Skill Scanner

[jdegre]

hi,

i've made a small program to extract skills info from screenshots; the same as EUComp or NRF did but, hopefully, it should work with the new skill screens.

please find detailed info here:

Skill Scanner

the program has been in beta testing for several months now, and it is getting quite ready to a "release" status. still, if you find out bugs or parsing errors, please let me know by PM, email, or posting here.

any other kind of suggestions, impressions, recommendations, etc... are extremely welcome.

main features:
- the scanner parses your in-game skill screens and shows skill levels, skill tt value and skill market value
- market values are calculated from peauction.com prices
- profession levels are calculated (not parsed from in-game screens); there are some high-level unlockable skills, whose influence in the professions is unknown, so some professions might not be 100% accurate. (texture-related professions are completely unknown yet).
- you can export skill and profession levels to a text file formatted in Comma-Separated-Values (CSV), to be used with other applications.

additional features (only if you have an account in the chipping optimizer; read below, in the "donation-ware" paragraph, about how to get an account):
- you can save your skills with 1-click in the chipping tool
- you can load your skills from the chipping tool and check your progress comparing your current skill set and your last saved skill set
- same progress tracking, but applied to profession levels


known issues:
- the scanner relies on the windows clipboard to get the content of the in-game screenshots. there are issues related to the way the Java system handles this clipboard that makes it difficult to co-exist with other clipboard-reading applications (such as PEAss, AutoScreen, and such). if you experience problems with it, please turn off temporarily any of these applications while you run the skill scanner
- the gamma setting (when you're running in full-screen mode) should be set around the center point of the slider. most probably, it will work ok with other settings too, but if you move it too much to the extremes (i.e. too dark or too bright), the parser may fail



Note: this software is donation-ware. what does this mean?:

i was asked some time ago (in the chipping optimizer thread) if the scanner program was going to be released "for free", or for "subscribers only", and my answer was that it was going to be "donation-ware", the same model as the chipping optimizer web app.

this means that it is absolutely free to use and people have no "obligation" whatsoever to donate, if they choose so. but, it is very appreciated if those people that find the application useful, and use it frequently, decide to support its development with a small donation.

this model has been "reasonably" successful with the chipping optimizer (which means that, in terms of donated money, it barely covers the web site hosting fees, but you get the idea, i appreciate people's gesture of contributing to it).

i have included a "donate" link, so people can donate via paypal. other than that, if people wants to donate ingame peds, that's also ok.

donators to the scanner tool, will also be "rewarded" with an account in the chipping optimizer.

some people seem to be a little bit confused about this model, and they usually ask me what's the purpose of paying for a feature (the account) that does not really add a lot of functionality to the tool (it is more of a "convenience", to save you some typing and stuff...). well, the answer is that i'm not "selling" the account functionality; instead, i'm simply asking for donations to support the development of these (and future) tools; the "account" thing, is simply a kind of reward for donators.

if you want a simple comparison, it is similar to the old EF forum donations, where people contributed to support the forum, and they were rewarded with minor forum function enhancements (such as a rep boost), but they were not "buying" these features. hope it is clearer now



cheers,
/jdegre.

[Xen]

It rocks!

Thank you!

[alh]

Great! I was afraid I would have to do get my half working code going for it

Will you (maybe when you've tested it a bit more) share it with the NRF guy?
He is looking for help with code for this.

Would be great if also NRF would start working again.

Darn, can +rep you now, another time

[Witte]

Great job

I have a feature request for you right away (wihout even ever trying the tool ). If its not too much work to make, it would be great if we could also read chat, so that the result is in seqeuntial order, without double or missing lines. Would be very usefull to test attachments.

Will give it a try soon, tnx for making this.

[safara]

Originally Posted by Xen

It rocks!

Thank you!

That it does

Nice work jdegre

[jdegre]

Originally Posted by alh

Great! I was afraid I would have to do get my half working code going for it

Will you (maybe when you've tested it a bit more) share it with the NRF guy?
He is looking for help with code for this.

Would be great if also NRF would start working again.

Darn, can +rep you now, another time

i've been in contact with squee, and we've been sharing our respective approaches and comparing some ideas.

however, sharing code does not seem feasible since NRF is C++ and my app is Java (not mentioning that my code maturity is very low atm, which makes it very unsuitable for reusing).

[alh]

Ok I see. Maybe I should dust off my code then anyhow.

My code works for the trade window currently, (except for "alferix" for some reason), by comparing font samples taken from screen shots to extract the text.
also has character withs, and tries with different offsets, to cope with un-even kerning.

It should be possible to adapt for the skill view by adding a filter function to filter out only the pixels with high luminosity (text is pretty bright), and a new font sample table...

Do yours work along the same lines?

My code is available in C#, easily portable to C/C++, and an old version in plain C that probably works for the chat still

[Vedder]

Originally Posted by jdegre

...and my app is Java (not mentioning that my code maturity is very low atm, which makes it very unsuitable for reusing).

Ooh... Nice with a Java app
Any chance of making it Open Source?

[Galax]

Thx works perfectly

[jdegre]

Originally Posted by alh

Ok I see. Maybe I should dust off my code then anyhow.

My code works for the trade window currently, (except for "alferix" for some reason), by comparing font samples taken from screen shots to extract the text.
also has character withs, and tries with different offsets, to cope with un-even kerning.

It should be possible to adapt for the skill view by adding a filter function to filter out only the pixels with high luminosity (text is pretty bright), and a new font sample table...

Do yours work along the same lines?

My code is available in C#, easily portable to C/C++, and an old version in plain C that probably works for the chat still

yeah, that's the basic approach of my program too. i think that squee was investigating a pure OCR approach (character recognition, instead of bitmap matching), which is much more flexible, but extremely harder to implement.