AutoHotkey Community

Loop, Read, ListA.txt
{
   CurrentLine := A_LoopReadLine
   Loop, Read, ListB.txt, NewList.txt
   {
      If (InStr(A_LoopReadLine, CurrentLine))
         FileAppend, %A_LoopReadLine%`n
   }
}

list1=
(
about adv., prep.
above prep., adv.
abroad adv.
absence n.
absent adj.
)
list2=
(
about prep S1, W1
about adv S1, W1
above adv, prep S2, W1
above adj W3
abroad adv S2, W3
absence n S3, W2
absolute adj S2, W3
)
obj1:=Object()
obj2:=Object()
While A_Index<3    ,    idx:=A_Index
   Loop,Parse,List%A_Index%,`n,`r
   {
      obj%idx%[(in:=InStr(A_LoopField," "))?SubStr(A_LoopField,1,in-1):A_LoopField]:=A_LoopField
   }
for k in obj1
   If !obj2[k]
      out.="List 1: " k "`n"
for k in obj2
   If !obj1[k]
      out.="List 2: " k "`n"
MsgBox % out

#NoEnv
#SingleInstance force
SetBatchLines -1

listA =
(
about adv., prep.
above prep., adv.
abroad adv.
absence n.
absent adj.
)

listB =
(
about prep S1, W1
about adv S1, W1
above adv, prep S2, W1
above adj W3
abroad adv S2, W3
absence n S3, W2
absolute adj S2, W3
)

listA := TidyList(listA)
listB := TidyList(listB)

; add items from listA which are not in listB
Loop, Parse, listA, `,
  { if A_LoopField not in %listB%
      result .= A_LoopField ","
  }
; add items from listB which are not in listA
Loop, Parse, listB, `,
  { if A_LoopField not in %listA%
      result .= A_LoopField ","
  }
StringTrimRight, result, result, 1   ; remove trailing comma

Msgbox % result

ExitApp

; ===============================

; get a comma-separated list of just the words
TidyList(list) {
; Note: Could use VarSetCapacity() here to boost performance?
Loop, Parse, list, `n, `r
  { ; get rid of the clutter - we just want the word
    StringLeft, word, A_LoopField, InStr(A_LoopField, A_Space)-1
    tidyList .= word ","
  } StringTrimRight, tidyList, tidyList, 1   ; remove trailing comma
; remove dupliates - this should help performance I would think
Sort, tidyList, U D,
return tidyList
}

Loop, Read, FileA.txt
{
   unique = 1
   RegExMatch(A_LoopReadLine, "^\w+", FileAWord)
   
   Loop, Read, FileB.txt
   {
   RegExMatch(A_LoopReadLine, "^\w+", FileBWord)
   
      If(FileBWord == FileAWord)
      {
         unique = 0
      }
   }
   
   If(unique == 1)
   {
      FileAppend, %A_LoopReadLine%`n, Output.txt
   }   
}

#NoEnv
#SingleInstance force
SetBatchLines -1

listA =
(
about adv., prep.
above prep., adv.
abroad adv.
absence n.
absent adj.
all right adj., adv., exclamation
businessman, businesswoman n.
)

listB =
(
about prep S1, W1
about adv S1, W1
above adv, prep S2, W1
above adj W3
abroad adv S2, W3
absence n S3, W2
absolute adj S2, W3
good morning interjection S2
good night interjection
)

listA := TidyList(listA)
listB := TidyList(listB)

; add items from listA which are not in listB
Loop, Parse, listA, `,
  { if A_LoopField not in %listB%
      result .= A_LoopField ","
  }
; add items from listB which are not in listA
Loop, Parse, listB, `,
  { if A_LoopField not in %listA%
      result .= A_LoopField ","
  }
StringTrimRight, result, result, 1   ; remove trailing comma

Msgbox % result

ExitApp

; ===============================

; get a comma-separated list of just the words
TidyList(list) {
static noise := "adv|prep|n|adj|exclamation|interjection"
; Note: Could use VarSetCapacity() here to boost performance?
Loop, Parse, list, `n, `r
  { ; get rid of the clutter - we just want the word
    ; regex: keep matching chars a-z, spaces and commas, until we find
    ; a "noise" sequence followed by a comma, period or space
    haystack := A_LoopField . " "  ; indicate end of line
    if !RegExMatch(haystack, "iU)^([a-z ,]+) (" noise ")[,. ]", m)
      msgbox ERROR - No match for: %A_LoopField%
    words := m1
    ; e.g. "businessman, businesswoman" becomes 2 separate entries
    Loop, Parse, words, `,
      { word = %A_LoopField%
        tidyList .= word ","
      }
  } StringTrimRight, tidyList, tidyList, 1   ; remove trailing comma
; remove dupliates - this should help performance I would think
Sort, tidyList, U D,
return tidyList
}