Is there a way to automatically/batch extract image files from formatted notes? I’d like to batch convert the image contents of multiple files into PDF then run OCR.
If you convert the whole note to PDF the images are “printed” onto a page or multiple pages which isn’t ideal.
That’s not possible currently, neither via the interface nor via AppleScript.
Many thanks for swift response.
Any third party options / work arounds you can think of?
A script converting the notes to rich text first should actually work:
tell application id "DNtp"
set theSelection to the selection
set tmpFolder to path to temporary items
set tmpPath to POSIX path of tmpFolder
repeat with theRecord in theSelection
try
if type of theRecord is formatted note then
set theRTF to convert record theRecord to rich
if type of theRTF is rtfd then
set thePath to path of theRTF
set theGroup to parent 1 of theRecord
tell application "Finder"
set filelist to every file in ((POSIX file thePath) as alias)
repeat with theFile in filelist
set theAttachment to POSIX path of (theFile as string)
if theAttachment does not end with ".rtf" then
-- Importing skips files inside the database package,
-- therefore let's move them to a temporary folder first
set theAttachment to move ((POSIX file theAttachment) as alias) to tmpFolder with replacing
set theAttachment to POSIX path of (theAttachment as string)
tell application id "DNtp" to import theAttachment to theGroup
end if
end repeat
end tell
end if
delete record theRTF
end if
end try
end repeat
end tell
1 Like
Amazing. Thank you!
Any way to name the extracted images after the original file name sequentially?
Thanks for posting this script which I thought might help me.
Unfortunately when applying I get the error message "on performSmartRule (Expected end of line, etc.but found end of script.)
I’m trying to separate mixed items then have them grouped from merged notes imported from Evernote into DEVONthink Pro. These items are mostly JPG’s and PDF’s.
Is there a way the script could be altered to run correctly?
The above script doesn’t contain this line actually. Where & how exactly did you use the script’s code?
Hi, Thanks for replying.
I made a smart rule and applied the script to a test folder containing four html and rtfd files.
The html files are the result of an import from Evernote and the rtfd files are the result of a test conversion with DEVONthink.
My problem is that I have thousands of mixed notes which I have imported from Evernote.
Each of these notes is a merged note containing JPG’s and PDF’s and a text timestamp.
I would like to extract the items, group them and then make the items searchable.
I am able to extract the items successfully by exporting from DEVONthink to and external folder but of course they ar not grouped.
Is there a way I could automate the extraction, grouping and OCR within DEVONthink?
If so this could be a useful solution for others migrating from Evernote.
The source of the smart rule’s script doesn’t seem to be valid and causes the error message.
Does this mean that I have set the rule incorrectly?
Below is the Smart Rule I created with the script you provided earlier in the thread.
To try and avoid the source problem I created a new folder and a single RTFD file to test the script.
The error message I now receive is "on performSmart Rule (Error -1708)”
Is there something simple that I could be doing wrong?
Thanks for helping me.
The syntax of the script used by the Execute Script action is not correct according to the error. Press the Edit Script… button and then the small preview button in the lower left corner to check the syntax.
Thanks. I found the below which is a copy and paste of the Smart Rule posted earlier. Should I be changing parts of it to suit my situation? I’m using DEVONthink Pro 3.8.
tell application id “DNtp”
set theSelection to the selection
set tmpFolder to path to temporary items
set tmpPath to POSIX path of tmpFolder
repeat with theRecord in theSelection
try
if type of theRecord is formatted note then
set theRTF to convert record theRecord to rich
if type of theRTF is rtfd then
set thePath to path of theRTF
set theGroup to parent 1 of theRecord
tell application "Finder"
set filelist to every file in ((POSIX file thePath) as alias)
repeat with theFile in filelist
set theAttachment to POSIX path of (theFile as string)
if theAttachment does not end with ".rtf" then
-- Importing skips files inside the database package,
-- therefore let's move them to a temporary folder first
set theAttachment to move ((POSIX file theAttachment) as alias) to tmpFolder with replacing
set theAttachment to POSIX path of (theAttachment as string)
tell application id "DNtp" to import theAttachment to theGroup
end if
end repeat
end tell
end if
delete record theRTF
end if
end try
end repeat
end tell
1 Like
You can’t use it like that in a smart rule. Please read the documentation, section “Automation”. Make sure you include your script in a on performSmartRule handler. In this context, accessing the selection makes no sense because the handler receives the “selected” records (i.e. those matching the smart rule’s criteria) as a parameter.
If you search around in this forum, you’ll find a ton of examples for smart rule scripts that you can use as a starting point.
Slightly off-topic: Since images are apparently included as data URLs in a formatted note, it should be possible to extract them using JavaScript (or AppleScript, but I wouldn’t want to try that 
These data URLs could then be converted to proper images with online tools (your favorite search engine will tell you more about those).
What exactly is it that you’re trying to achieve?
My EN workflow over the years involved scanning, (with a Fujitsu ScanSnap), documents individually as JPG files then merging them to create an EN note. I then relied on EN’s servers for searching.
On importing my EN files (ENEX files via Import > Files and Folders not DT’s built in feature) I’m left with numerous HTML documents in DT with the only searchable text being the timestamp created by the scanner.
Additionally, the HTML documents have other files “embedded” eg. Searchable PDF’s, XLS files etc.
My aim was to use a smart rule to extract and group the individual files then OCR the non-searchable documents
My current solution involves converting the HTML files to RTFD then exporting (Export > As Website) to an external folder which separates the embedded files successfully. I’ll then import the files into DT, sort the relevant non-searchable files via a smart folder and perform OCR on them. The searchable files won’t be grouped but this should be acceptable for my purposes. I might experiment with DT’s “Group Similar Items” a little more but have found the results a little hit and miss.
I’m quite comfortable scanning and importing into DT and will make sure that everything is searchable going forward!
Thanks again for your help and have a good weekend.
In case anyone is interested in this topic, I whipped up an example JavaScript function (using some ObjC, too) to do just this: Extract embedded files from Evernote (or possibly formatted) notes.
1 Like
Very nice. Thanks for sharing it 
Thanks! Certainly helped reducing the steps required for my migration and may help others with the same problem 