Devonthink/PDFExpert.scpt

(*
Batch-OCR selected PDFs in DEVONthink using PDF Expert
Originally by Alex Wellerstein – adapted by ChatGPT, May 2025
Released to the public domain: absolutely no warranties, guarantees, promises, *anything*.
Tested on macOS 14.6.1 • DEVONthink 3.9.x • PDF Expert 3.10.10
*)

------------------------------------------------------------
-- 1. COLLECT THE SELECTION FROM DEVONTHINK
------------------------------------------------------------
tell application id "DNtp" -- DEVONthink’s four-letter bundle ID
	activate
	set theSelection to selection
end tell

if theSelection is {} then
	display dialog "Nothing is selected in DEVONthink." buttons {"Cancel"} default button 1 with icon caution
	return
end if

-- Build a simple AppleScript list of POSIX paths for any *PDF* items.
set theFiles to {}
repeat with rec in theSelection
	try
		tell application id "DNtp"
			if type of rec is PDF document then -- skip anything that isn’t a PDF
				set end of theFiles to POSIX path of (path of rec as string)
			end if
		end tell
	end try
end repeat

if theFiles is {} then
	display dialog "None of the selected items is a PDF." buttons {"Cancel"} default button 1 with icon caution
	return
end if

------------------------------------------------------------
-- 2. SET UP A PROGRESS BAR
------------------------------------------------------------
set totalFiles to length of theFiles
set progress total steps to totalFiles
set progress completed steps to 0
set progress description to "Processing PDFs with PDF Expert…"
set progress additional description to "Starting…"

------------------------------------------------------------
-- 3. LOOP THROUGH EACH PDF AND RUN YOUR ORIGINAL OCR ROUTINE
------------------------------------------------------------
repeat with i from 1 to totalFiles
	set thePath to item i of theFiles
	set progress completed steps to i
	set progress additional description to "Processing " & i & " of " & totalFiles & "…"
	
	my pdfexpert_ocr(thePath) -- run your existing handler
	
	-- Refresh the record inside DEVONthink so its thumbnails & metadata update
	try
		tell application id "DNtp"
			set recList to lookup records with path thePath -- returns a list (usually length 1)
			repeat with r in recList
				update record r without asking -- silent refresh
			end repeat
		end tell
	end try
	
	-- Free up RAM every ten files (same logic you had)
	if i mod 10 is 0 then
		tell application "PDF Expert" to quit
		delay 1
	end if
end repeat

log "All done!"
display dialog "OCR finished for " & totalFiles & " PDF(s)." buttons {"OK"} default button 1

--------------------------------------------------------------------------------
-- ↓↓↓  EVERYTHING BELOW HERE IS YOUR ORIGINAL CODE, UNCHANGED  ↓↓↓
--------------------------------------------------------------------------------
on pdfexpert_ocr(filePath)
	log ("Starting OCR for " & filePath)
	do shell script "open " & quoted form of filePath -- assumes PDF Expert is the default PDF viewer
	delay 0.5
	activate application "PDF Expert"
	delay 0.5
	set fileTitle to getFilename(filePath, ".pdf")
	tell application "System Events"
		tell process "PDF Expert"
			
			-- locate the window that matches the file name
			set win to false
			repeat with w in every window
				if (title of w) is fileTitle then set win to w
			end repeat
			
			if (win is false) then
				delay 2 -- give PDF Expert a moment more
				repeat with w in every window
					if (title of w) is fileTitle then set win to w
				end repeat
			end if
			
			if (win is false) then
				display dialog "Could not find window for “" & fileTitle & "” in PDF Expert." buttons {"Cancel"} default button 1 with icon stop
				error number -128
			end if
			
			-- click “Recognize Text” in Scan & OCR
			click my getElementByClassAndName("Recognize Text", false, entire contents of win)
			
			-- click “Recognize…” in the side bar
			tell my getElementByClassAndName("Recognize...", false, entire contents of win)
				click
				-- choose “All” pages when the dialog appears (if >1 page)
				if exists radio button "All" of sheet 1 of win then
					click radio button "All" of sheet 1 of win
					click button "Apply" of sheet 1 of win
				end if
			end tell
			
			delay 1 -- short head-start before polling
			
			repeat -- poll until the progress sheet disappears
				try
					if not (exists win) or (count sheet of win) = 0 then exit repeat
					if not (exists progress indicator 1 of sheet 1 of win) then exit repeat
				on error
					exit repeat -- any UI error = assume finished
				end try
			end repeat
			
			delay 1 -- extra cushion
			
			-- save & close
			set fileMenu to menu "File" of menu bar item "File" of menu bar 1
			click my getElementByClassAndName("Save", false, entire contents of fileMenu)
			delay 1
			if my isSaved(front window) then
				click my getElementByClassAndName("Close Window", false, entire contents of fileMenu)
			else
				log "WARNING: file might not have saved!"
			end if
		end tell
	end tell
end pdfexpert_ocr

-- utility handlers (unchanged from your original)
on getFilename(theFile, theExtension)
	tell application "Finder" to set fName to name of (POSIX file theFile as alias)
	if (theExtension is not "") then
		return trimText(fName, theExtension, "end")
	else
		return fName
	end if
end getFilename

on trimText(theText, theCharactersToTrim, theTrimDirection)
	set theTrimLength to length of theCharactersToTrim
	if theTrimDirection is in {"beginning", "both"} then
		repeat while theText begins with theCharactersToTrim
			try
				set theText to characters (theTrimLength + 1) thru -1 of theText as string
			on error
				return ""
			end try
		end repeat
	end if
	if theTrimDirection is in {"end", "both"} then
		repeat while theText ends with theCharactersToTrim
			try
				set theText to characters 1 thru -(theTrimLength + 1) of theText as string
			on error
				return ""
			end try
		end repeat
	end if
	return theText
end trimText

on getElementByClassAndName(targetName, targetClass, targetElements)
	repeat with uiElem in targetElements
		try
			if (targetClass is false and name of uiElem is targetName) ¬
				or (name of uiElem is targetName and class of uiElem is targetClass) then ¬
				return uiElem
		end try
	end repeat
	return false
end getElementByClassAndName

on isSaved(win)
	try
		set winName to name of win
		if winName begins with "*" then return false
	end try
	return true
end isSaved