Page 2 of 8

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 01:11
by heero
@kali, not much since real life work is keeping me occupied. Only took a quick look at pRO to ascertain if they did anything at all. So far the login packer is the main change judging by the packet but you can bypass that by using xKore, for now I will just focus on xKore since my work is keeping me occupied. After all bills have to be paid. Well back to work for me, will check pRO again later when the server goes up.

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 02:14
by islarky88
@heero, how about my solution for x0... i think...

this is to be added to receive/pRO.pm

Code: Select all

'0009' => ['account_id', 'a1 a2 a4', [qw(buffer packet accountID)]],
'0010' => ['skills_list', 'a1 a2 a4', [qw(buffer packet skillslist)]],
buffer (a1) is the third byte of the length, packet is the 2bytes, 3rd is the packet msg

but all this still won't matter because i still get timed out in connecting to map server...

does the master_version matter when connecting to map?

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 02:42
by heero
islarky88 wrote:@heero, how about my solution for x0... i think...

this is to be added to receive/pRO.pm

Code: Select all

'0009' => ['account_id', 'a1 a2 a4', [qw(buffer packet accountID)]],
'0010' => ['skills_list', 'a1 a2 a4', [qw(buffer packet skillslist)]],
buffer (a1) is the third byte of the length, packet is the 2bytes, 3rd is the packet msg

but all this still won't matter because i still get timed out in connecting to map server...

does the master_version matter when connecting to map?
Sadly that wont work since there are packets which have the same size but different packet switches. It will only make the bot fail to understand those packets properly.

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 04:59
by fjx
The received packets with structure <length><header><content> are only being sent by the server when you are inside the game with the exception of 'map_loaded' packet.

The problem after this, is - the packets being sent by the client are encrypted. Also, the code for the encryption is virtualized ( vmp/safengine ) so it's a pain in the ass to figure it out. As far as I know, someone has fully emulated the encryption already.

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 05:04
by benedictbenedict
heero wrote:.... Well back to work for me, will check pRO again later when the server goes up.
region blocked? :| :cry: I can offer you free VPN access using my router if you have problems connecting to pRO (20mbit/s up&down)

Read pRO Announcement
https://www.ragnarokonline.com.ph/news/ ... ver-update

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 05:34
by xypie
fjx wrote:The received packets with structure <length><header><content> are only being sent by the server when you are inside the game with the exception of 'map_loaded' packet.

The problem after this, is - the packets being sent by the client are encrypted. Also, the code for the encryption is virtualized ( vmp/safengine ) so it's a pain in the ass to figure it out. As far as I know, someone has fully emulated the encryption already.
Would this be also the case if you're on xKore 1?

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 09:31
by heero
fjx wrote:The received packets with structure <length><header><content> are only being sent by the server when you are inside the game with the exception of 'map_loaded' packet.

The problem after this, is - the packets being sent by the client are encrypted. Also, the code for the encryption is virtualized ( vmp/safengine ) so it's a pain in the ass to figure it out. As far as I know, someone has fully emulated the encryption already.
We tackle one problem at a time for each coding issue we have in openkore. This so called encryption you mentioned is not yet tackled since we are first trying to make openkore understand the new packet structure. If and when we finish coding the new structure then we will tackle the next issue if there is an encryption in there. I have only done a quick look of the packets and have not inspected them fully but rest assured it will be done one step at a time.

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 11:49
by fclose
Hello,

Here's the (testing) fixed for packet tokenization problem. replace it in src\Network.

btw it's still very dirty fix for just the received packet thus still leave sent packet untouched. also i'm not sure if it works or not as my pc can't run the ragnarok client right now.

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 12:32
by d4xyjen
fclose wrote:Hello,

Here's the (testing) fixed for packet tokenization problem. replace it in src\Network.

btw it's still very dirty fix for just the received packet thus still leave sent packet untouched. also i'm not sure if it works or not as my pc can't run the ragnarok client right now.
Got disconnected on connecting to map server

Re: Packet Tokenizer issue with users of XKore 1

Posted: 05 Jul 2017, 18:02
by dawnvillain
fclose wrote:Hello,

Here's the (testing) fixed for packet tokenization problem. replace it in src\Network.

btw it's still very dirty fix for just the received packet thus still leave sent packet untouched. also i'm not sure if it works or not as my pc can't run the ragnarok client right now.
It won't connect to the map server.

I think one problem is that the recvpackets.txt are not updated?
I've been following the topic since day 1, and i believe that most people are having difficulty extracting the packets from the Ragexe.

PS: It will connect if I'm using the default MessageTokenizer.pm

Also, you might find this useful.
https://github.com/lututui/bRO-Updates

That person is lututui (one of the devs that manages bRO).
You might want to check his latest bRO connection update and perhaps use it as a guide/basis to reverse engineer pRO's kore settings? I just thought you guys might find it useful.


Image