The smaller one. I THOUGHT I had deleted the other. My bad! Fixed now.

Oracle, thanks for the test. I broke it last night trying to get fancy and was too tired. I got an extra hour so I fixed and partly tested. I did not test the trash sub, but the bank subs work fine. I updated the script to recalling straight through all locations for the beta file below. I did find a couple of specials ID's I had to add. If you find any, let me know and I'll get the types added. Thanks for testing!! If You find any I missed, post em here and I'll get them added.

still finding stuff I don't have in the specials yet. I know I'm missing the north/south webs. The Type is in the %loot list, but since I don't know what it is, I haven't moved it yet. THis could create a problem if you are dumping the candy, as they are getting dumped too.

Another minor issue is blocked runes. When I change the end location variable to +1, it tries to recall the the blocked rune, then stops. When I have it = the start location, it recalls over and over, then stops. Welcome any advice on this! (hint hint nudge nudge Gaderian)

GOTCHA!!!
I only used the random because that was the only one i could get working. Which was the issue. I couldn't get any of the others to change runes, that is what I was really after.

Lets say I want to do sequential recall then, which is what I tried to do, but couldn't figure out how from the header.
I would much rather randomize my runes in the book at this point and recall sequentially 1- 16. Maybe when I get rid of my scripter pacifier I'll try the random multi book thingy.

While I figure THAT part out, which won't take a lot of code rearranging i think, I have gone ahead and fleshed out the bank and trash routines, with the option (hardcode for now) to save candy or trash it.
This is totally untested. I will not have the time until late tomorrow. If someone brave wants to take a stab, here it is.
Need one rune book with 16 trick or treat target runes named TrickOrTreat1.
One rune book with 16 bank runes named Bank
One rune book with 6 runes to trash barrels named Trash
2 empty bags in your bank (1 if you want to trash the candy)
It'll take about 12 hours to fill a 125 spots, or about 10 specials an hour. ish. I don't have a bank is full option coded, so it's up to you to watch that.

First, Great news. Neo granted me permission to scavenge subs from his trick or treater, so it will be much easier to do the bank and trash subs!! Thanks NEO!!!!

I finally understood what you were getting at there Gaderian. I was misreading the instructions in the travel sub header, which lead to my confusion. I ASSUMED that the numbers represented the range of the runes in the book, not the range of runes to be used for the recall attempt.
When I ran the script with your suggested changes, I noticed it was skipping every other rune n the book, because of :
set %RuneLocStart %RuneLocStart + 2Yet, it worked flawlessly except for not hitting all the runes.
When I changed it to :
set %RuneLocStart %RuneLocStart + 1I would always get stuck on the third rune and not change to the fourth.

Here is a decently working version, but you have to dump it every couple of hours.

Need 16 runes to places with 2 or more valid trick or treat targets that are static. Targets must be within 4 tiles for Trick Or Treat to work.

Ill work on adding in NEO's subs later today. One tip, Do not start it standing on your first recall spot. It will get stuck. Start from your home or anywhere and it will be fine.

*heads up for prep* Ill probably call for a 16 bank rune book and a 16 trash rune book as well. Lots of set up I know, but beats getting paged. I just gotta fix the skipping thing.

OKay, I took all the suggestions and rethought my process using ghosts simplification. I instead went with TM's recall sub. The script seems to work, but the runebook wont change pages, instead it stays on the starting number in the range, no matter what I put it at. I will be adding other subs, weight check, bank, trash, menu, etc once I get this working and recalling around. I tried recall from targeted runebook, and recall from named runebook. same result. Random runebook travel works though. However, it doesn't appear to detect blocked spots. If it randomly chooses the same spot I occupy, it recalls continously.

HI! I'm baby stepping my way through making a robust trick or treater. First, I want to ensure I can reliably target, trick or treat, detect a teleport, and recall back to the spot. The script will "eventually" be a recall based script with bank/trash support. BUT I am having issues getting it to detect the teleport. This is what I have for one static NPC. The targetting and trick or treat and timing are spot on so far.

Display ok Target your recall rune or book
set #targcurs 1
  while #targcurs = 1
    wait 0
  set %rune #ltargetid
wait 5
;----------
 Display ok Target your First Animal.
     set #targcurs 1
  while #targcurs = 1
    wait 0
  set %heal1 #ltargetid
wait 5
;-----------
mainloop:
gosub TM_AdvJournalSync TRICK 100
gosub trickortreat1
wait 60s
return
;----------
Sub trickortreat1
set #ltargetid %heal1
set #ltargetkind 1
msg $trick or treat
 if  #targcurs = 0
{
 repeat
}
target
Event macro 22 0
gosub trickcheck
return
;-----------
Sub trickcheck
gosub TM_AdvJournalScan TRICK VALID_Advance teleports
 If #result = #true
          {
          msg $trick detected
           }
  return
;=================================================================
; Script Name: TrailMyx's Advanced Journal Scanner
; Author: TrailMyx     gosub TM_AdvJournalScan heal VALID_ADVANCE
; Version: 1.2
; Shard OSI / FS: OSI / FS?
; Revision Date: 10/20/2007
; Purpose:
;   Use these subs to quickly find text in your #journal entries.  These subs
; use #jindex for flawless journal scanning and is much more reliable than using
; standard indexing of #journal and #SYSMSG.
;
;   Now it is possible to manage separate journals based on unrelated text.  It's now possible
; to monitor spellcasting, bandaging, stealing, or anything else without a TM_AdvJournalSync
; potentially removing text needed for another UO funciton.
;
;   New is the ability to either gosub or call these functions without the need to change the
; header!  When calling, a limit of 10 arguments is allowed, but more can be added by editing the
; call interface section.
;
;  Examples:
;     gosub TM_AdvJournalSync speech 100 ; sync "speech" journal space, set #LPC to 100 from default of 1000
;     gosub TM_AdvJournalScan speech VALID Find_this_text and_find_this_too ; will not advance copy of #jindex
;     gosub TM_AdvJournalScan heal VALID_ADVANCE you_heal_what that_patient_is_not ; advances pointer after scan
;     gosub TM_AdvJournalScan spellcast NONE fizzle ; no spam checking and doesn't advance #jindex copy automatically
;
;  Subs included:
;     TM_AdvJournalGetTrigger - retrieve the triggering string (#FALSE for no trigger)
;     TM_AdvJournalSync - Must call this in initialzation
;     TM_AdvJournalScan - see header for details....
;
;  Release history:
;    v1.0 - Initial release.
;    v1.1 - debug code added, fixed small possibility to miss an incoming journal entry
;
;  Requirements:
;    Nothing special
;
;  Credit:
;    If you use these subs, please leave credit for me somewhere in your script header.
;=================================================================
;------------------------  Call interface  -----------------------
;=================================================================
set !TM_FunctionCalled #FALSE
if %0 = 1
  gosub %1
if %0 = 2
  gosub %1 %2
if %0 = 3
  gosub %1 %2 %3
if %0 = 4
  gosub %1 %2 %3 %4
if %0 = 5
  gosub %1 %2 %3 %4 %5
if %0 = 6
  gosub %1 %2 %3 %4 %5 %6
if %0 = 7
  gosub %1 %2 %3 %4 %5 %6 %7
if %0 = 8
  gosub %1 %2 %3 %4 %5 %6 %7 %8
if %0 = 9
  gosub %1 %2 %3 %4 %5 %6 %7 %8 %9
if %0 = 10
  gosub %1 %2 %3 %4 %5 %6 %7 %8 %9 %10
if %0 = 11
  gosub %1 %2 %3 %4 %5 %6 %7 %8 %9 %10 %11
if %0 > 11
{
  display ok Too many arguments for "call", edit file.
  stop
}

if !TM_FunctionCalled = #TRUE
  exit
if %0 = N/A
  display ok You may not run this script directly.
else
  display ok Function " , %1 , " not found.
stop
;-------------------------------------------------------------------------------
sub TM_AdvJournalGetTrigger
  namespace push
  namespace local TM_AdvJS_ , %1
  set #RESULT !trigger
  namespace pop
  set !TM_FunctionCalled #TRUE
return #RESULT
;-------------------------------------------------------------------------------
; %1 - Journal Name
; %2 - #LPC setting (optional)
; Brings !_jindex up to the most recent #journal entry
sub TM_AdvJournalSync
  namespace push
  namespace local TM_AdvJS_ , %1
  set !_jindex #jindex + 1
  if %0 > 1
    set !lpc_set %2
  namespace pop
  set !TM_FunctionCalled #TRUE
return
;-------------------------------------------------------------------------------
; %1 - Journal Name
; %2 - NONE, ADVANCE , ( _VALID ) - advances jindex pointer, anything else
; %3, %4, %5, etc strings to match
; returns #TRUE for match, #FALSE for no match
;  Will not advance !_jindex pointer to allow for scanning journal history for more than one search.
;  Also searches for : , #SPC in journal entry to be sure someone isn't spamming the text
;  About %2 arguments:
;    NONE: defaults to basic journal scan (no SPAM checking, no #jindex pointer copy advancing)
;    ADVANCE: no spam checking, advances #jindex copy
;    VALID: invokes SPAM filtering, no advance of #jindex copy
;    VALID_ADVANCE, VALIDADVANCE, ADVANCE_VALID, etc.: invokes SPAM filtering, advances of #jindex copy
sub TM_AdvJournalScan
  namespace push
  namespace local TM_AdvJS_ , %1
  set !args %2
  set !temp_lpc #LPC
  if !lpc_set = N/A
    set #LPC 1000
  else
    set #LPC !lpc_set
  set !num_args %0
  set !first_arg 3
  set !sampled_jindex #JINDEX
  if !_jindex = N/A
    set !_jindex !sampled_jindex
  if !charname = N/A
  {
    set !charname #CHARNAME
AdvJournalScan_loop1:
    str pos !charname #SPC
    if #STRRES <> 0
    {
      set !val #STRRES - 1
      str left !charname !val
      set !left #STRRES
      set !val !val + 1
      str del !charname 1 !val
      set !charname !left , _ , #STRRES
      goto AdvJournalScan_loop1
    }
  }
  set !index !first_arg
  repeat
    set !temp_jindex !_jindex
    set !text % . !index
    while !temp_jindex <= !sampled_jindex
    {
      scanjournal !temp_jindex
      str pos #JOURNAL !charname 1
      set !namepos #STRRES
      str count #JOURNAL !charname
      set !namecnt #STRRES
      str pos #JOURNAL :_ 1
      set !smcpos #STRRES
      str pos #JOURNAL !text 1
      set !textpos #STRRES
      if !textpos < !smcpos && !smcpos <> 0 || !smcpos = 1 || :_ notin #JOURNAL || VALID notin !args
        set !pass #TRUE
      else
        set !pass #FALSE
      if ( !text in #journal && ( ( !namepos = 1 && !namecnt <= 1 ) || !pass ) )
      {
        set !temp_jindex !temp_jindex + 1
        if ADVANCE in !args
          set !_jindex !temp_jindex
        set !trigger !text
        set #LPC !temp_lpc
        namespace pop
        set !TM_FunctionCalled #TRUE
        return #TRUE
      }
      set !temp_jindex !temp_jindex + 1
    }
    set !index !index + 1
  until !index - !first_arg > !num_args - !first_arg
  set %10 !sampled_jindex - !_jindex
  set %10 %1 , _ , %10 ; for debugging purposes
  set #LPC !temp_lpc
  set TM_AdvJournalGetTrigger #FALSE
  namespace pop
  set !TM_FunctionCalled #TRUE
return #FALSE
;-------------------------------------------------------------------------------



Thanks for the script. Works well. I added in ignore #findrep 2-7 because of all the invulnerable npc's now.
    if #findrep = 2_3_4_5_6_7
       ignoreitem #FINDID

I just used madd dog's jhelom trainer, and I had no issues with the script. Another guy was there running the same script. If it's crashing, did you kill something in the pen before starting? One hot tip, trining from 50 to 70 is slowwwww. If you jewel up to tame bulls, then take them off, you gain much faster to 90.