Using convert image in JavaScript

DEVONthink Automation
1

After reading some older posts here on the AppleScript method convert image, I wrote a tiny little thingy in JavaScript:

function performsmartrule(records) {
	var app = Application("DEVONthink 3");
	app.includeStandardAdditions = true;

	records.forEach (r => {
	  if (r.wordCount() === 0) {
 	    const newRecord = app.convertImage (r, {record: r,
			type: "PDF document", waitingForReply: false});
       }
	})
}

which does nothing but produce weird error messages:

  • If I leave out the first parameter of convertImage, i.e. r, I get a ā€œmissing parameterā€ error
  • The same happens if I leave out the record: r part from the second parameter
  • If I call the method as shown above, I get an Error -1728, object canā€™t be found:
app = Application("DEVONthink 3")
	app.selectedRecords()
		--> [app.databases.byId(1).contents.byId(186284)]
	app.databases.byId(1).contents.byId(186284).wordCount()
		--> 0
	app.convertImage(app.databases.byId(1).contents.byId(186284), {record:app.databases.byId(1).contents.byId(186284), type:"PDF document", waitingForReply:false})
		--> Error -1728: Objekt kann nicht gefunden werden.

From the older posts, I concluded that in AppleScript the method would be called as
convert image record r, without an unnamed first parameter (which goes against the description in the scripting library).

Bonus discovery: using app.ocr(r.path(), { file: r.path(), waitingForReply: true}); at least works, but also insists on using the fileā€™s path twice. And of course it does not remove the original file after OCRing. On the other side, I donā€™t know if convertImage would remove it, were it working as advertised.

Why do I even bother? Until recently, I had to use Hazel to watch a folder for incoming documents and change their names according to my wishes, since DT could not run JavaScript inside a smart rule. Now it can, and Iā€™d like to get rid of the Hazel step, in analogy to

2

I think part of the problem is that there shouldnā€™t be an unnamed parameter to any of the OCR commands. Whilst the automatically generated documentation in Script Editor shows that parameter it is not as far as I can see defined internally in the application.

Thanks for the full description, I will take a look.

1 Like