As I’ve recently discovered Scrivener don’t accept the e-mail files. The emails need to be edited in a pdf file before.

So I was wondering if it couldn’t be possible to get Scrivener a little bit Automator scriptable. That allows the user to creat an Automator service which it make automatically the different actions: edit a pdf file of the email, and then import the pdf into the blinder.

Of course many other functions could be make with Automator actions.

Automator is unfortunately a bit limited in what it can accomplish, being a subset of what applications can do with Applescript (Scrivener doesn’t have an Applescript dictionary for various reasons, but might some day) and not much else. I used to try and use it to automate, but found that general-purpose tools work a bit better. Keyboard Maestro is the one that I settled on. It lets me automate Scrivener and just about every other program, thanks to its wide variety of tools that can trigger buttons, menus and even manipulate the keyboard and mouse. With a program like Scrivener which has a huge amount of keyboard accessibility, you can do a great deal! I even have scripts that automatically fill in custom meta-data fields based on found content in the editor. :slight_smile:

At any rate, as noted, supported Automator means supporting Applescript, which is all a great idea, but we don’t have the resources to do it at this time.

Ooh, intriguing. Can you expand on that a bit?

I’m kinda doing something similar, but at the moment I have KM clicking into the Editor, move cursor over X number of spaces/words, select next three characters, copy, paste into meta-data, next document, repeat. But if I could get that happening a bit more automagically it would be great.

…I really need to set aside some time to play with KM a bit more, got some ideas, now it’s just a matter of testing/implementing.

Ha, well, it’s not much more than that for me as well. I do use Find to get around in the document and then arrow keys if necessary to select and copy text. Since I use MultiMarkdown, meta-data I want make more visible is pretty easy to work with since they all follow the “Key:{Spaces}Value” approach. I can search for “Key:”, then use the right number of keys to select the value part.

Something I’d like to figure out is a way to reliable select regular expressions mode in the Find panel. Tracking mouse position and click and drag to manipulate the option tool only has spotty success for me, for some reason—it seems to equate “relative to frontmost window” with the main project window in some cases, but in other cases it properly identifies the Find palette as the frontmost. If I could get regex working, that would eliminate much of the cursoring around and resulting fragility.