ABBYY Finereader And Snow Leopard – File Not Created With ScanSnap

One issue with the Fujitsu ScanSnap and OSX 10.6 Snow Leopard that I forgot to mention the other day is the ABBYY FineReader that comes bundled with it.

When scanning with the version of Finereader that ships with the ScanSnap S510M and S1500M, you may get an error message like “File not created with ScanSnap”.

This is a known issue and according to this bulletin from Fujitsu Support, it will be fixed “within 2009″.

Fujitsu has assured me that they’re working on it, so hopefully we’re not talking December 31 here!

I personally do not use FineReader.. anyone have any workarounds for the Snow Leopard issue that they use? Leave a note in the comments.

Update: Thanks to reader Spike in the comments for the tip, ABBYY has released a version of FineReader Express Edition that supports Snow Leopard. More info here.

Update #2 Nov 19/09: The ABBYY FineReader for ScanSnap Snow Leopard Update is now available.

Tags: , , , , , , ,

50 Responses to “ABBYY Finereader And Snow Leopard – File Not Created With ScanSnap”

  1. Michael F August 31, 2009 at 2:53 pm #

    It would be interesting to see whether the problem is with Scansnap Manager (not setting the PDF metadata "Creator" field properly under Snow Leopard) or with Finereader (not reading the field properly from the PDF under Snow Leopard).

    A good test of case 2 would be to take a PDF scanned under Leopard, and try dropping it onto the "Scan to OCR" app in the "Finereader for Scansnap" folder (in Snow Leopard), and see what happens.

    Another test (case 1) would be to scan a PDF under Snow Leopard, then copy it to a Leopard machine and drop it onto "Scan to OCR".

    This bug is personally my one showstopper in upgrading to Snow Leopard — if anyone has already upgraded and would be interested in troubleshooting with me, leave a comment and we can be in touch privately.

    • jordanbrock September 6, 2009 at 5:03 am #

      I have done some (basic) testing on this.

      A document scanned by ScanSnap under MacOSX 10.5.8 has the following details:

      Content Creator: ScanSnap Manager

      Encoding Software: Mac OS X 10.5.8 Quartz PDFContext

      It is possible to drag this document onto ABBYY FineReader and have it OCR'd.

      A scan created under MacOSX 10.6 has the following details:

      Content Creator: ScanSnap Manager

      Encoding Software: Mac OS X 10.6 Quartz PDF Context

      It is not possible to drag this onto ABBYY FineReader. So I can only presume that FineReader is checking the "Encoding Software" field and the 10.6 is causing the problem. There's no other (visible) difference.

      A strange, and annoying problem.

      • Michael F September 6, 2009 at 6:29 am #

        Hmph. And I was hoping to be able to fix this with an Automator script. But while the "Get PDF Metadata" action shows the "PDF Producer" field, the "Set PDF Metadata" action does not have access to it. As I said, Hmph.

        • Michael F September 6, 2009 at 7:05 am #

          Got it. You need to use the package "pdftk", which is available here: http://www.accesspdf.com/pdftk/

          The linked Panther installer works fine for me on Snow Leopard (go figure).

          Then, make a text file (bugfix.txt, for example) containing the following:
          InfoKey: Producer
          InfoValue: Mac OS X 10.5.8 Quartz PDFContext

          Finally, run the following command line on a PDF to fix it:

          pdftk INPUT.PDF update_info bugfix.txt output OUTPUT.PDF

          (where, obviously, you use the correct names of your input and output files).

          Confirmed to make a PDF scanned by the Scansnap under Snow Leopard OCRable in Finereader.

          Should be able to put it in an Automator workflow and run this as a service (to fix legacy 10.6 PDFs), and also turn it into an app that fixes the metadata then hands it off to Finereader. Then, you should be able to scan to this new Automator app from Scansnap Manager, and everything should work OK.

          Cool — thanks for finding that.

          • Paul Healey September 6, 2009 at 9:51 pm #

            Michael, that works great. Just one thing to add, you'll need to install Rosetta in order for pdftk to run. This can be found in the Optional Installs section of your Snow Leopard DVD. Once that it was installed, I was able to follow your instructions.

          • BrooksD September 8, 2009 at 12:55 pm #

            Michael, thanks so much for posting this!

          • Shawn September 8, 2009 at 7:30 pm #

            Is there a way to create something like this for the none Mac OSX programmer? I just want to download and have something set it up for me? Much like an actual update from ABBY would do.

          • BrooksD September 8, 2009 at 7:49 pm #

            Hi Shawn, I was talking to abbyy about this on Twitter and they are
            definitely working on the issue. Hopefully the \”real\” fix will be
            soon.

  2. Doug Ransom August 31, 2009 at 3:41 pm #

    Totally incompetent on Fujitsu ScanSnap (in their responsibility to manage a relationshiip with their suppliers) that the fix wasn't ready before Snow Leopard left beta.

    • Michael F August 31, 2009 at 9:27 pm #

      well, to be fair, everyone was led by Apple to believe that Snow Leopard would not be shipping for another month. Fujitsu may have thought their schedule was OK until 5 days beforehand…

  3. Jason Luros September 8, 2009 at 8:30 am #

    I encounter the following error dialog anytime I throw a pdf at ABBYY FineReader: Error
    Problem was encountered while loading component files. Can't proceed.

    • BrooksD September 8, 2009 at 9:02 pm #

      Hi Jason,

      Not sure about that but I know Abbyy is on Twitter (http://twitter.com/abbyy_usa) so they might be able to point you in the right direction?

      • Greta December 22, 2011 at 4:44 pm #

        A simple and itnleliegnt point, well made. Thanks!

  4. Frank September 9, 2009 at 2:15 am #

    Chalk me up as another dissatisfied user w/ the 'File Not Created with ScanSnap' error.

  5. Frank September 9, 2009 at 12:07 pm #

    I get the 'File Not Created with ScanSnap' error too. I too am surprised they didn't anticipate this during Snow Leopard beta.

  6. Paul September 10, 2009 at 1:04 pm #

    I too was screwed by this incompatibility. I told Abbyy a month or so ago about the issue and explained exactly what the problem is to them (the format of the PDF that Mac OS X generates changed slightly and their software doesn't handle it correctly). I have an alternative fix for this which I'm happy to share if you contact me directly at pmarcos@comcast.net. It involves installing what's called an "Input Manager" bundle which patches their software to avoid this issue.

    • Tony Denson October 16, 2009 at 3:36 pm #

      I would be interested in the fix for FineReader and Snow Leopard please.

      • Paul October 17, 2009 at 3:32 am #

        If you're looking for my fix, just drop me an email…

        • Ralph Elliott October 18, 2009 at 8:32 pm #

          I would be quite interested in your fix for the software

      • sznn November 14, 2009 at 6:20 pm #

        Could you please send me your fix?

    • sathish October 21, 2009 at 6:17 am #

      abbyy is not exactly coverting ocr sheet to word.can you give me the suggestions.

  7. BoilingConcrete September 12, 2009 at 2:13 am #

    I created a new Automator Service, added the "Run Shell Script" action, kept the shell at "/bin/bash", changed the Pass Input to "as arguments", then created the following shell script:

    for f in "$@"
    do
    /usr/sbin/pdfauxinfo -k "ScanSnap Manager" -i "$f" -o "$f"
    /usr/local/bin/pdftk "$f" update_info /usr/local/lib/PDFInfoString.txt output ~/Downloads/-Scripts/TempPDFtkFile.pdf
    mv -f ~/Downloads/-Scripts/TempPDFtkFile.pdf "$f"
    g="${f##*/}"
    g="${g%.*}"
    open -a 'FineReader for ScanSnap.app' "$f"
    while [ ! -e "${f%.pdf} processed by FineReader.pdf" ]; do
    sleep 30
    done
    sleep 15
    mv -f "${f%.pdf} processed by FineReader.pdf" "$f"
    /usr/local/bin/growlnotify -a "FineReader for ScanSnap" -t "FineReader for ScanSnap" -m "FineReader has finished OCRing $g"
    done

    -where pdfauxinfo is a command line utility that can set the creator/author/password/etc of a pdf. It's free from http://msyk.net/macos/pdfauxinfo/
    -where /usr/local/lib/PDFInfoString.txt is the "bugfix.txt" file Michael F described above.
    -where ~/Downloads/-Scripts/ is just some random directory into which I can temporarily store a file. PDFtk doesn't allow the output file to overwrite the input file.
    -where growlnotify is the command line version of the popular notification utility. It's free at http://growl.info/ That line can be removed if desired.

    I'm not a shell script writer, but the above script works for me. It works especially well in Snow Leopard as a Service. I can right-click on a bunch of PDFs and select "OCR PDFs with FineReader" (which I named my service) and it works like a champ.

    • Hugo October 2, 2009 at 3:52 am #

      Perfect! Thanks, just tried this – my first Automator experience – and it worked fine.

      Some notes for others: start automator, pick the "service" template. At the top of the right hand pane, alter the popups so it says:

      service receives selected *PDF files* in *Finder*

      Next, scroll down the big list of actions (that starts with "activate fonts") until you get to the "run shell script", then drag this one over into the blank right hand pane.

      Change the "pass input" to "as arguments", then select the text that appears in the box below and delete it, replacing it with the script boilingconcrete provided above.

      I don't have growl so removed the growl line at the end.

      Save the service as "OCR PDFs" or similar.

      Then, install the pdfauxinfo and pdftk from the links above, create a "-Scripts" folder in your downloads folder, and finally open a terminal and do this:

      sudo bash (then enter password)
      cd /usr/local
      mkdir lib
      echo "InfoKey: Producer" >lib/PDFInfoString.txt
      echo "InfoValue: Mac OS X 10.5.8 Quartz PDFContext" >>lib/PDFInfoString.txt

      …then close the terminal window.

      You should now, at the bottom of the right-click menu over a freshly scanned PDF, see an "OCR PDF" option. Pick this and finereader will start and then OCR the file just as it used to on 10.5.

      Now I can clear the backlog of documents I was waiting to OCR :)

      • nms24 October 3, 2009 at 1:05 pm #

        I got this to work. Thank you all very much.

        • Jason October 6, 2009 at 4:52 am #

          Is it possible for someone to post a step by step for someone very unfamiliar with Automater and Terminal?

          • Jason October 6, 2009 at 5:05 am #

            okay I tried to follow the steps above, but get this error that it will only work on PDFs created by ScanSnap (which I'm trying to use).

            Am I supposed to create a PDFINfoSTring.txt file?

            Do the files I'm trying to convert need to go into the -Scripts folder? Do I put something in there?

          • Paul October 6, 2009 at 5:15 am #

            Drop me a line at my email address above. I have a double-clickable installer package that will install a workaround for the File Not Created With ScanSnap issue. A good number of folks have installed it and it's working for them.

          • JPT October 12, 2009 at 11:10 pm #

            Is this still available to download? I would love a double click installer to make this work. I'm all thumbs on automator.

            -JP

          • Paul October 13, 2009 at 1:17 am #

            As I said above, drop me a line at pmarcos@comcast.net and I'll send you the installer package.

      • BoilingConcrete October 11, 2009 at 1:18 am #

        Hugo – Thanks for clarifying what I wanted to say!

        One note: if you want to delete the lines relating to using Growl, you can also delete the following two lines:
        g="${f##*/}"
        g="${g%.*}"
        It doesn't harm anything to keep those lines in, but it'll make things a little cleaner.

        Lastly, here's a commented version so people can see what's going on behind the scenes and allow them to tinker a little more easily:

        # For every document…
        for f in "$@"
        do

        # Tell the utility PDFAuxInfo to change the PDF's creator to "ScanSnap Manager".
        # FineReader will only work on pdfs created by ScanSnap
        /usr/sbin/pdfauxinfo -k "ScanSnap Manager" -i "$f" -o "$f"

        # Tell the PDFtk utility to add the InfoKey and InfoValue tags from the text file in the -Scripts directory
        # The PDFtk will create a new temporary PDF in your temporary directory.
        # FineReader will only work on pdfs it thinks have been created under 10.5
        /usr/local/bin/pdftk "$f" update_info /usr/local/lib/PDFInfoString.txt output ~/Downloads/-Scripts/TempPDFtkFile.pdf

        # Overwrite the original pdf with the pdf altered by the PDFTk utility.
        # This also deletes the temporary pdf
        mv -f ~/Downloads/-Scripts/TempPDFtkFile.pdf "$f"

        # Get the name of the file without the full UNIX path and store it in the variable g
        # This will just make the Growl notification easier to read
        g="${f##*/}"
        g="${g%.*}"

        # Tell FineReader to open the pdf, and thereby start OCRing it.
        open -a 'FineReader for ScanSnap.app' "$f"

        # Since FineReader is not directly scriptable, check the directory every 30 seconds and wait until
        # there is a file named "[yourPDFname] processed by FineReader.pdf". That's how you know FinderReader is done.
        while [ ! -e "${f%.pdf} processed by FineReader.pdf" ]; do
        sleep 30
        done

        # I don't remember why I wanted to wait another 15 seconds.
        # I must have had a good reason at the time. Usually this just helps handle glitches.
        # You can probably delete this without ruining Christmas (or your favorite holiday).
        sleep 15

        # Overwrite the original pdf with the one created by FineReader.
        # This deletes the original and replaces it with the OCR'd version
        mv -f "${f%.pdf} processed by FineReader.pdf" "$f"

        # Tell Growl to notify you. Use the icon from FineReader, use "FineReader for ScanSnap" as the title of the notification,
        # and use "FineReader has finished OCRing [yourPDFname]" as the message
        /usr/local/bin/growlnotify -a "FineReader for ScanSnap" -t "FineReader for ScanSnap" -m "FineReader has finished OCRing $g"

        # Exit
        done

  8. Peter Hawkes September 23, 2009 at 7:59 pm #

    According to Abbyy, Fine Reader is not compatible with Snow Leopard! They are working on a new version but until ……………..

  9. wealthychef October 15, 2009 at 5:41 am #

    How will we know when the new version is out? I found this with Google, must I keep pinging Fujitsu, or Abbyy, or…? Thanks

  10. wealthychef October 15, 2009 at 6:00 am #

    @Paul — you rock! Incredibly simple fix — shame on ABBYY and Fujitsu for not fixing this. It is pathetic.

  11. Michael F October 16, 2009 at 2:19 pm #

    Thanks for this. A quick note — you can add two lines to the PDFInfoString.txt file to make pdftk change the PDF's creator, too. Then you can get rid of the 'pdfauxinfo' step.

    My file reads:

    InfoKey: Creator
    InfoValue: ScanSnap Manager
    InfoKey: Producer
    InfoValue: Mac OS X 10.5.7 Quartz PDFContext

  12. BrooksD October 19, 2009 at 11:56 pm #

    Hi all, just wanted to thank you so much for the awesome workarounds you've been posting to this issue. You rule!!

  13. BoilingConcrete October 22, 2009 at 6:20 am #

    Well, we got an update from Fujitsu! They say "Fujitsu is still working hard to complete the ScanSnap Snow Leopard compatibility update. We expect to have it ready to share with all of you by the end of November…"

  14. noddy67 November 4, 2009 at 2:21 pm #

    Quick question if I may. I'm trying to use the scansnap in snow leopard to regognise highlighted text, incorporate that into the pdf's metadata and then automatically generate the file name from that metadata keyword information. I have no problems getting the highlighter pen to work and to create an automator script to see that metadata. Where I'm having no luck is how to get automator to incorporate that info into the file nae.
    Looked everywhere and only have a very very basic knowledge of automator (and none of applescript) so any help with this would be much appreciated.
    Thanks in advance.

  15. Spike November 11, 2009 at 6:09 pm #

    FineReader update announced. Here's a summary from MacNN:

    http://www.macnn.com/articles/09/11/11/gains.snow

    • Tamara November 13, 2009 at 6:37 pm #

      Curious, does the updated version from ABBYY apply to the version that came with ScanSnap or is it a separate purchase? i.e. is it a free upgrade? Wondering if anyone has tried this. I have the ScanSnap s1500m, haven't upgraded our iMac to Snow Leopard, in part due to this.

  16. BoilingConcrete November 14, 2009 at 3:13 am #

    Fujitsu has announced an update for ScanSnap Manager for S1500M, S300M, S510M and S500M that will work with Snow Leopard. Check out the following website:

    <http://www.fujitsu.com/us/services/computing/peri

    At the bottom of that page, they state that ABBYY will be hosting the update to FineReader for ScanSnap, but the link isn't available yet. I imagine that it'll be posted somewhere soon…

    • Corey Ward February 16, 2010 at 3:38 pm #

      Your link is broken because you have a ">" after the link that's being included. If you can edit or delete it that might be helpful. ;)

  17. Brad January 28, 2010 at 4:20 am #

    I've been reading the fixes, but the pdftk script did not work for me. However, I ran across a very cool utility – "a-pdf". Select a whole folder of PDF files, type in the Creator and Producer to the ScanSoft settings, and off you go…PERFECT!
    For those that don't want to fool with scripts, give this a try. They have a 15-day trial.

    http://www.a-pdf.com/rename/?product-rn

    A nice side benefit to finding a fix for the bug…I am moving over from Paperport to the ScanSnap solution. I have 7 years of PDF scan files that are not in searchable PDF format (Paperport creates it's own proprietary index). Using this utility, I just reset all the metadata for these PDF files and ScanSnap Organizer is busy converting them all to searchable PDF files!

  18. Corey Ward February 16, 2010 at 3:36 pm #

    For anybody that hasn't figured this out already that is landing on this article, you can download the updated SnowLeopard version of FineReader for ScanSnap at the bottom of Fujitsu's Snow Leopard Compatibility page. Link below.

    http://www.fujitsu.com/us/services/computing/peri

    • BrooksD February 16, 2010 at 3:47 pm #

      Thanks for the comment Corey!

  19. matthew_k March 4, 2010 at 4:22 pm #

    Though the new versions work, you're still much better off scripting Fine Reader for any volume of scanning. I modified the script a little to reflect not needing to hack around incompatibilities, and now kick off my OCR jobs every night before going to sleep. It's sped up my scanning dramatically.

    My revision of the script is available here:
    http://gist.github.com/321839

    • BrooksD March 4, 2010 at 4:50 pm #

      Thanks for sharing the script Matthew!

  20. Ardath Reinwald January 17, 2011 at 4:17 pm #

    hopefully this put up doesn’t seem a number of times (it seems to freeze once i try to submit my comment.. not sure if it’s really posting), however all I really wished to say was good submit and thanks for sharing.

  21. chale February 2, 2011 at 8:58 am #

    I own a ScanSnap S510M with Abby FineReader. However I want to use fine reader to also convert a few pdfs that weren't created or scanned by the ScanSnap and am getting a message that it wont do it because it wasn't created by it. What is the easiest way for a mac user to change this to work. I have seen a few info file programs for the mac to look at the creator type but if I drop a pdf that has been created from the scanner, it doesn't give me any info as to what the creator type is or what I might need to change.
    Thanks for your ideas

    • BrooksD February 2, 2011 at 10:55 am #

      Hi chale, there are some tips from BoilingConcrete and Michael F in the comments above that take you through changing the creator, but since you have a ScanSnap S510M, you might find it easier to just use Adobe Acrobat that came with it to do the OCR for you for those non-ScanSnap-scanned PDFs.

      It is probably less headache than messing around with the PDF metadata.

  22. Son Mckentie May 3, 2011 at 2:14 pm #

    We’re a group of volunteers and opening a new scheme in our community. Your website offered us with valuable information to work on. You’ve done a formidable job and our whole community will be thankful to you.

Leave a Reply