I can't search by content

[Iwen]

I install okm 3.0.
I import lots of PDFs(Chinese) which is scaned documents and dealed with OCR.
I can\'t search them by content?
But it\'s ok using okm 2.0
Why?:blink:

[pavila]

Are the PDF composed of images or text? Can you post here a sample PDF?

[Iwen]

The PDF only composed by text.
When I open the pdf file using adobe reader, I can select text in it and search word in pdf.

[jllort]

Put seomeone on OpenKM demo, all tell us the location and the search you\'re doing to test it.

[Iwen]

I login OpenKM demo using user9 and create a folder in My documents named Iwen which I put one file in. The file name is \"2.09 明达投资顾问有限公司简介.pdf\". I search content by \"明达\" which is behind \"2.09\". But no result.
I open this file in adobe reader, I can search \"明达\" in pdf.

[pavila]

OpenKM uses a generic index algorithm, and it should be tunned to work with chinese to get better search results. By the way, the document you mention is a PDF composed by images. It won\'t be indexed, only PDF with text can be indexed.

[joako]

So, I am wondering if a "searchable PDF" of a scanned image can not be searched for by content? If an OCR process is run on the PDF file I can search it in Acrobat and other applications and even select the text and copy & paste it around. I can provide a sample but right now the ones I am testing have sensitive data.

[pavila]

Sorry, but I need a sample PDF to check why the text extraction fails. If you documents have sensitive data, maybe you want to contact us and become a customer.

[joako]

I've scanned a document that couldn't have any private data in it and I've posted it over here: http://forum.openkm.com/viewtopic.php?f=4&t=4581

[pavila]

Answered in the other thread.