Noodlesoft Forums

How do I get Hazel to tag files for devonthink?

You'll need to use Automator or AppleScript to interface with Devonthink. Search the forums as someone probably has already done that.

I did have a look but it was like jumping into the deep end. Please could you point me in the right direction?

Have you used Automator before? If not, I suggest playing with that outside of Hazel. Not using DevonTHINK, I don't know if it has Automator actions but if it does, that would be a place to start.

abetterhope wrote:How do I get Hazel to tag files for devonthink?

1) What version of DevonThink do you have?
2) What exactly do you want to do? (a) Tag files that are already in DevonThink, (b) Tag files before importing them into DevonThink or (c) Both these things?
3) Have you asked the guys in the DevonThink Forum? (http://forum.devontechnologies.com/index.php)

I have just found this in the DevonThink Forum which might be of interest:
http://forum.devontechnologies.com/view ... ng#p115095

I'm the "Hugh" who asked the question referenced by Allsop above. I've absolutely no inside knowledge. But, as my up-to-date copy of DevonThink Pro Office is Version 2.9.11, if I had to guess I'd say that the "actions" promised in Christian Grunenberg's answers to me are likely to arrive at some point in Version 3.x - in other words, relatively soon (although I've seen no sign yet of betas being tested).

However, it sounds as if Hazel is unlikely to be involved in those "actions". And Hazel can't effectively, currently, "see" into DT's databases. But the promised DT "actions" won't necessarily provide the only way to respond to the OP's wish for automation to be used to tag files for DevonThink. Hazel could still play a role, acting outside DT's databases.

DevonThink is capable of transforming "Finder" tags into the tags that it uses internally. So Hazel could be used to tag files with "Finder" tags externally before they are imported into, or indexed by, DevonThink; the tags would carry over.

In the simplest case, with one DT database and files that already contain text, a folder could be created outside DevonThink that is monitored by Hazel. A series of Hazel rules could be created in order to tag files placed in the folder depending on their content, and then move or copy them into the Inbox of the DT database, or into folder(s) outside DT but indexed by DT. As I say, the tags would carry over.

In a slightly more complex case, where some files may contain no text and need OCR-ing, something like steps 1 and 2 of this procedure could be used to OCR the files first (so that Hazel could search their content and apply tags accordingly).

Moving up the tree of complexity, with more than one DT database, what I'd do is to pair each database with a newly created Hazel-monitored folder to carry out the tagging as above, with a single separate Hazel-monitored Inbox folder with Hazel rules to parcel out incoming files to the database-paired folders as appropriate.

As you can see, the system of folders quickly becomes quite complex. There may be simpler ways (probably involving scripting: there are AppleScripts for selecting particular DT databases elsewhere on the Web).

The highest layer of complexity would involve Hazel placing the tagged files not in the databases' Inbox folders, but in named folders, or "groups" as DT calls them, within the databases. That's when you'd almost certainly need Hazel with DevonThink's Automator rules (if you have DevonThink and have installed its "Extras", open up Automator and they should already be there), or Hazel with Applescripts (there is one set of such scripts further down the thread I referred to above), or, in the final analysis, DevonThink's promised "actions". And if your use-case is as complicated as this, possibly it would be worth waiting for them!