I fixed seq...but what is the policy on posting it now?

[codepig666]

I finally got all the packet structures fixed and got showeq fully working again. I was gonna post the changes, but I have to ask this question:

Am I supposed to post them?

I got on the irc channel and got slammed mercilessly and banned for attempting to start up a discussion about packet structures. The op who kicked me claimed that they had a working version and didn't care about anyone else.

I also noticed that Ratt has been posting tid-bits, but not full code releases.

So the big question is: is this project now living at a figure-the-rest-out-yourself level, or are we supposed to contribute when we make it work.

There has always been some counter-open-source logic here, so I don't want to piss anyone off.

If one of the developers could respond, I'd gladly post the new code (or not post it as they require).

-piggy.

[fester]

I would wait until Tuesday, because there is a patch Monday
and there will be little point in posting a patch (if that is what you wish to do) if it is broken Monday.

In the end, do as you wish (post or keep private.)

I would suspect the issue with irc, when you got booted, is that there are far too many people there who cause problems (whine.) It is sometimes difficult to weed out the trouble users from people who (like you) legitimately have knowledge or have code to contribute. Mainly because the trouble users often say they want to help, they have a fix, or they know the algorithm used. You may have followed someone 30 minutes prior and they may have assumed the "trouble user" created a new name.

This combined with the people who feel it is my job to supply them with a patch makes one have a too harsh knee jerk response at times.

[BlueAdept]

Congrats on figuring it out.

I would have to agree that waiting until after the servers come back up on monday morning might be prudent. That way there wont be a string of "Why doenst my SEQ work any more" posts if they do break it again. :)

[Ratt]

Please post a diff in thep patches section. I will incorporate it into the CVS when I get a chance.

No sense in waiting on it, regardless... as the patch will break it either way. May as well have 1 day of working SEQ *shrug*.

As for the IRC... I hear that complaint a lot. I don't hang out in IRC much anymore, so I'm not sure why everyone over there is being so supposedly vitrolic. I haven't seen it myself, just hear about it on the boards.

I will talk to Casey and see what's up. I'm trying to put a kinder, gentler face on SEQ, as some of you may have noticed. There's no longer any point in keeping it away from the unwashed masses, with the advent of LEQM and MySEQ. Like I said in my now infamous rant... it can't be stopped and it was inevitable until a Windows version of SEQ came out. (Contrary to popular belief, I never said that we, the LSEQ team would put out a WSEQ, I just said there would be one out).

So it's time to ride the wave into shore, might as well have some fun surfing it in in the meantime.

[codepig666]

Ok. Here's the deal. I'm not used to sourceforge and cvs. I could tar up the files that I have changed, but I'm not sure how to go about making proper patch files or clean diffs.

Also, I had to shoe-horn the libEQ into the makefile cause I am clueless there too.

If I post a tarball like that, will you be kind enough to walk it the rest of the way for me, Ratt?


-piggy

[codepig666]

I can't even figure out how to post a patch.

Gods willing, this attachment will work

-piggy



This stuff goes in the src directory.
There is a readme.piggy explaining what is what.
Do a make clean; make; make install from your src directory and it should work.

[Amadeus]

Thanks for the patch :)

I havn't tested yet, but since you did the work no reason for you to worry about making patches, just post it like this and one of us will take care of you.

Here is a patch file made with 'cvs -z3 diff -c > piggys_update.diff'. Note that it does not include the patch for libeq.cpp since it's not in CVS yet, and it does not include a patch for the Makefile since, well, that should be obvious ;)

[Amadeus]

And...for those that prefer this format...this is the SAME patch created with 'cvs diff -u > piggys_update.patch'

[codepig666]

Thanks Amadeus =)

Note that you need to change to libEQ.cpp tho. the buffer size in the one Ratt posted is not quite large enough for some crowded zones.

Someone please test this so I know if I forgot something.

-piggy

[Amadeus]

also...where is it announced there is another *boggle* patch this week?

[cavemanbob]

On the official site in support, scheduled downtime:

On Monday, May 19th, at 3AM PDT, (11AM GMT), all EverQuest servers will be brought down for a brief update. The estimated downtime is 2 hours.

The EverQuest Operations Team

[Kimbler]

I dont want to hijack this post so continue on along it's started lines but some food for thought about Piggy's treatment at:

http://seq.sourceforge.net/forums/showthread.php?s=&threadid=3473

[sequser0067]

ok not sure on how to apply this patch if someone could please help me out if they can. I used the search and read thru archives to see if I could find anything about patching ect.

Do I need to compile a new libeq.a? if so When i try to g++ libEQ.cpp I get errors about qapplication.h/qdatastream.h and qfile.h not found. so it doesnt compile

let me know if I should just stfu and wait for CVS to have the update and working version. =p

Thanks for your time

[e@tme]

I have tried both the attacment and the cvs and both datasets give me the following error

make[2]: Entering directory `/seq/showeq/src'
make[2]: *** No rule to make target `m_libEQ.cpp', needed by `all'. Stop.
make[2]: Leaving directory `/seq/showeq/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/seq/showeq'
make: *** [all] Error 2


anyone got any ideas ?

Cheers

[e@tme]

I realised that libEQ.cpp and m_libEQ.cpp are the same things - so i cp'd the file and make now works

[fester]

Originally posted by sequser0067
ok not sure on how to apply this patch if someone could please help me out if they can. I used the search and read thru archives to see if I could find anything about patching ect.

Do I need to compile a new libeq.a? if so When i try to g++ libEQ.cpp I get errors about qapplication.h/qdatastream.h and qfile.h not found. so it doesnt compile

let me know if I should just stfu and wait for CVS to have the update and working version. =p

Thanks for your time

use "patch < filename" to install the patch. If it cries about "looks like a reverse patch", use "patch -R < filename" instead.

You do need to recompile libeq.cpp into libeq.a library. Don't do g++ libEQ.cpp, do "g++ -c libeq.cpp" and then "ar cq libeq.a libeq.o" to make libeq.a file. As far as your qapplication.h errors, they sound like you have your qt include files in a non standard location. In which case, you would need to add to the g++ command something like "-Idirectory" where directory is the location of the root of your QT include files (/usr/include/qt3.03, e.g.)

As far as stfu, it certainly would be easier to wait, but sometimes, the waiting is painful. Good luck.

Edit: made a tad bit more clear.

[sequser0067]

Thank you fester

I got seq working with what e@tme posted

notice a few bugs in it but it works for the most part

*bugs*
Guild tags not showing up (or incorrectly ie mobs are in guilds now!)
spell list not working
map not updating like it normally did

just a few that I noticed right away.

Thanks piggy for doing what you did

thanks to everyone who puts effort into getting this to work again.

wish I could be more help
I now my basics and that is about it.

[quackrabbit]

Originally posted by codepig666
I got on the irc channel and got slammed mercilessly and banned for attempting to start up a discussion about packet structures. The op who kicked me claimed that they had a working version and didn't care about anyone else.NONE of the ops in #showeq still play EQ anymore, they all do other things. They also feel that since they've moved on from EQ you should also.

Perhaps we should start another IRC channel on trifocus.

[datadog]

i downloaded a clean CVS, compiled libeq.cpp and moved the libeq.a to /usr/lib....

did

make -f Makefile.dist
./configure
make

on 'make' i get an error...

*** No rule to make target '`m_libeq.cpp', needed by `all'. Stop

stumped (not a difficult thing to do :-p )

Is the LibEQ.a source in the CVS now? is it compiled automagically??

Guess this should go to the help forum.. /shrug

[fester]

Originally posted by datadog
on 'make' i get an error...

*** No rule to make target '`m_libeq.cpp', needed by `all'. Stop

stumped (not a difficult thing to do :-p )

Is the LibEQ.a source in the CVS now? is it compiled automagically??


I don't profess to be a QT/MOC expert, but because it is likely getting forced because of the "m_%.cpp: %.h" rule. You would need to remove all reference to libEQ.cpp from the sources and add libEQ.a as a dependency to showeq, then add the rules to build a libEQ.a from libEQ.cpp.

LibEQ.a source is available, but not yet in CVS. It will likely be soon, as it is a required recompile to get seq working right now.

The Makefile is not currently properly coded to have seq compile libeq.a automatically.

[codepig666]

If you look in the Makefile in your src directory, you can just remove the LIBS reference to libEQ.a and add libEQ.cpp to the other places where things like packet.cpp are being compiled.

You really don't need a library.

[monster69]

Very nice work codepig.

One more detail to the list of bugs posted before though. The z-axis is fubar also. I am trying to figure out why, but, I am just a network guy, programing isn't on my radar. /sigh


Monster69

[dogmeat]

Thanks codepig, nice work.

I think that having a working version of SEQ for any amount of time could be useful to people who may wish to investigate/fix post-patch, because it would presumably allow them to compare SEQ's packet dumps and working code from before patch, with new dumps and potentially broken code post-patch.

Thanks again for your work and for posting it.

[Amadeus]

strange...I'm getting very weird things with this patch. Things are decoding..but getting garbage as if the structs are wrong.

Let me know when cvs is updated :)

[LordCrush]

Hmm can someone please post the md5 and date / size of the eqgame.exe that works with this patch ? - I am playing on the german server and i am not sure if we have the same version.



/cheer Codepig =) Thank you

[lullimarka]

patch: *** Only garbage was found in the patch input.



tried both with -u and -c ......

[Zeppo]

Just a question here.
If this is how a person trying to help the project gets treated:



Originally posted by codepig666
I got on the irc channel and got slammed mercilessly and banned for attempting to start up a discussion about packet structures. The op who kicked me claimed that they had a working version and didn't care about anyone else.


Then what are other people contributing to when they get their penguin on the side?
What's the incentive to contribute if there is no urge to help everyone?

I know I have no room to talk since I have not contributed, but this will definately make me never even think about it if that is how others are treated that are "below" the ops and devs.

[fester]

Originally posted by lullimarka
patch: *** Only garbage was found in the patch input.



tried both with -u and -c ......

Taking a stab. Is the file still .gz compressed?

[fester]

Originally posted by Zeppo
What's the incentive to contribute if there is no urge to help everyone?

I know I have no room to talk since I have not contributed, but this will definitely make me never even think about it if that is how others are treated that are "below" the ops and devs.

Would be best to take this post to another thread.

I can not (will not) speak to your question about contribution. I can speak to it about the irc channel. It is virtually useless for discussing showeq code now for several reasons.

[seqmage]

this is how i compiled.. on my redhat 8 box..
hope it helps someone..
i donwnloaded the seqpatch.tar.gz to my home dir..

[ice@Carpet]cd /seq/showeq
[ice@Carpet showeq]$ su
Password: (note use your root password)
[root@Carpet ice]# mv /home/ice/seqpatch.tar.gz /seq/showeq
[root@Carpet showeq]# gunzip seqpatch.tar.gz
[root@Carpet showeq]# tar -xf seqpatch.tar
[root@Carpet showeq]# g++ -c libEQ.cpp -I /usr/lib/qt3-gcc3.2/include
[root@Carpet showeq]# ar cq libEQ.a libEQ.o
[root@Carpet showeq]# ls libEQ*
libEQ.a libEQ.cpp libEQ.o
[root@Carpet showeq]# md5sum libEQ.a
***note*** your md5sum maybe different then what i get?
9b4c33225ae6b42c14568c7ed4535992 libEQ.a
[root@Carpet showeq]#cd /usr/lib/
[root@Carpet lib]# mv libEQ.a libEQ.a.19may
[root@Carpet lib]# cp /seq/showeq/libEQ.a /usr/lib/
[root@Carpet lib]# cd /seq/showeq
[root@Carpet showeq]# make clean
[root@Carpet showeq]# make -f Makefile.dist
[root@Carpet showeq]# ./configure && make

its compiling now.. i hope it works.. if someone can help make this a little clearer.. please let me know.

[user387]

Originally posted by fester
I don't profess to be a QT/MOC expert, but because it is likely getting forced because of the "m_%.cpp: %.h" rule. You would need to remove all reference to libEQ.cpp from the sources and add libEQ.a as a dependency to showeq, then add the rules to build a libEQ.a from libEQ.cpp.

LibEQ.a source is available, but not yet in CVS. It will likely be soon, as it is a required recompile to get seq working right now.

The Makefile is not currently properly coded to have seq compile libeq.a automatically.

actually, it is.

touch seq/libEQ.h fixes this problem..

we should either rename libeq.h to libEQ.h or rename libEQ.cpp to libeq.cpp, and this problem is history...

[BlueAdept]

They extended the patch time...I have a bad feeling. Sucks because I took today off too....

[powerspike]

ok a couple of things i have found with this patch =)


i had to run the moc line by hand (missed in the configure/make stuff - ran fine after this)

/<PATH TO QT>/bin/moc -nw libeq.h -o m_libEQ.cpp

that's all i had to, no playing with anything else from memory.




When run, Eq give the message that decoding is disable.
If started from console gives alot of the following 2 lines (asuming debug stuff)
clean uncompress on 0xXXXX:
decoding 0xXXXX with 0xXXXXXXXX:

*Bad Guild tags
*Directional movement lines aren't working (trying turning off and on and still no working)
*Odd mob is showing up in a differcolor - like when it should be Dark blue it's showing up pink, rare but happens =)
*Mob timers don't seem to be working (ie mob will spawn in this location in X mins - static spawns)


This is what i have found so far.


(This is for both this path and cvs version from 20/5)

[user387]

Originally posted by powerspike
ok a couple of things i have found with this patch =)


i had to run the moc line by hand (missed in the configure/make stuff - ran fine after this)

/<PATH TO QT>/bin/moc -nw libeq.h -o m_libEQ.cpp

that's all i had to, no playing with anything else from memory.

use the search! there are a couple of solutions to this posted already!





When run, Eq give the message that decoding is disable.
If started from console gives alot of the following 2 lines (asuming debug stuff)
clean uncompress on 0xXXXX:
decoding 0xXXXX with 0xXXXXXXXX:

*Bad Guild tags
*Directional movement lines aren't working (trying turning off and on and still no working)
*Odd mob is showing up in a differcolor - like when it should be Dark blue it's showing up pink, rare but happens =)
*Mob timers don't seem to be working (ie mob will spawn in this location in X mins - static spawns)


This is what i have found so far.


(This is for both this path and cvs version from 20/5)

first 2 things are just messages, not harmfull. there's a patch floating here somewhere for those already.

[tonni]

i got up and runing /cheer the coders well done