FileMaker to DT Pro

I recently had some Support queries from someone who wants to export a large number of records in a FileMaker database to DT Pro. There have been discussions in the forum on how to do that, involving intermediate steps with Word, or using scripts. Those seem to have worked for some users.

But I was looking for tools already available in DT Pro. Here’s an approach that results in capturing individual documents per record.

[1] Export the database from FileMaker as a tab-delimited text file. Change the extension to ‘.tsv’.

[2] Import the .tsv file into DT Pro using File > Import > Files & Folders. The result will be a Sheet containing multiple records. Open the Sheet in its own window and delete any unnecessary columns, and perhaps edit column headers if desired.

[3] Select the Sheet and export by choosing File > Export > as OmniOutliner document. (See, that older .outliner file type can be useful.)

[4] Open and update the exported file under OmniOutliner 3. (A freebie on many Macs.)

[5] Export the oo3 file as OPML.

[6] Import that OPML file into DT Pro using File > Import > Files & Folders.

Result: A new group has been created, and each record is now an individual document. The names may or may not be meaningless, depending on the content of the first column in the Sheet.

Did you realize, by the way, that you can change the order of columns in the Sheet? Or change the name of the Sheet to an appropriate name for the first column?

This approach is of course general for splitting up any .tsv or .csy text file into multiple documents in your database. Perhaps not perfect for all needs, but often useful. No scripting required.


this is wonderful, thank you!

And rest a bit over Christmas!


I’m an historian with a FIleMaker database of archival notes (it has 750 records, one record per archival folder; all are text-only, so the total database is just 8.5 MB in size).

I’ve switched to DTPO2 and would like to import this FM database, ideally keeping each record separate. My FM field headings include several short fields (name of the archive, exact citation for the folder’s location, number of pages in the folder, title of the folder, date I read it, etc.) and then one long text field (for notes on the individual documents within the folder).

I’ve done a search of the forums for relevant threads, and this is the most relevant — but it’s five years old, hence wanted to check that the technique advised is still feasible/advisable.

My FM file is now in fp7 format, although I have access to version 11 and could upgrade the format if it matters.

In DEVONthink 2.x, .tsv or .csv files imported as Sheets no longer contain rows as individual Record files, as was the case in DEVONthink 1.x.

Thanks very much, Bill. Is there another technique you would recommend? Shall I simply export the FM database as csv or Excel or pdf, and then import the resulting file(s) to DTPO?

To keep the archival folders separate, I could do this manually for each record, although obviously that’s a pain. Unfortunately I’m new to this; specifically, I’m not versed in Applescript. I suppose I should treat this task as a spur to change that Applescript ignorance?, since I assume it would easily automate the process of exporting/importing the hundreds of records.