the rqreport
.txt file will be looked for in that directory. Also the full path and file
name is passed now as the first parameter on the command line, (Note if u
use {LN, then the path and filename become the second parameter. Every meta
u add pushes the path/filename up one slot).
2> Did some more work on the load sequence, as sometimes the msgareas loader
was failing to start, hopefully cured in this build.
2.57
1> Fixed it so srifx.txt is only created if a request list is actually received
instead of being created on all inbound mailer calls, the srif cmd will also
not be executed if no request list was received.
2> Discovered that the srif cmd field really does store up to 128 chars for
a full path and file specification, just be aware that if u go back to
look at it it will look truncated for some reason, this is purely a display
bug, the actual command will be executed, u can verify this by looking in
your logfiles for the node receiving the freq. Remember to use \\ instead
of \, and by the way all the metas except those that make no sense when a
user is not logged on are available for use on this command string.
2.56d update
1> Ok.. removed the srif path field and replaced it with Srif Cmd, in here
you put the command to start your external req processor up, adept will
now write the srifx.txt file to \adept\system\linex directory where x
is the nodenumber, it will then try to spawn the external process
defined in Srif Cmd field, note that adept will wait for this process
to complete before carrying on with the file request. If u are gonna
specify a drive and path to yur req processor u must use double backslashes
in that field, example cmd.exe /c \\adept\\extreq.cmd will invoke extreq.cmd
without the \\ it will not work. Currently the largest string u can have
in that field is 30 characters, but only cause i have not yet found where
they are limiting those fields to 30 chars or less, i intend to find it
and allow 128 chars for a command entry. Am releasing it this way cause
I'm outta time for today and would like to get some feed back on this feature.
BTW putting cmd.exe /c extreq.cmd {LN will pass the nodenumber on the command
line, just like in a door menu type, Port meta will also work so u have
some flexibility here... of course this whole thing is still controlled by
the bbs action flag enable external req processor, if that flag is not set
then none of the above will happen.
2.56c update
1> added a check to see if message bases have completed loading before starting
the local login bbs thread, fiddled with the startup order again cause on
occasion the large message areas file was not completely loading, still
working on this one...
2> Cut the srif delay time down to 6 seconds from 15, this should still be
enough time for the external request processor to do it's thing...
2.56b update
1> Corrected a trap condition during freq processing if srif was being used
and the requested file was not found..(I hope) BK i need yur feedback on
this one..
2.56a Update
1> Fixed Srif creation, it was not being created at all on an inbound
request, it now gets created..
2.56
1> Recompiled from scratch, full compiler, rexxlib and adept.
Note: if u were testing the earlier 1.11y builds and had to pull your
set adept=d:\adept statements from your config.sys to get it to work, be
aware that this breaks smtp2adept.exe and your inet e-mail will not be imported
becuase of that. Solution is to grab the updated smtp2adept.exe or simply
put the set statement back in your config.sys.
2.55g update
1> Fixed a prob with srif path not being saved, now saves properly..
2.55e update
1> Did some more work on the new files at login scan code, the thread that
counted new files was searching all areas, even if the area was marked
no new files scan in fileareas editor, it will now properly ignore those
areas (why search cdrom areas for new files?) and only scan areas without
the no new files scan flag.
2> Revisisted the netmail node/address entry thing, now if a name is entered
it will search the nodelist and pull up a match, if n is specified it will
pull up next match and so on, if u answer No to the last match, u will be
presented with an oportunity to enter the address yourself, names need no
longer be in the nodelist to send netmail to.
2.55d
1> Fixed the new files scan at login to only display the new files then quit
if all displayed stead of searching all the remain areas, I thought i did this
before but it turns out it was working only if all new files were in the same
area, it now does not care what area they are in.
2> Added a bit of code to close all files on exit, this should prevent logfiles
and other open files remaining in an open state if u have to close adept
from the upper left hand corener icon and also on shutdown from pulldown
menu or trap. Hopefully no more "file is in use by another process" type
messages from os2.
2.55c
1> Fixed menu type 78 (for real this time), looks like it is broken in the 1.xx
builds to start with, which is where this code started from. Neway it now
works as it did before in the old 2.xx series code.
2.55b update
1> Fixed menu type 78 (old list files option) that i broke earlier on
by accident, works now.. 178 was not affected..
2.55a update
1> DOH!!! somewhere along the line I broke TolOw.asc ToYoung.Asc etc, all those
nice displays when a user did not match a menuitems criteria, these are all
fixed now and i have added a new one, WrongGender.A?? will be displayed if the
users gender does not match that specified in the menu item.
2> Did some work on bbs watch thread to make it more reliable and less prone
to problems, I suspect that hang i got the other night was due to a fuqed
user.name.index file and not bbswatcher, so i'm running with these guys
turned on again, I will let you know how it goes.
2.55
1> Added safehex's latest filters to the edituser find dialog box, changed the
internal storage of the srif path var to match hexers builds so we'll all
be compatible now that he's added srif to the vac builds, this means ull have
to reneter the srif path in bbsconfig settings, nothing else should be affected
but chek all bbs config settings to be sure..
2.54
1> fixed a problem with the phone numbers displaying twice when prompting for
a phone number.
I had a weird hang here last night, traced it to the bbswatcher threads
looks like under some conditions these guys can cause problems, so i recommend
turning off the watcher threads till i can investigate further...
2> Added the allow users to keep mail on popserver flags to the bbs action flag
and the menutype 637 to allow user to toggle this flag on and off, {VU displays
the state of the flag for that user..
3> Made the nodelist lookup threadsafe so multi users can do it at the same time
there was a possibiltiy of a problem before if two or more users attempted
a nodelist lookup simultaneuosly
2.53
1> Implemented hexers message base handling improvments, this stuff speeds
message base handling by 2x or more, this is a massive change so please
report any message base bugs u come across immediatly...
I have tested the basics, (read/write/bwave export) and all seems well
however I cannot give this build the standard 5+hr or more runtime i try to give
all my builds before putting them out as i am 12+ hours into a test run
of build 2.52 to see if i've cured the traps with > 7000 msgareas and
32 nodes loaded problems, don't wanna restart the test at this point. Will
be satisfied if it goes 24+ hours without a problem..
2.52
1> Added safehex's locked/unverified filters to the user editor.
2.51 update
1> Took out a dossleep i had stuck in there for debugging, please let
me know if u have any startup troubles, ie, lines not starting,
lines stuck in initing state, blank entries on the system status
window for lines etc...
2.51
1> Fixed a variable allocation problem in the new srif processing code,
dunno how I missed this one, would only cause a problem if you got a
file request comeing in, I get no requests hence my system ran fine.
2.50d update
1> put hexers new file/msgbase load code in place that saves bytes of memory
by not storing the message base paths in mem if default path is used.
During testing of this version i somehow managed to fuq my lang file and
*.bin files in \menus dir, prolly caused by the quick succession of
start/shutdown/start/shutdown sequences, if u get weird shit happening
del the lang.text.compiled and .index and all yur *.bin files in \menus
and restart. should be all right then, prolly not needed, but thought i
would mention it just in case..
2.50c update
1> Did sum more work on the newfiles display routines for user files scanned
at login, should not go thru all areas now, should stop after last new file
is displayed...
2.50bupdate
1> Yet another attempt to fix the intermittent traps when loading large
message/file areas and lots of nodes. Cut down the sleep value some
cause at times it would only answer on 2nd or 3rd ring even if told to
answer on first ring, it should answer on first ring all the time now.
2.50
1> Ok, the compile lang prompts from the pulldown works properly now, but
same caveat applies, ensure there are no users online when u recompile
the prompts.
2> added a check for the nonewfileslist flag in the file area editor, if that
flag is set the area name will not be displayed oon a new files scan from
logon new files scan request, should speed up display of new files by showing
only the file areas which actually have new files in them..
2.49
1> Ohhhhhh, did I ever screw up, In my efforts to fix the stuck nodes
problem when starting > 10 nodes, i managed to introduce a huge
memory overwrite, this *WILL* cause adept to trap if run long enough,
it's much more evident on my 64 node test setup, but will show up on
a single node setup too if allowed to run long enough... Sorry folks...
2.48a update
1> Tied the atend watcher thread to the enable bb watch thread in bbs config
action flags so those who don't want it can turn it off, note enabling this
flag turns on *BOTH* bbswatcher and atendwatcher threads.
2> Pulled the recompile lang prompts from the pulldown till I can sort out
*WHY* it traps when running more than 24 nodes...
2.48
1> Enabled 64 nodes in this version, mainly to test cpu loading on my system
but i'm gonna leave it on for nebody else who may want/require more than
the 24 nodes that was the previous limit...
2> Fixed badpass.menu so user logging off after writing a message to sysop
does not crash adept nemore, this one got me a few times today.. same goes
for dumbuser.menu.
3> Fixed a startup anomly that on large systems could crash adept at load time.
4> Fixed another startup problem...
5> Loosened the memory allocations somewhat to allow for bigger systems, on
larger systems adept was being squeezed for memory...
2.47c update
1> Made it much more cpu friendly, 24 nodes was taking 30% cpu before this change
now it takes 5% and still answers on first ring.
2.47b updated archive
1> Fixed Wikk's pet peeve aout name/nodeaddress/subject editing if u screwed
up in the original entry, arrow keys, and delete now work as they are supposed
to.
2.47a updated archive
1> Ok I think I finally have this stuck node problem licked, at least it's
not happening nemore here, feedback would be appreciated...
2.47 updated archive
1> Fixed the hit return to logoff during logoff countdown after exported
mail was downloaded, was returning to the menu stead of logging off.
It now logs you off as expected..
2.47
1> Fixed the make line inactive from status window function, it now makes
the selected line inactive stead of the one with the focus.
2> Had another go at this stuck in starting state nodes on large systems,
lets see how it pans out..
2.46 Updated Archive...
1> Updated the cserver.exe file in the archive, otherwise no other changes
so this archive is named 2.46a but the adept version number is still
2.46
2> Ok, another update, the exe adeptx has been updated too in this update,
in an effort to solve nodes getting stuck in a "starting" state on systems
with large msg or file bases, the nodes will start and adept will be ready
for calls long before the msg and file areas have finished loading now,
this should allow all nodes to start properly, but be aware, u can screw
adept up by hitting shutdown before the msg/file bases are fullu loaded.
When adept gets confused like that it more often than not takes a reboot
to clean it out of memory.. moral of the story, is don't shutdown adept
before it's finished its msg/file area loading. Only u can tell how long this
takes on your machine, since it is highly dependent on disk speed, cpu speed
and size of msg file areas configured.
3> adu246c.zip the adeptx exe has been updated yet again, this time to support
the /Lockout IP# switch as per 2.34a code, u can have up to 20 /Lockout
statements in yur command line, the IP# specified in /lockout will not be
allowed to connect via telnet or vmodem to yur copy of adept.
4> adu246d.zip, the exe has been updated again, this time in an effort to resolve
chat induced traps in adept, seems memory may have been set way too tight
for most syschat functions, i have relaxed the memory assignements quite
a bit, lets see if it makes a difference..
5> adu246e.zip hrmmm maybe it's time to bump the version number, neway this build
removes the two times display of login.a?? file, but retains the feature of
redisplaying it if the user answers No to the are u a new user question.
6> Adu246f.zip update
Fixed the = selection at the more prompt for displaying msg/file areas, it
was working for local logins, but was totally broken for remote users. It
now works for both.
7> Fixed the {VM {VN metas, these two somehow got broken sometime ago...
2.46
1> Damn, forgot to adjust the stacks to match what I had in 2.44 and below
result seems to be sporadic traps, mainly when using type 51 calls to
display ans/asc files, not always but enough to know there is a problem.
This build addresses that issue.
2> The menu type 301 was not processing esc codes, it will now do so, and will
also process Meta Codes as before. Same For AdeptShow() function...
3> Chat was truncating messages at the 255 char mark, this has been removed.
4> Note: The /LOCKOUT IP# is not yet implemented in this build, (do we really
need it?) Will see about maybe putting it in the next release..
5> Got a report that the newest pop3 function menu type is not working, will
have to investigate this further, as I don't do pop3 here.
It's quite possible those functions are not in this build, will have to check
just when Hexie added the meta and menutype for that last pop3 function, sumthin
bout allowing the user to leave mail on the popserver if I remember right.
2.45
1> Snagged safehex's latest code tree and included all the mods listed in
the previous 2.x builds I released as private builds. this brings us
up to date with the vac codeline for now, note adeptget/putvarbyrecord
is now known as adeptget/putuservarbyrecord, I did this to stay compatible
with the vac builds, sorry if u have to modify yur srx files.
Note2: This is a drop in exe for any VAC build adept system and if u changed
line_x and *.menu.bin formats as mentioned in 2.43 yu'll have to change em
back to run this build. Simply delete all line_x_configuration files as
well as local_configuration and all *.bin in \adept\menus dir, while your
at it deleteing language.text.compiled and language.text.index is not a bad
idea either, the system will recompile these files on startup. Check your
bbs_configuration file by going to the config->bbs->general setup dialog
and see that it is ok, i made no changes so it should be all right, but check
it anyway..
2.44
1. Changed the origin line to store 78 chars instead of 65, this change
requires u reenter yur default origin line in bbs config dialog.
2. Added the below stuff from 2.34a series codeline so sysops who were using
these features now have them available again, the excerpt below is lifted
outta the old readme.txt from the other codeline.
- Fixed the Group option in the menudrvr files, this was never
working before, now if = that user will be allowed that menuitem,
fixed comm,tcp,pipe support, use PortFlags 1=commport,2=telnet, and
3 = named pipe for the portflags.
In addition if a user does not have the group access he will be shown
NotGroup.Asc/Ans/Avt if it exists in the \text dir (or language\text dir)
If a user tries accessing a port restricted item he will be shown
NotCom.Asc/Ans/Avt if it's restriced to com ports and NotTCP.ASC/ANS/AVT
and NotPipe.ASC/ANSAVT if its restricted to telnet or named pipes. Remmeber
PORTFLAGS ARE NUMERIC, ie 1=Com Ports only, 2= TelNEt ports only
and 3 = Named Pipe nodes only. The local login is considered a com port.
Fixed the starttime end time options on individual menu items, they
will now respect the STARTTIME and ENDTIME keywords in the menus.
They will also show NotStartTime.A?? for items restricted by a starttime
and NotStopTime.A?? for items restricted by an end time.
Went nuts and added in all the other docced menurestrictions on a
per item basis.
< MinOnlineTime displays NotMinTime.A??
> MaxOnlineTime displays NotMaxTime.A??
< MinCalls displays NotMinCall.A??
> MaxCalls displays NotMaxCall.A??
< MinPosts dislays NotMinPost.A??
> MaxPosts displays NotMaxPost.A??
< MinUploads displays NotMinUpl.A??
> MaxUploads displays NotMaxUpl.A??
< MinDownLoads displays NotMinDl.A??
> MaxDownLoads displays NotMaxDl.A??
That's all the documented restrictions as per the newmenuengine
docs earlier in this readme.txt, lemme know if you have a problem
with any of these, i put them in in one sitting, so it's
possible there may be mites with some of em.
Note: the min/max online time is for the current call, ie if
a user has been online this call for more than minonline time
he/she will have access to that menu key..
Further note, these files live in the text dir for the language
in question, if no languages file then they live in \adept\text
by default. The Asc is looked for first as with all the display
files, the .Asc must exist or looking stops. If u have only an
Ans/Avt file they will not be displayed till u put in a corresposnding
.Asc file, it can be a 0 byte file, but it *MUST* exist in order
to have the others display properly.
2.43
1. Changed the line_x_configuration storage format and the *.menu.bin
storage formats, these changes speed up the processing by an order
of magnitude, to upgrade to this version, simply delete all *.Bin
files in \adept\menus dir, delete all Line_X_Configuration files
as well as Local_Configuration file in \adept dir, fire up adept
with adeptx /nodes x /setup, edit the line configs to match yur
system, shutdown adept to save the newly edited configs. Fire up
adept normally and u should be ready for callers. Note that BBS_Configuration
file is *NOT* affected by these changes. If yu are upgrading from
any 1.xx version, you need to do the above setup...
2. added the adeptsmtpimport rexx function, but the nntp start I cannot
do easily, will try to work sumthin out for that.
3. Put Hexies fix for the msg not being saved if compression was turned on
for an area in place, I don't use compression here, so someone else
who does will have to test it.
2.42
1. Added per line load of addresses file ala safehex builds, if a file
called addresses.1 exists it will be used for line 1 addresses and
2 for line 2 etc, defaults to addresses if the line files do not exist.
2. Pulled all the parameters except for Srifx.Txt file name outta the
mail.cmd command line, since they are all in srif.txt i figured they
were no longer required...I'm flexible on this, i'll put them back if
you really think it's required.
3. Increased the chat buffers from 1536 to 3072 lets see if this makes any
difference at all. Works ok here, byt my board is not a busy chat board.
Matthew ?? Remarks/comments....
2.41
1. Added a new entry box to bbs->config->misc paths/files dialog, this box
called SRIF Path will determine where adept places the SRIF files and
reads the RqReport.Txt files from if enable ext req proccessor flag
is set in bbs action flags.
2. Implemented safehex's compression fixes, I don't use compressiom here so
this stuff is untested by, feedback would be good.
3. Hopefully fixed the ext req prozzer stuff so it will actually work now,
awaiting feedback on this from BK who is hot to trot on this with Mtic.
4. Feedback received, the SRIF stuff works as advertised.. :>
5. Added the BlueWave offline configuration options, Bwave users can now
tag/untag areas from their offline readers.
6. Fixed srif file systemstatus to properly reflect Listed/Unlisted status
of a requesting system, if it's not in the current nodelist, it's
status will show as Unlisted.
7. Added the name of the Srif file to the Mail.Cmd line so u can use %1 to
access it in mail.cmd
2.40
1. Added new bbs action flag, Use ext req processor, if this flag is turned
on in config->bbs->action flags, then the following occurs on a file request,
Srifx.Txt is written to the main \adept directory, we wait 10 seconds for the
processor to write RqReportx.Txt in the main adept directory, we read reqreportx.txt
and send all the files listed in that response file. Note that x is the node
that got the file request, ie Srif1.txt and rqreport1.txt for node 1 and so on.
Rqreportx.txt is deleted at the completion of the file request.
2. Added the IP# /Lockout Cmd line param ala 2.34a series code, recall, you can
lock an IP address from connecting with yur adept by specifying
adept /nodes 5 /lockout 206.248.278.156 /lockout 206.276.234.1 on the command line,
this example locks out two addresses, u can specify up to 20 on cmd line
if os2 will take a line that long, i dunno if it will or not.
2.39
1. Fixed a trap condition I intorduced while getting Hot/Cold input
routines to work.
2. Added Session Info to Mail.Cmd processing, the following parameters
will be passed to the mail.cmd proc line as %1 %2 %3 %4 %5
%1 = Sysop Name Of Requesting System
%2 = 5d Address of that system
%3 = AKA: followed by List of Aka's for that system enclosed in " chars
ie "AKA: 222/70 222/71 55:50/0 "
%4 = Secure/Insecure or Unlisted session type
%5 = .Req file submitted by that system (full path and name)
2.38
1. Fixed relogin function (again), It was not working if bbswatch threads
were enabled, it now works correctly, note that it does not work on
a local login, this is the way it was in previous builds too.
2. Fixed adeptget/putvarbyrecord, forgot that this build handles the user
files differently, they now work as advertised. Note that if u try to
use these functions on a user record that is currently logged in strange
things will happen cause these functions work on the DISK record, not the
memory record, so be careful, it is quite possible to fuq up the users
file with these functions, I know, I've done it.. actually its the put
function that's dangerous, the get is harmless.
3. Fixed the nntp drop down list selection, note nntp must be active for the
list to contain anything. Had a report that nntp was duping posts across
different newsgroups instead of just posting it to the active group, am
checking this out, as another guy reported this is not the case, so I've gotta
do some testing here with it. If you notice anything strange going on with
nntp get/put do e-mail me please.
4. Fixed the Hot/Cold input flag to really work in this code line too, works
just as it did in the previous 2.xx series builds, ie if a user chooses
cold input he will have to hit enter after ever menu selection/ more prompt
etc. Note as before the local login cannot have cold input enabled, hot
keys are *ALWAYS* on for a local login, regardless of the state of the
hot/cold input flag.
2.37 update
1. DOH!!! Added AdeptGet/PutVarByRecord and adeptgetusername to this build
from my old 2.34a code line, seems some people were actually using these
and missed them.
2. Did some work on the nntp engine, seems to work better now.
3. added the pop3 mail scan at login code..
4. added the pop3 menuchoices safehex has in his codeline, this build has
all the safehex options in it except for language buffering, mainly cause
i enabled buffering of both language.text and *.menu.bin long ago, there
is no option to turn it off...
5. Fixed adeptget/putvarbyrecord, forgot that this build handles the user
files differently, they now work as advertised.
2.36
1. Ok, I broke down and put in code to let smtp/nntp areas be squish/*/msg
base areas if you choose. Note that if you post in a squish/*.msg base
nntp or smtp area with an external editor like fleetstreet or whatever
don't expect it to go out, as adept has no way of knowing your trying
to post an internet type message. Posting from an adept login to a
squish/*.msg base area it will go out as expected.
2. Freq download counters were broken in x2, fixed in this release.
3. Problem with nntp posts not being sent out, fixed in this release, and by
the way nntp import smtp import to squish/msg base works, I tested them
and they appear to work as advertised..
4. Do *NOT* put this build up for general distribution anywhere, my builds
are meant for myself and a couple other people who have expressed interest
in running my watcom builds, I will try to stay current with safehex's
code, but he moves fast, so this may not always be possible... feel free
to give out copies of this archive to friends who want it, just don't go
putting it on hobbes or snoval or anywhere like that... I went back to the
2.xx numbering scheme, cause my experiments are over, i run this build
live on my system so it's not experimantal code anymore..
1.11x2
1. Corrected a trap condition where a user could start writing a message
even if he had no access to any msg areas, this would trap if running
without exception handling turned on.
2. Corrected a handle problem for the pop3 functions as per safehex's
instructions.
3. Moved the save offline/msg/file tags to the olr/adjustmsg/adjustfile
functions stead of blindly saving at logoff, these flags will now be saved
when user chooses "Permanent" selections in the appropriate menu choices.
1.11x1
-----
1. Cleaned up several trap conditions in BlueWave offline mail export, these
are prolly responsible for some of the unexplained traps during olr export.
2. Corrected a trap condition during bluewave import to smtp area, if u ran
without except handling on adept would go down.
3. Fixed the relogin option so it relogs u in stead of logging u off.
1.11x
-----
1. This build inlcludes everything
from 1.11f and down, as well as everything from the 2.34a and down
code lines, I have added the Cacheing i/o system to the 1.xx codeline
and it appears to be stable here so far, have also added the squish
*.msg base support to this codeline, and it too appears to be stable
so far. I have run this build on my system (live) for over 24 hours
without a trap (i do not run with the exception handler turned on)
so i'm letting this experimental build loose for those who care to try
it. Note: The vac versions have squish/*.msg base support too, and
safehex is in the process of revamping the enitre msgbase handling
system in the vac builds. To keep my wat builds seperated from his
numbering scheme, i will number mine as 1.11x, the next version
will be 1.11x1 then 1.11x2 etc, till i have a stable and tested
build... Latest official safehex build is (at time of this writing)
adu111j.zip, available at snoval.com.
1.11f
-----
1. Added a new user flag which will allow a user to toggle off any
alternate editors used by Adept via the ExtFSEx.Cmd or ExtLEx.Cmd.
A new menu option 634 can be used to toggle the flag and a new meta
{VS will show the current state.
Ok, so I liked the look of QuikEdit but it's DOS and it's kinda
slow, specially when compared to the bare bones internal Adept FSE...
some users cried out about it and some even liked it (amazing eh? :))
well...now the users have a choice...the sysop can configure the
external editors and the user has the option not to use them.
2. Removed the time zone display...it seems it was not only the VAC RTL
but also WTC that has problems in that area for now. At least, most
people should know their time zones and a whole lot might not even
have it setup. So...gone for now :)
1.11e
-----
1. Added the FTP Server startup to the BBS Action Flags. The /FTP command
line option has been removed.
2. Moved some other command line options to the BBS Action Flags. Do check
out the current settings for your BBS Action Flags.
3. Added a sysop mail indicator to the Adept status window.
4. Made the display of the sysop mail and the system and BBS info in the
secondary status window configurable via the BBS Action Flags. So...now
if you don't want to watch the system threads/process etc, or how many
calls the BBS got n stuff...just turn it off.
5. Enabled the forced password change from the BBS General Settings and
added a group wise password expiry which can be set from the groups
editor. In short, if you want all users in all groups to change
passwords after a fixed number of days, set it in the BBS General
Settings, otherwise, use the individual group password expiry.
The group settings will over-ride the BBS general settings if set
to any non zero value, otherwise the global setting will be active.
The Groups file uses the value of the new field "EXPPWDDAYS" to get
the number of days before a pwd expires for that group. This is just
for those who prefer to manually edit the files...If you use the
group editor you don't have to bother to remember this :)
The prompt #479 from the Language.Text file is display on expiry.
#479
[Cvt][Mta]\r\nIt has been {PW days since your last password change.
[Cvt]\r\nIt is time to change your password.\r\n
1.11d
-----
1. Corrected some potential GUI related problems. Mostly visual stuff
nothing serious...
2. The VAC RTL seems to have a problem returning the correct time zone
string. My WTC compile showed that up, so till the next FP for VAC
which will correct this problem (it's already listed for the next FP)
please don't tell me about the time zone display not being correct...
I already know :)
3. The send files dialog set colors that became black on black after the
recent color scheme change :) Fixed now.
4. Removed a non-required timer...was keeping time for nothing :)
5. You can now select to disable the status window from the BBS Action
Flags. The command line option /NOSTAT has been removed.
1.11c
-----
1. Fixed a small mailer related problem a cpl of people had.
2. Reduced the minimum for msgbase compression to 384 bytes from the
earlier 512 as it seems to work equally well. XMsg, GateKpr and
Adept now adhere to this new minimum. There is no problem as far
as the compatibility is concerned since the limit is down...not up :)
3. Corrected a small inconsistency in the userserver dll which might
have accounted for some of the handle related problems on rare
occasions. uNuke and ReIndexUsers have the same corrections as well
just in case. This was actually a prob someplace in Adept which was
corrected quite some time ago, but made these changes to make sure
of it :)
1.11b
-----
1. Expanded the message and file area tags to a maximum of 10240 now
although FSys retains the 4 digit limit for now, so keep the file
area numbers within the 9999 limit.
Nope, no change in the configs etc 'cept the BBS_Configuration,
to which Adept will adapt automatically when run the first time
and will still remain backward compatible. There is no real need
to backup anything, just backup for safety though :)
I kept it compatible to offer an easy upgrade path for anyone who
wishes to try a drop in replacement...Keep in mind though, that
this might not last forever :) Some changes are due in the forseeable
future then I will no longer be able to keep up the backward
compatibility anymore...
In case there is something broken with this build, you can go back to
any older 1.1x without any problems...
Now for the gory details for those interested...I have taken the same
approach that was taken for the Offline Mail tags in Adept a long time
ago, they were moved from the userbase to files in the
System\Offline_Tags. Similarly, the message and file area tags are now
saved in files in the System\MessageArea_Tags and System\FileArea_Tags.
2. Fixed some minor GUI inconsistencies...just visual...
3. Added the configuration of the main BBS file/msg Flags to the new user
defaults dialog for now. These flags decide what areas a new user will
have access to and over-ride the new user flags defaults. In other words,
the new user flags are applied to a new user and then cross-checked
against the main BBS flags for access. For example, if you turn on all
file areas for a new user and then turn off some areas in the main BBS
flags, those will be turned off for the new user as well.
1.11a
-----
1. The POP server name expanded to 127 characters from the earlier 15.
No, nothing to change...completely backward compatible :)
2. Corrected the local window display in which the top status line would
vanish if desktop windows were switched. (For example when using OD or
any other virtual windows kinda stuff).
3. Corrected the per msgbase area compression flags check. You must have
the compression in the BBS Action Flags for online compression to work.
For example, if you have some msg areas marked for compression, but
turn off the compression from the BBS Action flags, then the msg
compression will not happen for any online messages. GateKpr and XMsg
will still compress the messages for any marked areas from the outside
though...as before...
4. Removed the time display from the local bbs window and put in the user
handle there. Added the user handle and the current time to the main
Adept status window.
5. Changed the name and corresponding action of the "No Compression" in
the BBS Action Flags settings as it seemed to be causing some confusion.
It's now called "Enable Online Msg Compression" and if added to the BBS
Action enabled flags, will enable online msg compression in Adept.
6. Some minor GUI related changes...
7. The Colors ini file will now be of a fixed length. It will not keep
growing (if req) with everytime the settings were saved.
8. All files in the current build are neither compressed nor LxOpt'd.
I will do that only for a normal release version, ie., a version
that does not have any letter in it. If the programs are LxOpt'd
it will be mentioned in the readme...
9. None of the DLLs have changed from the 1.11 release...the incl DLLs
are fresh compiles with no LxOpt. This does *not* mean they are
slower or will require more RAM or any such thing...LxOpt only
optimizes the page faults and for the small DLLs that we use, will
not have a major impact, tho there will be some improvement if the
files are LxOpt'd :)
1.11
-----
1. The xxx.Desc file was not displayed unless the user chose to display
it from the read message prompt. Fixed!
The default keys in the language.text file for displaying the
xxx.Rules is U and for xxx.Desc is D.
2. Okay...this one is for me...wanted to do this for quite some time...:)
The Local login, i.e., Line #0 will now be as consistent as possible
with the other normal lines in the sense of logging etc. I was having
to go and view the Line0 log in my testing all the time, now press
PageUp to go to line 0 and the logging is there as it is for all other
lines. Some minor changes in the way the PageUp/PageDn worked to
accomodate this.
3. The POP passwords will be kept encrypted now, just in case some users
do not trust us SysOps for some reason :)
The encryption is quite reasonable and will take more than just a
casual effort to go through...
What this also means is that anyone using POP from any earlier build
will have to re-enter the POP password at the very least. Also, the
pop password will display '*' on user entry now.
4. Found some inconsistent code in the GUI, nothing serious, but corrected
anyway. Mild display changes...someone might even notice :)
5. The Offline Tags for user could not be fully edited (set/unset) from the
user editor, it would allow setting of upto area number 1024, now
corrected to go upto area #10240 which is the current Adept limit for
message and file areas.
Once again, keep in mind the fact that the file indexes use only 4
digits, and using area numbers over 9999 for any file area will cause
some file searching anomalies...
6. Added the date/time to the local bbs window header display.
7. Corrected a possible trap condition while reading messages.
8. The dates display in the user editor is now localized as well.
9. You can now drag-n-drop colors and fonts to any of the main Adept
windows and save the settings using Windows -> Save Status (^z).
The VAC build no longer uses the Colors.Ctl file.
The colors and fonts will be saved to a file called "NC_Colors.Ini"
for now and a sample file is included which was made on a 64k color
system so the colors might need to be tweaked a bit depending on
your display.
NOTE: Use the solid color palette to set up the colors if you are
using 256 color, non-palette aware modes, otherwise you might
experience the NetScape/2 effect, temporarily...on Adept
startup with some displays :)
There is one other thing to keep in mind, the Status and the
Transfer boxes (the top 2 windows in Adept out of the 4) will
not show the color change in the bottom half immediately. You
can either a) mininize the window and redisplay, or, b) Restart
Adept. The changes will NOT be there unless you saved the
setting via ^z first!
10. Changed the InActive Line Logic. It now waits on the system rather
than sit in a loop waiting and eating the CPU while waiting... :)
11. Fixed a trap condition in the user editor.
12. Found a few more lines of squish which slipped earlier, removed now.
13. Fixed some potential telnet/socket related problems.
14. Some more general cleaning and corrected some anomalies in some places
to make things more consistent.
15. Some of the debug logging removed for this build...
16. I think I held it back enough for now :) letting 1.11 go as it stands
for now...
1.10z
-----
1. Fixed a trap on shutdown which crept into 1.10y 'cause of the addition
of the action flags to make some things optional that were not
optional earlier. Nothing serious as all files were saved and comitted
and all allocated RAM was free'd before the trap. In fact, that is why
the trap occurred :)
2. The Prev button was after the Next button in the user editor :)
Corrected now...
3. Re-arranged the user editor a bit to accomodate the new POP mail
related fields.
4. The NNTP add/remove groups was no longer immediate since the NNTP
server no longer keeps checking the disk all the time for add/remove
drop lists...The addition/removal will be immediate now (as it was
earlier) unless, of course, the NNTP is already running a poll.
5. The online message compression is back now and is controlled by the
BBS Action Flags. The option name is "No Message Compression".
If you remove this from the active flags, Adept will compress all
messages written online and are greater than 512 bytes in size.
This includes offline mail uploads as well...
********************************************************************
The Default Value for this is ON. The option was not implemented in
Adept was a different issue. It is implemented and works now. Make
sure you turn it off by adding the "No Message Compression" to the
active BBS Flags if you do not want to use message base compression.
********************************************************************
Given the more powerful systems today, the online compression might
well be feasible enough for most people. From what I gather, there
seem to have been some problems with the online compression, let's
see if there still are, maybe we can get it working... :)
6. The POP Mail get will now use the prompts from the Language.Text as
follows:
#951
[Ans][Cvt][Str]\r\n\n
Your POP
Password
:
D;
[Asc][Cvt][Str]\r\n\nYour POP Password: ;
#952
[Ans][Cvt][Str]\r\n\n
Your POP
Server IP
:
D;
[Asc][Cvt][Str]\r\n\nYour POP Server IP: ;
#953
[Ans][Cvt][Str]\r\n\n
Your POP
Login
:
D;
[Asc][Cvt][Str]\r\n\nYour POP Login: ;
#954
[Cvt]\r\n\07Your POP mail has arrived.\r\n
#955
[Cvt]\r\n\07No New POP mail found.\r\n
#956
[Ans][Cvt][Str]\r\n\n
Do you want to save the POP information for future use
?
[
Y
]D
;
[Asc][Cvt][Str]\r\n\nDo you want to save the POP information for future use? [Y/n] ;
Prompts #951 - #953 and #956 are used by the menu option 931 when
asking for input from the user and #954 - #955 will be displayed if
the user has Monitor Mail processing off after the POP Mail thread
completes processing.
7. Currently the POP mail get has been changed to a foreground only
process and will display all related messages while getting mail.
8. The Accounting and Exception Handling will now take effect online
when changed via the BBS Action Flags. The current Accouting dialog
has been made smaller, since there was not much on it...
9. XMsg should now properly mark blank messages and remove them during
a pack operation. I had plenty os users saving blanks msgs by mistake
or design, specially the new users...this should fix it...hopefully...
In effect, XMsg will now remove and msgs which have a size <= 38 bytes
34 bytes are always there as the MSGID and another 1 for the eof, which
means that any msg text <= (approx) 4 is gone! These are generally
space/enter and save, enter and save or just save...
Just managed to remove 300+ messages from new users w/o having to read
or even look at them.
10. Ok, here is how the POP mail get will use the required input fields.
1) Any information entered via menu type 931, will over-ride any
previously saved info.
2) The information saved in the userbase will be used.
3) If a server is not supplied by the user, the server from the
INet config will be used.
4) If the user does not supply a login name for the POP server,
the user's handle will be used.
1.10y
-----
1. Fixed a problem in some cases when using compressed message bases
the VAC compiled build would freeze and the WTC compiled build
would trap. Also corrected a related, potential problem in the
bluewave mail export.
2. Corrected a problem in the Adept startup. It could trap or freeze
the system on rare occasions.
3. Cleaned up the mismatches between XMsg/GateKpr/Adept in the msgbase
compression. Should no longer see the "an error occured reading..."
kinda error message now...hopefully :)
NOTE: The minimum compression size is now set to 512 bytes for all
the programs. Make sure your config files (GateKpr and XMsg)
are consistent with this change.
4. Corrected some area tags related inconsistencies...and some more
general cleaning up...
5. Made the highest possible file and message area numbers consistent in
Adept and all related utils. The maximum supported file or area number
is 10240. Using any area numbers beyond this will cause Adept to
ignore them with an error msg. Once again, keep in mind the fact that
the current versions of FSys (0.99+) use only 4 digits of the file
area number in the index, using any number beyond 9999 will result in
inconsistent results on file search/list operations.
6. Added some more BBS Action Flags...I think we would have run out of
command line length soon, so before that happens...here is the
cleanup :)
a) UnVerified User Menu
This flag, if set, will call UnVerifiedMain.Menu instead of
the regular Main.Menu for all users not marked as verified.
This should help a lot of people to make things earier as
well as make external CBVs easier to manage since there is
no built-in CBV in Adept...yet :)
Of course, as with all other menus, the order remains the
same for this one as well...see the section for 1.10c
point #9 for the details.
b) No New User Alias
This flag, if set, will not allow a new user to use aliases for
a full name. See the section on 1.09.50r point #14 for details.
This makes the /NC_FULLNAME command line option reduntant and
is removed now.
c) Drop Carrier On TimeUp
This flag, if set, will force a carrier drop the moment time
runs out for a user regardless of what the user is doing!
This was added quite some time back, but is optional now...
d) BBS Watch Thread
This flag, if set, will start a Watch Dog thread for every
user login and kill the login in case of problems after a
timeout. I do not recommend using this option together with
the Carrier Drop on TimeUp, but...it can be used :)
e) No Timeout On Telnet
This flag, if set, will protect all logins via telnet/vmodem
from user timeouts. In effect, any Telnet/VModem login will
not have the user time limits enforced. The users can stay
logged in as long as they like...This was done on a special
request and is still there... :)
f) Enable Accounting Features
This flag, if set, will enable the accounting features of
Adept. This will make the /ACCT command line redundant altho
i have left the command line option in place for now, it will
be gone in the next release or so...
g) Enable Exception Handling
This flag, if set, will enable the exception handling features
of Adept. This is primarily a debugging tool in case of
problems. Not to be used normally.
h) All Files Commit On Timer
This flag, if set, will commit all open files on your system to
disk every minute.
i) System Log Commit EveryTime
This flag, if set, will commit all Adept log files to disk after
every line added to them.
NOTE: The Enable Accounting and Exceptions will not come into action if
changed online. You will have to re-start the BBS for that. This
will be corrected later.
1.10x
-----
1. Fixed some minor inconsistencies in XMsg and GateKpr...mostly
costmetic :)
XMsg takes a wee bit more RAM now, but i gave it that since the
processing speed almost jumped to 2x...plus XMsg will now accept
a command line like "XMsg -A xact.fil", i.e., you can have a space
or tab in between the -A and the filename if desired, the same
applies to the action file itself as well...it can have tabs now...
2. Completely overlooked the userbase auto-update function Tower had
put in while he was trying out the new user base system way back.
Removed now...just in case it caused a problem.
3. Some more defines made consistent to ensure proper values in all
places as there is so much of changing in progress...found a change
related bug in the process as well...fixed now, nothing big.
4. Fresh full compile again since i had made some changes to make
the code more consistent...just in case...well...I generally
do a full compile anyway before every release, but this is the
first time I have done it 2wice in a row within a few hours :)
This should be a fairly safe and stable version...as most of the
new additions/changes have now been verified as working.
Now to move on to the next batch of changes :)
1.10w
-----
1. Broke the NNTP poll when I corrected the msg area editor. Fixed!
1.10v
-----
1. Tried to correct the file u/l and d/l timings and ratios now that both
BFSMake and FileEd (purge will put in all the req permissions for an
existing filebase) update the file permissions in the FileBase according
to the current File Area Flags.
Although this should be more configurable, but currently the offline mail
will not be counted and the time given back, the file time/bytes will
depend on the d/l'ded files settings/perms, the u/l files, as usual will
pick up the factor to give back from the File Area settings.
For file uploads, the time factor to give back is checked first, if that
is zero, then the file area flags are checked for the up back attribute
before deciding on how much time to give back to the user.
Hopefully i got it right...maybe :)
2. When I changed the NNTP and the Addresses to drop down lists in the
message area editor...I completely forgot to change the save process
to reflect the changes...This did cause some grief to at least one new
Sysop using the msg area editor since then to save. For some strange
reason tho, this seemed to have no effect on the addresses, but the
usetag tag would never be saved...Fixed!
1.10u
-----
1. Completed some incomplete stuff in AdeptRexx. Might not really have
been noticed except that i was fiddling around as usual :)
2. Corrected a minor flaw/inconsistency in the user editor dialog which
would not accept next/prev after defaulting to the last user online
till you went into find. Works as expected now...
3. You can use tabs in the config files now in case you prefer manually
editing them...Adept would not accept tabs earlier.
4. Well...although I did make an attempt to correct the msg/file areas
selection for menu types 46/47, but I completely overlooked the fact
that it cannot be done properly w/o changing the userbase structure
as well as the BBS config. Currently it's back to the way it was, will
correct and change the structures after a 1.11 or 1.20 release version
so we have something which is completely backward compatible and stable
to use while changing over to the new user base and bbs config as it
might take some time to stabilize...
This correction, once done, would allow an easier implementation of the
File and Message area groups as it seems to have been planned in Adept.
Will also implement 2000 safe stuff, as well as enhance the filebase to
cater to Adept more closely as it should.
5. Added 2 new user manipulation functions to Adept REXX, similar to what
Hagar had added to the WC build earlier, as follows...
UserData = AdeptGetUserVarByRecord( Line, RecordNumber, UserVariable )
Call AdeptPutUserVarByRecord Line, RecordNumber, UserVariable, UserData
The UserVariable number is the same as defined in the AdeptREXX docs.
The RecordNumber is *not* the same as the user id, rather the index
of the user record in the userbase. The index starts from 1 *not* 0 :)
As before, note the fact that Adept does NOT cater to changing the user
name although you can change it via REXX or the user editor...If you must
change the user name, make sure you run uNuke /r right after the change.
Changing the user handle is ok :)
I know, I know I'm using fn names which differ from what Hagar used, but
AdeptGetVarByRecord and AdeptPutVarByRecord did not really make much sense
to a dummy like me, so I added the User in between... :)
6. Changed the way the NNTP Server used to time the polls for news. It now
uses the system timer instead of checking on the time by itself to decide
when to start the next poll. This will not only make the NNTP poll more
accurate, but will also eliminate the redundant file i/o calls it does
to check for it's drop files regardless of the poll time set in the Inet
config.
Ok, ok...net result is, NNTP no longer uses the CPU in between polls as
it did earlier...there, that should be a simpler version :)
7. I re-enabled the VAC specific opts which caused some probs on one system
only so far since there have been more changes and cleaming, it might not
cause a problem anymore...
1.10t
-----
1. Well...the new beta compiler FP seems to have fixed the locally scoped
vars prob, but still has the bit twiddling bugs...just for info :)
Back to the earlier release FP again...
2. Adept Status will now display the last caller's name on startup,
assuming there was any caller at all :) This does make a wee bit
of sense also from the user editor's view poitn as the user editor
now defaults to displaying the last caller for the line in focus.
3. You can now toggle the show birthday using menu type 633 for the
verbose user listing (menu type 43). The bday, if displayed, will
be localized. The setting is also available from the user
editor -> attributes.
4. Added a new flag to the BBS Action flags "Always Save BBS Status"
which will save the BBS window positions everytime Adept is shutdown
normally. In effect as if you presses ^z before shutting down Adept.
5. Changed the way the local and netmail file attaches worked making the
/NC_NOLOCALFILE and /NC_NONETFILE redundant. These command line
options no longer exist as the option of being able to attach files to
messages is now a part of the user attributes.
The new flags "Allow Netmail File Attach" and "Allow Local File Attach"
have been added to the user editor -> attributes settings.
This has been added to previously un-used user attribs and by default
should be off for all users unless someone had been fiddling with the
user file directly... :)
6. The Help -> Author dialog has some information filled in now...more
to come later.
1.10s
-----
1. Completely forgot to remove the \Adept hardcoded path from the DOS
door runner menu option. It uses the base Adept path to pick up
VX00.Sys and ANSI.Sys.
2. The local BBS window, would not appear in the tasklist, so if you
minimized it...well... :)
My CoSysop tried to clear the desktop to see behind the BBS...
3. Corrected a slight display problem with the line editor and some
other pesky stuff which i am not going to document as it does not
really matter to anyone else but me :)
4. Removed the ^A kludge from the new menu driver...it should no longer
be required (I hope...). I use ^A from neplace to get the allfiles
list for the leeches :)
5. Cut the mem reqirements back to the original slimmed down ones for
the VAC builds. WC required more and I forgot to change it back...
6. Added some more carrier checks, some code re-arrangements to optimize
stuff and added more logging (optionally) for the future use of the
existing extended logging features already in Adept. Now we Adept
Sysops should be able to show off our cool Adept threads :)
7. Added an Author Info Dialog to the Help Menu, which I will expand on
to display the currently active settings for Adept and allow changing
some that can be changed online before i plug them into the BBS action
flags or some such place where they should be after they are out of
a reasonable trial period.
8. Fixed fList for config file read/process problem and not prepending
files properly. You can now use spaces or tabs in the cfg file.
fList will now run at minimal normal priority by default rather than
the idle priority as earlier. You can use /L on the fList command
line to run it on idle priority as before if required.
9. Please note that since version 2.1x (not sure which x it is now), the
Hagar build is no longer backward compatible specially if you have
been running all of the updates. You might have to clean out the
BBS and Line config files and re-create them to use the VAC version
which is still completely compatible with all older versions.
In short, the old idea of drop in replacement no longer applies to
the Hagar builds and if you use that, do not try the VAC build with
the same config files.
As for the new menu driver's *.bin files, no need to bother about
them, the VAC build will auto-adapt for those if required...at least
for now, unless Hagar changes something else there which will make
the auto-adapt incompatible.
There seems to be some change in the mailer as well in Hagar builds
as wikk (bkirby@prismnet.com) tried out some polling which worked
fine when using VAC to VAC but failed on VAC to Hagar's WC builds.
I have no idea on this as all the other Adept Sysops here use the VAC
build and we have had no problems in polling. If indeed there is some
change in the WC version I am not aware of it and will not touch the
mailer in the VAC build for 2 reasons...
a) it works fine for me and most others who use it
b) There is a lot more to be done in Adept w/o going into something
that works reasonably well anyway.
Lastly, as already put up in the Adept mail list by Hagar, version
2.15+ use different config files, so do *not* try to use the WC/VAC
builds on the same set of configs. CServer has also been changed for
the Hagar (WC) build 2.18+ which means that you have to use the right
CServer.Exe with the WC/VAC versions.
Finally :) the whole point about #8 so far is, if you are using
both WC/VAC builds, and unless you are using clean configs for
both, please do not email me about bugs/problems as I cannot
re-produce or fix and garbled up config related probs as the
config files are assumed to be correctly set up by the SysOp.
Also, please note that I do not recommend using any of my VAC built
EXEs or DLLs with Hagar's builds as I am not aware of the changes
that might be there which could cause problems if any of my VAC
built stuff is used with the WC builds.
1.10r
-----
1. Corrected an old problem while saving a user record from the user
editor which would require running ReIndexUsers before the userbase
started functioning as it should.
NOTE: Adept does not cater to editing the user name from the user
editor! If you must change the user name, then you MUST run
ReIndexUsers right after that!!!
Make sure you run
uNuke /r /q
ReIndexUsers
before starting up with this build as it will also adapt to
the last user id on every startup instead of letting it run
wild which degrades the response time in general.
Would be best if you if you had these lines permanently in the
.Cmd (or whatever) that starts up Adept for you. Some parts of
Adept may not function as they should if the above 2 are not run
before starting up Adept...every time!!!
2. Corrected the Delete button behaviour for the User Editor when called
for the online user from the local login window view. It was
inconsistent with the changes to the normal user editor which were
made earlier. The online user is a wee bit redundant now as the normal
user editor will now, by default, select the last login (which is also
the current user if logged in) name to edit.
3. Finally fixed the ZedZap transfer trap on a carrier drop. The problem
is documented under 1.09.50v #6. Of course the forced handlers have been
removed now which did not help anyway since i did get the trap here...
Whew!!! :)
4. Added the POP3 get support to Adept. The interface for setting it up in
the GUI I had done earlier, now filled in the backing code :)
Here is how it works for now...
a. Add the POP3 server address to the Internet config and check the
POP3 receive option next to it.
b. The User Editor will allow you to enter a password for the POP3 server
for any given user.
c. For Logging into the POP3 server, the existing user's handle will be
used. Of course the user, or you, the SysOp, can change the user Handle
if required first.
d. The POP3 get is independent of the user carrier state, once started,
it will run to completion unless there is some POP3 related error.
e. Menu Type 930 will start the POP3 mail import for the user currently
online and 931 will allow the user to set/change the POP3 password.
The type 931 is there just in case if there is a requirement and uses
the normal password prompts (which might be slightly incorrect :))
f. The mail (if any) will be imported into the message area pointed to
in the Inet config for SMTP mail.
g. SMTP and POP3 gets can be used togther.
That's about it for now, till I get some more feedback on this. The user
who selects to get POP3 mail will not be notified in any way if there is
mail which has come in, a new mail check will do that anyway so I have
not put in any notification of errors/new mail got from the POP server
etc. for now...
1.10q
-----
1. There seem to be some problems in the VAC/WC mix, so, switching back to
VAC for all the DLLs and EXEs.
2. Fixed the VAC XMsg problem of buffers and not processing the action file
works fine now. It was yet another RTL difference. Optimized it some...
seems to be slightly faster for my small usage now...can't say much about
the performance for larger msgbase bbses.
Keep in mind the fact that this is still considered to be a beta ver
till i get confirmations from other people who use and verify it's
functionality...
3. Removed the extra code I put in to process the menu type 3 for the new
menu driver, it's now done in the existing (older) code saving some
space...
1.10p
-----
1. Corrected some inconsistencies in some of the display code which might
have caused problems for some people...rare case...
2. In case I forgot to mention this before, try to keep all message/file
area numbers within the 1-9999 limit. The file area indexes are built
using only 4 digits and using numbers > 9999 will have some undesired
effects. Hopefully this might not be a problem for most people :)
3. Optimized the bit-twiddling macros.
4. Menu Type 3 for the new menu driver should function the same as the
old classic driver now...hopefully :) Would appreciate if someone
could verify this...
5. More cleaning and pruning...nothing too serious :)
6. Added a drop down list for the usenet tags to the message area editor
and enhanced the editor save so that it will not save anything that is
not required based on the message area attributes. For example, the
editor will no longer save addresses for on echo/net areas and if the
no origin is set, then the origin line will not be saved, if the usenet
attribute is not set no usenet tags will be saved for that area etc etc :)
Do let me know if i missed something in here...Of course you can always
manually change/edit the message area file as usual...
1.10o
-----
1. Corrected the REXX sub-system for a compiler bug (bug is in both VAC
and Watcom). Neither compiler can handle aliased assignments of nested
structure pointers with or without optimization.
With Watcom this problem was not apparent as it validates for all free
calls, with VAC though, it was a gotcha...TRAP! :)
2. Added uploader name field to fList.
1.10n
-----
1. Made a cosmetic change to the user editor. The delete button will now
show "Restore" is a user is marked as deleted and vice-versa. Also,
the button will be disabled if the user has the no kill flag set which
will reduce the chances for related errors...Of course, you can still
manipluate the flags directly and do whatever from the Attributes/Flags
dialog, but made the main del button a wee bit safer to operate :)
So...next time the VAC build does not let the SysOp login, you know
why...visually...from the user dialog editor :)
2. Fixed the Groups pull down in the user editor, it would show an ever
growing list of the same groups...The list would keep growing every
time you changed the user rec, for example, clicking on Next 10 times
would show the same list 10 fold.
3. Corrected and Accounting related problem for people who might not be
using it...
4. Changed the default all files commit (flush to disk) on the status timer
to off. Specifying /NC_COMMIT on the command line will enable the all
open files flush every minute.
5. Corrected some more potential rexx related problem stuff...and fixed
the so called "telnet" script that most people seem to be using.
Please keep in mind, the telnetsnoop is not telnet, it's a character
get from user and block get from the other end, it's not a link as
telnet should be, block send/get both ways.
For actual telnet, use the original Telnet script, which is included
as Telnet-O.Cmd in the modified TelnetNC.Zip.
1.10m
-----
1. The automatic diz extraction was broken in file uploads...Fixed!
2. If a BadUpload.Cmd did not exist, Adept would not delete the bad
uploads...it will delete all bad uploads now unless it finds the
BadUpload.Cmd file which will be run instead of deleting the bad
uploads.
3. After an upload, the description editor would not check for user
time left, which made it possible for a user to keep a node hanging
for as long as they wanted...the editor will now check for the user
time online etc...
4. Added a SysOp mail indicator to the status info. It will come on if
there is any personal mail waiting for the SysOp along with the number
of the messages.
1.10l
-----
1. The Watcom 1.10k did bring out a problem in the SMTP send, I had reduced
the thread stack which did not work with the Watcom build, fixed now for
the Watcom builds. The default will be higher for Watcom builds...
I found something very interesting from the beta VAC fixpak fixes...
besides other stuff, there was one thing that affects Adept in
general..here is what it sez...
AP 72949 RT Whenever a DLL built with VisualAge C++ is loaded
and freed by a program not built with VisualAge C++,
the runtime will leak 64K of address space.
Since at least the gio and nntp (which incl smtp) dlls are VAC created,
I will be switching back to VAC for the Adept and CServer Exes, but will
make all other DLLs and utils in Watcom for now...
2. Yeah well...i did back out of the beta fix for the above reason :)
The Adept and CServer Exes are now VAC built again...DLLs and other
utils remain Watcom, till we get an official fix for the VAC DLL
problem.
3. As per wikk's suggestion, this section moved to the top of the readme :)
1.10k
-----
1. Cleaned up some more code...the Watcom build works equally well here...
over 24 hrs now...the one incl in the 1.10j archive might have had some
probs on some configs...
2. Corrected another cheat introduced as a side effect by one of Hagar's
earlier kludges which would allow users to d/l all they wanted...
The kludge is gone, but the functionality w/o the cheat remains...
The kludge itself was never really required as the original code had
all that was needed for the changed functionality.
The cheat is not documented right now for obvious reasons, but if you
are an AdeptXBBS sysop on the AdeptNet list, you can ask for more info
about it to check out on prior versions if req...
3. Put in a forced exception handler for SMTP since we have no control over
that for now, this should keep Adept from trapping on that...
4. Switching to Watcom for this and some future builds just to make sure I
have all the code cleaned up and working for either compiler...this might
sound experimental :) but trust me...the Watcom runtime libs are far more
safe than VACs, WTC is forgiving, VAC is not...so if the code was stable
under VAC, it should be even more so under WTC.
Also, since WTC does not add any extra localization/msg related stuff on
it's own, the files compiled by WTC will generally be smaller in size...
Of course this has nothing to do with the fact that i working on a
project where I have to use Watcom anyway, nor the fact that I applied
a beta (unofficial) fixpak to VAC which completely broke the localized
scope stack processing and am too lazy to back out and would rather wait
for the official fix... :)
5. Fixed the user file upload process...it would save the file in the file
base even if the user dropped carrier on enter description and anyplace
around that time...It should now call badupload...
6. Re-enabled the FTS-0001 menu option...will try to fix this as and when
I can get the hang of it...BTW, only the menu option is enabled :) as
this flag does effect the behaviour of YooHoo...
7. I have lxLite'd the Watcom exes (not the DLLs) to reduce the disk i/o
during program loading...might help, in case anyone has a problem,
try to unpack using lxLite 1.2.0 or e-mail me about it...
1.10j
-----
1. The minutes displayed in the download prompt was showing seconds...Fixed!
2. Localized the FList program. FList is an AllFiles.List kinda proggy for
Adept...
3. Recompiled FUtil with P5 opts. FUtil is a cmd line filebase manager for
Adept. Please refer to the original Adept docs for more info on these...
4. Included both Watcom (Adept-WTC.Exe) and VAC compiled (Adept.Exe)
files...both compilers differ in some respects so having both builds
might help further eliminate the vague compatibility problems in the
code if any are still left...The Watcom build, like the VAC one has
all optimizations turned on to bring out any possible bugs arising
due to the optimizer...
BTW, there is NO debug code in the Watcom build...the size difference
between Hagar's and my WC exe is that Hagar has compressed the exe
using LxLite and I have not...If LxLite'd exes have no problems on
your system, feel free to compress the WTC and/or the VAC builds...
Just remember to use the newest version of LxLite...currently 1.20
5. For those who might be interested, if you want to use the new beta FP #2
you must back out of the OS2KRNL file and comment out the
DEVICE=X:\OS2\MDOS\VW32S.SYS in the config.sys, otherwise....:)
Of course you might want to back out of VW32S.SYS as well if you use
it...(ugh :))
Interestingly...the in same code, the VAC build generally freezes OS/2
whereas the Watcom build re-boots OS/2...duh...talk about vague :)
6. Corrected the GateKpr PM version behaviour on startup...it would always
go behind other open windows regardless of the command line chosen to
run it...
1.10i
-----
1. Finally managed to figure out possibly the last invalid free memory
related trap that was proving to be quite elusive, ever since I
re-did the entire mem allocs... :)
This would generally occur in the delete operation of a tagged file
list.
1.10h
-----
1. The upload disk space prompt will now be picked up from #915
#915
[Cvt][Str]\r\nUpload disk has %s bytes free\r\n
Please add this to the Language.Text
2. Removed some more redundant and debug code...
3. Fixed a bug which came up when I added the validation checks for
msgbases which would not save the messages written online...
1.10g
-----
1. Re-arranged some more startup sequences...
2. Attempted to make some displayed numbers more readable :)
3. Added the password to the logging on a bogus password entry...
4. Corrected a potential problem with the address loader. It would fail
to correctly load up the BBS addresses under certain conditions
causing Adept to trap on startup.
5. Adept will no longer attempt to drop DTR on an inactive line on
shutdown...let's see if this really helps...
6. Made a slight change in the local window display routine...might help
someone :)
7. Adept searches for its files in the following order:
path -> apath -> dpath -> Adept\System -> Adept\Mailer
Please make sure that you do not have files like Addresses, Archivers
etc duplicated. Will try to change this so that Adept looks for its
files only in the Adept base dir and dirs under it...
8. Removed some more code that is not currently in use...
9. The time stamp in the log in the Recent events is now localized.
10. Some of the metas that return numbers, will now be local formatted
with commas (or dots whatever :))...will eventually try to complete
localizing all of them...
11. Changed the way Adept now searches for it's files...it will now look
for them firstly in it's base dir, then the system and then mailer,
failing this, it would then try the APATH and DPATH. Should speed up
the load time in general if nothing else...Plus this would minimize
the change of having a dupe file of the same name somewhere in the
path incorrectly picked up by Adept for itself...
12. Changed the date display prompt in the birthdate change (menu type 156)
to display the localized date format. This would require a slight change
in the Language.Text file for prompt #320 unless you don't mind the
dd or mm to display as 4 digits :)
#320
[Ans][Cvt][Str]\x1b[0m\r\n\n%02u/%02u/%02u
[Asc][Cvt][Str]\r\n\n%02u/%02u/%02u
Preferrably put in a 04 instead of a 02 where ever the year comes in
for your locale. This will most prolly again change once all the
localization is in place. This still has the seperator chars hardcoded...
13. Formatted the d/l prompt numbers as well, but they will now use the
prompts #912 and #913 instead to maintain compatibility with Hagar's
versions. Just Add the following to your language.text file...
#913
[Asc][Cvt][Str]Transfer Size: %s bytes\r\n
[Ans][Cvt][Str]
Transfer Size:
%s
bytes\r\n
#914
[Asc][Cvt][Str]Transfer Time: %s min, %s sec (appx)\r\n
[Ans][Cvt][Str]
Transfer Time:
%s
min,
%s
sec (appx)\r\n
14. Extended the display files in a few places i missed out earlier...
15. Added message validation to the online/offline/smtp/nntp mail as per
the XMsgBase specs. Might help solve some msg base probs...
1.10f
-----
1. Re-arranged the startup loading...seems better this way...let's see...
2. Cleaned up the shutdown...although the compiler libs would have done
it anyway, but in case there is some problem we would never get to know...
3. Added a valid user name check to the usersrvr.dll just in case...did the
same for uNuke earlier...
4. Fixed a SMTP problem carried over from the last version. Would cause
Adept to trap if there was any mail in the IMail dir on startup.
5. Fixed the irritating, automatic, msg to "All" in private msg areas when
enter was pressed on the To line to leave blank...It would force the user
to enter a To: line before allowing a quit...
6. Improved the PMI processing speeds...should be substantially faster...
specially if using SMTP. The related disk i/o also has been halved.
7. As for the date display...please let me know if i have missed out a
date display anywhere. The searching will prompt for YYMMDD as that
is far more logical, but other displays should be in your own local
format...
8. Releasing 1.10f now 'cause of the fixed #4.
1.10e
-----
1. Fixed a trap in the message area editor for 1.10d but is documented as
a part of 1.10e since only the exe was updated in the archive.
2. The User base files can now be copied etc if no user is logged in,
local or remote. This might well make the nuke user file option I
added earlier a bit redundant, but i'll leave it there for the
lazy people like me :)
3. Improved slightly more on the message threading...should be more
efficient now...
4. Localized the date/time stamp for the trap.log file as well...
5. Corrected a problem with the drop-down list for addresses in the
message area editor...selecting next/prev area would keep adding
the same addresses to the list over and over...no big prob, just
incorrect and irritating :)
6. Pinched mem requirements again...since I do have most problems fixed,
I guess it is high time we found out what the min req is that works
for most people...
The current mem settings never worked even for me earlier...they do
now :) let's hope and wait and see...
7. As per #6, just checking out this build with the beta linker again :)
Hopefully there would be no problems for anyone this time...
8. The user editor will now directly go to the last callers record for
the line that has the current focus rather than the first record all
the time as it did earlier...
9. Added the non-correcting modem callers option to write online messages
to the BBS Flags rather than forcing the issue for all :) By default,
it is off. Check out the flags in the Config->BBS->BBS Action Flags.
10. Fixed a rare case compiler library dependent problem thanks to Dutch :)
11. Localized the date display in the file listings...
12. Had not planned on releasing 1.10e yet, am doing so 'cause of #10...
1.10d
-----
1. Fixed two of my earlier blind global changes in the REXX sub system...
those 2 should not have been changed...duh...
2. Corrected the login new user yes/no puzzle...will correctly handle the
yes/no input now, will carry it over to all other such places
gradually...
3. Fixed another of my problems...Adept will no longer allow callers with
non-correcting modems or any non-correcting connect to write any message
online :)
It will now display the prompt #912 from the language file...
#912
[Cvt][Mta]\rSorry, Non Error Correcting Modem Callers cannot write online!\r\n{PC
Anyone using external editors for Adept will have to check the connects
flags there...this restriction is place only for the built-in Adept
Full Screen as well as the Line Editor...
4. Corrected a small potential problem in the NNTP section...
5. Added the archiver selection in the user editor -> More.
Fixed the protocol selection while I was at it....
6. Changed the format of the dates in the user settings to YYMMDD to
improve consistency with the rest of Adept.
7. Made a drop-down list for addresses in the Message Area Editor. Dutch and
one more (i forget who) suggested that it might make things easier...
8. Some more cosmetic changes...is it any better? dunno...you tell me :)
9. Localized the date/time stamp in the log files. Please let me know if
you get the incorrect format of the date or time in your log files now.
It should display in the format set for your OS/2 config.
Would also like to know if I should leave it localized or make it
consistent with the rest of Adept as YYMMDD...
10. Please make SURE that the the language file is corrected for the
following problem. The file included in the full VAC release is
incorrect as far as the message threading keys are concerned...
The thread to keys in the prompts #379, #380, #381, #382, #383,
#384 and #385 should match the key in the prompt #901
#901 (Next Page, Prev Page, Over, Abort, Reply, Forward-thread, Backward-thread, Quit-thread, Kill, Move, Copy, Locate, LocateNext, LocatePrevious, #Msg, Description, rUles, =Again, Scan, Without stopping, !Toggle Ctrl-A, LocateUser)
[Str]NPOAR+-.KMCL/\#DU=SW@! X*~:
The full install VAC version had the keys displayed as + and -, but
#901 had [] which would never work....sorry :)
Please check out your language files anyway and correct if required!
11. Improved on the message threading...should be substantially faster now...
and not eat up the CPU when users read threaded messages online. This
would generally be far more apparent in private or public/private mail
areas, specially if private only and the user had no messages to them.
1.10c
-----
1. Fixed FSys.DLL (now v 0.993) for a rare over-run problem.
2. Updated FileEd to 0.33b at the same time...
3. Made all the display/menu files checks consistent. Please let me know
it any particular menu/text file is not displayed when it should be,
possibly a typo on my part. Works for all I have on my board, so should
not really be very many problems...
4. Added an unsorted list option in the user editor find and made it the
default. To see a sorted user list (as before) just un-check the
unsorted check box. I always missed the ability to look at the newest
users 'cause of the sorted list, so... :)
The Find box will also display the number of users in the user base.
5. Just some more futurisitic UI changes and more fiddling as usual :)
6. Corrected a problem in uNuke which would cause it to trap in some cases
of grundged user name/handle. Corrected the same problem for XMsg as well.
********** WARNING **********
PLEASE NOTE: The included XMsg.Exe is something
that does seem to work all right here,
but use it on a backup first before
deciding to use it on a regular basis.
********** WARNING **********
7. Some people might not be quite aware of the fact that you can do all the
user and the message base maintainence from the nuke user file option in
the program menu.
When you select the nuke user file option, all lines are de-activated and
the user file close and AdeptUserNuke.Cmd is called. After this file
finished execution, the lines are activated and the user file re-opened.
8. Corrected some potential problems in the REXX sub system.
9. Started work on extending and standardizing the pre-defined display and
menu files. They will work in the following order now:
MainxxNC.Cmd -> MainNC.Cmd -> Mainxx.Cmd -> Main.Cmd -> Main.Menu
Mainxx.Asc -> Main.Asc
The only current exceptions are the Levelxx.Asc and the Groupxx.Asc.
1.10b
-----
1. Forgot to disable the exception handler in qwk related fns which would
cause Adept to trap in case of repeated errors...Fixed!
2. Removed some more un-used code...(just a few lines, not much :))
3. Pulled out a cpl of the forced exception handlers since they might not
be required anymore...
4. Added the Adept version number to the Trap.Log information. I got some
Trap logs in the past with no indication as to the version of Adept
that caused it.
5. Some more meddling and fiddling...let me know if you actually notice :)
1.10a
-----
1. Left a small problem in the shutdown menu not being enabled in some
cases...Fixed!
2. Corrected the security checks etc in the new menu driver as per the
original Adept documentation. The old menu driver was ok...the new
driver will function more or less in sync with the original now...
Will check out Hagar's port flags implementation before i add them
to ensure compatibility later...currently the port flags are not
implemented...
3. GateKpr fixes a TIC prob as pointed out by GrKnight...don't ask me
what...I have no idea on tics and fleas :)
1.10.00
--------
1. Well...this would have been 1.09.51f but since Hagar decided to jump
over to version 2.0, I guess it's best if I change to 1.10+ from now
on to avoid the previous confusion over 1.09.xxx :)
2. Cleaned out some more test code which i forgot about while testing the
file list display earlier. Would have caused a trap under some conditions
while displaying a file.
3. The 51e corrected probs remained in the REXX sub-system...Fixed!
4. The new beta VAC linker i used in the last cpl of versions seem to have
problems on some configurations and the DLLs were not always usable with
the Watcom build...switched back to the normal linker.
5. My old removed message base trap came up again...got it fixed this time
I think...:)
6. Changed the check for file presence logic to be faster now...will try to
change in all places in Adept with time...
7. Some more petty changes...:) Maybe someone else besides me will notice
'em as well...
8. Corrected the LMR problem when the messages exceed the max number
allowed to d/l. It used to set the LMR to the last message regardless
of the fact if the message was actually packed or not.
9. Adept will now play the file "AdeptSysOpPage.Wav" if found when the
page sysop option is selected, otherwise the normal beeping as before...
10. Some checks added to the BBS Watch thread...hopefully to avoid the
cases when it freezes the local display after a kill...happens on
rare occasions, till i figure out the exact conditions, this might
help...let's see...
1.09.51e
--------
1. Duh...my fault, I left some test code in 51c and 51d which would have
caused traps during message writing and qwk as well as text export. It
(the debug code) was not supposed to go out in a release :(
2. Some more code re-arrangements...nothing added..
1.09.51d
--------
1. Forgot to un-comment one Outbound re-scan statement...no big deal,
but...:)
2. Removed some useless stuff I had put in earlier and forgot :)
3. Reverted back to the VAC beta linker to re-confirm if that was ever
the real cause of probs some people had or was it just me :)
4. Modified FSys.DLL for something that *might* have caused an incorrect
index to be built and maintained based on any invalid area number or
file date.
5. Corrected GateKpr so that the pkt headers for netmail now have the
correct routes as per the specs.
6. Updated BFSMake to update the security levels in the filebase as Adept
itself does on any uploads. BFSMake will also update the costs field in
the Filebase if req. altho this is not currently in use...
7. Updated FileEd to .30b to account for most of the above mentioned pesky
probs to offer an easy path to update the filebase without shedding
tears :)
8. Added the FREE FILE check to Adept...a FREE file means to time/byte
check. There is a small problem with this though due to an earlier
addition...the status timer which checks for the user timeout and
drops carrier if the user times out will still throw out any user
d/l a FREE file if (s)he runs out of time during the d/l...
9. Some other related stuff yet to be plugged in...will do that a bit
later...
1.09.51c
--------
1. Corrected and invalid free in CServer which would cause it to trap the
moment there were more than 2 users using the chat...
2. Missed the phone number entry and the Address prompt/re-prompt after
help (TAB)...Fixed! It will now re-prompt properly
3. Removed the un-necessary kludges put in earlier for the user time cheat
and related stuff. The kludges were never required, only one incorrect
check in the original code caused this problem. The kludges are now
gone leaving the code cleaner and more functional.
Even on a carrier drop in the login procedure will not display news
all over again to the user...
Do let me know if any one finds any problem related to this...
4. Cleaned up some more display related kludges which were generally not
required AFAIK...maybe I am wrong...do let me know in case there is any
anomaly left...Lotsa delays/kludges gone...might have some display side
effects in case i missed anything...should not have any effect on the
BBS functionality...
5. The only display side-effects I have heard about so far is the recent
events and the outbound window not having any display on startup. This
is nothing to worry about as the events box will start logging the
moment anything happens and the outbound will be re-scanned within 5
mins (the default outbound re-scan time). There were kludges/delays in
the code to work around this which are now gone!
The kludges were system/config dependent anyway AFAIK...so, they
*could* cause the same effect given the right system config...
6. Corrected a small problem while reading messages...a hang up on the
prompt would take a while before coming out of the bbs.
1.09.51b
--------
1. Found a small problem in the display during the search file functions.
If you enter a single character to search for, and the character happens
to be the character M, the the highlight function fails due to m being
an integral part of the ANSI sequences and the M is splattered all over :)
IMO, searching for a single char is useless, so the file search functions
will no longer accept less than 2 chars to beging a file search.
2. Added the ability to stop the offline mail packing using either ^K or
the SysOp defined stop key in from the Language.Text file. The default
is the upper case S as shown below...
#900 Yes, No, Stop, Pause, Quit, Tag, View, Non-Stop, Edit-List
[Str]YNSPQTVCE
I changed it to the Escape key for myself...if you want to use the escape
key as the stop key, just cut-n-paste the lines below in your own Lang
file replacing the existing lines...
#900 Yes, No, Stop, Pause, Quit, Tag, View, Non-Stop, Edit-List
[Str]YNPQTVCE
3. Improved the breakout ability in the file search. For example, if your
search string is not found anywhere in all the file areas, you could not
break out of the search earlier. This is fixed now...:)
4. Changed the default new user birth date entry to YYMMDD to be consistent
with the rest of Adept. This would require a modification to the current
Language.Text file...change the existing #320 to read
#320
[Ans][Cvt][Str]\x1b[0m\r\n\n%04u/%02u/%02u
[Asc][Cvt][Str]\r\n\n%04u/%02u/%02u
5. Changed the order of the Date in the User Editor as well to the same
YYMMDD format for consistency. Also added shortcuts to the various
buttons in the user editor.
Ok Ok...why the date format changes?
a) to keep a consistent interface, and,
b) I do aim to localize them later :) but I do need to get them
consistent first...
1.09.51a
--------
1. Changed the default font for the Events and OutBound box... :)
2. Changed yet another mem alloc to be consistent...
3. I think i got 'em all now...all mem allocs have been visited and
revised where required...
4. Added the ability to change the outbound path in Adept via the
Config -> BBS -> Misc Paths menu...Although you could always have used
the environment variable to configure it, but seems some people had
probs finding the outbound dir...
This change uses an un-used part of the BBS_Configuration file (most
prolly reserved for such future extensions :)) and will be transparent
for all purposes...
The environment variable will still over-ride the Path set in the
outbound as it should...
5. Put back the line in-active menus on the status RMB as they were
earlier, but now the main mailer menu also reflects the change...
Also changed the menu text to be more accurate...i.e., Toggle :)
6. ************ NOTE *************
I forgot to document a required fact about the 51 release....If you use
the Archivers file from the full installation archive of 51 (VAC), then
a) either remove all the pkzip/2 2.50 related entries from it, or,
b) run pkzip/2 with /config /nozipextension FIRST and then make sure
you have PkZip/2 2.50 in the path...
1.09.51
-------
1. Found yet another small mem leak...fixed!
2. I am hoping that this version would be quite stable...:)
1.09.50z
--------
1. Oh well...after all those offline mail changes, i still forgot a little
something in another place :( Anyway, this time it should be fine!
2. Got a rare case trap in the message reply section, fixed on a temp basis
for now...till i can figure out why it got to that point in the first
place...
3. The Caller log will no longer save the local login info.
4. Some cleaning and re-shuffling as usual...experimenting as usual :)
1.09.50y
--------
1. Added the option to save the BBS and active line config files to disk
from the programs menu.
2. A re-prompt would generally not actually re-prompt :) i.e., would not
re-display the prompt, but would cycle back to gather the input. For
example, press the help key (TAB) while entering your name at login
and you won't see the prompt to enter the name again...this is basically
a language file prob...incorrect definitions...will try to correct all
such instance and include an updated file in the archive...
3. The Adept.Help can now have metas/ansi/adept codes in it.
4. The password would display after a re-prompt instead of *...fixed!
5. The caller logging is back...optionally...for now...specify
/NC_CallLog on the Adept command line to enable it...also back is
the corresponding menu option #925
This will require 2 changes to the Language.Text file...In addition
to #865 which is already in use for the caller log header
#866
[Cvt][Mta][Str] {B1{FG%-7lu {FF%-20.20s {FB%-2u %-5u {FC%-15.15s {FD%s {FC{FG
is now used for the lines, and,
#867
[Cvt][Mta][Str]───────────────────────────────────────────────────────────────────────────────\r\n
is the ending line...
If there is posivite feedback on this, I will try to enhance and expand
these functions...Will also include a Caller log maint (trunc) utility
if required...
6. The compile language file was not quite working...does now :) There is
a little something you should know about it though...no messages/prompts
will be updated till the compilation is completed and it works the same
way as on Adept startup...i.e., the Language.Text file must have a newer
date/time stamp than the compiled files. There is no message about this
in the display...it is only logged to the Adept.Log.
1.09.50x
--------
1. Tried to make file/message area searches a wee bit faster...are they?
Need to get some more feedback on this...if it traps during any
search etc...
1.09.50w
--------
1. Changed the logic for offline mail export checks a bit...dunno, but i
think it might be more robust this way...let's see...
2. Please note that fact that the compile language file option added in 50v
works just like when you re-start Adept, i.e., requires the Language.Text
file date/time stamp to be newer than the compiled file versions. This
will be the case if you edit the file anyway. To force a re-compile, just
open and save the Language.Text file so that it's date/time stamp is
newer than the compiled language files' stamp.
3. Plugged in GrK's Busy Flags code into gatekpr...might help solve some
flags related probs...BTW, Adept does properly clear flags on manual
clear as well as events clear...checked it...works! :)
4. Changed the offline import fn a wee bit as well...removed redundant
code from it...
5. Removed the non-functional FTS-0001 from the mailer menu till I can
figure out why it always causes grief for me (or anyone else with bad
line conditions...noise etc...)
6. Duh...forgot to put in an exception handler around the other GIO call
to stop the GIO probs...done now...'cepting in the case #7 below...
7. Made RealTime transfer info the default for the transfer status window,
seems the lazy update time runs into major probs specially on local
vmodem and telnet i/o. This default will apply only if u re-create the
BBS_Configuration file, otherwise, just click both mouse buttons in the
transfer box, and if u see the option "Real Time Data Updates" just
click on it (otherwise you will see "Lazy Data Updates" which means
you are already set up for real time updates which is fine).
This might also have some effect on high speed net/modem connects as
well...keep in mind, in case of high speed i/o probs, turn on the real
time data updates option!
8. The Full Screen Editor allowed the subject/to to be changed even when
it was forced...somewhat fixed...altho it will still allow the change
in case of the area attributes being forced, but when forced by a menu
item (or rexx script) it will remain forced :)
9. The poll dialog in the outbound window will now default to the last
mailer node from the line that has the current focus, instead of none
or the last you chose...just something I reqired for my small setup...
did not want to enter the node # to poll after a shutdown...:)
11. As usual...some more meddling :)
1.09.50v
--------
1. Corrected the date stamp in the AdeptPostTextMessage fn.
2. Added the compile language prompts option to the Programs Menu.
Although not *very* critical, but please make sure no users are
logged in while trying this option :)
3. Added the 3 new metas Hagar added to the .69 version, but with a small
difference...the new metas are:
{VP -> Date and Time of the first ever login of the current user
{VQ -> Date and Time of the last login of the current user
{VR -> Number of users based on the current userbase file
NOTE: the {VP, {VQ and {SD metas all display the date/time in the
same format, have no text attatched to them and will *not*
tag in the newline after the date/time string. This is
consistent with all the other displays in the BBS.
4. Added the menu type 1003 (as per the Hagar .69 build) to change the
language type from a menu option.
5. Added a new REXX function to copy a file. The Syntax is...
Call AdeptFileCopy AdeptLineNumber, 'c:\config.sys', 'c:\config.bak'
i.e., Copy the c:\config.sys file to c:\config.bak
6. I finally got a fix on the GIO trap problem, but, unfortunately,
nothing much I can do about it...it's in GIOMT.DLL itself. Here
is what happens...
Tag a cpl of files for d/l and select ZedZap and start the d/l...drop
carrier in the mid or try to hang up on the user in the mid from the
local Adept display...it waits, thinks about it...and...TRAP!
(No...it's not a VAC build prob :) happens with any WC/VAC build)
This does not happen with ZModem tho...and, does not happen with the
older GIOMT.DLL (from 1.07f). Since using the older GIOMT.DLL requires
you to use the older AdRTL.DLL and the older AdNNTP.DLL (and maybe the
older ConnLib.DLL and AdFTP.DLL as well) I have, for the time being,
put in a forced exception handler around the GIOMT calls in Adept...
Also re-enabled the forced exception handler in the SMTP 'cause of the
above...so, till we get hold of the GIOMT, AdNNTP etc src, this will
most probably remain...:(
7. Some more meddling and cleaning as usual :)
1.09.50u
--------
1. Corrected a small prob in the AdeptToss.Log code...would generally
fail to dump the Log when Adept was shutdown if there were any areas
to dump...
2. Updated to UserSrvr DLL to use an OS/2 specific function for flushing
the disk writes...should be more efficient...
3. Changed the shutdown a bit...Adept should be killable now...hopefully
if I got it right...otherwise I will just have to try again :)
4. Saving of line configs if now logged to the Adept.Log
5. Added a user file nuke function to the Programs menu..."Nuke User File".
The shortcut for this is ^o. This will run a file called:
AdeptUserNuke.Cmd
if present in your Adept base directory. Before calling this file, it
will toggle DTRs for all nodes, make all lines in-active and then close
the userbase file.
I have added this option for those who want to manage their userbases
and yet, not give up the high uptimes that one sees with the newer
Adept builds :)
It is *your* responsibility to ensure that no users are logged in while
attempting to run this option...A DTR toggle may not work all the time...
To be extra safe :) you can manually set all nodes inactive first and
then try this option.
My AdeptUserNuke.Cmd looks something like this...
@echo off
uNuke /r /q /e30 /d10
ReIndexUsers
WARNING! You are on your own with this. It works for me...you gotta
check it out for yourself...:) Just make *sure* no users are
logged in when running this.
Also, if there is no AdeptUserNuke.Cmd file, then it does nothing!
BTW, the AdeptUserNuke.Cmd could also contain just a copy statement,
which would allow you to make a backup copy of the userbase...
6. Added a system wide file flush to the status timer (for now) which
is on by default. To switch it off, just re-start Adept with
/NC_NOCOMMIT
The flush is on it's own thread so as not to block the GUI...
1.09.50t
--------
1. I corrected and removed what I thought was a typo "MainNC.Cmd" :)
It's back now...
2. In case you have any grundged messages in the Internet\SMTPOut dir
Adept will trap right after startup...this occurs from AdNNTP.DLL
so not much we can do about it for the time, but will try to add
some checking in Adept itself to try to avoid these errors...
3. Fixed the outbound not scanning on startup...
1.09.50s
--------
1. Fixed the reply to trap over telnet in NetMail...was in the new
Netmail file attach code ...and just in case... added 2 new command
line parms...
/NC_NONETFILE -> Will disable the NET FILE ATTACH feature
/NC_NOLOCALFILE -> Will disable the LOCAL FILE ATTACH feature
2. Cleaned up some more debug code...
3. Corrected a type (yeah...my doing :)) in the NNTP which might have
caused a problem on some specific systems...
4. Meddled with some display and prompts...pesky things...:)
5. Please note that all command line parms starting with NC_ are on
a temporary basis and might be removed unless I get some feedback
that they are required/used.
1.09.50r
--------
1. Tried to correct the occasional local display corruption.
2. Removed the additional logging by default, can be enabled by
/NC_VERBOSE for now if req. Will be removed completely in the
future unless some people want it.
3. Some more un-used code removed...
4. Improved the file d/l counter updation...
5. Improved slightly on the AdeptToss.Log logic...
6. Globally changed some code sequences to be more efficient and smaller
7. Added the SysOp Next to the Mailer Menu, but not quite sure of what
was actually req...forgot :) Might not work as expected till I find
out what exactly was req here...
8. Fiddled some more with the timings and the DCD logic as every GIO trap
I have seen so far, starts there. Since we do not have the GIO source,
we can only experiment and see how best to avoid this...
9. Added yet another command line parameter to control the SMTP thread
stack size, i.e.,
/SMTPSTACK 262144
The default is 64k, in case there are any probs with SMTP packets
you can increase the SMTP thread stack size.
10. Removed some more un-used code and cleaned up some more prototypes.
11. Fixed a problem in the SMTP...when I made all the paths consistent,
forgot the hardcode in the SMTP module...found this out thanks to
dutch :) In the process...found and removed some more un-used code...
12. Since the paths in Adept are now consistent, you should be able to
run Adept (only the VAC build for now) from any base directory and
a) the SET ADEPT= and the set APATH= are no longer
*required* for Adept (you can keep those if you must :))
b) No paths require the ".\" in the start...for eg, in the Adept
Config->BBS Settings->Misc Path/Files Menu. You can remove
the leading ".\" in the paths...no longer required...This,
actually, never was required anyway...even earlier...AFAIK...
13. More fiddling...as usual :) Made some more messages consistent...
there were (and maybe still are) a few messages that were repeated
but had *slightly* different text...
14. Added a new command line switch /NC_FULLNAME (for now) which will
not allow a new user to use a single word (alias if u plz :)) for
a name, or, even 2 single chars seperated by a space...this is good
enough for me...I used to do the same in my startup rexx script...
do let me know if this can be improved on in other ways...ne1? :)
15. Added Hagar's new file list display code. Here is a cut-n-paste about
it from his readme...
Scott Drake of isonline.com fame did not like the default adept
files lister display, he says rest of world is using the Quickbbs/
RemoteAccess/Proboard/Lora display style, so ok here it is, if
you add 100 to any of the file display menu commands (ie 73
becomes 173) you will now invoke the new display routines. For
this to work though you need to add two lines to your
language.text, make lines 870 and 871 look like this:
#870
[Mta]\r\n
╚{FF{B2 Name
{B1════════════{FC{B2 Size
{B1══{FB{B2 Date
{B1═══{FB{B2 Description\r\n
#871 actual file name date size [Mta][Str]{B1%u\r\n └> {FF{B1%-30s {FE%5luk {FB%s {FC[%5.5lu] {FD%s
\r\n
[Asc][Mta][Str]{B1{C1\r\n └> {FF{B1{C3 {FE{C5k {FB{C0 {FC{FD{CB {CA{FC\r\n
[Ans][Mta][Str]{B1{C1\r\n └> {FF{B1{C3 {FE{C5k {FB{C0 {FC{FD{CB {CA{FC\r\n
I suggest cutting and pasting the above directly into your
Language.Text. Course if you like the old display style you don't
really need to modify language.text, however u should just to stay
current.
16. Finally got hold of the SMTP and other related problems...duh...some
optimization and incorrect prototyping probs...hopefully most, if not
all, fixed now...
1.09.50q
--------
1. Added the 2 new metas Hagar added to the Official WC build
a) {VN = Handle of last user on this node
b) {VO = FullScreen Editor ON/OFF.
2. Hagar changed the GateKpr switch to /X instead of the /T used in the VAC
build to force GateKpr to quit if no AdeptToss.Log was found. The VAC
build will now accept both /T as well as /X for now...also added the /S
option to force GateKpr to ignore the AdeptToss.Log completely.
3. Added a few more metas...please keep in mind that these may not be
supported in the Hagar version and might be removed from here as well
if Hagar decides to use any of these codes for any other purpose to
maintain compatibility.
{SL -> Free Physical Memory
{SM -> Free Virtual Memory
{SN -> Number Of New User Calls For The Current UpTime
{SO -> Number Of Actual BBS Logins Completed For The Current UpTime
{SP -> Number Of OutGoing Mail Calls For The Current UpTime
{SQ -> Number Of Incoming Mail Calls For The Current UpTime
{SR -> Number Of FAX Calls For The Current UpTime
{SS -> Number Of Times Adept Got a Connect String For The Current UpTime
{ST -> Number Of Socket Calls For The Current UpTime
{SU -> Number Of Telnet Calls For The Current UpTime
{SV -> Number Of VModem Calls For The Current UpTime
{SW -> Number Of Modem Calls For The Current UpTime
{SX -> Number of the currently running Tasks
{SY -> Number of the currently running Processes
{SZ -> Number of the currently running Threads
4. The occasional traps in the file area list/d/l should now be fixed...
as well as the telnetting from the BBS...hopefully :)
5. If a user uploads an incorrect extension offline mail packet, Adept will
display prompt #911 from the Language file.
#911
[Cvt][Mta]\rCould not find *.Rep or *.New\r\n{PC
6. Enhanced the file tag dupe check...
7. Fixed a possible trap condition in the NNTP
8. The NNTP got messed up a bit in the earlier build...just to remind me
that there's a whole lot I still don't know about Adept I guess :)
Fixed!
9. Removed the menu item logging...hopefully no longer required :)
1.09.50p
--------
1. Removed some more un-used code...
2. Made a small change in the view archive function in the file listing
and updated the Archivers file for PkZip 2.50 and other InfoZip/UnZip
formats (mostly the sfx kinds).
3. PkZip 2.50 for OS/2 is quite a lame one :) don't try using the /extra
option with /add...hehe :)
4. The new user menu exec got messed up in the earlier changes...would
not be called if rexx was enabled, only the NewUser.Cmd was used...Fixed!
5. Some more re-arrangement of code...should make this build faster than
the earlier ones...hopefully :)
6. Added a message to the user if an offline mail import extract fails for
any reason as well as if the user u/ls a file which does not have an
extension of .Rep (QWK) or .New (BWave)
1.09.50o
--------
1. Just some general cleaning...nothing special...:)
2. Under a certain condition, Adept could write out a 0 byte
AdepToss.Log, this is now corrected...Altho GateKpr would have dealt
with it without any probs, but fixed in Adept as well...
3. Changed the file name of the toss log to AdeptToss.Log, which is what
the Hagar version will also use from now on...well...since Adept does
require HPFS...:)
4. Those of you who are using compressed message bases, please be aware
that this can cause probs as has been the experience of most Adept
SysOps I have talked to...try to uncompress the message bases and then
re-try running Adept in case of problems...
5. Disabled some VAC specific code which was causing probs for some ppl
1.09.50n
--------
1. Missed the correction of some spawn settings for the mailer...Fixed!
2. Setting Call Collision off was not implemented...should work now...
3. And yet again...this time import offline mail...duh...if a large
packet was u/l and then the carrier dropped...well...hopefully fixed!
4. Finally found and fixed the GateKpr/VAC prob...made the VIO and PM
version messages slightly more consistent...
1.09.50m
--------
1. Forgot to fix the spawn flags in the Events...and i thought i had
already done that...duh...:)
2. Royally screwed up the last cpl of rare case offline mail fixes :(
was fixed so only the first user on any line could d/l...duh..
Re-Fixed...works now :)
3. Fixed a cpl of semaphores and got rid of un-used ones...
4. Please remember to use the DLLs from 1.09.50l
1.09.50l
--------
1. Had not intended to hand out 50k to anyone, but...:) To avoid confusion
shifting to 50l for this build. I am now going to drop the VAC beta
linker for this and the future builds as some people have reported some
problems due to OS/2 versions etc (I think...)
Please, try not to mix-match the Hagar WC and my VAC DLLs or EXEs...
might cause probs...
The beta linker did make real compact executables tho...the normal linker
makes them slightly bigger...:(
2. Changed the outbound icons' size as suggested by dougk. Sorry Doug, I
cannot really store anything else in the INI for now as there is a whole
lot of Adept I still don't know much about and am still in the cleaning
process, but yes, I do have the same thing in mind...someday...:)
3. Changing the default menu type to the old type as the new ones are more
easy to detect. It will still auto-switch on the first login, if req.,
but the default startup option is now OldMenu.
Oh well...the truth? :) I find 'em easier to use, specially while testing
etc and they are faster...at least for my setup...so...
4. Made some more corrections to the spawning routine calls...
5. Corrected another rare condition offline mail export prob...
6. Duh...how could i forget an option to view the system log? Added! :)
Shortcut is ^y and it will run the file AdeptLog.Cmd if found, else
will open the Adept.Log file in EPM.EXE
7. Duh...why me? Again got a rare condition in offline mail exports
(no traps tho :))...hopefully fixed for now.
8. A small mod to the outbound re-draw process...
9. Added a shared semaphore to drop the AdepToss.Log to Adept and GateKpr.
Adept will now drop the AdepToss.Log whenever GateKpr is run. Will
include the full specs of this functionality as soon as I have confirmation
that this indeed works as expected :)
BTW, Adept will still drop the AdepToss.Log on ShutDown.
1.09.50k
--------
1. There is some prob in still left in the spawning routines, till I can
find and fix that, I have plugged in work-arounds to stop the traps
related to this...
2. BackGround QWK mail import would either not complete, or, trap Adept in
case of a carrier drop. Fixed!
BTW, this was ok in bwave import, no probs there...
3. Changed Text and Fido Export as well now to be consistent with the QWK
and BWave Processing.
4. There have been lots of quick fix mods to Adept which generally defeat
the idea of multi-lingual stuff and some similar things...trying to find
and correct all such instances...I just might break some stuff at places
but I guess it has to be done if the program has to be maintained...
I am depending on all people using the VAC builds to point out any such
problems (in case I break any menu/script path etc.), it won't be
anything very serious and nothing that cannot be fixed right back in the
very next build...let's hope for the best and see... :)
I will be making these changes in blocks, not all at one shot, so even if
something does go broke, it will not be serious and we will end up with
a far more robust model in future...
************ NOTE *** NOTE *** NOTE *** NOTE *** NOTE ************
5. Okay, as a result as #4 above, we also have and extended AtLoginX.Cmd or
AtLoginX.Menu, WelcomeX.Asc, NewsX.Asc and BulletX.Asc. A VERY important
point to keep in mind is that fact that although you may never require
or use the plain *.ASC files, they MUST be present!
Most places pre-check for the presense of the *.ASC before actually
trying to find *.ANS or *.AVT etc. If the corresponding *.ASC file is
not present in the Adept\Text directory, some files may NEVER be
displayed even though you might have the ANS/AVT file for it.
I will most probably make this behaviour consistent thoughout Adept...
already started actually :) Just keep in mind that any Standard
(Pre-Defined) Adept display file MUST have at least a .ASC file and
if should not be a 0 byte file. You can place 1 byte dummy files if
you never use the *.ASC, but they have to be there to display a
corresponding ANS/AVT etc...
************ NOTE *** NOTE *** NOTE *** NOTE *** NOTE ************
6. Offline mail export again...:) Seems that most work and updates were
done on the bwave export, so I corrected a few inconsistencies still
left between bwave and other offline mail exports. Will do a proper
cleanup later, for now, they are more consistent than before...
Also, any background mail export (monitor off) should now work equally
well in local mode.
7. Files in the System\Line0 directory, i.e., left over files from a
local login will no longer be deleted...at least we can see those
later...:)
8. Well...can't blame me for not trying...:( I tried to use the variable
paths that were a part of Adept to begin with, only to find a few
places depended on hard coded dir names...so...I made it consistent in
that fashion :) Altho I removed all references to the base dir being
Adept (yeah, now you should be able to run it from any dir), but, the
dir names inside the base Adept dir will remain hard-coded.
9. BTW, the difference in speed shows up even in comparison to the 50j
version...not bad eh? :) The cleaning process does seem to be making
Adept faster...let's hope it stays as stable as well...:)
For those of you who might be interested...the old menus run almost
twice as fast as the new menus for now...at least on my setup (very
plain and simple). Both share the same menu item processing routines,
it's only the basic menu display/structure/parsing that is different.
10. The shortcuts for init, answer and next call were not quite working
as they should...Fixed!
Alt-P would bring up the Poll Dialog box instead of pulling down the
program menu...Fixed!
^P still brings up the poll dialog box.
Changed the shortcut for next call to ^N rather than ^R.
Changed the shortcut for Get Files Dialog from ^G to Alt-G
Added ^G shortcut for the Group Editor.
Added ^U shortcut for the User Editor.
Added Alt-X shortcut for ShutDown-On Hook
Added ^X shortcut for ShutDown-Off Hook
Added ^C shortcut for Clear All Calls
Removed the Extra Seperator between Clear Calls and Make Line InActive
from the Mailer Menu
Added ^t shortcut to read a file called GateKpr.Log from the LogFiles
dir, or, run a file called GateLog.Cmd if present.
This is similar to the line log view command now which would run a
file called ReadLog.Cmd if present, else view the log for the
focussed line. These commands will no longer try to run/view a 0
byte file...
Added a new "Run Manager" command to the outbound window which will
run a file called OutBound.Cmd if present. This can be used to run
any external outbound manager (like bill-e's for instance :))
Pulled the InActive lines options from the Adept Status for now as
they were not co-ordinated with the resetting back to active. If u
tried the make in-active option from the Adept Status RMB menu, you
would have to select Make Inactive from the main Adept Menu to get
to the option to make it active again.
11. Tried to clean up a huge (for me :)) mess in the OutBound window
processing...it would never get refreshed on any other option on the
RMB menu until you actually forced a re-scan by selecting the ReScan
option from there. There is a lot more to do here before it's finally
cleaned out...seems more functional for now though...
Also added a command line option to control the outbound re-scan timer.
/OUTBOUNDSCANTIME 300000 was, and is, the default. The time is in
milli-seconds. Feel free to experiment :)
12. Found something *very* interesting while trying to run 1024 nodes...
yeah...again :) Adept traps while trying to check out port 1000. Tried
this while SIO was installed...is it a SIO limitation/prob or OS/2...?
Anyone have any ideas on this one?
14. Changed the logic for the shutdown menu items enable/disable. Should be
faster now.
1.09.50j
--------
1. Adept will now drop the current AdepToss.Log on all of the following
a) Any Event Command is run
b) Select the Mail Proc (^M) from the Adept Menu
c) Adept is ShutDown
The AdepToss.Log format should make it simple for any 3rd party util
like GoldEd to incorporate it since the information in the AdepToss.Log
is just the Massage Area number in plain ASCII format and has only one
Message Area Number per line.
2. GateKpr's version number has been bumped up to 2.30.00 to make it
different from earlier versions which do not support the AdepToss.Log.
3. Added a new switch to GateKpr to make it exit if no AdepTosss.Log file
is found instead of processing all the areas (as earlier). The switch
is /T (as is Toss)
If a /T is specified, GateKpr will import all mail, but will not export
any area unless it finds the AdepToss.Log file, in which case, it will
export only the areas specified in the AdepToss.Log.
If a /T is *not* specified, GateKpr will import and export all mail,
unless it finds the AdepToss.Log file, in which case, it will export
only the areas in the AdepToss.Log.
4. Fixed GateKpr's problem of shouting "Probably insufficient diskspace"
etc. if you specified a relative path in the Area.Control file.
5. Cleaned up some more code in Adept. Re-Checked all the / and \ stuff,
found a few probs that might have resulted in incorrect display or
traps. Hopefully Fixed!
Re-arranged all such / and \ code to check for \ first, should be more
efficient (faster) now. Made similar chamges in the GCU.DLL code as well.
6. Backed out of the beta FP #2, OS2KRNL, PMMERGE and PMWP DLLs...no more
pesky traps so far...:)
7. If a user entered an invalid date in the new file search, Adept would
trap. This trap would also occur when ever the same get date was called
from any other place as well...(FTP etc). This will now default to the
current date in case of any invalid input. This should fix some of the
vague traps we have been getting while not watching :)
8. Added yet another checkpoint in all background offline mail exports
just to avoid a potential prob...dunno if that condition will ever
occur, but just in case...:)
9. Forced an exception handler where some people where getting traps for
now in order to get a hold of the exact location and to keep Adept up
and running for this known prob situation :) This will create a Trap.Log
file in the Adept dir in case of any probs. Please send over any Trap.Log
file(s) from your Adept dir if u get any...
10. Added one more log line in the file d/l couter update to check for any
probs etc. for instance a file d/l counter not updated when it should
and possibly could have been...If you notice any probs, plz send me that
portion of the linex.log file...Plus methinks it is called one more time
than is really req...let's see...
11. Fixed a problem in the d/l filelist...it always added an additional \
after the path and that is what was displayed as well...This was what
was causing the time delay in the file d/l counter update routine as
it had to search through the entire filebase for a pathname with 2 \s
in it...duh... :)
This would happen for all files in file areas that were marked for
copy to HDD from CD...
12. Have left some additional logging on for now...will remove it in the
next build...should know better by then...hopefully... :)
13. This build, in general, should be very resistant to the vague traps
that we sometimes got...let's hope and wait and see...:)
1.09.50i
--------
1. Duh...missed some more un-used code...related to squish...removed!
2. Fixed a nasty user generated trap in all background offline mail
exports. User selected not to monitor mail packing and w/o waiting
for it to complete, changed the option to monitor mail packing and
BANG! TRAP!...GRRR....Fixed!
3. Forgot to switch off the notify the user flag on the mail proc (^m)
the user would see the prompt "loading external door..." if you tried
^m...duh...Fixed!
4. Added a rough support for EchoToss.Log in Adept...
5. Duh...it was'nt only the APM...it was DosCalls.DLL in the beta FP #2
I guess...backed out of the DosCalls.DLL for now...
6. Okay...serialized the addition of message areas for the EchoToss.Log
from Adept using a semaphore...seems to be working great...for now :)
For now, Adept will dump the EchoToss.Log file on shutdown only...
7. Now all I need to do is to have common link between GateKpr and Adept,
so that when ever GateKpr is run, it will ask Adept to dump the currenty
posted to message areas to the EchoToss.Log file and then scan only
those areas....
8. Backing out of DosCalls and CMD.exe from the beta FP #2 seems to work
much better...altho there are still probs...Did help me to catch a bug
in the Adept Spawn routing tho :) The size of the buffer passed to OS/2
was hard-coded to 50...it actually turns out to be 60 now...duh...Fixed!
9. Okay, I dropped the idea of EchoToss.Log...no point in it and since at
least I don't know of any third party util which supports this for Adept,
Adept now has it's own AdepToss.Log :)
The EchoToss.Log format meant a huge Kludge in Adept as the only link
between Adept and the GateKpr areas is the Area Number itself and not
the name or the echo tag.
The AdepToss.Log has a plain ASCII format and just contains the Adept
area number, one per line that is to be exported. GateKpr 2.01.04 will
check for the presence of AdepToss.Log and scan only those areas if it
is found, otherwise it will scan all areas (as it already does).
Now...instead to having Adept write out to the AdepToss.Log all the time,
I have currently chosen a faster method (can be changed if req l8r).
If the running event command line has the words "Mail" or "GateKpr" in
them, or, if you select the mail proc from the Adept Menu, the current
AdepToss.Log file will be written out to disk. Hopefully this should
work fine for most people, but let's see...
The EchoToss.Log kinda support can also be plugged in GateKpr if
required...
BTW, Adept will also dump the AdepToss.Log on shutdown...
10. GateKpr.Exe Version 2.01.04 (PM) is included in the update archive
for now...
1.09.50h
--------
1. Changed the "Change Type" option in the outbound manager a bit. It
would appear on the menu even if nothing was under the RMB and if
clicked in this condition, would cause Adept to trap. Fixed!
2. Fixed the no timer check on telnet/vmodem calls. Should work fine now.
3. Fixed the netmail address guesser trapping on rare occasions.
4. Fixed a possible mem leak under very rare conditions.
5. The Archiver and Protocol selection would never default to anything
even though the prompt said so...Fixed!
Both will now default to Z, i.e., ZModem and Zip/UnZip
6. The Mailer Proc (in the menu ^M shortcut) would not run REXX CMD
files...at least, it did not for me...Fixed!
7. Removed a cpl of logging lines in the new menu driver which were
basically being repeated...all I now have to check is why the repeat...:)
8. For those interested, the probs I had with the Beta FP #2 were related
to APM. I turned APM off and works like before...solid! :)
1.09.50g
--------
1. Reduced the number of tries on getting CRC errors "YooHoo CRC Error" is
the message for those of you who might have seen it :)
2. There was a problem in the OS/2 Startup Date/Time display...the uptime
was correct. Anyway, I rewrote the damn thing to be far more quick and
and small as well as accurate...hope it remains that way :)
The System Startup Date/Time is now computed only once as we can safely
assume that Adept will not keep running while the system is re-started :P
Also made the computations for it consistent so that the compiler can lift
the non-variant parts making it faster...everytime...Should make it
slightly faster in the conversion Metas as well...{SD and {SE
3. Extended an already existing fn to display the formatted uptimes etc in a
consistent and pleasing manner :)
4. Think the prob was with the APM in FP #2 beta...can't be abs sure for now,
but if u r not using the beta FP #2 for Warp 4, go ahead and use the
/BBSWATCH switch...:)
5. Some other minor changes here-n-there :)
6. Made a small fix to the get index for message replies...got a trap there :(
Let's see....
7. Added a command line option to disable time checking for Telnet/VModem
calls on request :) Should not affect any other normal operation...
Just start Adept with /NOTELNETTIMER and it will not check the time on
Telnet/VModem calls...
8. Adept directly checks for the time left in too many places, so the above
switch will work, but with some limitations...hopefully we should be able
to work them out later...currently, it will just keep increasing the user
time left...
1.09.50f
--------
1. Made the BBS Watch thread a command line option. It is now disabled by
default. The option is /BBSWATCH to enable it. I applied the beta FP #2
and the Watch thread causes a trap on Doscalls, but other programs also
seem to have a similar prob, so I can't be sure if it's the Beta FP that
is the problem or something in the watch thread. So...till we are sure
of the cause...it remains off by default, but, can be enabled from the
cmd line if required.
2. Globally changed all '/'s in pathnames to '\' to be more consistent as
a mix of both were being used.
3. Some Squish code was still enabled in places...might have caused probs
in some cases...as there was no backing code for it. Removed!
4. Added a flimsy change pkt type to the outbound manager on the RMB as per
dougk's suggestion. It is currently flimsy in the sense that it will
cycle through the pkt types on every click. C -> H -> N for now :)
Will improve on this later if this is what was required....
5. Added logging for locked out and deleted marker users. Trixer could not
figure out why this build threw him out on a local login...his a/c was
marked as deleted :)
6. Pulling files from 1.07f and Hagar's full version to make a full version
for the VAC build as many people seem to have problems "finding" a full
archive...Will try to make a better, updated full archive later...
1.09.50e
--------
1. Extended the user logoff check for offline mail export to both cases,
Monitor off and on...had a slight prob here, so this should fix it :)
2. Fiddled a bit more with the answer of RINGs...
1.09.50d
--------
1. Some users still had probs exporting mail d/l packets in the background
so removed the not so required checks which I thought I had removed
earlier as well...:)
2. Fixed the Mail Incoming/Outgoing Counters on the Status.
3. Fixed a problem with GateKpr in addressing. If it could not for any
reason make a guess at a matching zone/net/node etc. it would set the
number to 0. At least, AFAIK, Zone cannot be 0, so it will now default
to the originating Zone in case it cannot guess the correct one.
4. Removed the "Hit Enter" after the "offline mail is ready for d/l"
prompt when a user selects offline mail packing in the background.
Causes a loong loopback and not really required AFAIK...
5. Updated the status with some more info...:)
6. Corrected the display for the Metas K6, K7, K9 and KA...They would show
huge (out of bound) values in some cases. For example a user d/l a file
marked free or bytes free which is greater than his/her d/l limit, the
corresponding d/l left Meta would display well...all you have on the BBS
as the d/l Kb left...The difference was being printed as an unsigned
number...duh :) Fixed!
7. The New/Re-Arragned Status Line now looks something like:
"[CONN: 0] [MDM D/V: 0/0] [TEL: 0] [SOC: 0] [FAX: 0] [BBS: 0] [NEW: 0] [MAIL I/O: 0/0]"
Where:
CONN -> Total connects to Adept
MDM D/V -> Modem Calls (Dial up/V Modem)
TEL -> Normal Adept Telnet Calls
SOC -> Normal Adept Socket Calls
FAX -> Fax connections
BBS -> Actual logins to the BBS
NEW -> New Users logged in
MAIL I/O -> Mailer Calls (Incoming/Outgoing)
I guess this is enough info to get at a glance to see what all happened :)
1.09.50c
--------
1. Made some changes to the incoming call loop. Might help solve the pesky
RING1 etc answering problems. Should not affect anything else otherwise
excepting to *possibly* remove the multiple RINGs that one sees at times
before Adept actually decides to answer...let's see...
2. Fiddled with some more timing factors for the mailer EMSI sessions,
might improve some things...then again, might not :)
3. Added check in all background offline mail export type for user logoff.
It could bring down the BBS in some cases...did here in my local
checking anyway...hopefully fixed!
4. Added a number of mail calls counter to the status...let's see it it
works :)
5. Added dumping of the Adept Status stats to the Adept.Log on shutdown.
6. Moved the BBS Name from the status to the Status window Title Bar...
gives us more room for junk on the status :) Only drawback is that if
you change the BBS Name in the config, the change will not be reflected
till you re-start Adept as the Title is set on Startup and not changed
after that...
7. Enabled a feature that was earlier causing a prob on dutch's display
screens. Should no longer cause any probs and increase the display
speed.
8. Removed the currently useless SMTP Stats button...
9. Changed the default NNTP Server stack default from 128k (50b) to 96k
10. Changed the Status display to be more compact...it uses NM for NewMenus
and OM for OldMenus now...
1.09.50b
--------
1. Cleaned up the background offline mail export LMR updates code...should
be more robust now...Also added the updation of LMRs to the line log for
now...
2. Increased the NNTP stack size as bjoolav was getting a trap which shows
an outta stack prob. Most probably stems from the NNTP dll, which we
don't have the src for as yet, so trying an increase in the Adept NNTP
thread stack size...let's see...
3. Added a command line parameter to set the NNTP Stack size for the NNTP
server thread. /NNTPSTACK xxxx
The NNTP Server thread now defaults to 128 kb of stack as do the Mailer
and BBS threads. IMO, even assuming a 64kb+ msg in the NNTP now, it
should not run out of stack space as the earlier size was 64kb which
is now doubled to 128kb.
Will ask bjoolav to determine the optimal stack size...perhaps 96kb
ought to do it, if so, i'll switch the default to 96kb, till such time
tho, the default remains at 128kb.
4. Added some more logging to the menu driver and menu items used, and added
the logging of the offline mail packet type selected by a user.
5. Duh...yeah, will try to make a full install pack for the VAC builds...
soon :)
1.09.50a
--------
1. Version numbers jumping tooo fast, so will now keep version numbers for
only tried and stable builds, otherwise it will be 1.09.50x, where x is
gonna be from a-z :)
2. Another LMR update prob reported in the background mail export. Fixed!
1.09.50
-------
1. GRRR...the file d/l counter would really take CPU after a mail d/l...
searching for the file thru the entire file base to match and update
the counter. Fixed!
2. Added checks for mail packet files (polling etc) in the File d/l
update. That should do it :)
3. Corrected a minor inconsistency between how the mailer added files
to the xfer list and how the normal d/l and tag did it. Nothing big
just made the behaviour consistent to avoid possible probs in future.
4. Hrrrm, if u have files in CD, and the area also has the copy to hdd
flag set, the d/l counter for those files will not be updated for now
as it requires a kludge and some stuff is tied to other places in the
BBS which will have to be re-worked before this can be done.
Alternatively, if I can assume that there are no dupe files on a CD,
the counter can be updated, let's have some feedback on this issue...:)
5. Ok, in case a file has the attributes of copy and kill after transfer
(typical for files on CD), the d/l counter will be updated based on
the first matching filename in the filebase for now...It will still
match the complete path in case of files on HDD (as long as they are
not marked copy to HDD :)) Have added logging for the file d/l counter
update for now...will remove it later after we know there are no probs.
6. Minor changes to FTP and File listing search fns. Just tried removing
a kludge, nothing special.
7. Found another possible source of probs/leak in the file tagging. Fixed!
8. Added the initials in the msg reply in the FullScreen Editor.
When a user selects to reply to a message and selects to quote from the
original msg, Adept used to put in just " > ", it will now plug in the
initials as in " XX> " as is generally done by most such editors.
9. Just some more general re-arrangements...and added a small fix for dougk
so that the reply to in a netmail msg will not have a null addr. It will
be 0 rather than completely absent.
10. Just a mention about READLOG.CMD. This file, if present, will be run
when you select the view log menu option and it will be passed the
current line number as a parameter.
11. Re-compiled the new Fsys, GCU and UserSrvr with the single threaded
libs as it should have been in the first place. No probs, just adds
un-required bulk. Same for uNuke and ReIndexUsers...
12. Made a small change to the BBS Watcher thread. Let's see how it
works out.
1.09.49
-------
1. Changed the cheat check a bit to get around the CBV bug for now...let's
see...
2. Changed the last 2 stack allocations I had left out earlier to heap
allocs. Should solve some rexx call probs (if any).
3. Changed the logging in BWave/QWK to be consistent. Both were logging
differently earlier.
4. Please note the fact that the earlier added menu types 8/9 are part of
the base menu driver and cannot be directly accessed via rexx using
AdeptMenu calls. (Why would anyone need 8/9 in rexx anyway :))
Only menu types above 10 are accessible via the AdeptMenu Rexx command.
5. I am now gonna increase the logging, so that we know what's going on...
Please...someone...lemme know if I overdo the logging :)
6. QWK Export was using a background session to pack mail, whereas BWave
was using a child session if the mail monitoring was on. Changed the
QWK to be consistent with BWave.
7. Slightly optimized the thread and process count display, required two
seperate calls earlier for similar processing, now makes only one call.
8. Removed a VAC kludge from FSys.DLL, does make it slightly smaller,
perhaps slightly faster as well :)
9. Added the functionality to update the file d/l counter on a successful
file d/l. Now we can find out which file was leeched the most :)
Oh well...let's just say we will now get to know the most popular one
on our boards...the file I mean :P
10. Made a minor change for Adept to be able to answer to RINGxx as
documented. Cannot be sure of this, but should not make any diff
to normal answering...Anyone who checks this out, please let me
know if Adept now responds/answers to RING1, RING2 etc when you
set it in the Active Line Config.
11. Everything seems to be working and in place for now...the CBV carrier
drop no longer causes a prob, but will improve on it so all relevent
user stuff is preserved, not just the time (in case of the cheat).
1.09.48
-------
1. GRRR...in correcting the prob pointed out by Mario in the bwave export,
I broke the current bwave normal message import. Fixed!
2. Also changed the multiple mem alloc/free's in the offline import to be
the same as the export procedures. This improves the consistency as
well as the speed.
3. Releasing this now 'cause of the offline import prob...
1.09.47
-------
1. I had made a modification to free mem in email/nntp posting which turned
out to be incorrect as pointed out by Mario. This would have caused Adept
to trap in case a user u/l email/nntp from a bwave packet. Fixed!
2. Think I found the prob spot for dutch's trap. Was an allocation on the
stack, changed it to normal heap alloc.
3. Added the latest Hagar (Mario's) GateKpr compiled for PM to the archive.
4. Made some minor changes to the exception handlers.
1.09.46
-------
1. Thanks to dougk for getting the user server (and some more) source from
JM the user file flush function is now a part of the new usersrvr.dll.
This is how it works now...when ever Adept makes a call to any functions
in the user server which change/add/modify any user record, after the
update, the user file is flushed to disk (closed using a dup handle).
This will ensure that the userbase is always correct on disk. The only
problem that remains is the user index file which I have not done the
same to (since we don't have the source to that), but that's not a prob
as the index file can be re-created in case of any problems...In fact I
always run reindexusers from my AdeptStart.Cmd before starting Adept to
make sure it is in good condition.
This way, there will be no performance hit and is transparent to Adept
itself. The only difference will be that the user file will be updated
on disk after any change.
2. The timer i fiddled with is back to what it was since i could not make
out the diff...so better leave it as was for now...:)
3. Also, 'cause of #1, pulled the flush user file I added to Adept itself..
4. Checked out the menu functions 820/821 (DOS Door Runner) over VModem
Telnet ports 5 to 16 using a game called Lord. Works fine!
5. Re-compiled FSys and GCU DLLs for using P5 optimization...:) No other
change in them otherwise for now...BTW, the User Server DLL was also
compiled using P5 opts tho it won't make much of a diff there as it
mostly consists of calls to Btree.DLL.
6. Would need feedback on this build since a whole lot is changed...seems
to work just fine here...for now...:) and yes, it is faster...
7. Re-compiled uNuke and ReIndexUsers using P5 opts. No Change otherwise...
8. Had to do some work on FileEd, but got side-tracked with all of this
for now...will get around to FileEd probs as well l8r :)
1.09.45
-------
1. Added a user base flush function. When ever a user record is now saved,
Adept will also flush the user base file to disk, preventing the loss
of any user data and the need to shutdown to actually flush the user
base file to disk.
2. This is just in case...since i changed the main display fonts to WarpSans
anyone using Warp 3 w/o WarpSans installed, just drag-n-drop any font of
your choice to anyplace on the BBS Status display to change the default
if required.
3. Someone said the "kill log" function did not work. It does...now :)
1.09.44
-------
1. I finally got the file list display prob. Fixed!
Dunno if it has any side effects for now tho...
2. Some more mem alloc changes and checks added at some places.
3. Found and fixed yet another mem leak in the new menu driver. Musta
been outta my mind to have missed this one :)
4. Mario pointed out the fact that the DOS door runner I plugged in
does not work for COM ports beyond COM4. This is a limitation. You
have to have a port and irq configured for vmodem and the same setup
in the door...(menu types 820/821)
5. Changed the frequent mem allocs in the new driver to a single one,
hopefully no probs 'cause of that and menus should react faster.
Of course, I am assuming that no one uses more than a cpl of thousand
hots keys per menu screen....:)
6. Re-arranged the repreated mem allocs/frees in all offline mail
exports to alloc only if req. This alloc/free was happening for
every msg. Should make a diff...let's see :)
7. Changed most of the 8.Helv fonts to 9.WarpSans...looks neater, and,
of course, it always was available for Warp 3, so... :)
8. Changed the timings between rings a bit...let's see...was 35000 ms.
9. Generally improved consistency in checks where ever i went, so that
most of them are equiv now when checking for the same thing.
10. In case I forgot to mention it earlier, the BBS Watch thread is enabled
again, with logging.
1.09.43
-------
1. Nope, the file list display fix is some thing else. What I did as a trial
earlier is now removed.
2. Trying to fix the NNTP prob. Globally changing file permission flags.
Let's see...seems to work here for now...
3. Added Task, Process, Threads, Physical and Virtual Memory Free display
to the Adept Status.
4. The Adept Status Window's Status was not being updated on a call
collision. Only the Status in the main window was...Tring a "fix" :)
There were 2 individual checks for "NO DIALTONE" and "NO DIAL TONE",
currently changed it to one and will check only the first 7 chars, i.e.,
"NO DIAL". Let's see....
5. Cleaned up some discrepencies in the offline mail export code and
sorted out the prob associated with some setups regarding offline
mail export. You must *NOT* have [Mta] in the language file for
74, 112, 114, 646, i.e., Do not use any metas for these lines like
{PC etc for the background export to work properly. Using any metas
in these makes a kinda complex recursive loop in the input routines
which also checks for the mail export semaphores.
6. Some more minor changes to reduce mem requirements even further...
7. Using the new beta of the VAC linker, seems faster and better...
1.09.42
-------
1. Added Mario's (Hagar) local file u/l attach code and the fix for the
Event Editor. Documented in the Hagar ReadMe for the 1.09.49 release.
2. Tried a fix for the CBV dropping DTR not saving the user record. Let's
see if this works....
3. Tried out the problem reported by dougk, it's a prob i reproduced here
as well...seems to have something to do with the nodelist compiler not
putting the cost in the compiled nodelist while i was using the normal
cfg. I really had to add the cost everywhere before it actually showed
up in Adept. The short of this is, it's a nodelist compiler configuration
problem rather than a problem with Adept.
4. Neway, while searching for dougk's prob, found some more redundant debug
code...removed!
5. Releasing this now for 2 reasons.
a) This release is at par with the official "Hagar" 1.09.49 sans the
file caching, and,
b) need to check out the time cheat fix "fixed" for the CBV DTR drop
problem :)
1.09.41
-------
Skipped, as I already used the 41 in the last release archive :)
1.09.40.1
---------
1. GRRR...I broke the LMR update in the offline mail export while fixing
the mem leaks in the export. Actually it was a piece of debug code
which I forgot to re-compile. Fixed!
Built this exe from scratch...
2. Removed more of the original debugging code from Adept which had no
use from the user's view point other than increase the exe file size
and decrease performace by a wee bit (almost negligible).
3. Seems like my file area listing corruption fix worked :)
4. Still trying to track down the text export prob as I have not been
able to re-produce it and no one else seems to have had it besides
me :) Also trying to track down yet another elusive mem leak
someplace as I did get a slight drain with 24 hrs of runtime...
It's no longer as massive as it was though...:)
5. Releasing this build now 'cause of the LMR bug introduced earlier.
1.09.40
-------
Ah....at laaast :)
I think this build is now stable enough to let loose in general....
Please go through this entire file for details on what's new/fixed.
This archive also includes the fixed versions of FileEd 0.25beta as
well as BFSMake. Please keep in mind the fact that this is only an
update. You need to have a installed and working Adept System in
order to use this.
The current build includes Hagar's file attach code but not the file
caching system as yet. Also, this is a 24 node version to preserve
memory as most SysOps may not need more. In case required, a build
with a higer number of nodes can be provided.
Feel free to e-mail and problems/suggestions/bugs to:
nitin@poboxes.com
1.09.39.3
---------
1. Changed the file attach code a bit. It will no longer ask to attach a
file is the msg area is marked as private and the subject and the name
to fields are forced. I was having a prob with my CBV. I have a
seperate msg area for new user to leave some introduction about 'em
and...well...it puzzled 'em as to what file to attach :)
2. Rmoved some redundant code from the Adept REXX cache processing.
3. Trying another fix for the file display list displaying junk at some
places instead of blank. Think I found the prob...let's see...:)
4. I switched over to the new menus and discovered a mem leak there as
well. This would effect only the Clear All Gosubs command. Fixed!
5. Found a mem leak in the Adept REXX chat function ADEPTSYSCHATGETMSG.
The mem allocated for *every* line of chat was not being free'd.
Fixed!
6. Added a log line in a user selects text mail export to enable me to
pinpoint the fact first before i even attempt to fix it. Altho the
forced handler works fine for now :)
7. Found a mem leak in the write message fn in certain cases. Fixed!
8. Found a mem leak in the archive view fn in certain cases. Fixed!
1.09.39.2
---------
1. Added a forced exception handler for file transfers which are done by
GIOMT dll and nothing we can do about it for now. This would cover the
last trap I got in recent times...let's hope :)
2. Found and fixed quite a few mem leaks in the file listing, tagging fns.
Though small (pathname size which would be max 260 bytes (approx) per
leak, it would make a big difference if any BBS kept running for days
w/o being shutdown. Fixed!
3. Fixed the check in file listing and searching. It would not show files
marked as not d/lable but would show the files marked as not listable :)
Now, it will show files marked as not d/lable but not the ones marked
as not listable. I already fixed BFSMake (0.96) to reflect the perms,
and will get around to fixing FileEd to default flags to the file area
flags...soon :)
4. CServer was trying to free already free'd mem, which the VAC libs do not
like :) Fixed! (hopefully)
5. Removed delays from offline mail message writing. IMO it was not really
really required. Let's see...
6. The file list display on the SysOp side sometimes got mucked up on the
upper half depending on the displayed file descriptions length. Dunno
exactly what causes that...yet, but a tried fix anyway that would not
change anything else :) If it works, it stays, otherwise it goes away..
I think it was the description buffer not being cleared...which is what
I added for now...
7. Added a new command line option /WILDFILES. If /WILDFILES is specified
on the Adept command line, it will allow the Download command to process
wildcard characters which otherwise it prompts the user to
"be more specific" :)
8. GRRR...there remained a small possible mem leak in CServer. Fixed!
9. Reduced the number of file handles allocated by Adept for itself to be
smaller, based on the number of nodes it is compiled for. I have taken
and average of 8 files per node which should be quite ok...let's see...
10. Fixed BFSMake (now ver 0.96.1) to not join the first 2 description
lines of any file.
11. Just in case I forgot to mention it earlier, i really got peeved when
I deleted a user a/c and the user could still get in...so, now Adept
checks for the deleted flag before the locked out flag and will throw
out the user after displaying the file DeletedAccount.ASC (if it exists)
1.09.39.1
---------
1. Added Hagar's file attach code.
2. Tried 512 nodes :) came up, but took a whopping 2 mins (approx) to start
up. Got adventurous, max my system could take was 750 :( but at least we
know Adept is capable of it :)
3. Switching to a 24 Line version from now on, since most people do not use
more. This will further reduce the mem requirements of Adept. Anyone who
needs a higer number of lines, can req a version for more lines.
4. Discovered a bug in the FileEd 0.25 beta. Fixed! Documented in the FileEd
beta readme.
1.09.39
-------
I think whatever was done till now is now stable enough to call this
a usable release :) Just keep in mind the BBSStack Option documented
earlier in case of problems.
1. Duh...found another problem common to all mail exports under certain
conditions which would trap Adept. Fixed!
2. CServer had some severe mem problems. Took some doing, but I think it's
finally fixed...i tried it with the menu and the rexx functions, but only
on 2 lines so far...no probs and no more mem leaks.
The CServer has 2 command line switches in case of problems...
/EXCEPT to enable exception handling, and,
/DEBUG to create a debug log of what's happening in CServer.
As usual...CServer now runs at less than half the previous requirements
per thread :)
3. I left a forced exception handler for a particular routine in the offline
text export as it caused a trap i have not been able to figure out yet...
4. Thinks this build is stable enough for general use, so I am going to
document the stack size switches once again...
The three user definable stacks in this build of Adept are:
a) Mailer
b) BBS
c) OffLine Mail
The mailer and BBS stacks now default to 128kb and the Offline mail
import/export to 96kb. The original were 256k for Mailer, 320k for
the BBS and 128k for the offline mail. In case of any problem with
either of the above, try increasing this value using the following
command line options to restore the previous defaults.
/MAILERSTACK 262144
/BBSSTACK 327680
/OFFLINESTACK 131072
In short, run Adept as:
Adept.Exe /MAILERSTACK 262144 /BBSSTACK 327680 /OFFLINESTACK 131072
besides any other option you use normally.
1.09.38.9
---------
1. Added my DOS door runner into Adept itself as menu types 820 and 821.
Why 2? well...820 keeps the session in the background and 821 keeps
it in the foreground. I like foreground and Hagar likes background :)
Syntax is the same as for type 801 etc...eg:
1,820, Some Vague Door, VagueDor.Bat {LN,0
2,821, Yet Another Vague Door, YAVD.Bat {LN,0
I can think of some improvements on this, but after I get some
feedback as to what other sysops feel/think about this option.
Currently the door runner assumes that VX00.SYS and the OS/2 ANSI.SYS
is in your Adept dir (if req) AND that your Adept dir is \Adept.
I will fix this later...if req :)
2. Minor irritant left in the auto-menu switch. Since the default was to
start off with old menus, during the auto-check for the first login,
a 0 byte .bin file was created and left in the menu dir. It is now
removed.
3. Found that the ChatServer required more Stack for threads on account
of the printf fn in the WC libs using more stack. Increased it to 80k
from 64k for now. Seems to work fine now. The VAC printf had no such
hangups.
4. Clubbed a few mem alloc calls in the new menu driver to a single one
and divided internally.
5. The vastly reduced mem requirements now seem to work quite fine at
least for me. Both, VAC and WC builds of the same seem to be equally
good. Maybe I finally did get it right eh? :) Well...let's see...
1.09.38.8
---------
1. GRRR...introduced a mem leak in the old driver. Fixed!
Thought of trying the VAC _heapmin() again...seems to have some
kinda prob. Removed from menu drivers for now, called only when
a user logs off.
2. Currently using 128k for mailer and BBS. In case the BBS traps on
startup or something, please use the command line option:
/BBSStack 262144
This will set the BBSStack up for 256k. For the mailer its the same
syntax, the option is /MailerStack.
3. Pulled out the BBSWatch thread for now as I could not really see a
use for it....:)
6. Tried a reduced mem requirement for qwk/bwave import/export...failed :)
I put it down to less than 1/4 by mistake (typo), neway it seems to work
fine in 96k as compared to 128k earlier.
7. Seem to have sorted out most of the mem alloc related probs, but i have
a feeling there is still the odd one out there :)
8. The exception handling in BWave/QWK will now honour the command line
/EXCEPT or /NOEXCEPT. I have done the same at all places (unless I
missed something) where it was forced. Let's see what happens :)
9. The FTP Server was setting the exception handler but not freeing it. Fixed!
10. Added exception handling to get_reply_idx fn. Got a cpl of traps there
some time ago. Don't recall what exactly happened, but plugged it in
for now...
11. Please note the fact that exception handlers (enabled by the command
line switch /EXCEPT) are meant only for testing purposes or for SysOps
who want the BBS to keep up and running, regardless!
1.09.38.7
---------
1. Discovered a prob in the qwk import, would cause a trap, must have
been my own creation in the earlier mods. Fixed!
2. Found a potential prob in bwave import, in case of low mem, it would
freeze. This is similar to what was happening in rexx. Fixed!
3. Found an anomaly in the bwave/qwk import which could have, in cases,
been the source of yet another mem leak. Fixed!
4. Changed the multiple allocations in the old menu driver to a single
call and divided it internally. Much faster and smaller now.
5. Reduced the mem requirement of the mailer from the original
256k to 128k and the BBS req from 320k to 128k as well. This
is per thread of the Mailer and the BBS. A substantial gain
specially for multi-line systems. Seems to be working well
enough so far...:)
6. This is for those who might be interested. There seems to be a
substantial differnce in the Watcom and VAC mem alloc/de-alloc
functions. The one that was causing most grief to VAC was the
alloca (allocate on stack). Watcom handled this as documented and
quite well, but VAC invariably trapped at times. Also, Watcom does
not seem to mind a invalid pointer in the free call, it most prolly
checks, but VAC on the other hand, traps!
7. Found a prob in the Old Menu driver. A misplaced break would cause
the goto menu command never to be executed. Fixed!
8. Found and fixed another mem leak in ANSI display routines in case of
a carrier drop during display.
9. Found a bug in the ChatServer. Was assuming the value of a free'd
pointer at one place. Dunno what it might have screwed up. Fixed!
10. Possible mem leak under some conditions in the new menu driver. Fixed!
11. Found quite a few mem leaks in the Chat Server. Fixed!
12. Removed TSYNCH (FTS-0001) code left in the mailer to make it smaller.
FTS never did work ok and no one actually uses that anymore, so...:)
13. According to Hagar, the BBS Watch thread works fine, so, enabled that
as well, though we already have the Status timer doing a similar job.
This might make it a doubly robust kick out check :) let's see...
1.09.38.6
---------
1. Duh...changing a few of those stack vars did muck up the status
display at times...no prob, just incorrect display. Fixed..again :)
2. Now that I have a better idea on the mem usage and most mem leaks
have been fixed...(hopefully :)) reverting back to the original
form of the stack usage as the existing experimental usage would
in cases trap or freeze. Hopefully multi-line systems will no
longer have any problems now with the VAC build.
3. Although the BBS was setting the File can't be listed or d/l when
any file was u/l on the BBS based on the file area flags, it was
not setting the bytes/free files flags in the FileBase. Fixed!
3. Forgot to metion earlier...the menu processing is done entirely by
the new engine (actually, it was never changed from the old one).
Only the basic menu engine call differs for the Original/New styles.
4. Added 2 new MenuItems for the Old Menu Style, which I have not
currently included in the new menu engine as it might not be really
required by most sysops. If there are a sufficient number of SysOps
wanting that in the new menu style engine, I will try to add it to
that as well. The new menu funtions added to the Old Style menus are:
Menu Type 8 - Switch to a given Message area and then Gosub a Menu
Menu Type 9 - Switch to a given File area and then Gosub a Menu
For example:
A,8,[1] Switch to Message Area,Message.Menu;Message Area Name,0
B,9,[2] Switch to File Area ,Files.Menu;File Area Name,0
Keep in mind that the Area Name should NOT be in quotes even if it
has spaces in between. Oh well...I was using 2 line rexx scripts
just to do this...the other option was to use the AutoMenu as
suggested by Hagar, but I guess I am too lazy :) so I plugged in a
single command for it.
5. There seems to be a slight problem in the qwk import regarding areas
marked as e-mail/usenet, have to talk to some people who know more
about this before I can fix it (if it needs to be fixed at all :))
6. Bill-E pointed out the fact that AdeptRecv() was not working as
documented. It is not and is not fixed as yet. The function was
never completed AFAIK, but if there is a great need for it, it can
be "Klugded" in. For the time being tho, altho the syntax remains
the same, *only* the upload file path parameter is used by
AdeptRecv() function.
7. What the heck...added the menu type options 8 and 9 to the new menu
driver as well, works...for now at least :)
Hopefully this build might be more usable and stable than 1.09.38.5 as
most of the experimentation mem related code is now gone. Those who
missed the .5 update notes, please read about the changes to BFSMake.
1.09.38.5
---------
1. Possible mem leak spawning apps from Adept in some cases...fixed!
2. Checks added for mem alloc in places where there wasn't as i go along
for now...never know, it might turn up something :) Anyway, no harm
in the checking which is generally not required for small amounts...
3. Discovered an in-consistency in BFSMake and the BBS. Some details are
documented in the BFSMake 0.96 now. Please get it.
4. Lotsa itty-bitty changes...too many to document...no bugs/fixes, just
trying to make the code more consistent...But you never know, it might
have an impact on some parts of the BBS...good or bad is yet to be seen :)
5. The BBS was not checking for file d/l permission from the file area
flags. It checked from the FileSystem and BFSMake never did set
those flags. The result was, anyone who re-built a FileBase using any
earlier version of BFSMake can have files d/l from an area which has
no d/l flag set. This is now fixed!
Part of the fix includes having to re-build the filebase using the new
BFSMake...so that file permissions can be properly set based on the
area flags when building a filebase.
I would appreciate some feedback on these...BFSMake as well as this
BBS beta update.
1.09.38.4
---------
This is at the same level as the Official 1.09.37 as far as features
are concerned. Please consult the official "Hagar" build readme for
all theeee older details. This readme will document changes since the
official Watcom 1.09.37 build from Hagar.
It also includes the new quickscan fix from Hagar for the Official
1.09.40 build. Some of the fixes documented below are already
included in the official AdeptX build, and some that are not, are
already documented in the official readme anyway...:)
1. The Menu System is now auto-detected, so, it does not matter what
menu style you use...the old classic one or the new style. The default
startup is with the new style menus but will switch to old style, if
required, on the first login and the status will be updated to reflect
the change. The earlier, officially documented, /OLDMENU command line
switch can still be used to start up with the old style menus as the
default if required.
2. Re-Fixed the LMR bug to be more robust
3. QWK, TEXT, FIDO export would leave locked files in case of a carrier
drop...fixed (BWave had been fixed earlier, most prolly we missed
something someplace...according to Hagar, all had been fixed earlier)
4. The problem with Meta chars in ANS/ASC (Adept codes) files trapping
sometimes during a fast enough display has been fixed (hopefully :))
This problem would be apparent when displaying large/complex ANS/ASC
files with Metas in between...more specifically if they also have the
{PC meta as well. I have tried to fix this, but cannot be sure as it
is not something I am able to reproduce everytime, but as things get
fixed (or broken :)) we will find out the cause for this..hopefully..
5. Couple of memory leaks (would happen at times) in AdeptREXX have
been fixed.
6. A Memory leak in BWave Import have also been fixed.
7. BlueWave/QWK imports would not sometimes close files in case of
some problems/errors, leaving files open and therefore...locked.
8. Mem Leaks in email and nntp posting and a possible locked file
prob is now corrected (AFAIK :))
9. Adept would display a message of "not implemented" in case some user
dropped carrier at any file/msg list prompt. Minor irritant, but fixed.
10. A lot of mem alloc code has been changed...
11. Stack requirements reduced by a substantial amount (I hope :)), lots
of changes in the code to make that possible...seems to work for now.
This might also help us catch the vague traps which left no address
in the POPUP.LOG (OS/2 System Error Log, if enabled) or even the
Trap.Log (created by Adept on exceptions when run with the /EXCEPT
command line switch), generally 'cause of "not enough stack to handle
the exception" (as per the OS/2 official help/docs :))
On the negative part, this also means that the basic mem requirement
for Adept is up slightly, but, the good part is that all those places
would also function slightly faster...let's see...
12. Displaying an ANS file with no literals would leave the file open.
Dunno what it was, but...fixed :)
13. The Events SubSystem would either freeze or trap if a valid events
file was not found...fixed!
14. On a seek failure in the nodelist, it would leave the file open. Fixed!
15. The File Search should be easier on the CPU now...added a delay between
file area switching.
16. Most of the original debugging code has been removed from this
update build.
17. Some places in Adept had exception handling turned on regardless of
the command line /EXCEPT or /NOEXCEPT specification. Some have been
corrected and others will be to help find out the exact problems
associated with those parts.
18. Quick Scan fix from Hagar (refer to the official readme for more
information)
That should do it for now....hopefully the current build should work
fine for most people...<g>.
Note: Had a request from a guy wanting to know what the numerical values
were for each of the msg and file area flags in the message editor, these
values are simply added together to form the FLAGS number in message_areas
and File_areas file, so i have tacked on the hexadecimal values to the very
end of this readme.txt. You'll have to convert em to decimal as the FLAGS
entry u see in the file is a decimal number, rememeber its the sum of all the
flags checked in the editor.
/*----------------------------------------------------------*/
/* Message area flags (areaflags) */
/*----------------------------------------------------------*/
#define M_NOORIG 0x00000001 /* No Origin or tear */
#define M_ANSI 0x00000002 /* Allow Ansi Messages */
#define M_PRIVATE 0x00000004 /* Force Msgs Public */
#define M_PUBLIC 0x00000008 /* Force Msgs Private */
#define M_ECHO 0x00000010 /* Is a Echomail */
#define M_NET 0x00000020 /* It a Netmail area */
#define M_MCI 0x00000040 /* Allows MCI Commands */
#define M_ANON 0x00000080 /* All Msgs Annon. */
#define M_REAL 0x00000100 /* Requires Real Name */
#define M_ASSOC 0x00000200 /* Msg area allows assoc */
#define M_FORCE 0x00000400 /* Always force */
#define M_COMPRESS 0x00000800 /* Allow LZSS comp here */
#define M_GROUP 0x00001000 /* groupmail */
#define M_USENET 0x00002000 /* usenet conference */
#define M_QWK 0x00004000 /* qwk conference */
#define M_SQUISH 0x00008000 /* Squish MSG Base */
#define M_EMAIL 0x00010000 /* Internet E-Mail */
#define M_FIDO 0x00100000 /* Star . msg Format */
/*-----------------------------------------------------------*/
/* File area flags (areaflags) */
/*-----------------------------------------------------------*/
#define F_FREE 0x00000001 /* */
#define F_NOUP 0x00000002 /* */
#define F_NODOWN 0x00000004 /* */
#define F_LEECH 0x00000008 /* */
#define F_DOWNBACK 0x00000010 /* */
#define F_UPBACK 0x00000020 /* */
#define F_CHARGE 0x00000040 /* */
#define F_CANTLIST 0x00000080 // Uploads not listable
#define F_CANTDL 0x00000100 // Uploads not downloadable
#define F_NOFBBS 0x00000200 /* no FILES.BBS at all */
#define F_COMMONONLY 0x00000400 /* use only common FILES.BBS */
#define F_UPBLIND 0x00000800 // UL get desc. afterwards
#define F_COPYFILE 0x00001000 // copy download from CD to HD
#define F_NODIR 0x00002000 // copy download from CD to HD
#define F_NOCOMMON 0x00004000 // copy download from CD to HD
#define F_NOSEARCH 0x00008000 // copy download from CD to HD