Memory reading : export item tooltips

[infotech1]

Well finding Objectmanager offset is done by pattern matching as well, just manually in this case. Pattern matching to find the containers in objman be easy as well what with the nice string name at the start and reliable layout. But your right, i had just assumed you used set offsets, hadnt read the code to see.
Nice app btw

[extern]

I'd love to see a quick output for this tool that could, say, copy the stats to the clipboard in an easily splittable way, or writing to a temp file that could be read by another program. Is that something you could do in the future?

sunblood, why would you like to use another software to deals with captured tooltips through temp file ? The source code is available, just modify it to fit your needs.

could you go into more detail for reading the random stats of the item? stuff like +57 int etc

Honski, random stats are in "Root.TopLayer.item 2.stack.frame body.stack.stats", you would find a string pointer pointing at something like this "{icon:bullet} {c:ff6969ff}+4 intelligence{/c}"
I'm searching for that hash {0xB7, 0xAB, 0xAD, 0xB5, 0x84, 0xD3, 0xFB, 0xFC} followed 0x210 bytes after by that hash {0x7E, 0x4E, 0xD4, 0x05, 0xEE, 0x32, 0x5C, 0x37}.
After checking constant values at offsets (0x458, 0x490, 0xAD8, 0xB10, 0xB60, 0xBB0 and 0xBEC), I assume that the string pointer is at offset 0xA98.

With CE, search for that string "Root.TopLayer.item 2.stack.frame body.stack.stats". You would have 4 or 5 matches. Add 0xA90 and you have your string pointer.

[206221]

nice job . ~

[kingwa5]

v0.2 is out, with new ToolTips support like sockets. Available on github.

where is the new version?

[sandcho]

Hi extern i think there is bug in your 0.2v - after some item has sockets you don't clear the sockets variable or i don't know but each next item has the same sockets until next item with socket
i have not debugged much the 0.2v because i was using 0.1 because i had a software witch imports the XML and didn't have time to fix for v0.2

[extern]

Hi sandcho, thank you for highlighting that point. I forgot to remove hidden fields {QtEBUG:hidden} from exported data. An issue has been added on github, and I am going to fix it next monday. By the way, I am sorry for having changed the XML schema from v0.1 to v0.2.

[extern]

where is the new version?

It's on github, I pasted the link on the first page of that thread. If you are not familiar with github, you can download a zip file here : https://github.com/d3dev/d3_tooltips/zipball/master
The compiled binary is in release/D3_ToolTips_v0.2.exe and you may need to install the Microsoft Visual C++ Redistributable Package as said in the readme file.

[Venk2012]

From today, im always run into this error:



i did not change anything, yesterday it works like a charm.

Any chance to get that fixed?

Version: D3_ToolTips_v0.3.exe

[extern]

From today, im always run into this error:



i did not change anything, yesterday it works like a charm.

Any chance to get that fixed?

Version: D3_ToolTips_v0.3.exe

Hmm, this happens when a field (name, durability, etc) is not found in memory, or if it's found more than once. That was a common error in v0.1 when searched patterns were overlapping memory regions. It may also happens if patterns are located outside of scanned memory range [0x06000000 - 0x30000000].

//D3_ToolTips.cpp:110-111
unsigned int min = 0x06000000; //(unsigned int)si.lpMinimumApplicationAddress; //0x06000000
unsigned int max = 0x30000000; //(unsigned int)si.lpMaximumApplicationAddress; //0x30000000

In order to investigate, we have to add debug code on D3_ToolTips.cpp lines 214,215, find out what field is missing, and then browse the memory to understand why the pattern did not mach.

This error is based on random memory layout, so it's difficult to trigger it. I just run the game, and it works for me.
Try to restart the game, I can not do anything until I got the bug.

Anyway, thank you for reporting that bug.

[extern]

I just try to run D3_ToolTips on PTR, and it does not work. Tooltips structures have been modified. (I tought it was unlikly to happen).
I don't know if these modifications are specific to PTR client or if it will be include in 1.0.5 patch.

I pushed a v0.4 release with a slightly wider memory range (from [0x06000000 - 0x30000000] to [0x06000000 - 0x40000000]), I do not expect to fix the bug with that.
I also add a new branch for PTR 1.0.5, so you can download D3_ToolTips_v0.4 for PTR : https://github.com/d3dev/d3_tooltips...s_v0.4_PTR.exe

[Venk2012]

Thanks for the new Version.
I will give it a try

[sandcho]

is this ready for 1.05 ?

[Venk2012]

Try the PTR Version:

https://github.com/d3dev/d3_tooltips...s_v0.4_PTR.exe

It works good for me

[extern]

Just to say that I merged PTR branch with master, so no need to bother with that again.

D3_Tooltips is working on patch 1.0.8 without any code/binary update, so the patch agnostic feature is pretty reliable.
No issue has been reported since patch 1.0.5 (6+ months ago) so I decided to release a stable (perhaps final) version.
https://github.com/d3dev/d3_tooltips