PDA

View Full Version : Implicit Length Opcode Changes 5/26



ksmith
05-26-2004, 12:33 PM
// Generated Wed May 26 14:25:23 2004 from eqgame.exe (2293759 bytes)
// Function found at 0x509fb

$implicitlen_table = array(
0x0021 => 0x08,
0x0027 => 0x12,
0x003e => 0x0c,
0x00bf => 0x0c,
0x00e2 => 0x17,
0x00f3 => 0x04,
0x0100 => 0x88,
0x0101 => 0x1f,
0x012c => 0x09,
0x012f => 0x08,
0x0140 => 0x04,
0x022d => 0x03,
0x022e => 0x03,
0x0243 => 0x06,
0x0244 => 0x06,
);

How I find the implen table (http://eqitems.13th-floor.org/misc/implen.pl)

The reason it generates a php array is because I use this script for the itemcollector.

CybMax
05-26-2004, 01:06 PM
And this we edit in what file?

ksmith
05-26-2004, 01:54 PM
People using the beta, I'd like you to try this updated zoneopcodes.xml file.

monster69
05-26-2004, 02:00 PM
looks good on spawns, I can see them and other players...but

It's not decoding me so (with defaults set) I look like a lvl 1 in the zone


Thanks for your work on this ksmith!

Monster

onied
05-26-2004, 02:02 PM
seems to be working nicely

onied
05-26-2004, 02:05 PM
Warning: 16:04 - OP_PlayerProfile (0x6b) (dataLen: 10400) doesn't match: sizeof(charProfileStruct):10392
Warning: 16:04 - OP_ZoneEntry (0x23a) (dataLen: 291) doesn't match: sizeof(ServerZoneEntryStruct):267


newest errors i get

but it works non-the less

ksmith
05-26-2004, 02:05 PM
I don't have an active everquest account anymore, so I need people to copy and past any error messages they're getting, most importantly size mismatch errors.

bonkersbobcat
05-26-2004, 02:13 PM
It's an improvement. That seems to get GPS and skittles working again. I didn't pick up the zone change and have to manually load maps. (Will probably have to go edit the zone struct)

Warning: OP_PlayerProfile (0x6b) (dataLen: 10400) doesn't match: sizeof(charProfileStruct):10392
Warning: OP_ZoneEntry (0x23a) (dataLen: 291) doesn't match: sizeof(ServerZoneEntryStruct):267

Looks like a bunch of the op code numbers got shifted. How many new codes did you find?

BB

monster69
05-26-2004, 02:20 PM
It's not always catching the map either. Here is some data:


arning: error on zone-client: size > left (size=244, left=24, opcode=0x00e2)
Warning: error on zone-client: size > left (size=244, left=34, opcode=0x00e2)
Warning: error on zone-client: size > left (size=248, left=57, opcode=0x00e2)
Warning: error on zone-client: size > left (size=244, left=24, opcode=0x00e2)
Zone: LogoutCode: Client logged out of server
Info: EQPacket: SEQClosing detected, awaiting next zone session, pcap filter: EQ Client 192.168.254.36
Zone: EntryCode: Client
Info: EQPacket: SEQStart detected, pcap filter: EQ Client 192.168.254.36, Client port 1855
Warning: OP_PlayerProfile (0x6b) (dataLen: 10400) doesn't match: sizeof(charProfileStruct):10392
Warning: OP_ZoneEntry (0x23a) (dataLen: 291) doesn't match: sizeof(ServerZoneEntryStruct):267
Time: Wed Mar 01,3178 - 01:14 pm
Warning: OP_NewZone (0xec) (dataLen: 0) doesn't match:
Player: Exp: Set: 0 total, with 0 (85/330) into level with 900 left, where 1/330 = 2
Info: Your player's id is 251


Monster

ksmith
05-26-2004, 02:21 PM
I just moved opcodes below f0 down by one, opcodes below 200 by two, and opcodes > 200 by 4.

ksmith
05-26-2004, 02:27 PM
monster: try downloading the zoneopcodes.xml file again, I just updated it and it should get rid of the size > left messages.

monster69
05-26-2004, 02:39 PM
Okay, its catching the map evertime and it did get rid of the size > left msgs.

Still not decoding the player info:

arning: OP_PlayerProfile (0x6b) (dataLen: 10400) doesn't match: sizeof(charProfileStruct):10392
Warning: OP_ZoneEntry (0x23a) (dataLen: 291) doesn't match: sizeof(ServerZoneEntryStruct):267


Monster

ksmith
05-26-2004, 02:46 PM
We're working on the charProfileStruct in irc. It will require changes to everquest.h and a recompile.

keendevil
05-26-2004, 02:49 PM
what do you do with the zoneopcodes.xml file?

Where does it go?

:-/

ksmith
05-26-2004, 03:01 PM
Fix for charProfileStruct attatched.

monster69
05-26-2004, 03:35 PM
That patch broke it, it seems. No decode of anything shows on screen. Data:

Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: OP_GroupDisband (0x101) (dataLen: 31) doesn't match: sizeof(groupDisbandStruct):128
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: error on zone-client: size > left (size=173, left=26, opcode=0x003e)
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: error on zone-client: size > left (size=108, left=74, opcode=0x0021)
Warning: error on zone-client: size > left (size=107, left=38, opcode=0x003e)
Warning: OP_GroupDisband (0x101) (dataLen: 31) doesn't match: sizeof(groupDisbandStruct):128
Warning: error on zone-client: size > left (size=71, left=34, opcode=0x012f)
Warning: error on zone-client: size > left (size=66, left=26, opcode=0x003e)
Warning: error on zone-client: size > left (size=199, left=21, opcode=0x003e)
Warning: error on zone-client: size > left (size=233, left=26, opcode=0x003e)
Warning: OP_GroupDisband (0x101) (dataLen: 65) doesn't match: sizeof(groupDisbandStruct):128
Warning: OP_GroupDisband (0x101) (dataLen: 31) doesn't match: sizeof(groupDisbandStruct):128
Warning: error on zone-client: size > left (size=71, left=34, opcode=0x012f)
Warning: error on zone-client: size > left (size=74, left=34, opcode=0x0243)
Warning: OP_GroupDisband (0x101) (dataLen: 35) doesn't match: sizeof(groupDisbandStruct):128
Segmentation fault

cornholio
05-26-2004, 03:40 PM
Please forgive my ignorance, but could I please ask how to impliment these changes? I realize you guys are very busy, so if you do not have time that would be ok as well.

Thank you in advance.

reaver
05-26-2004, 03:46 PM
You can apply this diff patch, or edit your everquest.h file manually and make the following 2 changes:

line 447 in everquest.h : change unknown260[7] to unknown260[31]
line 576 of everquest.h change: unknown4380[5348] to unknown4380[5356]

Recompile, then put the new zoneopcodes.xml file in your /usr/local/share/showeq directory

Confirmed working.

-reaver

cornholio
05-26-2004, 03:51 PM
Thank you very much Reaver.

recompiling now, your help is appreciated.

NeedHelp
05-26-2004, 03:56 PM
Someone mind posting the new everquest.h file.

I'm a bit new to this and while I can read a few of the changes you made I do not understand the last part of the file.

Thanks,

NH

onied
05-26-2004, 04:03 PM
since it was asked for

uRit1u2CBBA=
05-26-2004, 04:08 PM
Originally posted by keendevil
what do you do with the zoneopcodes.xml file?

Where does it go?

:-/

mine was in /usr/local/share/showeq

If it isn't there, do a find:

find / -name zone*.xml -print

and it will tell you where it exists.

monster69
05-26-2004, 04:53 PM
Ignore my "It's still broke" post above.

I am such a bonehead. I applied the patch and recompiled but forgot to recopy the zoneopcodes.xml file after "make install" overwrote it.

monster

NeedHelp
05-26-2004, 05:11 PM
I replaced the everquest.h file a few times. Download the CVS code but I'm now coming up with an error that I'm hoping someone can explain. The error happens when I go to make the program.

[root@gizmo showeq]# make
make all-recursive
make[1]: Entering directory `/seq/showeq'
Making all in maps
make[2]: Entering directory `/seq/showeq/maps'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/seq/showeq/maps'
Making all in src
make[2]: Entering directory `/seq/showeq/src'
/usr/bin/moc -nw packet.h -o m_packet.cpp
/usr/bin/moc -nw packetstream.h -o m_packetstream.cpp
/usr/bin/moc -nw spawnshell.h -o m_spawnshell.cpp
/usr/bin/moc -nw spawnlist.h -o m_spawnlist.cpp
/usr/bin/moc -nw spellshell.h -o m_spellshell.cpp
/usr/bin/moc -nw spelllist.h -o m_spelllist.cpp
/usr/bin/moc -nw editor.h -o m_editor.cpp
/usr/bin/moc -nw interface.h -o m_interface.cpp
/usr/bin/moc -nw compass.h -o m_compass.cpp
/usr/bin/moc -nw map.h -o m_map.cpp
/usr/bin/moc -nw experiencelog.h -o m_experiencelog.cpp
/usr/bin/moc -nw combatlog.h -o m_combatlog.cpp
/usr/bin/moc -nw msgdlg.h -o m_msgdlg.cpp
/usr/bin/moc -nw player.h -o m_player.cpp
/usr/bin/moc -nw decode.h -o m_decode.cpp
/usr/bin/moc -nw skilllist.h -o m_skilllist.cpp
/usr/bin/moc -nw statlist.h -o m_statlist.cpp
/usr/bin/moc -nw filtermgr.h -o m_filtermgr.cpp
/usr/bin/moc -nw category.h -o m_category.cpp
/usr/bin/moc -nw compassframe.h -o m_compassframe.cpp
/usr/bin/moc -nw group.h -o m_group.cpp
/usr/bin/moc -nw guild.h -o m_guild.cpp
/usr/bin/moc -nw netdiag.h -o m_netdiag.cpp
/usr/bin/moc -nw logger.h -o m_logger.cpp
/usr/bin/moc -nw xmlpreferences.h -o m_xmlpreferences.cpp
/usr/bin/moc -nw seqwindow.h -o m_seqwindow.cpp
/usr/bin/moc -nw seqlistview.h -o m_seqlistview.cpp
/usr/bin/moc -nw zonemgr.h -o m_zonemgr.cpp
/usr/bin/moc -nw spawnmonitor.h -o m_spawnmonitor.cpp
/usr/bin/moc -nw spawnpointlist.h -o m_spawnpointlist.cpp
/usr/bin/moc -nw spawnlistcommon.h -o m_spawnlistcommon.cpp
/usr/bin/moc -nw spawnlist2.h -o m_spawnlist2.cpp
/usr/bin/moc -nw itemdb.h -o m_itemdb.cpp
/usr/bin/moc -nw datetimemgr.h -o m_datetimemgr.cpp
/usr/bin/moc -nw spawnlog.h -o m_spawnlog.cpp
/usr/bin/moc -nw packetlog.h -o m_packetlog.cpp
make all-am
make[3]: Entering directory `/seq/showeq/src'
if g++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/lib/qt-3.1/include -I/usr/X11R6/inclu de -DMAPDIR=\"/usr/local/share/showeq\" -DLOGDIR=\"/usr/local/share/showeq\" -I /usr/include/pcap -D_REENTRANT -O2 -Wall -g -ggdb -DDEBUG -finline-functions -DQT_THREAD_SUPPORT=1 -DDISPLAY_ICONS=false -DICON_DIR=\"/eq-icons/\" -MT mai n.o -MD -MP -MF ".deps/main.Tpo" \
-c -o main.o `test -f 'main.cpp' || echo './'`main.cpp; \
then mv -f ".deps/main.Tpo" ".deps/main.Po"; \
else rm -f ".deps/main.Tpo"; exit 1; \
fi
In file included from interface.h:34,
from main.cpp:34:
spawnshell.h:114: error: syntax error before `*' token
spawnshell.h:115: error: syntax error before `*' token
In file included from main.cpp:34:
interface.h:130: error: syntax error before `*' token
interface.h:131: error: syntax error before `*' token
interface.h:134: error: syntax error before `*' token
interface.h:135: error: syntax error before `*' token
make[3]: *** [main.o] Error 1
make[3]: Leaving directory `/seq/showeq/src'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/seq/showeq/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/seq/showeq'
make: *** [all] Error 2
[root@gizmo showeq]#


If you could give any help that would be great. Also if I need to download anything from the CVS could you please give me the command. I have tried the commands listed and for some reason unless I use the command from the CVS page I keep getting stuck when it ask for the password to the CVS system.

NH

ieatacid
05-26-2004, 05:20 PM
Thanks Ksmith, worked like a charm.

Hobo
05-26-2004, 05:21 PM
At the risk of exposing my REALLY bad memory, I have a question to ask. Once I get the new xml file can I just move it to usr/local/share/showeq directory and then just use the gunzip command? Or do I need to add gunzip -XXX (where X = a trigger or flag of some sort.) I didn some searching and wasn't entirely sure since I've seen some tarbals that require flags to both tar and gunzip a file.

Hope this makes sense, I'm just barely above the idiot level when it comes to Linux.


Thanks!

KaL
05-26-2004, 05:25 PM
Rock on. Thanks guys.

eViL_nEcRo_01
05-26-2004, 06:10 PM
works like a charm )

thanks guys!

Hobo
05-26-2004, 09:12 PM
Works great. Thanks!



Bo

Choofer
05-26-2004, 09:58 PM
anyone hapen to have a tarball of 5.0.0.8 i simply cannot download showeq off the server i keep getting timed out

Choof

acher1
05-26-2004, 10:39 PM
Choofer, if you're using the server name cvs.seq.sourceforge.net, change it to just cvs.sourceforge.net and it will work better.

Hope that helps.

Choofer
05-26-2004, 10:41 PM
thanks mate that was the mistake i have only just found that one out but thanks for replying

Choof

wyldkrd
05-27-2004, 02:37 AM
still having this error message after applying the everquest.h and xml fixes... running SuSe

Warning: OP_TimeOfDay (0x27) (dataLen: 18) doesn't match: sizeof(timeOfDayStruct):8
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: OP_TimeOfDay (0x27) (dataLen: 18) doesn't match: sizeof(timeOfDayStruct):8
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:
Warning: OP_TimeOfDay (0x27) (dataLen: 18) doesn't match: sizeof(timeOfDayStruct):8
Warning: OP_TimeOfDay (0x27) (dataLen: 30) doesn't match:

perlmonkey
05-27-2004, 05:40 AM
Originally posted by keendevil
what do you do with the zoneopcodes.xml file?
If you are unclear on how to apply these fixes, you are probably not the intended audience, and should wait for a new version that supports this patch to be available. This is a development forum, and items being posted here are for the purpose of CREATING a fix, not publishing it.... yet.

ThanosOfTitan
05-27-2004, 07:07 AM
Originally posted by acher1
Choofer, if you're using the server name cvs.seq.sourceforge.net, change it to just cvs.sourceforge.net and it will work better.

Hope that helps.
Acher1, do you mean this:

export CVSROOT=:pserver:[email protected]:/cvsroot/seq
cvs login
cvs co -r pre_5_0_beta showeq

Have to try this when I get home was getting the same as Choofer last night

onied
05-27-2004, 07:10 AM
NEWBIES GUIDE can't go wrong.

1: Download the everquest.h tar and untar it into your /root/showeq/src/ directory.

2: Download the zoneopcodes.xml tar and un-tar that into your
/root/showeq/conf/ directory. *this make it so you can't posssably forget to put it where it belongs later. make install will do it for you.*

3:
cd /showeq

4:
make -f Makefile.dist
*this may crap out the first time just do it again.*

5:
./configure

6:
make

7:
make install

help_plz
05-27-2004, 08:28 AM
I am receiving the same problem as the previous user. When I do the exact steps posted above and attempt to make, I receive the following:


[root@localhost showeq]# make
make all-recursive
make[1]: Entering directory `/showeq'
Making all in maps
make[2]: Entering directory `/showeq/maps'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/showeq/maps'
Making all in src
make[2]: Entering directory `/showeq/src'
make all-am
make[3]: Entering directory `/showeq/src'
source='main.cpp' object='main.o' libtool=no \
depfile='.deps/main.Po' tmpdepfile='.deps/main.TPo' \
depmode=gcc3 /bin/sh ../depcomp \
g++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/lib/qt3-gcc3.2/include -I/usr/X11R6/include -DMAPDIR=\"/usr/local/share/showeq\" -DLOGDIR=\"/usr/local/share/showeq\" -I/usr/include/pcap -D_REENTRANT -O2 -Wall -g -ggdb -DDEBUG -finline-functions -DQT_THREAD_SUPPORT=1 -DDISPLAY_ICONS=false -DICON_DIR=\"/eq-icons/\" -c -o
main.o `test -f 'main.cpp' || echo './'`main.cpp
In file included from interface.h:34,
from main.cpp:34:
spawnshell.h:114: parse error before `*' token
spawnshell.h:115: parse error before `*' token
In file included from main.cpp:34:
interface.h:130: parse error before `*' token
interface.h:131: parse error before `*' token
interface.h:134: parse error before `*' token
interface.h:135: parse error before `*' token
make[3]: *** [main.o] Error 1
make[3]: Leaving directory `/showeq/src'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/showeq/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/showeq'
make: *** [all] Error 2

CybMax
05-27-2004, 08:35 AM
Unless you just made a new CVS download of the source, remember to make distclean before doing

make -f Makefile.dist
./configure
make

Just to make sure you did not just edit the file, and tried to make without actually cleaning a previous compile..

help_plz
05-27-2004, 08:41 AM
My apologies, I figured out my problem. My SEQ was version 4.3.2.1 I simply downloaded the latest beta release using the instructions posted above, then followed the steps again. It is currently finishing the compiling now.

Thanks.

CybMax
05-27-2004, 08:44 AM
Should still compile without the error above, but its a different method of implementing the opcodes to actually get it to work tho...

But doing a clean CVS of the latest beta (wich is the one actually being supported atm) is always good :)

help_plz
05-27-2004, 08:53 AM
hmm,

Okay, here are my steps.

1 ) Clean Cvs Install
2 ) Placed the everquest.h file in the showeq/src
3 ) Placed the zoneopcodes in the showeq/conf
4 ) make - f Makefile.dist
5 ) ./configure
6 ) make && make install

I received the following error message again. I must be missing something?


g++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/lib/qt3-gcc3.2/include -I/usr/X11R6/include -DPKGDATADIR=\"/usr/local/share/showeq/\" -I/usr/include/pcap -D_REENTRANT -O2 -Wall -g -ggdb -DDEBUG -finline-functions -DQT_THREAD_SUPPORT=1 -DDISPLAY_ICONS=false -DICON_DIR=\"/eq-icons/\" -c -o filtermgr.o `test -f 'filtermgr.cpp' || echo './'`filtermgr.cpp
source='m_filtermgr.cpp' object='m_filtermgr.o' libtool=no \
depfile='.deps/m_filtermgr.Po' tmpdepfile='.deps/m_filtermgr.TPo' \
depmode=gcc3 /bin/sh ../depcomp \
g++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/lib/qt3-gcc3.2/include -I/usr/X11R6/include -DPKGDATADIR=\"/usr/local/share/showeq/\" -I/usr/include/pcap -D_REENTRANT -O2 -Wall -g -ggdb -DDEBUG -finline-functions -DQT_THREAD_SUPPORT=1 -DDISPLAY_ICONS=false -DICON_DIR=\"/eq-icons/\" -c -o m_filtermgr.o `test -f 'm_filtermgr.cpp' || echo './'`m_filtermgr.cpp
source='mapcore.cpp' object='mapcore.o' libtool=no \
depfile='.deps/mapcore.Po' tmpdepfile='.deps/mapcore.TPo' \
depmode=gcc3 /bin/sh ../depcomp \
g++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/lib/qt3-gcc3.2/include -I/usr/X11R6/include -DPKGDATADIR=\"/usr/local/share/showeq/\" -I/usr/include/pcap -D_REENTRANT -O2 -Wall -g -ggdb -DDEBUG -finline-functions -DQT_THREAD_SUPPORT=1 -DDISPLAY_ICONS=false -DICON_DIR=\"/eq-icons/\" -c -o mapcore.o `test -f 'mapcore.cpp' || echo './'`mapcore.cpp
mapcore.cpp: In member function `void MapData::loadSOEMap(const QString&,
bool)':
mapcore.cpp:968: no matching function for call to `QString::remove(QRegExp&)'
/usr/lib/qt3-gcc3.2/include/qstring.h:500: candidates are: QString&
QString::remove(unsigned int, unsigned int)
/usr/lib/qt3-gcc3.2/include/qstring.h:607: `QString& QString::replace(const
QString&, const QString&)' is private
mapcore.cpp:1094: within this context
/usr/lib/qt3-gcc3.2/include/qstring.h: In member function `void
MapData::saveSOEMap(const QString&) const':
/usr/lib/qt3-gcc3.2/include/qstring.h:607: `QString& QString::replace(const
QString&, const QString&)' is private
mapcore.cpp:1361: within this context
make[3]: *** [mapcore.o] Error 1
make[3]: Leaving directory `/showeq/src'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/showeq/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/showeq'
make: *** [all] Error 2
[root@localhost showeq]#


Thank You.

BlueAdept
05-27-2004, 09:26 AM
For those using the non-beta, I updated the opcodes.h file (along with decode.cpp and everquest.h) It also includes the fix so that light sources are displayed again (yes Im doing another shawl :p)

I noticed that ksmith had duplicated one opcode. It was SendAAStats. I renamed one SendAAStats1 (which means that the AAStats might be wrong).

Update: I tested it at home and it seems to work good.

ksmith
05-27-2004, 09:36 AM
Originally posted by BlueAdept
I noticed that ksmith had duplicated one opcode. It was SendAAStats. I renamed one SendAAStats1 (which means that the AAStats might be wrong).

I just ran the zoneopcodes.xml through a perl script to re-number the opcodes based on the changes to the implicitlen table.

BlueAdept
05-27-2004, 09:53 AM
Originally posted by ksmith
I just ran the zoneopcodes.xml through a perl script to re-number the opcodes based on the changes to the implicitlen table.

Not saying anything is wrong with the xml file and really appreciate the help.

It just gave me warnings when I duplicated the settings in the xml to the opcodes.h

This is the duplication that I saw:

<opcode id="002b" name="OP_SendAAStats" updated="05/26/04">
<comment></comment>
</opcode>

and

<opcode id="01c8" name="OP_SendAAStats" updated="05/26/04">
<comment></comment>
</opcode>

It gave me a warning when I had the duplicate in the opcodes.h file.

reaver
05-27-2004, 04:45 PM
Here's a new error I saw pop up also related to struct:

OP_BazaarSearch (0x1e7) (dateLen: 20) doesn't match: modules of sizeof(bazaaarSearchResponsesStruct):88

You get this when you go in or out of Trader mode. I am not getting it when using the /baz command, only /trader

-reaver

Radical
05-27-2004, 07:09 PM
Anyone able to through the patch updates into the CVS?

bonkersbobcat
05-27-2004, 07:16 PM
Ksmith's patches have been applied to CVS. Thanks ksmith.

It may take a bit for CVS to make them available. If they are not there, check back later.

BB

SeqTester
05-27-2004, 10:19 PM
/bow... nice work guys. :)

jarula
05-27-2004, 10:25 PM
Thanks alot guys. you have done a superb job again!

monklett
05-27-2004, 11:21 PM
Originally posted by BlueAdept
For those using the non-beta, I updated the opcodes.h file (along with decode.cpp and everquest.h) It also includes the fix so that light sources are displayed again (yes Im doing another shawl :p)

I noticed that ksmith had duplicated one opcode. It was SendAAStats. I renamed one SendAAStats1 (which means that the AAStats might be wrong).

Update: I tested it at home and it seems to work good.

Worked fine for me too; running 4.3.21 on RH 7.3.

keendevil
05-28-2004, 10:12 AM
Originally posted by perlmonkey

quote:Originally posted by keendevil
what do you do with the zoneopcodes.xml file?


If you are unclear on how to apply these fixes, you are probably not the intended audience, and should wait for a new version that supports this patch to be available. This is a development forum, and items being posted here are for the purpose of CREATING a fix, not publishing it.... yet.


It wasnt a hard question, in fact if you care to know I figured it out myself
after a little while. The purpose of creating the fix is so people can get their
programs working again before waiting for a published fix, and I'm one of those
people that didn't want to wait.

Even though I dont know shit about linux, I applied the fixes..confirm it works, and I congratulate those who helped fix it, good job guys



p.s. - wait for a published fix?? Was there even a published fix from the 4/5 patch? I'm still using the fixes from these boards after that patch...You've got to be kidding

Caustic
05-28-2004, 02:27 PM
You guys are awesome....Hats off for a job well done!!!....:D

ksmith
05-28-2004, 02:30 PM
Originally posted by keendevil
p.s. - wait for a published fix?? Was there even a published fix from the 4/5 patch? I'm still using the fixes from these boards after that patch...You've got to be kidding

The fix for the 4/5 patch was put in the beta branch of CVS on Thu May 6 22:22:13 2004 UTC (3 weeks ago) by bonkersbobcat.

If you aren't using the beta, I would suggest it. It always gets fixed first.

Asterof
05-28-2004, 08:15 PM
I dont have much trouble with the patches cept this time
ver 5.0.07 on RH been working well
I patched the .h file and then copied the opcodes
When I run it I get a seq fault
there is so much chat about the fix's im confused
I cleaned and ran compiled and installed
So I need some direction
Thanks

jarula
05-28-2004, 08:31 PM
downloaded the zoneopcodes.xml.gz file from the forum here and gunzipped it

then downloaded the everquest.h.tar.gz file from the forum here and gunzipped and untarred it.


then moved the everquest.h file to /seq/showeq/src directory

performed a clean compile

then moved the zoneopcodes.xml file to /usr/local/share/showeq directory and have not had a problem.

I am using RH9

Hope this helps someone.

Choofer
05-28-2004, 11:16 PM
Im still having one fault i hope you guys can fix. I am only getting the Benificial spells cast on me in the spell window, has anyone got a fix to show all other spells.


Choof

zeroman
05-29-2004, 01:03 PM
removed my old showeq dir
did a fresh checkout of 5.0

got this as a result:

bazaarlog.cpp: In member function `void BazaarLog::bazaarSearch(const uint8_t*,
unsigned int, unsigned char)':
bazaarlog.cpp:51: `isdigit' undeclared (first use this function)
bazaarlog.cpp:51: (Each undeclared identifier is reported only once for each
function it appears in.)
bazaarlog.cpp:60: warning: int format, time_t arg (arg 3)
bazaarlog.cpp:50: warning: `char*p' might be used uninitialized in this
function
make[2]: *** [bazaarlog.o] Error 1
make[2]: Leaving directory `/root/showeq/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/root/showeq'
make: *** [all-recursive-am] Error 2

any thoughts?

wyldkrd
05-29-2004, 02:30 PM
Lemme guess running Suse..

In the bazaarlog.cpp file add in

#include <ctype.h>

and search is still your friend :)

trustno1
05-31-2004, 10:26 AM
I had the exact same error as Zeroman, I'm running RH 7.2.

Been searching for a fix, or possibly something wrong with the CVS version, so waiting for a tarball :P

EDIT: I did wyldkrd's suggested fix, and that worked with RH 7.2, thanks man!

Choofer
05-31-2004, 02:52 PM
Can anyone please tell me before i waste CVS system time and download the CVS, can you see detrimintal dots in the spell windows.

I have done all the suggested changes and patches and can only see the benificial buffs cast on me and i mainly use SeQ for timing of my dots.

Thanks in advance

Choof.

T0bai
05-31-2004, 04:31 PM
I'm currently using the newest cvs beta.

After a completely clean install, with the fixes, everything works except for guildtags.

Fedora RC2, complete new reinstall, cvs, patches, compile.

Here is a more detailed link in help forum (http://seq.sourceforge.net/forums/showthread.php?s=&threadid=4859)

Choofer
05-31-2004, 06:24 PM
just did a complete new install from the CVS and it's still broken. The fault i have is


** found out this has nothing to do with the fault **Warning: Error on zone-client: size > left (size253, left=123, opcode=0x00e2) **

thats what i get after casting a detrimental spell and i have no spell times in the spell window.

I hope this helps someone who can fix it.

Choof

update :- what is happening in the spell window is it recognises that the spell has been cast but acts like it is a DD with no timer, there are no errors reported in the main window so i cant fill you in on anything more sorry. Any help will be greatly appreciated.

newport
06-02-2004, 02:37 AM
Choof, did you copy across the updated spells_us.txt and eqstr_us.txt ?

Not sure wheather or not you posted that already but i don't have time to check previous pages right now.

My apologies if you did.

Chuggy
06-02-2004, 12:12 PM
I used Blue's files to patch 4.3.21

Everything decodes on zone in fine. But repops will not show up at all. Odd thing is, corpses do show up. In the past when repops not appearing was a problem, corpses didn't show up either. Anyone know what I might can do to fix this?

Thanks!

BlueAdept
06-02-2004, 12:38 PM
If you use the one from the cvs (non beta) and apply the 3 files from my other post, it should work. It works for me and for others.

There has been a problem with SEQ occasionally losing synch for a while. It is caused, we think, by a bad hub, network card, or because of some program running on the windows machine. Zoning will fix it.

Im going to submit a patch and hopefully get the 3 files updated in the cvs.

EDIT: I didnt realize there was a patch today. I dont know if it broke anything or not.

Chuggy
06-02-2004, 10:05 PM
It's not losing sync I don't think, because the new corpses still pop up.. dunno though.. I'm just a luser, not a programmer :)

ksmith
06-02-2004, 10:35 PM
The implen table hasn't changed. The packet structures may have, but you should see size mismatch messages from that.

Choofer
06-03-2004, 03:04 PM
Newport, that was it exactly, i'm sure that i have read and re-read all the posts on fixing the latest patch problems and i couldnt see it anywhere. I suppose it was one of those " cant see the forest for the tree's ".

Anyhow thanks again

Choof

jarula
06-04-2004, 07:23 AM
I think the CHANGES file in the download from CVS may have mentioned the needing to moving copies of the latest spells_us.txt and eqstr_us.txt.
watch your eq downloads, if it shows either of those two updating, you'll probably need to recopy them.

jarula
06-04-2004, 07:33 PM
I only have one error showing up and that is:

Warning: OP_NewZone (0xec) (datalen: 0) doesn't match:


I am using RH9 and the latest beta 5.0.0.9


other than that SEQ appears to be working fine.


thanks

Chuggy
06-05-2004, 12:34 PM
Ok I did some more checking. I am in fact losing sync.

This appears to be a similar problem that popped up a year or so ago. I don't remember the specifics, but here is what I do remember to jog your memories. The testers never saw it because it had to do with a structure or something like that changing in certain items. When the people that had the problem would stay naked after a rez (zone in) the problem wouldn't happen until they put their gear back on. Because it was only happening to some people it was thought for a long time that it was a flakey hardware issue, such as network card or router problem. But once the structure (I think that's what it was) problem was fixed, it was all good.

I will do some testing today with a naked wood elf. How bad can that be?

BlueAdept
06-05-2004, 12:43 PM
I havent had any problems with it losing synch (occasionally but not very often). It sounds like a bad net card or hub if it is doing it every time.

The reason why it loses synch is that it is missing packets. It has to be able to see all the packets or it will lose synch.

Chuggy
06-05-2004, 01:53 PM
Ok did the test..

With the naked chick it held sync alot longer.. but still failed.

I don't think it's a hardware problem. Like I said, it's the same thing that was said last year, and it's the same hardware I have now. Also, it didn't do this until the recent patch.. so. Too many coincidences.

Here is what looks like the error if it is of any benefit:
--------------------------------------------------------
SEQ: Giving up on finding arq 739a in stream zone-client cache, skipping!
SEQ: Giving up on finding arq 74b0 in stream zone-client cache, skipping!
Lost sync, relog or zone to reset
uncompress failed on 0xf2e8: -3 - data error
no further attempts will be made until zone on stream zone-client.
SEQ: Giving up on finding arq 769d in stream zone-client cache, skipping!
--------------------------------------------------------

R0x0r_Mc0wnage
06-08-2004, 07:52 PM
Edit: Nevermind..

Side note.. What could cause "General: Unknown: 0"?

ksmith
06-08-2004, 10:47 PM
Some of you may find your solution here:

http://seq.sourceforge.net/forums/showthread.php?s=&threadid=4873