Metadata after OCR in DT3b3

Also no change on my part after complete restart…

After further checking different avenues of importing documents in DT3b3 (eg. printing a PDF and importing directly) it has become clear to me that the issue is tied to the way ScanSnap Home treats the scanned document.
Will investigate further.

The only hint relevant to ScanSan I have found in the system log after a scan procedure that I have found indicates the following (and I am not sure whether there is pointer to DT…):

ul 17 10:34:25 iMac-Gert ScanSnapHomeMain[7032]: objc[7032]: Class FIFinderSyncExtensionHost is implemented in both /System/Library/PrivateFrameworks/FinderKit.framework/Versions/A/FinderKit (0x7fffa40163d8) and /System/Library/PrivateFrameworks/FileProvider.framework/OverrideBundles/FinderSyncCollaborationFileProviderOverride.bundle/Contents/MacOS/FinderSyncCollaborationFileProviderOverride (0x11484df50). One of the two will be used. Which one is undefined.
Jul 17 10:34:28 iMac-Gert ScanSnapHomeMain[7032]: assertion failed: 18F132: libxpc.dylib + 90677 [9A0FFA79-082F-3293-BF49-63976B073B74]: 0x89

I’m not using ScanSnap at all. Using ExactScanPro.

I just updated to DT3b4 and the bug is still there: it replaces whatever title/name text I’ve entered with text grabbed from near the beginning of the scanned document. I’m using a ScanSnap iX500.

I don’t have a Scansnap but I just searched on line and note that SN’s organiser quick menu has a feature which can rename the pdf automatically. Might be worth looking at.

Same here, upgraded to DT3b4. I’m using ExactScanPro and not using ScanSnap software, and problem persists. I save a copy of the scan outside of DT3 and it keeps the name ExactScanPro gives it.

@mog: This feature is also available in the ScanSnap software for macOS, and I’ve been using it successfully in DT2–and up through DT3b2–to rename my files immediately after scanning. The problem is that DT3b3+ is overriding that naming downstream of the import process and OCR, using the first few words in the OCR’d document to replace whatever Title/Name had originally been entered in ScanSnap dialog.

I have the same problem, but described it less well.

I was able to correct this problem by uninstalling DT3 (now at DT3b5)–using the instructions on page 6 of the DT3 user manual–then creating a new user account (for testing) and reinstalling DT3b5 in the new account. Following that, the new filenames were once again preserved throughout the import and OCR conversion process.

Perhaps there may have been some adverse interaction caused by installing DT3 alongside DTPO (v2.11.3) in the same account. However I’ve no idea why it would begin manifesting after updating DT3 to beta3.

There was a smart rule downloading automatically that was causing the behavior. That was disabled in public beta 5.

But I’m glad to see you’re running DEVONthink 3 on a test account.

I’m on beta 6 and am still having this problem. Is there a smart rule that I should try to disable to prevent this from happening?

I deleted all smart rules, after which everything worked perfectly…

Control-click the Incoming Scans smart rule and either Remove it or Edit it and set the event trigger to On demand.