En,
That 2x2 normal_gump means you are on the login screen. Once that is present, just go forward with setting accountname and password. You could get clever and use the offsets from #contposx/#contposy for that 2x2 gump (which lives along the lower left edge of the screen each time). You would have relative offsets that are positive for X and negative for Y.

If you try to eliminate the normal_gump (via ignorecont or cycling via event conttop), you might be able to eliminate the "normal_gump" and 2x2 size, but then I think you are left with a size of 640x480 sized gump. Several of the gumps are that size during the login procedure, so I do not see it as more beneficial to "skip" over the 2x2 gump. It shows up and you know you have arrived at the login screen.

Gaderian

The code you want uses the #findindex, because it updates all the the #find... variables on each iteration through the loop. So #findx/#findy/#findid/#findcol/#findmod all get updated.

Using a variable other than #findindex will not update the #find... variables. %1 will hold the values of 1, 2, 3, ... up to #findcnt. None of the #find... variables will change until either you use the #findindex loop or issue a new finditem statement.

Gaderian

The Ghost,
Your description sounds like you only run the code at start up. Are you including this check in a cycle?

For instance a menu with buttons, we have a loop that checks menu button variable and if it is set, then it will take action. Do you have a gosub to your routine where you pass POF every cycle (or possibly a timer based cycle)?

Notice in Freddy’s code he has a label Mloop and a bunch of tests before the go to back to that label.
He uses a test on %training_mde <> #true where he gosub to his weight check.
In the weight check he calls his bow logic routine.

Are you doing something structurally similar?
Gaderian

So once or twice each year I spend a little time trying to figure out how to clean up the report option. I like the character report because it can show me where I have skills or characters that maybe I may bring across shards or that I may wish to develop.

There are 3 things with the reporting that I have changed:

• totals line is misaligned (decimal doesn't line up)
• when an account has no character (or sometimes the login fails - maybe because the previous character was not logged out in a safe spot), the report data is all wrong on that character pass.
• all characters are reported as 'female'

Totals alignment:
I modified the format for the !Skill variable formatting from "PadWithSpaces LEFT 7" to "6" like this:

If a character was not available, then the reporting would show up like this if there was not a character on that shard for this account, taking data from whatever the previous character in the inventory list was:
****************************************************** 
 Shard: Lake Austin Characters: 1 Of 0 
 ****************************************************** 
 CharName: N/A 
 Title:  ****** $No significant properties...$ 
 . 
 Sex: Female Str: N/A Dex: N/A Int: N/A MaxStat: N/A 
 . 
 Current  Real    Cap  Locked  Skill 
 ======= =====  =====  ======  ======================== 
   21.2   21.2  100.0  up      Focus 
   55.0   55.0  100.0  up      Magery
    3.2    3.2  100.0  up      Meditation 
    1.0    1.0  100.0  down    Tailoring 
   36.8   36.8  100.0  up      Veterinary 
   52.3   52.3  100.0  up      Animal Taming 
   59.4   59.4  100.0  up      Animal Lore 
 ======= ===== 
  228.9  228.9   Totals 
 ======= =====  Here is how I adjusted the reporting on the character in the following sub:


Testing the %charloop is not more than %ShardChars then there is no more reported characters with bad statistics. 

All characters report as "Female" in the following subroutine:
Altered the gender test:

Gaderian

No one has a modern Remove Trap training script here, so I am cross posting mine from the other site. Someone may benefit from it.

;==================================================
; Script Name: Remove Trap Cylinder Training
; Author: Gaderian
; Version: 2.10
; Client Tested with: 7.0.85.15
; EUO version tested with: 1.60 ver. 335
; Shard OSI
; Revision Date: 06/16/20
; Public Release: 07/03/19
; Cylinder Trap Kits are available for purchase from the Oddities Trader at the Lycaeum.
; Start Script up to GM
; Credits to TM's journal scan, Onkelatze for kit open, wait and click routines
;
; version 1.0 7/3/2019 initial cylinder trap trainer
; version 2.0 12/21/19 significant speed enhancement:
;                        *version 1.0 average solve time per puzzle about 100 seconds
;                        *version 2.0 average solve time per puzzle 35-45 seconds
;                                 Note:average < 2 minutes on an initial reset of a partially solved puzzle
;                                      which could also apply to suffering very poor internet connection
;                     - only tests 7 colors - calculates 8th
;                     - only tests 2 to as many slots as needed to determine count
;                     - resets on partially solved puzzle (should improve any lag issues; fixes restarts)
;                     known issue: a puzzle with all 5 slots the same color may break the script, but with the
;                                  current reset it may continue. It can happen 1/32,768 puzzles - so too rare to test.
;                                  The work around if it doesn't self correct is to stop and restart the script.
;                                  If you encounter that, it is more rare than a Blaze Cu Sidhe - consider yourself
;                                  privileged. :)
; version 2.1 6/16/2020 added number of attempts to set each box to assist lag correction
;                       much slower than without it, but can cleanup the testing when the internet isn't playing nice ;)
;==================================================


This was inspired from work from Onkelatze's Remove Trap Circuit trainer. Along with The Ghost, the 3 of us tried to extend his original script to train above 75 skill. The solution finder script I had created gets too slow to work through all the possibilities, so I was searching for another way to train through a script. This was my solution.

I used it to train from 75 to 100.0. While this could train from 0 - GM, I recommend Onkelatze's script to get you to 75, because that is so much quicker than this method. I did test this to rerun it on another character from low levels successfully. I recommend buying up the skill to what is possible though to push it along - if your shard allows it (non-Siege ruleset).

The requirements are to have one of the Cylinder traps (purchase at the Lycaeum from the Oddities Trader) in the top of your backpack with your backpack open (the script does none of that for you. Press play.

I was working on streamlining some of the conditions to make it a little quicker, but ran into some weird states where it would get stuck never completing the puzzle. So I went back to this (the original) and it has been chugging along.

The script works through investigating the trap for clues and then assembles the answer.

I am uncertain how it would behave if a trap's solution was all the same color. It may choke on that or it may reset. If it chokes on it, then it would require stopping and restarting the script. The rarity of that happening though is 1/32768. That would happen approximately once in 800 hours. I am reaching GM in roughly 36-40 hours from 75 skill. So maybe every 20 characters that train might see it happen once. If it gets into a bad cycle... stop/restart works. I did not experience the situation, but walked through the logic to predict what would happen.

Those hour calculations apply to Version 1 of the script.
Version 2 runs in about 1/3 the time. So 75-GM should take about 13-16 hours. The estimate for the 800 hour occurrence gets reduced to 320 hours.

Please give any feedback or success postings.

In order for this to function, it also requires the latest (ver 6.6 or higher) of KalOCR.txt which can be found here:
http://www.easyuo.com/forum/viewtopic.php?f=3&t=53735&p=434545#p434545
Alternatively, here is a non-maintained version of the KalOCR.txt script below that can be used (maybe someone wouldn't be able to access the other one, so I am including it here).

7/11/2020 I added the version 2.1 of the Remove Trap script and version 6.10 of kalOCR here.
The major difference from 2.0 to 2.1 was adding how many attempts it would go through the click settings - because with recent internet congestion while we live in Global Pandemic mode was causing clicks to be missed. In the script it is set to 1 which runs pretty much like 2.0 did, but if you find your clicks are being missed you can increase the number and it will repeat it's attempts to click and can cleanup some behavior to leave this as AFK as possible.

All the versions of the script use pixel scans to do OCR - so you can't run it minimized. If you run into problems, that is the first thing I would verify.

Gaderian

So the problem with the logout is probably because of the following conflict:
YesNo_gump (used for the logout) is currently set to UULC.

Notice this set of ignorecont statements:
;--------------------------------------------------------------------------------------------------
;-----------------  Add this since  Build 248  by Cheffe courtosy of The Ghost --------------------
;--------------------------------------------------------------------------------------------------
;  Ignores the little translucent tool tip gumps that popup when you mouse over the buttons
;ignorecont reset
ignorecont UULC
ignorecont YCLD
ignorecont GXJC
ignorecont UGQB
ignorecont GVKC
ignorecont SVP
ignorecont SXQ
ignorecont IGHC
ignorecont QCPD
ignorecont MSMC
That first "ignorecont UULC" statement means that the YesNo_gump will be ignored by the current client.

I just commented out that line and with cybercasper's vars it is logging in and out smoothly.

Gaderian

Would it be possible to cycle through all bumps with event control conttop?

That might allow finding the gimp if it is there but not recognized by the script.

Gaderian