Problems and error messages when importing OCR

I have two problems when scanning and importing documents with Import -> From Scanner - With Capture

1 - sometimes, the first page looks fine at the preview and immediately after scanning, but then after OCR is performed, the page looks like it has been stretched enormously vertically. The text is unreadable, looks like barcodes. I usually just hit Add Page and it scans the page again, and this time its OK. Then I delete the first. It only happens to the first page, all successive pages are fine.

2 - after I have added the scanned document to the database (final step), I always get a log window up with a “Failed” message. but everything looks fine, the document is scanned OK, and it is available in the database. what is the error message trying to tell me? here is an example, I saved the log file, here is the contents from the last scan:

10:47:44: /var/tmp/folders.501/TemporaryItems/Playfilms Avtale.pdf Failed

Please note; problem 1 and 2 are independent from each other (2 can happen if 1 does not)

I am using Devonthink Pro office 1.3.1, OSX 10.4.9 and a Canon MP 600

Re. 1: I have no idea why that happens just the first time. It looks like an intermittent bug in the IRIS OCR program.

Re. 2: That is harmless, for some reason the PDFKit needs multiple tries to properly interpret your document. If it really failed, you would get a warning panel with some useful options.

Thanks annard, good to know 2) it’s not serious. I will just disregard the log window.

Would you like me to send you an example of a stretched scan, or should I just consider it unfixable for now… Its not a big problem except a re-scan takes some time, if it happens a lot it can get annoying.

It’s always good to send me a sample, both of the scan before and after. (So you’d have to scan it again without OCR enabled so we get the “raw” data as well.)