Working Parser?

[Motenten]

Since I think it's fairly solid now, I'll go ahead and add my own parser to the growing list in this thread: http://code.google.com/p/kparser/

Note on requirements: You need to have .NET 3.5 and SQLCE 3.5 installed. The readme.txt on the google site has the link addresses of where to download the installations from Microsoft if you need them.

While DirectParse is more summary-oriented, KParser is more details-oriented. The potential amount of data is huge. I've written a number of tab/plugins, and am hoping to solidify an API so others can do so as well if they wish. Unfortunately that comes at a price.

Report pages can get long, and if you have the tabs active while parsing it can get very resource intensive (the worst at the moment is the Buffs page), possibly slowing your system during a parse. I'm still working on ways to better optimize code speed without making the code itself overly cumbersome, however you can turn off tabs you don't want to view at the moment and they won't eat up time on an active parse. Note that the vast majority of the processing cost is in the datamining and display; the parsing itself is quite fast.

Side note: I know there exist some issues with using it on Vista/Vista x64, at least for reading from RAM, but I'm limited in what I can do there at the moment since I don't have that OS. Switch it to log-reading mode if RAM mode doesn't work. Also applicable in case a patch changes the base RAM address.

Comments welcome.

[Breemer]

Does anyone have the new memloc to add to the DvsConf.txt to get us running again after the patch?

[Zhais]

didn't know it was broken yet :D haven't played since I d/led the patch as of yet.

got a school project that I need to work on the next few nights, so nothing from me for a bit. Still haven't packed up a new release >.>

[Motenten]

New memloc should be 0x0056DA48.

[Zhais]

New memloc should be 0x0056DA48.

and indeed it is

[Breemer]

Thanks :3

[Jetpack]

How do you find out the MEMLOC because it's not working again after March 17th update.

[retice]

New Memloc for Mar 17th update MEMLOC=0x0056DA50

[Zhais]

New release @ sourceforge website.

check out changelog for all the updates relenaseraph did to improve a LOT of parsing accuracy issues

This release includes the most recent memloc change.

[mallister]

many thanks!

[Motenten]

KParser also updated with today's new memory address. Available on the Google page.

In addition, new build config should (hopefully) make it work properly on Vista x64. (SQLCE is 32-bit only, so had to force it to x86 compat mode only, rather than AnyCPU.)

Finally, added an import mechanism to read DirectParse database files (only had limited testing, though). Hopefully comparisons between the two can help pinpoint parsing bugs.

[Breemer]

Thanks on the new release!

[Tekki]

Thanks a million for the update,

[Elipse]

I don't know if this is the place to post. I had a feature request so feel free to move it where ever this post belongs or direct me where to go.

Anyway, what I'd love to see catagory totaling of certain spells like sleeps "total sleep 1" "Total Lullaby" "Total Sleep II's" etc. Broken down by player. Not necessary to see if the sleep landed or was resisted, just want to see who was casting them in general.

It seems an odd request but I was using DirectParse today to monitor our dynamis and unfiltering cures really showed me whos "pulling" their weight in terms of team play. 1 brd for example cured 28,000+ hp while another cured 2.8k. So one brd did 10 times the curing of another. I was hoping to be able to see the relation of things like that when combined with sleeps. However, it didn't give me the full picture.

Did Brd 1 cure a ton but barely cast a lullaby? Was Brd 2 so focused on sleeping things, that he didn't have time to cure anyone? Did Blm #1 cast 150 sleeps, while the other blm casted 15? Really would help to identify where the team is weak in terms of crowd control. This type of thing comes in very handy with "low man" dynamis type runs with 18-20 people where you really need everyone pulling their weight.

I never would have been able to predict 1 brd did that much more than another, without DirectParse. So I was really hoping something like this would be a possible feature you'd consider adding. Thanks again and forgive me if this is not the place to post something like this.

[Motenten]

May want to try KParser. Check the Buff/Debuff pages and see if that's what you're looking for.

[Tekki]

add effect fixed btw <3

[Elipse]

May want to try KParser. Check the Buff/Debuff pages and see if that's what you're looking for.

Perfect, thanks ^^

[Khajit]

is there any way to view trick attack and sneak attack damage?
Can unstacked WS, SA+WS ,and TA+WS be seperated on the parse?

[Motenten]

It's probably doable, but it'll be tricky to code. I'll put it on my list of things to datamine and see if I can get it working.

[goshtin]

cant get kparser to load
"The application failed to initilize properly (0xc0000135)"

first time in ages I used a parser, so no idea whats wrong