Nope....

Hi Chris,

There shouldn't be any OCR process going on here, it's just doing a direct text extraction operation. Can you speak more of why you need to extract text here? Full text search should be able to update the search index of the document even if you don't update text, provided that text is not already present. 

Hi Justin,

Please look at case 196369 this might give you some more info

Ok, so basically the issue is the context hits? 

Hmmmm, not quite. So basically, sometimes when you import an Excel document with OCR switched on which it is by default, the users sees the import "taking a long time", when in fact what's actually happening in the background is Laserfiche is generating thousands of pages for a single page Excel document. I haven't managed to pinpoint the exact cause of these, but the solution from Laserfiche is to turn off the option above, which obviously isn't ideal as people want to content search documents saved from MS Office.

Interestingly, it's only Excel that this issue affects, Emails & Word documents don't see affected.

P.S. I did see this issue in the past affect emails, but this was back in 8.2.x but haven't seen it since then.

Oh, I saw that that was the bug - I was wondering why you needed to generate text in the first place, since the documents will be full-text searchable without it. I was curious if it was so there will be context hits. 

Yes, the full text search is useful at returning the documents, but the context hits gives that extra layer for the user (when it works). For now we just tell them to turn it off, as it's on by default (might be useful to change this to off by default).

Do we know what's going on with this, and when a fix might be available? We see this issue pop up often on the front line.

I agree Chris this should be turned off by default.  But as you state it is not ideal because you would like to see the full context hit capability.  This issue seems to be persistent. 

The problem I see Chris is having it turned off is that no office documents are OCR'd and given we can not run an overnight OCR routine, we are just building up the number of word and excel documents that someone has to manually OCR and for which they may not have access to.

We can not ask the users to once they have finished for the day, find all docs they have worked on and OCR them

Hi Paul,

I think there may be crossed wires here, so even with the setting turned off, the documents are still text searchable, it just doesn't include the context hits (so where LF puts the word in the sentence).

If you text search for these documents they will still be returned.

This of course doesn't excuse the fact that this needs to be resolved, but at least there is some functionality. 

Cheers!

Just wanted to add my voice to this issue, which we're also experiencing. Hoping for a resolution but will turn off the setting for now.