OK, it usually does the Tricks of the Trade though it seemed to skip a couple times when it would have given me an extra quality round (I think; I might have missed durability being low). On Armorer, it stops every round or two just before the final progress step (low CPs). The console makes it look like it's checking the buffs then it throws "failed with [string "%local directory%\FFXIV Minion\LuaMods..."]:557: attempt to index local 'bestSkill' (a nil value).
There are other possible minor quirks but the craft stopping is the current critical bug.
Thanks!
Edit: When it stops it's always on D='Checking Buff: 253' with the above error. Also, this is a low level armorer (lvl 13) with only a couple cross-class skills if that matters. Nor does it stop on Checking Buff: 253 every time...I'm still trying to come up with a common point.
(10-29-2013, 10:03 PM)Hairyogre Wrote: [ -> ]OK, it usually does the Tricks of the Trade though it seemed to skip a couple times when it would have given me an extra quality round (I think; I might have missed durability being low). On Armorer, it stops every round or two just before the final progress step (low CPs). The console makes it look like it's checking the buffs then it throws "failed with [string "%local directory%\FFXIV Minion\LuaMods..."]:557: attempt to index local 'bestSkill' (a nil value).
There are other possible minor quirks but the craft stopping is the current critical bug.
Thanks!
Edit: It's stopping on D='Checking Buff: 253' with the above error. Also, this is a low level armorer (lvl 13) with only a couple cross-class skills if that matters. (Buff 253 is Steady Hand)
Interesting, I've been testing with Steady Hand II, but that shouldn't matter. I'll look into that.
I've found the problem, and it has to do with the fundamental design of checking CP before choosing the skill (this should happen in the original, too, does it?), which is something that I was trying to figure out but thought I could put it on the backburner. Looking into it now.
(10-29-2013, 08:51 PM)dannuic Wrote: [ -> ]I've modified the code a bit.
Added:- Does Hasty Touch spam fairly well (had to change minimum CP for touches to 0, this needs to be generalized)
- Uses Tricks of the Trade on Good procs
- Uses Cross Class Buffs (ie SH2)
- Added a progress formula instead of just hitting it to find out progress gain
- Added a Quality formula (planned on doing some intelligent moves like quality finishers)
- Quality and Progress Formulas need craftsmanship, control, and recipe level, so added text box to enter it in the GUI (NEEDS TO BE FILLED IN FOR BOT TO WORK)
- Added efficiency to the skills (need to add length, too)
Need:- Player stats (control and craftsmanship) need to be exposed
- Recipe Level needs to be exposed
- A way to use HQ ingredients (this is probably hard)
- Working on some logic to use synth/touch efficiency to plan out some moves
Would love some feedback, and I would like to thank ymko for starting this up and giving me a framework to work in (beats writing from scratch....)
edit: had a bug, replaced file
Could you use GIT and push it please? I want you to have credit on GITrepo, rather than me pushing it
If it helps, I've run 60 40-durability crafts tracking this. It failed 17 times. Every time it failed it was on 5 CP so obviously I was wrong that low CP had nothing to do with it. Interestingly the bot never got stuck when I ended with 7 or 9 CP...just when I had 5. I upped the wait time to 4000 ms and it seemed to help though the problem didn't stop...maybe it's lag related? I heard people complaining about lag affecting their crafting today.
I don't recall the previous version failing in this way but if it's lag exposing the problem that could explain things.
Forked your repo and pushed my changes to the new fork (new to git, I've always used older versioning tools).
(10-29-2013, 10:35 PM)Hairyogre Wrote: [ -> ]If it helps, I've run 60 40-durability crafts tracking this. It failed 17 times. Every time it failed it was on 5 CP so obviously I was wrong that low CP had nothing to do with it. Interestingly the bot never got stuck when I ended with 7 or 9 CP...just when I had 5. I upped the wait time to 4000 ms and it seemed to help though the problem didn't stop...maybe it's lag related? I heard people complaining about lag affecting their crafting today.
I don't recall the previous version failing in this way but if it's lag exposing the problem that could explain things.
It may well be lag since it seems to fail to recognize a buff or your CP properly. What I pushed to github has a temporary workaround for that; see if it works.
(10-29-2013, 11:26 PM)dannuic Wrote: [ -> ]Forked your repo and pushed my changes to the new fork (new to git, I've always used older versioning tools).
You have to make a pull request
(10-29-2013, 11:38 PM)ymko Wrote: [ -> ] (10-29-2013, 11:26 PM)dannuic Wrote: [ -> ]Forked your repo and pushed my changes to the new fork (new to git, I've always used older versioning tools).
You have to make a pull request
I see. Done.
Dannuic, using your new code I tested it 15 times. 2 failures (both on 6cp this time) so a lower failure rate but it's hard to draw a conclusion from such a small sample. I left the WaitTime at 3000 as I figured we wanted to break it if it could be broken...
By the way, I wouldn't read too much into the failure happening after checking buff 253; that's the last buff check reflected to the console so it's what the mod displays when the next crafting step starts...which is where the failure is probably really occurring.
(10-30-2013, 12:03 AM)Hairyogre Wrote: [ -> ]Dannuic, using your new code I tested it 15 times. 2 failures (both on 6cp this time) so a lower failure rate but it's hard to draw a conclusion from such a small sample. I left the WaitTime at 3000 as I figured we wanted to break it if it could be broken...
By the way, I wouldn't read too much into the failure happening after checking buff 253; that's the last buff check reflected to the console so it's what the mod displays when the next crafting step starts...which is where the failure is probably really occurring.
This is quite strange. I put nil checks so that shouldn't be able to happen. Are you still getting the same error (possibly different line) that bestSkill is nil? Also, did you grab from my fork, or ymko's original? I don't think it's been committed to the main fork yet.
Yes on grabbing the forked code (I cut and paste it as there was no download option on the fork). But...erm...I didn't save after cutting and pasting.
Let me log in and try that again.